2018 Book DietaryPatternsAndWholePlantFo

Nutrition and Health
Series Editors: Adrianne Bendich · Connie W. Bales
Mark L. Dreher
Dietary Patterns
and Whole Plant
Foods in Aging
and Disease
Series editors:
Adrianne Bendich, Ph.D., FACN, FASN
Wellington, FL, USA
Connie W. Bales, Ph.D., R.D.
Human Dev’t, Ctr Aging Blue Zone Rm 2508
Duke Univ Med Ctr, Ctr Study of Aging
Durham, NC, USA
The Nutrition and Health series has an overriding mission in providing health
professionals with texts that are considered essential since each is edited by
the leading researchers in their respective fields. Each volume includes: 1) a
synthesis of the state of the science, 2) timely, in-depth reviews, 3) extensive,
up-to-date fully annotated reference lists, 4) a detailed index, 5) relevant
tables and figures, 6) identification of paradigm shifts and consequences, 7)
virtually no overlap of information between chapters, but targeted, inter-
chapter referrals, 8) suggestions of areas for future research and 9) balanced,
data driven answers to patient/health professionals questions which are based
upon the totality of evidence rather than the findings of a single study.
Nutrition and Health is a major resource of relevant, clinically based nutrition
volumes for the professional that serve as a reliable source of data-driven
reviews and practice guidelines.
More information about this series at http://www.springer.com/series/7659

Mark L. Dreher
Dietary Patterns and

Whole Plant Foods in
Aging and Disease
Mark L. Dreher
Chief Science Officer
Nutrition Science Solutions LLC
Wimberley, TX
USA

ISBN 978-3-319-59179-7 ISBN 978-3-319-59180-3 (eBook)
https://doi.org/10.1007/978-3-319-59180-3
Library of Congress Control Number: 2017959638
© Springer International Publishing AG 2018

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, express or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Printed on acid-free paper
This Humana Press imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
For most people, aging and chronic diseases rates are not genetically prede-
termined. It is estimated that as much as 70–80% of aging and chronic dis-
ease rates are associated with lifestyle choices. Unhealthy aging and chronic
disease rates are largely associated with excessive intake of Western energy
dense, poor nutritional quality diets and sedentary lifestyles, which lead to
overweight and obesity being the normal phenotype. Overweight and obesity
lead to increased risk of cardiovascular diseases, type 2 diabetes, nonalco-
holic fatty liver disease and chronic kidney disease, reduced healthy life
expectancy, and premature mortality. Excessive body fat is not inert but
accelerates metabolic dysfunction processes by stimulating increased sys-
temic, organ and tissue inflammation. Healthy aging and lower chronic dis-
ease rates are largely associated with adherence to healthy dietary patterns,
increasing physical activity most days of the week, achieving and maintain-
ing lean body weight and waist circumference. Unhealthy aging and higher
chronic disease rates are largely associated with adherence to Western dietary
patterns, relatively sedentary or inactive lifestyles, and a positive energy bal-
ance leading to weight gain and central obesity. Among cancer survivors a
study indicated that healthy diets are associated with a reduced mortality rate
of 20% compared to an over 40% increased mortality rate for those consum-
ing Western diets. Over the last 30–40 years, the expansion of the Western
diet and sedentary lifestyle has led to pandemic levels of obesity, chronic
disease rates, and unhealthy life expectancy in children and young and older
adults, which is progressively moving toward a global healthcare and eco-
nomic crisis, which is expected to be a critical global issue in the next 20 years.
Over the last decade, dietary and nutrition science has advanced significantly
with thousands of prospective cohort, case-control, and cross-sectional
studies, and randomized controlled trials to identify dietary and specific
healthy foods that can help healthcare professionals and government policy
makers to limit this current and growing global health crisis, but it will require
major changes in our food systems and polices and increased investment in
high-quality food and nutrition research.
The objective of this book is to comprehensively review published research
on the effects of dietary patterns and whole plant foods in aging and disease
to help increase awareness of the thousands of studies in a comprehensive,
condensed resource for health professionals to help them in addressing cur-
rent global health issues associated with unhealthy diets and lifestyles. This
book is divided into five parts focusing on the effects of dietary patterns and
v
vi Preface
whole plant foods. The first part provides an overview of lifestyle and dietary
factors associated with aging and chronic disease. The second part addresses
specific effects of dietary patterns and especially fiber-rich diets on gastroin-
testinal tract health and disease. The third part examines the effects of dietary
patterns and whole plant foods on body weight and composition regulation,
type 2 diabetes, and nonalcoholic fatty liver disease. The fourth part provides
an in-depth analysis of the effects of dietary patterns on coronary heart dis-
ease, hypertension, chronic kidney disease, stroke, and age-related cognitive
decline and dementia. The fifth part comprehensively reviews the evidence
related to dietary patterns, whole plant foods, nutrients, and phytochemicals
on breast and colorectal cancer risk and survival. Figures are extensively used
to highlight important findings. Tables provide summaries of many published
observational and intervention trials and their meta-analyses to provide spe-
cific details on the health effects of dietary patterns and whole plant foods.
This book is for dietitians, physicians, nurses, nutritionists, pharmacists, food
industry scientists, academic researchers and educators, naturopathic doctors,
health professionals, graduate and medical students, policy makers, and all
others interested in the role of healthy plant-based diets and the benefits of
individual foods in aging and disease risk.
Wimberley, TX Mark L. Dreher

Series Editor
The great success of the Nutrition and Health Series is the result of the con-
sistent overriding mission of providing health professionals with texts that are
essential because each includes (1) a synthesis of the state of the science,
(2) timely, in-depth reviews by the leading researchers and clinicians in their
respective fields, (3) extensive, up-to-date fully annotated reference lists,
(4) a detailed index, (5) relevant tables and figures, (6) identification of para-
digm shifts and the consequences, (7) virtually no overlap of information
between chapters, but targeted, interchapter referrals, (8) suggestions of areas
for future research, and (9) balanced, data-driven answers to patient as well
as health professionals’ questions that are based upon the totality of evidence
rather than the findings of any single study.
The Series volumes are not the outcome of a symposium. Rather, each
editor has the potential to examine a chosen area with a broad perspective,
both in subject matter and in the choice of chapter authors. The international
perspective, especially with regard to public health initiatives, is emphasized
where appropriate. The editor(s), whose trainings are both research and prac-
tice oriented, have the opportunity to develop a primary objective for their
book, define the scope and focus, and then invite the leading authorities from
around the world to be part of their initiative. The authors are encouraged to
provide an overview of the field, discuss their own research, and relate the
research findings to potential human health consequences. Because each
book is developed de novo, the chapters are coordinated so that the resulting
volume imparts greater knowledge than the sum of the information contained
in the individual chapters.
“Dietary Patterns and Whole Plant Foods in Aging and Disease,”
edited as well as written by Mark L. Dreher, Ph.D., is a very welcome addi-
tion to the Nutrition and Health Series and fully exemplifies the series’
goals. This volume represents a critical, in-depth review of the recent develop-
ments in the studies of dietary patterns and their role in health and disease and
concentrates on their role in affecting the health of seniors. Evaluations of the
major dietary patterns that have been examined in large survey studies and in
certain clinical trials are discussed in depth including, but not limited to, the
Mediterranean diet, Dietary Approaches to Stop Hypertension (DASH), new
Nordic, and vegetarian diets. The volume also examines the importance of
whole plant foods in disease prevention and reduction in disease severity in the
aging population. The volume is designed as an important resource for physicians
vii
viii Series Editor
in many clinical fields, nutritionists and dietitians, research and public health
scientists, and related health professionals who treat senior adult patients. The
volume provides objective and relevant information for professors and lectur-
ers, advanced undergraduates and graduates, researchers and clinical investi-
gators who require extensive, up-to-date literature reviews, instructive tables
and figures, and excellent references on all aspects of diet and health as well
as nutrition’s role in treatments of the major chronic diseases affecting the
aging population. This volume is especially relevant as the number of research
papers and meta-analyses in the clinical nutrition arena increases every year
and clients and patients are very interested in dietary components such as
whole plant foods and their components for disease prevention. Certainly, the
obesity epidemic remains a major concern especially as the comorbidities,
such as the metabolic syndrome, type 2 diabetes, hypertension, and hyperlip-
idemia, are seen more frequently in older individuals whose diets contain
lower than recommended levels of fiber-rich, whole plant foods.
Dr. Dreher, who has written every chapter in this volume, has made every
effort to provide health professionals with the most up-to-date and compre-
hensive volume that highlights the key, well-accepted nutrition information
available to date on the importance of plant-based dietary patterns for many
aspects of health and wellness. Clear definitions and distinctions are made
concerning commonly asked patient questions such as what are the differences
between the various types of dietary patterns and how do these affect the risk
of developing the major diseases including obesity, cardiovascular and cere-
brovascular diseases, and cancers. Explanations are also provided for the
numerous types of vegetable-based diets that are often questioned by clients
and patients and discussed by health professionals even among themselves as
there are many findings in this field of nutrition research that are complex.
The author of the volume, Dr. Mark L. Dreher, Ph.D., is an internationally
recognized expert in the field of dietary fiber research. Currently, he serves as
President and Chief Science Officer of Nutrition Science Solutions, LLC. He
received his education in biochemistry and agricultural biochemistry and
nutrition at UCLA and the University of Arizona. Dr. Dreher started his
career as a research scientist in medical food product development at McGaw
Laboratories in Irvine, California. Later, he served as Assistant Professor in
Food and Nutrition at North Dakota State University leading research on sun-
flower seeds, dry edible beans, and emerging grains. During his subsequent
30+ year career in the food, agricultural, and pharmaceutical industries, he
held key roles in over 150 new healthy product development projects and
food-based clinical research trials. He has authored or coauthored over 50
research journal articles and book chapters. Dr. Dreher has authored and
edited the impressive 2017 volume entitled “Dietary Fiber in Health and
Disease” that is part of the Nutrition and Health Series. Earlier, he authored
the Handbook of Dietary Fiber in 1987 and was an editor and contributor to
the second edition of the Handbook of Dietary Fiber and the Complex
Carbohydrates in Foods book. He served as the chair of the International Life
Science Institute—North American Food, Nutrition and Safety and
Series Editor ix
Carbohydrate Committees and vice chair of the Functional Foods for Health
Committee. Dr. Dreher was a member of the 1997 Joint FAO/WHO Expert
Consultation on Carbohydrate and Human Nutrition. He was a Fellow in the
National Center for Food and Agricultural Policy and Resources Leadership
Program for the Future in Washington DC. Dr. Dreher is an active member of
the American Society for Nutrition, American Heart Association, Academy
of Nutrition and Dietetics, the Nutrition Society (UK), American Chemical
Society, and Institute of Food Technology. Dr. Dreher is actively engaged in
projects and research related to the role of healthy diets, whole foods, and
phytochemicals in health, chronic disease prevention, and optimal aging.
Dr. Dreher provides extensive summaries and assessments of the major prospec-
tive cohort studies, randomized controlled trials, and relevant meta-analyses on the
role of dietary patterns and lifestyles on the health and disease status in aging. All
20 chapters contain Key Points and Key Words as well as targeted references,
useful tables and figures, and a listing of recommended readings. In addition, the
volume contains an extensive index and helpful Appendices. The volume chap-
ters are organized in a logical progression so that the reader can identify the areas
most relevant for their needs. All chapters and the entire volume are available
online and are downloadable. The book focuses on the effects of whole plant
foods on the colonic microbiota, body weight regulation, and digestive health in
the aging population. Diseases and syndromes reviewed include those of the
digestive tract including irritable bowel syndrome, inflammatory bowel disease,
and diverticular disease. The interactions between diet and metabolic diseases,
coronary heart disease, hypertension, chronic kidney disease, nonalcoholic fatty
liver disease, stroke as well as reduced cognitive functions, and cancers of the
colon and breast are examined in individual chapters.
Part I: Overview of Aging and Disease
The three introductory chapters in the first part provide readers with a broad
review of the major lifestyles and phenotypes of the aging population and then
examine the most prevalent dietary patterns and whole plant foods that are
consumed. Chapter 1 reviews the effects of aging on tissues, organs, and organ
systems that increase vulnerability to disease and give rise to the characteristic
manifestations of aging including, but not limited to, the loss of muscle (sar-
copenia) and bone mass, a decline in reaction time, compromised hearing and
vision, reduced kidney function, and elasticity of the skin. The chapter includes
key tables that identify the aging processes that affect the mind and body. We
also learn that the rate of aging varies by individual with 25–30% programmed
by heredity factors and 70–75% due to external factors such as lifestyle factors
and random events. Thus, it appears to be possible for individuals to signifi-
cantly influence their rate of aging. However, data are reviewed that indicate
that the gains in life expectancy seen during the twentieth century are decreas-
ing globally due to the rapid rise in obesity and its related diseases as well as
the increase in sedentary long-term work environments and non-exercise-
x Series Editor
related relaxation lifestyles. The chapter concludes with an in-depth examina-

tion of the components of healthy aging lifestyles and dietary patterns and
includes 17 tables and figures and over 100 relevant references.
The second chapter provides up-to-date definitions of healthy aging,
healthy diets, Western diets, and the role of each in reducing the risks of age-
related chronic diseases. Healthy aging is the absence of premature chronic
disease, lack of physical disability, and the presence of good social engage-
ment and mental health. Of the lifestyle choices, the type of dietary pattern
followed appears to have a significant effect on aging and longevity including
a lower risk of obesity and overweight, cardiovascular disease, hypertension
as well as type 2 diabetes, cognitive declines, and reductions in the risk of
developing certain cancers. The chapter includes a discussion of the major
health, environment, and lifestyle challenges that make meeting dietary rec-
ommendations difficult for the elderly. Some of the impediments to consum-
ing healthy nutrient-dense diets include losses of the ability to taste and smell,
loss of appetite, dental and chewing problems, less family support, and limi-
tations in mobility for accessing high-quality fresh whole foods. The aging-
related inefficiencies in essential nutrient absorption and utilization that result
in increased nutrient requirements are reviewed. The major research studies
are illustrated in the 21 tables and figures included in this comprehensive
chapter that also contains over 80 targeted references.
Chapter 3 describes the role of whole plant foods in maintaining the health of
aging populations. Overall, healthy dietary guidelines recommend eating
2 1/2 cups of a variety of vegetables/day; two cups of fruits, especially whole
fruits/day; six servings of total grains at ≥3 servings of whole grains/day and ≤3
servings of refined grains/day, ≥4 weekly servings of legumes (dietary pulses or
soy), and/or ≥5 weekly servings of nuts, all of which belong to the category of
whole plant foods. The chapter describes in detail the functions of the macro-
and micronutrients found in these foods and their major health effects. This
comprehensive chapter includes 29 tables and figures, as well as extensive
appendices and almost 200 excellent references that provide the reader with an
extensive body of data from well-controlled and large survey studies that consis-
tently point to the health value of whole plant foods for the aging population.
Part II: Gastrointestinal Tract
Part II contains three chapters that examine the critical issues of the importance
of the human microbiome and its role in laxation and constipation, irritable
bowel syndrome, inflammatory bowel disease, and diverticular disease. Chapter
4 reviews the key role of fiber in maintaining an optimal microbiome. Fiber is
the primary dietary energy source of the microbiota bacteria that produce short
chain fatty acids such as butyrate. Butyrate is the major energy source of nor-
mal colon cells and thus the microbiome is actively involved in maintaining the
colonic barrier defenses associated with protecting the colon from bacterial
infections such as C. difficile, decreasing the risk of inflammatory bowel
Series Editor xi
d isease, and reducing colon cancer risk. The chapter also reviews other critical
functions of the microbiota including aiding the absorption of nutrients, the
synthesis of certain vitamins, fermentation of fiber to metabolically active mol-
ecules linked to optimizing colonic and systemic immune function, and improv-
ing cardiometabolic health and glycemic control. References, tables, and
figures provide the reader with summaries of key studies in aging populations.
Chapter 5 provides more detailed descriptions of the fiber-rich dietary patterns
associated with improved laxation and reduced constipation by increasing stool
weight and bulk volume. Adequate intake of fiber from whole cereal, fruits
(including dried fruits), and vegetables and common fiber-rich food ingredients
including polydextrose, psyllium, konjac glucomannan, guar gum, and inulin
are discussed with regard to providing constipation relief. The seven compre-
hensive tables and figures provide valuable data for the reader.
Irritable bowel syndrome (IBS), examined in Chap. 6, is the most common
gastrointestinal disorder occurring in people younger than 45 years. IBS is a
chronic and relapsing functional colonic disorder characterized by abdominal
pain, bloating, distension, and other changes in bowel habits that lack visible
structural or anatomic abnormalities. Emerging research shows that the colon
of the IBS patient contains low-grade inflammation and neuronal hyperexcit-
ability. Often, there is also reduced bacteria diversity including lower levels of
butyrate producing bacteria and increased levels of pathogenic bacteria.
Certain dietary patterns and foods can be triggers for IBS symptoms. Avoidance
of certain food components, called FODMAP (fermentable oligosaccharides,
disaccharides, monosaccharides, and polyols), may reduce acute IBS symp-
toms. Chapter 6 also examines the influence of dietary patterns, foods, and
fiber on diverticular disease. Diverticulae or colonic submucosal herniated
pouches, and/or diverticulosis’ incidence increases with age affecting 5–10%
of adults under 40 years, 30% by age 50 years, and 70% by the age of 85 years.
The chapter summarizes the data from the intervention trials that looked at the
effects of fiber-rich diets, foods, and/or supplements. Six RCTs showed ben-
eficial effects of different fiber sources including fiber-rich diets, bran, bran
crisps, psyllium, and methylcellulose on symptoms and/or bowel function.
The chapter contains 144 relevant references and 16 tables and figures.
Part III: Weight Management and Related Diseases
The four chapters included in the third part review the roles of dietary pat-
terns and whole plant foods as well as dietary fiber in affecting the risk of
becoming overweight or obese during the aging process. Observational
studies have consistently shown that from 50 years of age onward, healthy
dietary patterns are inversely associated with weight gain and central obe-
sity in both men and women and a small positive energy balance of 50
kcals/day can lead to an average increase in weight of 1 pound/year; at the
same time, height and activity levels are decreasing adding further to the
increased risk of becoming overweight or obese during the sixth, seventh,
xii Series Editor
and eighth decades of life. We learn, in Chap. 7, that obesity is a complex

multifactorial disease resulting from chronic increased energy intake and
insufficient energy expenditure that is caused by many factors including,
but not limited to, genetic, environmental, lifestyle, and emotional factors
as well as age and sex of the individual. The chapter includes details of the
major healthy diets including the Mediterranean diet, the DASH diet, the
Healthy Eating Index (HEI), and other healthy eating diets and summary
tables of studies utilizing these diets and others in which obese and over-
weight individuals consumed higher than their normal levels of fiber. In
addition to the 21 informative tables and figures and over 140 references,
the Appendix reviews the foods and portions used in each of the major
healthy diets discussed in the chapter.
Chapter 8 concentrates on the composition of diets containing whole plant
foods and their role in weight control. Whole plant foods are generally associ-
ated with lower energy density, reduced obesity, and decreased chronic disease
risk than highly processed plant foods. However, these foods, including grains,
fresh and dried fruits, vegetables, nuts, and pulses, vary widely in nutrient com-
position, energy density, and physical properties. Detailed analyses of whole
plant foods and their effects on body composition are included. As examples,
randomized control intervention studies indicate that whole grains are more
effective in reducing body fat and waist circumference than in reducing body
weight or body mass index. For fruits and vegetables, cohort studies found an
association with a lower risk of weight, waist circumference, body fat gain, and
obesity especially when diets included healthier varieties of fruits and vegeta-
bles. However, higher energy dense, lower fiber fruits and vegetables promoted
weight gain. Eating lower energy dense, higher fiber and flavonoid rich fruits
and vegetables was associated with lower risk of weight gain or modest weight
loss, promoted additional weight loss in a hypocaloric diet, and helped support
weight maintenance after weight loss. The chapter includes over 100 refer-
ences, 14 tables and figures, and an informative appendix that describes the
nutrient composition of the major categories of whole plant foods.
Chapter 9 looks at the data that link certain dietary patterns that emphasize
whole plant foods reducing the risk of developing type 2 diabetes with benefi-
cial effects in older individuals with type 2 diabetes. The emphasis of this
chapter is on the senior population. The chapter reviews the prospective
cohort studies that consistently show that increased intake of diet patterns
identified as healthy that include higher than average intakes of whole foods
and provided lower glycemic loads was effective in reducing diabetes risk.
The emphasis on dietary patterns is due to their influence on various diabetes
and cardiometabolic risk factors including controlling body weight, visceral
fat, glucose-insulin homeostasis, oxidative stress, inflammation, and endo-
thelial health, lipoprotein concentrations, and blood pressure. High-quality
diets that lower diabetes and cardiovascular risks can help to control body
weight and composition. The mechanisms of action include better control of
whole body inflammatory responses that are linked to improved insulin sen-
sitivity and vascular endothelial function, reduced risk of diabetes-related
atherosclerosis, and other cardiovascular comorbidities. There are over 130
Series Editor xiii
references, 21 tables and figures, and two additional appendices that help the
reader understand the details of the major dietary patterns and nutrient con-
tent of important whole foods that are of value to patients at risk of/or with
type 2 diabetes.
Chapter 10, containing seven relevant tables and figures and 84 refer-
ences, provides an in-depth review of another chronic disease seen more
frequently in older obese/overweight individuals. Nonalcoholic fatty liver
disease (NAFLD) can progress to a more severe, related disease, nonalco-
holic steatohepatitis (NASH), which is characterized by hepatocellular
injury or hepatic steatosis and associated hepatocyte injury, inflammation,
and fibrosis. Approximately two-thirds of obese adults have NAFLD and
20% have NASH. NASH occurs more frequently in older people especially
those with sarcopenia. The primary clinical risk factors for NAFLD are
excess body weight especially abdominal fatness, insulin resistance associ-
ated with both prediabetes and type 2 diabetes, and cardiovascular disease.
Both NAFLD and NASH are also associated with the Western lifestyle and
dietary patterns including high energy dense, low fiber and nutrient dense
diets, and inactivity. The chapter includes a review of the nutrients and diets
associated with reducing the risk of these liver diseases as well as improving
treatment.
art IV: Cardiovascular and Cerebrovascular Diseases

P
and Age-Related Cognitive Function
The eight chapters in Part IV review the effects of dietary patterns and con-
sumption of whole plant foods on the aging population’s risk of developing
cardiovascular and cerebrovascular diseases including coronary heart dis-
ease, hypertension, kidney disease, stroke, and cognitive decline. Chapter 11
emphasizes coronary heart disease, as this disease is most prevalent in indi-
viduals >50 years of age, and reviews the totality of the evidence that healthy
dietary patterns with greater concentrations of whole plant foods are associ-
ated with decreased risk of coronary heart disease (CHD). These dietary pat-
terns are characterized by greater intakes of vegetables, fruits, whole grains,
low-fat dairy, and seafood, and limited intakes of red and processed meat,
refined grains, and sugar-sweetened foods and beverages compared to the
increased CHD risk associated with greater adherence to Western dietary pat-
terns. These healthy dietary patterns, including the Elderly Dietary Index, are
associated with CHD protective effects because they are reduced in energy
density and higher in fiber, healthier fatty acid profiles, essential nutrients,
antioxidants, and electrolytes, and are anti-inflammatory compared with
Western diets. Chapter 12 examines the role of whole plant foods in the
reduction of primary and secondary risks of CHD. The major causative agent
of CHD is higher than recommended levels of cholesterol in the blood that
has been implicated in the development of atherosclerotic plaques in coro-
nary vessels. Lifestyle changes recommended for those with high cholesterol
levels include adopting a diet low in saturated and trans fatty acids, incorporating
xiv Series Editor
fiber, antioxidants, plant sterols and stanols into the diet, exercising regularly,
not smoking, and maintaining a healthy weight. The prospective cohort stud-
ies are reviewed and consistently show that higher intakes of whole and mini-
mally processed plant foods (whole plant foods) including whole grains,
fruit, vegetables, legumes, and nuts and seeds are associated with reduced
CHD risk compared with lower intakes. Chapter 12 contains 12 important
tables and figures and over 150 references.
There are strong links between hypertension (elevated blood pressure
[BP]) and subsequent CHD that may be the result of high cholesterol levels
seen in both conditions. Moreover, both conditions are also found in patients
with excess body weight. Overweight and obesity are associated with
increased activity of the renin-angiotensin-aldosterone system, insulin resis-
tance, and reduced kidney function associated with salt-sensitive hyperten-
sion. Rates of hypertension are twice as likely to occur in obese (40%) vs.
normal weight (20%) individuals. Chapter 13 looks at the data related to
dietary patterns and hypertension and includes 12 tables and figures as well
as 77 references. The major factors associated with elevated BP are aging,
especially unhealthy aging associated with overweight and obesity, poor
dietary habits, inactivity or lack of exercise, and ineffective stress manage-
ment. Adherence to healthy dietary patterns, including the Dietary Approaches
to Stop Hypertension (DASH), the Mediterranean (MedDiet), Nordic Diet,
dietary guidelines-based, and vegetarian diets, is an effective strategy to help
prevent elevated BP and important adjuncts in the treatment of hypertensive
populations. Chapter 14 looks at the role of whole plant foods in controlling
hypertension. The chapter examines foods and dietary constituents associated
with hypertension and foods that are associated with risk reduction. Foods
that are low in fiber density and high in energy density and contain a higher
ratio of sodium to potassium, and have high saturated fat to polyunsaturated
fat ratios are associated with increased prevalence of hypertension and prehy-
pertension. In contrast, whole plant foods and minimally processed foods
contain higher concentrations of fiber, potassium, magnesium, carotenoids,
polyphenols, unsaturated fat, and plant proteins and are lower in sodium and
sugar compared to highly processed plant foods; these food components are
associated with normal blood pressures in older individuals as well as mod-
estly reducing elevated blood pressures. The data are compiled in 11 tables
and figures and the chapter includes over 130 references.
Hypertension and obesity are major risk factors for chronic kidney disease
(CKD) that is discussed in Chap. 15. The chapter concentrates on patients
with CKD and its stages and reviews the literature concerning the dietary pat-
terns, foods, beverages, and specific nutrient needs of CKD patients including
low phosphorus intakes. Relevant topics reviewed with regard to clinical data
available include protein sources, fibers especially from whole grains and
fruits and vegetables, minerals including sodium, phosphorous, and magne-
sium, and beverage data from studies on sugar-sweetened versus artificially
sweetened sodas, alcohol, and coffee consumption.
The next two chapters concentrate on stroke risk. Strokes are caused by a
disruption of the blood supply to the brain due to either vessel blockage as
seen in almost 90% of strokes that are defined as ischemic strokes or rupture
Series Editor xv
of a blood vessel that results in a hemorrhagic stroke. With regard to clinical

data, most are based on findings from studies involving patients with isch-
emic stroke. Chapter 16 examines the dietary patterns associated with increas-
ing the risk of stroke as well as the patterns associated with reducing its risk.
Stroke risk is linked to poor diet, low physical activity, smoking, high systolic
blood pressure, high body mass index and obesity, high fasting plasma glu-
cose, and above normal total cholesterol. The chapter examines the data link-
ing the American Heart Association’s “Life’s Simple 7” plan for ideal
cardiovascular health and lower stroke risk that includes (1) nonsmoking or
quit >1 year ago; (2) BMI < 25; (3) blood pressure (BP) <120/80 mmHg;
(4) ≥ 150 min/week of physical activity; (5) healthy dietary pattern (high in
fruits and vegetables, fish, fiber-rich whole grains), (6) low intake of sodium,
and (7) limiting or avoiding sugar-sweetened beverages, and reduced risk of
stroke seen in clinical studies. Specific healthy dietary patterns are also
reviewed and findings are presented in 11 informative tables and figures; 75
references are included in this chapter. Chapter 17, with over 100 references,
looks at the role of whole plant foods in reducing the risk of stroke. Whole
plant foods contain a variety of macro- and micronutrients and phytochemi-
cals including fibers, antioxidant vitamins, potassium, magnesium, carot-
enoids, flavonoids, and phytosterols that have been associated with reducing
stroke risk in prospective as well as intervention studies that are reviewed and
presented in 11 tables and figures. The mechanisms that have been postulated
to reduce stroke risk include promoting vascular health by attenuating ele-
vated blood pressure, lowering LDL-cholesterol levels and systemic inflam-
mation associated with atherosclerosis, and promoting better insulin
sensitivity, blood glucose control, weight control, and microbiota health com-
pared to less healthy or highly processed plant foods.
Chapter 18 reviews the studies that have included dietary patterns, whole
foods, and beverages and the risk of cognitive impairments including demen-
tia and provides over 100 references and 14 tables and figures. The chapter
includes data from systematic reviews, randomized, placebo-controlled trials,
and prospective cohort studies that support the benefits of high polyphenolic
fruits and vegetables, dairy (especially yogurt), 100% vegetable and fruit
juices (polyphenol rich), coffee, tea, flavanol-rich cocoa beverages, and
low-moderate wine consumption (1 glass/day) on improving age-related cog-
nitive performance and reducing risk of dementia. In addition to diet, exercise
is also linked to reduced risk of cognitive dysfunction as well as Alzheimer’s
disease; inversely, sedentary lifestyle and poor nutrient diets are associated
with increased risk of cognitive decline and Alzheimer’s disease.
Part V: Cancer Prevention and Survival
Chapters 19 and 20 provide objective, up-to-date reviews of the associations

between dietary patterns, whole plant foods, and the nutrients and phyto-
chemicals contained within these foods, on the risks associated with the
development of precancerous colon adenomas and/or colorectal cancer
(Chap. 19), and breast cancer (Chap. 20). The dietary risk factors for colorectal
xvi Series Editor
cancer reviewed in Chap. 19 include higher than recommended intakes of

alcohol, total dietary fat, and red meat and lower intakes of dietary fiber,
calcium and folate, isoflavones, flavonoids, antioxidant vitamins, carotenoids,
magnesium, and selenium. The chapter examines the mechanisms by which
soluble fibers may lower colorectal cancer risk including the ability of fer-
mentable fiber to lower colonic pH and inhibit pathogenic bacteria, increase
butyrogenic bacteria to promote healthy colonic mucosal cells, reduce colon
inflammation, and inhibit cancer cell proliferation and facilitate apoptosis.
Insoluble fiber may reduce the colon’s exposure to carcinogens by bulking
stools and binding carcinogens. The chapter, containing over 100 important
references, tabulates the convincing evidence that higher intakes of calcium
and fiber-rich foods reduce colorectal cancer risk and that lower intakes are
associated with an increased risk of colorectal cancer. Data are organized in
14 informative tables and figures.
Chapter 20 looks at the studies linking dietary patterns and foods with
breast cancer (BC) risk. Both positive and negative studies are included
and tabulated in 23 relevant tables and figures. The chapter, with more than
130 references, examines the types of breast cancer and the genetics behind
these differences, clinical studies on breast cancer primary prevention as
well as secondary prevention of recurrence, and importantly, dietary man-
agement for the breast cancer patient. Dietary patterns associated with
lower BC risk include diets with 45–65% energy from fiber-rich sources
including whole grains, fruits, vegetables, and legumes; 10–35% energy
from healthy dietary fats low in saturated fats; and 10–35% energy from
protein sources which are very low or devoid of processed meats. Unhealthy
dietary patterns and obesity, especially among postmenopausal women,
are associated with negative changes in biomarkers such as insulin, lipo-
proteins, and estradiol, which are risk factors for BC. Biological mecha-
nisms associated with dietary intake and increased risk of BC and breast
cancer recurrence include exposure to heterocyclic amines, lipid peroxida-
tion and systemic inflammation, and other causes of oxidative stress and
low antioxidant status. These factors are reviewed as well as lifestyle indi-
cators that are associated with increased breast cancer risk, recurrence, and
mortality.
Conclusions
Of importance to physicians, nutritionists, dieticians, researchers, nurses, and

allied health professionals who provide advice concerning diet, foods, nutri-
tion, and clinical management of nutritionally related conditions and/or dis-
eases is the identification of reputable sources of nutrition information.
“Dietary Patterns and Whole Plant Foods in Aging and Disease” provides
20 valuable chapters that review and integrate these relevant and objective
resources. The volume examines the major dietary patterns that have been
identified as containing the components of a healthy diet and contrasts these
with dietary patterns, such as Western diets and other patterns that are consid-
ered as unhealthy based upon a totality of the epidemiological evidence available
Series Editor xvii
to date. Moreover, the volume includes extensive reviews of the whole plant
foods associated with the healthy dietary patterns, and each chapter contains
detailed analyses of the nutrients and bioactive molecules contained within
these foods.
This comprehensive volume examines patient-related topics including
chapters on the major changes in lifestyle and dietary patterns, and organ
functions during the aging process and the potential for whole plant foods to
help reduce certain of the health risks associated with aging. Topics included
in this comprehensive volume include the aging effects on the gastrointestinal
tract including laxation and constipation and diseases of the colon including
irritable bowel syndrome and diverticular disease and cancer. Weight man-
agement during aging often includes dealing with being overweight or obese
and the resulting diabetes, heart, liver, and kidney diseases are reviewed in
depth. The potential for dietary patterns to reduce the risks of cognitive
decline and Alzheimer’s disease is examined and studies are tabulated. Two
cancers relevant to aging and dietary choices are reviewed in depth in indi-
vidual chapters on breast and colon cancer.
The volume contains over 300 data-rich tables and figures and appendices
as well as more than 2200 up-to-date references that give physicians and
health providers important tools that can help to alter patient dietary habits
that may be less than ideal, and chapters also review the many types of whole
plant foods that may enhance their patients’ and clients’ diets and health.
Patients and consumers are concerned about many claims that are made for
common nutrients and fibers found in foods, such as juice drinks, soy prod-
ucts, novel fruits, and organically grown vegetables. The chapters in this vol-
ume examine these and other provocative areas of diet information. There are
more than a dozen chapters that provide clinically relevant information on
risk reduction of the major chronic diseases associated with the aging pro-
cess. The 20 chapters within this valuable volume provide a wealth of timely
information for health providers, medical students, graduate students, nurses,
dietitians, and other related health professionals.
Dr. Mark L. Dreher is an internationally recognized leader in the field of
human nutrition with more than 30 years of research in the importance of
whole food intake for the reduction in risk of obesity and other critical clini-
cal outcomes reviewed in this comprehensive volume. Dr. Dreher is a proven
excellent communicator and has worked tirelessly to develop this volume that
is destined to be the benchmark in the field of clinical nutrition because of its
extensive coverage of the most important aspects of the complex interactions
between diet, foods, nutrients, bioactive food components, and health and
disease. Hallmarks of all of the chapters include complete definitions of terms
with the abbreviations fully defined for the reader and consistent use of terms
between chapters. Useful features of this comprehensive volume include the
informative Key Points and Keywords that are at the beginning of each chap-
ter and relevant references at the end of each chapter.
In conclusion, “Dietary Patterns and Whole Plant Foods in Aging and
Disease,” edited as well as written by Mark L. Dreher, Ph.D., provides
health professionals in many areas of research and practice with the most up-
to-date, organized volume on the clinically researched and documented
xviii Series Editor
healthy dietary patterns that are linked to reducing the risks of the major
chronic diseases associated with the aging process. Of great importance,
these are also the major chronic diseases that are often discussed by patients
with their healthcare providers. Thus, the data provided in this book enables
the reader to answer their patient or client questions with the confidence that
their answers are based upon the totality of the evidence from well-accepted,
data-driven nutrition research. This volume serves the reader as the bench-
mark in this complex area of interrelationships between the major dietary
patterns that have been associated with reducing the age-related risks of
unhealthy body weight, type 2 diabetes, cancer, cardiovascular and cerebro-
vascular disease, diseases of the gastrointestinal tract, liver, and kidney, and
reduced brain function. Dr. Dreher is applauded for his efforts to develop this
volume with the firm conviction that nutrition research serves as an essential
source of important data for all health professionals. This excellent text is a
very welcome addition to the Nutrition and Health series.
Morristown, NJ Adrianne Bendich, Ph.D., FACN, FASN

About the Series Editors
Adrianne Bendich, Ph.D., F.A.S.N.,

F.A.C.N., has served as the “Nutrition
and Health” Series Editor for more than
20 years and has provided leadership and
guidance to more than 200 editors that have
developed the 80+ well-respected and
highly recommended volumes in the
series.
In addition to “Dietary Patterns and Whole
Plant Foods in Aging and Disease,” edited
as well as written by Mark L. Dreher, Ph.D.,
major new editions published in 2012–2017
include the following:
1. Dietary Fiber in Health and Disease, edited as well as written by Mark

L. Dreher, Ph.D., 2017
2. Clinical Aspects of Natural and Added Phosphorus in Foods, edited by
Orlando M. Gutierrez, Kamyar Kalantar-Zadeh, and Rajnish Mehrotra,
2017
3. Nutrition and Fetal Programming, edited by Rajendram Rajkumar,
Victor R. Preedy, and Vinood B. Patel, 2017
4. Nutrition and Diet in Maternal Diabetes, edited by Rajendram
Rajkumar, Victor R. Preedy, and Vinood B. Patel, 2017
5. Nitrite and Nitrate in Human Health and Disease, Second Edition,
edited by Nathan S. Bryan and Joseph Loscalzo, 2017
6. Nutrition in Lifestyle Medicine, edited by James M. Rippe, 2017
7. Nutrition Guide for Physicians and Related Healthcare Professionals,
Second Edition, edited by Norman J. Temple, Ted Wilson, and George
A. Bray, 2016
8. Clinical Aspects of Natural and Added Phosphorus in Foods, edited by
Orlando M. Gutiérrez, Kamyar Kalantar-Zadeh, and Rajnish Mehrotra,
2016
9. L-Arginine in Clinical Nutrition, edited by Vinood B. Patel, Victor
R. Preedy, and Rajkumar Rajendram, 2016
xix
xx About the Series Editors
10. Mediterranean Diet: Impact on Health and Disease, edited by Donato

F. Romagnolo, Ph.D. and Ornella Selmin, Ph.D., 2016
11. Nutrition Support for the Critically Ill, edited by David S. Seres, MD,
and Charles W. Van Way, III, MD, 2016
12. Nutrition in Cystic Fibrosis: A Guide for Clinicians, edited by
Elizabeth H. Yen, M.D., and Amanda R. Leonard, MPH, RD, CDE, 2016
13. Preventive Nutrition: The Comprehensive Guide For Health
Professionals, Fifth Edition, edited by Adrianne Bendich, Ph.D., and
Richard J. Deckelbaum, M.D., 2016
14. Glutamine in Clinical Nutrition, edited by Rajkumar Rajendram,
Victor R. Preedy, and Vinood B. Patel, 2015
15. Nutrition and Bone Health, Second Edition, edited by Michael
F. Holick and Jeri W. Nieves, 2015
16. Branched Chain Amino Acids in Clinical Nutrition, Volume 2, edited
by Rajkumar Rajendram, Victor R. Preedy, and Vinood B. Patel, 2015
17. Branched Chain Amino Acids in Clinical Nutrition, Volume 1, edited
by Rajkumar Rajendram, Victor R. Preedy, and Vinood B. Patel, 2015
18. Fructose, High Fructose Corn Syrup, Sucrose and Health, edited by
James M. Rippe, 2014
19. Handbook of Clinical Nutrition and Aging, Third Edition, edited by
Connie Watkins Bales, Julie L. Locher, and Edward Saltzman, 2014
20. Nutrition and Pediatric Pulmonary Disease, edited by Dr. Youngran
Chung and Dr. Robert Dumont, 2014
21. Integrative Weight Management, edited by Dr. Gerald E. Mullin, Dr.
Lawrence J. Cheskin, and Dr. Laura E. Matarese, 2014
22. Nutrition in Kidney Disease, Second Edition, edited by Dr. Laura
D. Byham-Gray, Dr. Jerrilynn D. Burrowes, and Dr. Glenn M. Chertow,
2014
23. Handbook of Food Fortification and Health, Volume I, edited by Dr.
Victor R. Preedy, Dr. Rajaventhan Srirajaskanthan, and Dr. Vinood
B. Patel, 2013
24. Handbook of Food Fortification and Health, Volume II, edited by Dr.
Victor R. Preedy, Dr. Rajaventhan Srirajaskanthan, and Dr. Vinood
B. Patel, 2013
25. Diet Quality: An Evidence-Based Approach, Volume I, edited by Dr.
Victor R. Preedy, Dr. Lan-Ahn Hunter, and Dr. Vinood B. Patel, 2013
26. Diet Quality: An Evidence-Based Approach, Volume II, edited by Dr.
Victor R. Preedy, Dr. Lan-Ahn Hunter, and Dr. Vinood B. Patel, 2013
27. The Handbook of Clinical Nutrition and Stroke, edited by Mandy
L. Corrigan, MPH, RD, Arlene A. Escuro, MS, RD, and Donald F. Kirby,
MD, FACP, FACN, FACG, 2013
28. Nutrition in Infancy, Volume I, edited by Dr. Ronald Ross Watson, Dr.
George Grimble, Dr. Victor Preedy, and Dr. Sherma Zibadi, 2013
29. Nutrition in Infancy, Volume II, edited by Dr. Ronald Ross Watson, Dr.
George Grimble, Dr. Victor Preedy, and Dr. Sherma Zibadi, 2013
30. Carotenoids and Human Health, edited by Dr. Sherry A. Tanumihardjo,
2013
About the Series Editors xxi
31. B ioactive Dietary Factors and Plant Extracts in Dermatology, edited

by Dr. Ronald Ross Watson and Dr. Sherma Zibadi, 2013
32. Omega 6/3 Fatty Acids, edited by Dr. Fabien De Meester, Dr. Ronald
Ross Watson, and Dr. Sherma Zibadi, 2013
33. Nutrition in Pediatric Pulmonary Disease, edited by Dr. Robert
Dumont and Dr. Youngran Chung, 2013
34. Nutrition and Diet in Menopause, edited by Dr. Caroline J. Hollins
Martin, Dr. Ronald Ross Watson, and Dr. Victor R. Preedy, 2013.
35. Magnesium and Health, edited by Dr. Ronald Ross Watson and Dr.
Victor R. Preedy, 2012.
36. Alcohol, Nutrition and Health Consequences, edited by Dr. Ronald
Ross Watson, Dr. Victor R. Preedy, and Dr. Sherma Zibadi, 2012
37. Nutritional Health, Strategies for Disease Prevention, Third Edition,
edited by Norman J. Temple, Ted Wilson, and David R. Jacobs, Jr., 2012
38. Chocolate in Health and Nutrition, edited by Dr. Ronald Ross Watson,
Dr. Victor R. Preedy, and Dr. Sherma Zibadi, 2012
39.
Iron Physiology and Pathophysiology in Humans, edited by Dr.
Gregory J. Anderson and Dr. Gordon D. McLaren, 2012
Earlier books included Vitamin D, Second Edition, edited by Dr.

Michael Holick; “Dietary Components and Immune Function” edited by
Dr. Ronald Ross Watson, Dr. Sherma Zibadi, and Dr. Victor R. Preedy;
“Bioactive Compounds and Cancer” edited by Dr. John A. Milner and Dr.
Donato F. Romagnolo; “Modern Dietary Fat Intakes in Disease
Promotion” edited by Dr. Fabien De Meester, Dr. Sherma Zibadi, and Dr.
Ronald Ross Watson; “Iron Deficiency and Overload” edited by Dr.
Shlomo Yehuda and Dr. David Mostofsky; “Nutrition Guide for
Physicians” edited by Dr. Edward Wilson, Dr. George A. Bray, Dr. Norman
Temple, and Dr. Mary Struble; “Nutrition and Metabolism” edited by Dr.
Christos Mantzoros; and “Fluid and Electrolytes in Pediatrics” edited by
Leonard Feld and Dr. Frederick Kaskel. Recent volumes include “Handbook
of Drug-Nutrient Interactions” edited by Dr. Joseph Boullata and Dr.
Vincent Armenti; “Probiotics in Pediatric Medicine” edited by Dr. Sonia
Michail and Dr. Philip Sherman; “Handbook of Nutrition and Pregnancy”
edited by Dr. Carol Lammi-Keefe, Dr. Sarah Couch, and Dr. Elliot Philipson;
“Nutrition and Rheumatic Disease” edited by Dr. Laura Coleman;
“Nutrition and Kidney Disease” edited by Dr. Laura Byham-Grey, Dr.
Jerrilynn Burrowes, and Dr. Glenn Chertow; “Nutrition and Health in
Developing Countries” edited by Dr. Richard Semba and Dr. Martin
Bloem; “Calcium in Human Health” edited by Dr. Robert Heaney and Dr.
Connie Weaver; and “Nutrition and Bone Health” edited by Dr. Michael
Holick and Dr. Bess Dawson-Hughes.
Dr. Bendich is President of Consultants in Consumer Healthcare, LLC,
and is the editor of ten books including “Preventive Nutrition: The
Comprehensive Guide for Health Professionals, Fifth Edition,” co-edited
with Dr. Richard Deckelbaum (www.springer.com/series/7659). Dr. Bendich
serves on the Editorial Boards of the Journal of Nutrition in Gerontology and
xxii About the Series Editors
Geriatrics and Antioxidants and has served as Associate Editor for Nutrition,
the International Journal; served on the Editorial Board of the Journal of
Women’s Health and Gender-Based Medicine; and served on the Board of
Directors of the American College of Nutrition.
Dr. Bendich was Director of Medical Affairs at GlaxoSmithKline (GSK)
Consumer Healthcare and provided medical leadership for many well-known
brands including TUMS and Os-Cal. Dr. Bendich had primary responsibility
for GSK’s support for the Women’s Health Initiative (WHI) intervention
study. Prior to joining GSK, Dr. Bendich was at Roche Vitamins Inc. and was
involved with the groundbreaking clinical studies showing that folic acid-
containing multivitamins significantly reduced major classes of birth defects.
Dr. Bendich has coauthored over 100 major clinical research studies in the
area of preventive nutrition. She is recognized as a leading authority on anti-
oxidants, nutrition and immunity and pregnancy outcomes, vitamin safety,
and the cost-effectiveness of vitamin/mineral supplementation.
Dr. Bendich received the Roche Research Award, is a Tribute to Women
and Industry Awardee, and was a recipient of the Burroughs Wellcome
Visiting Professorship in Basic Medical Sciences. Dr. Bendich was given the
Council for Responsible Nutrition (CRN) Apple Award in recognition of her
many contributions to the scientific understanding of dietary supplements. In
2012, she was recognized for her contributions to the field of clinical nutri-
tion by the American Society for Nutrition and was elected a Fellow of
ASN. Dr. Bendich is Adjunct Professor at Rutgers University. She is listed in
Who’s Who in American Women.
Connie W. Bales, Ph.D., R.D., is a
Professor of Medicine in the Division of
Geriatrics, Department of Medicine, at the
Duke School of Medicine and Senior Fellow
in the Center for the Study of Aging and
Human Development at Duke University
Medical Center. She is also Associate Director
for Education/Evaluation of the Geriatrics
Research, Education, and Clinical Center at
the Durham VA Medical Center. Dr. Bales is a
well-recognized expert in the field of nutri-
tion, chronic disease, function, and aging. Over the past two decades, her
laboratory at Duke has explored many different aspects of diet and activity as
determinants of health during the latter half of the adult life course. Her cur-
rent research focuses primarily on enhanced protein as a means of benefiting
muscle quality, function, and other health indicators during geriatric obesity
reduction and for improving perioperative outcomes in older patients. Dr.
Bales has served on NIH and USDA grant review panels and is Past-Chair of
the Medical Nutrition Council of the American Society for Nutrition. She has
edited three editions of the Handbook of Clinical Nutrition and Aging, is
Editor-in-Chief of the Journal of Nutrition in Gerontology and Geriatrics,
and is a Deputy Editor of Current Developments in Nutrition.
Authors’ Biography
Mark L. Dreher, Ph.D., is president and

chief science officer of Nutrition Science
Solutions, LLC. He received his education in
biochemistry and agricultural biochemistry
and nutrition at UCLA and the University of
Arizona. Dr. Dreher started his career as a
research scientist in medical food product
development at McGaw Laboratories in
Irvine, California. Later, he served as assis-
tant professor in food and nutrition at North
Dakota State University, leading research on
sunflower seeds, dry edible beans, and emerg-
ing grains. During his subsequent 30-plus-
year career in the food, agricultural, and pharmaceutical industries, he held
key roles in over 150 new healthy product development projects and food-
based clinical research trials. He has authored or coauthored over 50 research
journal articles and book chapters. Dr. Dreher has authored or coauthored two
handbooks on dietary fiber, one book on complex carbohydrates and the 2017
Dietary Fiber in Health and Disease book. He served as the chair of the
International Life Sciences Institute—North American Food, Nutrition and
Safety, and Carbohydrate Committees—and vice chair of the Functional
Foods for Health Committee. Dr. Dreher was a member of the 1997 Joint
FAO/WHO Expert Consultation on Carbohydrate and Human Nutrition. He
was a fellow in the National Center for Food and Agricultural Policy and
Resources for the Future Leadership Program in Washington, DC. Dr. Dreher
is involved in the American Society for Nutrition, American Heart Association,
Academy of Nutrition and Dietetics, UK Nutrition Society, Institute of Food
Technologists. American Chemical Society and the American Association for
the Advancement of Science. Dr. Dreher is actively engaged in projects and
research related to the role of healthy diets, whole foods, and p hytochemicals
in health, chronic disease prevention, and optimal aging.
xxiii
Acknowledgments
I am profoundly appreciative to the hundreds of investigators who have stud-

ied and published on the effects of dietary patterns and whole plant foods in
aging and chronic disease which made this book possible.
I want to thank Dr. Adrianne Bendich, editor of Preventative Nutrition:
The Comprehensive Guide for Health Professionals, for her support, and
critical guidance and insights that inspired me at each phase of this book
project.
Finally, I am indebted to my wife Claudia, who provided love and support,
constructive criticism and insights, and for space and time that was essential
for completing this book.
xxv
Contents
Part I Overview of Aging and Disease
1 Major Lifestyles and Phenotypes in Aging and Disease�� 3

2 Dietary Patterns in Aging and Disease�� 29
3 Whole Plant Foods in Aging and Disease�� 59
Part II Gastrointestinal Tract
4 Fiber-Rich Dietary Patterns and Colonic Microbiota

in Aging and Disease�� 119
5 Fiber-Rich Dietary Patterns and Foods in Laxation
and Constipation�� 145
6 Dietary Patterns, Foods and Fiber in Irritable
Bowel Syndrome and Diverticular Disease�� 165
Part III Weight Management and Related Diseases
7 Dietary Patterns and Fiber in Body Weight and Composition

Regulation�� 195
8 Whole Plant Foods in Body Weight and Composition
Regulation�� 233
9 Dietary Patterns and Whole Plant Foods in Type 2 Diabetes
Prevention and Management�� 257
10 Dietary Patterns, Foods, Nutrients and Phytochemicals
in Non-Alcoholic Fatty Liver Disease�� 291
Part IV Cardiovascular and Cerebrovascular Diseases,

and Age-Related Cognitive Function
11 Dietary Patterns and Coronary Heart Disease�� 315

12 Whole Plant Foods and Coronary Heart Disease �� 337
xxvii
xxviii Contents
13 Dietary Patterns and Hypertension�� 371

14 Whole Plant Foods and Hypertension�� 391
15 Dietary Patterns, Foods and Beverages in Chronic
Kidney Disease �� 417
16 Dietary Patterns and Stroke Risk�� 435
17 Whole Plant Foods and Stroke Risk�� 451
18 Dietary Patterns, Foods and Beverages in Age-Related
Cognitive Performance and Dementia �� 471
Part V Cancer Prevention and Survival
19 Dietary Patterns, Whole Plant Foods, Nutrients

and Phytochemicals in Colorectal Cancer Prevention
and Management�� 521
20 Dietary Patterns, Whole Plant Foods, Nutrients
and Phytochemicals in Breast Cancer Prevention
and Management�� 557
Index �� 611

Part I
Overview of Aging and Disease
Major Lifestyles and Phenotypes
in Aging and Disease 1
Keywords
Healthy aging • Mortality • Healthy dietary patterns • Western diet •
Obesity • Body mass index • Central obesity • Physical activity • Metabolic
syndrome • Type 2 diabetes • Prediabetes • Sarcopenia
Key Points • The concept of healthy aging includes

• Since an estimated 70 to 80% of the rate of healthy life expectancy (e.g., absence or
aging is related to lifestyle choices, it is pos- delay of chronic diseases and the mainte-
sible for individuals to significantly influence nance of cognitive, physical, and other
their odds of healthy aging and longevity, even functions with limited dependence on fam-
if healthy lifestyles are adopted later in life. ily members or extended care assistant liv-
• Unhealthy or premature aging, which is largely ing) and longevity. Since a higher percentage
associated with excessive intake of Western of people worldwide are surviving to older
energy dense diets and sedentary lifestyles, ages, it is critical to promote optimal
involves a complex interplay between obesity healthy aging lifestyle habits to assure qual-
and related metabolic dysfunctional effects ity of life for aging individuals and their
leading to increased risk of chronic disease and families, and for sustainable healthcare cost
mortality, and reduced healthy life expectancy. management.
• In general, obesity, especially central adipos- • The probability of healthy aging can be sig-
ity, represents a state of accelerated aging as nificantly increased by up to 80% by follow-
adipose cells produce adipokines, which can ing a healthy lifestyle even if it is adopted in
lead to increased systemic and tissue inflam- middle age adulthood or older. These lifestyle
mation and peripheral insulin resistance. choices include: adhering to a healthy dietary
• Metabolic syndrome, type 2 diabetes and pre- pattern, increasing physical activity most days
diabetes, and sarcopenia are major unhealthy of the week, achieving and maintaining a
aging phenotypes, which can be prevented by healthy body weight and waist size, and smok-
appropriate lifestyle choices. ing avoidance.
© Springer International Publishing AG 2018 3

M.L. Dreher, Dietary Patterns and Whole Plant Foods in Aging and Disease, Nutrition and Health,
https://doi.org/10.1007/978-3-319-59180-3_1
4 1 Major Lifestyles and Phenotypes in Aging and Disease
1.1 Introduction During the twentieth century, life expectancy

at birth increased from 50 years to ≥80 years in
Aging is the biological progressive deterioration countries such as Japan, Western Europe, the
of physiological functions and metabolic pro- USA, Canada, Australia, and New Zealand [2, 6].
cesses leading to chronic diseases such as neuro- Much of the early improved life expectancy was
degenerative disorders, cardiovascular disease due to reduced childhood mortality and treatments
(CVD), type 2 diabetes (diabetes) and cancer, for infectious diseases. Since the mid-twentieth
and ultimately to death [1–3]. Aging is the accu- century, life expectancy gains have been due
mulation of random damage to the body’s DNA, mainly to reduced mortality at birth, childhood
structural and regulatory proteins such as hor- and older age, cleaner drinking water, advances in
mones and enzymes, or from excessive metaboli- health care (e.g., vaccines, antibiotics, preventive
cally dysfunctional adipose tissue, that begins and therapeutic medicines, devices and surgeries),
early in life and eventually exceeds the body’s and better hygiene, nutrition, housing and educa-
self-repair capabilities [2, 3]. This damage gradu- tion. With this increased life expectancy, the world
ally impairs the functioning of cells, tissues, population is expected to double from 6 billion
organs, and organ systems, thereby increasing the people in 2000 to 11.5 billion people by 2100 with
risk of chronic diseases/frailty and quality of life increasing aging demographic changes in every
markers of aging, such as loss of muscle and region and across most socioeconomic groups [7,
bone mass, a decline in reaction time, and reduced 8]. The proportion of the population that is
hearing, vision, and skin elasticity. However, ≥65 years old has been rising and is projected to
aging processes are highly malleable and influ- increase from 8% (550 million people) in 2010 to
enced by lifestyle factors that can improve the 21% (>1.5 billion people) by 2050. Increased
odds of healthy aging, even if they are adapted in rates of chronic disease, disability and frailty are
mid- or older age [2]. Since the rate of aging var- common in this age group with unhealthy aging.
ies by individual with 20–30% programmed by There is the potential for a significant global
heredity factors (e.g., rate of cellular senescence), decline in life and healthy life expectancy with a
and 70–80% due to external factors such as life- major global health crisis expected by 2040 or
style factors and random events, it is possible for sooner, as today’s children and young adults
individuals to significantly influence their rate of develop pandemic rates of obesity, diabetes,
aging [1]. Aging is measured by: (1) chronologi- chronic liver diseases, and other cardiometabolic
cal age (longevity, life expectancy), the absolute diseases that normally occur later in life [1, 6–15].
years a person lives and (2) biological age (health The US annual rate of increased life expectancy is
span or healthy life expectancy), the healthy forecasted to decline by half between 2015 and
quality of the aging process. Common physical 2040 [6, 7]. Currently, Americans have shorter and
and physiological changes associated with aging less healthy lives than their peer populations in
are summarized in Table 1.1 [1]. An overview of most other high-income countries with the US
the gradual transition during aging from homeo- ranked 32nd [14]. Worldwide, the primary chronic
stasis to loss of function due to age-related dis- diseases are CVD (e.g., heart attacks and stroke),
eases are presented in Table 1.2 [1–5]. Major cancer, chronic respiratory disease and diabetes,
biological processes linked to the rate of aging which account for 80% of all chronic disease
include: elevated oxidative and inflammatory deaths and disability [11–15].
stress, genomic instability, telomere attrition, Ideal healthy life expectancy would be to
epigenetic alterations, mitochondrial dysfunc- delay for as long as possible chronic disease
tion, cellular senescence, stem cell exhaustion, and physical or mental disability, which have
progenitor cell dysfunction, and altered intercel- detrimental effects on individual’s and their
lular communication [5]. family’s quality of life, and adds burden and
1.1 Introduction 5
Table 1.1 Common physical and physiological changes occurring during the aging process [1]
Category Body function Changes with aging
Physical capability Strength, locomotion, – Physical capability declines progressively in later life with men
balance and dexterity performing better than women at all ages
– Low performance on grip strength, walking speed, chair rise time
and standing balance tests are associated with higher mortality
rates
Physiological Lung function, body – Beginning at about 25 years of age, forced expiratory volume
function composition (including declines at approximately 32 mL/year in men and 25 mL/year in
bone mass and skeletal women
muscle), cardiovascular – Bone mass declines with age, and bone mass or density predicts
(CV) function and risk for future fractures and mortality
glucose metabolism – Large waist size, greater BMI and weight-gain in middle age are
all associated with higher mortality or poor health status
– Declining skeletal muscle mass is associated with increased
functional impairment and disability (sarcopenia)
– Elevated blood pressure and blood lipids are the strongest
predictors of CV morbidity and mortality
– Diminished glucose homeostasis is associated with raised fasting
blood glucose, CV events and mortality
Cognitive function Memory, processing – Executive function is affected by aging, with an inverted U-shape
speed and executive pattern across the lifespan
function – Processing speed declines progressively with age and is
associated with greater mortality risk, and cardiovascular and
respiratory diseases
– Episodic memory is sensitive to brain aging and declines with
mild cognitive impairment and neurodegenerative diseases
– Higher blood pressure in midlife is associated with cognitive
decline in senior years
Endocrine system Hypothalamic, – Decreases with aging: aldosterone, calcitonin, growth hormone,
function pituitary, adrenal axis; renin; in women estrogen and prolactin and in men testosterone
thyroid, sex and growth – The pituitary gland gradually becomes smaller with aging
hormones – Thyroid hormone is often reduced especially in women
– Increases with aging: follicle-stimulating hormones, luteinizing
hormone, norepinephrine, and parathyroid hormone
Immune function Immunosenescence Aging is associated with increased systemic inflammatory
cytokines (plasma concentrations of IL-6 and TNF-α), which are
associated with lower grip strength and gait speed
Sensory function Hearing, vision, smell, – Most sensory functions, with the exception of pain, decrease
and pain across the lifespan
– Sensory changes may overlap with changes in cognitive and
motor functions with loss of audition and vision being the most
prominent
– Smell acuity declines with age, especially in men, and is thought
to be an indicator of brain integrity in older people as smell
dysfunction is among the earliest signs of neurodegenerative
diseases such as Alzheimer’s disease and Parkinson’s disease, and
is associated with mortality
cost to health care systems and long-term care intake and sedentary habits (positive energy
facilities [9, 16]. However, since the 1970s and balance), the global obesity pandemic has been
1980s, as a result of increased adherence to a primary cause of a significant reduction in the
unhealthy lifestyles leading to excessive energy odds of healthy life expectancy. This global
Table 1.2 Overview of the gradual loss of function and transition to aging related diseases [1–5]
Diminishment of functions: Age-related diseases:
Cardiometabolic dysfunction Atherosclerosis
Extracellular and intracellular damage Stroke
Oxidative DNA damage Cancer
Degraded regenerative and cell Diabetes
proliferation ability Macular degeneration
Decrease or loss of beta-cell function Aging biochemical, physiological Arthritis
Reduced immune function and pathology transition Osteoporosis
Increased fibrosis Alzheimer’s disease
Loss of bone, muscle and physical Sarcopenia
degradation Metabolic syndrome
Age-related cognitive decline Non-alcoholic fatty liver disease
Weight and body composition changes Chronic kidney disease
Shorter telomeres and epigenetic changes
trend in excessive weight gain, especially when involves a complex interplay between obesity
seen as central obesity, has been a major factor and related metabolic dysfunctional adverse
in the increased rates of chronic diseases and effects leading to increased risk of chronic dis-
related disabilities not only just in older adults ease and mortality risk, and reduced healthy life
but also in children and young adults [8–20]. A expectancy and is especially associated with sev-
2016 systematic review (164 cohort studies; 5 eral specific accelerated aging phenotypes
to 36 years of follow-up) found consistent evi- including metabolic syndrome, diabetes and pre-
dence that higher adherence to healthy behav- diabetes, and sarcopenia [1–15].
iors such as healthy diets and recommended
levels of physical activity were associated with
successful aging and reduced disability, demen- 1.2.1 Obesity
tia and frailty in later life [13]. A Greek study of
older adults (2663 adults aged 65–100 years) 1.2.1.1 Metabolic Dysfunction
found an inverse association between high Metabolic dysfunction associated with the over-
energy intake and successful aging [17]. weight and obesity pandemic is a major global
Adipose cells are not inactive fat storehouses of concern now and if not addressed it will become
excessive energy intake, but part of an active a major global crisis by 2040 [1, 6–13]. It is esti-
endocrine organ that can speed up the aging mated that in the US almost 40% of adults and
process [15, 18, 19]. The Merck Manual identi- 20% of children are obese [21]. In general, the
fied three primary strategies that might help excessive fat mass associated with obesity leads
people live healthier and longer lives including: to a state of accelerated aging as adipose cells
body weight control, energy controlled healthy produce adipokine cell signaling messengers,
diets, and exercising most days of the week such as leptin, adiponectin, IL-6, and TNF-α,
[20]. The objective of this chapter is to compre- which can influence systemic and tissue inflam-
hensively review the association of lifestyle mation, oxidative stress, and insulin resistance
factors and related phenotypes on aging, dis- [15, 18, 19, 22]. This metabolic stress can lead to
ease and mortality risk. cellular and systemic dysregulation associated
with increased risk of developing insulin resis-
tance, β-cell dysfunction, diabetes, atherosclero-
1.2 Unhealthy Aging sis, tumorigenesis, or neurodegenerative
disorders such as age related cognitive decline or
Unhealthy or premature aging, which is largely Alzheimer’s disease [1, 4, 5, 18, 19, 22]. A
associated with excessive intake of Western French prospective study (2,733 participants;
energy dense diets and sedentary lifestyles, baseline age range 45 to 60 years; 23 years of
1.2 Unhealthy Aging 7
follow-up) found that increased body mass index follow-up with first 5 years excluded; 66,552
(BMI) above normal levels was negatively asso- deaths) found that BMI and all-cause mortality
ciated with healthy aging [22]. Compared to nor- risk followed a U-shaped curve with extreme
mal BMI, individuals with obese BMIs in leanness and obesity associated with increased
mid-life had a significantly 32% higher risk of mortality risk for both genders [28, 29].
unhealthy aging compared to 9% for those who Compared to the normal BMI (18.5–24.9), each
were overweight at mid-life. Emerging research increased 5 BMI units was on average associated
suggests that increased BMI in the obese range is with about 30% higher all-cause mortality with
associated with structural brain changes includ- median survival age reduced for stage 1 obesity
ing reduction in gray matter and white matter (BMI 30–34.9) by 2–4 years and stage 2–3 obe-
alterations of the prefrontal regions, which could sity (BMI >35) by 8–10 years. Several other
increase risk for cognitive decline during aging meta-analyses have projected that an increase in
[23]. Also, several systematic reviews and meta- 5 BMI units can significantly increase the risk of
analyses have concluded that generally all obese diabetes during early weight gain by 207% and
individuals are at an increased risk of CVD com- later weight gain by 112% [30]. Also, an increase
pared with normal-weight healthy individuals, in 5 BMI units was associated with a 20% higher
especially in studies with ≥10 years of follow-up risk of prostate cancer death or recurrence by
[24–26]. Some studies suggest that there is an 20% [31]. Another meta-analysis (8 prospective
intermediate phenotype known as metabolically studies published between 1999–2014; 5.8 mil-
healthy obese, without metabolic syndrome and lion participants; 582,000 deaths) found for all-
the CVD risk typically associated with metaboli- cause mortality rates for normal weight
cally unhealthy obese [27]. One potential mecha- individuals (18.5–24.9 BMI) were higher than
nism associated with the “metabolically healthy those who were overweight (25–29.9 BMI) or
obese” phenotype may be a lower fasting respira- grade 1 obesity and as expected the highest sig-
tory quotient, which may lower the risk of insulin nificantly increased rates were observed for
resistance. However, a meta-analysis (8 cohort those with grades 2 and 3 obesity (35–40 plus
studies; 61,386 participants; 3988 CVD events) BMI) [32]. This analysis has been critically
found that overweight and obese metabolically reviewed by other investigators who suggest that
healthy individuals still had a 21–24% increased there are possible methodological confounders
risk for CVD events compared with metaboli- in this analysis, especially related to smoking,
cally healthy normal-weight individuals [26]. reverse causation due to existing chronic dis-
Consequently, most obese individuals appear to ease, and non-specific loss of lean mass and
be at increased risk for adverse long-term function in the frail elderly [18, 32]. A system-
aging health outcomes. atic review and meta-analysis (230 cohort stud-
ies; 3.74 million deaths among 30.3 million
1.2.1.2 All-Cause Mortality Risk participants) found a 15% increase in all-cause
mortality for each 5-unit increase in BMI among
ody Mass Index (BMI)
B never smokers [33]. Overall this analysis of
The effect of BMI on mortality risk has been cohort studies showed that both overweight and
extensively evaluated in prospective cohort stud- obesity increase the risk of all-cause mortality
ies. Generally, BMIs at the extremes of under- with a J shaped dose-response relation with the
weight or obesity are associated with increased lowest risk at 23–24 BMI range among never
all-cause and disease specific mortality risk, smokers. The 2016 pooled analysis of the
with some variability in older age, ethnicity or Nurses’ Health Study (1980–2012) and Health
timing of the weight gain or loss. A collaborative Professionals Follow-up Study (1986–2012)
analysis (57 prospective studies; 894,576 par- (74,582 women and 39,284 men; up to 32 years
ticipants; mostly in Western Europe and North of follow-up) found a U-shaped relation
America; 61% male; mean recruitment age between BMI and premature mortality with the
46 years; mean BMI 25; mean 13 years of lowest all-cause mortality risk between a BMI of
Number of Healthy Lifestyles Adopted

None One Two 3 or 4
1.5
Hazard Ratio of All-Cause Mortality

1.3
1.1
0.9
0.7
0.5
0.3
18.5-22.4 22.5-24.9 25.0-29.9 >= 30
BMI Range
Fig. 1.1 Risk of all-cause mortality as a function of combined body mass index (BMI) and number of healthy lifestyle
factors* from the pooled data of the Nurses’ Health Study (1980–2012) and Health Professionals Follow-up Study
(1986–2012) (adapted from [34]). *No smoking, exercise ≥30 min/day at moderate or vigorous intensity, healthy diet
index, and low to moderate alcohol intake (5-15 g in women and 5–30 g in men)
22.5–27.4 [34]. The higher the number of low attributed to increased central adiposity, which
risk lifestyle behaviors adapted the greater the is not measured by BMI and not a result of
reduction of mortality risk across the BMI range some protective physiological benefit of higher
(Fig. 1.1) [34]. This analysis found that those body fat.
who were underweight or obese in mid-life had
an increased risk of premature death over 10 to 1.2.1.3 Body Composition
30 years of follow-up. A 2014 meta-analysis of BMI can only estimate body composition and
older adults (32 studies; 197,940 subjects; base- does not accurately reflect central abdominal fat
line age ≥ 65 years; average 12-year follow-up) mass or fat-free mass, which includes skeletal
demonstrated a U-shape curve for the effect of muscle mass [35, 36]. A high BMI reflects high
BMI on all-cause mortality risk with a relatively fat free mass and fat mass in women but only
broad lower risk base of 24–30 BMI with the high fat mass in men. Additionally, body com-
lowest risk between a BMI of 27.0–27.9 [35]. position and central obesity may affect mortality
Also, BMIs of ≤20 had at least a 28% greater risk independently of BMI [19]. A loss of fat
mortality risk than those at the reference BMI free mass significantly increases the risk of mor-
of 23.0–23.9. A 2017 UK cohort study (130,473 tality in adults ≥65 years by 102% with greater
participants; mean baseline age 64 years and effects in men than women [37]. The Women’s
BMI 27; 52% women; mean follow-up 6.5 Health Initiative (10,525 postmenopausal
years) found that non-smokers with increased women; age 50–79 years; 13.6 years of follow-
central adiposity and normal or overweight up) observed that evaluating body composition
BMIs had an increased in mortality risk of 33% provides a more robust assessment of mortality
and 41%, respectively [36]. This study suggests risk among postmenopausal women than BMI
that the paradoxical increase mortality risk in [38]. Specifically, among women 50–59 years,
normal BMI participants later in life compared higher fat mass increased risk of death by 144%
to overweight or moderately obese BMI partici- and higher lean body mass reduced the risk of
pants observed in some studies may be largely death by 59%, whereas this relationship begins
All-cause CVD Cancer

2.2
2
Risk Ratios for Mortality in Women
1.8
1.6
1.4
1.2
0.8
<28 28-29 30-31 32-34 > 35
Waist Circumference (inches)
Fig. 1.2 Association between increasing waist circumference and mortality risk in women from the Nurses’ Health
Study (p-trend <0.001 for all; multivariate adjusted) (adapted from [39])
to be reversed in women >70 years with higher defined by an absence of major chronic diseases
fat mass lowering mortality risk. The Nurses’ and physical, cognitive, and mental limitations
Health Study (44,000 women; mean age (Fig. 1.3) [43]. The Osteoarthritis Initiative
52 years; 16 years of follow-up; 3507 deaths) study (2378 participants; mean age 68 years;
found that higher abdominal adiposity (waist 62% women; 6-year follow-up) observed that
circumference) was strongly and positively obese participants had worse physical activity
associated with all-cause, CVD, and cancer scores, lower quality of life, and higher risk of
mortality independently of BMI (Fig. 1.2) [39]. disability than those who were overweight or
Additionally, central obesity was a better indica- normal weight [44]. A Finnish prospective
tor than BMI of increased risk of nonalcoholic study (6542 middle-aged Helsinki city employ-
fatty liver disease [40], aortic stiffness [41] and ees; mean age 50 years; 32% of women and
atrial fibrillation [42]. 46% of men were overweight and 15% of both
women and men were obese; mean follow-up of
7.8 years) observed that obese employees had a
1.2.2 Unhealthy Life Expectancy significantly elevated risk for disability at
retirement, especially for musculoskeletal
Elevated BMI or excessive abdominal fat dur- causes of lowered physical functioning
ing early and middle adulthood increases the (Fig. 1.4) [45]. The Women’s Health Initiative
odds of unhealthy life expectations with signifi- study (36,611 participants; mean age 72 years;
cant limitations or disability in older ages. The follow-up 14–19 years) observed a significant
Nurses’ Health Study (17,065 women who sur- increased risk of mobility disability in over-
vived until at least the age of 70) found that weight women by 60% and obese women by
weight gain after age 18 years, especially in 220–570% compared to normal weight women
early adulthood and middle age, was signifi- [46]. Other studies in adults age ≥ 60 years
cantly associated with a linearly decreased show that elevated waist circumference is asso-
probability of healthy survival at age ≥ 70, as ciated with lower quality of life and a decline in
1.4
Odds of Healthy Survival at 70 Years or Older

1.2
0.8
0.6
0.4
0.2
0
Loss of > 4 Stable 4.0-9.9 10-14.9 15-19.9 >20
(Gain) (Gain) (Gain) (Gain)
Weight Change (kg)
Fig. 1.3 Odds of healthy survival at age 70 years or older associated with weight change since age 18 years from the
Nurses’ Health Study (p-trend <0.001) (adapted from [43])
All-cause disability Musculoskeletal diability

2
1.8
1.6
Hazard Ratio
1.4
1.2
0.8
20-24.9 25-29.9 30-34.5 > 35
BMI
Fig. 1.4 Association between BMI and odds of disability at retirement (p-trend <0.01; multivariate adjusted) (adapted
from [45])
physical function [47] and weight loss is asso- trolled trial (RCT) [67 obese adults; mean BMI
ciated with improved mobility [48]. Although 37; mean age 68 years; traditional vs. high pro-
obesity is associated with significant functional tein (30 g per meal) weight loss regimen;
limitations in older adults, there have been con- 6-months] showed that both diets significantly
cerns that weight loss diets could diminish reduced weight, but the protein group also sig-
muscle mass along with fat mass to increase nificantly improved physical function by over
frailty complications [49]. A randomized con- twofold.
Healthy (p-trend =.33) Western (p-trend <.001)

1.6
Relative Risk of Diabetes in Women 1.4
1.2
0.8
0.6
Q-1 Q-2 Q-3 Q-4 Q-5
Dietary Pattern Adherence Score
Fig. 1.5 Association between dietary pattern score and type 2 diabetes (diabetes) risk in women from the Nurses’
Health Study, multivariate adjusted (adapted from [50])
1.2.3 Western Dietary Patterns odds of healthy aging by about 50% for moderately
energy restricted healthy dietary patterns, after full
Several large prospective studies show that high multivariate adjustments [53]. Also, high adher-
adherence to Western dietary patterns, especially ence to a Western dietary pattern is associated with
those with the lowest dietary fiber levels, increase poorer cognitive outcomes such as lower process-
the risk of weight gain, unhealthy aging and prema- ing speed and executive function and an overall
ture death especially when consumed during increased rate of cognitive decline compared to
midlife and older years compared to healthy dietary diets rich in fruit, vegetables, and other plant-based
patterns [50–53]. The Nurses’ Health Study food items which conferred cognitive benefits [54].
(69,554 women; age 38–63 years; 14 years of fol-
low-up; 2699 incident cases of type 2 diabetes)
found that high adherence to a Western pattern sig- 1.2.4 Sedentary Lifestyle
nificantly increased risk of diabetes (Fig. 1.5) [50],
and increased risk of cardiovascular disease, can- A sedentary lifestyle throughout adulthood is
cer, and all-cause mortality by 20% [51]. The associated with an increased risk of obesity,
British Whitehall II cohort study (5350 partici- chronic diseases, disability, frailty and premature
pants; mean age 51 years; 71% men; 16 years of death with aging [55–59]. A meta-analysis (6
follow-up) observed that high adherence to cohort studies; 595,086 adults; mainly female,
Western-type diets significantly lowered odds of middle-aged or older adults from high-income
ideal aging (a composite of cardiovascular, meta- countries; 2.8 to >8 years of follow-up; 29,162
bolic, musculoskeletal, respiratory, mental, and deaths) found a non-linear association with sitting
cognitive functions), independently of other health time and all-cause mortality risk with a 2%
behaviors (Fig. 1.6) [52]. In the French SU. increase for each hour of daily sitting time up to
VI.MAX ((SUpplementation en Vitamines et 7 h/day and 5% increase for each 1-h increment in
Mineraux AntioXydants) study (2796 participants; daily sitting above 7 h/day, after adjusting for the
mean age 52 years; mean 13.3 years of follow-up), protective effects of physical activity [56]. A large
Western diets produced a 20% lower odds of Taiwan prospective cohort study (416,175 adults;
healthy aging compared to significantly improved 52% women; various ages; 8-year follow- up)
Healthy diet (p =.68) Western diet (p =.02)

1.2
1.1
Odds Ratio of Ideal Aging
1
0.9
0.8
0.7
0.6
0.5
0.4
Low Moderate High
Dietary Pattern Adherence Score
Fig. 1.6 Association between dietary pattern type and the odds of ideal aging from the British Whitehall II Study
(multivariate adjusted) (adapted from [52])
Mobility disability All-cause mortality

6
4
Relative Risk
0
High Moderate Low Inactive
Recalled Physical Activity Level
Fig. 1.7 Level of physical activity during early and middle adulthood and risk of mobility disability and all-cause
mortality at 75 years of age (p > 0.01; multivariate adjusted) (adapted from [58])
found that individuals who were inactive had a participants; mean age 66–70 years; 6-year follow-
17% increased risk of all-cause mortality com- up) found that more physical activity maintained
pared with individuals exercising for only 15 min/ healthy physical function through increased mus-
day [57]. The 2015 Finnish InCHIANTI study cle strength in older adults with or at risk for osteo-
(1149 participants; mean age 75 years; 10 years of arthritis [59]. A meta-analysis (92,986 participants;
follow-up) observed that recalled physical inactiv- approx. 90% men; mean age 51 years; mean
ity during younger and middle adulthood was at 11.4 years of follow-up) demonstrated that unfit
age 75 years significantly associated with higher individuals had twice the mortality risk regardless
rates of mobility disability and premature death of BMI as more fit participants [60]. Also, each 1
compared to those who had been physically active metabolic equivalent decrease in cardiorespiratory
(Fig. 1.7) [58]. The Osteoarthritis Initiative (2252 fitness increased mortality risk by 13%.
1.3 Major Unhealthy Aging Phenotypes 13
1.3 ajor Unhealthy Aging

M in reducing the risk of metabolic syndrome [70,
Phenotypes 71]. Metabolic syndrome is also associated
with increased risk of non-alcoholic fatty liver
Three major phenotypes responsible for acceler- disease which is another important factor asso-
ated aging and premature death are: (1) the meta- ciated with unhealthy aging.
bolic syndrome; in which excess abdominal body
and ectopic fat drive cardiometabolic dysfunc-
tion, (2) type 2 diabetes (diabetes) and prediabe- 1.3.2 T
ype 2 Diabetes
tes due to relative insulin deficiency or impaired and Prediabetes
effectiveness of insulin action, and (3) sarcopenia
or sarcopenia plus obesity; age-related loss of The prevalence of prediabetes and diabetes has
lean muscle mass due to inactivity and inade- increased globally in parallel with the rising lev-
quate protein intake [61–67]. els of obesity in adults and children, a phenome-
non sometimes called diabesity [66, 67, 72]. If
this global trend continues, by 2030 about one
1.3.1 Metabolic Syndrome billion people will have prediabetes or diabetes.
Diabetes is a threat to healthy aging as its preva-
Metabolic syndrome occurs in 20–40% of the lence increases during aging, with 26% of
worldwide adult population [64]. It is a major Americans 65 years or older having diabetes
and escalating public-health challenge as a compared with about 9.0% in the general popula-
result of urbanization, surplus energy intake, tion. Diabetes is a major risk factor for premature
increasing obesity, and sedentary life habits onset of multiple age-related conditions, includ-
[68]. Metabolic syndrome causes a twofold ing renal dysfunction, cardiovascular disease,
increase in risk of CVD, CVD mortality, and stroke, impaired wound healing, infection,
stroke, and a 1.5-fold increase in risk of all- depression, and cognitive decline [66]. Diabetes
cause mortality over 5–10 years compared with is preventable through the practice of a healthy
those without the syndrome, regardless of a lifestyle, including weight loss/control, regular
previous history of cardiovascular events [69]. exercise, and modification of an unhealthy to a
Its features are central obesity and ectopic fat healthy diet, but weight control offers the greatest
infiltration into the skeletal muscle, liver, heart, benefit [67, 73, 74]. A meta-analysis of prospec-
pancreas, and kidney; atherogenic dyslipid- tive cohort studies found that compared to healthy
emia (elevated triglycerides, elevated apolipo- normal-weight adults, the risk for diabetes
protein B and reduced high-density lipoprotein); was increased by 300% to 800% depending on
elevated blood pressure; elevated glucose lev- obesity classification or degree of metabolic
els; and proinflammatory and pro-thrombotic dysfunction [75].
states, which are strongly associated with the Individuals with prediabetes have blood glu-
positive energy balance often found with the cose levels above normal but below the thresh-
Western lifestyle [61, 69]. One of the defects old for diabetes [66, 67]. Prediabetes is
associated with ectopic fat in metabolic syn- associated with subclinical transition to insulin
drome is an excess production of reactive oxy- resistance and β-cell dysfunction, and early
gen species generated by mitochondria, or from complications associated with damage to kidney
other sites within or outside the cell, causing and nerves. For prediabetic individuals, identifi-
damage to mitochondrial components and initi- cation and treatment with a healthy diet and
ating degradative processes including increased physical activity to prevent weight gain or pro-
inflammation. Metabolic syndrome is strongly mote weight loss is the cornerstone of diabetes
related to unhealthy aging. Weight loss, exer- prevention. A 2017 meta-analysis (50 RCTs)
cise and healthy dietary patterns, including found that adapting healthy lifestyle habits
fiber-rich diets, have been shown to be effective reduced the risk of progression from prediabetes
to diabetes by 36% (range 28% to 43%) in inter- are a growing portion of the worldwide popula-
ventions with durations lasting from 6 months tion, it is critical to figure out how to promote
to 6 years [76]. guidance for healthy aging lifestyles in this popu-
lation to assure better quality of life and to help
control the expected surge in healthcare expenses.
1.3.3 Sarcopenia From a clinical and public health perspective,
maintaining a healthy weight through diet and
Adults, especially those who are inactive and physical activity is the foundation of the preven-
consume inadequate protein, tend to have a tion of chronic diseases and the promotion of
3–8% reduction in lean muscle mass per decade healthy aging. In Western countries, such as the
after the age of 30 years, [64, 77–81]. If this US, only a small fraction (< 10%) of the popula-
muscle mass loss progresses unchecked, sarco- tion is both at normal weight and practicing mul-
penia (loss of muscle and function, body water, tiple healthy lifestyles (i.e., eating a healthy diet,
and bone density) develops in approximately exercising regularly, and not smoking) [19]. The
30% of individuals over 60 years of age and adherence to healthy lifestyles throughout adult-
≥50% of those over 80 years [79]. In the later hood can promote healthy aging processes by
stages, sarcopenia or sarcopenia obesity (fat helping to increase lean muscle mass, healthy
replaces much of the muscle) results in accel- mitochondrial biogenesis (e.g., in the liver and
erated aging including increased physical muscle cells) and genome stability. Age protec-
dependence, risk of chronic disease, inflamma- tive nutrition and phytochemical dietary quality
tion, insulin resistance, and endocrine dysfunc- are important in lowering diabetes and other met-
tion [64, 77]. Even short term reductions in abolic, cardiovascular and neurological disease,
activity, such as a reduction of daily step counts and oxidative and inflammatory stress risks [82].
from 6000 to 1500, over time can reduce insu- The Finnish nutritional guidelines emphasize that
lin sensitivity and lean mass, and increase fat due to the impact of good nutrition on health and
mass in both young and older adults within well-being in later life, nutrition among older
2 weeks [79]. A minimum of 7000 steps/day people should be given more attention [83].
and adequate protein intake is recommended to These guidelines emphasize the importance of
lower sarcopenia risk [80]. Although a high- adequate intake of energy, protein, fiber, other
quality protein supplementation of about 25 g nutrients such as vitamin D supplementation and
divided between breakfast and lunch/day for fluids plus regular physical activity. In addition,
24 weeks in healthy older adults resulted in a weight changes, oral health, and constipation
positive (+0.6 kg) difference in lean tissue should be monitored.
mass compared with an isoenergetic, non-pro-
tein control [80], 25–30 g of high quality pro-
tein per meal has been proposed to maximize 1.4.1 Healthy Body Weight
muscle protein synthesis and lean body mass
with aging [77, 78]. Prospective cohort studies generally demonstrate
that maintaining normal body weight is impor-
tant for healthy aging, but this can vary with age
1.4 Healthy Aging and body composition. A systematic review and
meta-analysis (19 prospective studies; 1.5 mil-
The concept of healthy aging includes healthy lion white adults; median age 58 years; median
life expectancy (e.g., absence or delay of chronic BMI 26; median 10-year follow-up; 160,087
diseases and the maintenance of cognitive, physi- deaths) demonstrated that all-cause mortality
cal, and other functions with limited dependence was lowest with a normal BMI range of 20.0–
on family members or extended care assisted liv- 24.9 whereas overweight and obese BMI range
ing) and longevity [19]. Since older individuals individuals were associated with significantly
1.4 Healthy Aging 15
increased death rates [84]. However, a meta- Maintaining normal weight in midlife (45–
analysis of older adults (32 observational stud- 65 years) is an important predictor of healthy
ies; 197,940 adults ≥65 years of age; average aging as obesity is one of the underlying causes
follow- up of 12 years) found that in older of frailty. A prospective cohort study within the
adults being overweight was not signifi- Honolulu Heart Program/Honolulu Asia Aging
cantly associated with an increased risk of mor- Study (5820 healthy Japanese-American men,
tality [85]. Also, the risk of mortality increased mean age 54 years; 40 years of follow-up)
in older people with a BMI <23.0, which sug- showed that leaner men had greater odds of sur-
gests the importance of monitoring weight loss vival at age 85 years than those who were over-
in older age groups as an indicator of some seri- weight or obese [87]. The Uppsala Longitudinal
ous health issue or sarcopenia. Compared to a Study (2293 Swedish men; mean age 50 years; 4
reference BMI of 23.0–23.9, there was a 12% decades of follow-up; 38% survived to age
higher mortality risk for individuals in the BMI 85 years) observed that the maintenance of nor-
range of 21.0–21.9 and a 19% greater risk for a mal weight at age 50 significantly improved odds
range of 20.0–20.9. In the obesity range mortal- of survival by 20% and independent aging by
ity risk began to increase for BMI ≥ 33 with only 66% [88]. Similarly, the Nurses’ Health Study
a slight attenuation after adjustment for interme- (40,000 US women; mean age 58.5 years; 4-year
diary factors. Exclusion of early deaths or preex- follow-up) found weight maintenance and/or
isting disease did not markedly alter the weight loss (5–20 lb) in overweight and obese
associations. In a large, population-based women was associated with significantly
California cohort study, the Leisure World improved physical function and vitality and
Cohort Study (13,451 participants; average aged reduced body pain [89]. The Helsinki
73 years; women 64%; average 13 years of Businessmen Study (1114 participants, mean age
follow-up; 11,203 deaths) it was observed that 47 years; 26 years of follow-up; 425 deaths)
compared to normal weight individuals those observed that men who were overweight or obese
who were either underweight or obese had at a mean of 47 years had a significantly higher
significantly higher all-cause mortality risk risk of developing frailty at 73 years of age com-
(Fig. 1.8) [86]. pared with men at constant normal weight with
1.6
Relative Risk of All-cause Mortality
1.4
1.2
0.8
0.6
0.4
0.2
0
Underweight Normal Overweight Obese
(BMI 17.8) (BMI 22.4) (BMI 26.5) (BMI 31.6)
BMI Category
Fig. 1.8 Association between BMI and all-cause mortality risk over 23 years of follow-up from the California Leisure
World Cohort Study of adults with a mean baseline age of 73 years (adapted from [86])
Odds of Developing Frailty

4
0
Normal weight Overweight Obese
BMI Category at Middle Age
Fig. 1.9 Association between BMI in mid-life (mean age 47 years) and odds of developing frailty at age 73 years from
the Helsinki Businessmen Study (p < 0.001; multivariate adjusted) (adapted from [90])
aging (Fig. 1.9) [90]. In this study, each 1% patterns (e.g., rich in vegetables, fruits, whole
increase in body weight was associated with grains, low- or non-fat dairy, seafood, legumes,
increased risk of developing both coronary artery and nuts and low in red and processed meats,
disease and frailty by 16%. A meta-analysis (15 sugar-sweetened foods and drinks and refined
prospective studies; 25,624 participants; 3.2– grains), and healthy aging [93, 94]. Several
36 years of follow-up) suggested a U-shaped meta- or pooled analyses report that higher
relationship between midlife BMI and late-life adherence to any healthy dietary pattern, includ-
risk of Alzheimer’s disease (AD) with normal ing the Healthy Eating Index (HEI)-2010,
BMI in midlife having the lowest risk of AD and Alternative Healthy Eating Index (AHEI)-2010,
obese BMI in midlife having the highest risk Alternate Mediterranean Diet (aMED), and
[91]. Compared with midlife normal BMI, (1) Dietary Approaches to Stop Hypertension
overweight BMI in midlife is associated with (DASH) converge to have a similar inversely
35% increased risk of developing AD and a 26% associated significant reduction in the risk of all-
increased risk of any dementia later in life and (2) cause mortality, CVD, and type 2 diabetes by
obese BMI in midlife is associated with twice the 22% and cancer mortality and incidence by 15%
risk of AD and a 64% increased risk for any because these diets all have relatively similar
dementia later in life. Overweight and obese core dietary components [93, 94]. The French
women were more susceptible to AD and demen- SU.VI.MAX study found that people with adher-
tia than were overweight and obese men. A meta- ence to a healthy diet in midlife combined with a
analysis (17 cohort studies; 563,277 participants) regulated energy intake below the threshold of
found a significantly reduced risk of adiposity for about 2500 kcal/day in men and 1820 kcals/day
fruit or vegetables intake by 17% (highest vs. in women, had significantly improved odds of
lowest intakes) and fruit intake was inversely healthy aging by 46% compared to those con-
associated with waist circumference [92]. suming above the threshold who had an insignifi-
cant 7% increase in the odds of healthy aging
(Fig. 1.10) [53]. This study supports a controlled,
1.4.2 Healthy Dietary Patterns moderate energy healthy diet as a potentially
important factor in healthy aging. A dietary
Prospective studies generally show a positive mechanism related to healthy aging with the
association between fiber-rich healthy dietary consumption of healthy diets such as the
1.4 Healthy Aging 17
Western pattern (p=0.17) Healthy pattern < median kcal intake (p=0.01)
Heathy pattern > median kcal intake (p=0.59)
1.5
Odds for Healthy Aging 1.4
1.3
1.2
1.1
0.9
0.8
Low Moderate High
Dietary Pattern Adherence
Fig. 1.10 Odds of healthy aging for healthy dietary pattern above and below the median energy level (median energy
intake of 2500 kcal/day in men and 1820 kcals/day in women) and the Western dietary pattern (adapted from [53])
Mediterranean diet (MedDiet) or other healthy allowed (equivalent to 1 teaspoon) per day (as
plant-based diets is that they appear to maintain iodized salt) [93]. Several analyses have evalu-
or promote longer telomere length [95]. In con- ated the effect of the WHO diet on longevity and
trast, the Western diet with high saturated fat, cardiovascular disease mortality in adults
refined grains, meat and meat products, and >60 years. The first pooled analysis of cohort
sugar-sweetened beverages intake appears to be studies (11studies; 400,000 participants; 42%
associated with shorter telomeres. women) estimated that greater adherence to the
The World Health Organization (WHO) has WHO guidelines increased life expectancy by at
formulated global guidelines for a healthy diet least 2 years in European and United States men
plan to prevent chronic diseases and premature and women >60 years old [97]. The second
death [96–98]. This plant based diet consists WHO pooled analysis (10 cohort studies;
primarily of fruits, vegetables, legumes 280,000 men and women; aged >60 years)
(e.g. lentils, beans), nuts and whole grains found that higher adherence to the WHO dietary
(e.g. unprocessed maize, millet, oats, wheat, guidelines was inversely associated with CVD
brown rice) and specific dietary guidance mortality observed in older populations in
including: (1) at least 400 g (5 portions) of fruits southern Europe and the United States but sig-
and vegetables a day (excluding potatoes, sweet nificant reductions were not observed across
potatoes, cassava and other starchy roots); (2) other European regions [98].
less than 10% of total energy intake from free
sugars, which is equivalent to 50 g (or around 12
level teaspoons) per daily 2000 calories; (3) 1.4.3 Physically Active Lifestyle
healthy fats (e.g. found in fish, avocados, nuts,
sunflower, canola and olive oils) being prefera- The role of physical activity in promoting healthy
ble to saturated fats (e.g. found in fatty meat, life expectancy (e.g., chronic disease risk reduc-
butter, palm and coconut oil, cream, cheese, tion) and reducing premature mortality is well
ghee and lard); (4) avoidance of industrial trans established [99–103]. Over 5 decades of system-
fats found in processed food, fast food, snack atic epidemiological and intervention research
food, fried food, frozen pizza, pies, cookies, supports the US and WHO minimum physical
margarines and spreads; and (5) <5 g of salt is activity guidelines of 150 min of moderate
a ctivity, 75 minutes of vigorous activity, or some 1.5 ultiple Healthy Lifestyles

M
combination of moderate and vigorous activity Factors and Mortality Risk
per week [99, 100]. A 2015 pooled analysis
(6 cohort studies; 661,137 participants; median Prospective studies generally find that following
aged 62 years; median 14.2 years of follow-up; multiple healthy lifestyle factors has accumula-
116,686 deaths) found that individuals meeting at tive benefits for longer life and healthy life
least the recommended minimum physical activ- expectancy (Table 1.3) [34, 105–111]. A system-
ity had a 31% lower mortality risk compared with atic review and meta-analysis showed that a
those reporting no leisure time physical activity, combination of at least 4 of the following healthy
whereas increased physical activity above the lifestyle habits: low-moderate consumption of
recommended level yielded relatively small addi- alcohol; not being overweight or obese, not
tional benefits [101]. A meta-analysis (33 cohort smoking; healthy diet; and regular physical exer-
studies; 102,980 participants; mean age cise was associated with a reduction of all-cause
37–57 years; 1.1–26 years of follow-up; 6910 mortality risk by ≥66% [105]. The pooled analy-
deaths) showed that an increased aerobic activity sis of the Nurses’ Health Study and Health
of 15 minutes/day or 90 minutes/week extended Professionals Follow-up Study with over
life expectancy by an average of 3 years [102]. A 100,000 men and women evaluated over 26 to
dose response meta-analysis (80 prospective 32 years showed that an increased number of
studies; 1,338,143 participants; 85% US/EU,15% healthy lifestyle factors significantly reduced
Asian; median age 56 years; median 10.7 years all-cause mortality rate in normal, overweight
of follow-up; 118,121 deaths) found that higher and obese populations (Fig. 1.1) [34]. A combi-
levels of total and specific types of physical activ- nation of at least 3 low risk lifestyle factors and
ity were associated with reduced all-cause mor- BMI between 18.5 and 22.4 was associated with
tality with the largest reductions observed with the lowest risk of all-cause mortality by 71%.
vigorous activity, supporting the message that The Physicians’ Health Study (1163 men with
‘some is good; more is better’ (Fig. 1.11) [103]. type 2 diabetes; average baseline age 69 years;
The beneficial effects of physical activity on all- mean 9 years of follow-up) found an inverse
cause mortality risk tends to be independent of association between the number of healthy life-
other lifestyle risk factors such as BMI [104]. style factors and total mortality risk with a
Vigorous exercise (sports) Moderate/vigorous (leisure time)

Moderate activities (daily living)
1
0.9
All-cause Mortality Risk
0.8
0.7
0.6
0.5
60 150 300
Minutes Per Week
Fig. 1.11 Association between type and intensity of physical activity and all-cause mortality risk (multivariate
adjusted) (adapted from [103])
1.5 Multiple Healthy Lifestyles Factors and Mortality Risk 19
Table 1.3 Summaries of studies on adherence to multiple healthy lifestyle behavior and all-cause and disease
mortality
Objective Study details Results
Meta-analysis
Loef et al. (2012) 15 cohort studies; 531,804 people; A combination of at least 4 healthy
Investigate combined effects of mean follow-up of 13 years lifestyle factors including: low-
lifestyle behaviors on all-cause moderate consumption of alcohol;
mortality (German) [105] not being overweight or obese; not
smoking; healthy diet; and regular
physical exercise is associated with a
reduction of the all-cause mortality
risk by 66%
Prospective cohort studies
Veronese et al. (2016) 74,582 women and 39,284 men; up In each of the four categories of BMI
Evaluate the combined associations of to 32 years of follow-up; 30,013 studied (18.5–22.4, 22.5–24.9,
diet, physical activity, moderate deaths (including10,808 from 25–29.9, ≥30), people with one or
alcohol consumption, and smoking cancer and 7189 from more healthy lifestyle factors had a
with body weight on risk of all cause Cardiovascular disease); lifestyle significantly lower risk of total
and cause specific mortality (Nurses’ factors included body mass index mortality (Fig. 1.1), and
Health Study and Health Professionals (BMI), score on the alternate cardiovascular and cancer mortality
Follow-up Study; US) [34] healthy eating index, level of than individuals with no low risk
physical activity, smoking habits, lifestyle factors. A combination of at
and alcohol drinking least 3 low risk lifestyle factors and
BMI between 18.5–22.4 was
associated with the lowest risk of
all-cause mortality by 71%
Patel et al. (2016) 1163 men with type 2 diabetes; An inverse relationship was found
Examine the association of healthy average baseline age 69 years; between the number of healthy
lifestyle factors on mortality in mean 9 years of follow-up; lifestyle lifestyle factors and total mortality;
people with type 2 diabetes factors consisted of currently not compared with participants who had
(Physicians Health Study; US) [106] smoking, moderate drinking ≤1 healthy lifestyle factor, the risk of
(1–2 drinks/day), vigorous exercise death was 42% lower for those with
(1+/week), BMI < 25 kg/m2, and 2 healthy lifestyle factors, 41% lower
being in the top 2 quintiles of the for those with 3, and 44% lower for
alternate healthy eating index-2010 those with 4 or more healthy lifestyle
(AHEI-2010) factors
Prinelli et al. (2015) 974 Italian completers; 51% A high adherence to the MedDiet,
Evaluate the association of the women; mean age 56 years; mean non-smoking and physical activity
Mediterranean diet (MedDiet), BMI 27; median 17.4 years of were strongly associated with a
smoking habits and physical activity follow-up; 193 deaths reduced risk of all-cause mortality in
with all-cause mortality (Italian) healthy subjects after long-term
[107] follow-up (Fig. 1.12). All-mortality
risk was reduced with high
adherence to the MedDiet Score by
38%, non-smokers by 29% and
physically active subjects by 45%.
Each point increase in the MedDiet
score was associated with a
significant 5% reduction of death
risk. The average risk of death was
significantly reduced for 1, 2 or 3
healthy lifestyle behaviors by 39%,
56%, and 73%, respectively
(continued)
Table 1.3 (continued)
May et al. (2015) 33,066 participants; mean age Non-smoking, maintaining a lean
Assess the effect of several healthy 50 years; 12 years of follow-up; BMI (<25 kg/m2), being physically
behaviors on disability-adjusted life 6647 disease incidences and 1482 active, and consuming a MedDiet
years (DALYs) (EPIC-Netherland deaths were all associated with a
cohort) [108] significantly lower disease burden.
Individuals adhering to all four
healthy lifestyle factors lived a
minimum of 2 years longer in good
health than those not following any
healthy lifestyles. Due to challenges
in follow-up, the total number of
DALYs, and consequently the
estimates, are underestimated.
Therefore, true lifetime health
benefits of a healthy lifestyle were
under reported
Petersen et al. (2015) 51,521 subjects (27,256 women and Among men, adherence to one
Investigate the combined impact of 24,265 men); mean baseline age additional health recommendation
adherence to five lifestyle factors 55 years; 14 years follow-up; 6768 was associated with an adjusted
(smoking, alcohol intake, physical deaths (3941 men and 2827 mortality risk reduction for all-cause
activity, waist circumference and women); of these, 43% of men and by 27%, for cancer by 26% and for
diet) on all-cause, cancer and 51% of women died due to cancer, CV by 30%. Among women, the
cardiovascular (CV) mortality based and 20% of men and 12% of adjusted risk was reduction for
on health recommendations women died due to CVD all-cause by 28%, cancer by 24%
(Denmark) [109] and CV by 37%. Figs. 1.13 and 1.14
summarizes the effects of the
number of healthy lifestyle choices
on all-cause, cancer and CV
mortality risk in men and women
Behrens et al. (2013) 170,000 participants; age Adherence to all 4 low risk lifestyle
Determine the effect of healthy 51–71 years; mean 13 years of factors, including achieving healthy
lifestyle behaviors on mortality risk follow-up; 20,903 deaths waist size, recommended physical
in older US adults (National activity guidelines, non-smoking,
Institutes of Health-AARP Diet and and adherence to a healthy diet
Health Study; US) [110] significantly reduced mortality risk
by 73% vs. following none of these
risk factors
Van den Brandt et al. (2011) 120,852 men and women; baseline Adherence to the MedDiet was
Investigate the association between age 55–69 years; 10 years of significantly related to lower
adherence to the Mediterranean diet follow-up; 18,091 deaths mortality in women but not in men.
(MedDiet) and total mortality and The healthy lifestyle score was
estimate the overall impact of a strongly inversely related to
combined healthy lifestyle on mortality in women and men. The
premature death (The Netherlands) least-healthy compared to the
[111] healthiest lifestyle scores increased
premature death in women by
4.1-fold and in men by 2.6-fold
(p-trend = 0.001)
reduction of up to 41–44% depending on the MedDiet and physical activity were significantly
number of healthy lifestyle factors [106]. In an associated with reduced all-cause mortality in
Italian cohort (974 completers; 51% women; middle aged healthy adults over 17 years of fol-
mean age 56 years; mean BMI 27; median low-up (Fig. 1.12) [107]. The Rotterdam Study
17.4 years of follow-up) high adherence to the (5974 participants from the general population;
1.5 Multiple Healthy Lifestyles Factors and Mortality Risk 21
MedDiet score Physical activity

1.2
Harzard Ratios of All-cause Mortality

1
0.8
0.6
0.4
0.2
0
Low Medium High Inactive Active
Level of Adherence
Fig. 1.12 Association between the Mediterranean diet (MedDiet) Score and physical activity on all-cause mortality
average age 69 years; 595 women; mean related epigenetics affecting genomic mainte-
15.1 years of follow-up; 3174 deaths) observed nance and functional factors, which may influ-
that smoking status (pack-years), energy intake, ence aging mortality risk [112]. A meta-analysis
blood pressure, BMI, waist size, C-reactive pro- (4 cohorts; 4658 subjects; 662 deaths; mean
tein level, total cholesterol, bone mineral density baseline ages range from 66–79 years) found
of the femoral neck, aortic calcification, diabe- that a 5-year higher DNA methylation age than
tes, cardiac diseases, cancer and cognitive func- chronological age was associated with a 16%
tion were independent indicators of mortality higher mortality risk [113].
risk [108]. A Danish cohort analysis (51,521
subjects; mean baseline age 55 years; 14-year Conclusions
follow-up) found that the adherence to multiple Aging is the biological progressive deteriora-
healthy lifestyle factors significantly lowered tion of physiological functions and metabolic
all-cause, cancer and cardiovascular mortality processes that can lead to chronic diseases
by up to about 80% compared to no healthy life- such as neurodegenerative disorders, CHD,
style factors in men (Fig. 1.13) and women type 2 diabetes and cancer, and ultimately to
(Fig. 1.14) [109]. The US National Institutes of death. Since it is estimated that 70 to 80% of
Health-AARP study (170,000 subjects; 13 years the rate of aging is related to external factors
of follow-up) observed a 73% lower morality such as lifestyle choices, it is possible for
risk in subjects adhering to all four low risk life- individuals to significantly influence their
style factors of healthy waist circumference, rec- odds of healthy aging and longevity, even if
ommended physical activity, adherence to healthy lifestyles are adopted later in life.
healthy diets and non-smoking compared to sub- Unhealthy or premature aging, which is
jects adhering to none of these healthy lifestyles largely associated with excessive intake of
[110]. A Dutch cohort analysis found that the Western energy dense diets and sedentary life-
least-healthy vs. the healthiest lifestyle scores styles, involves a complex interplay between
increased premature death in women by 4-fold obesity and related metabolic dysfunctional
and in men by 2.6-fold [111]. Lifestyle factors effects leading to increased risk of chronic
can influence aging related DNA methylation disease and mortality, and reduced healthy life
All-cause Cancer Cardiovascular

1.2
Men
Hazard Ratio for Mortality 1
0.8
0.6
0.4
0.2
0
0 1 2 3 4+5
Number of Lifestyle Choices*
Fig. 1.13 Association between number of healthy lifestyle choices adopted and all-cause mortality in men (mean age
55 years; followed for 14 years; p-trend <0.001; multivariate adjusted) (adapted from [109]). *Healthy lifestyle: smok-
ing cessation, limited alcohol intake, physically active, healthy waist circumference, and healthy diet
All-cause Cancer Cardiovascular

1.2
Women
1
Hazard Ratio for Mortality
0.8
0.6
0.4
0.2
0
0 1 2 3 4+5
Number of Lifestyle Choices*
Fig. 1.14 Association between number of healthy lifestyle choices adopted and all-cause mortality in women (mean
age 56 years; followed for 14 years; p-trend <0.001; multivariate adjusted) (adapted from [109]). *Healthy lifestyle:
smoking cessation, limited alcohol intake, physically active, healthy waist circumference, and healthy diet
References 23
expectancy. In general, obesity, especially United States in the 21st century. N Engl J Med.
central adiposity represents a state of acceler- 2005;352(11):1138–45.
8. Global Burden of Disease Study (GBD). Mortality
ated aging as adipose cells produce adipo- and Causes of Death Collaborators. Global,
kines, which can lead to increased systemic regional, and national age-sex specific all-cause
and tissue inflammation and peripheral insu- and cause- specific mortality for 240 causes of
lin resistance. Metabolic syndrome, type 2 death, 1990-2013: a systematic analysis for the
Global Burden of Disease Study. 2013. Lancet.
diabetes and prediabetes, and sarcopenia are 2015;385(9963):117–71.
major unhealthy aging phenotypes, which can 9. Beltran-Sanchez H, Soneji S, Crimmins EM. Past,
be prevented by adapting healthy lifestyle present, and future of healthy life expectancy. Cold
choices. The concept of healthy aging includes Spring Harb Perspect Med 2015;5(11). doi: https://
doi.org/10.1101/cshperspect.a025957.
healthy life expectancy (e.g., absence or delay 10. Murray CJ, Vos T, Lozano R, Naghavi M, et al.
of chronic diseases and the maintenance of Disability-adjusted life years (DALYs) for 291 dis-
cognitive, physical, and other functions with eases and injuries in 21 regions, 1990–2010: a sys-
limited dependence on family members or tematic analysis for the Global Burden of Disease
Study 2010. Lancet. 2013;380:2197–223.
extended care assisted living) and longevity. 11. World Health Organization. WHO global status
Since a higher percentage of people world- report on noncommunicable diseases. Geneva:
wide are surviving to older ages, it is critical World Health Organization Press; 2010. p. 2010.
to promote optimal healthy aging lifestyle 12. Murray CJ, Lopez AD. Measuring the global burden
of disease. N Engl J Med. 2013;369(5):448–57.
habits to assure quality of life for aging indi-
13. Lafortune L, Martin S, Kelly S, et al. Behavioural
viduals and their families and for sustainable risk factors in mid-life associated with success-
healthcare cost management. The odds of ful aging, disability, dementia and frailty in later
healthy aging can be significantly increased life: a rapid systematic review. PLOS ONE.
2016:11(2):e0144405.
by following a healthy lifestyle even if it is
14. Avendano M, Kawachi I. Why do Americans have
adopted in middle age adulthood or older. shorter life expectancy and worse health than people
These lifestyle choices include; adhering to a in other high-income countries? Annu Rev Public
healthy dietary pattern, increasing physical Health. 2014;35:307–25.
activity most days of the week, achieving and 15. Nevalainen T, Kananen L, Marttila, et al. Obesity
accelerates epigenetic aging in middle-aged but
maintaining a healthy body weight and waist not in elderly individuals. Clin Epigenetics. 2017;
size, and smoking avoidance. 9:20.
16. Lunenfeld B, Stratton P. The clinical consequences
of an ageing world and preventive strategies. Best
Pract Res Clin Obstet Gynaecol. 2013;27(5):
643–59.
References 17. Tyrovolas S, Haro JM, Mariolis A, et al. The role
of energy balance in successful aging among elderly
1. Mathers JC. Impact of nutrition on the ageing pro- individuals: The Multinational MEDIS Study. J
cess. Br J Nutr. 2015;113:S18–22. Aging Health. 2015;27(8):1375–91.
2. Olshansky SJ, Hayflick L, Carnes BA. No truth to 18. Fontana L, Hu FB. Optimal body weight for health
the fountain of youth. Sci Am. 2008;14:98–102. and longevity: bridging basic, clinical, and popula-
3. Kirkwood TBL. Why and how are we living longer? tion research. Aging Cell. 2014;13:391–400.
Exp Physiol. 2017;102(9):1067–74. 19. Newman AB. Is the onset of obesity the same as
4. Zhavoronkov A, Bhupinder B. Classifying aging aging? PNAS. 2015. doi-https://doi.org/10.1073/
as a disease in the context of ICD-11. Front Genet. pnas.1515367112.
2015;6(3262):1–8. 20. Beers MH. The Merck manual of health & aging.
5. Lopez-Otin C, Blasco MA, Partridge L, et al. The Whitehouse Station: Merck Laboratories; 2004.
hallmarks of aging. Cell. 2013;153:1194–217. p. 2.
6. Olshansky SJ. Has the rate of human aging already 21. Ogden CL, Carroll MD, Fryar CD, Flegal
been modified? Cold Spring Harb Perspect Med. KM. Prevalence of obesity among adults and youth:
2015;5(12). https://doi.org/10.1101/cshperspect. United States, 2011–2014. NCHS Data Brief.
a025965. 2015;219:1–7.
7. Olshansky SJ, Passaro DJ, Hershow RC, et al. 22. Verdile G, Keane KN, Cruzat VF, et al. Inflammation
A potential decline in life expectancy in the and oxidative stress: the molecular connectivity
between insulin resistance, obesity, and Alzheimer’s 35. Winter JE, MacInnis RJ, Wattanapenpaiboon N,
disease. Mediators Inflamm. 2015;2015:105828. Nowson CABMI. All-cause mortality in older adults:
https://doi.org/10.1155/2015/105828. a meta-analysis. Am J Clin Nutr. 2014;99:875–90.
23. Kullmann S, Callaghan MF, Heni M, et al. Specific 36. Bowman K, Atkins JL, Delgado J, et al. Central adi-
white matter tissue microstructure changes asso- posity and the overweight risk in aging: follow-up of
ciated with obesity. Neuroimage. 2016;125: 130,473 UK Biobank participants. Am J Clin Nutr.
36–44. 2017;106:130–5.
24. Eckel N, Meidtner K, Kalle-Uhlmann T, et al. 37. Graf CE, Herrmann FR, Spoerri A, et al. Impact
Metabolically healthy obesity and cardiovascu- of body composition changes on risk of all-
lar events: a systematic review and meta-analysis. cause mortality in older adults. Clin Nutr 2016;
Eur J Prev Cardiol. 2016;23(9):956–66. https://doi. 35(6):1499–1505. doihttps://doi.org/10.1016/j.clnu.
org/10.1177/2047487315623884. 2016.04.003.
25. Roberson LL, Aneni EC, Maziak W, et al. Beyond 38. Bea JW, Thomson CA, Wertheim BC, et al. Risk of
BMI: the “metabolically healthy obese” phenotype mortality according to body mass index and body
& its association with clinical/subclinical cardio- composition among postmenopausal women. Am J
vascular disease and all-cause mortality—a system- Epidemiol. 2015;182(7):585–96.
atic review. BMC Public Health. 2014. https://doi. 39. Zhang C, Rexrode KM, van Dam RM, et al.
org/10.1186/1471-2458-14-14. Abdominal obesity and the risk of all-cause,
26. Kramer CK, Zinman B, Retnakaran R. Are metaboli- cardiovascular, and cancer mortality: a sixteen
cally healthy overweight and obesity benign condi- years of follow-up in US women. Circulation.
tions? A systematic review and meta-analysis. Ann 2008;117:1658–67.
Intern Med. 2013;159:758–69. 40. Pang Q, Zhang J-Y, Song S-D, et al. Central obesity
27. Pujia A, Gazzaruso C, Ferro Y, et al. Individuals and nonalcoholic fatty liver disease risk after adjust-
with metabolically healthy overweight/obesity have ing for body mass index. World J Gastroenterol.
higher fat utilization than metabolically unhealthy 2015;21(5):1650–62.
individuals. Forum Nutr. 2016;8:2. https://doi. 41. Wohlfahrt P, Somers VK, Cifkova R, et al.
org/10.3390/nu8010002. Relationship between measures of central and general
28. Whitlock G, Lewington S, Sherliker P, et al. Body- adiposity with aortic stiffness in the general popula-
mass index and cause-specific mortality in 900,000 tion. Atherosclerosis. 2014;235(2):625–31. https://
adults: collaborative analyses of 57 prospective stud- doi.org/10.1016/j.atherosclerosis.2014.05.958.
ies. Lancet. 2009;373:1083–96. 42. Aronis KN, Wang N, Phillips CL, et al. Associations
29. Lorenzini A. How much should we weigh for a long of obesity and body fat distribution with incident
and healthy life span? The need to reconcile caloric atrial fibrillation in the biracial health aging and
restriction versus longevity with body mass index ver- body composition cohort of older adults. Am Heart
sus mortality data. Front Endocrinol. 2014;5(121):1– J. 2015;170(3):498–505.
8. https://doi.org/10.3389/fendo.2014.00121. 43. Sun Q, Townsend MK, Okereke OI, et al. Adiposity
30. Kodama S, Horikawa C, Fujihara K, et al. and weight change in mid-life in relation to healthy
Quantitative relationship between body weight gain survival after age 70 in women: prospective cohort
in adulthood and incident type 2 diabetes: a meta- study. BMJ. 2009;339. https://doi.org/10.1136/bmj.
analysis. Obes Rev. 2014;15(3):202–14. b3796.
31. Cao Y, Ma J. Body mass index, prostate cancer–spe- 44. Batsis JA, Zbehlik AJ, Barre LK, et al. Impact of
cific mortality, and biochemical recurrence: a sys- obesity on disability, function, and physical activ-
tematic review and meta-analysis. Cancer Prev Res. ity: data from the Osteoarthritis Initiative. Scand J
2011;4(4):486–501. Rheumatol. 2015;44(6):495–502.
32. Flegal KM, Kit BK, Orpana H, Graubard 45. Roos E, Laaksonen M, Rahkonen O, et al. Relative
BI. Association of all-cause mortality with overweight and disability retirement: a prospec-
weight and obesity using standard body mass index tive cohort study. Scand J Work Environ Health.
categories: a systematic review and meta-analysis. 2013;39(3):259–67.
JAMA. 2013;309:71–82. 46. Rillamas-Sun E, LaCroix AZ, Waring ME, et al.
33. Aune D, Sen A, Prasad M, et al. BMI and all-cause Obesity and survival to age 85 years without major
mortality: systematic review and non-linear dose- disease or disability in older women. JAMA Intern
response meta-analysis of 230 cohort studies with 3.74 Med. 2014;174(1):98–106.
million deaths among 30.3 million participants. BMJ. 47. Batsis JA, Zbehlik AJ, Barre LK, et al. The impact of
2016;353:i2156. https://doi.org/10.1136/bmj.i2156. waist circumference on function and physical activity
34. Veronese N, Li Y, Manson JE, et al. Combined asso- in older adults: longitudinal observational data from
ciations of body weight and lifestyle factors with the osteoarthritis initiative. Nutr J. 2014;13:81.
all cause and cause specific mortality in men and https://doi.org/10.1186/1475-2891-13-81.
women: prospective cohort study. BMJ. 2016;355: 48. Beavers KM, Beavers DP, Nesbit BA, et al. Effect
i5855. doi.org/10.1136/bmj.i5855. of an 18-month physical activity and weight loss
References 25
intervention on body composition in overweight the prevention and treatment of abdominal. Obesity.
and obese older adults. Obesity (Silver Spring). 2014:141–50.
2014;22(2):325–31. 63. Weiss EP, Fontana L. Caloric restriction: powerful
49. Porter Starr KN, Pieper CF, Orenduff MC, et al. protection for the aging heart and vasculature. Am J
Improved function with enhanced protein intake Physiol Heart Circ Physiol. 2011;301:H1205–19.
per meal: a pilot study of weight reduction in frail, 64. Ribeiro SML, Kehayias JJ. Sarcopenia and the anal-
obese older adults. J Gerontol A Biol Sci Med Sci. ysis of body composition. Adv Nutr. 2014;5:260–7.
2016;71(10):1369–75. https://doi.org/10.1093/ https://doi.org/10.3945/an.113.005256.
gerona/glv210. 65. Grundy SM. Adipose tissue and metabolic syn-
50. Fung TT, Schulze M, Manson JE, et al. Dietary pat- drome: too much, too little or neither. Eur J Clin
terns, meat intake, and the risk of type 2 diabetes in Invest. 2015;45(11):1209–17.
women. Arch Intern Med. 2004;164:2235–40. 66. Ley SH, Hamdy O, Mahan V, Prevention
51. Heidemann C, Schulze MB, Franco OH, et al. HFB. Management of type 2 diabetes: dietary
Dietary patterns and risk of mortality from cardio- components and nutritional strategies. Lancet.
vascular disease, cancer, and all causes in a pro- 2014;383:1999–2007.
spective cohort of women. Circulation. 2008;118: 67. Tabák AG, Herder C, Rathmann W, et al. Prediabetes:
230–7. a high-risk state for developing diabetes. Lancet.
52. Akbaraly T, Sabia S, Hagger-Johnson G, et al. 2012;379(9833):2279–90.
Does overall diet in midlife predict future 68. Kaur J. A comprehensive review on metabolic syn-
aging phenotypes? A cohort study. Am J Med. drome. Cardio Res Pract. 2014;2014:943162. https://
2013;126:411–9. doi.org/10.1155/2014/943162.
53. Assmann KE, Lassale C, Andreeva VA, et al. A 69. Mottillo S, Filion KBJ, Genest J, et al. The meta-
healthy dietary pattern at midlife, combined with a bolic syndrome and cardiovascular risk: a system-
regulated energy intake, is related to increased odds atic review and meta-analysis. J Am Coll Cardiol.
for healthy aging. J Nutr. 2015;145:2139–45. 2010;56:1113–32.
54. Ashby-Mitchell K, Peeters A, Anstey KJ. Role of 70. Ma Y, Olendzki BC, Wang J, et al. Single-component
dietary pattern analysis in determining cognitive versus multicomponent dietary goals for the meta-
status in elderly Australian adults. Forum Nutr. bolic syndrome. A randomized trial. Ann Intern
2015;7:1052–67. Med. 2015;162:248–57.
55. Bouchard C, Blair SN, Katzmarzyk PT. Less sitting, 71. Esposito K, Marfella R, Ciotola M, et al. Effect of
more physical activity, or higher fitness? Mayo Clin a Mediterranean-style diet on endothelial dysfunc-
Proc. 2015;90(11):1533–40. tion and markers of vascular inflammation in the
56. Chau JY, Grunseit AC, Chey T, et al. Daily sit- metabolic syndrome. A randomized trial. JAMA.
ting time and all-cause mortality: a meta-analysis. 2004;292:1440–6.
PLoS One 2013; 8(11):e80000. doi: https://doi. 72. Murray MT. Diabetes mellitus (Chapter 161). In:
org/10.1371/journal.pone.0080000. Pizzorno JE, Murray MT, editors. Textbook of natu-
57. Wen CP, Wai JPM, Tsai MK, et al. Minimum amount ral medicine. 4th ed. Philadelphia: Elsevier; 2013.
of physical activity for reduced mortality and p. 1320–48.
extended life expectancy: a prospective cohort study. 73. FB H, Manson E, Stampfer MJ. Diet, lifestyle and
Lancet. 2011;378:1244–53. the risk of type 2 diabetes mellitus in women. N Engl
58. Stenholm S, Koster A, Valkeinen H, et al. Association J Med. 2001;345(11):790–7.
of physical activity history with physical function 74. Alhazmi A, Stojanovski E, McEvoy M, Garg
and mortality in old age. J Gerontol A Biol Sci Med ML. The association between dietary patterns
Sci. 2016;71(4):496–501. https://doi.org/10.1093/ and type 2 diabetes: a systematic review and
gerona/glv111. meta-analysis of cohort studies. J Hum Nutr Diet.
59. Batsis JA, Germain CM, Vasquez E, et al. Physical 2014;27:251–60.
activity predicts higher physical function in older 75. Bell JA, Kivimaki M, Hamer M. Metabolically
adults: The Osteoarthritis Initiative. J Phys Act healthy obesity and risk of incident type 2 diabetes:
Health. 2016;13(1):6–16. https://doi.org/10.1123/ a meta-analysis of prospective cohort studies. Obes
jpah.2014-0531. Rev. 2014;15:504–15.
60. Barry VW, Baruth M, Beets MW, et al. Fitness vs. 76. Barry E, Roberts S, Oke J, et al. Efficacy and effec-
fatness on all-cause mortality: a meta-analysis. Prog tiveness of screen and treat policies in prevention
Cardiovasc Dis. 2014;56:382–90. of type 2 diabetes: systematic review and meta-
61. Bonomini F, Rodella LF, Rezzani R. Metabolic syn- analysis of screening tests and interventions. BMJ.
drome, aging and involvement of oxidative stress. 2017; 356:i6538. doi: org/https://doi.org/10.1136/
Aging Dis. 2015;6(2):109–20. bmj.i6538.
62. Rossi AP, Fantin F, Bertassello P, et al. Chapter 77. Jackman RW, Kandarian SC. The molecular basis of
13. Visceral fat predicts ectopic fat accumulation skeletal muscle atrophy. Am J Physiol Cell Physiol.
mechanisms and health consequences. Nutrition in 2004;287:C834–43.
78. Paddon-Jones D, Rasmussen BB. Dietary protein tality in the multiethnic cohort: the dietary patterns
recommendations and the prevention of sarcopenia: methods project. Am J Clin Nutr. 2015;101:587–97.
protein, amino acid metabolism and therapy. Opin 94. Schwingshackl L, Hoffmann G. Diet quality as
Clin Nutr Metab Care. 2009;12(1):86–90. assessed by the Healthy Eating Index, the Alternate
79. Moore DR. Keeping older muscle “young” through Healthy Eating Index, the Dietary Approaches to
dietary protein and physical activity. Adv Nutr. Stop Hypertension score, and health outcomes:
2014;5(5):599S–607S. a systematic review and meta-analysis of cohort
80. von Haehling S, Morley JE, Anker SD. From muscle studies. J Acad Nutr Diet. 2015;15(5):780–800.e5.
wasting to sarcopenia and myopenia: update 2012. https://doi.org/10.1016/j.jand.2014.12.009.
J Cachexia Sarcopenia Muscle. 2012;3:213–7. 95. Freitas-Simoes TM, Ros E, Sala-Vila A. Nutrients,
81. Norton C, Toomey C, McCormack WG, et al. foods, dietary patterns and telomere length: update
Protein supplementation at breakfast and lunch for of epidemiological studies and randomized trials.
24 weeks beyond habitual intake increases whole- Metabolism. 2016;65(4):406–15.
body lean tissue mass in healthy older adults. 96. WHO. Healthy diet. 2015; http://www.who.int/medi-
J Nutr. 2016;146(1):65–9. https://doi.org/10.3945/ acentre/factsheets/fs394/en. Accessed 10.27.2015.
jn.115.219022. 97. Jankovic N, Geelen A, Streppel MT, et al. Adherence
82. North BJ, Sinclair DA. The intersection between to a healthy diet according to the World Health
aging and cardiovascular disease. Circ Res. Organization guidelines and all-cause mortality in
2012;110:1097–108. elderly adults from Europe and the United States.
83. Suominen MH, Jyvakorpi SK, Pitkala KH, et al. Am J Epidemiol. 2014;180(10):978–88.
Nutritional guidelines for older people in Finland. 98. Jankovic N, Geelen A, Streppel MT, et al. WHO
J Nutr Health Aging. 2014;18(10):861–7. guidelines for a healthy diet and mortality from
84. de Gonzalez AB, Hartge P, Cerhan JR, et al. Body- cardiovascular disease in European and American
mass index and mortality among 1.46 million white elderly: the CHANCES project. Am J Clin Nutr.
adults. N Engl J Med. 2010;363(23):2211–9. 2015;102:745–56.
85. Winter JE, MacInnis RJ, Wattanapenpaiboon N, 99. US Department of Health and Human Services.
Nowson CABMI. All-cause mortality in older adults: 2008 Physical activity guidelines for Americans.
a meta-analysis. Am J Clin Nutr. 2014;99:875–90. Washington, DC: US Dept of Health and Human
86. Corrada MM, Kawas CH, Mozaffar F, Paganini- Services; 2008.
Hill A. Association of body mass index and weight 100. World Health Organization. Global recommenda-
change with all-cause mortality in the elderly. Am J tions on physical activity for health. World Health
Epidemiol. 2006;163(10):938–49. Organization website. 2010; http://www.who.int/
87. Willcox BJ, He Q, Chen R, et al. Midlife risk dietphysicalactivity/publications/9789241599979/
factors and healthy survival in men. JAMA. en/. accessed November 23, 2015.
2006;296:2343–50. 101. Arem H, Moore SC, Patel A, et al. Leisure time
88. Franzon K, Zethelius B, Cederholm T, Kilander physical activity and mortality: a detailed pooled
L. Modifiable midlife risk factors, independent analysis of the dose-response relationship. JAMA
aging, and survival in older men: report on long-term Intern Med. 2015;175(6):959–67.
follow-up of the Uppsala longitudinal study of adult 102. Kodama S, Saito K, Tanaka S, et al.
men cohort. J Am Geriatr Soc. 2015;63(5):877–85. Cardiorespiratory fitness as a quantitative predictor
89. Fine JT, Colditz GA, Coakley EH, et al. A prospec- of all-cause mortality and cardiovascular events in
tive study of weight change and health -related qual- healthy men and women: a meta-analysis. JAMA.
ity of life in women. JAMA. 1999;282:2136–42. 2009;301(19):2024–35.
90. Strandberg TE, Sirola J, Pitkata KH, et al. Association 103. Samitz G, Egger M, Zwahlen M. Domains of physi-
of midlife obesity and cardiovascular risk with old cal activity and all-cause mortality: systematic
age frailty: a 26-year follow-up of initially healthy review and dose-response meta-analysis of cohort
men. Int Obes (Lond). 2012;36(9):1153–7. studies. Int J Epidemiol. 2011;40:1382–400.
91. Anstey KJ, Cherbuin N, Budge M, et al. Body mass 104. Katzmarzyk PT, Janssen I, Ardern CI. Physical
index in midlife and late-life as a risk factor for inactivity, excess adiposity and premature mortality.
dementia: a meta-analysis of prospective studies. Obes Rev. 2003;4(4):257–90.
Obes Rev. 2012;12:e426–37. 105. Loef M, Walach H. The combined effects of
92. Schwingshackl L, Hoffmann G, Kalle-Uhlmann T, healthy lifestyle behaviors on all-cause mortality:
et al. Fruit and vegetable consumption and changes a systematic review and meta-analysis. Prev Med.
in anthropometric variables in adult populations: a 2012;55:163–70.
systematic review and meta-analysis of prospective 106. Patel YR, Gadiraju TV, Gaziano JM, Djoussé L.
cohort studies. PLoS One. 2015;10(10). https://doi. Adherence to healthy lifestyle factors and risk of
org/10.1371/journal.pone.0140846. death in men with diabetes mellitus: The Physicians’
93. Harmon BE, Boushey CJ, Shvetsov YB, et al. Health Study. Clin Nutr. 2016; doi: 10.1016/j.
Associations of key diet-quality indexes with mor- clnu.2016.11.003.
References 27
107. Prinelli F, Yannakoulia M, Anastasiou CA, et al. 110. Behrens G, Fischer B, Kohler S, et al. Healthy life-
Mediterranean diet and other lifestyle f actors style behaviors and decreased risk of mortality in
in relation to 20-year all-cause mortality: a a large prospective study of U.S. women and men.
cohort study in an Italian population. Br J Nutr. Eur J Epidemiol. 2013;28(5):361372. https://doi.
2015;113(6):1003–11. org/10.1007/s10654-013-9796-9.
108. May AM, Struijk EA, Fransen HP, et al. The
111. van den Brandt PA. The impact of a Mediterranean
impact of a healthy lifestyle on disability- diet and healthy lifestyle on premature mortality in
adjusted life years: a prospective cohort study. men and women. Am J Clin Nutr. 2011;94:913–20.
BMC Med. 2015;13:39. https://doi.org/10.1186/ 112. Walter S, Mackenbach J, Vokó Z, et al. Genetic,
s12916-015-0287-6. physiological, and lifestyle predictors of mortal-
109. Petersen KEN, Johnsen NF, Olsen A, et al. The com- ity in the general population. Am J Public Health.
bined impact of adherence to five lifestyle factors 2012;102:e3–e10.
on all-cause, cancer and cardiovascular mortality: 113. Marioni RE, Shah S, McRae AF, et al. DNA meth-
a prospective cohort study among Danish men and ylation age of blood predicts all-cause mortality
women. Br J Nutr. 2015;113(5):849–58. in later life. Genome Biol. 2015;16:25. https://doi.
org/10.1186/s13059-015-0584-6.
Dietary Patterns in Aging
and Disease 2
Keywords
Aging • All-cause mortality • Disease-specific mortality • Chronic disease
• Cardiovascular disease • Cancer • Cognitive function • Alzheimer’s
disease • Dietary patterns • Healthy diets • Western diets • Mediterranean
diet • Telomeres
Key Points aging and frailty whereas higher adherence to

• Dietary pattern habits have a significant effect Western diets increases the odds. Simple
on aging, chronic disease risk, and longevity dietary rules for healthy aging are to limit the
but in developed countries like the US mean intake of energy dense, low nutrient quality
diet quality scores are only about half of the foods, red and processed meats and added salt
optimal healthy eating diet score. Healthy and sugar, and increase the intake of fiber, and
dietary patterns include adequate whole-grain nutrient and phytochemical-rich whole or
cereals, legumes, fruits, vegetables, and nuts, minimally processed plant foods.
and lower intakes of red and processed meats • Healthy dietary patterns, especially the
and sugar-sweetened beverages. Mediterranean diet, are associated with
• Prospective cohort studies with middle and reduced risk of age-related decline in cogni-
older age adults consistently show that high tive performance and Alzheimer’s disease,
adherence to Western dietary patterns and longer telomere length compared to
increases cardiovascular disease (CVD), dia- Western diets.
betes and cancer, and premature death risk • High adherence to healthy dietary patterns
whereas high adherence to healthy/plant- supports healthy aging and reduces chronic
based dietary patterns lowers CVD, diabetes disease and mortality risk through beneficial
and cancer, and premature mortality risks. effects on body weight regulation and adipos-
• Among cancer survivors, healthy diets are ity, energy metabolism, lipoprotein concen-
associated with reduced overall mortality risk trations and function, blood pressure,
by 20% and Western diets are associated with glucose-insulin homeostasis, oxidative stress,
increased risk by over 40%. inflammation, endothelial health, hepatic
• For older adults, higher adherence to healthy function, cardiac and cognitive function, telo-
dietary patterns reduces the odds of unhealthy mere length, and the microbiota ecosystem.

30 2 Dietary Patterns in Aging and Disease
2.1 Introduction meats and sugar-sweetened beverages, generally

promote healthy aging and lower of risk prema-
By 2050, the world’s aging population (≥60 years) ture chronic disease and mortality [7, 8]. For
is forecasted to double to two billion making up example, these healthy diets reduce risk of com-
21% of the population and exceeding the number mon age-related diseases such as cardiovascular
of young children for the first time [1, 2]. The disease, dementia, and cancers by increasing
growing aging population will increasingly stress dietary levels of major shortfall nutrients such as
health care systems and present challenging eco- dietary fiber (fiber), calcium, vitamin D and potas-
nomic implications with major effects on the sium. Healthy diets also reduce intake of choles-
labor force, health, social security and family sup- terol raising saturated and trans fats, sodium and
port systems. It has been estimated that poor life- refined carbohydrates, which are of public health
style choices account for about 70–80% of the concern because of their association in the scien-
risk of chronic diseases in later life and the quality tific literature to adverse health outcomes,
of aging, successful vs. premature [1–6]. Globally, unhealthy aging and premature death rates [5–13].
the high rates of Western diets and sedentary life- The objective of this chapter is to comprehen-
style choices, especially since the 1980s have sively review the effects of dietary patterns on
resulted in pandemic rates of obesity, metabolic aging and chronic disease risk.
syndrome, type 2 diabetes (diabetes) and other
related health conditions leading to sub-optional
aging [5–11]. Healthy aging is the absence of pre- 2.2 ietary Patterns in Aging
D
mature chronic disease, lack of physical disabil- and Disease
ity, and the presence of good social engagement
and mental health [3]. Of the lifestyle choices, the Suboptimal dietary patterns, with poor nutrition
type of dietary pattern followed appears to have quality and/or excess energy intake, are a leading
one of the most significant effects on aging and risk factor leading to unhealthy aging, increased
longevity but in developed countries like the US chronic disease risk, and premature death in the
the mean dietary quality score is only about half US and worldwide [14]. Overconsumption of
of the optimal diet score (53 vs. 100) [3–13]. unhealthy foods and under-consumption of
Older adults often have major health, environ- healthier foods remains a major global public
ment and lifestyle challenges that make meeting health challenge. Western dietary patterns, espe-
dietary recommendations difficult [1]. Some of cially those with excess calories, contribute to
the impediments to consuming healthy nutrient- unhealthy levels of saturated fat, added sugars,
dense diets include changes in taste and smell, high sodium foods, red and processed meats,
loss of appetite, dental and chewing problems, refined grains, and energy dense desserts and bev-
less family support and limitations in mobility for erages, which are associated with an increased
accessing high-quality fresh whole foods. Also, risk for overweight, obesity, chronic disease, such
aging-related inefficiencies in essential nutrient as hypertension, dyslipidemia, insulin resistance
absorption and utilization tends to increase nutri- and colorectal cancer, disability, and premature
ent requirements, which leads older adults to death [14]. Table 2.1 provides a list of food cate-
commonly fall below recommendations for intake gory substitutes for a Western dietary pattern to
of a number of nutrients, including protein, achieve a healthy dietary pattern, which can help
omega-3 fatty acids, dietary fiber, carotenoids in the maintenance of a healthy body weight and
(vitamin A precursors), calcium, magnesium, help to avoid the development of unhealthy aging
potassium, and vitamins B-6, B-12, D, and E, the and chronic disease [14, 15]. In that context,
lack of which are related to elevated health risk. under consumption of whole grains, vegetables,
Healthy dietary patterns, containing adequate fruits, and nonfat and low-fat dairy by the vast
whole-grain cereals, legumes, fruits, vegetables, majority of the population in developed coun-
and nuts, and lower intakes of red and processed tries has resulted in inadequate intakes of dietary
2.2 Dietary Patterns in Aging and Disease 31
Table 2.1 Diet modifications to change from a Western to a healthy dietary pattern (adapted from [14, 15])
Western dietary pattern (foods to discourage) Healthy dietary pattern (foods to encourage)
Vegetables mixed with sauces or fried Vegetables (e.g., fresh, frozen, dried with low to no salt;
roasted, microwaved, stir-fried in healthy oils or steamed)
Fruit pies, jams, jellies or fruit juices with added sugar Whole or minimally processed fruits (fresh/frozen/
dried/canned without added sweetener, canned in
natural juice)
Refined-grains (e.g., white bread or rice, cookies, Whole grains and grains high in dietary fiber
granola bars, crackers, sugar rich cereals, donuts, cakes)
with added sugar and solid fats or little to no fiber.
French fries, hash brown potatoes or white rice Legumes (beans, peas), sweet potatoes
Salted or candy-coated nuts or chips Unsalted nuts and seeds; cut vegetables or fruit; baked,
low-sodium chips; unsalted popcorn; low sodium dried
seaweed
Full-fat dairy Low-fat and nonfat milk, dairy products, calcium-
fortified nondairy milks
Poultry with skin or fried Poultry (skinless; grilled/baked/broiled)
Fish (battered and fried) Fish and seafood (grilled with unsaturated oils/baked/
broiled)
Processed meats (e.g., sausage and hot dogs) Lean meat
Sugar sweetened beverages (soda, sweet teas, fruit Water and beverages without added sugars
drinks, sports drinks or energy drinks
fiber (current intakes are only half the recom- habits have a strong direct metabolic influence on
mended (28–30 g/day), potassium, calcium, and aging and chronic disease risk factors, including;
vitamin D, all considered major shortfall nutri- obesity, lipoprotein concentrations and function,
ents of global public health concern. An over- blood pressure, glucose-insulin homeostasis, oxi-
weight and obesity epidemic currently affects the dative stress, inflammation, endothelial health,
majority of the US population, including an esti- hepatic function, adipocyte metabolism, cardiac
mated 33% of all US children. There are many function, metabolic expenditure, pathways of
reasons for poor adherence to healthy dietary rec- weight regulation, visceral adiposity, and the
ommendations: (1) inadequate knowledge or pri- microbiota ecosystem. Worldwide, chronic dis-
ority focus on nutrition guidance, cost factors, or eases are forecasted to cause an estimated $17
lack of motivation to change; and (2) lack of trillion of cumulative economic loss between
access to or availability of healthy foods, and easy 2011 and 2030 from healthcare expenditures,
access to unhealthy foods are common, especially reduced productivity, and other economic losses.
in certain geographic areas. Also, healthcare pro- Human research on the effects of dietary pat-
viders typically lack financial incentive, interest, terns on health and longevity has increased sev-
knowledge, and office visit time to discuss health eral fold over the past decade or so. Some of these
dietary and other lifestyle guidance with their findings show that simple dietary rules can help
patients [15]. A comparison of the foods and promote healthy aging including limiting the
nutrient content of the Western and common intake of energy dense, low nutrient quality foods,
healthy dietary patterns is found in Appendix A. red and processed meats and added salt and sugar,
Over the last two decades, the critical impor- and increasing the intake of fiber, nutrient and
tance of nutrition science in health management phytochemical -rich whole or minimally pro-
has been elevated by a significant increase in cessed plant foods [2, 5, 7, 8, 16]. Important dif-
more rigorous evidence from well-designed pro- ferences between the Western diet and healthy
spective cohort studies and randomized controlled dietary patterns including those based on the US
trials (RCTs) [14]. It is now evident that dietary Dietary Guidelines, Dietary Approaches to Stop
Hypertension (DASH), Mediterranean diet lowered all-cause and CVD mortality by 23% and
(MedDiet), lacto-ovovegetarian or healthy vegan cancer mortality by 17% [18]. A 2017 pooled anal-
diets are summarized in Appendix A. All of these ysis of the Nurses' Health Study and Health
dietary patterns meet or exceed the US adequate Professionals Follow-up study found that a 20%
intake of fiber (≥28–30 g/day compared to the improvement in diet quality was significantly asso-
usual diet which contributes about half of the ade- ciated with lower all-cause mortality by 8 to 17%
quate fiber intake) and other major shortfall nutri- and CVD mortality by 7 to 15% compared to no
ents including potassium and calcium by change in diet quality [19]. A 2016 meta-analysis
increasing intake of whole or minimally pro- of cancer survivors (117 cohort studies; 209,597
cessed plant foods (e.g., fruits, vegetables, whole- subjects) found adherence to healthy dietary pat-
grains, legumes and nuts/seeds), reducing levels terns to be inversely associated with overall mortal-
of energy dense food, and limiting levels of ity, whereas a Western dietary pattern was positively
refined carbohydrates and saturated fats found in associated with overall mortality (Fig. 2.1) [20]. A
the Western diet. These healthy dietary patterns 2015 meta-analysis (13 cohorts; 338,787 partici-
are associated with healthier aging quality of life, pants) demonstrated that healthy dietary patterns
physical function and cognitive function [2] and significantly lowered mortality risk for all-cause
lower chronic disease risk and premature death mortality by 24% and cardiovascular disease
[14]. The effect of dietary patterns will now be (CVD) mortality by 19% whereas the Western
assessed based on mortality risk, chronic disease dietary pattern insignificantly increased the risk of
risk, effects on healthy aging and physical func- all-cause, CVD and stroke mortality risk by 1–7%
tion, age-related cognitive function and dementia, [21]. A meta-analysis of a number of healthy
and telomere length (cellular senescence). dietary patterns (15 cohorts; 1,020,642 partici-
pants) showed that subjects with the highest adher-
ence to the Healthy Eating Index (HEI), the
2.2.1 Mortality Risk Alternate Healthy Eating Index (AHEI), and the
Dietary Approaches to Stop Hypertension (DASH)
Twenty-five prospective cohort studies or their had similar significantly lowered risk for all-cause
meta- or pooled analyses on dietary patterns and mortality by 22%, CVD mortality by 22%, and
mortality risk include three 2017 meta- or pooled cancer mortality by 15% compared to the lowest
analyses [17–19] and 22 meta-or pooled analyses adherence scores [22]. The Consortium on Health
and cohort studies from 2016 or earlier and are and Ageing: Network of Cohorts in Europe and the
summarized in Table 2.2 [20–41]. US (CHANCES) pooled analysis (10 cohorts;
281,874 adults ≥60 years; 5–15+ years of follow-
2.2.1.1 Meta-analyses and Pooled Data up) showed lowered CVD mortality risk in
Ten meta-analyses and pooled prospective data Southern Europeans by 13% and US participants
analyses provide strong evidence that healthy by 15% with higher adherence to the WHO dietary
dietary patterns reduce the odds of premature mor- guidelines [23]. A second CHANCES pooled anal-
tality [17–26]. A 2017 meta-analysis (225 cohort ysis (11 cohorts; 396,391 adults ≥60 years;
studies) found that the optimal intake of healthy 5–15 years of follow-up) found significantly
foods (e.g., whole grains, fruits, vegetables, reduced all-cause mortality by about 10% (or an
legumes, nuts, and fish) significantly reduced all- average of 2 years increased longevity) for each
cause mortality by 56% compared to a 2-fold 10-point improvement in adherence to the WHO
increased risk with the high intake of unhealthy dietary guidelines [25]. A meta-analysis of vegetar-
foods (e.g., red and processed meats and sugar- ian diets (7 cohort studies; 124,706 adults) showed
sweetened beverages along with low intake of that vegetarian diets lowered risk for mortality for
healthy foods) [17]. Another 2017 meta-analysis all-cause by 9%, CVD by 16%, ischemic heart dis-
(12 cohort studies) showed that the highest adher- ease by 29% and stroke by 12%, and overall cancer
ence to the Healthy Eating Indices significantly incidence by18% compared to nonvegetarians
Table 2.2 Summary of dietary pattern prospective cohort studies on all-cause and chronic disease mortality risk
Systematic reviews and meta-analyses or large pooled data sets
Schwedhelm et al. (2016) 117 cohort studies; 209,597 In cancer survivors, adherence to a
Investigate the association cancer survivors; 60,134 high-quality diet index and a prudent/healthy
between dietary patterns and deaths. dietary pattern is inversely associated with
overall mortality among cancer overall mortality, whereas a Western dietary
survivors [20]. pattern is positively associated with overall
mortality (Fig. 2.1).
Li et al. (2015) 13 cohort studies; 338,787 Healthy dietary patterns were significantly
Assess the effects of adherence to participants. inversely associated with reduced mortality
dietary patterns on all-cause, risk for all-cause by 24% and CVD by 19%,
CVD, and stroke mortality risk whereas stroke mortality risk was
[21]. insignificantly reduced by 11% (for highest
vs. lowest adherence). Western diets
insignificantly increased all-cause, CVD,
and stroke mortality risk by 1–7%.
Schwingshackl and Hoffmann 15 cohort studies; 1,020,642 Diets that scored highly on the HEI, AHEI,
(2015) subjects; healthy dietary and DASH were associated with a
Examine the associations of diet patterns including the Healthy significant reduction in mortality risk for
healthy dietary quality score and Eating Index (HEI), Alternate all-cause by 22%, CVD by 22%, and cancer
mortality risk [22]. Healthy Eating Index (AHEI), by 15%.
and DASH diets.
Jankovic et al. (2015) 10 prospective cohort studies; Greater adherence to the WHO dietary
Examine the effects of WHO 281,874 healthy men and guidelines was insignificantly associated
dietary guidelines on CVD women aged ≥60 years; with reduced CVD mortality risk by 6%,
mortality in persons aged WHO healthy dietary after multivariable adjustments. However, the
≥60 years (Consortium on Health guidelines; median follow-up results varied by region with significant
and Ageing: Network of Cohorts of 5 to 15 years. inverse CVD associations observed in
(CHANCES) in EU and US) [23]. southern Europe by 13% and the US by 15%.
Huang et al. (2012) 7 cohort studies; 124,706 Vegetarians had significantly lower ischemic
Estimate the effect of vegetarian individuals. heart disease mortality risk by 29% and
dietary patterns on CVD mortality overall cancer incidence by18% than
and cancer incidence [24]. nonvegetarians.
Jankovic et al. (2014) 11 cohort studies; 396,391 Greater adherence to the WHO guidelines
Investigate the association participants; 42% women; was associated with greater longevity in
between the adherence to WHO baseline age ≥ 60 years; older men and women in Europe and the US
dietary guidelines, and all-cause WHO healthy dietary with an average lower risk of premature
mortality in men and women guidelines; median follow-up death for men by 10% and women by 11%
≥60 years of age from Europe and of 5 to 15 years. for each 10-point increase in adherence
the US (CHANCES) [25]. score, after multivariable adjustments. These
estimates translate to an increased mean life
expectancy of 2 years after age ≥ 60 years.
Sofi et al. (2010) 18 cohorts; 2,190,627 A 2-point increase in adherence to the
Assess the effect of adherence to follow-up of 4 to 20 years. MedDiet was associated with a significant
the Mediterranean diet (MedDiet) reduced mortality risk from all-causes by
on disease incidence [26]. 8%, CVD by 10% and from cancer by 6%.
Gopinath et al. (2016) 1609 healthy adults; baseline Higher adherence to dietary guidelines had
Examine the relationship between age ≥ 49 years; 10-year 20% lower odds of premature death after
overall diet quality (reflecting follow-up; 610 (37.9%) died; 10 years compared to poor adherence to
adherence to dietary guidelines) 249 (15.5%) aged guidelines. Adherence to recommended
and mortality and successful aging successfully. dietary guidelines significantly increases the
in older adults (Blue Mountains probability of reaching old age disease free
Eye Study; Australia) [27]. and fully functional.
(continued)
Yang et al. (2015) 22,071 male physicians; Comparing men in the highest versus the
Explore the effect of diet on baseline age 40–84 years; lowest quartile of the Western pattern, there
prostate mortality risk mean diagnosis age 68 years; was a significant 150% increased risk for
(Physicians’ Health Study I or II; 926 men diagnosed with prostate cancer-specific mortality and 67%
US) [28] non-metastatic prostate cancer, for all-cause mortality. The healthy pattern
333 men died, 17% of prostate was associated with a significantly lower
cancer mortality; median post diagnosis all-cause mortality by 36%.
9.9-years of follow-up.
Harmon et al. (2015) 215,782 participants; mean All healthy indices were significantly
Assess the effect of 4 healthy diet baseline age 59 years; mean associated with reduced risk of mortality
quality indices On risk of age death 75 years; Healthy from all causes, CVD, and cancer in both
mortality from all causes, CVD, Eating Index-2010 (HEI- men and women (Fig. 2.2). The highest vs.
and cancer in men and women 2010), the Alternative lowest adherence of these healthy diets
(Multiethnic Cohort; US) [29]. HEI-2010 (AHEI-2010), the lowered risk of mortality from all causes,
alternate Mediterranean diet CVD, or cancer for men by 17–26%, and
score (aMED), DASH diets; women by 11–24%.
13–18 years of follow-up;
34,430 deaths, 11,919 deaths
were from CVD, and 10,883
deaths were from cancer.
Reedy et al. (2015) 492,823 participants; mean Higher diet quality adherence scores were
Examine the effects of major baseline age 62 years; dietary significantly and consistently associated
healthy dietary patterns on patterns including HEI-2010, with a significantly reduced risk of death due
mortality risk (NIH-AARP Diet AHEI-2010, aMed, and to all causes, CVD, and cancer compared
and Health Study, the Multiethnic DASH diet scores; 15-year with the lowest adherence, independent of
Cohort, and the Women’s Health follow-up; 86,419 deaths, known confounders. In men, mortality risk
Initiative including 23,502 CVD and was reduced for all cause by 17–24%, CVD
Observational Study; US) [30]. 29,415 cancer-specific deaths, by 15–26%, and cancer by 18–24%. In
on all-cause, CVD, and women, mortality was reduced for all-cause
cancer mortality. by 22–24%, for CVD by 21–28% and for
cancer by 12–21%.
Liese et al. (2015) 492,823 participants; mean In women, a high diet quality score was
Evaluate the relationships between baseline age early 60s; dietary associated with a lower risk of all-cause
4 diet quality indices and all cause, pattern score - HEI-2010, mortality by 18–26%, CVD mortality by
CVD, and cancer mortality AHEI-2010, aMED, and 19–28%, and cancer mortality by 11–23%.
(Dietary Patterns Methods Project/ DASH; 15 years of follow-up. In men, high diet quality score was
NIH-AARP Diet and Health associated with a lower risk of all-cause
Study; US) [31]. mortality by 17–25%, CVD mortality by
14–26%, and cancer mortality by 19–24%.
Shi et al. (2015) 8959 participants; baseline Fruit and vegetable intakes were inversely
Estimate the effects of food habits, age > 80 years; follow-up associated with mortality risk with a 15%
lifestyle factors and all-cause home visits over 6.5 years for reduction for fruit and 26% reduction for
mortality in elderly adults men and 12 years for women; vegetables. However, the intake of salt-
(Chinese Longitudinal Healthy 6626 deaths. preserved vegetables increased premature
Longevity Survey) [32]. death risk 10%.
George et al. (2014) 64,000 participants; mean Across indices and after adjustment for
Study the relationships between 4 baseline age 63 years; dietary multiple covariates, having high diet quality
diet quality indices and all cause, pattern scores -HEI-2010, scores was significantly associated with a
CVD, and cancer mortality among AHEI-2010, aMED, and 18%–26% lower all-cause and CVD
postmenopausal women (Women’s DASH 12.9 years of mortality risk. Higher HEI, aMED, and
Health Initiative Observational follow-up; 5692 deaths, DASH (but not AHEI) scores were
Study; US) [33]. including 1483 from CVD associated with a significant reduction in
and 2384 from cancer. risk of cancer mortality by approx. 20%.
Zazpe et al. (2014) 16,008 middle-aged Spanish The traditional Mediterranean diet was
Evaluate the effect of dietary University graduates; mean associated with a significant reduction in the
patterns on all-cause mortality baseline age 38 years; 60% risk of all-cause mortality among middle-
(Seguimiento Universidad de women, 12.9 years of aged adults by 47% whereas the Western
Navarra Project; Spain) [34]. follow-up. diet was not significantly associated with
all-cause mortality.
Yu et al. (2014) 61,239 men and 73,216 A higher pagoda score was associated with
Assess the effect of adherence to women; mean baseline age significantly lower total mortality with
the Chinese Food Pagoda on total early to mid-50s; 2954 deaths multivariable-adjusted risk in men by 33%
and cause-specific mortality in men and 4348 deaths in and in women by 13% (high vs. low
(Shanghai Men’s Health Study and women; mean follow-ups of adherence. Significant lower risks for CVD,
Shanghai Women’s Health Study) 6.5 and 12.0 years. cancer, and diabetes mortality, particularly
[35]. in men were associated with higher pagoda
scores.
Martinez-Gonzalez et al. (2014) 7216 high CVD risk Among omnivorous subjects those with food
Examine the effect of a pro- participants, 57% women; patterns that emphasized whole plant-
vegetarian dietary pattern on mean baseline age 67 years; derived foods (pro-vegetarian) were
all-cause mortality risk median 4.8 years of associated with a 41% reduced risk of
(Prevencion con Dieta follow-up; 323 deaths. all-cause mortality.
Mediterranea (PREDIMED)
sub-study; Spain [36].
Orlich et al. (2013) 96,500 Seventh-day Adventist The adjusted all-cause mortality risk was
Evaluate the effect of vegetarian men and women; mean significantly reduced for all vegetarians
dietary patterns on mortality follow-up of 5.8 years; 2570 combined by 12% vs. non-vegetarians. The
(Adventist Health Study 2; US) deaths. adjusted risk for all-cause mortality was
[37]. reduced for vegans by 15%, lacto-ovo–
vegetarians by 9%, pesco-vegetarians by
19%, and in semi-vegetarians by 8% vs.
nonvegetarians.
Akbaraly et al. (2011) 7319 participants; mean Higher adherence to the AHEI was
Examine the effects of adherence baseline age 49.5 years; 30% significantly associated with 25% lower
to the Alternate Healthy Eating women; follow-up of all-cause mortality, 40% lower CVD
Index (AHEI) on all-cause and 18 years. mortality and an insignificant 20% lower
cause-specific mortality cancer mortality.
in a British working population
(Whitehall II Study; UK) [38].
Anderson et al. (2011) 3075 adults; mean baseline High-fat dairy, sweets and desserts dietary
Assess the effect of dietary age 74 years; approx. 40% intake was associated with a 1.4-fold higher
patterns in older adults on survival men; follow-up of 10 years. risk of mortality compared to a healthy
(The Health, Aging, and Body foods cluster, characterized by higher intake
Composition Study; US) [39]. of low-fat dairy products, fruit, whole grains,
poultry, fish, and vegetables.
Buckland et al. (2011) 40,622 adults; mean baseline A high MedDiet score compared with a low
Explore the effects of adherence to age 48 years; mean BMI 28; score was associated with a significant
the MedDiet on mortality risk 38% males; mean follow-up reduction in mortality from all causes by
(EPIC-Spain) [40]. of 13.4 years; 1855 deaths. 21%, from CVD by 34%, but not from
overall cancer by 8%. A 2-unit increase in
MedDiet score was associated with a
significant 6% lower risk of all-cause
mortality.
(continued)
Heidemann et al. (2008) 72,000 women; mean baseline Higher adherence to a healthy dietary pattern
Evaluate the effect of healthy vs. age of 50 years; 18-years of significantly lowered mortality risk of
Western dietary patterns on the follow-up. all-cause by 17% and CVD by 28% vs.
mortality risk of women (The higher adherence to a Western dietary
Nurses’ Health Study; US) [41]. pattern which significantly increased risk of
mortality from all causes by 21%, CVD by
22%, and cancer by 16% (Fig. 2.3).
Barnia et al. (2007) 74,607 adults; baseline Higher adherence to a plant-based diet was
Investigate the effect of dietary age ≥ 60 years; 3–21 years of associated with a significant lower overall
patterns on overall survival of follow-up. mortality by 14% per one standard deviation
older Europeans (EU EPIC- increment.
Elderly cohort) [42].
Cancer Survivors
1.6
1.4
Risk Ratio of Overall Mortality
1.2
0.8
0.6
0.4
0.2
0
Diet Quality Index Prudent/Healthy Western
Dietary Pattern
Fig. 2.1 Association between the type of dietary pattern consumed and overall mortality in cancer survivors (adapted
from [20])
[24]. For the Mediterranean diet (MedDiet), a 9.9-year follow-up) the Western diet was shown to
meta-analysis (18 cohorts; 2,190,627 subjects) significantly increase prostate cancer mortality by
showed that higher adherence significantly reduced 150% and all-cause death by 67% compared to a
mortality from all-causes, CVD, and cancer by 36% lower risk of all-cause mortality after diagno-
6–10% for each 2-point increase in adherence [26]. sis for healthy patterns [28]. In a prospective study
of older US adults (3075; mean age of 74 years at
2.2.1.2 Specific Cohort Studies baseline; 10 years of follow-up), high-fat dairy,
Fourteen prospective studies report consistent sweets and desserts dietary intake was associated
beneficial effects of high adherence to healthy with a 1.4-fold higher multivariate risk of mortal-
dietary patterns on all-cause and disease specific ity compared to a healthy foods, characterized by
mortality risk [27–41]. higher intake of low-fat dairy products, fruit,
whole grains, poultry, fish, and vegetables [39].
estern Dietary Patterns
W
Studies consistently show that high adherence to ealthy Dietary Patterns
H
the Western dietary pattern has adverse effects on Non-US Cohorts. A number of prospective cohort
mortality risk. In the Physicians’ Health Study studies from Australia, the EU and China consis-
(22,071 men; mean age at diagnosis 68 years; tently show that high adherence to a variety of
healthy dietary patterns significantly reduced mor- reduced all-cause mortality risk compared to those
tality risk or promoted longevity [27, 32, 34, 35, consuming diets low in whole plant foods [36].
37, 40, 41]. The Australian Blue Mountains Eye US Cohorts. Several large cohort studies show
Study (1609 participants; ≥ 49 years; 10 years of that higher adherence to healthy dietary patterns,
follow-up) observed that higher adherence to including the HEI-2010, AHEI-2010, the alter-
healthy dietary patterns reduced risk of mortality nate Mediterranean Diet (aMedDiet), and DASH
by 20% [27]. The Chinese Longitudinal Healthy diets have similar effects on all-cause and disease
Longevity Survey (8959 participants; ≥80 years; specific mortality because all these patterns have
follow-up of 6.5 years for men and 12 years for similar core dietary components such as empha-
women) reported that fruit and vegetable intake sizing whole plant foods with limits on red and
was inversely associated with all-cause mortality processed meats, refined grains and sugar sweet-
[32]. Also, the Shanghai Women and Men Health ened beverages [29–31]. The US Multiethnic
Studies (61,239 men and 73,216 women; age early Cohort (215,782; mean age 58 years; followed
to mid-50s; 6.5 and 12 years of follow-up for men for 13–18 years) found that high adherence to all
and women, respectively) found that a higher of the healthy dietary patterns lowered all-cause
adherence to the Chinese (Healthy) Food Pagoda mortality in men by 19–25% and in women by
lowered total mortality by 33% in men and 13% in 20–22% (Fig. 2.2) [29]. These healthy dietary
women along with lower CVD, cancer and diabe- patterns also decreased disease specific mortality
tes mortality [35]. Three EU studies observed that for CVD in men by 17–26% and in women by
the consumption of healthy dietary patterns in 19–24%, and for cancer in men by 17–24% and
mid-life (45–65 years) significantly reduced all- in women by 11–15%. The 2015 NIH-AARP
cause mortality in both men and women, which Diet and Health Study found that high adherence
was primarily as a result of lower CVD mortality to major healthy dietary patterns significantly
[38, 40, 42]. In the British Whitehall II Study reduced premature death by 12–28% [30]. In
(7319 mid-age adults, 30% women; 18 years of women, mortality was reduced for all-cause by
follow-up) higher adherence to the AHEI dietary 22–24%, for CVD by 21–28%, and for cancer by
pattern (rich in fruit, vegetables, nuts, soy, white 12–21%. In men, mortality was reduced for all-
meat, fiber, polyunsaturated fat and multivitamin cause by 17–24%, for CVD by 15–26%, and for
use and lower in red meat, saturated and trans-fat, cancer by 18–24%. Similar findings were
and alcohol) lowered mortality risk from all-cause observed in the Dietary Patterns Methods/NIH-
by 25% and CVD by 40% [38]. A 2011 European AARP and the Women’s Health Initiative
Prospective Investigation into Cancer and Observational Study [30, 31, 33]. The Adventist
Nutrition (EPIC)-Spanish cohort (40,622 middle- Health Study II observed that all-cause mortality
aged adults, mean follow-up of 13.4 years) found was reduced for vegans by 15%, lacto-
that a higher MedDiet score was associated with a ovovegetarian by 9%, pesto-vegetarians by 19%,
significant reduction in mortality from all causes and semi-vegetarians by 8% vs. nonvegetarians
by 21% and from CVD by 34% compared to a low [37]. The Nurses’ Health study (72,113 women;
score [40]. A 2-unit increase in MedDiet score was mean age of 50 years; 18 years of follow-up)
associated with a significant 6% lower risk of all- observed that total mortality risk, especially CVD
cause mortality. A 2007 EPIC study of a cohort of mortality, was reduced with adherence to a
older adults age ≥60 years observed that each one healthy diets and significantly increased with
standard deviation increase in plant foods in com- adherence to Western diets (Fig. 2.3) [41].
bination with the avoidance of margarine, sugar-
sweet beverages and potatoes, was associated with
a significant 14% lower all-cause mortality [42]. A 2.2.2 Chronic Disease Risk
Prevencion con Dieta Mediterranea (PREDIMED)
sub-study demonstrated that among omnivorous Chronic diseases such as CVD, stroke, and type 2
older adults with high CVD risk, those with food diabetes become more common with aging and
patterns rich in whole plant foods had a 41% the rates of these chronic diseases are increasing
HEI-2010 aMedDiet DASH Diet AHEI-2010
All-cause Mortality Hazard Ratio

1.1
Men
1.0
0.9
0.8
0.7
Q-1 Q-2 Q-3 Q-4 Q-5
Dietary Pattern Adherence Quintile
HEI-2010 aMedDiet DASH Diet AHEI-2010

All-cause Mortality Hazard Ratio
1.1
Women
1.0
0.9
0.8
0.7
Q-1 Q-2 Q-3 Q-4 Q-5
Dietary Pattern Adherence Quintile
Fig. 2.2 Association between dietary pattern adherence and all-cause mortality risk among US middle-aged men and
women during 18 years of follow-up (p < 0.001 for all; multivariate adjusted) (adapted from [29])
Prudent/Healthy Dietary Pattern Western Dietary Pattern

1.3
1.2
Relative Risk of Total Mortality
1.1
0.9
0.8
0.7
0.6
Q-1 Q-2 Q-3 Q-4 Q-5
Dietary Pattern Adherence (Quintiles)
Fig. 2.3 Association between adherence to dietary patterns and total mortality risk in women (p-trend <0.001 for all
diets; multivariate adjusted) (adapted from [41])
in mid-age because of the global increased intake 2.2.2.1 Healthy vs. Western Dietary
of Western diets, sedentary lifestyles and the obe- Patterns
sity pandemic effects on accelerating the aging Four meta-analyses comparing the effects of
processes [1, 14]. Eight systematic reviews and healthy and unhealthy (Western) dietary patterns
meta- and pooled analyses of observational stud- found that high adherence to healthy diets
ies, and RCTs evaluating the effects of dietary reduces risk of major chronic disease and
patterns on chronic disease risk are summarized Western diets increased this risk [44–47]. A
in Table 2.3 [22, 26, 43–48]. meta-analysis of colorectal cancer (CRC) studies
Table 2.3 Summary of systematic reviews, meta- and pooled analyses of prospective studies and randomized con-
trolled trials (RCTs) on dietary patterns and chronic disease risk
Ndanuko et al. (2016) 17 RCTs; 5014 adults; The pool results of the 4 healthy diets, DASH,
Assess the effect of healthy dietary average mid-age or Nordic diet, MedDiet, and Tibetan diet
patterns on blood pressure (BP) in older; duration significantly lowered systolic BP by 4.3 mm Hg
adults [43]. 6 weeks to 2 years; 13 and diastolic BP by 2.4 mm Hg. The specific BP
RCTs allowed use of lowering effects are summarized in Fig. 2.7.
BP meds.
Feng et al. (2016) 40 studies; 22 cohorts, For the highest vs. lowest adherence: ‘healthy’
Identify the association between 17 case-controls and 1 dietary patterns decreased CRC risk by 25%
dietary patterns and the risk of cross-sectional; mean (p < .00001]; Western style dietary patterns
colorectal cancer (CRC) [44]. subject age mid-age or increased CRC risk by 40% (p < .00001);
older. alcohol-consumption pattern is characterized by
high consumption of beers, wines, and white spirits
increased CRC risk by 44% (p = .003) (Fig. 2.4).
Maghsoudi et al. (2015) 10 cohort studies; For highest vs. lowest adherence: (1) ‘healthy’
Review cohort studies about the 404,528 subjects. dietary patterns significantly lowered diabetes risk
association between dietary by 14% (no difference between men and women)
patterns and type 2 diabetes and (2) ‘unhealthy’ western dietary patterns
incidence [45]. increased diabetes risk by 30% (men were more
significantly affected than women). Subgroup
analysis showed that unhealthy dietary patterns
containing foods with high phytochemical content
insignificantly increased risk by 6% (Fig. 2.5).
Rodriguez-Monforte et al. (2015) 19 cohort studies; Comparing the highest vs. the lowest category: (1)
Estimate the association between number of cases healthy dietary patterns reduced risk of CVD, CHD
empirically derived dietary patterns ranged from 449 to and stroke and unhealthy/Western diets slightly
and cardiovascular disease (CVD), 74,942. increased risk for these vascular chronic diseases
coronary heart disease (CHD) and (Fig. 2.6).
stroke risk [46].
Schwingshackl and 15 cohort studies; Higher healthy diet scores were associated with a
Hoffmann (2015) 1,020,642 subjects; significantly reduced risk of CVD and type 2
Examine the effects of diet quality healthy dietary patterns diabetes by 22% and cancer by 15%, but the effects
scores on aging health outcomes including the Healthy for neurodegenerative diseases were insignificant.
[22]. Eating Index (HEI),
the Alternate Healthy
Eating Index (AHEI),
and DASH diets.
Hou et al. (2015) 12 cohort studies; There was an inverse association between healthy
Evaluate the effects of dietary 409,780 subjects; dietary patterns and CHD risk with a 20% lower
patterns on CHD risk [47]. 4.6–13 years of risk for highest vs. lowest adherence. The Western
follow-up; 6298 CHD diet globally resulted in a 5% increased CHD risk
cases. but in the US cohorts there was a significant 45%
increase in CHD risk.
(continued)
Sofi et al. (2010) 18 cohorts; 2,190,627; A 2-point increased adherence to a MedDiet was
Assess the effect of adherence to the follow-up 4 to associated with significant reduction in incidence
Mediterranean diet (MedDiet) on 20 years. of CVD by 10% and cancer by 6%.
incidence of chronic diseases [26].
Chiuve et al. (2012) 71,495 women, 26,759 The AHEI-2010, which emphasizes high intakes of
Evaluate the association between the chronic disease events, whole grains, PUFA, nuts, and fish and reductions in
Alternative Healthy Eating Index- 24 years of follow-up; red and processed meats, refined grains, and sugar-
2010 (AHEI-2010) and risk of major 41,029 men. 15,558 sweetened beverages, was associated with significantly
chronic diseases (Nurses’ Health chronic disease events, multivariate lower risk of a composite of major chronic
Study and Health Professionals 22 years of follow-up. diseases by 19% (p < .001). Risk reduction for specific
Follow-Up Study; US) [48]. disease is summarized in Fig. 2.8.
50
40
30
Colorectal Cancer Risk (%)
20
10
0
Healthy DP Western DP Alcohol DP
(p <.00001) (p <.00001) (p =.003)
-10
-20
-30
Fig. 2.4 Association between high adherence to dietary pattern (DP) types and colorectal cancer risk from meta-
analysis of 40 observational studies (adapted from [44])
(40 studies; 22 cohorts, 17 case-controls and 1 found that high adherence to healthy dietary pat-
cross-sectional) found that higher adherence to terns reduced incidence of CVD, coronary heart
healthy dietary patterns significantly reduced disease (CHD) and stroke compared to Western
CRC risk by 25% whereas adherence to dietary patterns (Fig. 2.6) [46]. Another meta-
unhealthy/Western diets increased risk by 40% analysis (12 cohort studies; 409,780 subjects)
(both p < .0001; Fig. 2.4) [44]. A meta-analysis reported that healthy patterns lowered CHD risk
(10 cohort studies; 404,528 subjects) showed by 20% and Western patterns increased CHD
that high adherence to healthy dietary patterns risk by 5% globally but US participants with the
lowered type 2 diabetes (diabetes) risk by 14% highest adherence to the Western diet had a sig-
and unhealthy/Western diets increased risk by nificant 45% increase in CHD risk [47]. The
30% (Fig. 2.5) [45]. However, subjects that Reasons for Geographic and Racial Differences
enriched their Western dietary pattern with high in Stroke (REGARDS) prospective study (17,418
phytochemical-rich plant foods reduced their subjects; national, population-based, longitudi-
increased diabetes risk from 30 to 6% [45]. For nal study of adults aged ≥45 years; median fol-
vascular diseases, a meta-analysis (19 cohorts) low-up of 5.8 years) found that high adherence to
35
30
25
20
Type 2 Diabetes Risk (%)
15
10
0
Healthy DP Unhealthy DP Unhealthy DP plus
-5 (p =.60) (p =.03) Foods with High
Phytochemical
-10 Content (p >.05)
-15
-20
Fig. 2.5 Association between high adherence to dietary patterns (DP) and type 2 diabetes risk from meta-analysis of
10 cohort studies (adapted from [45])
Healthy Dietary Pattern Western Dietary Pattern

1.2
0.8
Relative Risk
0.6
0.4
0.2
0
Cardiovascular Disease Coronary Heart Disease Stroke
Fig. 2.6 Association between dietary patterns quality and vascular chronic disease incidence risk from a meta-analysis
of 19 cohort studies (adapted from [46]
the US Southern dietary pattern (characterized 2.2.2.2 Healthy Dietary Patterns

by added fats, fried food, eggs, organ and pro- Adherence
cessed meats, and sugar-sweetened beverages) Four meta-analyses consistently found that
significantly increased risk of acute CHD by increased adherence to a range of healthy dietary
36% compared to low adherence (or a healthier patterns significantly reduced incidence rates of
diet) [49]. CVD, coronary heart disease (CHD) and cancer
Systolic BP Diastolic BP
1
Healthy Dietary Patterns
0
DASH Diet Nordic Diets MedDiet Tibetan Diet
Blood Pressure (mm Hg)
-1
-2
-3
-4
-5
-6
Fig. 2.7 Effect of dietary pattern (DP) type on lowering blood pressure from a meta-analysis of 17 RCTs (adapted from
[43])
[22, 26, 43, 48]. A meta-analysis on four healthy are summarized in Table 2.4 [2, 27, 50–55]. A
diets (17 RCTs; 5014 adults) found that the DASH 2016 systematic review of observational studies
and Nordic diets were most effective in signifi- in older adults (23 longitudinal and 11 cross-
cantly lowering both systolic and diastolic blood sectional studies) concluded that higher adher-
pressure (Fig. 2.7) [43]. A meta-analysis of major ence to healthy dietary patterns or diet quality
healthy dietary patterns (15 cohorts; 1,020,642 indexes was associated with improved health
subjects) found that higher adherence to these related quality of life [2]. This review also showed
diets reduced incidence for CVD and diabetes by high variability between study outcomes and
22% and cancer by 15% [22]. In a meta-analysis indicated that more, larger studies are required to
of MedDiets (18 cohorts; 2,190,627 participants), better understand the diet and aging relationship.
it was shown that a 2-point increase in adherence The Australian Blue Mountains Eye Study in
reduced incidence of CVD by 10% and cancer by adults ≥49 years (1609 participants; 10 years of
6% (9 point scale) [26]. The Nurses’ Health Study follow-up) observed that higher adherence to
and Health Professionals Follow-up Study healthy dietary patterns improved the odds of
(71,495 women and 41,029 men; age 30–75 years healthy/successful aging by 58% (Fig. 2.9) [27].
at baseline; 22–24 years of follow-up) found that In a French study (2796; mean age 52 years; fol-
the Healthy Eating Index-2010 based on the 2010 lowed for 13 years), subjects with high adherence
US Dietary Guidelines was inversely associated to a Western dietary pattern had 17% higher odds
with the risk of CVD, stroke, diabetes, and cancer for unhealthy aging compared to significantly
(Fig. 2.8) [48]. better healthy aging odds by 46% for a moderate
energy healthy dietary pattern (Fig. 2.10) [52]. In
a study of older adults (1872 subjects; mean age
2.2.3 H
ealthy Aging and Physical 69 years; 3.5 years of follow-up), those consum-
Function (Frailty) ing a healthy dietary pattern had an inverse rela-
tionship with the risk of frailty and those
One systematic review of observational studies consuming a Westernized pattern had an increased
and seven cohort studies on the effects of dietary frailty risk by 61% (Fig. 2.11) [53]. The British
patterns on general health and physical function Whitehall II study (5350 adults; mean age at
CVD Stroke Type 2 Diabetes Cancer

1.1
1
Relative Risk
0.9
0.8
0.7
0.6
Q-1 Q-2 Q-3 Q-4 Q-5
AHEI-2010 Adherence Quintile
Fig. 2.8 Association between 2010 Alternative Healthy Eating Index (AHEI-2010) and chronic disease risk from the
pooled analysis of the Nurses’ Health Study and Health Professionals Follow-Up Study (p-trend <.001; for all except
cancer with p-trend = .003; multivariate adjusted) (adapted from [48])
Table 2.4 Summary of dietary patterns studies on healthy aging and physical function (frailty)
Systematic reviews
Milte et al. (2016) 11 cross-sectional and 23 In older adults, this analysis found that higher
Examine dietary patterns and longitudinal studies. adherence to healthy dietary patterns or diet
quality of life, physical quality indexes was associated with improved
function, cognitive function health related quality of life. This review also
and mental health among showed high variability between study
older adults [2]. outcomes and indicated that more larger studies
are required to better understand the diet and
aging relationship.
Prospective studies
Pilleron et al. (2017) 972 initially non-frail and Men in the ‘pasta’ pattern and women in the
Analyze the association of nondemented participants; 336 ‘biscuits and snacking’ pattern had a significant
dietary patterns and frailty in men and 636 women; average 2-fold higher risk of frailty compared to those in
community-dwelling elderly baseline age 73 years; 12-year the ‘healthy’ pattern (p = .09 for men and
(France) [50]. follow-up; 78 men and 221 p = 1.3 for women). Men had a 3-fold greater
women became frail. risk for muscle weakness with the ‘biscuit and
snacking’ pattern vs. the healthy pattern
(p = .003).
Gopinath et al. (2016) 1609 healthy adults; baseline High adherence to healthy dietary guidelines
Examine the relationship aged ≥49 years; 10 years of resulted in 58% higher odds of successful aging
between overall diet quality follow-up; 610 (37.9%) died; after 10 years compared to poor adherence to
(reflecting adherence to 249 (15.5%) aged successfully. guidelines (Fig. 2.9). Adherence to
dietary guidelines) and recommended dietary guidelines significantly
mortality and successful increases the probability of reaching old age
aging in older adults (Blue disease free and fully functional.
Mountains Eye Study;
Australia) [27].
(continued)
Hagan et al. (2016) 54,762 healthy women; mean Women with higher AHEI-2010 scores were
Examine the association baseline age 56 years; 18 years significantly13% less likely to have physical
between the Alternative of follow-up. impairment than those with low scores
Healthy Eating Index-2010 (multivariable-adjusted). Further, significantly
(AHEI-2010), a measure of reduced physical impairments were also
diet quality, with impairment observed for higher intake of vegetables and
in physical function with fruits; lower intake of sugar sweetened
aging (Nurses’ Health Study; beverages and sodium intake. The strongest
US) [51]. relations were found for increased intake of
oranges, orange juice, apples and pears, romaine
or leaf lettuce, and walnuts.
Assmann et al. (2015) 2796 subjects; mean baseline High adherence at midlife to a healthy dietary
Investigate healthy vs. age 52 years; mean 13.2 years pattern with lower energy intake (≤ 2500 kcal/
Western dietary patterns in of follow-up time; healthy day in men and ≤1820 kcal/day in women)
midlife on healthy aging aging was defined as not resulted in significantly higher odds of healthy
(SUpplementation en developing any major chronic aging by 46% compared to those consuming
Vitamines et Mineraux disease, good physical and above the energy thresholds who had only an
AntioXydants Study; France) cognitive function, no insignificant 7% improved odds of healthy
[52]. limitations in activities of daily aging (Fig. 2.10). Those consuming a Western
living, no depressive symptoms, diet in midlife had 17% lower odds of healthy
no health-related limitations in aging (highest vs. lowest adherence), which was
social life, good overall insignificant after adjusting for potential
self-perceived health, and no confounders.
function-limiting pain over
13 years of follow-up.
Leon-Munoz et al. (2015) 1872 free-living adults; mean High adherence to a healthy dietary pattern had
Study the association of baseline age 70 years; 50% significantly lower risk of frailty by 60%
healthy vs. Western dietary women; 3.5 years of follow-up. compared to high adherence to a Western
patterns on frailty risk in older pattern, which increased risk by 61%
adults (Spanish Seniors- (Fig. 2.11).
ENRICA cohort) [53].
Samieri et al. (2013) 10,670 women; median Higher midlife adherence to healthy dietary
Explore the relation of baseline age 59 years; patterns significantly improved the multivariate
dietary patterns in midlife to Alternative Healthy Eating odds for healthy aging: AHEI-2010 improved
the prevalence of healthy Index-2010 (AHEI-2010) and odds by 34% and the alternative MedDiet
aging; no major chronic Alternate MedDiet scores; improved odds by 46%.
diseases or cognitive, average 15 years of follow-up;
physical function or mental healthy aging was based on
health impairments (Nurses’ survival to 70+ years.
Health Study; US) [54].
Akbaraly et al. (2013) 5350 healthy adults; mean Subjects adhering to the Western pattern
Assess the effects of the baseline age 51 years; 29% (characterized by high intakes of fried and
healthy (Alternative healthy women; mean 16 years of sweet food, processed food and red meat,
eating index [AHEI]) pattern follow-up; participants were refined grains, and high-fat dairy products) had
vs. the Western pattern on screened every 5 years for ideal a significant 42% lower odds of ideal aging (for
healthy aging (Whitehall II aging, composite of metabolic, top vs. bottom tertile), independently of other
study; UK) [55]. cardiovascular, musculoskeletal, health behaviors. The healthy diet improved
respiratory, mental, and trend for ideal aging odds by 8% (Fig. 2.12).
cognitive functions.
51 years; 29% women; mean 16 years of follow- healthy pattern directionally improved the odds
up) showed that high adherence to a Western pat- of ideal aging by 7% (Fig. 2.12) [55]. Two Nurses’
tern at mid-life significantly lowered the odds of Health Studies showed that high adherence to
ideal aging by 42% and high adherence to a healthy dietary patterns in midlife significantly
1.8
1.6
Odds Ratio for Successful Aging
1.4
1.2
0.8
<9.3 9.3 to 10.9 10.9 to 12.6 > 12.6
Total Diet Quality Score
Fig. 2.9 Association between total diet quality score on odds of successful aging (adapted from [27])
Western DP (p=.17) Healthy DP < Median Energy

Healthy DP > Median Energy Intake (p=.01)
Intake (p=.59)
1.5
1.4
Odds for Healthy Aging
1.3
1.2
1.1
0.9
0.8
1 2 3
Fig. 2.10 Association between healthy dietary patterns (DP), stratified by energy median intake level, compared to a
Western DP and the odds for healthy aging (multivariate adjusted; median energy intake 2500 kcal/day for men and
1820 kcal for women) (adapted from [52])
reduced multivariate risk of physical impairment overall healthy aging by 34–46% [54]. A 2017
by 13% (with increased fruits and vegetables French cohort (972 healthy subjects (336 men
being beneficial and increased sugar sweetened and 636 women); mean baseline age 73 years;
beverage and sodium being detrimental) [51] and 12 years of follow-up) found that a dietary pattern
significantly improved the multivariate odds for characterized as high biscuits and snacking
Healthy Pattern (p-trend =.009) Western Pattern (p-trend =.14)

1.8
1.6
1.4
Relative Risk of Frailty
1.2
1
0.8
0.6
0.4
0.2
0
1 2 3
Fig. 2.11 Association between dietary pattern adherence and frailty risk in adults ≥60 years after a 3.5 years of follow-
up (adapted from [53])
Western Pattern (p =.02) Healthy Pattern (p=.68)

1.3
1.2
Odds Ratio of "Idea Aging"
1.1
0.9
0.8
0.7
0.6
0.5
0.4
Tertile 1 Tertile 2 Tertile 3
Fig. 2.12 Effect of dietary pattern adherence and the odds of “ideal aging” in midlife adults after 13-year follow-up
(adapted from [55])
increased the risk of becoming frail by twofold in patterns and age-related cognitive performance
men and women and significantly increased the and Alzheimer’s disease (Table 2.5) [26, 56–68].
risk of muscle weakness in men [50].
2.2.4.1 Mediterranean Diet (MedDiet)
The protective effects of the MedDiet on age-
2.2.4 Age-Related Cognitive related cognitive decline and Alzheimer disease
Function and Dementia risk have been extensively investigated [26, 56–
59, 64, 67, 68]. Three meta-analyses concluded
Three meta-analyses, eight cohort studies, one that the high adherence to a MedDiet can signifi-
large cross-sectional study and two RCTs pro- cantly improve age-related cognitive perfor-
vide important insights on the effects of dietary mance and reduce Alzheimer’s disease risk
Table 2.5 Summary of dietary pattern studies on age-related cognitive performance and Alzheimer’s disease risk
Systematic reviews and meta-analyses
Wu and Sun (2017) 9 cohort studies; 34,168 Compared with the lowest category, the
Evaluate the association and participants; baseline pooled analysis showed that the highest
dose-response of the MedDiet on age ≥ 45 years (45–98 years); MedDiet scores were inversely
cognitive function [56]. 2–12 years of follow-up. associated with the developing of
cognitive disorders by 21%. Subgroup
analysis showed that the MedDiet was
inversely associated with mild cognitive
impairment by 17% and Alzheimer’s
disease by 40% (Fig. 2.13).
Singh et al. (2014) 6 cohorts; 8019 subjects, mean Subjects in the highest MedDiet tertile
Determine whether there is an baseline age mid-70s; cognitively had significantly lower adjusted risk of
association between the normal; follow-up of 4–8 years. cognitive impairment by 33% as
Mediterranean diet (MedDiet) compared to the lowest tertile. Among
and risk of cognitive impairment cognitively normal individuals, higher
(57). adherence to the MedDiet was associated
with significantly reduced risk of mild
cognitive impairment by 27% and
Alzheimer’s disease by 36%.
Sofi et al. (2010) 18 cohorts; 2,190,627 follow-up of A 2-point increase in adherence to the
Assess the effect of adherence to 4 to 20 years. MedDiet was associated with a
the MedDiet on significant reduction in
neurodegenerative diseases [26]. neurodegenerative diseases by 13%.
Prospective studies
Haring et al. (2016) 6425 postmenopausal women; There was no significant relationship
Determine the effect of dietary baseline age 65 to 79 years; between the adherence to a healthy
patterns on cognitive function in alternate MedDiet score, HEI- dietary pattern and cognitive function,
older women (Women’s Health 2010), the AHEI-2010, or DASH especially in women with hypertension.
Initiative Memory Study; diet score; median follow-up of
Germany) [58]. 9.1 years; 499 cases of mild
cognitive impairment and 390
probable dementia cases.
Galbete et al. (2015) 823 participants; mean age Although high MedDiet adherence was
Evaluate the effect of adherence 62 years; 6 to 8 years of follow-up. associated with a small decline
to the MedDiet on cognitive (−0.6 units) in cognitive function, there
function (Spanish) [59]. was a significantly slower rate of
cognitive decline compared to those with
low or moderate MedDiet adherence.
Jacka et al. (2015) 255 healthy adults; mean baseline Increased adherence to a healthy dietary
Examine the association between age 63 years; follow-up of 4 years. pattern was associated with a larger left
dietary patterns and hippocampal hippocampal volume, while higher
volume in humans (Personality adherence to an unhealthy Western
and Total Health Through Life dietary pattern was (independently)
project; Australian) [60]. associated with a smaller left
hippocampal volume (Fig. 2.14). These
relationships were independent of
covariates including age, gender,
education, labor-force status, depressive
symptoms and medication, physical
activity, smoking, hypertension and
diabetes.
(continued)
Shakersaina et al. (2015) 2223 healthy participants; mean Higher adherence to the Western dietary
Assess the association between baseline age 71 years; 39% men; pattern was associated with a higher
dietary patterns and cognitive 6-years of follow-up. mini-mental state examination
changes with aging (Swedish (cognitive) decline than the lowest
National adherence to this pattern. In contrast,
study on Aging and Care- higher adherence to a healthy dietary
Kungsholmen) [61]. pattern was inversely associated with
cognitive decline.
Tsai (2015) 1926 Chinese men and 1744 A Western dietary pattern was positively
Examine the effect of dietary Chinese women; baseline associated with an 8-fold increased risk
patterns on cognitive decline in age > 65 years; 8 years of of cognitive decline over 8 years
older Taiwanese (Taiwan follow-up. (adjusted), whereas traditional and
Longitudinal Study of Aging healthy dietary patterns were not. Diets
[62]. rich in meats and infrequent
consumption of fish, beans/legumes,
vegetables and fruits may adversely
affect cognitive function in older
Taiwanese.
Gardener (2015) 527 subjects; mean baseline age Of the diets, higher adherence to the
Investigate the association of 69 years; 40% male; 3 years of MedDiet was significantly associated
three well-recognised dietary follow-up. after 36 months with better performance
patterns with cognitive change in the executive function in
(Australian Imaging, Biomarkers apolipoprotein E (APOE) ε4 allele
and Lifestyle study of Ageing) carriers compared to higher Western diet
[63]. adherence, which was significantly
associated with greater cognitive decline.
Tangney et al. (2014) 826 participants; mean baseline age The subject mean global cognitive score
Estimate the effects of DASH 81.5 years; 26% men; 4.1 years of at baseline was 0.08 units. Adherence to
and MedDiet on age-related follow-up. the DASH and MedDiets significantly
cognitive function (US Memory slowed the rate of global cognitive
and Aging Project) [64]. decline by 0.002 to 0.007 units, which is
equivalent to between 1.3 and 4.4 years
of younger brain age.
Ozawa et al. (2013). 1006 healthy community-dwelling A dietary pattern with high intakes of
Investigate the effect of dietary Japanese subjects; baseline age soybeans and soybean products,
patterns on risk of dementia in 60–79 years; median follow-up of vegetables, seaweed, milk and dairy
older Japanese (Hisayama 15 years; 144 Alzheimer disease products and a low intake of rice was
Study) [65]. cases, 88 vascular dementia cases. associated with 35% lower risk of
Alzheimer’s disease and a 55% lower
risk of vascular dementia.
Cross-sectional study
Pearson et al. (2016) 18,080 black and white subjects; Dietary patterns including plant-based
Evaluate associations between baseline age ≥ 49 years. foods were associated with higher
empirically derived dietary cognitive scores and reduced risk of
patterns and cognitive function cognitive impairment, and a pattern
in the Southeast USA known as including fried food and processed meat
the stroke belt. (Reasons for typical of a southern diet was associated
Geographic and Racial with lower scores and significantly
Differences in Stroke increased risk of cognitive impairment
(REGARDS) cohort; US) [66]. (Fig. 2.15).
RCTs
Valls-Pedret et al. (2015) Parallel RCT: 447 cognitively The MedDiet supplemented with olive
healthy volunteers with high CVD
Investigate the effect of MedDiet oil or mixed nuts was associated with
on the decline of cognitive risk; mean baseline age 67 years; significantly improved cognitive
function in older adults 51% women; randomly assigned to function. Compared with the low-fat
(PREDIMED trial; Spain) [67]. a MedDiet supplemented with control diets, MedDiet + nuts improved
extra-virgin olive oil (1 L/week) or memory (p = 0.04) and MedDiet + olive
mixed nuts (30 g/day) vs. a control oil improved frontal and global
diet (advice to reduce dietary fat); cognition.
median of 4.1 years.
Martinez-Lapiscina et al. Parallel RCT: The MedDiet supplemented with olive
(2013) 522 participants at high vascular oil or mixed nuts significantly enhanced
Assess effects of MedDiets on risk; mean baseline age 75 years; cognitive function as measured by
cognitive function (PREDIMED- 45% men; randomly assigned to a mini-mental state exam and clock draw
NAVARRA; Spain) [68]. MedDiet supplemented with test vs. a low-fat diet.
extra-virgin olive oil (1 L/week) or
mixed nuts (30 g/day) vs. a control
diet (advice to reduce dietary fat);
6.5 years of follow-up.
MedDiet in Cognitive Performance and Dementia

0
Total Cognitive Mild Cognitive Alzheimer’s Disease
-5 Disorders (p = .0003) Impairment (p = .001) (p = .0001)
-10
-15
Risk (%)
-20
-25
-30
-35
-40
-45
Fig. 2.13 Association between high adherence to the Mediterranean diet (MedDiet) and risk of developing cognitive
disorders in adults (≥45 years) from a meta-analysis of 9 cohort studies (adapted from [56])
[56–58]. A 2017 meta-analysis (9 cohort studies; significantly reduced multivariate age-related

34,169 subjects; baseline age >45 years; cognitive decline by 27% and Alzheimer’s dis-
2–12 years of follow-up) found that high adher- ease risk by 36% [57]. A 2010 meta-analysis (18
ence to a MedDiet significantly lowered multi- cohorts; 2,190,627 subjects; 4 to 20 years of
variate age-related mild cognitive impairment by follow-up) demonstrated a 13% reduction in
17% and risk of Alzheimer’s disease by 40% neurodegenerative diseases for each 2-point
(Fig. 2.13) [56]. Also, a 2014 meta-analysis (6 improved adherence to the MedDiet [26]. The
cohorts; 8019 subjects; 4–8 years of follow-up) effect of the MedDiet on cognitive performance
showed that high adherence to the MedDiet appears to vary with the type of MedDiet
guidance and/or traditional dietary culture and 2.2.4.2 Western vs. Healthy Diets
food availability; as the German Women’s Health Six observational studies consistently support the
Initiative Memory Study (6425 postmenopausal adverse effects of Western dietary patterns and
women; baseline age 65–79 years; 9.1 years of the beneficial effects of healthy or traditional
follow-up) found no significant effect of the dietary patterns in protecting against age related
aMedDiet on slowing cognitive decline, espe- cognitive decline [60–63, 65, 66]. An Australian
cially in women with hypertension [58] com- longitudinal investigation of older adults (255
pared to Spanish prospective cohort (823 subjects; aged 60–64 years at baseline; 4 years of
subjects; mean baseline age 62 year, 6–8 years follow-up) showed that persons with a high
follow-up) which showed a high adherence to a adherence to unhealthy diets or moderate adher-
more traditional MedDiet and culture resulted in ence to healthy diets had significantly smaller left
a significantly slower rate of cognitive decline hippocampal volumes than those with high
compared to low or moderate adherence [59]. adherence to healthy diets (Fig. 2.14) [60].
The US Memory and Aging Project cohort of Another Australian study (527 subjects with apo-
elderly adults (826 participants; mean baseline lipoprotein E (APOE) ε4 allele; mean baseline
age 81.5 years; 4.1 years of follow-up) demon- age 69 years; 3 years of follow-up) suggested that
strated that high adherence to the DASH or high adherence to the MedDiet significantly
MedDiets significantly slowed the rate of cogni- improved cognitive performance compared to
tive decline by 1.3–4.4 years [64]. Two Spanish high adherence to the Western dietary pattern
PREDIMED trials in older adults (447–522 sub- [63]. Two cohort studies from Sweden and
jects with elevated CVD risk; mean baseline age Taiwan show that the high adherence to Western
67–75 years; median 4.1–6.5 years of follow-up) diets by older adults accelerates cognitive decline
showed that the MedDiet supplemented with a [61, 62]. The Japanese Hisayama Study (1006
liter per week of extra virgin olive oil or 30 g tree community-dwelling subjects; 60–79 years at
nuts/day significantly improved frontal and baseline; 15 years of follow-up) found that a
global cognitive and mini-mental state exam dietary pattern rich in soy, vegetables, seaweed,
measures compared to lower fat control diets and dairy products and low in white rice was
[67, 68]. associated with a significantly lower risk of
Baseline 4 Year Follow-up

3200
Left hippocampus volume (mm3)
3000
2800
2600
2400
2200
2000
Western Diet Healthy Diet Healthy Diet
(Moderate Adherence) (High Adherence)
Fig. 2.14 Association between dietary pattern and left hippocampus volume in adults aged 60–64 years at baseline
(p = 0.008 high healthy diet adherence vs. other diets) (adapted from [60])
Plant-based/Healthy DP (p-trend =.23) Southern/Unhealthy DP (p-trend =.05)

1.2
1.15
1.1
Odds of Cognitive Impairment
1.05
0.95
0.9
0.85
0.8
0.75
0.7
1 2 3 4 5
DP Adherence Quintiles
Fig. 2.15 Association between dietary pattern (DP) type and odds of cognitive impairment from the Reasons for
Geographic and Racial Differences in Stroke (REGARDS) cohort (adapted from [66])
Alzheimer’s disease by 35% and vascular demen- helping to maintain telomere length compared to
tia by 55% [65]. A large cross-sectional analysis the intake of saturated fat, refined flour cereals,
of the US REGARDS cohort (18,080 black and meat and meat products, and sugar-sweetened
white subjects; age ≥ 49 years) showed that high beverages which are associated with shorter telo-
adherence to a healthy plant based diet improved meres. Being overweight and obese appears to
cognitive function, and Southern diets rich in accelerate the telomere shortening process
fried foods and processed meats, lowered cogni- whereas weight loss or maintaining a healthy
tive function (Fig. 2.15) [66]. weight slows the telomere shortening process
[69–71]. Individuals from two Danish prospec-
tive cohort studies: The Copenhagen City Heart
2.2.5 T
elomere Length (Cellular Study and the Copenhagen General Population
Senescence) Study found that short telomeres in peripheral
blood leukocytes were associated with increased
2.2.5.1 Overview of Telomere Length mortality risk by 40% for individuals in the short-
and Aging est vs. the longest decile of telomere length [72].
The shortening of telomeres destabilizes the A meta-analysis (24 studies; 43,725 participants)
genome, leading to cell senescence, accelerated found a 54% increased CHD risk for those sub-
aging and potential increased chronic disease risk jects with the shortest versus the longest leuko-
[16]. Telomere length is independent of chrono- cyte telomere lengths in the general population,
logical age as it appears to be influenced by mod- independently of conventional vascular risk fac-
ifiable factors, such as lifestyle variables tors [73]. A meta-analysis of three prospective
including diet, adiposity, and physical exercise. studies found that shorter leukocyte telomere
There is some limited human evidence that diet length is independently associated with an
can influence telomere length with antioxidant increased pooled risk of diabetes incidence by
and fiber-rich, plant based diets and whole foods 31% [74].
2.2.5.2 Dietary Patterns and Leukocyte cellular senescence during mid-life and elderly
Telomere Length years [76–78]. A Spanish PREDIMED cohort
Nine studies, including one systematic review, (520 participants; mean baseline age 67 years;
four cross-sectional studies, two cohort studies 5 years of follow-up) observed that MedDiets
and two RCTs, on the effect of dietary patterns with low inflammatory index scores were associ-
on telomere length are summarized in Table 2.6 ated with longer telomere length compared to
[75–83]. The MedDiet was shown to protect Western type proinflammatory diets, which
against the shortening of telomere length in seven increased risk of telomere shorting by 80% [80].
of nine of these studies [77–82]. A systematic Two Spanish RCTs found that MedDiets are pro-
review (17 observational studies) concluded that tective of telomere length compared to lower fat
the MedDiet was most consistently associated and higher carbohydrate diets, especially in
with longer telomere length with fruits and veg- women, but more RCTs are needed to better
etables being protective and processed meats and understand the MedDiet effect on telomere length
sugar sweetened beverages having adverse effects [82, 83]. Also, two non-MedDiet cohort studies
[75]. Three cross-sectional studies all show that confirmed the protective effect of other healthy
higher adherence to the MedDiet is significantly dietary patterns on telomere length vs. the
protective of longer telomere length and slower Western diet [79, 81].
Table 2.6 Summary of studies on dietary patterns and telomere length (TL)
Systematic review
Rafie et al. (2017) 1 cohort, 3 case-control, and This review concluded that high MedDiet
Evaluate the effect of dietary 13 cross-sectional studies; 56 adherence was associated with longer
pattern and foods on TL [75]. to 5862 subjects. TL. Fruit and vegetable intake was more
positively associated with TL than other
dietary factors. Also, processed meat and
sugar-sweetened beverages appear to be
associated with shorter TL.
Cross-sectional studies
Gu et al. (2015) 1743 multi-ethnic In elderly subjects, higher adherence to a
Examine the relation between community residents of MedDiet was associated with longer
MedDiet and leukocyte TL (The New York; mean age leukocyte TL among whites compared to
Washington Heights-Inwood 78 years; 68% female. blacks and Hispanics.
Community Aging
Project study; US) [76].
Crous-Bou et al. (2014) 4676 healthy middle aged Greater adherence to the MedDiet was
Examine whether adherence to the and older women; mean age associated with longer telomeres after
MedDiet was associated with 59 years. adjustment for potential confounders. Least
longer TL, a biomarker of aging squares mean TL z scores were −0.038 for
(Nurses’ Health Study; US) [77]. the lowest MedDiet score groups and 0.072
for the highest score group (p-trend = 0.004).
Boccardi et al. (2013) 217 elderly subjects; mean The high adherence MedDiet group was
Assess the effect of adherence to 78 years; stratified according associated with significantly longer
the MedDiet on TL and telomerase to MedDiet score in low leukocyte TL, and higher telomerase activity
activity (Sothern Italy) [78]. adherence, medium compared to lower adherence.
adherence and high
adherence groups.
Sun et al. (2012). 5862 women; mean age High adherence to a healthy lifestyle, such as
Examine healthy lifestyle practices 57 years. healthy diet and physical activity, was
on leukocyte TL (US Nurses’ associated with longer TL in leukocytes.
Health Study) [79].
Prospective studies
Garcia-Calzon et al. (2015) 520 participants at high CVD MedDiets with low dietary inflammatory
Examine if inflammation is risk; mean baseline age index scores were inversely associated with
associated with telomere attrition 67 years; 45% males; 5 years leukocyte TL in older adults at high risk of
rate (PREDIMED- of follow-up. CVD. Less healthy high proinflammatory
NAVARRA; Spain) [80]. diets were associated with an 80% higher
risk of shorter TL.
Lee et al. (2015). 1958 subjects; baseline age Healthy dietary patterns including whole
Determine the association between 40 to >60 years; 10 years of grains, seafood, legumes, vegetables and
dietary patterns or consumption of follow-up. seaweed were associated with longer
specific foods and leukocyte TL in leukocyte TL. In contrast, Western dietary
adults (Korean Genome patterns, especially rich in red meat or
Epidemiology Study) [81]. processed meat and sweetened carbonated
beverages, were associated with shorter TL.
RCTs
Garcia-Calzon et al. (2016). Parallel RCT: Better adherence to MedDiets was associated
Assess the effect of MedDiets on 520 participants; mean age with significantly longer basal telomeres in
TL in older adults at increased risk 67 years; 55% women; women but not in men.
of CVD (Spain- PREDIMED- randomly assigned to 3 diets:
NAVARRA trial) [82]. a low-fat control or one of 2
MedDiets, one supplemented
with extra virgin olive oil
and the other with mixed
nuts; 5 years.
Marin et al. (2012). Crossover RCT MedDiet significantly protected against
Evaluate the effect of dietary 20 subjects; ≥ 65 years; 10 endothelial cell senescence as shown by
patterns on cellular senescence males and 10 females; lower intracellular oxidative stress, less
(Spain) [83]. saturated fatty acid diet, a shortening of telomeres, and lower apoptosis
lower-fat diet, higher- compared to lower-fat and higher-
carbohydrate diets, and carbohydrate diets.
MedDiet; each for 4 weeks.
Conclusions score. Healthy dietary patterns include ade-

In the US and worldwide, high rates of quate whole- grain cereals, legumes, fruits,
Western diets and sedentary lifestyle habits, vegetables, and nuts, and lower intakes of red
especially since the 1980s have led to pan- and processed meats and sugar- sweetened
demic rates of obesity, nonalcoholic fatty beverages. Prospective cohort studies with
liver disease, metabolic syndrome, type-2 middle and older age adults consistently show
diabetes, chronic kidney disease, and other that high adherence to Western dietary pat-
related health conditions associated with terns increases CVD, diabetes and cancer,
unhealthy aging. Dietary pattern habits have and premature death risks whereas high
a significant effect on aging, chronic disease adherence to healthy/plant-based dietary pat-
risk, and longevity but in developed countries terns lower CVD, diabetes and cancer risk,
like the US mean diet quality scores are only and premature mortality risk. Among cancer
about half of the optimal healthy eating diet survivors, healthy diets are associated with
reduced overall mortality risk by 20% and associated with reduced age-related decline
Western diets are associated with increased in cognitive performance and Alzheimer’s
risk by over 40%. For older adults, higher disease, and longer telomere length compared
adherence to healthy dietary patterns reduces to Western diets. High adherence to healthy
the odds of unhealthy aging and frailty dietary patterns support healthy aging and
whereas higher adherence to Western diets reduce chronic disease and mortality risk
increases the odds. Simple dietary rules for through beneficial effects on body weight
healthy aging are to limit the intake of energy regulation and adiposity metabolism, lipo-
dense, low nutrient quality foods, red and protein concentrations and function, blood
processed meats and added salt and sugar, pressure, glucose-insulin homeostasis, oxida-
and increase the intake of fiber, and nutrient tive stress, inflammation, endothelial health,
and phytochemical-rich whole or minimally hepatic function, cardiac and cognitive func-
processed plant foods. Healthy dietary pat- tion, telomere length, and the microbiota
terns, especially the Mediterranean diet, are ecosystem.
ppendix A: Comparison of Western and Healthy Dietary Patterns

A
per 2000 kcal (Approximated Values)
Healthy
Healthy vegetarian
Western dietary USDA Base DASH diet Mediterranean pattern (Lact-ovo
Components pattern (US) pattern pattern pattern based) Vegan pattern
Emphasizes Refined Vegetables, Potassium Whole Vegetables, Plant foods:
grains, low fruit, rich grains, fruit, whole- Vegetables,
fiber foods, whole- vegetables, vegetables, grains, fruits,
red meats, grain, and fruits, and fruit, dairy legumes, nuts, whole
sweets and low-fat milk low fat products, seeds, milk grains, nuts,
solid fats milk olive oil, and products, and seeds, and
products moderate soy foods soy foods
wine
Includes Processed Enriched Whole- Fish, nuts, Eggs, Non-dairy
meats, sugar grains, lean grain, seeds, and non-dairy milk milk
sweetened meat, fish, poultry, pulses alternatives, alternatives
beverages, nuts, seeds, fish, nuts, and vegetable
and fast foods and and seeds oils
vegetable
oils
Limits Fruits and Solid fats Red meats, Red meats, No red or No animal
vegetables, and and added sweets, and refined white meats, or products
whole-grains sugars sugar- grains, and fish; limited
sweetened sweets sweets
beverages
Estimated nutrients/components
Carbohydrates (% 51 51 55 50 54 57
Total kcal)
Protein (% Total 16 17 18 16 14 13
kcal)
Total fat (% Total 33 32 27 34 32 30
kcal)
(continued)
References 55
Healthy
Healthy vegetarian
Western dietary USDA Base DASH diet Mediterranean pattern (Lact-ovo
Saturated fat (% 11 8 6 8 8 7
Total kcal)
Unsat. Fat (% Total 22 25 21 24 26 25
kcal)
Fiber (g) 16 31 29+ 31 35+ 40+
Potassium (mg) 2800 3350 4400 3350 3300 3650
Vegetable oils (g) 19 27 25 27 19–27 18–27
Sodium (mg) 3600 1790 1100 1690 1400 1225
Added sugar (g) 79 (20 tsp.) 32 (8 tsp.) 12 (3 tsp.) 32 (8 tsp.) 32 (8 tsp.) 32 (8 tsp.)
Plant food groups
Fruit (cup) ≤ 1.0 2.0 2.5 2.5 2.0 2.0
Vegetables (cup) ≤ 1.5 2.5 2.1 2.5 2.5 2.5
Whole-grains (oz.) 0.6 3.0 4.0 3.0 3.0 3.0
Legumes (oz.) − 1.5 0.5 1.5 3.0 3.0+
Nuts/seeds (oz.) 0.5 0.6 1.0 0.6 1.0 2.0
Soy products (oz.) 0.0 0.5 − − 1.1 1.5
U.S. Department of Agriculture, Agriculture Research Service, Nutrient Data Laboratory. 2014. USDA National
Nutrient Database for Standard Reference, Release 27. https://www.ars.usda.gov/nutrientdata. Accessed Feb 17, 2015
Dietary Guidelines Advisory Committee. Scientific Report. Advisory Report to the Secretary of Health and Human
Services and the Secretary of Agriculture. Appendix E-3.7: Developing vegetarian and Mediterranean-style food pat-
terns. 2015;1–9
U.S. Department of Agriculture and U.S. Department of Health and Human Services. Dietary Guidelines for Americans,
2010. 7th Edition, Washington, DC: U.S. Government Printing Office. 2010; Table B2.4; http://www.choosemyplate.
gov/. Accessed Aug 22, 2015
References Part D. Chapter 1: Food and nutrient intakes, and

health: current status and trends. 2015. p. 1–78.
1. Shlisky J, Bloom DE, Beaudreault AR, et al. 8. Dietary Guidelines Advisory Committee (DGAC).
Nutritional considerations for healthy aging and Scientific report. Advisory report to the Secretary
reduction in age-related chronic disease. Adv Nutr. of Health and Human Services and the Secretary
2017;8:17–26. https://doi.org/10.3945/n116.3474. of Agriculture. Part D. Chapter 2: Dietary patterns,
2. Milte CM, McNaughton SA. Dietary patterns and foods and nutrients and health outcomes. 2015.
successful ageing: a systematic review. Eur J Nutr. p. 1–35.
2016;55:423–50. https://doi.org/10.1007/s00394-015- 9. World Health Organization. WHO global status report
1123-7. on noncommunicable diseases, vol. 2010. Geneva:
3. Rowe JW, Kahn RL. Human aging: usual and suc- World Health Organization Press; 2010.
cessful. Science. 1987;237:143–9. 10. WHO/FAO. Joint Expert Consultation on Diet.

4. Beers MH, et al. The Merck manual of health & aging. Nutrition and the prevention of chronic diseases.
Whitehouse Station: Merck Laboratories; 2004. p. 2. WHO Technical Series. 2003. p. 916.
5. Fontana L, Optimal HFB. Body weight for health and 11. http://health.gov/dietaryguidelines/2015/guidelines/
longevity: bridging basic, clinical, and population message. Accessed January 17, 2016.
research. Aging Cell. 2014;13:391–400. 12. Avendano M, Kawachi I. Why do Americans have
6. Péter S, Saris WHM, Mathers JC, et al. Nutrient status shorter life expectancy and worse health than people
assessment in individuals and populations for healthy in other high-income countries? Annual Rev Public
aging - statement from an expert workshop. Forum Health. 2014;35:307–25.
Nutr. 2015;7:10491–500. 13. Millen BE, Abrams S, Adams-Campbell L, et al.

7. Dietary Guidelines Advisory Committee. Scientific The 2015 Dietary Guidelines Advisory Committee
report. Advisory Report to the Secretary of Health Scientific Report: development and major conclu-
and Human Services and the Secretary of Agriculture. sions. Adv Nutr. 2016;7:438–44.
14. Mozaffarian D. Dietary and policy priorities for car- 27. Gopinath B, Russell J, Kifley A, et al. Adherence to
diovascular disease, diabetes, and obesity a compre- dietary guidelines and successful aging over 10 years.
hensive review. Circulation. 2017;133(2):187–225. J Gerontol A Biol Sci Med Sci. 2016;71(3):349–55.
https://doi.org/10.1161/CIRCULATIONAHA. 28. Yang M, Kenfield SA, Van Blarigan EL, et al. Dietary
11519.018585. patterns after prostate cancer diagnosis in relation to
15. Van Horn L, Carson JA, Appel LJ, et al. Recommended disease-specific and total mortality. Cancer Prev Res
dietary pattern to achieve adherence to the American (Phila). 2015;8(6):545–51.
Heart Association/American College of Cardiology 29. Harmon BE, Boushey CJ, Shvetsov YB, et al.

Guidelines: A Scientific Statement from the American Associations of key diet-quality indexes with mor-
Heart Association. Circulation. 2016;134(22): tality in the Multiethnic Cohort: the Dietary Patterns
505–29. Methods Project. Am J Clin Nutr. 2015;101:587–97.
16. Freitas-Simoes TMF, Ros E, Sala-Vila A. Nutrients, 30. Reedy J, Krebs-Smith SM, Miller PE, et al. Higher
foods, dietary patterns and telomere length: update diet quality is associated with decreased risk of all-
of epidemiological studies and randomized trials. cause, cardiovascular disease, and cancer mortality
Metabolism. 2016;65(4):406–15. among older adults. J Nutr. 2014;144:881–9.
17. Schwingshackl L, Schwedhelm C, Hoffmann G, et al. 31. Liese AD, Krebs-Smith SM, Subar AF, et al. The
Food groups and risk of all-cause mortality: a system- dietary patterns methods project: synthesis of findings
atic review and meta-analysis of prospective studies. across cohorts and relevance to dietary guidance. J
Am J Clin Nutr. 2017;105:1462–73. Nutr. 2015;145:393–402.
18. Onvani S, Haghighatdoost F, Surkan PJ, et al.
32. Shi Z, Zhang T, Byles J, et al. Food habits, lifestyle
Adherence to the Healthy Eating Index and factors and mortality among oldest old Chinese: The
Alternative Healthy Eating Index patterns and mortal- Chinese Longitudinal Healthy Longevity Survey
ity from all-causes, cardiovascular disease and cancer: (CLHLS). Forum Nutr. 2015;7:7562–79.
a meta-analysis of observational studies. J Hum Nutr 33. George SM, Ballard-Barbash R, Manson JE, et al.
Diet. 2017;30(2):216–26. Comparing indices of diet quality with chronic dis-
19. Sotos-Pieto M, Bhupathiraju SN, Mattei J, et al.
ease mortality risk in postmenopausal women in
Association of changes in diet quality with total the Women’s Health Initiative Observational Study:
and cause-specific mortality. N Engl J Med. 2017; evidence to inform national dietary guidance. Am J
377(2):143–53. Epidemiol. 2014;180(6):616–25.
20. Schwedhelm C, Boeing H, Hoffmann G, et al. Effect 34. Zazpe I, Sánchez-Tainta A, Toledo E, et al. Dietary
of diet on mortality and cancer recurrence among can- patterns and total mortality in a Mediterranean
cer survivors: a systematic review and meta-analysis cohort: The SUN Project. J Acad Nutr Diet. 2014;114:
of cohort studies. Nutr Rev. 2016;74(12):737–48. 37–47.
21. Li F, Hou L, Chen W, et al. Associations of dietary 35. Yu D, Zhang X, Xiang Y-B, et al. Adherence to dietary
patterns with the risk of all-cause, CVD and stroke guidelines and mortality: a report from prospective
mortality: a meta-analysis of prospective cohort stud- cohort studies of 134,000 Chinese adults in urban
ies. Br J Nutr. 2015;113:16–24. Shanghai. Am J Clin Nutr. 2014;100:693–700.
22. Schwingshackl L, Hoffmann G. Diet quality as
36. Martınez-Gonzalez MA, Sanchez-Tainta A, Corella
assessed by the healthy eating index, the alternate D, et al. A provegetarian food pattern and reduc-
healthy eating index, the dietary approaches to stop tion in total mortality in the Prevencion con Dieta
hypertension score, and health outcomes: a systematic Mediterranea (PREDIMED) study. Am J Clin Nutr.
review and meta-analysis of cohort studies. J Acad 2014;100(suppl):320S–8S.
Nutr Diet. 2015;115:780–800. 37. Orlich MJ, Singh PN, Sabate J, et al. Vegetarian

23. Jankovic N, Geelen A, Streppel MT, et al. WHO
dietary patterns and mortality in the Adventist Health
guidelines for a healthy diet and mortality from Study 2. JAMA Intern Med. 2013;173(13):1230–8.
cardiovascular disease in European and American 38. Akbaraly TN, Ferrie JE, Berr C, et al. Alternative
elderly: the CHANCES project. Am J Clin Nutr. healthy eating index and mortality over 18 y of fol-
2015;102:745–56. low-up: results from the Whitehall II cohort. Am J
24. Huang T, Yang B, Zheng J, et al. Cardiovascular dis- Clin Nutr. 2011;94:247–53.
ease mortality and cancer incidence in vegetarians: 39. Anderson AL, Harris TB, Tylavsky FA, et al. Dietary
a meta-analysis and systematic review. Ann Nutr patterns and survival of older adults. J Am Diet Assoc.
Metab. 2012;60:233–40. 2011;111(1):84–91.
25. Jankovic N, Geelen A, Streppel MT, et al. Adherence 40. Buckland G, Agudo A, Travier N, et al. Adherence
to a healthy diet according to the World Health to the Mediterranean diet reduces mortality in
Organization guidelines and all-cause mortality in the Spanish cohort of the European Prospective
elderly adults from Europe and the United States. Am Investigation into Cancer and Nutrition (EPIC-Spain).
J Epidemiol. 2014;180(10):978–88. Br J Nutr. 2011;106:1581–91.
26. Sofi F, Abbate R, Gensini GF, Cesari F. Accruing evi- 41. Heidemann C, Schulze MB, Franco OH, et al. Dietary
dence on benefits of adherence to the Mediterranean patterns and risk of mortality from cardiovascular dis-
diet on health: an updated systematic review and ease, cancer, and all causes in a prospective cohort of
meta-analysis. Am J Clin Nutr. 2010;92:1189–96. women. Circulation. 2008;118:230–7.
References 57
42. Bamia C, Trichopoulos D, Ferrari P, et al. Dietary 55. Akbaraly TN, Sabia S, Hagger-Johnson G, et al. Does
patterns and survival of older Europeans: The EPIC- overall diet in midlife predict future aging pheno-
Elderly Study (European Prospective Investigation types? A cohort study. Am J Med. 2013;126:411–9.
into Cancer and Nutrition). Public Health Nutr. 56. Wu L, Sun D. Adherence to Mediterranean diet and
2007;10(6):590–8. risk of developing cognitive disorders: an updated
43. Ndanuko RN, Tapsell LC, Charlton KE, et al. Dietary systematic review and meta-analysis of prospective
patterns and blood pressure in adults: a system- cohort studies. Sci Rep. 2017;7:41317. https://doi.
atic review and meta-analysis of randomized con- org/10.1038/srep41317.
trolled trials. Adv Nutr. 2016;7:76–89. https://doi. 57. Singh B, Parsaik AK, Mielke MM, et al. Association
org/10.3945/an.115.009753. of Mediterranean diet with mild cognitive impair-
44. Feng YL, Shu L, Zheng PF, et al. Dietary patterns and ment and Alzheimer’s disease: a systematic review
colorectal cancer risk: a meta-analysis. Eur J Cancer and meta-analysis. J Alzheimers Dis. 2014;39(2):
Prev. 2017;26(3):201–11. https://doi.org/10.1097/ 271–82.
CEJ.0000000000000245. 58. Haring B, Wu C, Mossavar-Rahmani Y, et al. No asso-
45. Maghsoudi Z, Ghiasvand R, Solehi-Abargouei A.
ciation between dietary patterns and risk for cognitive
Empirically derived dietary patterns and incident type decline in older women with 9-year follow-up: data
2 diabetes mellitus: a systematic review and meta- from the Women’s Health Initiative Memory Study. J
analysis on prospective observational studies. Public Acad Nutr Diet. 2016;116:921–30.
Health Nutr. 2015;19(2):230–41. 59. Galbete C, Toledo E, Toledo JB, et al. Mediterranean
46. Rodríguez-Monforte M, Flores-Mateo G, Sánchez
diet and cognitive function: the SUN project. J Nutr
E. Dietary patterns and CVD: a systematic review Health Age. 2015;19(3):305–12.
and meta-analysis of observational studies. Br J 60. Jacka FN, Cherbuin N, Anstey KJ, et al. Western diet
Nutr. 2015;114(9):1341–59. https://doi.org/10.1017/ is associated with a smaller hippocampus: a longitu-
S0007114515003177. dinal investigation. BMC Med. 2015;13:215. https://
47. Hou L, Li F, Wang Y, Ou Z. Association. Between doi.org/10.1186/s12916-015-0461-x.
dietary patterns and coronary heart disease: a meta- 61. Shakersaina B, Santonia G, Larssonb SC, et al.
analysis of prospective cohort studies. Int J Clin Exp Prudent diet may attenuate the adverse effects of
Med. 2015;8(1):781–90. western diet on cognitive decline. Alzheimer’s.
48. Chiuve SE, Fung TT, Rimm EB, et al. Alternative Dementia. 2016;12(2):100–9. https://doi.
dietary indices both strongly predict risk of chronic org/10.1016/j.jalz.2015.08.002.
disease. J Nutr. 2012;142:1009–18. 62. Tsai HJ. Dietary patterns and cognition decline in
49. Shikany JM, Safford MM, Newby P, et al. Southern Taiwanese aged 65 years and older. Int J Geriatr
dietary pattern is associated with hazard of acute Psychiatry. 2015;30(5):523–30.
coronary heart disease in the Reasons for Geographic 63. Gardener SL, Rainey-Smith SR, Barnes MB, et al.
and Racial Differences in Stroke (REGARDS) Dietary patterns and cognitive decline in an Australian
Study. Circulation. 2015;132:804–14. https://doi. study of ageing. Mol Psychiatry. 2015;20(7):860–6.
org/10.1161/CIRCULATIONAHA. 114.014421. 64. Tangney CC, Li H, Wang Y, et al. Relation of DASH-
50. Pilleron S, Ajana S, Juland MA, et al. Dietary pat- and Mediterranean-like dietary patterns to cognitive
terns and 12-year risk of frailty: results from the decline in older persons. Neurology. 2014;83:1410–6.
three-city Bordeaux study. J Am Med Dir Assoc. 65. Ozawa M, Ninomiya T, Ohara T, et al. Dietary pat-
2017;18(2):169–75. https://doi.org/10.1016/j. terns and risk of dementia in an elderly Japanese
jamda.2016.09.014. population: The Hisayama Study. Am J Clin Nutr.
51. Hagan K, Chiuve SE, Stampfer MJ, et al. Greater 2013;97:1076–82.
adherence to the Alternative Healthy Eating Index 66. Pearson KE, Wadley VG, McClure LA, et al. Dietary
is associated with lower incidence of physical func- patterns are associated with cognitive function in the
tion impairment in the Nurses’ Health Study. J REasons for Geographic And Racial Differences in
Nutr. 2016;146(7):1341–7. https://doi.org/10.3945/ Stroke (REGARDS) cohort. J Nutr Sci. 2016;5:e38.
jn.115.227900. https://doi.org/10.1017/jns.2016.27.
52. Assmann KE, Lassale C, Andreeva VA, et al. A
67. Valls-Pedret C, Sala-Vila A, Serra-Mir M, et al.
healthy dietary pattern at midlife, combined with a Mediterranean diet and age-related cognitive decline.
regulated energy intake, is related to increased odds A randomized clinical trial. JAMA Intern Med.
for healthy aging. J Nutr. 2015;145(9):2139–1245. 2015;175(7):1094–103.
https://doi.org/10.3945/jn.115.210740. 68.
Martinez-Lapiscina EH, Clavero P, Toledo R,
53.
León-Muñoz LM, García-Esquinas E, López- et al. Mediterranean diet improves cognition: the
García E, et al. Major dietary patterns and risk of PREDIMED-NAVARRA randomized trial. J Neurol
frailty in older adults: a prospective cohort study. Neurosurg Psychiatry. 2013;84(12):1318–25.
BMC Med. 2015;13:11. https://doi.org/10.1186/ 69. Garcia-Calzon S, Gea A, Razquin C, et al. Longitudinal
s12916-014-0255-6. association of telomere length and obesity indices in
54. Samieri C, Sun Q, Townsend MK, et al. The relation an intervention study with a Mediterranean diet: the
of midlife diet to healthy aging: a cohort study. Ann PREDIMED-NAVARRA trial. Int J Obes (London).
Intern Med. 2013;159(9):584–91. 2014;38:177–82.
70. Garcia-Calzon S, Moleres A, Marcos A, et al.

77.
Crous-Bou M, Fung TT, Prescott J, et al.
Telomere length as a biomarker for adiposity changes Mediterranean diet and telomere length in Nurses’
after a multidisciplinary intervention in overweight Health Study: population based cohort study. BMJ.
adolescents: The EVASYON Study. PLoS One. 2014;349:g6674. https://doi.org/10.1136/bmj.g6674.
2014;9(2):e89828. https://doi.org/10.1371/journal. 78.
Boccardi V, Esposito A, Rizzo MR, et al.
pone.0089828. Mediterranean diet, telomere maintenance and health
71. Kim S, Parks CG, DeRoo LA, et al. Obesity and status among elderly. PLoS One. 2013;8(4):e62781.
weight gain in adulthood and telomere length. Cancer https://doi.org/10.1371/journal.pone.0062781.
Epidemiol Biomark Prev. 2009;18(3):816–25. 79. Sun Q, Shi L, Prescott J, et al. Healthy lifestyle and
72. Rode L, Nordestgaard BG, Bojesen SE. Peripheral leukocyte telomere length in U.S. women. PLoS One.
blood leukocyte telomere length and mortality among 2012;7(5):e38374. https://doi.org/10.1371/journal.
64,637 individuals in the general population. J Natl pone.0038374.
Cancer Inst. 2015;107(6). https://doi.org/10.1093/ 80. García-Calzón S, Zalba G, Ruiz-Canela M, et al.

jnci/djv074. Dietary inflammatory index and telomere length
73. Haycock PC, Heydon EE, Kaptoge S, et al. Leukocyte in subjects with a high cardiovascular disease risk
telomere length and risk of cardiovascular dis- from the PREDIMED-Navarra study: cross-sectional
ease: systematic review and meta-analysis. BMJ. and longitudinal analyses over 5 y. Am J Clin Nutr.
2014;349:g4227. https://doi.org/10.1136/bmj.g4227. 2015;102:897–904.
74. Willeit P, Raschenberger J, Heydon EE, et al.
81. Lee J-Y, Jun N-R, Yoon D, et al. Association between
Leukocyte telomere length and risk of type 2 diabetes dietary patterns in the remote past and telomere
mellitus: new prospective cohort study and literature- length. Eur J Clin Nutr. 2015;69(9):1048–52. https://
based meta-analysis. PLoS One. 2014;9(11):e112483. doi.org/10.1038/ejcn.2015.58.
https://doi.org/10.1371/journal.pone.0112483. 82. Garcia-Calzon S, Martinez-Gonzalez MA, Razquin C,
75. Rafie N, Hamedani SG, Barak F, et al. Dietary patterns, et al. Mediterranean diet and telomere length in high
food groups and telomere length: a systematic review cardiovascular risk subjects from the PREDIMED-
of current studies. Eur J Clin Nutr. 2017;71:151–8. NAVARRA study. Clin Nutr. 2016;35(6):1399–405.
https://doi.org/10.1038/ejcn.2016.149. https://doi.org/10.1016/j.clnu.2016.03.013.
76. Gu Y, Honig LS, Schupf N, et al. Mediterranean
83. Marin C, Delgado-Lista J, Ramirez R, et al.

diet and leukocyte telomere length in a multi-ethnic Mediterranean diet reduces senescence-associated
elderly population. Age. 2015;37:24. https://doi. stress in endothelial cells. Age. 2012;34:1309–16.
org/10.1007/s11357-015-9758-0. https://doi.org/10.1007/s11357-011-9305-6.
Whole Plant Foods in Aging
and Disease 3
Keywords
Whole-plant foods • Healthy aging • Mortality risk • Cardiovascular dis-
ease • Cancer • Type 2 diabetes • Cognitive performance • Lipoproteins •
Blood pressure • Cancer • Stroke • Telomeres • Whole-grains • Fruits •
Vegetables • Legumes • Soybeans • Nuts
Key Points lipoprotein concentrations and function, blood

pressure, glucose-insulin homeostasis, oxida-
• The rate and quality of the aging processes can tive stress, inflammation, endothelial health,
be modified by consuming healthy diets overall hepatic function, adipocyte metabolism,
and specific types of uniquely healthy foods. visceral adiposity, brain neurochemistry and
• Healthy dietary guidelines generally recom- the microbiota ecosystem.
mend eating: 2 1/2 cups of a variety of vegeta- • For whole-grains, β-glucan-rich oats and bar-
bles/day; 2 cups of fruits, especially whole ley lower total and LDL-cholesterol better
fruits/day; 6 servings of total grains at ≥3 than other cereal grains and whole-grain bread
servings of whole grains/day and ≤3 servings tends to be more beneficial than white bread in
of refined grains/day, ≥4 weekly servings of controlling weight gain and abdominal fat.
legumes (dietary pulses or soy), and/or ≥5 • For fruits and non-starchy vegetables, low energy
weekly servings of nuts, and limiting con- dense and flavonoid and/or carotenoid rich vari-
sumption of red or processed meats, added eties including apples, pears, berries, citrus fruits,
saturated and trans-fat, sugar or sodium for cruciferous vegetables, and green leafy vegeta-
improved odds for healthy aging and reduced bles are especially associated with improved
chronic disease and premature mortality risk. odds of healthy aging, cognitive performance
• Whole plant foods range widely in their health and weight control, and reduced risk of chronic
effects because of their variation in level and disease and premature death. Legumes (dietary
type of fiber, nutrients and phytochemicals, pulses or soy) are associated with reduced weight
which can have differential effects on aging, gain, chronic disease and mortality risks.
chronic disease risk, cognitive function and • All nuts tend to have similar effects on manag-
longevity by their impact on weight regulation, ing body weight, and glycemic, lipoprotein

60 3 Whole Plant Foods in Aging and Disease
and inflammatory profiles, but among nuts healthy whole plant food diets both directly add
walnuts appear to be uniquely effective in pro- healthy nutrients and phytochemicals to the diet
moting better vascular endothelial function and also displace a portion of high energy, low
such as flow mediated dilation, which helps to fiber, high added sugar, sodium, and saturated and
reduce the rate of vascular aging. trans-fat rich-foods typical of the Western diet.
Adherence to this type of healthy dietary pattern
would significantly improve the odds of healthy
3.1 Introduction aging and reduce risk of weight gain, obesity,
chronic diseases, and premature mortality com-
The global adoption of the Western lifestyle, pared to Western dietary patterns [11, 12]. The
especially since the 1980s, has resulted in a pan- nutrient and phytochemical compositions of whole
demic of sub-optimal aging with increased rates plant foods are summarized in Appendix A. Whole
of obesity, metabolic syndrome, type 2 diabetes plant foods tend to be lower in energy density and
and other related health conditions in both adults higher in nutrient quality than more highly refined
and children [1–7]. US adults have shorter and foods; they are lower in saturated and trans-fatty
less healthy lives than populations in 32 other acids, sodium, and added sugars and richer in
high-income countries [8]. As a result, popula- essential nutrients and phytochemicals like potas-
tion longevity forecasts suggest a stagnation or sium and antioxidants such as vitamin C and E,
decline in life expectancy over the next 25 years carotenoids, and polyphenols [13–17]. Whole
[4–8]. The aging process rate and quality is only plant foods influence a range of healthy biological
partly genetically pre-determined as it can be mechanisms including the improvement of weight
modified by lifestyle habits including dietary pat- regulation, lipoprotein concentrations and func-
terns, level of activity or exercise, personal habits tion, blood pressure, glucose-insulin homeostasis,
or risk-taking, and psychosocial and stress man- oxidative stress, systemic inflammation, endothe-
agement factors [9]. The habitual consumption of lial health, hepatic function, adipocyte metabo-
recommended levels of whole and minimally lism, visceral adiposity, brain neurochemistry and
processed plant foods (whole plant foods) and the microbiota ecosystem, which can help to atten-
avoiding excess energy intake are among the uate the risk of obesity, cardiovascular diseases
most important lifestyle factors associated with (CVD), metabolic syndrome, type 2 diabetes (dia-
healthy aging and reduced chronic disease and betes), certain cancers and cognitive dysfunction
premature mortality risk [9–26] but only a small [18–21]. However, there are a spectrum of whole
fraction (<10%) of western populations consume plant foods that range from healthy to less healthy
adequate whole foods [10, 11]. According to the options which is illustrated by the following
2015 US Dietary Guidelines, ≥3/4 of the popula- example; a pooled analysis of three US prospec-
tion consume very low levels of whole-grains, tive cohort studies (69,949 women from the
vegetables and fruits, legumes and nuts [11]. Nurses’ Health Study, 90,239 women from the
Healthy dietary consumption guidelines, such Nurses’ Health Study 2, and 40,539 men from the
as the 2015 US dietary guidelines, primarily focus Health Professionals Follow-Up Study) showed
on increasing intake of whole plant foods with that the consumption of a plant-based (vegetarian)
general recommendations for eating: 2 1/2 cups of diet that emphasized specifically healthy whole
a variety of vegetables/day; 2 cups of fruits, espe- plant foods was associated with a decrease in dia-
cially whole fruits/day; 6 servings of total grains at betes risk by 34%, while consumption of a plant-
≥3 servings of whole grains/day and ≤3 servings based diet high in less healthy processed and
of refined grains/day; alternatives to animal pro- refined plant foods was associated with a 16%
tein including legumes (beans and peas), and nuts, increased diabetes risk (Fig. 3.1) [21]. The objec-
seeds, and soy products including ≥4 weekly serv- tive of this chapter is to comprehensively assess
ings of legumes (dietary pulses or soy), and/or ≥5 the effects of whole (minimally processed) plant
weekly servings of nuts, along with limited con- foods in general and specific uniquely effective
sumption of red or processed meats, added satu- options on promoting healthy aging and reducing
rated/trans-fat, sugar or sodium [11]. These the risk of chronic disease and premature deaths.
3.2 Whole Plant Foods in Aging and Disease 61
Overall Plant-based Diet Index Healthy Plant-based Diet Index

Unhealthy Plant-based Diet Index
1.2
1.1
Hazard Ratios for Type 2 Diabetes
0.9
0.8
0.7
0.6
0.5
0.4
1 2 3 4 5 6 7 8 9 10
Deciles of Plant-based Dietary Indices*
Fig. 3.1 Association between plant-based diet indices and type 2 diabetes risk from the pooled analysis of Nurses’
Health Studies and Health Professionals Follow-Up Study with approx. 20 years of follow-up; baseline age 35–55 years
(p < 0.001 for all) (adapted from [21]).
*Plant-based diet index, where healthy plant foods (whole grains, fruits, vegetables, nuts, legumes, vegetable oils, tea/
coffee) received positive scores, while less healthy plant foods (fruit juices, sweetened beverages, refined grains, pota-
toes, sweets/desserts) and animal foods received reverse scores
3.2 hole Plant Foods in Aging

W 3.2.1 Whole-Grains
and Disease
3.2.1.1 Background
The US National Center for Health Statistics esti- Whole-grains consist of the intact, ground,
mates that about 45% of cardiometabolic deaths cracked or flaked kernel after removal of the
are associated with suboptimal intake of ten inedible parts such as the husk and hull, and
dietary factors including low intakes of fruits, includes the starchy endosperm, germ and bran
vegetables, nuts and seeds, whole-grains, and fiber in similar proportions that are found in the
seafood omega 3 and other polyunsaturated intact kernel compared to refined grains which
(PUFA) fats, and excessive intake of red and pro- consist primarily of the starchy endosperm with
cessed meats, sugar sweetened beverages, and low amounts of fiber, vitamins, minerals and phy-
sodium (Table 3.1) [22]. The optimal intake of tonutrients. Common examples of whole-grain
these dietary factors is at least 300 g of non-juice foods and ingredients include brown rice, oat-
fruit/day, 400 g vegetables including legumes/ meal, whole oats, bulgur (cracked wheat), pop-
day, 5 weekly servings of nuts, 2.5–3 servings of corn, whole rye, graham flour, and whole wheat.
whole-grains/day, 11% kcal from PUFA as a Reviews of whole-grain human studies recom-
replacement for carbohydrate and saturated fat mend that dietary patterns containing ≥3 servings
kcal, and 250 mg of seafood omega-3 fat, to limit of whole-grains and ≤3 serving of refined grains,
red meat to 100 g/week and sodium to 2000 mg/ without significant added saturated/trans-fat, sugar
day, and to avoid or rarely consume processed or sodium, are associated with healthy aging
meat or sugar sweetened beverages. and reduced chronic disease risk [11, 23–25].
Table 3.1 Association of sub-optimal intake of specific cular disease and cancer mortality [28–32]. Three
food groups on annual increased specific cardiometabolic
dose response meta-analyses show consistent and
disease related deaths among US adults (≥25 years) based
on 702,308 deaths in 2012 (adapted from [22]) significant protective effects of whole-grains in
lowering premature all-cause and disease specific
Total annual disease
increased number of mortality. The first dose-response meta-analysis
Suboptimal related cardiometabolic (11 cohort studies on all-cause mortality; 705,253
Food group dietary intake disease deaths (%) participants) found that 90 g (or 3 servings)
Fruit <300 g/day CHD (6%) whole-grains/day significantly reduced mortality
Stroke (22%) risk for all-cause by 17%, CVD by 29% and dia-
Vegetables <400 g/day CHD (7%)
betes by 51% [28]. The second dose-response
Stroke (22%)
analysis (10 cohort studies; 782,751 subjects;
Nuts and <5 servings/ CHD (15%)
Seeds week Diabetes (7%) 5.5–26 years of follow-up) showed that one
Whole-grains <2.5–3 CHD (4%) whole-grain serving/day significantly reduced
serving/day Stroke (11%) mortality for all-cause by 7%, CVD by 5% and
Diabetes (17%) CHD by 8% but diabetes morality was not signifi-
Red meats >100 g/week Diabetes (4%) cantly reduced, which is likely due to requiring a
Processed >0 g/day CHD (12%) higher effective threshold level above 1 serving/
meats Diabetes (18%)
day [31]. The third dose-response analysis (20
Sugar >0 g/day CHD (11%)
sweetened Stroke (1%) cohort studies; 2,282,603 subjects; follow-up of
beverages Diabetes (15%) 5.5–26 years) found that the daily consumption of
PUFA as a <11% kcal/ CHD (4%) three servings (90 g) whole-grains significantly
replacement day reduced risk of all-cause mortality by 17% and
for reduced CVD mortality by 25% and total cancer
carbohydrates
and saturated mortality by 10% [32]. Two other non-dose-
fat response meta-analyses provide similar signifi-
Seafood <250 mg/day CHD (15%) cant supporting evidence for the protective effect
omega 3 fat of ≥3 whole grain servings/day in lowering the
Sodium >2000 mg/ CHD (10%) risk of premature mortality [29, 30].
day Stroke (11%)
pecific Prospective Cohort Studies

S
However, only about 1% of Americans follow the Five large prospective studies demonstrate
recommendation for whole-grain intake as the that increased whole-grain intake significantly
average American intake is <1 serving of whole reduced all-cause and disease specific mortality
grains/day and ≥70% exceed the recommended [33–37]. For ready to eat cereal (RTEC) intake,
intake for refined grains [11, 26, 27]. the National Institutes of Health (NIH)-AARP
Diet and Health Study (367,442 subjects; aver-
3.2.1.2 Mortality Risk age 14-year follow-up) found that participants
Five systematic reviews and meta-analyses of who consumed high RTEC with whole-grain or
prospective cohort studies and five representative high cereal fiber had a significant 15% lower
cohort studies assessing the effects of whole- risk of all-cause mortality and 10–30% lower
grains intake on all-cause and disease specific risk of specific chronic diseases compared to
mortality are summarized in Table 3.2 [28–37]. non-consumers [33]. The US Nurses’ Health
and the Health Professionals Follow-Up studies
Meta-analyses (118,000; mean age 50–53 years; 25 years of
All five of these meta-analyses of prospective follow-up) showed that whole-grain intake was
studies published in 2016 strongly support the inversely associated with total and CVD mor-
protective effects of a daily intake of ≥3 servings tality (Fig. 3.2) [34]. Each serving of whole
of whole-grains in reducing all-cause, cardiovas- grains (28 g/day) was significantly inversely
Table 3.2 Summary of whole-grain prospective cohort studies on all-cause and disease specific mortality risk
Systematic Reviews and Meta-Analyses
Aune et al. (2016). 11 cohort studies; whole A 90 g/day increase in whole grain intake equivalent to
Quantify the dose- grain bread, whole grain 3 servings (for example, two slices of bread and one
response relation between breakfast cereals, and added bowl of cereal or one and a half pieces of pita bread
consumption of whole bran; 705,253 participants; made from whole grains) significantly reduced risk of
grain or specific types of 100,726 deaths. mortality from all causes by 17%, CVD by 29%, and
grains and all cause and diabetes by 51%. Reductions in risk were observed up
cause specific mortality to an intake of 225 g whole-grain products (7.5
[28]. servings/day)/day.
Wei et al. (2016). 11 prospective studies; total On the basis of the highest vs. the lowest categories of
Determine the effects of of 816,599 subjects; 89,251 intake, whole grains were associated with a lower risk
whole-grain consumption deaths. of mortality from all causes by 13%, CVD by 19%.
and the risk of all-cause, and all cancers by 11%. For each 3 servings/day
CVD and cancer increase in whole-grain intake, there was a 19%
mortality [29]. reduction in the risk of all-cause mortality, a 26%
reduction in CVD mortality and a 9% reduction in
cancer mortality.
Zong et al. (2016). 14 cohort studies plus Comparing highest vs. lowest intake of whole-grain
Evaluate the effect of unpublished results from intake there was a significant reduced risk for total
whole-grain intake on National Health and mortality by 16%, for CVD by 18%, and for cancer by
mortality from all-cause, Nutrition Examination 12%. For each 16-g/day increase in whole-grains
CVD and cancer [30]. Survey (NHANES) III and intake morality risk reductions were 7% for total, 9%
NHANES 1999–2004; for CVD, and 5% for cancer.
786,076 participants, 97,867
deaths.
Li et al. (2016) 10 cohort studies; 782,751 Per increment of 1 serving (30 g) a day of whole grain
Investigate the dose– subjects; 92,647 deaths; intake, mortality risk was significantly reduced for
response correlation 5.5–26 years of follow-up. all-cause by 7%, CVD by 5% and CHD by 8%. Higher
between consumption of consumption of whole grains was not appreciably
whole grains and the risk associated with risk of mortality from stroke and
of all-cause, CVD, and diabetes.
diabetes-specific
mortality [31].
Benisi-Kohansal et al. 9 cohort studies on total For an increase of 3 servings total whole grains/d
(2016) conduct a whole-grain intake and 11 (90 g/d), there was a significantly reduced risk of
meta-analysis of cohort studies on specific all-cause mortality by 17%. Also, each additional daily
prospective cohort studies whole-grain food intake; 3 servings total whole grains reduced CVD mortality
to summarize the relation 2,282,603 subjects; by 25% and total cancer mortality by 10%.
between whole-grain follow-up of 5.5–26 years;
intake and risk of 191,979 deaths (25,595
mortality from all-causes, from CVD, 32,746 from
CVD and total and total cancers, and 2671 from
specific cancers [32]. specific cancers).
Prospective Cohort Studies
Xu et al. (2016). Assess 367,442 participants; In multivariate models, participants in the highest
the associations of ready average 14 years of intake of RTEC, compared to non-consumers of RTEC,
to eat cereal (RTEC) follow-up; 46,067 deaths had a 15% lower risk of all-cause mortality and
intakes on all causes and (multivariate adjusted). 10%–30% lower risk of disease specific mortality.
disease- Within RTEC consumers, total fiber intakes were
specific mortality risk associated with reduced risk of mortality from
(US National Institutes of all-cause mortality and deaths from CVD, all cancers,
Health (NIH)-AARP Diet and respiratory disease (all p- trend <0.005).
and Health Study) [33].
(continued)
Wu et al. (2015). 74,341women and 43,744 Whole-grains were inversely associated with total and
Examine the association men; mean age 50–53 years; CVD mortality (Fig. 3.2). Every serving (28 g/day) of
between dietary whole followed for 24–26 years; whole grain was associated with a 5% lower total
grain consumption and 26,920 deaths (multivariate mortality or a 9% lower CVD mortality, whereas the
risk of mortality (US adjusted). same intake level was non-significantly associated with
Nurses’ Health and the lower cancer mortality.
Health Professionals
Follow-Up studies) [34].
Huang et al. (2015). 367,442 participants; On the basis of highest to lowest intake, whole grains
Assess the association of average 14 years of showed significant reduction in mortality risk for
whole grains and cereal follow-up; 46,067 deaths all-cause by 17% and 11–48% lower risk of disease-
fiber intake with all (multivariate adjusted). specific mortality. The highest intake of cereal fiber
causes and cause-specific had a 19% lower risk of all-cause mortality and
mortality (US NIH- 15–34% lower risk of disease-specific mortality (all
AARP Diet and Health p < 0.05).
Study) [35].
Johnsen et al. (2015). 120,010 adults; median age Whole-grain products were associated with a
Investigate the association 52 years; mean BMI 25; significantly lower all-cause mortality of 11% (highest
of whole-grain intake and 11–17 years of follow-up vs. lowest intake) with the strongest associations being
all-cause and cause- (multivariate adjusted). for breakfast cereals and non-white bread, especially
specific mortality whole-grain oat, wheat and rye.
(Scandinavian HELGA
cohort; Norwegian
Women and Cancer
Study, the Northern
Sweden Health and
Disease Study, and the
Danish Diet Cancer and
Health Study) [36].
He et al. (2010). 7822 US women with type 2 For the highest versus the lowest intakes, lower
Investigate whole grain and diabetes mellitus; mean age all-cause mortality risk for whole grain was reduced by
its components cereal fiber, 47 years; mean BMI 30; 11%, cereal fiber by 19%, bran by 25%, and germ by
bran, and germ in relation 26 years of follow-up; 852 5%. Only the association for bran intake was
to all-cause and CVD- all-cause deaths and 295 significant. Whole-grain and with bran intakes were
specific mortality in CVD deaths (multivariate associated with a significant lowering in CVD-specific
patients with type 2 adjusted). mortality in women with diabetes.
diabetes mellitus (US
Nurses’ Health Study) [37].
associated with all-cause m ortality by 5% and cause mortality of 11% (highest vs. lowest
CVD mortality by 9%. The US NIH-AARP intake) with the strongest associations observed
Diet and Health Study (367,442; mean age for breakfast cereals and non-white bread, espe-
62 years; 14 years of follow-up) observed that cially whole-grain oat, wheat and rye [36]. A
higher intake of whole- grains significantly Nurses’ Health Study evaluation of women with
reduced the risk of all-cause mortality by 17% type 2 diabetes (7822 women; mean 47 years at
and mortality risk from diabetes by 48%, CVD baseline; 26-year follow-up) showed that the
by 17% and cancer by 15% [35]. The large intake of whole-grain was inversely associated
Scandinavian HELGA cohort (120,010; median with all-cause and CVD-specific mortality [37].
age 52 years; mean BMI 25; 11–17 years of In each of these studies, the mortality protective
follow-up) found that whole-grain products effects of whole-grains were primarily due to its
were associated with a significantly lower all- bran (fiber) component.
Total mortality (p trend <0.01) CVD mortality (p trend <0.01)

1.1
1.05
1
Hazard Ratios
0.95
0.9
0.85
0.8
4 to 6 10 to 14 14 to 22 22 to 31 33 to 48
Whole-grain Intake (g/day)
Fig. 3.2 Effects of whole-grain foods intake on mortality risk of US men and women approx. 50 years of age at base-
line and followed for 24–26 years (adapted from [34])
3.2.1.3 Chronic Disease Risk review of RCTs (10 RCTs with 8 of these on
Whole-grain systematic reviews and meta- whole oats; 4–8 week durations) showed that oat-
analyses, and individual prospective cohort stud- meal accounted for virtually all the total and
ies and RCTs on chronic disease risk and risk LDL-cholesterol lowering effects of whole-
biomarkers are summarized in Table 3.3 [28, grains [40].
38–53].
Hypertension/Blood Pressure (BP)
Cardiovascular Diseases (CVD) One cohort study [41] and two RCTs [42, 43]
Five systematic reviews and meta analyses sup- support the protective effects of ≥3 daily whole-
port the protective effects of ≥3 daily servings of grain servings in reducing the risk of elevated
whole-grains in lowering the risk of developing BP or hypertension. The Health Professionals
CVD or biomarkers [28, 38–41]. A dose-response Follow-up Study (31,684 men; baseline age
meta-analysis (17 cohort studies; 1 million par- range 45–75; 18 years of follow-up) reported
ticipants; 33,300 cases) found that three daily that men with higher whole-grain and bran
whole grain servings (90 g) significantly reduced intake had significantly lower risk of hyperten-
risk for CVD by 22% and CHD by19% [28]. A sion by 19% (multivariate adjusted) [41]. A dou-
systematic review of 24 RCTs found that ble-blind, crossover RCT (40 overweight/obese
increased intake of whole oats was primarily men and women; mean age 39 years; mix of
responsible for most of the total and LDL choles- whole grains vs. refined grains at 50 g/1000 kcal;
terol lowering effects of whole grains (Fig. 3.3) 8 weeks) found that the whole grain supple-
[38]. Another meta-analysis (45 prospective mented diet significantly lowered diastolic BP
cohort studies and 21 RCTs) found in cohort that by 4.2 mm Hg more than a similar diet supple-
48–80 g whole-grains/day significantly lowered mented with refined grains [42]. A second RCT
CVD risk by 21% compared to never or very low (233 adults; mean age 52 years; 12 weeks) dem-
intake and RCTs showed that higher whole-grain onstrated that three daily servings of mixed
intake significantly lowered mean fasting glu- whole-grain foods significantly lowered systolic
cose, and total and LDL cholesterol compared to BP compared to diets with refined grain foods
refined-grains [39]. A 2007 Cochrane systematic (Fig. 3.4) [43].
Table 3.3 Summary of whole-grain prospective cohort studies and randomized controlled trials (RCTs) on chronic
disease risk
Cardiovascular Disease (CVD)
Aune et al. (2016). CHD (7 cohort studies; Per 3 whole-grain daily servings (90 g), there
Quantify the dose-response relation 316,491 subjects; 7068 was a significant reduced risk for CHD by
between consumption of whole cases); CVD (10 cohort 19% and CVD by 22%.
grain or specific types of grains and studies; 704,317 subjects;
the risk of CHD and CVD [28]. 26,243 cases).
Hollænder et al. (2015). 24 RCTs. Specifically, overall whole -grains significantly
Assess the effect of whole-grain lowered LDL cholesterol by 0.09 mmol/L and
compared with non–whole-grain TC by 0.12 mmol/L compared with the
foods on changes in total cholesterol control. Whole-grain oats had the greatest
(TC), LDL cholesterol, HDL effect, lowering TC by 0.17 mmol/L Fig. 3.3).
cholesterol, and triglycerides [38]. No effect of whole-grain foods on HDL
cholesterol or triglycerides was found.
Ye et al. (2012). 45 cohort studies; 21 RCTs. Cohort studies, which compared never/rare
Systematically examine human consumers of whole-grains with those
studies investigating whole-grain consuming 48–80 g whole grain/day showed
and fiber intake and CVD risk [39]. lower risk of CVD by 21% for the whole-grain
consumers. In RCTs, whole-grains
significantly lowered TC by 0.83 mmol/L and
LDL-cholesterol by 0.72 mmol/L vs. control
diets (multivariate adjusted).
Kelly et al. (2007). 10 RCTs with 8 of these Whole-grain oats significantly lowered TC by
Assess the effects of whole-grain testing whole-grain oats; 0.20 mmol/L and LDL cholesterol by
intake on CHD risk factors in ranged in duration from 4 to 0.18 mmol/L (all, p < 0.0001) vs. control diets.
participants previously diagnosed 8 weeks. However, there was a lack of evidence for
with CHD or with existing risk significant lowering effects for other non-oat
factors for CHD (Cochrane whole grains.
Systematic Review) [40].
Hypertension/Blood Pressure
Prospective Cohort Study
Flint et al. (2009) 31,684 men free of In multivariate-adjusted analyses, whole-grain
Evaluate the association between hypertension, cancer, stroke, intake was inversely associated with risk of
whole-grain intake (g/day) and risk or coronary heart disease; hypertension by 19% (highest vs. lowest
of hypertension in men (The Health baseline age 45–75 years; quintile; p-trend <0.0001). A similar inverse
Professionals Follow-up Study; US) 18 years of follow-up association was shown for total bran with a
[41]. (multivariate adjusted). risk reduction of 15% (p-trend =0.002).
RCTs
Kirwan et al. (2016). Double-blind, Crossover The whole-grain supplemented diet
Evaluate the efficacy of whole RCT: significantly lowered diastolic blood pressure
grains compared with refined grains 40 overweight or obese men (BP) by 4.2 mm Hg more than the refined
on body composition, hypertension, and women; mean age grain control diet (p = 0.01). This decreased
and related mediators of CVD in 39 years with no known whole-grain diastolic BP was correlated with
overweight and obese adults (US) history of CVD; whole-grain higher circulating adiponectin levels (r = 0.35,
[42]. and refined-grain (control) p = 0.04).
diets (50 g/1000 kcal in each
diet); 8-week periods;
10-week washout.
Tighe et al. (2010). Parallel RCT: After 6 weeks of the intervention, the wheat
Assess the effects of the intake of 3 233 middle age healthy and oats group systolic BP was significantly
daily portions of whole-grain foods subjects; mean age 52 years; lowered by 3.7 mmHg more than the refined
(provided as only wheat or a after a 4-week run-in period grain group. By 12 weeks, the decrease in
mixture of wheat and oats) on with a refined grain diet, systolic BP was significantly lowered in both
markers of CVD risk in relatively subjects were randomly whole-grain groups compared to the refined
high-risk individuals (UK) [43]. allocated to a control (refined grain group (P = 0.01; Fig. 3.4).
grain diet); 3 daily portions
wheat, or combined wheat
and oats; 12 weeks.
Ischemic Stroke
Systematic Reviews and Meta-analyses
Chen et al. (2016). 7 cohort studies; 446,451 High intake of whole grains was inversely
Evaluate the effect of whole and subjects; 5892 stroke events. associated with ischemic stroke risk by 25%
refined grains on stroke risk [44]. vs. low whole-grain intake. Consumption of
refined grains was not associated with risk of
stroke or its subtypes.
Fang et al. (2015). 6 cohort studies; 247,487 Highest whole grain intake vs. lowest intake
Examine the association between subjects; 1635 stroke events. was significantly associated with reduced risk
whole-grain intake and stroke of total stroke by 14%. In subgroup analysis
risk [45]. inverse stroke associations were also found in
the American population by 19% and in
women by 22%.
Liu et al. (2000). 75,521 women; baseline age For the highest vs. lowest whole-grain intake,
Examine the relationship between 38–63 years; 12 years of the risk of ischemic stroke was significantly
whole-grain intake and the risk of follow-up; 352 cases of reduced by 31% and for never smokers the
ischemic stroke in women (Nurses’ ischemic stroke (multivariate risk was reduced by 50%.
Health Study; US) [46]. adjusted).
Type 2 Diabetes (Diabetes)
Chanson-Rolle et al. (2015). 8 observational studies; 7 The daily consumption of 45 g of whole grains
Quantitative evaluation of the cohort studies; 316,051 significantly reduced diabetes risk by 20%
relationship between whole grain participants; 15,573 diabetes compared to the daily intake of 7.5 g.
intake and diabetes risk [47]. cases.
Aune et al. (2013). 16 cohorts; whole grain (10 Per 3 daily whole grain servings diabetes risk
Assess the dose–response cohorts; 385,868 participants was reduced by 32% and for refined grain the
relationship between whole grain 19,829 cases); refined grain diabetes risk was reduced by 5%. A significant
intake and diabetes risk [48]. (6 cohorts; 258,078 nonlinear association was observed for whole
participants; 9545 cases). grains but not for refined grains. Fig. 3.5
summarizes the diabetes risk lowering effects
of specific whole grain foods.
Specific Prospective Studies
Sun et al. (2010). 39,765 men and 157,463 Replacing 50 g/day (cooked, equivalent to 1/3
Examine the effects of white and women; baseline age range serving/day) of white rice with the same
brown rice on US men and women 26–87 years; 14–22 years of amount of brown rice was associated with a
(Nurses’ Health Studies 1 and II and follow-up; 2359–5500 16% lower risk of diabetes, whereas the same
the Health Professionals Follow-up diabetes cases (multivariate replacement with other types whole grains
Study) [49]. adjusted). lowered diabetes risk by 36%.
(continued)
de Munter et al. (2007). 161,737 women; baseline The daily intake of 31–40 g vs. about 5 g
Evaluate intakes of whole grain, age 26–65 years; 12–18 years whole-grains significantly lowered diabetes
bran, and germ in relation to risk of of follow-up; 6486 diabetes risk by 14–25% with the largest risk reduction
diabetes in prospective cohort cases. in the older women. Associations for bran
studies (Nurses’ Health Studies plus 6 cohort studies; 286,125 intake were similar to those for total whole
a meta-analysis) [50]. participants; 10,944 diabetes grain intake, whereas no significant
cases (multivariate adjusted). association was observed for germ intake after
adjustment for bran. In the meta-analysis, 2
daily whole grain servings were associated
with a 21% decrease in diabetes risk after
adjustment for potential confounders and BMI.
Colorectal Cancer (CRC)
Aune et al. (2011). 6 cohorts This dose-response analysis found that CRC
Assess the whole-grain dose- risk per 3 whole-grain increment servings
response effect on CRC risk [51]. daily was reduced by 17%.
Haas et al. (2009). 11 cohort studies, 1,719,590 In the multivariate analysis, the highest
Evaluate the effectiveness of participants; baseline age quintile of whole-grain intake was associated
whole- grain in preventing CRC 25–76 years; 6–16 years of with 6% lower risk of CRC. A sub-analysis by
[52]. follow-up; 7745 colorectal tumor location showed reduced risk in the
cancer cases. colon by 7% and rectum by 11%.
Kyro et al. (2013). 108,000 participants; median Intake of whole-grain products was associated
Assess the dose-response baseline age mid-50s; with a lower incidence of CRC per 50-g
association between whole-grain median 11 years of increment by 6%. Intake of whole-grain wheat
intake and CRC risk (Scandinavian follow-up; 1123 colorectal (highest vs. lowest quartile) was associated
cohort HELGA consisting of cancer cases (multivariate with a lower incidence of colorectal cancer by
Danish, Swedish, and Norwegian adjusted). 34% but the effect was non-linear.
persons [53].
Total Cholesterol LDL Cholesterol

0
Mixed (p >.05) Wheat (p >.05) Oat (<.001)
Type of Whole Grain

-0.05
-0.1
mmol/L
-0.15
-0.2
-0.25
Fig. 3.3 Association between wheat, oats and a mixture of the two whole-grains on total and LDL-cholesterol from a
meta-analysis of 24 RCTs (adapted from [38])
Type of Grains (per 3 Daily Servings)
0
Refined Whole Wheat Whole Wheat & Oats
-1
-2
-3
mm Hg
-4
-5
-6
-7
Fig. 3.4 Effect of type of grains consumed on blood pressure (BP) in 233 UK men and women with mean age of
52 years and BMI of 28 (adapted from [43])
Ischemic Stroke showed a significant linear inverse relationship

Two meta-analyses of prospective cohort studies between whole grain intake and diabetes inci-
[44, 45] and one representative cohort study dence; 45 g of whole grains/day significantly
example [46] provide support evidence for the reduced diabetes risk by 20% compared to con-
protective effects of ≥3 daily whole-grain serv- suming 7.5 g of whole grains/day [47]. In a dose-
ings on reducing the risk of ischemic stroke. A response meta-analysis (16 cohort studies), 3
2016 meta-analysis (7 cohort studies; 446,451 servings of whole-grains significantly reduced
subjects) found that higher whole-grain intake diabetes risk by 32% whereas the same number
reduced ischemic stroke risk by 25% whereas of servings of refined grains slightly reduced risk
refined grains were not associated with lower by 5% [48]. The effects of various grain foods on
stroke risk [44]. A 2015 meta-analysis (6 cohort diabetes risk are summarized in Fig. 3.5. A sys-
studies; 247,487 subjects) reported that higher tematic review (6 cohort studies; 286,125 partici-
intake of whole-grains reduced total stroke risk pants; 10,944 diabetes cases) found that two
by 14% in the total population and by 22% when servings per day of whole-grains was associated
only women were analyzed [45]. The Nurses’ with a 21% lower risk of diabetes, after adjusting
Health Study (75,521 women; baseline age range for potential confounders including BMI [50].
38–63; 12 years of follow-up) showed that higher The 2010 pooled data from the Nurses’ Health
whole-grain intake significantly lowered isch- Study I and II and Health Professionals Follow-up
emic stroke risk in all women by 31% and in non- Study (39,765 men and 157,463 women; baseline
smoking women by 50% (highest vs. lowest age range 26–87 years; 14–22 years follow-up)
quintile of intake; multivariate adjusted) [46]. showed that replacing 50 g/day cooked white rice
with the same portion of brown rice was associ-
ype 2 Diabetes (Diabetes)
T ated with a 16% reduced risk of diabetes, whereas
Three meta-analyses of cohort studies [47, 48, the same portion of a variety of whole grains fur-
50] and one representative large pooled cohort ther lowered diabetes risk by 36% [49].
study [49] support the protective effects of ≥3
daily whole-grain servings on reducing the risk olorectal Cancer (CRC)
C
of diabetes. A systematic review and meta-analy- Two meta-analyses of cohort studies [51, 52] and
sis (7 prospective studies and 1 case control a representative cohort study [53] support the
study; 316,051 adults; 15,573 diabetes cases) protective effects of ≥3 daily whole-grain
1.2
Relative Risk for Type 2 Diabetes

1
0.8
0.6
0.4
0.2
0
Whole grain Whole grain Brown rice White rice Wheat bran Wheat germ
bread breakfast
cereal
Fig. 3.5 Subtypes of grains on type 2 diabetes risk from meta-analysis of 16 cohort studies (adapted from [48])
s ervings in reducing the risk of colorectal cancer. bacteria and their endotoxins reducing chronic
A meta-analysis (6 cohort studies) showed that 3 systemic and acute episodes of colonic inflam-
daily servings of whole-grains significantly mation, (2) reduce the risk of excessive levels of
reduced colorectal cancer risk by 17% [51]. visceral fat, a known source of systemic inflam-
Another meta-analysis (11 cohort studies; mation, and (3) protect against periodontitis, a
1,719,590 subjects) found that higher whole- source of increased systematic inflammation with
grain intake reduced colorectal cancer risk by 6% aging. In adults >65 years higher IL-6 and CRP
[52]. The Scandinavian HELGA cohort study levels are especially associated with poorer cog-
(108,000 adults; age 52 years at baseline; 11 years nitive and/or functional performance, and a
of follow-up) observed that whole-grain intake higher risk of mortality [59].
was associated with a lower incidence of colorec-
tal cancer risk for total whole grains by 6% per Microbiota Ecosystem
25 g intake [53]. A healthy microbiota is an important barrier to
inflammaging processes and helps to promote
3.2.1.4 Aging healthy aging and reduced chronic disease risk
Elevated levels of proinflammatory cytokines and and progression to frailty in later life [60–65].
acute phase proteins, such as interleukin (Il)-6, Whole-grains provide a major source of fiber,
tumor necrosis factor (TNF)-α, and C-reactive which is the main energy source for colonic bac-
protein (CRP) have been associated with subopti- teria and resulting fermentation products, which
mal aging, known as inflammaging, and associ- among other functions, are important in main-
ated with increased chronic disease risk [54–57]. taining low systemic inflammation. In contrast,
A review of observational studies suggests that refined grain products are processed to remove
there is an inverse association between diets high most of their fiber and other bioactive compo-
in whole-grains and CRP concentrations [58]. nents [66, 67]. Increased intake of fiber- rich
Higher intake of fiber-rich whole-grains may pro- whole-grain foods can increase the relative num-
mote healthy aging by modulating systemic bers of bifidobacteria and lactobacilli bacteria,
inflammation and helping to (1) maintain a and fecal butyrate concentration and lower
healthy colonic microbiota ecosystem which pro- colonic pH, which promote colon barrier defenses
motes colon barrier defenses against pathogenic against pathogenic bacteria and their endotoxins
to reduce chronic systemic inflammation, aid in 2013 PREvencion con DIeta MEDiterranea
healthier aging, and reduce chronic disease risk. (PREDIMED) trial on 4 year changes in adipos-
Three RCTs demonstrate the beneficial effects of ity, which found that subjects with the highest
whole-grains on the microbiota and inflammatory white bread intake gained 0.8 kg more weight
markers [68–70]. One parallel RCT (49 men and and increased waist circumference by 1.3 cm
32 postmenopausal women; age range more than those with the lowest intake, which
40–65 years; 6 weeks) found that diets with whole impaired weight control by reducing their odds
grains (16 g whole grain/1000 kcal) significantly of losing weight by 33% and waist circumference
increased butyrate producing bacteria, total fecal by 36%. In the Framingham Heart Study (2834
short chain fatty acids (SCFAs) concentrations adults; mean age 51 years; 49% women), increas-
and acute innate immune response, and reduced ing whole-grain intake was associated with
colonic pro-inflammatory Enterobacteriaceae lower visceral adipose tissue volumes whereas
bacteria levels compared to refined grain diets higher intakes of refined grains were associated
[68]. A crossover RCT (28 healthy adults; daily with higher visceral fat levels (Fig. 3.6) [75]. A
dose of 60 g whole grain barley, brown rice or 2017 RCT (81 adults; age range 40–65 years;
mixture of the two; 4 weeks) showed improve- 6 weeks) found that subjects on a whole-grain
ments in the colonic microbiota ecosystem that rich diet (207 g/day) had a significantly higher
coincided with significantly reduced plasma IL-6 net daily energy loss by 92 kcal/day compared to
and peak postprandial blood glucose [69]. A par- those on a refined-grain diet (0 g whole-grains/
allel, single blind RCT (79 healthy; mean age day), which provides mechanistic support for
52 years, mean BMI 29; 3 daily whole-grain or whole-grain observational findings for lower
refined grain servings in the form of bread and risk of weight and fat gain via increased resting
breakfast cereal added into habitual diets; metabolic rate and metabolizable energy excre-
4 weeks) demonstrated that the whole-grain diet tion in the stool [76].
significantly lowered plasma CRP levels [70].
Periodontal Disease
Visceral Fat Periodontal disease is a set of inflammatory dis-
Unhealthy aging is often associated with inflam- eases affecting the tissues that surround and sup-
maging, characterized by a relative increase in port the teeth [77]. Approximately half of adults
visceral fat and loss of peripheral subcutaneous age ≥30 years in the US have mild, moderate
fat, which is associated with increased systemic or severe periodontal disease [78]. The risk of
free fatty acid and inflammatory biomarkers such periodontal disease increases with age and is
as CRP [58, 71]. People consuming the recom- more common in men than women. Periodontal
mended levels of whole-and refined-grains have disease and CVD share a common etiology of
lower odds of central obesity, including visceral chronic inflammation. NHANES analyses from
fat and waist circumference gains [26, 72–76]. 2009–2010 and 2011–2012 (6052 US adults;
A meta-analysis (26 RCTs, ≤16 weeks) found age ≥30 years) found an inverse relation between
that increased intake of whole-grains was shown fiber intake and periodontal disease among US
to modestly but significantly lower percentage of adults [78]. Also, periodontal disease was associ-
body fat by 0.48% compared with a refined grain ated with low whole-grain intake but not with low
control (in a relatively short time-frame) [73]. fruit and vegetable intake. The Health Professionals
A 2015 systemic review of observational stud- Follow-up Study (34,160 men; age range
ies and intervention trials concluded that whole- 45–75 years; 14 years of follow-up) showed that
grain bread was more beneficial than white bread men with a median intake of >3.4 daily servings of
in controlling weight gain and abdominal fat whole grains had 23% lower odds of developing
and there was a possible relationship between periodontal disease than those who consumed
high white bread intake and excess abdomi- <0.3 servings (multivariate; p-trend <0.001) [79].
nal fat [74]. Also, this review summarized a In contrast, refined-grain intake was associated
Whole-grains (p-trend <.001) Refined grains (p-trend <.001)

1950
Visceral Adipose Tissue Volumes (cm3)

1900
1850
1800
1750
1700
1650
Q-1 Q-2 Q-3 Q-4 Q-5
Daily Grain Serving Quintiles
Fig. 3.6 Association between increasing whole and refined grains and visceral adipose tissue volume in US adults
(whole grain: Q-5 [3 servings]; refined grains: Q-5 [4 servings]) (adapted from [75])
with an insignificant 4% increased odds of peri- v egetables [81]. The USDA MyPlate educational
odontal disease (multivariate; p-trend = 0.37). concept, devotes one-half the plate to fruit and
vegetables as a displacement of other foods of
higher energy density from the diet. This promotes
3.2.2 Fruit and Vegetables healthy aging, weight control and the prevention
of chronic disease risk and premature mortality,
3.2.2.1 Background especially if the fruits and vegetables consumed
The 1990 World Health Organization (WHO) are rich in flavonoids and fiber.
report recommended a minimum daily intake of
five daily servings of fruits and vegetables, which 3.2.2.2 Mortality Risk
provide aging and chronic disease protective levels It is estimated that the global intake of < 500g/
of micro and macronutrients, and phytochemicals day of fruit and vegetables was associated
such as flavonoids and carotenoids [80]. However, with 5.4 million premature deaths, including
globally only a small fraction of populations 710,000 CHD deaths, 1.5 million stroke deaths,
consume recommended intake of fruit and vegeta- and 560,000 cancer deaths [82]. Table 3.4
bles; for example, >85% of the US population falls summarizes the findings from one meta-anal-
short of meeting the daily recommendations [10, ysis of cohort studies and six representative
11]. Most Americans consume less than one cup of cohort studies on the effects of fruits and veg-
fruit and less than two cups of vegetables daily, etables on all-cause and disease specific mor-
with the primary contributors consisting of juice tality [82–89]. A 2017 systematic review and
and processed potatoes, compared to the current dose-response meta-analysis (94 cohort stud-
recommendations of 2 cups of fruit and 2.5 cups of ies; 2,123,415 participants) found significant
vegetables per day [11, 81]. A 2016 pooled analy- reduced all-cause mortality risk for combined
sis of 124,086 US men and women suggests that fruit and vegetable intake by 10%, for fruit
consuming high flavonoid fruits and vegetables, intake by 15% and vegetable intake by 13% per
such as apples, pears, berries, and peppers, may be 200 g/day [82]. Also, there was evidence that
especially helpful in preventing weight gain and high vs. low intake of apples, pears, berries, cit-
obesity compared to other types of fruits and rus fruits, 100% fruit juice, cooked vegetables,
Table 3.4 Summary of fruit and vegetable prospective studies on mortality risk
Reference/Objective Methods/Subjects Results
Systematic Review and Meta-Analysis
Aune et al. (2017) 94 cohort studies; 2,123,415 Per 200 g/day, there was a significant reduced
Assess the effects of total participants. risk of all-cause mortality for combined fruit and
fruits and vegetables and vegetable intake by 10%, for fruit intake by 15%
specific types on all-cause and vegetable intake by 13%. There was
mortality risk [82]. evidence that high vs. low intake of apples/pears,
berries, citrus fruits, 100% fruit juice, cooked
vegetables, cruciferous vegetables, potatoes and
green leafy vegetables/salads were inversely
associated with all-cause mortality and canned
fruits increased risk. Also, a dose-response
analysis found that 100% fruit juice, cruciferous
vegetables and green leafy vegetables/salads
were significantly associated with reduced
all-cause risk (Fig. 3.7).
Wang et al. (2014). 16 cohorts up to August Higher consumption of fruit and vegetables is
Examine and quantify the 2013; 833,234; 4.6–26 years associated with a reduced risk of all-cause
potential dose-response of follow-up; 56,423 deaths. mortality with an average 5% reduced risk per
relation between fruit and additional daily serving (Fig. 3.8). There was a
vegetable consumption and threshold at 5 servings/day, after which the risk
risk of mortality from of mortality did not reduce further. There was a
all-cause, CVD, and cancer significant inverse association for CVD mortality
[83]. by 4% for each added serving. However, there
was no appreciable association with cancer
mortality.
Bellavia et al. (2016). 74,645 men and women; In high red meat consumers, increased fruit and
Determine the effect of fruit baseline age 45–83 years; vegetable intake did not affect the increased
and vegetable intake on red 16 years of follow-up; 17,909 mortality risk of all-cause by 21% or CVD
meat mortality risk (the deaths occurred in the cohort mortality by 29%. This study found no
Swedish Mammography (mortality resulting from detectable affect of fruit and vegetables lowering
Cohort and the Cohort of CVD-related death 5495 and the elevated mortality risk in people with a high
Swedish Men) [84]. 4426 cancer deaths intake of red meat.
(multivariate adjusted).
Nguyen et al. (2016). 150,969 adults; mean The consumption of ≥7 servings/ day of total
Examine the association baseline age 60 years; 55% fruit and vegetables reduced all-cause mortality
between intake of vegetables women; average follow-up risk by 10% (p for trend =0.002; high vs. low
and fruits, and raw vs. cooked 6.2 years; 6038 all cause quartile). Fruit intake was inversely associated
vegetables consumption on the deaths (multivariate with all-cause mortality with a 16% risk
risk of all-cause mortality adjusted). reduction (p-trend ≤0.001; high vs. low intake).
(Australia) [85]. Consumption of total and cooked vegetables,
significantly reduced risk by 7% and 13%,
respectively (p-trend <0.05) but raw vegetables
reduced risk by 6% (p-trend =0.79).
Oyebode et al. (2014). 65,226 participants; baseline High daily fruit and vegetables intake (≥ 7 daily
Examine the association age 35+ years; median serving) was associated with a significant 33%
between fruit and vegetable 7.7 years of follow-up lower multivariate all-cause mortality risk, which
consumption and all-cause, (multivariate adjusted). was lowered to 48% lower risk after excluding
cancer and cardiovascular deaths during the first year of the study. Also,
mortality (Health Surveys for high fruit and vegetable intake was associated
England) [86]. with reduced cancer mortality by 25% and CVD
mortality by 31%. The effects of specific fruits
and vegetables on all-mortality risk are
summarized in Fig. 3.9.
(continued)
Reference/Objective Methods/Subjects Results
Leenders et al. (2013). 451,151 participants; range of Higher fruit and vegetables intake was
Evaluate the association 10–18 years of follow-up significantly associated with lower mortality risk
between fruit and vegetables (multivariate adjusted). for all-cause by 10% and for CVD by 15% (daily
intake and risk of mortality intake of >570 g of fruit and vegetables vs.
(European Prospective <250 g). Stronger inverse associations with both
Investigation into Cancer and cancer and CVD mortality were seen for raw
Nutrition [EPIC]) [87]. vegetables than for cooked vegetables.
Bellavia et al. (2013). 71,706 participants, 38,221 Those who rarely or never consumed fruit and
Examine the dose-response men and 33,485 women; vegetables lived shorter lives by 3 years and had
relation between fruit and mean baseline age a 53% higher mortality rate than those who
vegetable intake and mortality, 60 years;13 years of consumed ≥5 servings fruit and vegetables/day.
in terms of both time and rate follow-up; 11,439 deaths,
(The Swedish Mammography 6803 men and 4636 women
Cohort and the Cohort of (multivariate adjusted).
Swedish Men) [88].
Zhang et al. (2011). 134,796 Chinese adults; Fruit and vegetable intake was inversely
Investigate the associations of mean follow-up of 4.6– associated with risk of total mortality in both
cruciferous vegetables, 10 years; 3442 deaths among women and men. A significant dose-response
non-cruciferous vegetables, women; and 1951 deaths was observed for cruciferous vegetable intake
total vegetables, and total fruit among men (multivariate with a significant reduction in total morality risk
intake with risk of all-cause adjusted). by 22% (high vs. low quintiles) and 16% for
and cause-specific mortality total vegetables. Similar risk reduction was
(Shanghai Women’s Health observed for CVD mortality but not for cancer
Study and the Shanghai Men’s mortality.
Health Study) [89].
cruciferous vegetables, and green leafy veg- suggesting no beneficial interaction between
etables/salads were inversely associated with high red meat and increased fruit and vegetable
all-cause mortality and 100% fruit juice, intake. A large Australian study (150,969 adults;
cruciferous vegetables and green leafy veg- mean baseline age 60 years; 55% women; aver-
etables/salads were significantly associated age follow-up 6.2 years) observed that the con-
with reduced mortality risk per 100 g intake sumption of ≥7 servings/day of total fruit and
(Fig. 3.7). Another large dose-response meta- vegetables reduced all-cause mortality risk by
analysis (16 cohort studies; 833,234; 4.6– 10% (p for trend = .002; high vs. low quartile)
26 years of follow-up) found that total fruit [85]. Also, increased fruit intake was inversely
and vegetable intake was inversely associated associated with all-cause mortality by a signifi-
with all-cause mortality risk by 5% per daily cant 16% risk reduction (fully adjusted; high
serving (6% for fruits and 5% for vegetables) vs. low intake) and higher intake of total and
until a threshold of 5 servings/day was reached cooked vegetables in the fully adjusted mod-
(Fig. 3.8) [83]. Similar findings were observed els significantly reduced risk by 7% and 13%,
for CVD mortality, but not cancer mortality. A respectively, but higher raw vegetables insigni-
Swedish cohort study (74,645 adults; age range ficantly reduced risk by 6%. In English Health
45–83 years; 16 years of follow-up) observed Surveys (65,226; aged 35+ years; median
that high intakes of red meat (120–300 g/day), follow-up of 7.7 years) the consumption of
especially processed meats, were associated vegetables, salad, and fresh and dried fruit were
with a higher risk of mortality from all-causes associated with significantly lower all-cause
by 21% and CVD by 29% [84]. The increased mortality whereas the consumption of frozen or
risks were consistently observed in partici- canned fruit was associated with a significantly
pants with low, medium, and high fruit intake increased risk and 100% fruit juice was not
Per 100g/day Intake High vs Low Intake
Canned Fruits
Raw Vegetables
Cooked Vegetables
Root vegetables
Potatoes
Green leafy vegetables
Cruciferous vegetables
Onions/Allium vegetables
100% Citrus fruit juices
Citrus fruit
Bananas
Berries
Apples/Pears
-30 -25 -20 -15 -10 -5 0 5 10 15 20

All-cause Mortality Risk (%)
Fig. 3.7 Association between subtype of fruits and vegetables and all-cause mortality risk from a meta-analysis of 94
cohort studies (adapted from [82])
1.05
Hazard Ratio for All-cause Mortality
0.95
0.9
0.85
0.8
0.75
0.7
0 1 2 3 4 5 6+
Daily Servings of Fruits and Vegetables
Fig. 3.8 Frequency of daily fruit and vegetables consumption and the risk of all-cause mortality from a meta-analysis
of 16 cohort studies (p < 0.001) (adapted from [83])
1.20
Hazard Ratio for All-cause Mortality

1.10
1.00
0.90
0.80
0.70
0.60
0.50
Vegetables Salad Fresh fruit Dried fruit Frozen/ 100% Fruit
(p<.001) (p<.001) (p<.001) (p=.03) canned fruit juice
(p=.001) (p=.40)
Per Serving
Fig. 3.9 Association between specific fruit and vegetable foods and beverages and all-cause mortality risk in UK adults
aged ≥35 years (adapted from [86])
significantly related to mortality risk (Fig. 3.9) ardiovascular Diseases (CVD)

C
[86]. An EPIC study observed significantly Systematic Reviews and Meta-Analyses. Four
lower mortality risk for all cause by 10% and meta-analyses summarize the pooled data from
for CVD by 15% with the daily consumption of cohort studies and RCTs on the effects of increased
>570 g of fruits and vegetables vs. <250 g and fruit and vegetable intake on CVD risk [90–93]. A
there was a stronger inverse association in both meta-analysis of berries (22 RCTs; 1251 subjects;
cancer and CVD mortality seen for raw vege- 2–24 weeks) found that increased intake of berries
tables than for cooked vegetables [87]. A large significantly attenuated LDL cholesterol, systolic
Swedish dose-response cohort study (71,706 blood pressure (BP), fasting blood glucose, BMI
participants; mean baseline age 60 years; 60% and TNF-α levels, which are important factors
men; 13 years of follow-up) showed that those related to lower CVD risk [90]. A 2015 dose
participants with very low fruit and vegetable response meta-analysis (23 cohort studies; 937,665
intake had on average a 3-year shorter lifespan participants) showed a reduced CHD risk for 477 g/
and had a 53% higher mortality risk than those day of total fruit and vegetable intake by 12%, for
consuming 5 daily servings of fruit and vegeta- 300 g/day of fruit intake by 16% and for 400 g/day
bles [88]. The pooled data from the Shanghai of vegetable intake by 18% [91]. In the subgroup
Women’s Health and Shanghai Men’s Health analysis, there was a significant inverse association
studies (134,796; mean age mid-50 years; fol- observed in Western populations, but not in Asian
low-up of 5–10 years) suggests that crucifer- populations. Another 2015 meta-analysis (38
ous vegetables were particularly effective and cohorts; 1,498,909 participants; median 10.5 years
inversely associated with mortality risk for all- of follow-up) reported that those consuming 800 g
causes by 22% and CVD by 31% (>166 g vs. daily of fruits and vegetables reduced CVD risk by
<34 g/day) [89]. 17% [92]. A Cochrane systematic review (10 RCTs;
1730 subjects; ≥3-month duration) found increased
3.2.2.3 Chronic Disease Risk fruit and vegetables intake significantly lowered
Table 3.5 [90–117] summarizes the protective systolic and diastolic BP and LDL-cholesterol [93].
effect of fruits and vegetables on chronic disease Specific Studies. Three cohort studies and one
from cohort studies and RCTs. large RCT provide important insights on the effects
Table 3.5 Summary of fruit and vegetables prospective studies and randomized controlled trials (RCTs) on chronic
disease risk
Cardiovascular Disease (CVD)
Huang et al. (2016). 22 RCTs; 1251 subjects; The pooled results showed that berries intake
Estimate the effect of berries 2–24 weeks. significantly lowered LDL-cholesterol by
consumption on CVD risk 0.21 mmol/L, systolic blood pressure by 2.7 mmHg,
factors [90]. fasting glucose by 0.10 mmol/L, BMI by 0.36 kg/m2,
HbA1c by 0.20% and TNF-α by 0.99 ρg/mL vs.
control diets. Berries consumption appears to be
effective in prevention and control of CHD and CVD
risk.
Gan et al. (2015). 23 cohort studies; A dose–response analysis found a reduced CHD risk
Evaluate the dose-response 937,665 participants; of 12% per 477 g/day of total fruit and vegetable
relationship of fruit and 18,047 patients with intake; 16% per 300 g/day of fruit intake; and 18%
vegetable intake with CHD CHD. per 400 g/day of vegetable consumption. A
risk and quantify the dose– significant nonlinear association of CHD risk and
response relationship [91]. fruit or vegetable consumption was found
(p - nonlinearity <0.001). In the subgroup analysis,
there was a significant inverse association observed
in Western populations, but not in Asian populations.
Zhan et al. (2015). 38 cohort studies; For the highest versus lowest intake, pooled CVD
Examine the dose-response 1,498,909 participants; risk was reduced for total fruits and vegetables by
relation between fruit and median follow-up of 17%, fruits by 16%, and vegetables by 13%.
vegetable intake and CVD risk 10.5 years; 44,013 CVD Dose-response analysis showed that those eating
[92]. events. 800 g per day of fruits and vegetables had the lowest
CVD risk
Hartley et al. (2013). 10 RCTs; 1730 subjects; None of the trials reported clinical events or showed
Evaluate the effects of ≥ 3 months of duration. strong evidence for lowering CVD risk factors, but
increased fruit and vegetable trials were heterogeneous and short term. Increased
intake on the primary fruit or vegetable intake showed beneficial lowering
prevention of CVD (Cochrane effects on systolic blood pressure (BP) mean
Systematic Reviews) [93]. lowering by 3.0 mmHg, diastolic BP mean lowering
by 0.90 mmHg. Two studies showed significantly
lower LDL cholesterol.
Larsson, Wolk (2016). 69,313 healthy men and 3 to 7 servings/week of total potato consumption
Examine the association women; mean baseline (boiled potatoes, fried potatoes, and French fries)
between potato consumption age approx. 60 years; had no effect on CVD events, heart failure or
and risk of total and specific 13 years of follow-up; myocardial infarction (Fig. 3.10).
CVD events (Cohort of 10,147 major CVD events
Swedish Men and the Swedish (multivariate adjusted).
Mammography Cohort) [94].
Miedema et al. (2015). 2506 participants; 63% Higher intake of fruits and vegetables was associated
Assess the relationship women; baseline age with a lower prevalence of coronary artery calcium
between intake of fruits and 25 years; 20 years of by 26% (p-trend <0.001) in middle-aged adulthood,
vegetables during young follow-up (multivariate which was attenuated after adjustment for other
adulthood and coronary adjusted). dietary variables to a lower risk of 8%, but the trend
atherosclerosis later in life remained significant (p-value for trend <0.002). This
(Coronary Artery Risk reinforces the importance of establishing a high fruit
Development in Young Adults and vegetable intake early in life for CVD health.
[CARDIA] Study; US) [95].
(continued)
Bhupathiraju et al. (2013). 71,141 women and Participants in the highest quintile of fruit and
Examine the independent roles 42,135 men; mean vegetable intake had a 17% lower CHD risk. A
of quantity and variety in fruit baseline age approx. higher consumption of citrus fruit, green leafy
and vegetable intake in relation 50 years; 24 and 22 years vegetables, and β-carotene– and vitamin C–rich fruit
to CHD incidence (Nurses’ of follow-up; 2582 CHD and vegetables was associated with a lower CHD
Health Study and Health cases among women and risk (Fig. 3.11).
Professionals Follow-Up 3607 CHD cases among
Study) [96]. men (multivariate
adjusted).
RCT
CVD Prevention
Buil-Cosiales et al. (2016). Parallel RCT: This RCT found a significant inverse association
Investigate the association of 7216 elderly men and with CVD incidence and total daily fruit and
fiber, fruit, vegetable and women with high CVD vegetable consumption. Those Subjects consuming
whole-grain consumption with risk; mean baseline age ≥9 servings/day of fruits and vegetables had a 40%
CVD in a Mediterranean 67 years; up to 7-years lower CVD risk vs. those consuming <5 servings/
cohort of elderly adults at high follow-up; 342 confirmed day.
cardiovascular risk cases of CVD; yearly
(PREvención con DIeta repeated measurements of
MEDiterránea (PREDIMED) diet (multivariate
trial; Spain) [97]. adjusted).
Hypertension/Blood Pressure
Systematic Review and Meta-analyses
Li et al. (2016). 3 cohort, 2 case-control Comparing the highest with the lowest consumption,
Assess the hypertension risk and 20 cross-sectional the pooled hypertension risk was reduced by 19%
for highest vs. lowest fruit and studies; 334,468 subjects; for total fruits and vegetables, 27% for fruit, and 3%
vegetable consumption [98]. 41,713 hypertension for vegetables. A significantly inverse association
cases. between fruit consumption and hypertension risk by
30% was found in Asian studies.
Wu et al. (2016). 9 cohorts; 185,676 The risk of hypertension was decreased by 1.9% for
Evaluate dose-response participants. each daily fruit serving, and by 1.2% for each daily
association between the intake serving of total fruit and vegetables. For highest vs.
of fruit and/or vegetables and lowest intake, there was an inverse association with
the risk of developing hypertension risk for fruit by 13%, vegetables by
hypertension [99]. 12% and total fruit and vegetables by 10%.
Borgi et al. (2016). 187,453 participants; Compared with participants whose consumption was
Examine the independent baseline age range ≤4 servings/week, the pooled hypertension risk
association of whole fruit 43–55 years; >20 years of among those whose intake was ≥4 servings/ day was
(excluding juices) and follow-up; 77,373 reduced by 8% for total whole fruit intake and 5%
vegetable intake, as well as the hypertension cases for total vegetable intake. Analyses of individual
change in consumption of (multivariate adjusted). fruits and vegetables consumption levels of ≥4
whole fruits and vegetables servings/week (as opposed to <1 serving/month)
with hypertension incidence showed broccoli, carrots, avocado, tofu or soybeans,
(Nurses’ Health Study, Nurses’ raisins, and apples or pears were associated with
Health Study II, and Health lower hypertension risk (Fig. 3.12).
Professionals Follow-up
Study; US) [100].
Borgi et al. (2016). 187,453 participants; Compared with the intake of <1 serving a month, the
Determine whether higher baseline age range pooled hypertension risk for ≥4 servings a week was
intake of baked or boiled 43–55 years; >20 years of increased for baked or mashed potatoes by 11% (p
potatoes, French fries, or follow-up; 77,726 trend =0.05), French fries by 17% (p-trend =0.001)
potato chips is associated with hypertension cases and there was an insignificant risk reduction for
incidence of hypertension (multivariate adjusted). potato chips by 3% (p- trend =0.98) (Fig. 3.13). In
(Nurses’ Health Study, Nurses’ substitution analyses, replacing 1 daily serving of
Health Study II, and Health baked, boiled, or mashed potatoes with one serving a
Professionals Follow-up day of non-starchy vegetables was associated with
Study; US) [101]. decreased risk of hypertension by 7%.
RCT
Appel et al. (1997). Parallel RCT: A high fruit and vegetable diet with 8.5 daily
Examine the clinical effects of 154 subjects in each servings reduced systolic BP by 2.8 mm Hg
increased fruits and vegetables group; mean age approx. (p < 0.001) and diastolic BP by 1.1 mm Hg
on blood pressure (US) [102]. 45 years; mean systolic (p = 0.07) more than the control diet with 3.6 daily
and diastolic blood fruit and vegetable servings.
pressures were 131 and
85 mm Hg; 8 weeks.
Ischemic Stroke
Systematic Review and Meta-analysis
Hu et al. (2014). 20 cohort studies; For highest vs. lowest intake, stroke risk was
Summarize evidence from 760,629 participants; reduced for total fruits and vegetables by 21%, for
prospective cohort studies on mean follow-up ranged fruits by 23%, and vegetables by 14%. Citrus fruits,
the association of fruits and from 3 to 37 years; apples, pears, and leafy vegetables especially
vegetables consumption and 16,981 stroke events. contributed to the protection. The linear dose–
stroke risk [103]. response relationship showed that the risk of stroke
decreased for fruits by 32% and vegetables by 11%
per 200 g/day intake increment.
Cassidy et al. (2016). 43,880 healthy men; Higher intakes of flavonoids-rich fruits such as
Examine the relation between mean baseline age oranges, limes, lemons, apples and pears were
habitual anthocyanin and 53 years; 24 years of associated with a lower risk of ischemic stroke in
flavanone intake and coronary follow-up; 4046 men by 22% (p-trend =0.059), with the greatest
artery disease and stroke myocardial infarction and effects in participants aged ≥65 years (p-interaction
(Health Professionals 1572 stroke cases =0.04).
Follow-Up Study) [104]. (multivariate adjusted).
Oude Griep et al. (2011). 20,069 men and women; Green, orange/yellow, and red/purple fruits and
Examine associations between baseline average age vegetables were not related to stroke incidence.
consumption of fruit and 42 years; 10 years of Higher intake of white fruits and vegetables (e.g.,
vegetable color groups with follow-up; 233 stroke apples, pears, apple juice and sauce, bananas,
10-year stroke incidence cases (multivariate cauliflower, chicory, cucumbers, and mushrooms)
(Monitoring Project on Risk adjusted). were inversely associated with incident stroke (171
Factors and Chronic Diseases vs. 78 g/day) by 52%. Each 25-g/day increase in
in the Netherlands [MORGEN white fruit and vegetable consumption was
Study] [105]. associated with a 9% lower risk of stroke. Apples
and pears were the most commonly consumed white
fruit and vegetables (55%).
(continued)
Joshipura et al. (1999). 75,596 women; baseline Higher fruit and vegetable intake (5.1 servings for
Evaluate the associations age range 34–59 years; men and 5.8 servings for women) reduced ischemic
between fruit and vegetable 14 years of follow-up; stroke risk by 31% vs. the lowest intake. An increase
intake and ischemic stroke risk 366 ischemic stroke in 1 serving of fruits and vegetables significantly
(Nurses’ Health Study and cases; and 38,683 men; reduced risk by 6% (p = 0.01). Citrus fruit including
Health Professionals baseline age range juice, cruciferous vegetables, and green leafy
Follow-up Study; US) [106]. 40–75 years; 8 years of vegetables contributed the most to lower ischemic
follow-up; 204 ischemic stroke risk. Potatoes and legumes were not
stroke cases (multivariate associated with stroke risk.
adjusted).
Type 2 Diabetes (Diabetes)
Wang et al. (2016). 23 cohort articles; For the highest vs. lowest intake, the pooled lower
Evaluate the effects of higher baseline age range multivariate diabetes risk was reduced for total fruits
fruit and vegetables intake on 25–79 years; 4–24 years by 9%, blueberries by 25%, green leafy vegetables
diabetes risk [107]. of follow-up. by 13%, yellow vegetables by 28%, and cruciferous
vegetables by 18% for follow-ups of ≥10 years
(Fig. 3.14).
Li et al. (2014). 10 cohort studies, Evidence of a curve-a-linear association was seen
Clarify and quantify the 434,342 participants; between fruit and green leafy vegetables intake and
potential dose-response 24,013 diabetes cases. diabetes risk, without heterogeneity between studies.
association between the intake Each daily serving reduced diabetes risk for fruit by
of fruit and vegetables and 7% and for vegetables by 10%. For green leafy
diabetes risk [108]. vegetables, each 0.2 serving consumed/day reduced
diabetes risk by 13%.
Xi et al. (2014). 4 sugar sweetened fruit Higher intake of sweetened fruit juice was
Estimate the association juice cohort studies; significantly associated with increased diabetes risk
between sweetened and 100% 191,686 participants; and by 28% (p = 0.02), whereas intake of 100% fruit
fruit juice intake and diabetes 4100% fruit juice cohort juice was not associated with diabetes risk
risk [109]. studies; 137,663 (p = 0.62).
participants; follow-up
ranged from 5.7 to
25 years.
Mamluk et al. (2017). NIH-AARP 401,909 Comparing highest vs. lowest intake, fruit,
Assess the relationship subjects; mean baseline vegetables or green leafy vegetable intake showed
between fruit and vegetable age 62 years; mean no association with diabetes risk for the overall
intake and diabetes risk follow-up of 10.6 years; study. However, independent results from the
(Consortium on Health and 22,782 diabetes cases; NIH-AARP study showed significant lower diabetes
Aging Network of Cohorts in EPIC elderly 20,629 risk for fruit by 5% and for green leafy vegetables by
Europe and the United States subjects; mean baseline 13%.
[CHANCES]) [110]. age 63 years; mean
follow-up of 11.8 years;
1567 diabetes cases
Muraki et al. (2016). 87,739 women and Increased intake of French fries was positively
Determine the effect of potato 40,669 men; >20 years of associated with subsequent diabetes risk. Per each 3
intake on diabetes risk follow-up; 15,362 weekly servings diabetes risk was increased for
(Nurses’ Health Study, Nurses’ diabetes cases baked, boiled, or mashed potatoes by 4% and for
Health Study II, and Health (multivariate adjusted). French fries by 19%, independent of demographic,
Professionals Follow-up lifestyle, dietary factors, and BMI.
Study; US) [111].
Aune et al. (2011). 19 cohort studies. For highest vs. the lowest intake, CRC risk was
Summarize the evidence for lowered for fruits and vegetables combined by 8%,
fruit and vegetable intake and fruit by 10%, and vegetables by 9%. This analysis
CRC risk from cohort studies showed a non-linear association between fruit and
[112]. vegetable intake and lower CRC risk. For fruit, most
of the risk reduction occurred during the first 100 g/
day with higher intakes more modestly decreasing
risk up to 600 g/day. For vegetables, the greatest risk
reduction was between 100 and 200 g/day with
modest lowering up to 500 g/day.
Kunzmann et al. (2016). 57,774 subjects; mean Higher total fruit and vegetable intake was not
Evaluate the association baseline age 63 years; associated with reduced incident or recurrent
between fruit and vegetable mean 12.1 years of adenoma risk overall, but a protective association
intake and CRC development follow-up; 733 CRC was observed for lower risk of multiple adenomas by
by evaluating the risk of cases (multivariate 62%. Higher fruit and vegetable intakes were
incident and recurrent adjusted). associated with a borderline reduced risk of CRC by
colorectal adenoma and CRC 18%, which reached significance among individuals
(Prostate, Lung, Colorectal with high processed meat intakes with a 26% lower
and Ovarian Cancer Screening CRC risk.
Trial; US) [113].
Leenders et al. (2015). 442,961 participants from For highest vs. lowest intake, a lower risk of colon
Examine the effect of separate 10 EU countries; average cancer was observed with higher self-reported
subtypes and variety of fruits of 13 years of follow-up; consumption of fruits and vegetables combined by
and vegetables in the diet on 3370 participants were 13% (p-trend =0.02), but no consistent association
risk of colon and rectal cancer diagnosed with colon or was observed for separate consumption of fruits and
(EPIC; EU) [114]. rectal cancer (multivariate vegetables. No associations with risk of rectal cancer
adjusted). were observed.
Breast Cancer
Hui et al. (2013). 6 cohort studies and 6 BC was significantly reduced with high intake of
Summarize the association case-control studies; flavonols by 12% and flavones by 17%. When the
between flavonoids and its 190,000 subjects; 9513 data was stratified by menopausal status, higher
sub-classes (common in fruits cases. flavonols and flavones intake was associated with a
and vegetables) and breast significant reduced risk of BC in post-menopausal
cancer (BC) risk [115]. women but not in pre-menopausal women.
Aune et al. (2012). 15 cohort studies. For the highest vs. lowest intake, BC risk was
Summarize the evidence for an reduced for fruit and vegetables combined by 11%,
association between fruit and fruits by 8%, and vegetables by 1%. A dose-response
vegetable intake and BC [116]. analysis found a lower BC risk per 200 g/day for
fruit by 5%, fruit and vegetables combined by 4%
and no lowering for vegetables.
Aune et al. (2012). 25 cohort studies. For dietary intake studies, only intake of β-carotene
Summarize the evidence of was significantly associated with a reduced BC risk
dietary intake and blood by 5% per 5000 μg/day. However, blood
concentrations of carotenoids concentrations of carotenoids showed more robust
and BC risk [117]. BC risk reductions: total carotenoids reduced risk by
22% per 100 μg total carotenoids/dL; β-carotene
reduced risk by 26% per 50 μg β-carotene/dL;
α-carotene by 18% per 10 μg α-carotene/dL, and
lutein reduced risk by 32% per 25 μg lutein/dL
(Fig. 3.15). Blood concentrations of carotenoids are
more strongly associated with reduced BC risk than
are carotenoids assessed by dietary questionnaires.
Major CVD events (p-trend =.76) Myocardial infarction

Heart failure (p-trend =.51) (p-trend =.36)
1.04
1.02
1
0.98
Hazard Ratios
0.96
0.94
0.92
0.9
0.88
0.86
0.84
0-3.4 3.5-4.4 4.5-5.4 5.5-7.0 >7
Total Potato Intake Frequency
Fig. 3.10 Weekly potato servings and risk major and specific cardiovascular events (adapted from [94])
of specific fruits and vegetables on CHD and CVD with CVD incidence and total daily fruit and
risk [94–97]. A Swedish cohort study (69,313 vegetable consumption [97]. Subjects who con-
adults, mean age 60 years; 13-years of follow-up) sumed ≥9 servings/day of fruits and vegetables had
found that the consumption of 3 to 7 weekly serv- a 40% lower multivariate CVD risk in comparison
ings of potatoes boiled, fried or French fries had no with those consuming <5 servings/day.
significant adverse effects on CVD events, heart
failure or myocardial infarction (Fig. 3.10) [94]. ypertension and Blood Pressure (BP)
H
The US CARDIA Study (2506 adults, mean base- Meta-analyses. Two meta-analyses summarize
line age 25 years; 20 years of follow-up) demon- the findings from observational studies on the
strated that subjects with a higher intake of fruits effect of increased fruit and vegetables intake on
and vegetables as young adults had significantly hypertension risk [98, 99]. A meta-analysis com-
reduced coronary artery calcium in middle-age paring highest vs. lowest fruit and vegetable intake
adulthood [95]. The pooled analysis of the Nurses’ (3 cohort, 2 case-control, and 20 cross-sectional
Health and Health Professionals Follow-Up Studies studies; 334,468 subjects) found a lower risk of
(113,000 men and women; mean baseline age hypertension for total fruits and vegetables by
50 years; 22–24 years of follow-up) observed that 19%, for fruits by 27%, and for vegetables by 3%
participants in the highest intake quintile of fruits [98]. A dose response meta-analysis (9 cohorts;
and vegetables (excluding potatoes, legumes, and 185,676 subjects) found that the risk of hyperten-
fruit juices) had a 17% lower adjusted CHD risk sion was decreased by 1.9% for each daily fruit
[96]. Specifically, the consumption of 1–1.5 serving and by 1.2% for each daily serving of total
servings/day of green leafy vegetables, and
fruits and vegetables [99]. Also, high intake was
β-carotene and vitamin C rich fruits and vegetables associated with a lower risk of hypertension for
was associated with a significant 15–22% lower fruits by 13% and vegetables by 12%.
adjusted CHD risk (Fig. 3.11). The PREDIMED Specific Studies. Several prospective studies
trial (7216 elderly men and women with high CVD and one RCT show the effects of increased fruits
risk; mean baseline age 67 years; up to 7-years and vegetables on hypertension and blood pres-
follow-up) found a significant inverse association sure risk [100–102]. Two pooled analyses from
Green leafy vegetables (p-trend <.0001) Beta-carotene-rich F/V (p-trend =.002)

Vitamin C-rich F/V (p-trend =.003)
1.05
Relative Risk of CHD 1
0.95
0.9
0.85
0.8
0.75
0.7
0 -.24 .12-.50 .21-.72 .43-1.05 .86 -1.72
# Weekly Servings
Fig. 3.11 Risk of coronary heart disease (CHD) by fruit and vegetable (F/V) variety and number of weekly servings
based on multivariate pooled data from the Nurses’ Health Study and Health Professionals Follow-up Study (adapted
from [96])
Carrots (p-trend =.003)
Tofu or Soybeans (p-trend =.001
Broccoli (p-trend <.001)
Avocados (p-trend <.001)
Apples or Pears (p-trend <.001)
Raisins or Grapes (p-trend <.001)
-14 -12 -10 -8 -6 -4 -2 0

Hypertension Risk (%)
Fig. 3.12 Pooled hypertension risk reduction for individual fruits and vegetables from the Nurses’ Health Study and
Health Professionals Follow-up Study (≥ 4 weekly servings vs. ≤ 1 monthly serving) (adapted from [100])
the US Nurses’ Health Studies and Health [100, 101]. Also, higher intake (≥4 weekly serv-
Professionals Follow-up Study (187,453 ings) of broccoli, carrots, avocado, tofu or soy-
participants; mean baseline age range 43–55 years; bean, apples, pears and raisins significantly
>20 years of follow-up) observed that higher lowered hypertension risk (Fig. 3.12) [100].
intake (≥4 weekly servings) lowered hyperten- Potato products significantly increased hyperten-
sion risk for fruits by 8% and vegetables by 5% sion risk for baked or mashed potatoes by 11%
20
15
10
0
Baked, boiled or mashed French fries Potato chips
(p-trend =.05) (p-trend <.001) (p-trend =.98)
-5
Type of Potato Product
Fig. 3.13 Pooled hypertension risk for potato products from the Nurses’ Health Study and Health Professionals
Follow-up Study (≥4 weekly servings vs. ≤1 monthly serving) (adapted from 101])
and French fries by 17% whereas potato chips >65 years of age compared to men with the low-
did not increase the risk for hypertension est intake [104]. In a Dutch population cohort
(Fig. 3.13) [101]. A 1997 RCT (154 subjects in (20,069 participants; mean baseline age 42 years;
each group; mean age approx. 45 years; mean 10 years of follow-up) higher intake of white
systolic and diastolic BPs were 131 and 85 mm fruits and vegetables (55% apples and pears) was
Hg; 8-weeks) found that 8.5 daily fruit and veg- inversely associated with stroke risk with each
etable servings significantly reduced systolic BP 25 g/day increase associated with a significant
by 2.8 mm Hg and diastolic BP by 1.1 mm Hg 9% lower risk of stroke [105]. The pooled data
more than the control diet with 3.6 servings with from the Nurses’ Health Study and Health
greater BP reduction observed in hypertensive Professionals Follow-up Study found that per-
subjects [102]. sons in the highest quintile of fruit and vegetable
intake (median of 5.1 servings per day among
Ischemic Stroke men and 5.8 servings per day among women)
Meta-analysis. A meta-analysis (20 cohort stud- had a 31% lower ischemic stroke risk compared
ies; 760,629 participants) found a linear dose- with those in the lowest quintile [106]. An incre-
response relationship with reduced stroke risk for ment of one serving per day of fruits or vegeta-
fruits by 32% and for vegetables by 11% per bles (including juice) was associated with a 6%
200 g/day increment with citrus fruits, apple, lower risk of ischemic stroke with cruciferous
pears, and green leafy vegetables being the most vegetables, and citrus fruit being the most
effective in reducing stroke risk [103]. effective.
Specific Studies. Three cohort studies pro-
vide specific insights on the effect of fruits and Diabetes
vegetables on ischemic stroke risk [104–106]. Meta-analyses. Three meta-analyses of cohort
The Health Professionals Follow-up Study studies summarize the effects of increased intake
(43,880 men; mean baseline age 53 years; of fruits and vegetables on diabetes risk [107–
24-year follow-up) found that men with the high- 109]. A 2016 meta-analysis (23 cohort studies;
est intake of flavonoid-rich fruits, such as 4–24 years of follow-up) found that cohorts with
oranges, limes, lemons, apples and pears, had ≥10 years of follow-up had significantly lower
significantly reduced multivariate ischemic multivariate diabetes risk for total fruit, blueberries,
stroke risk in mean by 22%, especially in men green leafy, yellow and cruciferous vegetables intake

0.9
0.8
0.7
0.6
Total fruits Green Leafy Cruciferous Blueberries Yellow
Vegetables vegetables vegetables
Fig. 3.14 Association between specific fruit and vegetables and type 2 diabetes risk based on data from a meta-analysis
of cohort studies with follow-up of ≥10 years (adapted from [107])
(Fig. 3.14) [107]. A dose-response meta-analysis Cancer

(10 cohort studies; 434,342 participants) showed Colorectal cancer (CRC). One meta-analysis of
a curve-a-linear association for total fruit and cohort studies and two specific cohort studies
green leafy vegetables and diabetes with a daily assessed the effect of increased intake of fruits
serving of fruit lowering risk by 7% and 0.2 serv- and vegetables on CRC risk [112–114]. The
ing of green leafy vegetables lowering risk by meta-analysis (19 cohort studies) found modest
13%. A meta-analysis of fruit juice (8 cohort but significant reductions in CRC risk for fruits
studies; 4 on sweetened juices; 4 on 100% juices; and vegetables by 8–10%, for highest vs. lowest
191,686 participants) showed that sweetened intake [112]. This analysis showed a non-linear
fruit juice significantly increased diabetes risk by association between fruit and vegetable intake
28% whereas 100% fruit juices were not associ- and CRC risk. For fruits, most of the risk reduc-
ated with diabetes risk (p = 0.62) [109]. tion occurred during the first 100 g/day with
Specific Studies. Two pooled cohort studies higher intakes more modestly decreasing risk up
provide important insights on specific vegetables to 600 g/day. For vegetables, the greatest risk
and diabetes risk [110, 111]. The US NIH-AARP reduction was between 100 and 200 g/day with
cohort (401,909 participants; mean baseline age modest lowering up to 500 g/day. An analysis of
62 years; 10.6 years of follow-up) found a signifi- the US Prostate, Lung, Colorectal and Ovarian
cant lower diabetes risk for fruit by 5% and green Cancer Screening Trial (57,774 subjects; mean
leafy vegetables by 13% [110]. The pooled data baseline age 63 years; mean 12.1 years of follow-
from US Nurses’ Health Studies and Health up) showed that higher fruit and vegetable intake
Professionals Follow-up Study (87,739 women significantly reduced the risk of multiple adeno-
and 40,669 men; >20 years of follow-up) found mas by 62% and CRC risk among individuals
that 3 weekly servings of French fries signifi- with high intake of processed meat by 26% [113].
cantly increased diabetes risk by 19% compared The 2015 EPIC analysis (442,961 participants;
to a 4% increased risk for baked, boiled or 13 years of follow-up) observed that the partici-
mashed potatoes [111]. pants with the highest intake of total fruits and
0.9
Relative Risk for Breast Cancer
0.8
0.7
0.6
Total carotenoids Beta-carotene Alpha-carotene Lutein
Fig. 3.15 Association between carotenoids blood concentrations and breast cancer risk based on data from a meta-
analysis of 14 prospective studies (per total carotenoid 100 μg/dL; β-carotene 50 μg/dL; α-carotene 10 μg/dL; and lutein
25 μg/dL) (adapted from [117])
vegetables significantly reduced colon cancer better indicator of BC risk than dietary carot-
risk by 13% [114]. enoid intake (Fig. 3.15) [117].
Breast cancer (BC). Three meta-analyses,
primarily from cohort studies, summarize the 3.2.2.4 Healthy Aging
effects of increased fruit and vegetable intake on Table 3.6 summarizes the evidence from studies
BC risk [115–117]. A meta-analysis of flavo- on the effects of increased fruit and vegetable
noids and its subclasses common in fruits and intake on healthy aging with a focus on slowing
vegetables (6 cohort and 6 case-control studies; age-related cognitive performance and delaying
190,000 subjects) found that BC risk was signifi- general aging indicators and frailty [118–127].
cantly reduced with high intake of flavonols by
12% and flavones by 17% [115]. When the data Age-Related Cognitive Performance
was stratified by menopausal status, higher fla- The contribution of increased fruits and vegeta-
vonols and flavones intake was associated with a bles to the daily dietary intake of nutrient and
significant reduced risk of BC in post- phytochemical antioxidants may attenuate brain
menopausal women but not in pre-menopausal oxidative or inflammatory stress or elevate serum
women. A dose-response meta-analysis (15 levels of brain-derived neurotrophic factor
cohort studies) showed a lower BC risk per (BDNF) for possible protective effects on cogni-
200 g/day of fruits by 5% and fruits and vegeta- tive performance [118–123].
bles combined by 4% but no effect for vegeta- Systematic reviews. Several reviews have
bles alone [116]. A meta- analysis comparing analyzed the protective effect of fruit and vegeta-
dietary intake vs. blood concentration of carot- ble intake on age related cognitive function [118,
enoids, a common phytochemical in fruits and 119]. A 2014 systematic review found statisti-
vegetables (25 cohort studies), reported that cally significant benefits of fruits, vegetables, or
higher blood levels of carotenoids were associ- 100% juice consumption for improved cognitive
ated with lower BC risk by 18–32%, and were a function in older adults reported in 17 of 19
Table 3.6 Summary of fruit and vegetable prospective studies and randomized controlled trials (RCTs) on age related
cognitive performance and frailty risk.
Cognitive Function Performance
Systematic Reviews
Lamport et al. (2014). 19 observational studies and 6 17 observational studies and 3
Summarize the association intervention studies. intervention studies reported significant
between polyphenol intake from benefits of fruit, vegetable, or juice
fruit, vegetable, and juice consumption on cognitive performance.
consumption and cognition [118]. The data suggest that chronic intake of
fruits, vegetables, and juices are
beneficial for cognition in healthy older
adults. However, there was a high degree
of variability in cognitive effects
depending on the type of fruit, vegetable
or juice consumed.
Loef and Walach (2012). 6 cohort studies on fruits and Five of the 6 studies that analyzed fruit
Summarize the effects of fruit vegetables with a follow-up of and vegetable consumption separately
and vegetable intake on age 6 months or longer. found that higher consumption of
related cognitive function [119]. vegetables, but not fruit, was associated
with a decreased risk of cognitive
decline or dementia.
In these studies, the vegetables most
associated with slower cognitive decline
included cruciferous vegetables,
legumes, and green leafy vegetables,
particularly cabbage, zucchini, squash,
broccoli, and lettuce, at a daily intake of
3 servings (200 g) a day. The authors
suggest that these beneficial effects
might be due to higher intake of
flavonoids or antioxidants in both fruits
and vegetables, or increased vitamin E in
vegetables, compared to fruits, which
have more vitamin C. Furthermore,
people frequently consume vegetables
with added fats (e.g., oils) which may
aid in absorption of nutrients/
phytochemicals.
RCTs
Neshatdousta et al. (2016). Dose-response Parallel RCT: High-flavonoid intake from fruit and
investigate the link between 154 men and women; aged vegetables intake induced significant
changes in serum brain-derived 26–70 years; intervention diet improvements in cognitive performance
neurotrophic factor (BDNF) and averaged 3 portions of fruit and and increases in serum brain-derived
changes in human cognitive vegetables per day to deliver neurotrophic factor (BDNF) levels
performance following fruit and high-flavonoid (>15 mg/100 g) or (p = <0.001) compared to low flavonoid
vegetable flavonoid intake (UK) low-flavonoid (<5 mg/100 g). fruits and vegetables and habitual diets.
[120]. Intake was increased by 2 portions (Fig. 3.16 and 3.17).
every 6 weeks; control was Flavonoid rich fruits and vegetables
habitual diet; 18-week duration. include: apples, pears, berries, oranges,
Incrementally, 2, 4 and 6 portions peppers, broccoli, onions, cabbage.
of fruit and vegetable intake
delivered 3, 6 and 7 mg/d
(low-flavonoid intervention) and
49, 121 and 198 mg/d (high
flavonoid intervention) total
flavonoids.
(continued)
Kean et al. (2015). Double-blind, Crossover RCT: In healthy older adults, the daily
Investigate the effects of 37 healthy older adults; mean age consumption of flavanoid-rich 100
flavonoid rich orange juice on 67 years; high-flavanoid (305 mg) orange juice significantly improved
cognitive function (US) [121]. 100% orange juice and an global cognitive and executive function
equicaloric low-flavanoid (37 mg) compared to a low-flavanoid control
orange-flavored drink (500 mL) orange juice after 8 weeks.
were consumed daily; 8 weeks/4-
week washout.
Nooyens et al. (2011). 2613 men and women; baseline Higher vegetable intake was associated
Evaluate the effect of habitual age 43–70 years (mean 55 years); with smaller decline in information
fruit and vegetable intake during examined for cognitive function processing speed (p < 0.01) and global
mid-age on cognitive function twice, with a 5-year time interval cognitive function (p = 0.02) over
(Doetinchem Cohort Study; The (multivariate adjusted). 5 years. High intakes of some subgroups
Netherlands) [122]. of vegetables (i.e. cabbage and root
vegetables such as carrots, red beets,
mushrooms) were associated with
smaller decline in cognitive function.
Total intakes of fruits, legumes and
juices were not associated with change
in cognitive function.
Peneau et al. (2011). 2533 subjects; baseline age Higher intakes of fruit and vitamin
Examine the association between 45–60 years; mean age at C–rich fruits and vegetables were
fruit and vegetable intake and evaluation 66 years; 13 years of associated with better verbal memory. In
cognitive performance in a follow-up (multivariate adjusted). contrast, higher intakes of vegetables,
sample of adults and β-carotene-rich fruits and vegetables
(Supplementation with were associated with poorer executive
Antioxidant Vitamins and function. Further research is required to
Minerals 2; France) [123]. better understand the complex
associations between different groups of
fruits and vegetables and specific areas
of age related cognitive decline.
General Aging and Frailty
RCTs from the Ageing and Dietary Intervention Trial (UK)
Neville et al. (2013). Parallel RCT: Subjects consuming ≥5 servings of fruits
Examine the effect of increased 83 participants habitually and vegetables had significantly higher
fruit and vegetable intake on consuming ≤2 portions of fruits biomarkers of micronutrient status. At
measures of muscle strength and and vegetables/day; aged 16 weeks, there was a trend towards a
physical function among healthy, 65–85 years (mean age 71 years); greater improvement in grip strength in
free-living older adults [124]. usual diet (≤2 portions/day) vs. ≥5 the ≥5 portions/day to 2 kg vs. 0.1 kg for
fruit and vegetable portions/day; usual diet group (p = 0.06). Although
16 weeks. increasing F/Vs. intake to ≥5 portions/
day did not significantly improve
lower-extremity physical function, it is
possibly due to the study’s relatively
short duration.
Gibson et al. (2012). Parallel RCT: Antibody binding to pneumococcal
Determined whether increased 83 participants habitually capsular polysaccharides (total IgG)
fruit and vegetable intake consuming ≤2 fruit and vegetable increased more in the 5-fruit and
improves measures of immune portions/day; aged 65–85 years vegetable portion/day group than in the
function [125]. (mean age 71 years); usual diet 2-portion/day group with geometric
(≤2 portions/day) vs. ≥5 fruit and mean ratio of 3.1 vs. 1.7 (p = 0.005).
vegetable portions/day; 16 weeks.
Observational studies
Ribeiro et al. (2016). 432 late middle age African Vegetables other than carrots, salads and
Investigate the effect of fruit and Americans; 6-year follow-up potatoes were associated with improved
vegetable intake on different (multivariate adjusted). grip strength and lower frailty whereas
progressive aging disabilities high sugar sweetened fruit juices
(Longitudinal study) [126]. worsened grip strength and frailty.
Lian et al. (2015). 271 hypertensive patients and 455 Subjects with longer age-adjusted
Examine the potential effect of normotensive controls; aged peripheral leucocyte relative telomere
dietary factors on the association 40–70 years (mean age 57 years) length were associated with higher
between telomere length and habitually consuming ≤2 portions vegetable intake (p = 0.01). Individuals
hypertension risk (community- of fruits and vegetables/day with longer age-adjusted median relative
based case-control study; China) (multivariate adjusted). telomere length were 30% less likely to
[127]. have hypertension.
observational studies and 3 of 6 intervention tri- cognitive performance and BDNF compared to
als [118]. However, there was a high degree of the usual low fruit and vegetable diet or low fla-
variability in cognitive effects depending on the vonoid fruits and vegetables (Figs. 3.16 and 3.17)
type of fruit, vegetable or juice consumed. A [120]. A cross-over RCT (37 healthy adults;
2012 systematic review (six cohort studies; mean age 67 years; 8 weeks) showed signifi-
≥6 months of follow-up) showed in five of six cantly improved global cognitive and executive
studies that higher consumption of vegetables, function with high flavonoid orange juice
but not fruit, was associated with a decreased (305 mg) compared with low flavonoid orange
risk of cognitive decline or dementia [119]. In juice (37 mg) [121]. Two European cohort stud-
these studies, the vegetables most associated ies showed somewhat inconsistent findings for
with slower cognitive decline included crucifer- fruits and vegetables on cognitive performance
ous vegetables, legumes, and green leafy vegeta- [122, 123]. The Dutch Doetinchem Cohort Study
bles; particularly cabbage, zucchini, squash, (2613 adults; mean age 55 years; 5 years of fol-
broccoli, and lettuce, at a daily intake of three low-up) reported that higher vegetable intake,
servings (200 g) a day. The effect of vegetables especially from root vegetables, such as carrots,
may be associated with the fact that people fre- red beets and mushrooms, significantly slowed
quently consume vegetables with added healthy the decline in information processing speed and
oils, which aid in absorption of fat soluble global cognitive function [122]. A French cohort
antioxidants such as vitamins A and E, and study (2533 subjects; mean age at evaluation
carotenoids. 66 years; 13 years of follow-up) found that higher
Specific Studies. Four studies are representa- intake of fruit and vitamin C rich fruits and veg-
tive of the effects of increased fruit and vegetable etables was associated with better verbal memory
intake on cognitive function [120–123]. Two whereas higher intake of vegetables and
RCTs on flavonoid-rich fruits and vegetables or β-carotene rich fruits and vegetables was associ-
their juices show them to improve cognitive ated with poorer executive function [123].
performance in aging subjects [120, 121]. One
parallel RCT (154 adults; aged 20–70 years;
18 weeks) found that high flavonoid fruits and eneral Aging Indicators
G
vegetables such as berries, oranges, apples, pears, Two RCTs and two observational studies pro-
peppers, or broccoli significantly improved vide representative insights on the effects of
Control Low Flavonoid F/V High Flavonoid F/V

0.25
0.2
0.15
Cognitive Function (z-score)
0.1
0.05
0
Baseline 6 weeks 12 weeks 18 weeks
-0.05
-0.1
-0.15
-0.2
Fig. 3.16 Effect of fruits and vegetables (F/V) flavonoid level on cognitive function after 18 weeks from a randomized
controlled trial with 154 adults (p < 0.001 for high flavonoid F/V at 12 and 18 weeks) (adapted from [120])
Control Low Flavonoids F/V High Flavonoid F/V

450
400
350
300
BDNF (pg/ml)
250
200
150
100
50
0
Baseline 6 12 18
Weeks
Fig. 3.17 Effect of fruits and vegetables (F/V) flavonoid level on serum brain-derived neurotrophic factor (BDNF)
concentration over 18 weeks from a randomized controlled trial with 154 adults (p < 0.001 for high flavonoid F/V after
18 weeks) (adapted from [120])
increased fruit and vegetable intake on general nificantly improved antibody binding to pneu-
aging indicators [124–127]. Two publications mococcal capsular polysaccharides compared to
analyzing the British Ageing and Dietary RCT the intake of ≤2 portions [125]. A longitudinal
(84 subjects; mean age 71 years; 16 weeks) study (432 late middle age African Americans;
found that subjects consuming ≥5 fruit and veg- 6-year follow-up) showed that the increased
etable portions daily showed directionally intake of vegetables (exclusive of carrots, salads
improved grip strength (p = .06) [124] and sig- and potatoes) was associated with improved grip
strength and lower frailty whereas increased study (2820 men and 2950 women; 6.5 years of
sugar sweetened fruit juices decreased grip follow-up) showed that a bean-free diet signifi-
strength and increased frailty [126]. A Chinese cantly increased the risk of all-cause mortality by
community-based case-control study (272 98% in women [131]. The 2004 Food Habits in
hypertensive patients and 455 normotensive con- Later Life Study longitudinal study (785 adults
trols; mean age 57 years) observed that subjects from Japan, Sweden, Greece and Australia; aged
with higher vegetable intake had significantly 70+ years; 7 years of follow-up) found that higher
longer leucocyte telomere length and a 30% legume intake was the most protective dietary
lower risk of hypertension [127]. predictor of survival for the elderly subjects,
regardless of their ethnicity [132]. Specifically,
there was a 7–8% lower risk of all-cause mortal-
3.2.3 Legumes ity for each 20 g/day of legumes consumed, inde-
pendent of other mortality risk factors.
3.2.3.1 Background
Legumes, including pulses (e.g., pinto beans, 3.2.3.3 Chronic Diseases
split peas, lentils, chickpeas) and soybeans, are
rich in fiber, protein, B vitamins, iron, calcium ardiovascular Disease (CVD) and Stroke
C
and potassium and bioactive phytochemicals Systematic Reviews and Meta-analyses. Six
such as phenolics, saponins, and isoflavones, meta-analyses of prospective cohort studies and
which are especially concentrated in soy foods RCTs summarize the effects of increased legume
[16, 128]. Legumes are often consumed as a intake from soy, dietary pulses, and soy protein on
lower energy dense, lower saturated fat, and CVD, ischemic heart disease (IHD), blood lipid
higher fiber meat or milk replacer. Their intake profiles and stroke risk [133–138]. A 2017 meta-
has been in decline with the global shift to analysis (11 cohort studies; 367,000 subjects)
Western-style diets [129]. Legumes are infre- found that increased legume intake reduced the
quently consumed by North Americans and risk of both CVD and CHD by 10% but had no
northern Europeans, with <8% of Americans substantial effect on stroke [133]. Also, a 2014
consuming them on any given day. Also, between meta-analysis (5 IHD cohort studies, 199,000
the 1960s and 1990s, legume intake decreased by subjects; and 6 stroke studies, 255,000 subjects)
40% in India and by 24% in Mexico. Table 3.7 showed that 4 weekly servings of legumes reduced
summarizes the effects of increased intake of IHD risk by 14% but had no effect on stroke risk
legumes, both dietary pulses and soy, on mortal- [136]. A 2015 meta-analysis of soy foods (35
ity, chronic disease and age- related cognitive RCTs; 2670 subjects; 4 weeks to 1 year) demon-
function [130–154]. strated that increased intake of soy products sig-
nificantly reduced total and LDL cholesterol, and
3.2.3.2 Mortality Risk triglycerides, and increased HDL cholesterol
Legumes have been associated with longevity [134]. The LDL cholesterol lower effects were
and are important traditional food staples in many greater in hypercholesterolemic patients by
countries known for healthier diets; soy, tofu, 7.5 mg/dL compared to 3.0 mg/dL in healthy sub-
natto, and miso in Japan, brown beans and peas in jects. Also, the LDL cholesterol lowering effects
Sweden, and lentils, chickpeas and white beans were stronger for soy food products such as soy
in the Mediterranean countries [129]. Three pro- milk, soy beans or nuts than for soy extract sup-
spective studies frame-up the effects of high plements. A 2011 meta-analysis of soy protein
legume intake on reduced mortality risk [130– (43 RCTs) found that 30 g of soy protein/day sig-
132]. The Iranian Golestan Cohort Study (42,403 nificantly lowered LDL cholesterol by up to 5.5%
adults; 11 years of follow-up) found that increased and triglycerides by 10.7%, and increased HDL
legume intake significantly reduced total cancer cholesterol by 3.2% [138]. A 2014 meta-analysis of
mortality risk by 28% [130]. A Taiwanese cohort dietary pulses (26 RCTs; 1037 subjects; ≥3 weeks)
Table 3.7 Summary of legume (including soy) prospective studies and randomized controlled trial (RCTs) on mortal-
ity and chronic disease risk, age-related cognitive function, and telomere length.
Mortality Risk
Farvid et al. (2017). 42,403 participants; 11 years of The highest vs. lowest quintile of legume
Evaluate the effect of dietary follow-up; 3291 deaths intake was associated with a reduced total
protein source on mortality risk (multivariate adjusted). cancer risk of 28% (p-trend =0.004) and
(The Golestan Cohort Study; gastrointestinal cancer risk by 24% (p-trend
Iran) [130]. =0.05).
Chang et al. (2011). 2820 men and 2950 women; A bean-free diet significantly increased
Evaluate the associations of average baseline mid-40 years; adjusted (1) metabolic syndrome risk by 83%
all-cause mortality and a average follow-up of 6.5 years, in men and 45% in women and (2)
bean-free diets in adults 225 all-cause deaths significantly increased adjusted all-cause
(Taiwan) [131]. (multivariate adjusted). mortality among women by 98% but
insignificantly in men by 28%.
Darmadi-Blackberry et al. 785 elderly; 5 cohort studies The legume food group showed 7% reduction
(2004). from Japan, Sweden, Greece in mortality risk for every 20 g increase in
Assess protective dietary and Australia; up to 7 years of daily intake. Other food groups were not
predictors associated with follow-up (multivariate found to significantly predict survival among
long-lived elderly (Food Habits adjusted). these cohorts.
in Later Life Study) [132].
Cardiovascular Disease (CVD) and Stroke Risk
Marventano et al. (2017). 11 cohorts; 367,000 subjects; Compared with lower legume consumption,
Summarize the association 18,475 cases of CVD (7451 the highest category of intake was associated
between dietary legume CHD and 6336 stroke cases); with a decreased risk of 10% in both CVD
consumption and CVD risk, 4.9–26 years of follow-up; and CHD with no or little evidence of
including CHD and stroke intake <1 serving/d to 3–4 heterogeneity and no publication bias. There
[133]. weekly servings. was no significant effect of increased legume
intake on stroke risk.
Tokede et al. (2015). 35 RCTs (50 comparisons);; Intake of soy products resulted in a significant
Examine the effects of soy 2670 subjects (aged reduction in total cholesterol by 5.3 mg/dL,
consumption on the lipid 28–83 years and 82% women); LDL-cholesterol (LDL) concentration by
profiles using published RCTs average intake of soya protein 4.8 mg/dL, triglycerides by 4.9 mg/dL. There
[134]. was 30 g/day (range:14–50 g/ was also a significant increase in serum
day); 4 weeks to 1 year. HDL-cholesterol (HDL) by 1.4 mg/dL. LDL
reductions were greater in
hypercholesterolemic patients by 7.5 mg/dL
than in healthy subjects (3.0 mg/dL). LDL
lowering was stronger for whole soy products
(soy milk, soybeans and nuts) by 11.1 mg/dL
compared to processed soy extracts (3.2 mg/
dL).
Ha et al. (2014). 26 RCTs; 1037 subjects; Diets emphasizing dietary pulse intake at a
Assess the effect of dietary ≥3 weeks duration. median dose of 130 g/day (about 1 serving
pulse intake on established daily) significantly lowered LDL by a mean of
therapeutic lipid targets for 0.17 mmol/L compared with the control.
CVD risk reduction [135].
Afshin et al. (2014). IHD: 5 cohort studies; 198,904 Four weekly 100-g servings of legumes were
Investigate and quantify participants; 6514 events. associated with a 14% lower risk of IHD.
associations of legume intake Stroke: 6 cohort studies; Four weekly 100-g servings of legume intake
with ischemic heart disease 254,628 participants; 6690 were not significantly associated with total
(IHD) and stroke [136]. events. stroke or stroke subtypes,
Bazzano et al. (2011). 10 RCTs compared a non-soy Compared to control diets, pooled mean net
Summarize RCTs evaluating legume diet to control; 268 reduction in total cholesterol was 11.8 mg/dL
the effects of non-soy legume participants; ≥ 3-week and mean lowering of LDL was 8.0 mg/dL
consumption on blood lipids duration. These results indicate that a diet rich in
[137]. legumes other than soy decreases total and
LDL cholesterol.
Anderson et al. (2011). 20 parallel and 23 crossover Compared to control, 30 g/day soy protein
Assess the effect of soy protein RCTs. reduced LDL by 5.5% in parallel trials and by
on serum lipoproteins and 4.2% in crossover studies. In parallel trials soy
CHD risk [138]. increased HDL by 3.2% (p < 0.007) and
lowered fasting triglycerides by 10.7%
(p < 0.006).
RCTs
Liu et al. (2013). Double-blind, Parallel RCT: Subgroup analysis among 130 pre- and
Examine effects of soy protein 180 post-menopausal Chinese hypertensive women suggested that soy
with isoflavones or isoflavones women; 3-arms: 15 g soy protein and isoflavones significantly reduced
alone on blood pressure (BP) protein and 100 mg isoflavones systolic BP by 4.25% (p = 0.02).
(China) [139]. (Soy group), or 15 g milk
protein and 100 mg isoflavones
(Iso group), or 15 g milk
protein (placebo group);
6 months.
Pittaway et al. (2008). Parallel RCT: 45 free-living Serum total cholesterol and LDL cholesterol
Evaluate the effects of chickpea adults; 728 g chickpeas weekly were 7.7 mg/dL and 7.3 mg/dL lower,
supplementation on ad libitum (4 cans) as part of their habitual respectively, after the chickpea phase
nutrient intake, body weight, diet for 12 weeks (chickpea (p ≤ 0.01), fasting insulin was 0.75 μIU/mL
serum lipids, lipoproteins, and phase), followed by 4 weeks of lower (p = 0.045), and the homeostasis
other metabolic changes habitual diet without chickpeas assessment model of insulin resistance
(Australia) [140]. (usual phase). (HOMA-IR) was 0.21 lower (p = 0.01). In the
chickpea phase, mean dietary fiber intake was
6.8 g/day more and mean polyunsaturated
fatty acid consumption was 2.7% higher.
He et al. (2005). Double-blind Parallel RCT: Compared to control, soy protein significantly
Examine the effect of soybean 302 adults; aged 35–64 years; reduced systolic BP by 4.3 mm Hg (p < 0.001)
protein intake on BP (China) mean baseline systolic BP and diastolic BP by 2.8 mm Hg (p < 0.001).
[141]. 135 mm Hg and diastolic BP For subjects with hypertension the net reduced
85 mm Hg; 40 g isolated systolic and diastolic BP changes were 7.9
soybean protein daily; and 5.3 mm Hg, respectively. For
12 weeks. normotensive subjects, the BP changes were
more modestly reduced by 2.3 and 1.3 mm
Hg.
Cardiometabolic and Type 2 Diabetes Risk Management
Afshin et al. (2014). 2 cohort studies; 100,179 When these studies were pooled, legume
Investigate and quantify the participants. consumption insignificantly reduced diabetes
associations between legume risk by 22%.
intake and diabetes risk [136].
Liu et al. (2011). 24 RCTs; 1518 subjects; Soy consumption did not significantly affect
Evaluate the effects of soy 4–52 weeks; ≤ 40 g soy measures of glycemic control with a mean
intake on measures of glycemic protein/ day. lowering of fasting glucose by 0.7 mg/dL
control [142]. (p = 0.16) and of fasting insulin by 0.2 mg/dL
(p = 0.50). There was significant heterogeneity
in the results of fasting insulin levels and
HOMA-IR.
(continued)
Sievenpiper et al. (2009). 41 RCTs; 1674 participants; Pulses modestly improved glycemic control
Assess the clinical effects of ≥7 days. through a possible insulin-sparing mechanism;
dietary pulses on glycemic intake of chickpeas was most effective for
control [143]. individuals with diabetes with a study duration
of >4 weeks.
Ding et al. (2016). 142,176 women and 21,781 Compared to non-consumers of soy foods,
Evaluate the effect of soy food men; >20 years of follow-up those consuming >1 serving weekly had a
and isoflavones intake on 9185 diabetes cases trend for reduced diabetes risk by 7% (p-trend
diabetes risk in US adults (multivariate adjusted). =0.14) and high intake of total isoflavones
(Nurses’ Health Study I and II significantly reduced diabetes risk by 11%
and the Health Professionals (p- trend =0.009).
Follow-Up Study [144].
Villegas et al. (2008). 64,227 healthy women; mean Consumption of legumes, especially soybeans,
Examine the association baseline age 49 years; average was inversely associated with the risk of
between legume and soy food 4.6 years of follow-up diabetes (Fig. 3.18).
consumption and diabetes risk (multivariate adjusted).
(Shanghai Women’s Health
Study; China) [145].
RCT
Jenkins et al. (2012). Parallel RCT: The dietary pulse supplemented diet lowered
Evaluate the effect of 121 adults with diabetes; mean HbA1c levels by 0.2% more than the
low-glycemic index (GI) diets age 60 years; approx. 50/50 whole-grain wheat supplement diet
and CVD risk in people with men and women; low-GI pulse (p < 0.001). Also, dietary pulses significantly
type 2 diabetes, with a focus on diet: ≥ 1 cup dietary pulses lowered the absolute 10-year CHD risk vs. the
dietary pulses (e.g., beans, daily vs. increased insoluble whole grain wheat diet, primarily because of a
chickpeas or lentils) and fiber rich whole wheat greater effect on lowering of systolic and
whole-grain (Canada) [146]. products; 3 months. diastolic BP (p < 0.001).
Yu et al. (2016). 17 observational studies (13 Higher intake of soy isoflavone reduced CRC
Assess the association between case-control and 4 cohort risk by 22% (p = 0.024). Based on subgroup
soy foods isoflavone intake and studies); foods included soy, analyses, a significant CRC protective effect
CRC risk [147]. soy foods, soybeans, tofu, soy was observed with soy foods and in Asian
protein, miso, and natto. populations both by 21%.
Zhu et al. (2015). 14 cohort studies; 1,903,459 Higher legume consumption was associated
Investigate the association participants; 12,261 cases; with a decreased risk of CRC by 9%.
between dietary legume beans and soybeans. Subgroup analyses suggested that higher
consumption and risk of CRC legume consumption was inversely associated
[148]. with CRC risk in Asians by 18% and soybean
intake was associated with a decreased risk of
CRC by 15%.
Yang et al. (2009). 68,412 women; mean baseline Each 5-g/day increment intake of soy foods as
Investigate the effect of soy age 52 years; mean follow-up assessed by dry weight [equivalent to 1 oz.
food intake on CRC risk of 6.4 years; 321 CRC cases (28.35 g) tofu/day] was associated with an 8%
(Shanghai Women’s Health (multivariate adjusted). lower CRC risk. Women in the highest tertile
Study; China) [149]. of intake had a risk reduction by 33%
compared with those in the lowest tertile
(p-trend =0.008) (Fig. 3.19). This inverse
association was primarily for postmenopausal
women. Similar results were also found for
intakes of soy protein and isoflavones.
Breast Cancer (BC)
Chen et al. (2014). 10 cohort studies (4 studies in Pooled data from Asian country studies found
Systematically explore the Asian countries and 6 studies in soy isoflavones lowered BC risk in both
association between soy Western countries) and 21 pre- and post-menopausal women by 41%.
isoflavone from soy food and case-control studies (13 studies However, data on post-menopausal women
protein intake and BC risk for in Asia, and 8 studies in from Western countries suggested that soy
pre- and post-menopausal Western countries). isoflavone intake had a marginally significant
women [150]. protective effect by 8%.
Xie et al. (2013). 22 studies (7 cohort and 15 Higher soy isoflavone intake significantly
Examine the association case-control designs); reduced the BC risk in Asian women by 32%
between soy isoflavones intake menopausal status in 14 compared to an insignificant reduction of 2%
and BC risk by meta-analysis studies; 9 studies from Asian in Western women. Further analysis found that
[151]. populations and 5 studies from the intake of isoflavones reduced BC risk by
Western populations. 54% in postmenopausal Asian women and
37% in premenopausal women. The lack of
significant effects in the studies of Western
women appears to be related to their relatively
lower intake of soy products and isoflavones.
Age-related Cognitive Function
Prospective Study
Chen et al. (2012). Population-based, prospective There was an inverse association with
Investigate the association nested case-control study; 6911 cognitive decline in subjects frequently eating
between dietary habits and illiterate subjects; 76% women; vegetables by 34% or legumes by 22%
declines in cognitive function mean baseline age 83 years; (p < 0.0001 for both).
(Chinese Longitudinal Health 3 years of follow-up
Longevity Study) [152]. (multivariate adjusted).
RCTs
St John et al. (2014). Double-blind, Parallel RCT: After 2.5 years, there was no significant
Determine associations of soy 350 healthy postmenopausal difference in global cognition performance
protein and urine excretion of women; mean age 61 years, compared to milk protein control (p = 0.39). A
isoflavonoids with cognitive daily 25 g of isoflavone-rich secondary analysis, indicated that soy
performance (Women’s soy protein or milk protein- isoflavones decreased general intelligence by
Isoflavone Soy Health Clinical matched placebo; 2.5 years. >4 age-associated years compared to the milk
trial; US) [153]. control group.
Kreijkamp-Kaspers et al. Double-blind, Parallel RCT: After 1 year, cognitive function, bone mineral
(2004). 202 healthy postmenopausal density, or plasma lipids did not differ
Investigate whether soy protein women; mean baseline age significantly between the groups.
with isoflavones improves 67 years; 25.6 g of soy protein
cognitive function, bone containing 99 mg of isoflavones
mineral density, and plasma vs. milk protein daily; 1 year.
lipids in postmenopausal
women (The Netherlands)
[154].
found that each daily serving of pulses signifi- Specific RCTs. Three RCTs summarize car-
cantly lowered LDL cholesterol by 0.17 mmol/L diometabolic protective effects for soy products
vs. control food [135]. A 2011 meta-analysis of and protein and chickpeas [139–141]. The lipo-
non-soy legumes (10 RCTs; 268 participants; protein lowering effects of soy products is well
≥3 weeks) showed similar total and LDL choles- established [134, 138], and two double-blind
terol lowering effects as seen for soy products RCTs demonstrated that soy products and protein
[137]. are also effective in reducing systolic and
diastolic BP, especially for hypertensive patients women and 21,781 men; >20 years of follow-up)
over ≥12 weeks [139, 141]. The consumption of showed that the high consumption of soy foods
chickpeas (4 cans per week) was shown to sig- such as tofu or soy milk insignificantly lowered
nificantly lower total and LDL cholesterol, fast- diabetes risk by 7%, but higher intake of isofla-
ing insulin and insulin resistance (HOMA-IR) vones significantly lowered risk by 11% [144].
over 12 weeks [140]. The 2008 Shanghai Women’s Health Study
(64,227; 4.6 years of follow-up) found that Chinese
iabetes Risk and Management
D women with a high intake of pulses (e.g., beans,
Meta-analyses. Several meta-analyses of pro- lentils, and peas), soybeans and soymilk had sig-
spective studies and RCTs show beneficial effects nificantly lower diabetes risk (Fig. 3.18) [145]. A
for legumes on reducing diabetes risk and man- 2012 Canadian RCT (121 people with diabetes;
aging diabetes related health effects [136, 142, mean age 60 years; 3 months) found that dietary
143]. A 2014 meta-analysis of legumes (4 cohort pulses diets (≥1 cup daily of beans, chickpeas or
studies; 100,179 participants; 4 weekly100-g lentils) significantly reduced HbA1c values by
servings) showed high total legume intake low- 0.2% and lowered absolute 10-year CHD risk, and
ered diabetes risk by 22%, which did not reach systolic and d iastolic BP compared to high wheat
statistical significance after multivariate adjust- fiber diets [146].
ments [136]. Two meta-analyses of RCTs showed
that pulses and soy food intake have relatively Cancer
modest effects on lowering fasting blood glucose Four meta-analyses and one large cohort study
and insulin, and insulin resistance (HOMA-IR) provide insights on the effects of legumes on
compared to control diets [142, 143]. colorectal cancer (CRC) and breast cancer (BC)
Prospective Cohort Studies. Two cohort stud- risk [147–151].
ies in the US and China and one RCT provide Colorectal Cancer (CRC) Risk. Two meta-
insights on the effects of legumes (dietary pulses analyses evaluated the effects of soy foods/isofla-
and soy) on diabetes risk [144–146]. The pooled vone intake and total legumes on CRC risk and a
data from a 2016 Nurses’ Health Studies and Chinese cohort study provided insights on the
Health Professionals Follow-up Study (142,176 effects of legumes on CRC risk [147–149].
Soybeans Beans, lentils, peas Soy milk

1.1
0.9
0.8
0.7
0.6
0.5
0.4
Q-1 Q-2 Q-3 Q-4 Q-5
Legume Intake Quintile *
Fig. 3.18 Association between legumes and soy milk intake and type 2 diabetes risk in Chinese women from the
Shanghai Women’s Health Study (p < 0.0001 for all) (adapted from [145]). *Quintile 5: Beans, lentils and peas (37.1 g/
day) and soybeans (32 g/day); Quintile 4: soy milk (214 mL/day).
A 2016 meta-analysis (13 case-control and 4 isoflavone intake significantly reduced BC risk
cohort studies) found that higher soy foods/iso- in all Asian women by 32% compared to an
flavones intake was associated with a significant insignificant reduction of 2% in Western women
22% lower CRC risk [147]. A 2015 meta-analy- [151]. Further analysis found that the higher
sis of total legume intake (14 cohort studies; intake of soy isoflavones reduced BC risk in
1,903,459 participants) found that increased Asian postmenopausal women by 54%. This
legume intake lowered CRC risk by 9%; sub- apparent discrepancy between Asian and
group analysis showed a doubling of the CRC Western women is most likely the result of an
risk lowering to 18% in Asians populations and overall lower soy intake and frequency of con-
for soybean intake [148]. The Shanghai Women’s sumption in Western women compared to Asian
Health Study (68,412 women; mean baseline age women.
52 years; mean 6.4 years of follow-up) found that
each daily 1 ounce intake of tofu reduced CRC 3.2.3.4 Age-Related Cognitive
risk by 8% and higher soy products intake low- Function
ered CRC risk in postmenopausal women by The effect of increased total legume and soy
33% (Fig. 3.19) [149]. intake on age-related cognitive function is sum-
Breast Cancer (BC) risk. Two meta-analy- marized by one cohort study and two RCTs in
ses provide insights on the effects of soy foods or postmenopausal women [152–154]. The Chinese
protein sources on BC risk and recurrence in BC Longitudinal Health Longevity Study (6911 illit-
survivors [150, 151]. A 2014 meta-analysis (10 erate older adults; mean baseline age 83 years;
cohort studies and 20 case-control or cross- 76% women; 3 years of follow-up) observed that
sectional studies) found that increased isofla- frequent consumption of legumes significantly
vones from soy foods or proteins lowered BC reduced cognitive decline by 22% [152]. In post-
risk in both pre- and post-menopausal Asian menopausal women, two large double-blind
women by 41% whereas in Western women RCTs (202–350 women; mean baseline age
there was a marginally significant reduction by 61–67 years; isoflavone rich soy protein vs. milk
8% [150]. A 2013 meta-analysis (7 cohort and protein; 1–2.5 years of duration) found that the
15 case-control designs) showed that higher soy long-term intake of isoflavonoid rich soy protein
Postmenopausal Women
Soy foods (p-trend =.008) Soy protein (p-trend =.007)
Isoflavones (p-trend =.06)
1.1
1
Relative Risk for CRC
0.9
0.8
0.7
0.6
0.5
1 2 3
Tertiles of Intake
Fig. 3.19 Soy intake and colorectal cancer (CRC) risk from the Shanghai Women’s Health Study (adapted from [149])
did not improve cognitive performance compared 162, 166]. A 2016 dose-response meta-analysis
to milk protein [153, 154]. (20 cohort studies; 819,448 participants) found
per daily 28 g nut intake reduced mortality risk
for all-cause by 22%, respiratory disease by 52%,
3.2.4 Nuts and diabetes by 39% [161]. This analysis also
estimated that >4 million global deaths are attrib-
3.2.4.1 Background utable to daily nut intake below 20 g. Another
The consumption of a daily handful of nuts 2016 meta-analysis (20 cohort studies; 467,389
(approx. 45–60 g/day) can help to prevent or participants; primarily from North America and
reduce weight gain, cardiometabolic dysfunction Europe) showed that 4 weekly servings of nuts
and related chronic diseases, especially if eaten reduced mortality risk from all-cause by 19%,
as a replacement for unhealthful foods and snacks CVD by 28% and sudden cardiac death by 75%
[155, 156]. Nuts are compositionally low in satu- [162]. A 2015 meta-analysis (4 cohort studies)
rated fat and sugar, rich in unsaturated fats, pro- found that each daily serving of nuts reduced all-
tein and fiber, and contain a variety of healthy cause mortality by 17% [166].
vitamins, minerals, and phytonutrients including Specific Cohort Studies. Five specific cohort
carotenoids, polyphenols, and phytosterols [157]. studies provide examples of the effect of increased
Although nuts are energy dense (6 kcal/g), many nut intake on all-cause and disease specific mor-
nuts such as almonds have 25% lower metaboliz- tality risk [168–175]. The Health Professionals
able energy than estimated energy values used for Follow-up Study observed that the consumption
nutrition labeling, which is one of the main rea- of ≥5 weekly nut servings reduced after diagnosis
sons that nuts are not associated with the weight prostate cancer mortality risk by 34% [168]. A
gain risk concern of other energy dense foods large study of primarily peanut consuming
[158]. In the US >60% of men and women do not Southern Americans and Shanghai Chinese
consume any nuts on a given day and 14% of men (71,764 participants; 5.4–12.2 years of follow-up)
and 12% of women consume ≥1.5 ounces of found that total mortality risk was significantly
nuts/daily [159]. Frequent nut consumption has reduced in the Americans by 21% and in Chinese
been consistently associated with lower mortal- by 17%, which was mainly due to lower CVD
ity, CVD or weight gain risk and normal blood mortality [170]. A 2015 Australian cohort study
lipids profiles, which are all important for healthy (2893 participants; baseline age > 49 years;
aging [160]. Table 3.8 summarizes the findings 15 years of follow-up) showed that increased nut
from observational studies and RCTs on the intake by women had a larger lowering effect on
effects of increased nut intake on mortality, all-cause and disease specific mortality risk than
chronic disease risk and biomarkers, age-related for men [171]. The Physicians Health Study
cognitive function and telomere length (20,742 male physicians; mean baseline age
[161–194]. 67 years; mean 9.6 years of follow-up) found that
≥5 weekly servings of nuts lowered multivariate
all-cause mortality risk by 26% vs. <1 monthly
3.2.4.2 Mortality Risk serving [172]. Also, the pooled data from the
Nurses’ Health Studies and Health Professionals
rospective Cohort Studies
P Follow-up Study (120,000 men and women;
Meta-analyses. Three systematic reviews and/or 24–30 years of follow-up) observed that daily nut
meta-analyses of prospective cohort studies intake reduced total mortality by 20%. Disease
provide insights on the effects of nut intake on specific mortality risk preventative effects are
all-cause and/or disease specific morality [161, summarized in Fig. 3.20 [173].
Table 3.8 Summary of nut prospective studies and randomized controlled trials (RCTs) on mortality, chronic disease
risk and biomarkers, and age-related cognitive function.
Mortality and Chronic Disease Risk
Aune et al. (2016). 20 cohort studies (29 Per 28 g/day nut intake, there was
Conduct a systematic review and publications); 819,448 reduced risk for CHD by 29%, stroke by
dose-response meta-analysis of nut participants; 12,331 CHD cases, 7%, CVD by 21%, total cancer by 15%,
consumption and risk of CVD, 9272 stroke cases, 18,655 CV mortality for all-cause by 22%,
stroke, total cancer, and all-cause disease cases, 18,490 cancer respiratory disease by 52% and diabetes
and cause-specific mortality [161]. cases, and 85,870 deaths. by 39%. The results were similar for tree
nuts and peanuts. An estimated 4.4
million annual premature deaths in
America, Europe, Southeast Asia, and
the Western Pacific may be attributable
to a nut intake below 20 g/day.
Mayhew et al. (2016). 20 cohort studies; 467,389 High nut intake significantly lowered risk
Systematically review the literature participants; 13,226 CVD of mortality for all-cause by 19%, total
and quantify associations between outcomes including 10,120 deaths CVD by 27%, CHD by 30% and sudden
nut consumption and CVD/CHD from CVD; primarily based on cardiac death by 47%, as well as a lower
outcomes and all-cause mortality North American and European disease risk for total CVD by 44% and
[162]. studies. CHD by 34% vs. low nut intake. Also, 4
weekly servings of nut intake (similar to
the DASH diet) was associated with
significantly lower risk for all-cause
mortality by 19% and sudden cardiac
death by 75%; plus, lower risk of total
CVD by 28% and non-fatal CHD by
19%.
Wu et al. (2015) 36 observational studies, 30,708 High nut intake significantly decreased
Clarify the association between nut patients; 4.6–30 years of risk for total cancer by 15%. colorectal
consumption and risk of cancer or follow-up. cancer by 24%, endometrial cancer by
diabetes [163]. 42%, and pancreatic cancer by 32% vs.
very low or no nut intake. No significant
association was found with other cancers
or diabetes.
Guo et al. (2015). 9 cohort studies (3 hypertension Compared to never/rare consumers, the
Assess the effect of the intake of and 6 diabetes). consumption of >2 weekly servings of
nuts on hypertension and diabetes nuts significantly lowered hypertension
risk [164]. risk by 8% but there was an insignificant
reduction of diabetes risk by 2%.
Zhang et al. (2015). 6 cohort studies; 476,181 High nut intake reduced stroke risk by
Evaluate the effect of nuts on stroke participants. 10% (insignificant trend), but a subgroup
risk [165]. analysis showed a significant 12% lower
stroke risk in women,
Luo et al. 2014). Assess the relation 18 cohort studies; 12,655 Each incremental serving of nuts/day
between nut intake and incidence of diabetes, 8862 CVD, 6623 reduced risk for ischemic heart disease
diabetes, CVD, and all-cause ischemic heart disease, 6487 by 28%, CVD by 29%, and all-cause
mortality [166]. stroke, and 48,818 mortality mortality by 17%. The reduction in
cases. diabetes risk was insignificant after
adjusting for BMI.
(continued)
Zhou et al. (2014). Investigate the 23 prospective cohort studies; The consumption of each 1 serving of
association between nut 744,830 participants. nuts/day was significantly associated
consumption and risk of coronary with lower coronary artery disease by
artery disease, stroke, hypertension, 19%, and hypertension by 34%.
and diabetes [167]. However, there was no association
between the intake of each 1 serving of
nuts/day and the risk of stroke or
diabetes.
Wang et al. (2016). Evaluate the 47,299 men; 6810 incidences of There was no association between nut
association between nut intake and prostate cancer; 26 years of intake and prostate cancer incidence.
prostate cancer incidence and follow-up (multivariate adjusted). However, patients who consumed ≥5
survivorship (Health Professionals servings of nuts weekly after diagnosis
Follow-up Study; US) [168]. had a 34% lower rate of mortality than
those who infrequently consumed nuts
(<1 serving monthly).
Yang et al. (2016). 75,680 women; mid-age at Women consuming ≥2 servings nuts
Evaluate the association of baseline; >20 years of follow-up; weekly vs. those who rarely consumed
long-term nut intake on colorectal 1503 CRC cases (multivariate nuts had a 13% lower risk of CRC
cancer (CRC) risk (Nurses’ Health adjusted). (p-trend =0.06). There was no
Study II; US) [169]. association with peanut butter.
Luu et al. (2015). Examine the 71,764 participants; median Higher nut intake was significantly
association of nut consumption with follow-up of 5.4 years in the US inversely associated with risk of total
total and cause-specific mortality in and 6.5–12.2 years in China; mortality in all 3 cohorts with lower risk
Americans of African and European 14,440 deaths were identified; in the US cohort by 21% and the
descent who were predominantly of peanuts were the primary nut Shanghai cohort by 17% (highest vs.
low socioeconomic status and in consumed in this study lowest quintiles). This inverse association
Chinese individuals in Shanghai, (multivariate adjusted). was predominantly driven by CVD
China (Southeastern US -Southern mortality. There was significantly
Community Cohort Study, Shanghai reduced ischemic heart disease in the US
Women’s Health Study and the cohort by 38% and in the Shanghai
Shanghai Men’s Health Study) cohort by 30% and stroke risk by 23% in
[170]. both populations (highest vs. lowest
quintile).
Gopinath et al. (2015). Assess the 2893 participants; baseline Higher nut consumption significantly
association between nut intake and age ≥ 49 years; 15 years of reduced mortality risk for all-cause by
risk of total mortality and mortality follow-up; 1004 deaths 24%, for CVD by 24% and IHD by 23%.
from CVD (Australia) [171]. (multivariate adjusted). Associations were more marked in
women compared to men. Women with
higher nut intake had reduced risk of
death from all causes by 27%, CVD by
39%, IHD by 34% and stroke mortality
by 49%.
Hshieh et al. (2015). Test the 20,742 male physicians; mean age Compared to <1 serving nuts/month, ≥5
hypothesis that nut consumption is 67 years; mean 9.6 years of servings nuts/week reduced all-cause
inversely associated with the risk of follow-up; 2732 deaths mortality risk by 26% after adjustment
all-cause mortality (Physicians (multivariate adjusted). for age, BMI, alcohol use, smoking,
Health Study; US) [172]. exercise, prevalent diabetes and
hypertension, and intakes of energy,
saturated fat, fruit and vegetables, and
red meat. In a secondary analysis, higher
nut intake was suggestive for lower CVD
mortality and insignificant for CHD and
cancer mortality.
Bao et al. (2013). Examine the 76,464 women, 30 years of Compared with those who did not
association between nut follow-up, 16,200 women died; consume nuts, those who ate ≥7 nut
consumption and subsequent total 42,498 men, 24 years of servings/week had a significantly 20%
and cause-specific mortality follow-up, 11,229 men died lower total mortality risk. Significant
(Nurses’ Health Study and Health (multivariate adjusted). inverse associations were also observed
Professionals Follow-up Study; US) between nut consumption and deaths due
[173]. to cancer, heart disease, and respiratory
disease (Fig. 3.20).
Bao et al. (2013). 75,680 women; 30 years of 28 g serving of nuts 2 times week
Investigate the association between follow-up; 466 confirmed cases of significantly lowered risk of pancreatic
nut consumption and risk of pancreatic cancer (multivariate cancer by 35% (p-trend = 0.007) vs.
pancreatic cancer in a large cohort adjusted). rarely or non-nut consumers. The results
of women (Nurses’ Health Study; did not appreciably change after further
US) [174]. adjustment for BMI and history of
diabetes or adjustments for physical
activity, smoking, and intakes of red
meat, fruits, and vegetables.
Jenab et al. (2004) 478,040 subjects (141,988 men, There was no significant association
Determine the effects of nut and 336,052 women); mean 4.8 years between higher intake of nuts and seeds
seed intake on CRC risk (European of follow-up (multivariate and risk of colon cancer in men and
Prospective Investigation into adjusted). women combined, but a subgroup
Cancer and Nutrition (EPIC) Study; analysis showed that women with high
EU) [175]. nut intake had significantly lower colon
cancer risk by 31%.
RCTs
Mortality Risk
Guasch-Ferre et al. (2013). Parallel RCT: Increased nut intake was associated with
Investigate the association between 7216 men and women; mean a significantly reduced risk of all-cause
frequency of nut consumption and baseline age 67 years; MedDiets mortality (p-trend <0.05). Compared to
mortality in individuals at high supplemented with extra virgin non-consumers, subjects consuming nuts
CVD risk (PREDIMED; Spain) olive oil or nuts vs. lower fat >3 servings/week had a 39% lower
[176]. control diet; median follow-up of mortality risk. A similar protective effect
4.8 years; 323 total deaths, 81 against cardiovascular and cancer
CVD deaths and 130 cancer mortality was observed. Participants
deaths (multivariate adjusted). consuming more nuts at baseline and
allocated to a MedDiet supplemented
with nuts showed a significantly reduced
total mortality risk by 63%.
Chronic Disease Risk
Toledo et al. (2015). Parallel RCT: Compared to the lower fat control, the
Examine the effect of MedDiet on 7216 men and women with high MedDiet plus extra virgin olive oil group
primary prevention of breast cancer CVD risk; mean baseline age significantly reduced BC risk by 68%
(BC) (PREDIMED; Spain) [177]. 68 years; MedDiets supplemented and the MedDiet plus nuts group
with extra virgin olive oil or tree insignificantly lowered BC risk by 41%.
nuts vs. low-fat guidance control;
median follow-up of 4.8 years
Salas-Salvado et al. (2014). Parallel RCT: Compared with the lower fat control
Assess the effect of the MedDiet for 7216 men and women with group, diabetes risk was significantly
the primary prevention of diabetes increased CVD risk; mean reduced in the MedDiet plus extra virgin
(PREDIMED; Spain) [178]. baseline age 67 years; MedDiets olive oil group was significant reduced
supplemented with extra virgin diabetes risk by 40% vs. an
olive oil or tree nuts vs. lower fat insignificantly lowered risk in the
control diet; median follow-up of MedDiet plus tree nuts group by 18%.
4.1 years (multivariate adjusted).
(continued)
Estruch et al. (2013). Parallel RCT: Compared to the lower fat control, risk
Evaluate the effects of MedDiets 7216 men and women with of primary CVD events was significantly
increased CVD risk; mean
supplemented with tree nuts or extra reduced by MedDiets plus tree nuts by
baseline age 67 years; MedDiets
virgin olive oil vs. lower fat diets on 28% (p = 0.02) and MedDiets plus extra
the primary prevention of CVD supplemented with nuts or extra virgin olive oil by 30% (p = 0.009).
(PREDIMED; Spain) [179]. virgin olive oil vs. lower fat
control diet; median follow-up of
4.8 years; 288 primary CVD
events (multivariate adjusted).
Cardiometabolic Disease and Type 2 Diabetes Biomarkers
Systematic Reviews and Meta-Analyses of RCTs
Musa-Veloso et al. (2016). 18 RCT publications; 27 strata; Compared to control diets, almond
Determine the effects of almond approx. 1000 subjects; average supplemented diets significantly reduced
consumption on blood lipid levels daily intake of almonds ranged total cholesterol by 0.153 mmol/L,
[180]. from 20 to 113 g; 4 weeks to LDL-cholesterol by 0.124 mmol/L, and
18 months. triglycerides by 0.067 mmol/L, whereas
HDL-C was insignificantly lowered by
0.017 mmol/L.
Del Gobbo et al. (2015). 61 RCTs; 2582 subjects; Each daily nuts serving significantly
Evaluate the effects of tree nuts on dose-standardized to one 1-oz. lowered the following pooled mean
blood lipids, lipoproteins and other (28.4 g) serving/day walnuts, values: total cholesterol by 4.7 mg/dL,
cardiometabolic outcomes in adults pistachios, macadamia nuts, LDL cholesterol by 4.8 mg/dL, ApoB by
aged ≥18 years without CVD [181]. pecans, cashews, almonds, 3.7 mg/dL; and triglycerides by 2.2 mg/
hazelnuts, and Brazil nuts; dL. The dose-response between nut
3–26 weeks. intake and total cholesterol and LDL
cholesterol was nonlinear (p-nonlinearity
<0.001 all) with stronger effects
observed for ≥60 g nuts/day. Significant
heterogeneity was not observed by nut
type. There were insignificant effects of
tree nuts on HDL cholesterol, systolic or
diastolic blood pressure, and
inflammation such as C-reactive protein
(CRP).
Mohammadifard et al. (2015). 21 RCTs; 1652 adults; baseline This analysis indicates that nut intake
Estimate the effect of nut age 18–86 years; 2–16 weeks. leads to a very modest but significant
consumption on blood pressure reduction in systolic BP in participants
(BP) [182]. without diabetes by 1.3 mm Hg
(p = 0.02). Sub-group analyses of
different nut types suggest that pistachios
were the only nut that significantly
reduced systolic BP by 1.8 mm Hg
(p = 0.002).
Blanco Mejia et al. (2014). 49 RCTs; 2226 subjects healthy Tree nut diets modestly lowered
Assess effect of tree nuts on or with dyslipidemia, metabolic triglycerides by 0.06 mmol/L and fasting
metabolic syndrome criteria [183]. syndrome criteria or diabetes; blood glucose by 0.08 mmol/L compared
median nut intake approx. 50 g; with control diets. There was no effect of
median duration 8 weeks. nuts on HDL cholesterol or blood
pressure although there was a trend
toward lower waist circumference.
Banel and Hu (2009). 13 RCTs; 365 subjects; walnuts Compared with control diets, diets
Estimate the effect of walnuts on 10–24% of energy; 4–24 weeks. supplemented with walnuts resulted in a
blood lipids [184]. significantly greater decrease in total
cholesterol by 10 mg/dL and LDL-
cholesterol by 9.2 mg/dL (p = 0.001 for
both) whereas there were insignificant
decreases in HDL cholesterol by 0.2 mg/
dL and triglycerides by 3.9 mg/dL. Other
results show that walnuts provided
significant benefits for antioxidant
capacity and inflammatory markers and
had insignificant effects on body weight
and BMI.
Viguiliouk et al. (2014). 12 RCTs; 450 subjects. Compared to control diets, diets
Assess the effects of tree nuts on emphasizing tree nuts at a median dose
markers of glycemic control in of 56 g/day significantly lowered HbA1c
individuals with diabetes [185]. by 0.07% (p = 0.0003) and fasting
glucose by 0.15 mmol/L (p = 0.03).
Insignificant lowering effects were
observed for fasting insulin and
HOMA-IR with the consumption of tree
nuts.
Flores-Mateo et al. (2013). 28 RCTs, 1806 subjects. Pooled results indicated a nonsignificant
Evaluate the effect of nut intake on lowering of body weight by 0.47 kg,
adiposity measures [186]. BMI by 0.40, and waist circumference
by 1.25 cm. from diets including nuts
compared with control diets.
Xiao et al. (2017). 10 RCTs; 374 subjects Increased nut consumption significantly
Assess the effect of nut improved flow mediated dilation
consumption on vascular (p = 0.001) but subgroup analyses
endothelial function [187] indicated that the benefit was limited to
walnuts.
Intervention Trials
Gulati et al. (2017). Open Label Trial: 50 diabetic Almonds significantly improved waist
Evaluate the effects of almonds on subjects; 27 men and 23 women; circumference, waist-to-height ratio, total
cardiometabolic and diabetes mean age 46 years; mean BMI and LDL-cholesterol, serum
disease management biomarkers in 29; 20% of energy intake from triglycerides, glycosylated hemoglobin,
Asian Indians with type 2 diabetes almonds vs. nut free phase; and hs-CRP in Indian diabetic subjects.
(India) [188]. 24 weeks. This study supports the incorporation of
almonds in a well-balanced healthy diet
for multiple beneficial effects on the
management of glycemic and CVD risk
factors in Asian Indian patients with
diabetes.
Njike et al. (2015). Parallel RCT: The walnut diet significantly improved
Assess the effects of walnut intake 112 subjects at risk for diabetes; diet quality as measured by the Healthy
in subjects at risk for diabetes on walnut-included diet with daily Eating Index 2010, endothelial function,
cardiometabolic health and assess 56 g (366 kcal) of walnuts and a total and LDL cholesterol without
body composition in persons walnut-excluded diet; 6 months. significantly changing BMI, percent
consuming walnuts (US) [189]. body fat, visceral fat, fasting glucose,
glycated hemoglobin, and blood pressure
compared with a walnut-excluded diet.
(continued)
Sauder et al. (2015). Crossover, Controlled Feeding Pistachios supplemented diets
Examine the effects of daily RCT: significantly improved total cholesterol
pistachio consumption on the lipid/ 30 adults with type 2 diabetes and the ratio of total to HDL cholesterol
lipoprotein profile, glycemic (mean glycated hemoglobin and lowered triglycerides and
control, markers of inflammation, 6.2%); mean age 56 years; 50% fructosamine compared to the control
and endothelial function in adults women; pistachios 20% of energy diet. However, pistachio supplemented
with type 2 diabetes (US) [190]. vs. nut free control; 4 weeks on diets did not significantly improve
each diet; washout 2 weeks. fasting glucose and insulin or
inflammatory markers and endothelial
function.
Age-related Cognitive Function
Valls-Pedret et al. (2015). 447 cognitively healthy adults MedDiet plus nuts improved
Examine the effect of MedDiets with increased CVD risk; mean performance above the baseline in
supplemented with nuts or olive oil baseline age 67 years; median memory test (p = 0.04) whereas the
vs. lower fat diets on age-related follow-up 4.1 years (multivariate subjects on MedDiet plus extra virgin
cognitive decline (PREDIMED; adjusted). olive oil performed better in tests of
Spain) [191] frontal (p < 0.003) and global cognition
(p < 0.005). All cognitive tests on the
lower fat control diets were significantly
decreased (p < 0.05).
O’Brien et al. (2014). 16,010 women; mean baseline Women consuming ≥5 servings of nuts/
Examine the effects of long-term age 74 years; 6 years of follow-up week had higher cognitive scores than
nuts intake on cognition in older (multivariate adjusted). non-consumers with a mean
women (The Nurses’ Health Study; improvement by 0.08 standard units
US) [192]. (p-trend = 0.003). This mean difference
of 0.08 is equivalent to a mean 2-year
improved cognitive age.
RCT
Sanchez-Villegas et al. (2011). Parallel RCT: Participants assigned to MeDiet plus nuts
Assess the role of a MedDiet on 243 adults with increased CVD showed a significant lower risk of low
plasma brain-derived neurotrophic risk; mean baseline age 67 years plasma BDNF values (<13 μg/mL) by
factor (BDNF) levels (PREDIMED; from the Navarra center; 3 dietary 78% as compared to the low-fat control
Spain) [193]. interventions: control (low-fat) group. Among participants with prevalent
diet, MedDiet supplemented with depression at baseline, significantly
extra virgin olive oil, or MedDiet higher BDNF levels were found for those
supplemented with tree nuts. assigned to the MedDiet supplemented
Plasma BDNF levels were with nuts.
measured after 3 years of
intervention.
Telomere Length
Lee et al. (2015). 1958 middle-aged and older Higher intake of legumes, nuts, seaweed,
Determine the association between Korean adults from a population- fruits and dairy products and lower
dietary patterns or consumption of based cohort; semi-quantitative consumption of red meat or processed
specific foods and leukocyte food frequency questionnaire; meat and sweetened carbonated
telomere length (LTL) (Korean) 10 years of follow-up beverages were associated with longer
[194]. (multivariate adjusted). LTL (Fig. 3.21).
All-cause (p-trend <.001) Cancer (p-trend =.03)

CVD (p=trend <.001) Heart disease (p-trend <.001)
1
0.95
0.9
Hazard Ratios
0.85
0.8
0.75
0.7
0 <1 1 2-4 >5
Total Nut Weekly Servings
Fig. 3.20 Association between frequency of total nut intake (tree nuts and peanuts) on all-cause and disease specific
mortality risk in US men and women (mean baseline age 60 years; 24–30 years of follow-up) (adapted from [173])
andomized Controlled Trial (RCT)

R tional nut serving significantly lowered coronary
The PREDIMED trial (7216 adults; mean age artery disease (CAD) by 19%, hypertension by
67 years; high CVD risk; MedDiets supple- 34%, and IHD by 28% [165, 166]; (3) higher nut
mented with 30 g mixed nuts (15 g walnuts, 7.5 g intake significantly lowered colorectal cancer
hazelnuts, and 7.5 g almonds); median follow-up (CRC) by 24% [163]; (4) >2 weekly servings sig-
of 4.8 years) found that consuming >3 weekly nificantly lowered hypertension by 8% [164]; and
nut servings resulted in significantly reduced all- (5) higher nut intake significantly lowered stroke
cause mortality risk by 39% with similar effects risk by 12% in women [165].
for CVD and cancer mortality risks [176]. Prospective Studies. Four cohort studies pro-
vide insights on the effects of increased nut intake
3.2.4.3 Chronic Disease Risk on cancer risk [168, 169, 174, 175]. Two cohort
studies support a protective effect of nuts on CRC
Prospective Cohort Studies risk. The 2016 Nurses’ Health Study showed that
Meta-analyses. Seven systematic reviews and/or consuming ≥2 weekly nut servings lowered CRC
meta-analyses of prospective cohort studies pro- risk by 13% in women but there was no CRC
vide insights on the effects of increased nut intake protective effect with the consumption of peanut
on chronic disease risk [161–167]. A 2016 meta- butter [169]. A 2004 EPIC study found that
analysis (20 cohort studies; 467,389 participants) higher intake of nuts reduced CRC risk by 31%
found that each daily 28 g nut intake reduced risk in women but there was no effect for men [175].
for total CVD by 21%, CHD by 29%, stroke by For prostate cancer, the Health Professionals
7%, and total cancer by 15% [161]. A second Follow-up Study (47,299 men; 6810 cases;
2016 meta-analysis (20 cohort studies; 819,448 26 years of follow-up) showed that increased nut
participants) showed 4 weekly nut servings intake did not reduce prostate cancer incidence
reduced risk for CVD by 28% and CHD by 19%, but prostate cancer patients consuming ≥5
primarily based on North American and European weekly servings of nuts after diagnosis had a
studies [162]. The overall findings from five 34% lower mortality risk [168]. For pancreatic
meta-analyses concluded that: (1) higher nut cancer, the Nurses’ Health Study found that 2
intake insignificantly lowered diabetes risk vs. weekly servings of nuts significantly lowered
lower nut consumption [163–167]; (2) each addi- multivariate adjusted risk by 35% [174].
RCTs. Three large PREDIMED trials (7216 men (18 RCTs; 1000 subjects; ranging from 20 to
and women at high CVD risk; mean baseline age 113 g/day; approx. 10–20% of energy; 4 weeks
67–68 years; MedDiets supplemented 30 g mixed to 18 months) found that almond intake signifi-
nuts (15 g walnuts, 7.5 g hazelnuts, and 7.5 g cantly lowered total and LDL-cholesterol, and
almonds) or 1 L weekly of extra virgin olive oil triglycerides without significantly reducing HDL
vs. a lower fat control) found that compared to cholesterol compared to control diets [180]. For
the lower fat control diet, the MedDiets with nuts: walnuts, a 2009 meta-analysis (13 RCTs; 365
(1) significantly lowered CVD risk by 28% [179]; subjects; 10–24% energy; 4–24 weeks) showed
(2) lowered breast cancer risk by 41% [177]; that walnut intake significantly lowered total and
and (3) lowered type 2 diabetes by 18% [178]. LDL-cholesterol, but triglycerides and HDL
cholesterol were insignificantly lowered com-
3.2.4.4 Cardiometabolic Disease pared to control diets [184]. This analysis also
and Type 2 Diabetes showed that walnuts significantly improved anti-
Biomarkers oxidant capacity and lowered inflammatory bio-
Eight systematic reviews and/or meta-analyses of markers without increasing body weight or
RCTs [180–187] and three intervention trials on BMI. A meta-analysis of diabetic individuals (12
almonds, walnuts and pistachios [188–190] pro- RCTs; 450 patients) showed that diets with a
vide insights on the effects of increased nut intake median daily intake of 56 g of tree nuts signifi-
on cardiometabolic disease biomarkers. cantly lowered HbA1c and fasting blood glucose
compared to control diets [185]. Another meta-
ystematic Reviews and Meta-Analyses
S analysis (28 RCTs; 1806 subjects) found that
Eight systematic reviews and/or meta-analyses diets that included tree nuts insignificantly low-
summarize the effects of increased tree nut intake ered body weight, BMI and waist circumference
on cardiovascular ‘biomarkers’ [180–187]. A vs. control nut free diets [186]. For vascular
2015 dose-response meta-analysis of tree nuts endothelial function, a 2017 systematic review
(61 RCTs, 2582 subjects; 3–24 weeks) found and meta-analysis (10 RCTs; 374 subjects) found
that a one ounce (28.4 g) serving had significant that increased nut consumption significantly
lowering effects for tree total and LDL choles- improved flow mediated dilation but subgroup
terol, ApoB and triglycerides compared to con- analyses indicated that the benefit was limited to
trol diets [181]. This analysis showed the total walnuts [187]. This analysis suggests that wal-
and LDL cholesterol lowering effects of tree nuts nuts are more effective in reducing the rate of
was significantly non-linear with stronger lower- vascular aging than other tree nuts.
ing effects demonstrated for intake >60 g/day. In
general, all types of tree nuts had similar lower- pecific Intervention Trials
S
ing effects on blood lipid profiles. There were no Three intervention trials on almonds, walnuts and
significant effects from tree nut intake on HDL pistachios provide insights on their effects on
cholesterol, systolic or diastolic blood pressure, cardiometabolic and diabetes biomarkers [188–
and CRP. Another 2015 meta-analysis (21 RCTs; 190]. For almonds, a 2017 intervention trial (50
1652 subjects) showed that tree nuts modestly type 2 diabetes subjects; 27 men and 23 women;
but significantly lowered mean systolic BP mean age 46 years; mean BMI 29; 24 weeks)
by1.3 mmHg with pistachios having the best sys- showed that a diet with 20% of the energy intake
tolic BP lowering by 1.8 mmHg vs. control diets from almonds significantly improved waist cir-
[182]. A 2014 meta-analysis (49 RCTs; 2226 cumference, waist-to-height ratio, total and LDL-
subjects) reported that median tree nut intake of cholesterol, serum triglycerides, glycosylated
50 g/day modestly improved metabolic syn- hemoglobin, and hs-CRP [188]. For walnuts, a
drome factors including triglycerides, blood 2015 RCT (112 subjects; mean age 55 years; 31
glucose and waist circumference but not men and 81 women; 6 months) compared a
HDL cholesterol or BP compared to control walnut-excluded diet with a walnut-included
diets [183]. For almonds, a 2016 meta-analysis (56 g walnuts (366 kcal)/day) diet, and showed
that the walnut diet significantly improved diet follow-up) found that higher total nut intake was
quality as measured by the Healthy Eating Index associated with better average cognitive status for
2010, endothelial function, and total and LDL all cognitive attributes measured [192]. Women
cholesterol without significantly changing BMI, consuming ≥5 servings of nuts/week had signifi-
percent body fat, visceral fat, fasting glucose, cantly higher cognitive scores equivalent to a
glycated hemoglobin, and BP [189]. For pista- 2-year improved cognitive age vs. non-consumers.
chios, a 2015 cross-over RCT (30 adults with Two PREDIMED RCTs show that MedDiets sup-
type 2 diabetes; mean age 56 years; 50% women; plemented with nuts consumed by older men and
4 weeks) found that a diet with 20% of energy women with high CVD risk showed subjects had
from pistachios significantly improved total cho- improvements in memory and increased levels of
lesterol and the ratio of total to HDL cholesterol plasma brain derived neurotrophic factor (BDNF),
and lowered triglycerides compared to the con- known to improve brain function, compared to
trol diet [190]. However, pistachio supplemented lower fat control diets [191, 193].
diets did not significantly improve fasting glu-
cose and insulin or inflammatory markers and 3.2.4.6 Telomeres
endothelial function, although they did signifi- A Korean prospective study (1958 subject; age 40
cantly lower fructosamine (a measure of long- to >60; 10 years) observed that healthy dietary
term glycemic control in people with diabetes) patterns including nuts, whole grains, seafood,
compared to the control. legumes, vegetables and seaweed were associated
with longer leukocyte telomere length [194]. This
3.2.4.5 Age Related Cognitive study suggests that nuts are uniquely associated
Performance with longer telomere length compared to other
One prospective cohort study and two RCTs pro- healthy food sources (Fig. 3.21). In contrast,
vide indications that increased nut intake helps to higher consumption of red meat or processed
protect against age related cognitive decline meat and sweetened carbonated beverages, which
[191–193]. The Nurses’ Health Study (15,467 are common in the Western dietary pattern, were
women; mean age of 74 years; 15–21 years of associated with shorter telomere length.
Foods Associated with Longer Telomere Length
Coffee
Dairy products
Fruits
Seaweed
Legumes
Nuts
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14

Leukocyte Telomere Length Relative Units
Fig. 3.21 Association between consumption of specific “healthy” foods and leukocyte telomere length in middle-aged
and older adults from a Korean population-based cohort followed for 10 years (adapted from [194])
Conclusions endothelial health, hepatic function, adipo-

The rate and quality of the aging processes is cyte metabolism, visceral adiposity, brain
only partly influenced by genetic and it can neurochemistry and the microbiota ecosys-
be modified by consuming healthy diets tem. For whole-grains, β-glucan-rich oats
overall and specific types of uniquely healthy and barley lower total and LDL-cholesterol
foods. Healthy dietary guidelines generally better than other cereal grains and whole-
recommend eating: 2 1/2 cups of a variety of grain bread tends to be more beneficial than
vegetables/day; 2 cups of fruits, especially white bread in controlling weight gain and
whole fruits/day; 6 servings of total grains at abdominal fat. For fruits and non-starchy
≥3 servings of whole grains/day and ≤3 vegetables, low energy dense and flavonoid
servings of refined grains/day, ≥4 weekly
and/or carotenoid rich varieties including
servings of legumes (dietary pulses or soy), apples, pears, berries, citrus fruits, crucifer-
and/or ≥5 weekly servings of nuts, and limit- ous vegetables, and green le afy vegetables
ing consumption of red or processed meats, are especially associated with improved odds
added saturated and trans-fat, sugar or sodium of healthy aging, cognitive performance and
for improved odds for healthy aging and weight control, and reduced risk of chronic
reduced chronic disease and premature mor- disease and premature death. Legumes
tality risk. Whole plant foods range widely in (dietary pulses or soy) are associated with
their health effects because of their variation reduced weight gain, chronic disease, and
in level and type of fiber, nutrients and phyto- mortality risks. All nuts tend to have similar
chemicals, which can have differential effects effects on managing body weight, and glyce-
on aging, chronic disease risk, cognitive mic, lipoprotein and inflammatory profiles,
function and longevity by their impact on but among nuts walnuts appear to be uniquely
weight regulation, lipoprotein concentrations effective in promoting better vascular endo-
and function, blood pressure, glucose-insulin thelial function such as flow mediated
homeostasis, oxidative stress, inflammation, dilation.
ppendix A: Estimated Range of Energy, Fiber, Nutrient and Phytochemical

A
Composition of Whole or Minimally Processed Plant Foods/100 g Edible
Portion
Components Whole-grains Fresh fruit Dried fruit Vegetables Legumes Nuts/seeds
Nutrients and Wheat, oats, Apples, Dates, dried Potatoes, Lentils, Almonds,
phytochemicals barley, brown pears, figs, spinach, carrots, chickpeas, Brazil nuts,
rice, whole bananas, apricots, peppers, lettuce, split peas, cashews,
grain bread, grapes, cranberries, green beans, black hazelnuts,
cereal, pasta, oranges, raisins and cabbage, onions, beans, macadamias,
rolls, and blueberries, prunes cucumber, pinto pecans,
crackers strawberries, cauliflower, beans, and walnuts,
and avocados mushrooms, and soy beans peanuts,
broccoli sunflower
seeds, and
flaxseed
Energy (kcal) 110–350 30–170 240–310 10–115 85–170 520–700
Protein (g) 2.5–16 0.5–2.0 0.1–3.4 0.2–5.0 5.0–17 7.8–24
Available 23–77 1.0–25 64–82 0.2–25 10–27 12–33
Carbohydrate
(g)
References 109

Fiber (g) 3.5–18 2.0–7.0 5.7–10 1.2–9.5 5.0–11 3.0–27
Total fat (g) 0.9–6.5 0.0–15 0.4–1.4 0.2–1.5 0.2–9.0 46–76
SFAa (g) 0.2–1.0 0.0–2.1 0.0 0.0–0.1 0.1–1.3 4.0–12
MUFAa (g) 0.2–2.0 0.0–9.8 0.0–0.2 0.1–1.0 0.1–2.0 9.0–60
PUFAa (g) 0.3–2.5 0.0–1.8 0.0–0.7 0.0.0.4 0.1–5.0 1.5–47
Folate (μg) 4.0–44 <5.0–61 2–20 8.0–160 50–210 10–230
Tocopherols 0.1–3.0 0.1–1.0 0.1–4.5 0.0–1.7 0.0–1.0 1.0–35
(mg)
Potassium (mg) 40–720 60–500 40–1160 100–680 200–520 360–1050
Calcium (mg) 7.0–50 3.0–25 10–160 5.0–200 20–100 20–265
Magnesium 40–160 3.0–30 5.0–70 3.0–80 40–90 120–400
(mg)
Phytosterols 30–90 1.0–83 — 1.0–54 110–120 70–215
(mg)
Polyphenols 70–100 50–800 — 24–1250 120–6500 130–1820
(mg)
Carotenoids (μg) — 25–6600 0.6–2160 10–20,000 50–600 0.0–1200
Nutrient Database for Standard Reference, Release 27.
https://www.ars.usda.gov/nutrientdata. Accessed 17 February 2015
Services and the Secretary of Agriculture. Part D. Chapter 1: Food and nutrient intakes, and health: current status and
trends. 2015;1–78
Ros E, Hu FB. Consumption of plant seeds and cardiovascular health epidemiological and clinical trial evidence.
Circulation. 2013;128:553–565
USDA. What we eat in America, NHANES 2011–2012, individuals 2 years and over (excluding breast-fed children).
Available: www.ars.usda.gov/nea/bhnrc/fsrg
Slavin JL, Lloyd B. Health benefits of fruits and vegetables. Adv Nutr. 2012;3:506–516
Rebello CJ, Greenway FL, Finley JW. A review of the nutritional value of legumes and their effects on obesity and its
related co-morbidities. Obes Rev. 2014;15:392–407
Gebhardt SE, Thomas RG. Nutritive Value of Foods. 2002; U.S. Department of Agriculture, Agricultural Research
Service, Home and Garden Bulletin 72
Holden JM, Eldridge AL, Beecher GR, et al. Carotenoid content of U.S. foods: an update of the database. J Food Comp
An. 1999;12:169–196
Lu Q-Y, Zhang Y, Wang Y, et al. California Hass avocado: profiling of carotenoids, tocopherol, fatty acid, and fat con-
tent during maturation and from different growing areas. J Agric Food Chem. 2009;57(21):10,408–10,413
Wu X, Beecher GR, Holden JM, et al. Lipophilic and hydrophilic antioxidant capacities of common foods in the United
States. J Agric Food Chem. 2004;52:4026–4037
Dahl WJ, Steward ML. Position of the Academy of Nutrition and Dietetics: health implication of dietary fiber. J Acad
Nutr Diet. 2015;115(11):1861–1870
gov/ Accessed 8.22.2015
http://wholegrainscouncil.org/whole-grains-101/what-counts-as-a-serving. Accessed 12.26.2015
a
SFA (saturated fat), MUFA (monounsaturated fat) and PUFA (polyunsaturated fat)
References 3. WHO/FAO. Joint Expert Consultation on Diet,

Nutrition and the Prevention of Chronic Diseases.
Diet. Nutrition and the Prevention of Chronic
1. Olshansky SJ. Has the rate of human aging already
Diseases. WHO Technical Series. 2003:916.
been modified? Cold Spring Harb Perspect Med.
4. Fontana L, Hu FB. Optimal body weight for health
2015;5. https://doi.org/10.1101/cshperspect.a025965.
and longevity: bridging basic, clinical, and population
2. Olshansky SJ, Hayflick L, Carnes BA. No truth to the
research. Aging Cell. 2014;13:391–400.
fountain of youth. Sci Am. 2008;14:98–102.
5. Newman AB. Is the onset of obesity the same as 21. Satija A, Bhupathiraju SN, Rimm EB, et al. Plant-
aging? Proc Natl Acad U S A. 2015;112(52):E7163. based dietary patterns and incidence of type 2
https://doi.org/10.1073/pnas. 1515367112. diabetes in US men and women: results from
6. Beltran-Sanchez H, Soneji S, Crimmins EM. Past, three prospective cohort studies. PLoS Med.
present, and future of healthy life expectancy. Cold 2016;13(6):e1002039. https://doi.org/10.1371/jour-
Spring Harb Perspect Med. 2015;5. https://doi. nal.pmed.1002039.
org/10.1101/cshperspect.a025957. 22. Micha R, Peñalvo JL, Cudhea F, et al. Association
7. World Health Organization. WHO global status between dietary factors and mortality from heart
report on noncommunicable diseases 2010. Geneva: disease, stroke, and type 2 diabetes in the United
World Health Organization Press; 2010. States. JAMA. 2017;317(9):912–24. https://doi.
8. Avendano M, Kawachi I. Why do Americans have org/10.1001/jama.2017.0947.
shorter life expectancy and worse health than people 23. William PG. Evaluation of the evidence between
in other high-income countries? Annu Rev Public consumption of refined grains and health outcomes.
Health. 2014;35:307–25. Nutr Rev. 2012;70(2):80–99.
9. Rowe JW, Kahn RL. Human aging: usual and suc- 24. Seal CJ, Brownlee IA. Whole-grain foods and
cessful. Science. 1987;237:143–9. https://doi. chronic disease: evidence from epidemiological and
org/10.1126/science.3299702. intervention studies. Proc Nutr Soc. 2015;74:313–9.
10. Micha R, Khatibzadeh S, Shi P, et al. Global, regional 25. Slavin J. Why whole grains are protective: biological
and national consumption of major food groups in mechanisms. Proc Nutr Soc. 2003;62:129–34.
1990 and 2010: a systematic analysis including 266 26. Cho SS, Qi L, Fahey Jr GC, Klurfeld DM.
country-specific nutrition surveys worldwide. BMJ Consumption of cereal fiber, mixtures of whole
Open. 2015;5(9):e008705. https://doi.org/10.1136/ grains and bran, and whole grains and risk reduc-
bmjopen-2015-008705. tion in type 2 diabetes, obesity, and cardiovascular
11. Dietary Guidelines Advisory Committee. Scientific disease. Am J Clin Nutr 2013;98: 594–619.
Report. Advisory Report to the Secretary of Health 27. McGill CR, Fulgoni VL III, Devareddy L. Ten-year
and Human Services and the Secretary of Agriculture. trends in fiber and whole grain intakes and food
Part D. Chapter 1: Food and nutrient intakes, and sources for the United States population: National
health: current status and trends 2015. p. 1–78. Health and Nutrition Examination Survey 2001-
12. http://health.gov/dietaryguidelines/2015/guidelines/ 2010. Forum Nutr. 2015;7:1119–30.
appendix-3/ Accessed June 21, 2016. 28. Aune D, Keum N, Giovannucci E, et al. Whole grain
13. Wu X, Beecher GR, Holden JM, et al. Lipophilic consumption and risk of cardiovascular disease,
and hydrophilic antioxidant capacities of common cancer, and all cause and cause specific mortality:
foods in the United States. J Agric Food Chem. systematic review and dose-response meta-analy-

2004;52:4026–37. sis of prospective studies. BMJ. 2016;353:i2716.
14. Ros E, Hu FB. Consumption of plant seeds and car- https://doi.org/10.1136/bmj.i2716.
diovascular health epidemiological and clinical trial 29. Wei H, Gao Z, Liang R, et al. Whole-grain con-
evidence. Circulation. 2013;128:553–65. sumption and the risk of all-cause, CVD and can-
15. Slavin JL, Lloyd B. Health benefits of fruits and veg- cer mortality: a meta-analysis of prospective cohort
etables. Adv Nutr. 2012;3:506–16. studies. Br J Nutr. 2016;116(3):514–25. https://doi.
16. Rebello CJ, Greenway FL, Finley JW. A review of org/10.1017/S0007114516001975.
the nutritional value of legumes and their effects on 30. Zong G, Gao A, FB H, Sun Q. Whole grain intake
obesity and its related co-morbidities. Obes Rev. and mortality from all causes, cardiovascular dis-
2014;15:392–407. ease, and cancer a meta-analysis of prospective
17. Gebhardt SE, Thomas R. Nutritive value of foods. cohort studies. Circulation. 2016;133:2370–80.
Home and Garden Bulletin, Agriculture Research 31. Li B, Zhang G, Tan M, et al. Consumption of whole
Service, United States Department of Agriculture, grains in relation to mortality from all causes, car-
2002,72:36–68 diovascular disease, and diabetes. Dose–response
18. Dietary Guidelines Advisory Committee (DGAC). meta-analysis of prospective cohort studies.
Scientific Report. Advisory Report to the Secretary Medicine. 2016;95:33(e4229); doi: org/10.1097/
of Health and Human Services and the Secretary of MD.00000000000429.
Agriculture. Part D. Chapter 2: Dietary patterns, foods 32. Benisi-Kohansal S, Saneei P, Salehi-Marzijarani
and nutrients and health outcomes 2015. p. 1–35. M, et al. Whole-grain intake and mortality from all
19. Mozaffarian D. Dietary and policy priorities causes, cardiovascular disease, and cancer: a system-
for cardiovascular disease, diabetes, and obe- atic review and dose-response meta-analysis of pro-
sity. A comprehensive review. Circulation. spective cohort studies. Adv Nutr. 2016;7:1052–65.
2017;133(2):187–225. https://doi.org/10.1161/ https://doi.org/10.3945/an.11615.
CIRCULATIONAHA.11519.018585. 33. Xu M, Huang T, Lee AM, et al. Ready-to-eat cereal
20. Palmer AK, Tchkonia T, LeBrasseur NK, et al. consumption with total and cause -specific mortal-
Cellular senescence in type 2 diabetes: a therapeutic ity: prospective analysis of 367,442 individuals. J
opportunity. Diabetes. 2015;64:2289–98. Am Coll Nutr. 2016;35(3):217–23.
References 111
34. Wu H, Flint AJ, Qi Q. Association between dietary Eur J Epidemiol. 2013;28:845–58. https://doi.
whole grain intake and risk of mortality: two large org/10.1007/s10654-013-9852-5.
prospective studies in US men and women. JAMA 48. Chanson-Rolle A, Meynier A, Aubin F, et al.
Intern Med. 2015;175(3):373–84. Systematic review and meta-analysis of human
35. Huang T, Xu M, Lee A, et al. Consumption of whole studies to support a quantitative recommendation
grains and cereal fiber and total and cause-specific for whole grain intake in relation to type 2 diabe-
mortality: prospective analysis of 367,442 individu- tes. PLoS One. 2015;10(6):e0131377. https://doi.
als. BMC Med. 2015;13:59. https://doi.org/10.1186/ org/10.1371/journal.pone.0131377.
s12916-015-0338-z. 49. Sun Q, Spiegelman D, van Dam RM, et al. White
36. Johnsen NF, Frederiksen K, Christensen J, et al. rice, brown rice and risk of type 2 diabetes in US men
Whole-grain products and whole-grain types are and women. Arch Intern Med. 2010;170(11):961–9.
associated with lower all-cause and cause-specific 50. de Munter JSL, Hu FB, Spiegelman D, et al. Whole
mortality in the Scandinavian HELGA cohort. Br J grain, bran, and germ intake and risk of type 2 dia-
Nutr. 2015;114:608–23. betes: a prospective cohort study and systematic
37. He M, van Dam RM, Rimm E, et al. Whole review. PLoS Med. 2007;4(8):e261. https://doi.
grain, cereal fiber, bran, and germ intake and the org/10.1371/journal.pmed.0040261.
risks of all-cause and CVD-specific mortality 51. Aune D, Chan DSM, Lau R, et al. Dietary fibre,
among women with type 2 diabetes. Circulation. whole grains, and risk of colorectal cancer: system-
2010;121(20):2162–8. atic review and dose-response meta-analysis of pro-
38. Hollænder PLB, Ross AB, Kristensen M. Whole-grain spective studies. BMJ. 2011;343:d6617. https://doi.
and blood lipid changes in apparently healthy adults: org/10.1136/bmj.d6617.
a systematic review and meta-analysis of randomized 52. Haas P, Machado MJ, Anton AA, et al. Effectiveness
controlled studies. Am J Clin Nutr. 2015;102(3):556– of whole grain consumption in the prevention of
72. https://doi.org/10.3945/ajcn.115.109165. colorectal cancer: meta-analysis of cohort studies.
39. Ye EQ, Chacko SA, Chou EL, et al. Greater whole- Int J Food Sci Nutr. 2009;60(Suppl 6):1–13.
grain intake is associated with lower risk of type 2 53. Kyrø C, Skeie G, Loft S, et al. Intake of whole grains
diabetes, cardiovascular disease, and weight gain. J from different cereal and food sources and incidence
Nutr. 2012;142:1304–13. of colorectal cancer in the Scandinavian HELGA
40. Kelly SAM, Summerbell CD, Brynes A, et al. Cohort. Cancer Causes Control. 2013;24:1363–74.
Wholegrain cereals for coronary heart disease. https://doi.org/10.1007/s10552-013-0215-z.
Cochrane Database Syst Rev. 2007; 2:CD005051; 54. Milani C, Ferrario C, Turroni F, et al. The human gut
doi:10.1002/14651858.CD005051.pub2. microbiota and its interactive connections to diet. J
41. Flint AJ, FB H, Glynn RJ, et al. Whole grains and Hum Nutr Diet. 2016;29(5):539–46. https://doi.
incident hypertension in men. Am J Clin Nutr. org/10.1111/jhn.12371.
2009;90:493–8. 55. De Angelis M, Montemurno E, Vannini L, et al. Effect
42. Kirwan JP, Malin SK, Scelsi AR, et al. Whole-grain of whole-grain barley on the human fecal micro-
diet reduces cardiovascular risk factors in over- biota and metabolome. Appl Environ Microbiol.
weight and obese adults: a randomized controlled 2015;81:7945–56.
trial. J Nutr. 2016;146:2244–51. 56. vel Szic KS, Declerck K, Vidaković M, Vanden
43. Tighe P, Duthie G, Vaughan N, et al. Effect of Berghe W. From inflammaging to healthy aging by
increased consumption of whole-grain foods on dietary lifestyle choices: is epigenetics the key to
blood pressure and other cardiovascular risk markers personalized nutrition? Clin Epigenetics. 2015; 7:33.
in healthy middle-aged persons: a randomized con- doi: 10.1186/s13148-015-0068-2.
trolled trial. Am J Clin Nutr. 2010;92:733–40. 57. de Heredia FP, Gomez-Martınez S, Marcos
44. Chen J, Huang Q, Shi W, et al. Meta-analysis of A. Chronic and degenerative diseases, obesity,
the association between whole and refined grain inflammation and the immune system. Proc Nutr Soc.
consumption and stroke risk based on prospec- 2012;71:332–8.
tive cohort studies, Asia Pacific. J Public Health. 58. Lefevre M, Jonnalagadda S. Effect of whole grains
2016;28(7):563–75. on markers of subclinical inflammation. Nutr Rev.
45. Fang L, Li W, Zhang W, et al. Association 2012;70(7):387–96.
between whole grain intake and stroke risk: evi- 59. Puzianowska-Kuźnicka M, Owczarz M,
dence from meta-analysis. Int J Clin Exp Med. Wieczorowska-Tobis K, et al. Interleukin-6 and
2015;8(9):16978–83. C-reactive protein, successful aging, and mortality:
46. Liu S, Manson JE, Stampfer MJ, et al. Whole grain The PolSenior Study. Immun Ageing. 2016;13:21.
consumption and risk of ischemic stroke in women. https://doi.org/10.1186/s12979-016-0076-x.
JAMA. 2000;284(12):1534–40. 60. Jeffery IB, O’Toole PW. Diet-microbiota interac-
47. Aune D, Norat T, Romundstad P, Vatten LJ. Whole tions and their implications for healthy living. Forum
grain and refined grain consumption and the Nutr. 2013;5:234–52.
risk of type 2 diabetes: a systematic review and 61. Arora T, Backhed F. The gut microbiota and meta-
dose-response meta-analysis of cohort studies. bolic disease: current understanding and future
perspectives. J Intern Med. 2016;280(4):339–49. Nutr. 2017;105(3):589–99. https://doi.org/10.3945/

https://doi.org/10.1111/joim.12508. ajcn.116.139683.
62. Keenan MJ, Marco ML, Ingram DK, Martin 77. Alfakry H, Malle E, Koyani CN, et al. Neutrophil
RJ. Improving health span via changes in gut micro- proteolytic activation cascades: a possible mechanis-
biota and fermentation. Age. 2015;37(5):98. tic link between chronic periodontitis and coronary
63. O’Toole PWO, Jeffery IB. Gut microbiota and aging. heart disease. Innate Immun. 2016;22(1):85–99.
Science. 2015;350(6265):1214–5. https://doi.org/10.1177/1753425915617521.
64. Zapata HJ, Quagliarello VJ. The microbiota and 78. Nielsen SJ, Trak-Fellermeier MA, Joshipura K,
microbiome in aging: potential implications in Dye BA. Dietary fiber intake is inversely associ-
health and age-related diseases. J Am Geriatr Soc. ated with periodontal disease among US adults. J
2015;63(4):776–81. Nutr. 2016;146:2530–6. https://doi.org/10.3945/
65. Claesson MJ, Jeffery IB, Conde S. Gut microbiota jn.116.237065.
composition correlates with diet and health in the 79. Merchant AT, Pitiphat W, Franz M, Joshipura
elderly. Nature. 2012;488:178–85. KJ. Whole-grain and fiber intakes and periodontitis
66. Tuohy KM, Conterno L, Gasperotti M, Viola risk in men. Am J Clin Nutr. 2006;83:1395–400.
R. Up-regulating the human intestinal microbiome 80. World Health Organization. Diet, nutrition, and
using whole plant foods, polyphenols, and/or fiber. J the prevention of chronic diseases. Geneva: World
Agric Food Chem. 2012;60:8776–82. Health Organization; 1990. http://www.who.int/
67. Tap J, Furet JP, Bensaada M, et al. Gut microbiota nutrition/publications/obesity/WHO_TRS_797/en/
richness promotes its stability upon increased dietary index.html. Accessed 16 April 2015
fibre intake in healthy adults. Environ Microbiol. 81. Bertoia ML, Rimm EB, Mukamal KJ, et al. Dietary
2015;17(12):4954–64. flavonoid intake and weight maintenance: three pro-
68. Vanegas SM, Meydani M, Barnett JB, et al. spective cohorts of 124,086 US men and women fol-
Substituting whole grains for refined grains in a lowed for up to 24 years. BMJ. 2016;352:i17. doi.
6-wk randomized trial has a modest effect on gut org/10.1136/bmj.i17
microbiota and immune and inflammatory mark- 82. Aune D, Giovannucci E, Boffetta P, et al. Fruit
ers of healthy adults. Am J Clin Nutr. 2017;105(3): and vegetable intake and the risk of cardiovascular
653–0. https://doi.org/10.3945/ajcn.116.146928. disease, total cancer and all-cause mortality-a sys-
69. Martınez I, Lattimer JM, Hubach KL, et al. Gut tematic review and dose response meta-analysis of
microbiome composition is linked to whole grain- prospective studies. Int J Epidemiol. 2017:1–28.
induced immunological improvements. ISME J. https://doi.org/10.1093/ije/dyw319.
2013;7:269–80. 83. Wang X, Ouyang Y, Liu J. Fruit and vegetable
70. Price RK, Wallace JMW, Hamill LL, et al. Evaluation consumption and mortality from all causes, car-
of the effect of wheat aleurone-rich foods on markers diovascular disease, and cancer: systematic review
of antioxidant status, inflammation and endothelial and dose-response meta-analysis of prospective
function in apparently healthy men and women. Br J cohort studies. BMJ. 2014;349:g4490. https://doi.
Nutr. 2012;108:1644–51. org/10.1136/bmj.g4490.
71. Sepe A, Tchkonia T, Thomou T, et al. Aging and 84. Bellavia A, Stilling F, Wolk A. High red meat intake
regional differences in fat cell progenitors - a mini- and all-cause cardiovascular and cancer mortality: is
review. Gerontology. 2011;57:66–75. https://doi. the risk modified by fruit and vegetable intake? Am J
org/10.1159/000279755. Clin Nutr. 2016;104:1137–43.
72. Karl JP, Saltzman E. The role of whole grains in 85. Nguyen B, Bauman A, Gale E, et al. Fruit and vege-
body weight regulation. Adv Nutr. 2012;3:697–707. table consumption and all-cause mortality: evidence
73. Pol K, Christensen R, Bartels EM, et al. Whole grain from a large Australian cohort study. Int J Behav
and body weight changes in apparently healthy adults: Nutr Phys Act. 2016;13:9. https://doi.org/10.1186/
a systematic review and meta-analysis of randomized s12966-016-0334-S.
controlled studies. Am J Clin Nutr. 2013;98:872–84. 86. Oyebode O, Gordon-Dseagu V, Walker A, Mindell
74. Serra-Majem L, Bautista-Castaño I. Relationship JS. Fruit and vegetable consumption and all-cause,
between bread and obesity. Br J Nutr. 2015;113:S29– cancer and CVD mortality: analysis of health survey
35. https://doi.org/10.1017/S0007114514003249. for England data. J Epidemiol Community Health.
75. McKeown M, Troy LM, Jacques PF, et al. Whole- 2014;68:856–62.
and refined-grain intakes are differentially associ- 87. Leenders M, Sluijs I, Ros MM, et al. Fruit and
ated with abdominal visceral and subcutaneous vegetable consumption and mortality European
adiposity in healthy adults: the Framingham Heart Prospective Investigation into Cancer and Nutrition.
Study. Am J Clin Nutr. 2010;92:1165–71. Am J Epidemiol. 2013;178(4):590–602.
76. Karl JP, Meydani M, Barnett JB, et al. Substituting 88. Bellavia A, Larsson SC, Bottai M, et al. Fruit
whole grains for refined grains in a 6-wk random- and vegetable consumption and all-cause mor-
ized trial favorably affects energy-balance metrics in tality: a dose response analysis. Am J Clin Nutr.
healthy men and postmenopausal women. Am J Clin 2013;98:454–9.
References 113
89. Zhang X, Shu X-O, Xiang Y-B, et al. Cruciferous 102. Appel LJ, Moore TJ, Obarzanek E, et al. A clinical
vegetable consumption is associated with a reduced trial of the effects of dietary patterns on blood pres-
risk of total and cardiovascular disease mortality. sure. N Engl J Med. 1997;336:1117–24.
Am J Clin Nutr. 2011;94:240–6. 103. Hu D, Huang J, Wang Y, et al. Fruits and vegetables
90. Huang H, Chen G, Liao D, et al. Effects of berries consumption and risk of stroke. A meta-analysis of
consumption on cardiovascular risk factors: a meta- prospective cohort studies. Stroke. 2014;45:1613–9.
analysis with trial sequential analysis of randomized 104. Cassidy A, Bertoia M, Chiuve S, et al. Habitual
controlled trials. Sci Rep. 2016;6:23625. https://doi. intake of anthocyanins and flavanones and risk of
org/10.1038/srep23625. cardiovascular disease in men. Am J Clin Nutr.
91. Gan Y, Tong X, Li L, et al. Consumption of fruit and 2016;104:587–94.
vegetable and risk of coronary heart disease: a meta- 105. Oude Griep LM, Verschuren WMM, Kromhout D,
analysis of prospective cohort studies. Int J Cardiol. et al. Colors of fruit and vegetables and 10-year inci-
2015;183:129–37. dence of stroke. Stroke. 2011;42:3190–5.
92. Zhan J, Liu Y-J, Cai L-B, et al. Fruit and vegetable 106. Joshipura KJ, Ascherio A, Manson JE, et al. Fruit
consumption and risk of cardiovascular disease: a and vegetable intake in relation to risk of ischemic
meta-analysis of prospective cohort studies. Crit Rev stroke. JAMA. 1999;282:1233–9.
Food Sci Nutr. 2015. https://doi.org/10.1080/104083 107. Wang P-Y, Fang J-C, Gao Z-H, et al. Higher intake
98.2015.1008980. of fruits, vegetables or their fiber reduces the risk of
93. Hartley L, Igbinedion E, Holmes J, et al. Increased type 2 diabetes: a meta-analysis. J Diabetes Investig.
consumption of fruit and vegetables for the primary 2016;7:56–69. https://doi.org/10.1111/di.12376.
prevention of cardiovascular diseases. Cochrane 108. Li M, Fan Y, Zhang X, et al. Fruit and vegetable
Database Systematic Rev. 2013;6:CD009874; doi: intake and risk of type 2 diabetes mellitus: meta-
10.1002/14651858.CD009874.pub2. analysis of prospective cohort studies. BMJ Open.
94. Larsson SC, Wolk A. Potato consumption and risk of 2014;4(11):e005497. https://doi.org/10.1136/
cardiovascular disease: 2 prospective cohort studies. bmjopen-2014-005497.
Am J Clin Nutr. 2016;104:1245–53. 109. Xi B, Li S, Liu Z, et al. Intake of fruit juice and inci-
95. Miedema MD, Andrew Petrone A, Shikany JM, dence of type 2 diabetes: a systematic review and
et al. Association of fruit and vegetable consump- meta-analysis. PLoS One. 2014;9(3). https://doi.
tion during early adulthood with the prevalence of org/10.1371/journal.pone.0093471.
coronary artery calcium after 20 years of follow-up. 110. Mamluk L, O’Doherty MG, Orfanos P, et al. Fruit
The Coronary Artery Risk Development in Young and vegetable intake and risk of incident of type 2
Adults (CARDIA) Study. Circulation. 2015;132: diabetes: results from the consortium on health and
1990–8. ageing network of cohorts in Europe and the United
96. Bhupathiraju SN, Wedick NM, Pan A, et al. States (CHANCES). Eur J Clin Nutr. 2016;71:83–
Quantity and variety in fruit and vegetable intake 91. https://doi.org/10.1038/ejcn.2016.143.
and risk of coronary heart disease. Am J Clin Nutr. 111. Muraki I, Rimm EB, Willett WC, et al. Potato
2013;98:1514–23. consumption and risk of type 2 diabetes: results
97. Buil-Cosiales P, Toledo E, Salas-Salvadó J, et al. from three prospective cohort studies. Diabetes
Association between dietary fibre intake and fruit, vege- Care. 2016;39:376–84. https://doi.org/10.2337/
table or whole-grain consumption and the risk of CVD: dc15-0547.
results from the PREvención con Dieta MEDiterránea 112. Aune D, Lau R, Chan DSM, et al. Nonlinear reduc-
(PREDIMED) trial. Br J Nutr. 2016;116(3):534–46. tion in risk for colorectal cancer by fruit and veg-
https://doi.org/10.1017/S0007114516002099. etable intake based on meta-analysis of prospective
98. Li B, Wang L, Zhang D. Fruit and vegetable studies. Gastroenterology. 2011;141:106–18.
consumption and risk of hypertension: a meta-anal- 113. Kunzmann AT, Coleman HG, Huang W-Y, et al.
ysis. J Clin Hypertens (Greenwich). 2016;18(5): Fruit and vegetable intakes and risk of colorec-
468–76. tal cancer and incident and recurrent adenomas
99. Wu L, Sun D, He Y. Fruit and vegetable consump- in the PLCO cancer screening trial. Int J Cancer.
tion and incident hypertension: dose-response 2016;138:1851–61.
meta-analysis of prospective cohort studies. J 114. Leenders M, Siersema PD, Overvad K, et al.
Hum Hypertens. 2016;30(10):573–80. https://doi. Subtypes of fruit and vegetables, variety in con-
org/10.1038/jhh.2016.44. sumption and risk of colon and rectal cancer in the
100. Borgi L, Muraki I, Satija A, et al. Fruit and vegeta- European Prospective Investigation into Cancer
ble consumption and the incidence of hypertension and Nutrition. Int J Cancer. 2015;137(11):2705–14.
in three prospective cohort studies. Hypertension. https://doi.org/10.1002/ijc.29640.
2016;67:288–93. 115. Hui C, Qi X, Qianyong Z, Xiaoli P, et al. Flavonoids,
101. Borgi L, Rimm EB, Willett WC, Forman JP. Potato flavonoid subclasses and breast cancer risk: a meta-
intake and incidence of hypertension from three pro- analysis of epidemiologic studies. PLoS One.
spective US cohort studies. BMJ. 2016;353:i2351. 2013;8(1):e54318. https://doi.org/10.1371/journal.
https://doi.org/10.1136/bmj.i2351. pone.0054318.
116. Aune D, Chan DS, Vieira AR, et al. Fruits, veg- 130. Farvid MS, Malekshah AF, Pourshams A, et al.
etables and breast cancer risk: a systematic review Dietary protein sources and all-cause and cause-
and meta-analysis of prospective studies. Breast specific mortality: The Golestan Cohort Study in
Cancer Res Treat. 2012;134(2):479–93. https://doi. Iran. Am J Prev Med. 2017;52(2):237–48. https://
org/10.1007/s10549-012-2118-1. doi.org/10.1016/j.amepre.2016.10.041.
117. Aune D, Chan DS, Vieira AR, et al. Dietary compared 131. Chang W-C, Wahlqvist ML, Chang H-Y, et al. A
with blood concentration of carotenoids and breast bean-free diet increases the risk of all-cause mortality
cancer risk: a systematic review and meta-analysis of among Taiwanese women: the role of the metabolic
prospective studies. Am J Clin Nutr. 2012;96:356–73. syndrome. Public Health Nutr. 2011;15(4):663–72.
118. Lamport DJ, Saunders C, Butler LT, Spencer https://doi.org/10.1017/S1366890011002151.
JPR. Fruits, vegetables, 100% juices, and cognitive 132. Darmadi-Blackberry I, Wahlqvist ML, Kouris-
function. Nutr Rev. 2014;2(12):774–89. Blazos A, et al. Legumes: the most important dietary
119. Loef M, Walach H. Fruit, vegetable and preven- predictor of survival in older people of different eth-
tion of cognitive decline or dementia: a system- nicities. Asia Pac J Clin Nutr. 2004;13:217–20.
atic review of cohort studies. J Nutr Health Aging. 133. Marventano S, Pulido MI, Sánchez-González C,
2012;16(7):625–30. et al. Legume consumption and CVD risk: a system-
120. Neshatdousta S, Saunders C, Castle SM, et al. atic review and meta-analysis. Public Health Nutr.
High-flavonoid intake induces cognitive improve- 2017;20(2):245–54.
ments linked to changes in serum brain-derived 134. Tokede OA, Onabanjo TA, Yansane A, et al.
neurotrophic factor: two randomised, controlled tri- Soya products and serum lipids: a meta- analysis
als. Nutr Healthy Aging. 2016;4:81–93. https://doi. of randomised controlled trials. Br J Nutr.
org/10.3233/NHA-1615. 2015;114:831–43.
121. Kean RJ, Lamport DJ, Dodd GF, et al. Chronic con- 135. Ha V, John L, Sievenpiper JL, et al. Effect of dietary
sumption of flavanone-rich orange juice is associ- pulse intake on established therapeutic lipid tar-
ated with cognitive benefits: an 8-wk, randomized, gets for cardiovascular risk reduction: a systematic
double-blind, placebo-controlled trial in healthy review and meta-analysis of randomized controlled
older adults. Am J Clin Nutr. 2015;101:506–14. trials. CMAJ. 2014;186(8):252–62.
122. Nooyens ACJ, Bueno-de-Mesquita HB, van Boxtel 136. Afshin A, Micha R, Khatibzadeh S, Mozaffarian
MPJ. Fruit and vegetable intake and cognitive decline D. Consumption of nuts and legumes and risk of
in middle-aged men and women: the Doetinchem incident ischemic heart disease, stroke, and diabetes:
Cohort Study. Br J Nutr. 2011;106:752–61. a systematic review and meta-analysis. Am J Clin
123. Peneau S, Galan P, Jeandel C, et al. Fruit and Nutr. 2014;100:278–88.
vegetable intake and cognitive function in the 137. Bazzano LA, Thompson AM, Tees MT, et al. Non-
SU.VI.MAX 2 prospective study. Am J Clin Nutr. soy legume consumption lowers cholesterol levels: a
2011;94:1295–303. meta-analysis of randomized controlled trials. Nutr
124. Neville CE, Young IS, Gilchrist SECM, et al. Metab Cardiovasc Dis. 2011;21(2):94–103.
Effect of increased fruit and vegetable consump- 138. Anderson JW, Bush HM. Soy protein effects on
tion on physical function and muscle strength in serum lipoproteins: a quality assessment and meta-
older adults. Age. 2013;35:2409–22. https://doi. analysis of randomized, controlled studies. J Am
org/10.1007/s11357-013-9530-2. Coll Nutr. 2011;30(2):79–91.
125. Gibson A, Edgar JD, Neville CE, et al. Effect of fruit 139. Liua Z-M, Hob SC, Chen Y-M, Woo J. Effect of soy
and vegetable consumption on immune function in protein and isoflavones on blood pressure and endo-
older people: a randomized controlled trial. Am J thelial cytokines: a 6-month randomized controlled
Clin Nutr. 2012;96:1429–36. trial among postmenopausal women. J Hypertens.
126. Ribeiro SM, Morley JE, Malmstrom TK, Miller 2013;31(2):384–92. https://doi.org/10.1097/
DK. Fruit and vegetable intake and physical activ- HJH.0b013e32835c0905.
ity as predictors of disability risk factors in African- 140. Pittaway JK, Robertson IK, Ball MJ. Chickpeas may
American middle-aged individuals. J Nutr Health influence fatty acid and fiber intake in an ad libi-
Aging. 2016;20(9):891–6. https://doi.org/10.1007/ tum diet, leading to small improvements in serum
s12603-016-0780-4. lipid profile and glycemic control. J Am Diet Assoc.
127. Lian F, Wang J, Huang X, et al. Effect of vegetable 2008;108:1009–13.
consumption on the association between peripheral 141. He J, Gu D, Wu X, et al. Effect of soybean protein on
leucocyte telomere length and hypertension: a case– blood pressure. A randomized, controlled trial. Ann
control study. BMJ Open. 2015;5:e009305. https:// Intern Med. 2005;143(1):1–9.
doi.org/10.1136/bmjopen-2015-009305. 142. Liu ZM, Chen YM, Ho SC. Effects of soy intake on
128. McCrory MA, Hamaker BR, Lovejoy JC, glycemic control: a meta-analysis of randomized
Eichelsdoerfer PE. Pulse consumption, satiety, and controlled trials. Am J Clin Nutr. 2011;93:1092–101.
weight management. Adv Nutr. 2010;1:17–30. 143. Sievenpiper JL, Kendall CW, Esfahani A, et al.
129. Messina V. Nutritional and health benefits of dried Effect of non-oil-seed pulses on glycaemic control: a
beans. Am J Clin Nutr. 2014;100(suppl):437S–42S. systematic review and meta-analysis of randomised
References 115
controlled experimental trials in people with and from the 2008/09 New Zealand Adult Nutrition
without diabetes. Diabetologia. 2009;52:1479–95. Survey. Br J Nutr. 2016;115:105–12.
144. Ding M, Pan A, Manson JE, et al. Consumption of 158. Novotny JA, Gebauer SK, Baer DJ. Discrepancy
soy foods and isoflavones and risk of type 2 diabe- between the Atwater factor predicted and empiri-
tes: a pooled analysis of three US cohorts. Eur J Clin cally measured energy values of almonds in human
Nutr. 2016;70(12):1381–7. https://doi.org/10.1038/ diets. Am J Clin Nutr. 2012;96:296–301.
ejcn.2016.117. 159. Nielsen SJ, Kit BK, Ogden CL. Nut consumption
145. Villegas R, Gao YT, Yang G, et al. Legume and soy among U.S. adults, 2009-2010. NCHS data brief,
food intake and the incidence of type 2 diabetes in no 176. Hyattsville, MD: National Center for Health
the Shanghai Women’s Health Study. Am J Clin Statistics 2014.
Nutr. 2008;87:162–7. 160. Grosso G, Estruch R. Nut consumption and age-
146. Jenkins DJA, Kendall WC, Augustin LSA, et al. related disease. Maturitas. 2015;84:11–6.
Effect of legumes as part of a low glycemic index diet 161. Aune D, Keum NN, Giovannucci E, et al. Nut con-
on glycemic control and cardiovascular risk factors sumption and risk of cardiovascular disease, total
in type 2 diabetes mellitus. A randomized controlled cancer, all-cause and cause specific mortality: a
trial. Arch Intern Med. 2012;172(21):1653–60. systematic review and dose-response meta-analysis
147. Yu Y, Jing X, Li H, et al. Soy isoflavone consumption of prospective studies. BMC Med. 2016;14:207.
and colorectal cancer risk: a systematic review and https://doi.org/10.1186/s12916-016-0730-3.
meta-analysis. Sci Rep. 2016;6:25939. https://doi. 162. Mayhew AJ, de Souza RJ, Meyre D, et al. A system-
org/10.1038/srep25939. atic review and meta-analysis of nut consumption
148. Zhu B, Sun Y, Qi L, et al. Dietary legume consump- and incident risk of CVD and all-cause mortality. Br
tion reduces risk of colorectal cancer: evidence J Nutr. 2016;115:212–25.
from a meta-analysis of cohort studies. Sci Rep. 163. Wu L, Wang Z, Zhu J, et al. Nut consumption and
2015;5:8797. https://doi.org/10.1038/srep08797. risk of cancer and type 2 diabetes: a systematic
149. Yang G, Shu X-O, Li H, et al. Prospective cohort review and meta-analysis. Nutr Rev. 2015;73(7):
study of soy food intake and colorectal cancer risk in 409–25.
women. Am J Clin Nutr. 2009;89:577–83. 164. Guo K, Zhou Z, Jiang Y, et al. Meta-analysis of pro-
150. Chen M, Rao Y, Zheng Y, et al. Association between spective studies on the effects of nut consumption on
soy isoflavone intake and breast cancer risk for prehypertension and type 2 diabetes mellitus. J Diabetes.
and postmenopausal women: a meta-analysis of epi- 2015;7(2):202–12; doi: 1111/1753-0407.12173.
demiological studies. PLoS One. 2014;9(2):e89288. 165. Zhang Z, Xu G, Wei Y, et al. Nut consumption and
https://doi.org/10.1371/journal.pone.0089288. risk of stroke. Eur J Epidemiol. 2015;30(3):189–96.
151. Xie Q, Chen M-L, Qin Y, et al. Isoflavone consump- https://doi.org/10.1007/s10654-015-9999-3.
tion and risk of breast cancer: a dose-response meta- 166. Luo C, Zhang Y, Ding Y, et al. Nut consumption and
analysis of observational studies. Asia Pac J Clin risk of type 2 diabetes, cardiovascular disease, and
Nutr. 2013;22(1):118–27. https://doi.org/10.6133/ all-cause mortality: a systematic review and meta-
apjcn.2013.22.1.16. analysis. Am J Clin Nutr. 2014;100:256–69.
152. Chen X, Huang Y, Cheng HG. Lower intake of vege- 167. Zhou D, Yu H, He F, et al. Nut consumption in relation
tables and legumes associated with cognitive decline to cardiovascular disease risk and type 2 diabetes: a
among illiterate elderly Chinese: a 3-year cohort systematic review and meta-analysis of prospective
study. J Nutr Health Aging. 2012;16(6):548–52. studies. Am J Clin Nutr. 2014;100:270–7.
153. St. John JA, Henderson VW, Hodis HN, et al. 168. Wang W, Yang M, Kenfield SA, et al. Nut consump-
Associations of urine excretion of isoflavonoids with tion and prostate cancer risk and mortality. Br J
cognition in postmenopausal women in the Women’s Cancer. 2016;115(3):371–4. https://doi.org/10.1038/
Isoflavone Soy Health Clinical Trial. J Am Geriatr Soc. bjc.2016.181.
2014;62(4):629–35. https://doi.org/10.1111/jgs.12752. 169. Yang M, FB H, Giovannucci E, et al. nut con-
154. Kreijkamp-Kaspers S, Kok L, Gobbee DE, et al. sumption and risk of colorectal cancer in women.
Effect of soy protein containing isoflavones on cog- Eur J Clin Nutr. 2016;70(3):333–7. https://doi.
nitive function, bone mineral density, and plasma org/10.1038/ejcn.2015.66.
lipids in postmenopausal women a randomized con- 170. Luu HN, Blot WJ, Xiang Y-B, et al. Prospective eval-
trolled trial. JAMA. 2004;292:65–74. uation of the association of nut/peanut consumption
155. Jackson CL, Hu FB. Long-term associations of nut with total and cause-specific mortality. JAMA Intern
consumption with body weight and obesity. Am J Med. 2015;175(5):755–66.
Clin Nutr. 2014;100(suppl):408S–11S. 171. Gopinath B, Flood VM, Burlutksy G, Mitchell
156. Kris-Etherton PM, Hu FB, Ros E, Sabaté J. The role P. Consumption of nuts and risk of total and
of tree nuts and peanuts in the prevention of coro- cause-specific mortality over 15 years. Nutr Metab
nary heart disease: multiple potential mechanisms. J Cardiovasc Dis. 2015;25(12):1125–31.
Nutr. 2008;138(9):1746S–51S. 172. Hshieh TT, Petrone AB, Gaziano JM, Djousse L. Nut
157. Brown RC, Tey SL, Gray AR, et al. Nut consump- consumption and risk of mortality in the Physicians’
tion is associated with better nutrient intakes: results Health Study. Am J Clin Nutr. 2015;101(2):407–12.
173. Bao Y, Han J, FB H, et al. Association of nut con- a meta-analysis and systematic review. Am J Clin
sumption with total and cause-specific mortality. N Nutr. 2009;90:56–63.
Engl J Med. 2013;369(21):2001–11. 185. Blanco Mejia S, Kendall CWC, Viguiliouk E, et al.
174. Bao Y, FB H, Giovannucci EL, et al. Nut consump- Effect of tree nuts on metabolic syndrome criteria: a
tion and risk of pancreatic cancer in women. Br J systematic review and meta-analysis of randomised
Cancer. 2013;109:2911–6. controlled trials. BMJ Open. 2014;4(7):e004660.
175. Jenab M, Ferrari P, Slimani N, et al. Association of https://doi.org/10.1136/bmjopen-2013-004660.
nut and seed intake with colorectal cancer risk in 186. Flores-Mateo G, Rojas-Rueda D, Basora J, et al. Nut
the European Prospective Investigation into Cancer intake and adiposity: meta-analysis of clinical trials.
and Nutrition. Cancer Epidemiol Biomarkers Prev. Am J Clin Nutr. 2013;97:1346–55.
2004;13:1595–603. 187. Xiao Y, Huang W, Peng C, et al. Effect of nut
176. Guasch-Ferré M, Bulló M, Martínez-González consumption on vascular endothelial function: A
MA, et al. Frequency of nut consumption and systematic review and meta-analysis of random-
mortality risk in the PREDIMED nutrition inter- ized controlled trials. Clin Nutr. 2017. https://doi.
vention trial. BMC Med. 2013;11:164. https://doi. org/10.1016/j.clnu.2017.04.011.
org/10.1186/1741-7015-11-164. 188. Gulati S, Misra A, Pandey RM. Effect of almond
177. Toledo E, Salas-Salvado J, Donat-Vargas D, et al. supplementation on glycemia and cardiovascular
Mediterranean diet and invasive breast cancer risk risk factors in Asian Indians in North India with
among women at high cardiovascular risk in the type 2 diabetes mellitus: a 24-week study. Metab
PREDIMED trial. A randomized clinical trial. Syndr Relat Disord. 2017;15(2):98–105. https://doi.
JAMA Intern Med. 2015;175(11):1752–60. https:// org/10.1089/met.2016.0066.
doi.org/10.1001/jamainternalmed2015.48.38. 189. Njike VY, Ayettey R, Petraro P, et al. Walnut inges-
178. Salas-Salvado J, Bullo M, Estruch R, et al. tion in adults at risk for diabetes: effects on body
Prevention of diabetes with Mediterranean diets. composition, diet quality, and cardiac risk measures.
Ann Intern Ned. 2014;160:1–10. BMJ Open Diabetes Res Care. 2015;3:e000115.
179. Estruch R, Ros E, Salas-Salvadó J, et al. Primary pre- https://doi.org/10.1136/bmjdrc-2015-000115.
vention of cardiovascular disease with a Mediterranean 190. Sauder KA, McCrea CE, Ulbrecht JS, et al. Effects
diet. N Engl J Med. 2013;368(14):1279–90. of pistachios on the lipid/lipoprotein profile, glyce-
180. Musa-Veloso K, Paulionis L, Poon T, Lee HY. The mic control, inflammation, and endothelial function
effects of almond consumption on fasting blood lipid in type 2 diabetes: a randomized trial. Metabolism.
levels: a systematic review and meta-analysis of ran- 2015;64(11):1521–9. https://doi.org/10.1016/j.
domised controlled trials. J Nutr Sci. 2016;5:e34. metabol.2015.07.021.
https://doi.org/10.1017/jns.2016.19. 191. Valls-Pedret C, Sala-Vila A, Serra-Mir M, et al.
181. Del Gobbo LC, Falk MC, Feldman R, et al. Effects Mediterranean diet and age-related cognitive decline
of tree nuts on blood lipids, apolipoproteins, and a randomized clinical trial. JAMA Intern Med.
blood pressure: systematic review, meta-analysis, 2015;175(7):1094–103. https://doi.org/10.1001/
and dose-response of 61 controlled intervention tri- jamaiternmed.2015.1668.
als. Am J Clin Nutr. 2015;102:1347–56. 192. O’Brien J, Okereke O, Devore E, et al. Long-

182. Mohammadifard N, Salehi-Abargouei A, Salas- term intake of nuts in relation to cognitive func-
Salvadó J, et al. The effect of tree nut, peanut, and tion in older women. J Nutr Health Aging.
soy nut consumption on blood pressure: a systematic 2014;18(5):496–502.
review and meta-analysis of randomized controlled 193. Sánchez-Villegas A, Galbete C, Martinez-González
clinical trials. Am J Clin Nutr. 2015;101:966–82. MA. The effect of the Mediterranean diet on plasma
183. Viguiliouk E, Kendall CWC, Mejia SB, et al. Effect brain-derived neurotrophic factor (BDNF) levels:
of tree nuts on glycemic control in diabetes: a sys- The PREDIMED-NAVARRA Randomized Trial.
tematic review and meta-analysis of randomized Nutr Neurosci. 2011;14(5):195–201.
controlled dietary trials. PLoS One. 2014;9(7). 194. Lee J-Y, Jun N-R, Yoon D, et al. Association between
https://doi.org/10.1371/journal.pone.0103376. dietary patterns in the remote past and telomere
184. Banel DK, Hu FB. Effects of walnut consumption length. Eur J Clin Nutr. 2015;69(9):1048–52. https://
on blood lipids and other cardiovascular risk factors: doi.org/10.1038/ejcn.2015.58.
Part II
Gastrointestinal Tract
Fiber-Rich Dietary Patterns
and Colonic Microbiota in Aging 4
and Disease
Keywords
Dietary fiber • Dietary patterns • Microbiota • Short chain fatty acids
Butyrate • Symbionts • Pathobionts • C. difficile • Inflammatory bowel
disease • Colorectal cancer • Obesity • Type 2 diabetes • Metabolic
syndrome • Aging • Frailty • Centenarians • Mortality
Key Points healthy colonocytes and optimization of

colonic barrier defenses.
• Over the course of human evolution, a symbi- • Daily adequate fiber intake supports colonic
otic relationship was formed between fiber- microbiota health by increasing probiotic and
rich diets, the colonic microbiota and human decreasing pathogenic bacteria, lowering risk
health homeostasis. However, the emergence of endotoxemia and reducing colonic pH and
of the Western low fiber diet as a dominant bowel transit time, and contributing to greater
global dietary pattern has disrupted this sym- stool bulk to dilute potential toxic or carcino-
biotic relationship leading to an increased genic compounds or metabolites.
population risk for unhealthy aging, chronic • Fiber-rich healthy dietary patterns support the
diseases and premature death. health of colonic microbiota and their action in
• Fiber, a critical dietary factor in the mainte- protecting the colon from infections such as
nance of a healthy microbiota ecosystem, has C difficile, inflammatory bowel disease and
emerged in recent decades for its importance colorectal cancer; slowing the aging process by
in promoting colonic health, healthy aging, decreasing the risk of weight gain and obesity,
and reducing the risk of cardiometabolic type 2 diabetes, and metabolic syndrome; and
chronic disease and premature death. reducing the risk of frailty and premature death.
• Fiber is the primary dietary energy source of • In the elderly, high-fiber diets play an important
the microbiota bacteria and the breakdown role in establishing a healthy colonic microbiota
products of the resulting fiber fermentation associated with lower risk of frailty and longer
include short chain fatty acids such as butyrate life expectancy due in part to higher butyrate
which is the main energy source needed for production and lower risk of inflammaging.

120 4 Fiber-Rich Dietary Patterns and Colonic Microbiota in Aging and Disease
4.1 Introduction colonic anti-inflammatory functions. When fiber

becomes limiting, the colon pH increases to 6.5,
Over the last decade, there has been increased which coincides with a reduction of butyrate-pro-
human observational and clinical evidence sup- ducing bacteria and an increase in acetate- and
porting a strong relationship between the intake of propionate-producing Bacteroides-related bacte-
dietary patterns with adequate dietary fiber (fiber) ria. With the higher pH, there is an increased
for healthy microbiota and colonic function, opportunity for expansion of Proteobacteria,
which in turn supports the prevention of chronic which includes a wide variety of pathogens, such
disease risk (e.g., cardiovascular diseases, type 2 as Escherichia, Salmonella, Vibrio and increased
diabetes, cancer), and weight gain, and supports endotoxemia risk [21, 28]. Proteobacteria is
healthy aging and a longer life expectancy with emerging as a marker of chronic diseases risk
less frailty compared to results found with high (Fig. 4.1) [21]. The interplay between the level of
adherence to Western low fiber dietary patterns fiber intake and colonic microbiota health, includ-
[1–17]. The colonic microbiota serves a number ing a balance between symbionts (bacteria with
of important human biological functions includ- health-promoting functions) and pathobionts
ing aiding in the absorption of nutrients, synthesis (bacteria that potentially induce pathology), and
of vitamins, fermentation fiber to metabolically the increased colonic concentration of fermenta-
bioactive short chain fatty acids (SCFAs), promot- tion metabolites such as butyrate and their effects
ing barriers against pathogens, optimizing colonic on healthy aging and frailty are summarized in
and systemic immune function, and improving Fig. 4.2 [28–31]. The objective of this chapter is
cardiometabolic health and glycemic control [17– to comprehensively assess the effects of fiber-rich
22]. The symbiotic relationship between fiber-rich dietary patterns and colonic microbiota on healthy
diets, microbiota and human health developed aging and chronic disease.
over the course of human evolution [23–28].
There is a balance between microbiota health
and dysbiosis that depends on the level of fiber in 4.2 iber-Rich Dietary Patterns
F
the diet. Fiber is the primary source of microbiota- in Microbiota and Colonic
accessible carbohydrates for energy and fermen- Health
tation to metabolites such as SCFAs, primarily
butyrate, acetate, and propionate, which are cru- It has been hypothesized that individuals with high
cially involved in promoting a healthy microbiota adherence to the Western diet and sedentary life-
ecosystem [28]. The equilibrium between dietary styles are at increased risk of having colonic micro-
fiber, gut microbiota and SCFAs levels, especially biota dysbiosis (e.g., low bacterial gene count
butyrate are important in maintaining a healthy diversity, a higher pathogenic to healthy bacteria
colon (e.g., healthy colonic structure and acidic level, colonic and systemic inflammation) and
pH). Acetate and propionate tend to be rapidly unhealthy phenotype (e.g., overweight/obesity,
absorbed into circulation in exchange for bicar- metabolic syndrome, elevated cardiometabolic risk
bonate with a consequent rise in colonic pH. With biomarkers or prediabetes/diabetes) [9, 10, 23–31].
an adequate fiber intake, there is more likely to be An overview of healthy dietary patterns providing
higher levels of butyrate-producing bacteria (e.g., adequate fiber or 14 g per 1000 kcal, a Western
Roseburia spp., F. prausnitzii, Eubacterium rec- dietary pattern which provides approximately half
tale), which help maintain an acidic colon at 5.5 the adequate fiber level and a list of the top 50 fiber
pH as butyrate tends to maintain a presence in the rich foods are summarized in Appendices A and B.
colon [6]. Butyrate is also an important energy It is estimated that healthy fiber-rich dietary pat-
source for colonocytes and involved in the regula- terns generally contain 50–70% fermentable fiber
tion of cell proliferation and differentiation to available to help maintain a healthy microbiota
reinforcement of the colonic barrier to support ecosystem [28–32].
4.2 Fiber-Rich Dietary Patterns in Microbiota and Colonic Health 121
16
% Proteobacteria in the Colonic Microbiota

14
12
10
0
Healthy adult Gastric bypass Metabolic syndrome Clinical
surgery inflammation
and cancer
Fig. 4.1 Effect of health condition on the level of Proteobacteria in the colonic microbiota (adapted from [21])
Healthy Phenotype Colonic Microbial Ecology Dysbiotic Phenotype
Decreased Anti-inflammatory Pro-inflammatory Decreased

Colonic permeability species (Symbionts) species (Pathobionts) Fecal butyrate
Endotoxemia Fecal bulk
Inflammatory cytokines Bifidobacteria Escherichia coli Insulin sensitivity
Colonic pH Lactobacilli (Proteobacteria phyla) Satiety PYY/GLP-1
Lipogenesis Increased
F. prausnitzii Bacteroides spp
Insulin resistance Colonic permeability
(Clostridiaceae phyla) Clostridium difficile Endotoxemia
Increased
Fecal butyrate/bulk B. thethaiotamicron (Firmicutes phyla) Inflammatory cytokines
Insulin sensitivity Fecal calprotectin
Satiety PYY/GLP-1 Insulin resistance
Adiponectin Lipogenesis
Cancer cell cycle arrest Colonic pH
and cell apoptosis Secondary bile acids
Microbiota gene diversity
Bile acid deconjugation
Metabolic Health Metabolic Dysfunction

Decreased risk of colonic Western dietary pattern/lifestyle, low fiber Increased risk of colonic
infections/diseases, chronic intake, and excessive antibiotic use. infections/diseases (e.g.,
diseases (e.g., type 2 type 2 diabetes),
diabetes), overweight/ overweight/obesity,
obesity, metabolic metabolic syndrome, and
syndrome, and increased Healthy dietary pattern/ lifestyle, high unhealthy aging, frailty
odds of healthy aging and fiber intake, and prebiotics/probiotics. and premature death.
longevity.
Fig. 4.2 Overview of the effect of fiber intake and other lifestyle factors on human microbiota and metabolic and
immune health, chronic disease, aging and frailty (adapted from [28–31])
Table 4.1 summarizes major observational (EPIC) Study show that high adherence to
studies and RCTs on the effects of fiber-rich healthy dietary patterns rich in fiber are associ-
dietary patterns in improving colonic microbi- ated with reduced risk of colorectal cancer and
ota health and reducing risk of major intestinal inflammatory bowel disease compared to high
diseases [33–51]. Two large cohort studies, adherence to a Western or high sugar and soft
pooled data from the Nurses’ Health and Health drink pattern [33, 34]. Eight observational stud-
Professionals Follow-up Studies, and the ies show varying degrees of improved microbi-
European Prospective Investigation into Cancer ota ecosystem with higher adherence to plant
Table 4.1 Summary of healthy fiber-rich vs. Western dietary patterns observational studies and randomized
controlled trials (RCTs) on colonic microbiota health in aging and disease.
Observational Studies
Mehta et al. (2017). 137,217 subjects; 35% male; mean The prudent/healthy dietary pattern (rich
Examine the association age for men 54 years and women in whole-grain and fiber) significantly
between fiber rich prudent/ 46 years; 26–32 years of follow-up; lowered the multivariate risk of F.
healthy vs. Western dietary 1019 CRC cases with F. nucleatum. nucleatum positive CRC by 54%
patterns on colorectal cancer (p-trend = .003; highest vs. lowest diet
(CRC) risk in presence of score) but not F. nucleatum negative
intestinal F. nucleatum in tumor CRC, with insignificant lower risk by
tissue (Nurses’ Health Study 5% (p-trend =.47). This study supports
and Health Professionals the potential role of fiber-rich dietary
Follow-up Study; US) [33] pattern effects on the microbiota in
mediating the risk of CRC.
Racine et al. (2016). 366,351 participants with After excluding the first 2 years after
Investigate the association inflammatory bowel disease data; dietary analysis, there was a positive
between dietary pattern and 256 ulcerative colitis cases and 117 association for the ‘high sugar and soft
inflammatory bowel disease Crohn’s disease cases, with 4 drink’ pattern with a 68% increased risk
(European Prospective matched controls per case. of inflammatory bowel disease, which
Investigation into Cancer was only significant if there was both
(EPIC) study; nested matched high sugar and soft drink and low
case-control study) [34]. vegetable intake, which suggests a
relationship between vegetable fiber
intake and microbiota health.
Gutierrez-Diaz et al. (2016). 31 adults; 23 females/8 males; mean A higher MedDiet fiber level was
Assess the association age of 42 years; higher MedDiet directly associated with higher fecal
between the adherence to a score ≥ 4 [14 g fiber/1000 kcal] vs. SCFA concentrations (Fig. 4.3),
Mediterranean dietary lower MedDiet score < 4 increased abundance of phylum
(MedDiet) pattern, and its [11 g/1000 kcal]; all diets contained Bacteroidetes, family Prevotellaceae and
components, with fecal 850 mg phenolic/1000 kcal. genus Prevotella and decreased levels of
microbiota in a cohort of adults phylum Firmicutes and the genus
(Spanish) [35]. Ruminococcus. Higher cereal intake was
associated with higher levels of
Bifidobacterium and Faecalibacterium.
Wu et al. (2016). 21 urban adults;15 vegans/6 The differences in colonic microbiota
Compare measures of dietary omnivores; fiber intake for vegans between omnivores and vegans sampled
intake, gut microbiota (35 g/day) and omnivores (18 g/day); in an urban environment in the
composition and the plasma 3 × 24 hour dietary recalls. Northeastern USA were quite modest
metabolome between healthy but the vegan diet plasma metabolome
human vegans and omnivores differed markedly from omnivores
(US) [36]. because of the phytonutrients from
whole foods. Higher consumption of
fiber by vegans was not associated with
significantly higher levels of fecal
SCFAs.
De Filippis et al. (2015). 153 healthy adults; vegans, Subjects with good adherence to the
Evaluate the association vegetarians and omnivores had high MedDiet with higher fiber, fruits,
between the MedDiet, and the MedDiet adherence; 7-day weighed vegetables, and legumes intake had
microbiota and its metabolites food diary; fecal and urinary higher levels of fecal SCFAs, Prevotella
(Italian) [37]. samples. and some fiber degrading Firmicutes
compared to those with lower adherence
to the MedDiet. Western omnivore diets
with adequate fiber intake are not
necessarily detrimental to microbiota.
Low adherence to the MedDiet was
associated with higher urinary
trimethylamine oxide levels.
Matijasic et al. (2014). 60 adults; 31 vegetarians (11 Vegetarian diets were associated with
Examine the association lacto-vegetarians, 20 vegans); 29 higher levels of Bacteroides-Prevotella,
between omnivore vs. omnivores. Bacteroides thetaiotaomicron,
lacto-vegetarian and vegan Clostridium clostridioforme and
dietary patterns on Faecalibacterium prausnitzii compared
fecal microbiota composition to omnivores.
(Slovenia) [38].
Ou et al. (2013). 12 healthy African Americans, urban In their usual diets, African Americans
Examine if the influence of diet Western diets; 12 native Africans, consumed twice the protein, 3 times the
on colon cancer risk is mediated rural traditional diet; age- and dietary fat and less fiber than the native
by the microbiota through their sex-matched; mean age 58 years. Africans. The African Americans and
metabolites (African American native African adults had fundamentally
vs. Native African) [39]. different predominance of core
microbiota; Bacteroides vs. Prevotella,
respectively. The native Africans had
significantly higher total bacteria and
fecal SCFAs than the African
Americans. Stool butyrate
concentrations were significantly
correlated with the abundance of the
butyrate producers Clostridium cluster
IV and Clostridium cluster XIVa in the
native Africans. Fecal secondary bile
acid concentrations were higher in
African Americans.
Zimmer et al. (2012). 249 vegetarians or vegans; 249 Subjects on a vegan or vegetarian diet
Assess the effects of vegetarian control omnivores; matched for age showed significantly lower stool pH, and
vs. omnivorous diets on and gender; mean age 52 years. counts of E. coli and Enterobacteriaceae
microbiota composition than those on the omnivorous diet
(German) [40]. suggesting a healthy microbiota
environment (Fig. 4.4).
Kabeerdoss et al. (2012). 32 lacto-vegetarians and 24 Omnivores had an increased relative
Compare effects of lacto- omnivorous women from a similar abundance of Clostridium cluster XIVa
vegetarian and omnivorous diets social and economic background; bacteria, specifically Roseburia–E.
on the fecal microbiota of median age 19 years; median BMI rectale and butyryl-CoA-transferase
young women (Southern India) 21; macronutrient intake and gene, associated with microbial butyrate
[41]. anthropometric data were collected. production, compared with vegetarians.
Both diets had same median crude fiber
intake. The relative proportions of other
microbial communities were similar in
both groups.
(continued)
Wu et al. (2011). 98 healthy volunteers; collected diet The Bacteroides enterotype was highly
Investigate dietary pattern information using two food associated with animal protein, a variety
quality and its association with frequency questionnaires that of amino acids, and saturated fats, which
colonic microbiota profiles queried recent diet and habitual suggests a high level of meat intake
(US) [42]. long-term diet. (Western diet). In contrast, the
Prevotella enterotype was associated
with a higher carbohydrates/fiber based
diet more typical of agrarian societies or
a vegetarian diet.
Randomized Controlled Trials (RCTs)
Tap et al. (2015). Crossover RCT: Higher fiber diets increased microbiota
Assess the short-term effects of 19 healthy normal weight adults; 10 diversity, stability and promoted a
increased fiber intake on the females/9 males; age 19–25 years; higher Prevotella: Bacteroides ratio,
microbial composition (France) basal diet supplemented with 40 or increased fecal SCFAs and modulated
[43]. 10 g fiber/day; 5 days; 15-day the expression of microbiota metabolic
washout period. pathways such as glycan metabolism,
with genes encoding carbohydrate-active
enzymes for fiber, compared to the low
fiber diets, within 5 days. This was
particularly true when subjects switched
from their 10-g fiber diet to the 40-g
fiber/day diet.
O’Keefe et al. (2015). Crossover RCT: The food content of fiber and fat had
Evaluate the acute effects on 20 healthy middle aged African substantial effects on the subject’s
colonic microbiota after Americans and 20 rural South colonic microbiota and metabolome
switching diets of African Africans; first 2 weeks in their own within 2 weeks, which was associated
Americans (high colon cancer home environment, eating their usual with significant changes in mucosal
risk) and rural Africans (low food, and then again in-house they inflammation and proliferation
colon cancer risk) (US and switched diets so the African associated with potential colon cancer
South Africans) [44]. Americans were fed a high-fiber, risk. Diets with higher animal protein
low-fat African-style diet, and rural and fat and lower fiber consumption
Africans a high-fat low-fiber resulted in higher colonic secondary bile
Western-style diet under close acids, lower colonic short chain fatty
supervision for 2 weeks. acid quantities and higher mucosal
proliferative biomarkers of cancer risk in
the South Africans whereas higher fiber
and lower fat diets increased fecal
butyrate concentration and suppressed
secondary bile acid synthesis in the
African Americans.
David et al. (2014). Crossover RCT: Plant based diets increased saccharolytic
Compare the effects of plant vs. 10 US adults; 6 men and 4 women; bacteria and SCFAs fecal content
animal based diets on ages 21–33 years; BMI range from whereas animal-food based diets
microbiota (US) [45]. 19–32; 2 diets: a plant-based diet increased total count of bile-tolerant
(rich in whole-grains, legumes, microorganisms, decreased the levels of
fruits, and vegetables; 26 g Firmicutes able to ferment plant
fiber/1000 kcal); and an animal- polysaccharides, and increased levels of
based diet (consisting of meats, eggs, the products of amino acid fermentation
and cheeses; 0 g fiber/day); 5 days and Bilophila wadsworthia, known to
with a 6-day washout. elevate the risk of inflammatory bowel
disease. This study suggests that
microbiota can rapidly respond to large
changes in diet composition.
Fruits and Vegetables
Klinder et al. (2016). Dose-response, Parallel RCT: There was a dose effect for fruit and
Assess the impact of fruit and 122 UK participants; 60% female; vegetable intake on increasing C.
vegetable mean age 50 years; mean BMI 28; leptum-R. bromii/flavefaciens while a
intake on gut microbiota (UK) high-flavonoid and low flavonoid trend was reported for Bifidobacterium
[46]. fruit and vegetable intervention (p = .090) and Bacteroides/Prevotella
groups consumed 2, 4 and 6 portions (p = .070). Increased intake of fruit and
vs. habitual control diet; 6 weeks. vegetable portions high in flavonoids
were protective against the growth of
potentially pathogenic clostridia with a
negative correlation of (r = − 0.145) and
higher fiber intake was weakly
positively correlated with Bacteroides/
Prevotella (r = .091).
Whole-Grains
Heinritz et al. (2016). Pig-Human Model Parallel RCT: Significantly higher numbers of
Examine the effect of 2 diets 8 pigs were equally allotted to 2 lactobacilli, bifidobacteria and
with different levels of fiber and treatments, either fed a low-fat/ Faecalibacterium prausnitzii were found
fat on microbial composition high-fiber (whole wheat-grain type), in the feces of the whole-wheat grain
and metabolites (Netherlands) or a high-fat/low-fiber diet; 7 weeks; type (low-fat/high-fiber) diet fed pigs,
[47]. feces were sampled weekly. while pathogenic type
Enterobacteriaceae were significantly
increased in the high-fat/low-fiber diet
fed pigs. Significantly higher total and
individual fecal SCFAs levels, especially
butyrate, were found with whole-wheat
grain type diets vs. the low fiber diets
(Fig. 4.5).
Wang et al. (2016). Single Blind, Crossover RCT: The high MW β-glucan significantly
Evaluate the effect of β-glucan 30 hyperlipidemic adults; American increased Bacteroidetes and decreased
enriched breakfast cereals on Heart Association (AHA) diet plus 4 Firmicutes abundance compared to
microbiota composition and breakfasts containing 3 g high control. At the genus level, consumption
cardiovascular disease (CVD) molecular weight (MW) β-glucan, of 3 g/day high MW β-glucan increased
risk factors (Canada) [48]. 3 g and 5 g low MW β-glucan, vs. Bacteroides, tended to increase
refined wheat and rice (control); Prevotella but decreased Dorea, whereas
5-week study period; 4-week neither of the low MW β-glucan diets
washout. altered the microbiota composition. The
high MW β-glucan changes in
microbiota composition were
significantly correlated with shifts of
CVD risk factors, including reduced
BMI, waist circumference, blood
pressure, as well as triglyceride levels.
This study suggests the microbiota
health effects of high MW β-glucan.
Martinez et al. (2013). Crossover RCT: The barley whole-grain foods increased
Assess the effect 28 healthy subjects; 11 males and 17 overall microbiota diversity and
of whole-grains on both the females; mean age 26 years; mean specifically Roseburia, Bifidobacterium
colonic microbiome and human BMI 25; daily dose of 60 g of and Dialister, and the species
physiology (US) [49]. whole-grain barley (19 g fiber), Eubacterium rectale, Roseburia faecis
brown rice (4.4 g fiber), or an equal and Roseburia intestinalis. Additionally,
mixture of the two (11.5 g fiber); whole grain barley reduced IL-6
4-week treatments with 2-week associated with increased Dialister and
washout; fecal and blood samples decreased Coriobacteriaceae in the
were taken at baseline and after each microbiota. No significant differences
treatment period. were detected in fecal SCFAs but this
was because of colonic absorption.
(continued)
Carvalho-Wells et al. (2010). Double Blind Crossover RCT: Maize-whole grain breakfast cereal
Evaluate the effects of 32 subjects; 20 females/12 males; significantly increased levels of fecal
maize-whole grain and refined mean age 32 years; mean BMI 23; bifidobacteria compared with the control
breakfast cereal on the 48 g/day maize-whole grain cereal, which returned to baseline levels
microbiota (UK) [50]. breakfast cereal or refined grain after the washout period. There were no
cereal placebo; 3-week trial periods; statistically significant changes in fecal
3-week washout. SCFAs, bowel habit data, fasted lipids/
glucose, blood pressure, BMI and waist
circumference.
Costabile et al. (2008). Double Blind Crossover RCT: Whole grain wheat cereals significantly
Compare the effects of whole 31 volunteers; average age 25 years; increased the numbers of fecal
grain wheat breakfast cereal to 16 females/15 males; BMI 20–30; 2 bifidobacteria and lactobacilli compared
wheat bran on the human groups consuming daily either 48 g with wheat bran cereal. Ingestion of
microbiota (UK) [51]. of whole grain wheat or wheat bran both breakfast cereals resulted in a
breakfast cereals; 3-week study significant increase in ferulic acid
periods, 2-week washouts. concentrations in the blood but no
discernible difference in feces or urine.
No significant differences in fecal
SCFAs, fasting blood glucose, insulin,
total cholesterol, triglycerides or HDL-C
were observed upon ingestion of whole
grain wheat compared with wheat bran
breakfast cereals.
MedDiet Score < 4 MedDiet Score > = 4

3000
2500
Median Fecal SCFAs (µg/ml)
2000
1500
1000
500
0
Acetate Propionate Butyrate
Fig. 4.3 Effect of Mediterranean diet (MedDiet) score on fecal short-chain fatty acids (SCFAs) concentrations (adapted
from [35])
based diets compared to Western diets [35–42]. Western omnivore dietary pattern (Fig. 4.4)
A fiber-rich Mediterranean diet (MedDiet) pro- [40]. Nine RCTs strongly and consistently sup-
motes a higher ratio of probiotic to pathogenic port beneficial microbiota effects within 5 days
bacteria which functions by increasing fecal to 7 weeks of consuming fiber-rich healthy
concentrations of SCFAs (Fig. 4.3) [35]. Vegan dietary patterns compared to lower fiber
and vegetarian diets promote a higher ratio of Western diets [43–51]. A French crossover
probiotic to pathogenic bacteria compared to a RCT (19 adults; 40 g vs. 10 g fiber/day dietary
4.3 Fiber-Rich Dietary Patterns in Aging and Disease 127
Omnivore Vegetarian Vegan

10
9
Total germ Counts 108 8

7
6
5
4
3
2
1
0
Bifidobacteria E. Coli Enterobacc.
Fig. 4.4 Effect of vegetarian vs. omnivore dietary patterns on colonic microbiota bacteria composition (adapted
from [40])
patterns) showed that the 40-g fiber diets increased fecal SCFAs, especially butyrate,
increased microbiota diversity with a higher compared to a low fiber diet (Fig. 4.5) [47].
ratio of probiotic to pathogenic bacteria and
increased SCFAs concentrations compared to
10-g fiber diets within 5 days [43]. A US and 4.3 iber-Rich Dietary Patterns
F
South African crossover RCT (20 healthy mid- in Aging and Disease
dle aged African Americans and 20 rural South
Africans; diets were switched and African During the last few decades, there have been
Americans consumed high fiber and low fat numerous human studies showing a high
diets and rural Africans consumed high fat and degree of synergy between dietary patterns
low fiber diets for 2 weeks) found that after 2 with adequate fiber intake and healthy colonic
weeks the African Americans had increased microbiota with major beneficial effects on
fecal butyrate concentrations and suppressed promoting colonic health, weight control,
secondary bile acids synthesis and rural healthy aging, and lowering the risk of chronic
Africans had decreased butyrate concentrations disease and frailty.
and increased secondary bile acids synthesis
[44]. A UK fruit and vegetable dose response
RCT (122 participants mean age 50 years; 2, 4, 4.3.1 Colonic Health
and 6 servings daily vs. habitual control diet;
6 weeks) demonstrated that higher intake of 4.3.1.1 Clostridium difficile Infections
fruits and vegetables, especially associated Unnecessary use of antibiotics and excessive
with higher fiber and flavonoid intake, had dose hygienic precautions together with the Western
response type effects on promoting healthier diet have contributed to a decrease in adult
microbiota bacteria profiles, lower inflamma- colonic microbiota bacterial diversity [52, 53].
tion, and greater protection from pathogenic Over the past several decades, the growing aging
bacteria [46]. A pig-human colonic model RCT populations, increase in antibiotics usage and
(8 pigs; high fiber whole-grains and low fat diet resistance, high consumption of unhealthy low
vs. low fiber and high fat diet; 7 weeks) showed fiber Western diets, and growing prevalence of
that a higher whole-grain fiber diet significantly inflammatory bowel diseases and obesity has
High Fiber Diet Low Fiber Diet

600
550
Total SCFAs (mmol/kg dry fecal matter) 500
450
400
350
300
250
200
150
100
1 2 3 4 5 6 7
Duration (Weeks)
Fig. 4.5 Effect of fiber from wheat bran and cellulose on fecal short chain fatty acids (SCFAs) concentrations in pig
model study over 7 weeks (p = .002) (adapted from [47])
been correlated with recurrent Clostridium diffi- b iodiversity, especially healthy probiotic bacte-
cile (C. difficile) infections. C. difficile colonic ria, and increase fecal butyrate concentrations to
infections and associated gastrointestinal distress help protect against C. difficile associated dis-
and prolonged diarrhea rates are increasing in eases, colonic and systemic inflammatory and
both hospitals and communities worldwide, obesity-related metabolic syndromes [53]. Two
especially in elderly populations. C. difficile is an prebiotic fibers, fructo-oligosaccharides and
anaerobic, spore-forming bacterium with infec- polydextrose, have been shown to actively re-
tion symptoms varying among patients, ranging establish indigenous microbiota, particularly
from mild to severe diarrhea (>15 bowel those bacteria yielding large amounts of SCFAs,
movements/day) to death in severe cases [52]. C. and decrease gut pH, which can contribute to the
difficile transmission is a major problem in hospi- prevention of growth and toxin release by C. dif-
tals across the developed world, as its spores are ficile [56]. Also, the combination of prebiotic
highly resistant to routine cleaning agents, fiber with antibiotics appears to have a synergist
including alcohol-based hand washes and can effect in fighting C. difficile infections [57].
survive for months on aerobic surfaces (e.g.,
hospital walls, doors, surgical tools, cell phones, 4.3.1.2 Inflammatory Bowel Disease (IBD)
etc.) in spore form. When ingested, the multiple IBD, including Crohn’s disease and ulcerative
layers of the spore help protect it from stomach colitis, are chronic, life-long conditions that can
acids and digestive enzymes but in the colon the be treated but not cured and most cases are diag-
spores can be germinated into active cells by nosed in genetically susceptible individuals due
taurine-conjugated bile acid and colonize within to colonic microbiota dysbiosis and related
the gut microbiota to induce toxin associated inflammatory responses usually before age 35
intestinal damage and inflammation. [34, 58]. A 2016 EPIC nested match case-control
Low fiber diets and chronic antibiotic use are analysis (256 ulcerative colitis cases; 117 Crohns
commonly associated with C. difficile infected disease cases) observed that Western dietary pat-
patients, as both can lead to colonic microbiota terns rich in sugar and soft drinks increased risk
dysbiosis, resulting in low butyrate production of ulcerative colitis by 68% especially with low
[54, 55]. Healthy fiber rich dietary patterns and/ vegetable intake [34]. A 2015 meta-analysis
or prebiotic fiber can improve microbiotia (2 cohort studies, 1 nested case-control study, and
5 case-control studies) indicated a significant disease cases) observed that Western dietary pat-
inverse association between higher fiber intake terns rich in sugar and soft drinks increased risk
and Crohn’s disease risk by 56% and a margin- of ulcerative colitis by 68% especially with low
ally significant inverse association between vegetable intake [34].
higher fiber intake and ulcerative colitis risk by
20% [59]. In addition, a significant dose-response 4.3.1.3 Colorectal Cancer (CRC)
relationship was observed between fiber intake The potential importance of fiber and the colonic
and Crohn’s disease risk with a 13% lower risk microbiota in protecting against colorectal cancer
per 10 g of fiber intake. Mechanisms by which was first hypothesized in the early 1970s by Dr.
increased fiber intake may help in lowering the Burkitt, who observed lower rates of CRC among
risk of IBD include: improving colonic microbi- Africans who consumed a diet high in fiber [60].
ota health, which has a regulatory influence on Now, there are a number of strong biological
the colon immune response and maintenance of mechanisms associated with adequate fiber intake
immunological homeostasis; and effecting aryl and a lower risk of CRC. For example, the fiber
hydrocarbon receptor linking, a specific stimulus colon microbiota fermentation metabolite butyr-
to protect against IBD pathogenesis [59]. IBD is ate, a histone deacetylase (HDAC) inhibitor that
associated with decreased colonic microbiota can suppress the viability and progressive growth
diversity, reduced proportions of Firmicutes and of CRC cells [61, 62]. In 2011, a dose-response
increased levels of Proteobacteria and meta-analysis (16 cohort studies) showed each
Actinobacteria [58]. Specifically , with IBD pro- 10-g fiber increase in intake lowered risk of CRC
inflammatory bacteria (e.g., Escherichia and by 10% [63]. Also, the World Cancer Research
Fusobacterium) are increased and anti-inflam- Fund and American Institute of Cancer Research
matory bacteria (e.g., Faecalibacterium and continuous update concluded that there was con-
Roseburia) are decreased. The MedDiet has vincing evidence that increased fiber intake was
shown promise in improving IBD sysmptoms by protective against CRC risk [64]. A prospective
promoting strong immunomodulatory and epi- assessment of the Prostate, Lung, Colorectal, and
genetic effects that appear to normalize microbi- Ovarian Cancer Screening Trial (57,774 colorec-
ota in IBD patients by increasing DNA tal cancers, 16,980 adenomas, and 1667 recurrent
methylation in genes coding for inflammation adenoma cases; mean baseline age 63 years; flex-
[58]. A 2016 EPIC nested match case-control ible sigmoidoscopy at baseline; 3 or 5 years of
analysis (256 ulcerative colitis cases; 117 Crohns follow-up) found that participants consuming the
Any distal colon or rectal adenoma incidence (p =.003)

Distal colon cancer (p =.03)
1.1
0.9
Odds Ratio
0.8
0.7
0.6
0.5
0.4
< 9.9 g >= 9.9 to < 12.8 g >= 12.8 g
Total Fiber/1,000 kcal
Fig. 4.6 Association between total fiber density and distal colon or rectal adenoma or distal colon cancer risk (adapted
from [65])
Healthy controls Advanced colorectal adenoma

1.2
1
Fecal SCFAs (ug/L)
0.8
0.6
0.4
0.2
0
Acetate Propionate Butyrate
Fig. 4.7 Relationship between fecal short chain fatty acids (SCFAs) concentration in healthy subjects compared to
advanced colorectal adenoma subjects (p < .05) (adapted from [66])
highest fiber levels, especially cereal and fruit microbiome diversity [75–81]. The overweight and
fiber, had significantly reduced risk of colorectal obese phenotype is characterized by: more efficient
adenoma and distal colon cancer (Fig. 4.6) [65]. capacity for extracting metabolizable energy from
In a cross-sectional design study (688 subjects food; decreased abundance of butyrate-producing
eligible; 50% female; age > 50 years; healthy bacteria, in particular F. prausnitzii; increased pro-
control vs. advanced colorectal adenoma groups), inflammatory functions such as mucus degradation
it was observed that a high-fiber diet, higher fecal and production of endotoxins such as lipopolysac-
SCFAs and healthy gut microbiota were associ- charides (LPS) from gram negative endobacte-
ated with a reduced risk of advanced colorectal rium; and more up-regulated genes to manage
adenomas [66]. Healthy individuals with high oxidative stress, which are suggestive of increased
fiber intake had significantly higher levels of weight gain, inflammation, insulin resistance and
fecal butyrate than either healthy individuals with metabolic disease risk [76–83]. In contrast, the
low fiber intake or those individuals with lean, metabolically healthy phenotype is character-
advanced colorectal adenomas (Fig. 4.7). ized by: more diverse microbiota bacterial
gene functions; and increased abundance of
Bifidobacteria species and butyrate-producing bac-
4.3.2 Weight Control teria such as F. prausnitzii, a marker of a healthy
microbiota [75, 76, 82–88]. Increased fiber intake
Leaner individuals tend to consume higher fiber, is associated with a lean phenotype, despite fiber’s
healthier dietary patterns (low energy density and fermentation to SCFAs (an added source of energy),
fiber-rich foods such as fruits and vegetables) com- the evidence shows that increased fiber intake helps
pared to overweight and obese individuals who to protect against weight gain and obesity, as fiber
tend to consume higher energy dense foods and typically has only half the energy density of refined
Western lower fiber diets [67–74]. Higher fiber carbohydrates [85]. Colonic microbiota tend to be
diets are also associated with healthier, more clustered into two major enterotypes with the
diverse colonic microbiota ecosystems compared Bacteroides-dominate enterotype reported to be
to high energy dense, low fiber dietary patterns characteristic of individuals consuming more pro-
[28–52]. Studies consistently report that individu- tein and animal fat (Western diet), whereas the
als with low microbiota diversity are more often Prevotella-dominate enterotype appears to be char-
associated with higher body fat mass, insulin resis- acteristic of subjects consuming more fiber-rich
tance, dyslipidemia, and low-grade systemic whole plant foods [86]. A 2017 RCT (62 subjects
inflammation compared to individuals with higher with elevated waist size; ad libitum New Nordic
Diet high in fiber/wholegrain (43.3 g fiber) vs the such as retinopathy, neuropathy, shorter life
Average Danish Diet (28.6 g fiber); approx. 65% expectancy, and causes higher medical costs [93,
women; mean age approx 45 years; 26 weeks dura- 94]. Growing evidence suggests that colonic
tion) found individuals with a higher Prevotella microbiota may play a role in the pathogenesis
spp. to Bacteroides spp. (P/B ratio) had effective of both type 1 and 2 diabetes [95–97].
body fat loss on a diet rich in fiber and whole grain Overweight and obesity are initiating factors for
compared to the body fat loss observed in individu- diabetes risk because of increased low grade
als with a low P/B ratio [86]. Specifically, among inflammation, which reaches tissues involved in
individuals with high P/B the New Nordic Diet energy metabolism regulation, such as the liver,
resulted in a significant 3.15 kg larger body fat loss adipose tissue, and muscles and interferes with
compared to an average Danish diet. Among indi- cellular insulin signals leading to insulin resis-
viduals with low P/B ratio the New Nordic Diet tance. This low-grade inflammation and insulin
resulted in an insignificant 0.88 kg larger body fat resistance is known to be intimately linked to
loss compared to an average Danish diet. dysbiosis of the colonic microbiota [97].
Individuals with high P/B ratio were more suscep- Proteobacteria have been shown to be signifi-
tible to lose body fat on diets high in fiber diet by cantly higher in diabetic individuals compared
2.27 kg than subjects with a low P/B ratio. Potential to healthy persons and positively correlated with
mechanisms associated with the abundance of plasma glucose [98]. Two other studies also
Prevotella could include: differences in fiber-utili- showed that diabetic subjects were characterized
zation capacity and ratio of SCFAs produced, the by a reduction in the number of Clostridiales
secretion of gastrointestinal hormones affecting bacteria (Roseburia species and Faecalibacterium
appetite, and post-prandial blood glucose and insu- prausnitzii), which produce butyrate [89, 99].
lin responses [86]. Fiber directly or by its SCFA Adequate fiber intake may help to protect against
metabolites can modify some gut hormones that the adverse effects of microbiota dysbiosis on
regulate satiety and energy intake, thus also affect- insulin resistance associated with increased sys-
ing lipid metabolism and energy expenditure. temic inflammation related to colonic permea-
Increased fiber intake may affect gut hormones bility of lipopolysaccharides (LPS) from
such as ghrelin, glucagon-like peptide 1, peptide gram-negative pathogenic bacteria, low incretin
YY, and cholecystokinin. Dietary fiber is also secretion and fecal butyrate production, and
known to affect adipose tissue expression and macrophage influx into visceral fat tissue, and
secretion of a variety of adipocytokines, including activation of hepatic Kupffer macrophage cells
adiponectin, leptin, tumor necrosis factor-α, and [28–31, 89, 95]. Dietary fiber’s effects on the
interleukin-6, which can influence obesity, insulin microbiota plays a significant role in reducing
resistance, and hyperlipidemia. SCFAs, especially colonic mucosa permeability and regulation of
butyrate and propionate, can act to promote a lean energy homeostasis, thereby reducing the risk of
phenotype by: acting as ligands for free fatty acids metabolic endotoxemia or persistent low-grade
receptors, which increases expression and secre- inflammatory response, insulin resistance asso-
tion of satiety hormones glucagon-like-peptide 1 or ciated with diabetes or metabolic syndrome
peptide YY and leptin from adipocytes to reduce [100]. Four RCTs provide insights for the
hunger and promote higher adiponectin and lower actions of fiber-rich dietary patterns such as veg-
insulin resistance [89–92]. etarian diets [101, 102] and diets rich in resistant
starch in promoting a healthy microbiota eco-
system and controlling diabetes risk [103, 104].
4.3.3 Type 2 Diabetes (Diabetes)
Diabetes, a consequence of elevated blood glu- 4.3.4 Metabolic Syndrome

cose and a deficit in the secretion and action of
insulin, increases the risk for other chronic ill- Metabolic syndrome, which is associated with
nesses such as cardiovascular and renal disease, combinations of elevated blood pressure, dyslipid-
tuberculosis, and serious health complications emia (defined by increased triglycerides and
reduced high-density lipoprotein cholesterol), high [117]. Adequate fiber intake is associated with a
fasting glucose and/or central obesity, occurs in healthier microbiota diversity, including symbi-
20–40% of the worldwide adult population [105]. onts such as Prevotella, Lactobacillales,
The colonic microbiota can have a major influence Christensenellaceae, Bifidobacteria groups and
on the pathogenesis of metabolic syndrome, which the butyrate-producing bacteria such as Roseburia
can increase the rate of unhealthy aging [106]. spp., F. prausnitzii, and Akkermansia muciniphila
Prospective studies suggest that Western dietary [28–31, 115, 119]. This increased fiber intake pro-
patterns, rich in processed meats, snacks, baked vides the colon with the butyrate and lower pH
desserts and sugar sweetened beverages are associ- needed to maintain a strong barrier to fight patho-
ated with higher metabolic syndrome risk, whereas bionts and inflammaging; compared to a low fiber
healthy dietary patterns rich in vegetables, fruits, intake, which supports lower bacterial gene count
whole-grains, nuts and fish are associated with a diversity, fewer butyrate-producing bacteria,
reduced risk of metabolic syndrome [107]. Several increased acetate- and propionate-producing
RCTs suggest that a fiber-rich dietary pattern and Bacteroides-related bacteria, and optimal colonic
healthy microbiota ecosystem can protect against pH levels. Several studies show that individuals
metabolic syndrome [108–112]. For example, an with a low microbiota diversity are characterized
Italian RCT (54 adults with metabolic syndrome; by higher risk of weight gain or central adiposity,
31 females/23 males; mean age 58 years; mean insulin resistance, dyslipidemia and inflammation
BMI 31.5; whole-grain group with 29 g cereal when compared with high bacterial diversity indi-
fiber/day vs. control group with 12 g cereal fiber/ viduals [78, 79]. Increased fiber intake from the
day; 12 weeks) found that the whole-grain diet regular diet has been shown to be directly associ-
increased fasting plasma propionate, which corre- ated with healthier microbiota profiles in compari-
lated with lower postprandial insulin concentra- son to lower fiber diets, regardless of a person’s
tions [108]. Another study suggests that the age [119]. A meta-analysis including 14 RCTs
absorption of colonic propionate helps to prevent showed that an increase of 8 g/day of fiber com-
body weight gain and intra-abdominal fat accretion pared with lower fiber control diets significantly
in overweight adults by stimulating release of the reduced systemic C-reactive protein by 0.5 mg/L
satiety gut hormones peptide YY and glucagon like [120]. A 1999–2010 US National Health and
peptide-1 [109]. Although the pathogenesis of the Nutrition Examination Survey (NHANES) analy-
metabolic syndrome is complex and not fully sis observed that increasing levels of fiber intake
understood, adherence to fiber-rich dietary patterns were found to significantly reduce the risk of ele-
can help to reduce the risk of developing metabolic vated CRP levels and to reduce the risk of meta-
syndrome by promoting a healthy colonic micro- bolic syndrome and obesity, two microbiota
biota, preventing central obesity, and promoting related health conditions associated with acceler-
healthier systemic lipoprotein, inflammatory and ated inflammaging (Fig. 4.8) [121]. A cross-
blood pressure profiles and increasing insulin sen- sectional analysis of the Nurses’ Health Study
sitivity [112]. (2284 women; mean age 59 years; mean BMI 26;
87% postmenopausal) showed fiber was positively
associated with leukocyte telomere length with a
4.3.5 Healthy Aging and Longevity significant increase in telomere length by
0.19 units between the extremes of fiber intake,
The adherence to healthy fiber-rich dietary pat- after multivariate adjustment [122]. A Canadian
terns and/or prebiotic fiber supplementation sup- cost -of-illness analysis estimated that each addi-
ports a healthy microbiota, which can have a major tional 1 g fiber/day resulted in an annual $3–51
effect on the body’s age processes, to promote million in savings in type 2 diabetes care and
healthy aging and prevent premature death [113– $5–92 million in cardiovascular disease care
118]. Healthy aging may be defined as the absence [123]. A meta-analysis (25 cohort studies;
of chronic disease, lack of physical disability, 1,752,848 midlife individuals; average 12.4 years
good mental health, and good social engagement of follow-up) suggests that fiber is inversely
a ssociated with mortality risk (Fig. 4.9) [124]. The (multivariate adjusted) [125]. An EPIC prospec-
large US National Institutes of Health (NIH)- tive study (452,717 men and women; mean age
AARP Diet and Health Study (567,169 men and 51 years; mean BMI 25.5; mean 12.7 years of fol-
women; mean age 62; mean BMI 27; 9-years follow-up) found an inverse association with total
low-up; 20,126 deaths in men and 11,330 deaths in mortality and circulatory morality risk of 10%
women) found that increased fiber intake by 15 g/ reduction per 10 g fiber increased intake [126].
day significantly reduced all-cause mortality rates Fiber-rich foods such as whole-grains, fruits, and
by 22% in both men and women, and CVD mor- vegetables have been consistently shown to reduce
tality in men by 24% and women by 34% all-cause mortality risk [127, 128].
hs-CRP > 3.0 mg/L Metabolic Syndrome Obesity

1.1
0.9
Risk Ratios
0.8
0.7
0.6
< 8.1 8.1-12 12.1-16.2 16.3-22.4 > 22.5
Dietary Fiber Intake (g/day)
Fig. 4.8 Association between increasing fiber intake on risk of elevated hs-C-reactive protein (hs-CRP) levels
and risk of metabolic syndrome and obesity from the 1999–2010 US National Health and Nutrition Examination Survey
(p-trend <.001; multivariate adjusted) (adapted from [121])
All cause Total cardiovascular disease Cancer

1.1
Harzard Ratio for Mortality
0.9
0.8
0.7
<20 g 20-30 g > 30 g
Dietary Fiber Intake (g/day)
Fig. 4.9 Association between level of dietary fiber intake and all-cause, cardiovascular disease, and total cancer
specific mortality risks based on a meta-analysis of 25 prospective cohort studies (adapted from [124])
4.4 Frailty and Longevity lactobacilli, Bacteroides-Prevotella or F. prausnit-

zii, increases in Enterobacteriaceae and a major
In elderly adults, a fiber-rich dietary pattern, espe- shift from Firmicutes to Bacteroidetes phyla [130,
cially in long-stay care facilities, is important to 135–137].
reduce the risk of frailty by avoiding the adverse The centenarian phenotype has a unique and
effects of standardized low fiber meals, which can complex microbiota composition, which counter-
subsequently reduce microbiota diversity and balances inflammaging processes and is neces-
increase dysbiosis associated accelerated inflam- sary to establish a microbiota ecosystem for
maging related frailty [129, 130]. Table 4.2 sum- exceptionally healthy longevity. A cross-sectional
marizes the findings from nine cross- sectional study (24 semi-supercentenarians, average age
studies on microbiota composition profiles asso- 106 years, 18 females, 6 males); 15 young adults
ciated with unhealthy and frail elderly and lon- (average age 31 years, 8 females, 7 males)
gevity/centenarian phenotypes [131–137]. A observed that longevity adaptation involves an
cross-sectional study (371 elderly subjects; mean enrichment of health-associated microbiota [132].
age 78 years; stratified by community-dwelling, Extremely long-living people experience an
outpatient day hospitals, in short-term rehabilita- increase in several health-associated bacteria,
tion care or in long-term care facilities) observed especially from the family Christensenellaceae,
that elderly from either the community or long- which increases in prevalence in centenarians.
term care facilities consuming unhealthy diets Health-associated bacteria are inversely corre-
(e.g., low in fiber and high in sugars and fats) had lated with BMI, positively associated with
low microbiota diversity and increased signs of improved renal function, and significantly inter-
biological aging and frailty [131]. Additionally, act with the human genome. Additionally,
elderly in long-term care facilities had a gradual Akkermansia and Bifidobacterium, which have
change in their core healthier community-based well-known health benefits associated with
microbiota composition over 18 months to a new immunomodulation, protection against inflam-
core elderly type microbiota composition associ- mation, and promotion of a healthy metabolism,
ated with colonic microbiota dysbiosis and frailty are increased in centenarian’s microbiota. An
(in part due to lower fiber standardized diets). In Italian cross-sectional study observed that cente-
another cross-sectional study (178 elderly adults; narians had a unique microbiota species pattern
mean age 78 years; stratified by their current liv- that significantly differed from the typical adult-
ing situation: community-dwelling, outpatients, like pattern [134]. In this study, it was shown that
short-term hospitalized subjects, or long-term the centenarians, as all aging individuals, have an
care residents) it was observed that 83% of the increased microbiota pathogenic population asso-
long-term care elderly consumed less diverse and ciated with an increased risk of inflammaging.
lower fiber diets compared to elderly living in a However, they also have a uniquely effective
residential community [129]. The long-term care healthy microbiota bacterial population with
elderly experienced microbiota dysbiosis, and higher than normal anti-inflammatory activity.
accelerated frailty. The residential community- The major centenarian difference was found to be
living elderly consumed more diverse/higher fiber a marked restructuring of their anti-inflammatory
diets, had healthier microbiota with a higher butyrate producing bacteria population from F.
diversity index with a higher proportion of prauznitzii to E. limosum (Clostridium cluster
Firmicutes/Lachnospiraceae, and higher levels of XV), which was 15-times higher than levels
fecal SCFAs and lower rates of frailty. Other stud- found in typical older adults. In a Chinese study,
ies indicate that the microbiota composition pro- high-fiber diets were associated with changes in
file of unhealthy aging or frail elderly is the colonic microbiota of centenarians, suggest-
generally associated with reduced levels of ing that a high-fiber diet has a role in establishing
healthy probiotic bacteria and increased levels of a new structurally balanced microbiota that may
pathogenic bacteria; such as marked reductions in benefit the health of centenarians [133].
4.4 Frailty in Nonagenarian and Centenarian Phenotypes 135
Table 4.2 Summary of cross-sectional studies on microbiota composition and frailty in elderly individuals
Reference/Objective Cohort/Method Results
Jeffery et al. (2016). 371 elderly subjects; mean age Elderly from either the community or
Identify dietary intervention 78 years; ranging from 64 to long-term care facilities consuming unhealthy
and bacteriotherapy strategies 102 years; community- diets (e.g., low in fiber and high in sugars and
for promoting health in older dwelling, attending outpatient fats) had low microbiota diversity and
people (Ireland) [131]. day hospitals, in short-term increased signs of biological aging and frailty.
rehabilitation care (< 6 weeks) Elderly in long-term care facilities had a
or in long-term care facilities. gradual change in their core community-based
microbiota composition over approximately18
months to a new core elderly type microbiota
composition associated with dysbiosis and
frailty. Elderly living in the community had
more antibiotic associated microbiota loss but
also more recovery following antibiotic
treatment than long-term care elderly.
Biagi et al. (2016). 24 supercentenarians (average Longevity adaptation appears to involve
Assess the microbiota age 106 years,18 females, 6 enriched health-associated microbiota.
associated with extreme males); 15 young adults Extremely long-living people experience a
longevity (Italy) [132]. (average age 31 years, 8 parallel increase in several health-associated
females, 7 males). bacteria especially from the family
Christensenellaceae, which is increased in
terms of both relative abundance and
prevalence in centenarians as health-associated
bacteria inversely correlated with BMI,
positively associated with improved renal
function and with significant human genome
interaction. A characteristic of centenarian’s
microbiota are increased levels of
Akkermansia and Bifidobacterium, which
have well-known health benefits associated
with immunomodulation, protection against
inflammation, and promotion of healthy
metabolic homeostasis.
Wang et al. (2015). 24 subjects, 8 aged 100– This study suggests that high-fiber diets can
Examine the factors associated 108 years, 8 aged rural elderly establish a new structurally balanced colonic
with the microbiota 85–99 years, and 8 urban microbiota architecture that may benefit the
composition of centenarians elderlies, aged 80–92 years. health of centenarians. The abundance of
(China) [133]. Bacteroidales and Lachnospiraceae was lower,
but Ruminococcaceae was higher with
high-fiber diet intake.
Claesson et al. (2012). 178 elderly adults; mean age Elderly in long-term care facilities had
Assess the relationship 78 years; stratified by their significantly higher frailty test scores
between diet, microbiota and current living situation: compared to elderly living in a residential
health status in elderly subjects community-dwelling, community. The long-term care elderly
(Ireland) [129]. outpatients, short-term consumed less diverse/lower fiber diets, had
hospitalized subjects and microbiota dysbiosis (higher proportion of
long-term care residents, Bacteroidetes and lower fecal SCFAs), and
community-dwelling elderly accelerated frailty. The residential community-
showed significantly healthier living elderly consumed more diversity/higher
microbiota composition than fiber diets, had healthier microbiota (higher
long-term care residents; 98% diversity index with a higher proportion of
of community and day hospital Firmicutes/Lachnospiraceae, and high levels
subjects consumed low fat/high of fecal SCFAs) and lower rates of frailty.
fiber diets and 83% long-stay
subjects consumed high fat/low
fiber diets.
(continued)
Claesson et al. (2011). 161 subjects aged ≥65 years, The microbiota of each individual constituted
Examine the composition, approx. equal females and a unique profile that was separable from all
variability, and temporal males; 9 healthy younger others. In 68% of the elderly individuals, the
stability of the colonic control subjects, 5 female/4 microbiota was dominated by phylum
microbiota of the elderly males, with ages between 28 Bacteroides. The proportions of some phyla
(Ireland) [130]. and 46 years. and genera associated with disease or health
also varied dramatically, including
Proteobacteria, Actinobacteria, and
Faecalibacteria. The core microbiota of elderly
subjects was distinct from that previously
established for younger adults, with a greater
proportion of Bacteroides spp. and distinct
abundance patterns of pathobiontic
Clostridium groups.
Biagi et al. (2010) 84 subjects belonging to Young (30 years old) and older adults
Determine differences in the different age groups; 21 (70 years old) had very similar overall colonic
microbiota of adults with centenarians (20 women, 1 microbiota composition, dominated by
increasing age (Italy) [134]. man; average 101 years); 22 Bacteroidetes and Firmicutes (95% of the
elderlies (11 women, 11 men; microbiota), and smaller fractions of
average 73 years; genetically Actinobacteria, and Proteobacteria. Although
unrelated to the centenarians); centenarians’ microbiota was dominated by
21 elderly people (10 women, Bacteroidetes and Firmicutes (93% of
11 men; average 68 years; microbiota), the specific microbiota species
offspring of the centenarians); pattern significantly differed from the typical
20 young adults (9 women, 11 adult-like pattern. Some of the key differences
men; average 31 years). from typical adults are: lower diversity in
terms of species composition, a rearrangement
in the Firmicutes population, an enrichment in
facultative anaerobes, notably pathobionts
associated with increased systemic
inflammatory markers (inflammaging),
marked decrease in the anti-inflammatory F.
prauznitzii and different types of butyrate
producers such as E. limosum (Clostridium
cluster XV), which is approximately 15-fold
higher, which may be a primary anti-
inflammatory longevity bacteria. Centenarians
have a unique and complex microbiota
composition which counterbalances
inflammaging processes and is necessary to
establish a microbiota ecosystem for
exceptional healthy longevity.
Mariat et al. (2009). 21 adults (25–45 years old); 21 The ratio of Firmicutes to Bacteroidetes
Evaluate changes in human infants (3 weeks to 10 months evolves during different life stages with mean
microbiota Firmicutes and old); 20 elderly subjects ratios for infants, adults and elderly being 0.4,
Bacteroidetes ratio with aging (70–90 years old). 11 and 0.6, respectively. Elderly subjects
(France) [135]. exhibited higher levels of E. coli and
Bacteroidetes.
Van Tongeren et al. (2005). 23 elderly subjects; median age Fecal samples from elderly with high frailty
Examine the relationship 86 years; 83% females; living in scores showed a significant reduction in the
between fecal microbiota the same long-term care center number of lactobacilli by 26-fold, Bacteroides/
composition and frailty in the and receiving the same diet; 2 Prevotella by 3-fold and the Faecalibacterium
elderly (The Netherlands) groups low frailty vs. high prausnitzii by4-fold whereas the number of
[136]. frailty score; no antibiotics for Enterobacteriaceae pathogenic bacteria was
4 weeks. significantly higher by 7-fold.
4.4 Frailty in Nonagenarian and Centenarian Phenotypes 137
Bartosch et al. (2004). 35 healthy elderly subjects The primary difference between healthy
Characterize bacterial (mean age 71 years; all females; community living elderly and both patient
communities in feces from living in the local community); cohorts was a marked reduction in the
healthy 38 elderly hospitalized patients Bacteroides-Prevotella group following
elderly volunteers and (mean age 81 years; 60% hospitalization. Reductions in bifidobacteria,
hospitalized elderly patients males); 21 elderly hospitalized Desulfovibrio spp., Clostridium
(UK) [137]. patients receiving antibiotic clostridiiforme, and Faecalibacterium
treatment (mean age 81 years; prausnitzii were also found in the hospitalized
79% males). patients. Antibiotic treatment resulted in
further reductions in the numbers of bacteria
and their prevalence with complete
elimination of certain bacterial communities in
some patients. Enterobacteria counts increased
in the hospitalized patients who did not
receive antibiotics and Enterococcus faecalis
proliferated in antibiotic treated patients. Total
bacteria counts were lower in the stool
samples of the two groups of hospitalized
patients compared to the community living
elderly.
Conclusions Fiber is the primary dietary energy source of

The colonic microbiota provides a number the microbiota bacteria and the breakdown
of important human biological functions products of the resulting fiber fermentation
including aiding in nutrient absorption, syn- include short chain fatty acids such as butyr-
thesis of vitamins, fermentation of fiber to ate which is the main energy source needed
SCFAs, and promoting barriers against for healthy colonocytes and optimization of
pathogenic bacteria, secondary bile acid colonic barrier defenses. Daily adequate
transformation, and inflammation. The fiber intake supports colonic microbiota
colonic microbiota also defends against health by increasing probiotic and decreas-
colonic tumor initiation and progression, ing pathogenic bacteria, lowering risk of
and promotes cardiometabolic health (e.g., endotoxemia and reducing colonic pH and
increased insulin sensitivity and satiety hor- bowel transit time, and contributing to
mones, and lower cardiometabolic risk fac- greater stool bulk to dilute potential toxic or
tors). Over the course of human evolution, a carcinogenic compounds or metabolites.
symbiotic relationship was formed between Fiber-rich healthy dietary patterns supports
fiber-rich diets, the colonic microbiota and colonic microbiota health and their action in
human health homeostasis. However, the protecting the colon from infections such as
emergence of the Western low fiber diet as a C difficile, inflammatory bowel disease and
dominant global dietary pattern has dis- colorectal cancer; slowing the aging process
rupted this symbiotic relationship leading to by decreasing the risk of weight gain and
an increased population risk for unhealthy obesity, type 2 diabetes, and metabolic syn-
aging, chronic diseases and premature death. drome; and reducing the risk of frailty and
Fiber, a critical dietary factor in the mainte- premature death. In the elderly, high-fiber
nance of a healthy microbiota ecosystem, diets play an important role in establishing a
has emerged in recent decades for its impor- healthy colonic microbiota associated with
tance in promoting colonic health, healthy lower risk of frailty and longer life expec-
aging, and reducing the risk of cardiometa- tancy due in part to higher butyrate produc-
bolic chronic disease and premature death. tion and lower risk of inflammaging.
Appendix A
Comparison of Western and Healthy Dietary Patterns per 2000 kcal (Approximated Values)
Healthy Healthy vegetarian
Western dietary USDA base DASH diet Mediterranean pattern (Lact-ovo
Emphasizes Refined grains, Vegetables, Potassium rich Whole grains, Vegetables, fruit, Plant foods:
low fiber foods, fruit, whole- vegetables, vegetables, fruit, whole-grains, vegetables,
red meats grain, and fruits, and dairy products, legumes, nuts, fruits, whole
sweets, and solid low-fat milk low-fat milk olive oil, and seeds, milk grains, nuts,
fats products moderate wine products, and soy seeds, and soy
foods foods
Includes Processed meats, Enriched grains, Whole-grain, Fish, nuts, Eggs, non-dairy Non-dairy
sugar sweetened lean meat, fish, poultry, fish, seeds, and milk alternatives, milk
beverages, and nuts, seeds, and nuts, and seeds pulses and vegetable oils alternatives
fast foods vegetable oils
Limits Fruits and Solid fats and Red meats, Red meats, No red or white No animal
vegetables, added sugars sweets, and refined grains, meats, or fish; products
whole-grains sugar-sweetened and sweets limited sweets
beverages
Estimated Nutrients/Components
Carbohydrates 51 51 55 50 54 57
(% Total kcal)
Protein (% Total 16 17 18 16 14 13
kcal)
Total fat (% 33 32 27 34 32 30
Total kcal)
Total kcal)
Unsat. fat (% 22 25 21 24 26 25
Total kcal)
Fiber (g) 16 31 29+ 31 35+ 40+
Potassium (mg) 2800 3350 4400 3350 3300 3650
Vegetable oils 19 27 25 27 19–27 18–27
(g)
Sodium (mg) 3600 1790 1100 1690 1400 1225
Plant Food Groups
Fruit (cup) ≤1.0 2.0 2.5 2.5 2.0 2.0
Vegetables (cup) ≤1.5 2.5 2.1 2.5 2.5 2.5
Whole-grains 0.6 3.0 4.0 3.0 3.0 3.0
(oz.)
Legumes (oz.) − 1.5 0.5 1.5 3.0 3.0+
Nuts/Seeds (oz.) 0.5 0.6 1.0 0.6 1.0 2.0
Soy products 0.0 0.5 − − 1.1 1.5
(oz.)
Nutrient Database for Standard Reference, Release 27. https://www.ars.usda.gov/nutrientdata. Accessed 17 February
2015.
Services and the Secretary of Agriculture. Appendix E-3.7: Developing vegetarian and Mediterranean-style food
patterns. 2015;1–9.
gov/ Accessed 8.22.2015.
Appendix B 139
Appendix B
Fifty High Fiber Whole or Minimally Processed Plant Foods Ranked by Amount of Fiber per Standard
Food Portion Size
Calories Energy density
Food Standard portion size Dietary fiber (g) (kcal) (calories/g)
High fiber bran ready-to-eat-cereal 1/3–3/4 cup (30 g) 9.1–14.3 60–80 2.0–2.6
Navy beans, cooked 1/2 cup cooked (90 g) 9.6 127 1.4
Small white beans, cooked 1/2 cup (90 g) 9.3 127 1.4
Shredded wheat ready-to-eat cereal 1–1 1/4 cup 5.0–9.0 155–220 3.2–3.7
(50–60 g)
Black bean soup, canned 1/2 cup (130 g) 8.8 117 0.9
French beans, cooked 1/2 cup (90 g) 8.3 114 1.3
Split peas, cooked 1/2 cup (100 g) 8.2 114 1.1
Chickpeas (Garbanzo) beans, canned 1/2 cup (120 g) 8.1 176 1.4
Lentils, cooked 1/2 cup (100 g) 7.8 115 1.2
Pinto beans, cooked 1/2 cup (90 g) 7.7 122 1.4
Black beans, cooked 1/2 cup (90 g) 7.5 114 1.3
Artichoke, global or French, cooked 1/2 cup (84 g) 7.2 45 0.5
Lima beans, cooked 1/2 cup (90 g) 6.6 108 1.2
White beans, canned 1/2 cup (130 g) 6.3 149 1.1
Wheat bran flakes ready-to-eat cereal 3/4 cup (30 g) 4.9–5.5 90–98 3.0–3.3
Pear with skin 1 medium (180 g) 5.5 100 0.6
Pumpkin seeds. Whole, roasted 1 ounce (about 28 g) 5.3 126 4.5
Baked beans, canned, plain 1/2 cup (125 g) 5.2 120 0.9
Soybeans, cooked 1/2 cup (90 g) 5.2 150 1.7
Plain rye wafer crackers 2 wafers (22 g) 5.0 73 3.3
Avocado, Hass 1/2 fruit (68 g) 4.6 114 1.7
Apple, with skin 1 medium (180 g) 4.4 95 0.5
Green peas, cooked (fresh, frozen, 1/2 cup (80 g) 3.5–4.4 59–67 0.7–0.8
canned)
Refried beans, canned 1/2 cup (120 g) 4.4 107 0.9
Mixed vegetables, cooked from frozen 1/2 cup (45 g) 4.0 59 1.3
Raspberries 1/2 cup (65 g) 3.8 32 0.5
Blackberries 1/2 cup (65 g) 3.8 31 0.4
Collards, cooked 1/2 cup (95 g) 3.8 32 0.3
Soybeans, green, cooked 1/2 cup (75 g) 3.8 127 1.4
Prunes, pitted, stewed 1/2 cup (125 g) 3.8 133 1.1
Sweet potato, baked 1 medium (114 g) 3.8 103 0.9
Multi-grain bread 2 slices regular (52 g) 3.8 140 2.7
Figs, dried 1/4 cup (about 38 g) 3.7 93 2.5
Potato baked, with skin 1 medium (173 g) 3.6 163 0.9
Popcorn, air-popped 3 cups (24 g) 3.5 93 3.9
Almonds 1 ounce (about 28 g) 3.5 164 5.8
Whole wheat spaghetti, cooked 1/2 cup (70 g) 3.2 87 1.2
Sunflower seed kernels, dry roasted 1 ounce (about 28 g) 3.1 165 5.8
Orange 1 medium (130 g) 3.1 69 0.5
Banana 1 medium (118 g) 3.1 105 0.9
Oat bran muffin 1 small (66 g) 3.0 178 2.7
Vegetable soup 1 cup (245 g) 2.9 91 0.4

Dates 1/4 cup (about 38 g) 2.9 104 2.8
Pistachios, dry roasted 1 ounce (about 28 g) 2.8 161 5.7
Hazelnuts or filberts 1 ounce (about 28 g) 2.7 178 6.3
Peanuts, oil roasted 1 ounce (about 28 g) 2.7 170 6.0
Quinoa, cooked 1/2 cup (90 g) 2.7 92 1.0
Broccoli, cooked 1/2 cup (78 g) 2.6 27 0.3
Potato baked, without skin 1 medium (145 g) 2.3 145 1.0
Baby spinach leaves 3 ounces (90 g) 2.1 20 0.2
Blueberries 1/2 cup (74 g) 1.8 42 0.6
Carrot, raw or cooked 1 medium (60 g) 1.7 25 0.4
Dahl WJ, Stewart ML. Position of the Academy of Nutrition and Dietetics: health implications of dietary fiber. J Acad
Nutr Diet. 2015;115:1861–1870.
trends. 2015;97, 98; Table D1.8.
Slavin, J.L. Position of the American Dietetic Association: Health implications of dietary fiber. J. Am. Diet. Assoc.
2008;108:1716–1731.
gov/ Accessed 8.22.2015.
9. Sonnenburg ED, Sonnenburg JL. Starving our

References microbial self: the deleterious consequences of a diet
deficient in microbiota-accessible carbohydrates.
1. Arora T, Backhed F. The gut microbiota and meta- Cell Metab. 2014;20:779–86.
bolic disease: current understanding and future 10. Sonnenburg ED, Smits SA, Tikhonov M, et al. Diet-
perspectives. J Intern Med. 2016;280(4):339–49. induced extinctions in the gut microbiota compound
https://doi.org/10.1111/joim.12508. over generations. Nature. 2016;529(7585):212–5.
2. Keenan MJ, Marco ML, Ingram DK, Martin 11. Zhang C, Zhang M, Wang S, et al. Interactions
RJ. Improving healthspan via changes in gut micro- between gut microbiota, host genetics and diet rel-
biota and fermentation. Age (Dordr). 2015;37(5):98. evant to development of metabolic syndromes in
https://doi.org/10.1007/s11357-015-9817-6. mice. ISME J. 2010;4:232–41.
3. El Enshasy H, Malik K, Malek RA, et al. Anaerobic 12. Ley RE, Turnbaugh P, Klein S, Gordon JI. Microbial
probiotics: the key microbes for human health. ecology: human gut microbes associated with obe-
Adv Biochem Eng Biotechnol. 2016;156:397–431. sity. Nature. 2006;444:1022–3.
https://doi.org/10.1007/10-2015-5008. 13. Woodmansey EJ. Intestinal bacteria and ageing. J
4. Oozeer R, van Limpt K, Ludwig T, et al. Intestinal Appl Microbiol. 2007;102:1178–86.
microbiology in early life: specific prebiotics can 14. Tuohy KM, Fava F, Viola R. The way to a man’s
have similar functionalities as human-milk oligosac- heart is through his gut microbiota’—dietary pro-
charides. Am J Clin Nutr. 2013;98(suppl):561S–71S. and prebiotics for the management of cardiovascular
5. Nicholson JK, Holmes E, Kinross J, et al. Host- risk. Proc Nutr Soc. 2014;73:172–85.
gut microbiota metabolic interactions. Science. 15. Albenberg LG, Wu GD. Diet and the intestinal
2012;336:1262–7. microbiome: associations, functions, and impli-
6. Koh A, De Vadder F, Kovatcheva-Datchary P, cations for health and disease. Gastroenterology.
Backhed F. From dietary fiber to host physiology: 2014;146(6):1564–72.
short-chain fatty acids as key bacterial metabolites. 16. Logan AC, Jacka FN, Prescott SL. Immune–micro-
Cell. 2016;165:1332–45. biota interactions: dysbiosis as a global health issue.
7. Jia W, Li H, Zhao L, Nicholson JK. Gut microbiota: Curr Allergy Asthma Rep. 2016;16:13. https://doi.
a potential new territory for drug targeting. Nat Rev org/10.1007/s11882-015-0590-5.
Drug Disord. 2008;7(2):123–9. 17. Zeng H, Lazarova DL, Bordonaro M. Mechanisms
8. Mazidi M, Rezaie P, Kengne AP, et al. Gut micro- linking dietary fiber, gut microbiota and colon
biome and metabolic syndrome. Diabetes Metab cancer prevention. World J Gastrointest Oncol.
Syndr. 2016;10(2 Suppl 1):S150–7. https://doi. 2014;6(2):41–51.
org/10.1016/j.dsx.2016.01.024.
References 141
18. Hamer HM, Jonkers D, Venema K, et al. Review in tumor tissue. JAMA Oncol. 2017. https://doi.
article: the role of butyrate on colonic function. org/10.1001/jamaoncol.2016.6374.
Aliment Pharmacol Ther. 2008;27:104–19. 34. Racine A, Carbonnel F, Chan SS, et al. Dietary pat-
19. Meijer K, de Vos P, Priebe MG. Butyrate and other terns and risk of inflammatory bowel disease in
short-chain fatty acids as modulators of immunity: Europe: results from the EPIC study. Inflamm Bowel
What relevance for health? Curr Opin Clin Nutr Dis. 2016;2292:345–54. https://doi.org/10.1097/
Metab Care. 2010;13(6):715–21. MIB.0000000000000638.
20. Titgemeyer EC, Bourquin LD, Fahey GC, Garleb 35. Gutiérrez-Díaz I, Fernández-Navarro T, Sánchez B,
KA. Fermentability of various fiber sources by et al. Mediterranean diet and faecal microbiota: a
human fecal bacteria in vitro. Am J Clin Nutr. transversal study. Food Funct. 2016;;7(5):2347-56.
1991;53:1418–24. https://doi.org/10.1039/c6fo00105j.
21. Shin NR, Whon TW, Bae JW. Proteobacteria: micro- 36. GD W, Compher C, Chen EZ, et al. Comparative
bial signature of dysbiosis in gut microbiota. Trends metabolomics in vegans and omnivores reveal con-
Biotechnol. 2015;33(9):496–503. straints on diet-dependent gut microbiota metabolite
22. Marchesi JR, Adams DH, Fava F, et al. The gut production. Gut. 2016;65(1):63–72.
microbiota and host health: a new clinical frontier. 37. De Filippis F, Pellegrini N, Vannini L, et al. High-
Gut. 2015;65(2):330–9. level adherence to a Mediterranean diet beneficially
23. Deehan C, Walter J. The fiber gap and the disappear- impacts the gut microbiota and associated metabo-
ing gut microbiome: implications for human nutri- lome. Gut. 2015;65(11):1812–21. https://doi.
tion. Trends Endocrinol Metab. 2016;27(5):239–41. org/10.1136/gutjnl-2015-309957.
24. Blaser MJ, Falkow S. What are the consequences 38. Matijasic BB, Obermajer T, Lipoglavsek L, et al.
of the disappearing human microbiota? Nat Rev Association of dietary type with fecal microbiota in
Microbiol. 2009;7:887–94. vegetarians and omnivores in Slovenia. Eur J Nutr.
25. Jew S, Abumweis SS, Jones PJ. Evolution of the 2014;53(4):1051–64.
human diet: linking our ancestral diet to modern 39. Ou J, Carbonero F, Zoetendal EG, et al. Diet, micro-
functional foods as a means of chronic disease pre- biota, and microbial metabolites in colon cancer risk
vention. J Med Food. 2009;12(5):925–34. in rural Africans and African Americans. Am J Clin
26. Dominianni C, Sinha R, Goedert JJ, et al. Sex, Nutr. 2013;98:111–20.
body mass index, and dietary fiber intake influ- 40. Zimmer J, Lange B, Frick J-S, et al. A vegan or vege-
ence the human gut microbiome. PLoS One. tarian diet substantially alters the human colonic fae-
2015;10(4):e0124599. https://doi.org/10.1371/jour- cal microbiota. Eur J Clin Nutr. 2012;66(1):53–60.
nal.pone.0124599. 41. Kabeerdoss J, Devi RS, Mary RR, et al. Faecal
27. IOM (Institute of Medicine). Dietary refer- microbiota composition in vegetarians: compari-
ence intakes for energy, carbohydrate, fiber, fat, son with omnivores in a cohort of young women in
fatty acids, cholesterol, protein, and amino acids. southern India. Br J Nutr. 2012;108:9544.
2002/2005. National Academies Press, Washington 42. GD W, Chen J, Hoffmann C, et al. Linking long-
DC. term dietary patterns with gut microbial entero-
28. Milani C, Ferrario C, Turron F, et al. The human gut types. Science. 2011;334(6052):105–8. https://doi.
microbiota and its interactive connections to diet. org/10.1126/science.1208344.3-957.
J Hum Nutr Diet. 2016;29(5):539–46. https://doi. 43. Tap J, Furet JP, Bensaada M, et al. Gut microbiota
org/10.1111/jhn.12371. richness promotes its stability upon increased dietary
29. Parekh PJ, Balart LA, Johnson DA. The influence fibre intake in healthy adults. Environ Microbiol.
of the gut microbiome on obesity, metabolic syn- 2015;17(12):4954–64.
drome and gastrointestinal disease. Clin Transl 44. O’Keefe JD, Li JV, Lahti L, et al. Fat, fiber and can-
Gastroenterol. 2015;6:e91. https://doi.org/10.1038/ cer risk in African Americans and rural Africans.
ctg.2015.16. Nat Commun. 2015;6:6342. https://doi.org/10.1038/
30. Furusawa Y, Obata Y, Fukuda S, et al. Commensal ncomms7342.
microbe derived butyrate induces the differ- 45. David LA, Maurice CF, Carmody RN, et al. Diet
entiation of colonic regulatory T cells. Nature. rapidly and reproducibly alters the human gut micro-
2013;504:446–50. biome. Nature. 2014;505:559–63.
31. Conlon MA, Bird AR. The impact of diet and life- 46. Klinder A, Shen Q, Heppel S, et al. Impact of increas-
style on gut microbiota and human health. Forum ing fruit and vegetables and flavonoid intake on the
Nutr. 2015;7:17–44. human gut microbiota. Food Funct. 2016;7:1788–96.
32. Cummings JH, Englyst HN. Fermentation in the 47. Heinritz SN, Weiss E, Eklund M, et al. Intestinal
human large intestine and the available substrates. microbiota and microbial metabolites are changed
Am J Clin Nutr. 1987;45(5 suppl):1243–55. in a pig model fed a high-fat/low-fiber or a low-fat/
33. Mehta RS, Nishihara R, Cao Y, et al. Association high-fiber diet. PLoS One. 2016;11(4):e0154329.
of dietary patterns with risk of colorectal cancer https://doi.org/10.1371/journal.pone.0154329.
subtypes classified by Fusobacterium nucleatum
48. Wang Y, Ames NP, Tun HM, et al. High molecu- 62. Bultman SJ. Interplay between diet, gut micro-
lar weight barley β-glucan alters gut microbiota biota, epigenetic events, and colorectal cancer.
toward reduced cardiovascular disease risk. Front Mol Nutr Food Res. 2017; 61(1); doi: 10.1002/
Microbiol. 2016;7:129. https://doi.org/10.3389/ mnfr.201500902.
fmicb.2016.00129. 63. Aune D, Chan DSM, Lau R, et al. Dietary fibre,
49. Martinez I, Lattimer JM, Hubach KL, et al. Gut whole grains, and risk of colorectal cancer: system-
microbiome composition is linked to whole grain- atic review and dose-response meta-analysis of pro-
induced immunological improvements. The. ISME spective studies. BMJ. 2011;343:d6617. https://doi.
J. 2013;7:269–80. org/10.1136/bmj.d6617.
50. Carvalho-Wells AL, Helmolz K, Nodet C, et al. 64. World Cancer Research Fund, American Institute
Determination of the in vivo prebiotic potential of a of Cancer Research. Continuous Update Project.
maize-based whole grain breakfast cereal: a human Colorectal Cancer 2011 Report. Food, Nutrition,
feeding study. Br J Nutr. 2010;104:1353–6. Physical Activity, and the Prevention of Colorectal
51. Costabile A, Klinder A, Fava F, et al. Whole- Cancer. London; 2011.
grain wheat breakfast cereal has a prebiotic effect 65. Kunzmann AT, Coleman HG, Huang W-Y, et al.
on the human gut microbiota: a double-blind, Dietary fiber intake and risk of colorectal cancer
placebo-controlled, crossover study. Br J Nutr. and incident and recurrent adenoma in the Prostate,
2008;99:110–20. Lung, Colorectal, and Ovarian Cancer Screening
52. Reveles KR, Lee GC, Boyd NK, Frei CR. The rise in Trial. Am J Clin Nutr. 2015;102:881–90.
Clostridium difficile infection incidence among hos- 66. Chen HM, Y-N Y, Wang J-L, et al. Decreased dietary
pitalized adults in the United States: 2001-2010. Am fiber intake and structural alteration of gut micro-
J Infect Control. 2014;42:1028–32. biota in patients with advanced colorectal adenoma.
53. Van den Abbeele P, Verstraete W, Aidy SE, et al. Am J Clin Nutr. 2013;97:1044–52.
Prebiotics, faecal transplants and microbial network 67. Miller WC, Niederpruem MG, Wallace JP, Lindeman
units to stimulate biodiversity of the human gut AK. Dietary fat, sugar, and fiber predict body fat
microbiome. J Microbial Biotechnol. 2013;6(4):335– content. J Am Diet Assoc. 1994;94:612–5.
40. https://doi.org/10.1111/1751-7915.12049. 68. Epstein LH, Gordy CC, Raynor HA, et al. Increasing
54. Zhang I, Dong D, Jiang C, et al. Insight into altera- fruit and vegetable intake and decreasing fat and
tion of gut microbiota in Clostridium difficile infec- sugar intake in families at risk for childhood obesity.
tion and asymptomatic C. difficile colonization. Obes Res. 2001;9:171–8.
Anaerobe. 2015;34:1–7. https://doi.org/10.1016/j. 69. Epstein LH, Paluch RA, Beecher MD, Roemmich
anaerobe.2015.03.008. JN. Increasing healthy eating vs. reducing high
55. May T, Mackie RI, Fahey GC, et al. Effect energy-dense foods to treat pediatric obesity. Obesity
of fiber source on short-chain fatty acid produc- (Silver Spring). 2008;16(2):318–26.
tion and on the growth and toxin production by 70. Davis JN, Hodges VA, Gillham B. Normal-weight
Clostridium difficile. Scand J Gastroenterol. adults consume more fiber and fruit than their age-
1994;29(10):916–22. and height-matched overweight/obese counterparts.
56. Forssten SD, Roytio H, Ashley A, et al. The effect J Am Diet Assoc. 2006;106:833–40.
of polydextrose and probiotic lactobacilli in a 71. Davis JN, Alexander KE, Ventura EE, et al. Inverse
Clostridium difficile infected human colonic model. relation between dietary fiber intake and visceral
Microb Ecol Health Dis. 2015;26:27988. https://doi. adiposity in overweight Latino youth. Am J Clin
org/10.3402/mehd.v26.27988. Nutr. 2009;90:1160–6.
57. Johnson LP, Walton GE, Psichas A, et al. Prebiotics 72. Center for Disease Control and Prevention (CDC).
modulate the effects of antibiotics on gut micro- Eat more weigh less? How to manage your weight
bial diversity and functioning in vitro. Forum Nutr. without being hungry. http://www.cdc.gov/nccdphp/
2015;7:4480–97. dnpa/nutrition/pdf/Energy_Density.pdf. Accessed
58. Aleksandrova K, Romero-Mosquera B, Hernandez V. May 21, 2016.
Diet, gut microbiome and epigenetics: emerging links 73. Savage JS, Marini M, Birch LL. Dietary energy den-
with inflammatory bowel diseases and prospects for sity predicts women’s weight change over 6 years.
management and prevention, Nutrients. 2017;9:962. Am J Clin Nutr. 2008;88(3):677–84.
59. Liu X, Wu Y, Li F, Zhang D. Dietary fiber intake 74. Bertoia ML, Mukamal KJ, Cahill LE, et al. Changes
reduces risk of inflammatory bowel disease: result in intake of fruits and vegetables and weight change
from a meta-analysis. Nutr Res. 2016;2292:345–54; in United States men and women followed for up
doi: 20.1097/MIB00000000000638. to 24 years: analysis from three prospective cohort
60. Burkitt DP. Possible relationships between bowel studies. PLoS Med. 2015;12(9):e1001878. https://
cancer and dietary habits. Proc R Soc Med. doi.org/10.1371/journalpmed.1001878.
1971;64:964–5. 75. Barczynska B, Bandurska K, Slizewska K, et al.
61. Bultman SJ. The microbiome and its potential as Intestinal microbiota, obesity and prebiotics. Polish.
a cancer preventive intervention. Semin Oncol. J Microbiol. 2015;64(2):93–100.
2016;43(1):97–106. https://doi.org/10.1053/j. 76. Brahe LK, Astrup A, Larsen LH. Can we prevent
seminoncol.2015.09.001. obesity-related metabolic diseases by dietary
References 143
modulation of the gut microbiota? Adv Nutr. 2016;7:90– 92. Chambers ES, Morrison DJ, Frost G. Control
101. https://doi.org/10.3945/an.115.010587. of appetite and energy intake by SCFA: what
77. Turnbaugh PJ, Hamady M, Yatsunenko T, et al. are the potential underlying mechanisms? Proc
A core gut microbiome in obese and lean twins. Nutr Soc. 2015;74:328–336; doi: 10.1017/
Nature. 2009;457:480–4. S0029665114001657.
78. Cotillard A, Kennedy SP, Kong LC, et al. Dietary 93. Ley SH, Hamdy O, Mahan V, Prevention HFB.
intervention impact on gut microbial gene richness. management of type 2 diabetes: dietary com-
Nature. 2013;500(7464):585–8. ponents and nutritional strategies. Lancet.
79. Le Chatelier E, Nielsen T, Qin J, et al. Richness of 2014;383:1999–2007.
human gut microbiome correlates with metabolic 94. Tabák AG, Herder C, Rathmann W, et al. Prediabetes:
markers. Nature. 2013;500(7464):541–6. A high-risk state for developing diabetes. Lancet.
80. Lozupone CA, Stombaugh JI, Gordon JI, et al. 2012;379(9833):2279–90.
Diversity, stability and resilience of the human gut 95. Everard A, Cani PD. Diabetes, obesity and gut
microbiota. Nature. 2012;489:220–30. microbiota. Best Pract Res Clin Gastroenterol.
81. Miquel S, Martín R, Rossi O, et al. Faecalibacterium 2013;27:73–83.
prausnitzii and human intestinal health. Curr Opin 96. Murri M, Leiva I, Gomez-Zumaquero JM, et al. Gut
Microbiol. 2013;16:255–61. microbiota in children with type 1 diabetes differs
82. Geurts L, Neyrinck AM, Delzenne NM, et al. Gut from that in healthy children: a case–control study.
microbiota controls adipose tissue expansion, gut BMC Med. 2013;11:46.
barrier and glucose metabolism: novel insights into 97. Serino M, Fernandez-Real JM, Garcıa Fuentes
molecular targets and interventions using prebiotics. E, et al. The gut microbiota profile is associated
Benefic Microbes. 2014;5(1):3–17. with insulin action in humans. Acta Diabetol.
83. Karlsson FH, Tremaroli V, Nookaew I, et al. Gut 2013;50:753–61.
metagenome in European women with normal, 98. Larsen N, Vogensen FK, van den Berg FW, et al.
impaired and diabetic glucose control. Nature. Gut microbiota in human adults with type 2 dia-
2013;498:99–103. betes differs from non-diabetic adults. PLoS One.
84. Furet JP, Kong LC, Tap J, et al. Differential 2010;5:e9085. https://doi.org/10.1371/journal.
adaptation of human gut microbiota to bariatric pone.0009085.
surgery-induced weight loss: links with metabolic 99. Qin J, Li Y, Cai Z, et al. A metagenome-wide asso-
and low-grade inflammation markers. Diabetes. ciation study of gut microbiota in type 2 diabetes.
2010;59:3049–57. Nature. 2012;490:55–60.
85. Sánchez D, Miguel M, Aleixandre A. Dietary 100. Nagpal R, Kumar M, Yadav AK, et al. Gut micro-
fiber, gut peptides, and adipocytokines. J Med biota in health and dsiease: an overview focused
Food. 2012;15(3):223–30. https://doi.org/10.1089/ on metabolic inflammation. Benef Microbes. 2016;
jmf.2011.0072. 7(2):181–94.
86. Hjorth MF, Roager HM, Larsen TM, et al. Pre- 101. Kim MS, Hwang SS, Park EJ, Bae JW. Strict veg-
tretment microbial Prevotella-to-Bacteroides ratio, etarian diet improves the risk factors associated
determeines body fat loss success during 6-month with metabolic diseases by modulating gut micro-
randomized controlled diet intervention. Int J Obes. biota and reducing intestinal inflammation. Environ
2017; doi:10.1038/ijo.2017.220. Microbiol Rep. 2013;5:765–75.
87. Million M, Maraninchi M, Henry M, et al. Obesity- 102. Fallucca F, Fontana L, Fallucca S, Pianesi M. Gut
associated gut microbiota is enriched in Lactobacillus microbiota and Ma-Pi 2 macrobiotic diet in the
reuteri and depleted in Bifidobacterium animalis treatment of type 2 diabetes. World J Diabetes.
and Methanobrevibacter smithii. Int J Obes (Lond). 2015;6(3):403–11.
2012;36:817–25. 103. Karimi P, Farhangi MA, Sarmadi B, et al. The thera-
88. Kalliomäki M, Collado MC, Salminen S, Isolauri peutic potential of resistant starch in modulation of
E. Early differences in fecal microbiota composition insulin resistance, endotoxemia, oxidative stress and
in children may predict overweight. Am J Clin Nutr. antioxidant biomarkers in women with type 2 diabe-
2008;87:534–8. tes: a randomized controlled clinical trial. Ann Nutr
89. Moreno-Indias I, Cardona F, Tinahones FJ, Queipo- Metab. 2016;68(2):85–93.
Ortuño MI. Impact of the gut microbiota on the 104. Bodinham CL, Smith L, Thomas EL, et al. Efficacy
development of obesity and type 2 diabetes mellitus. of increased resistant starch consumption in human
Front Microbiol. 2014;5(190):1–10. type 2 diabetes. Endocr Connect. 2014;3:75–84.
90. Fernandes J, Su W, Rahat-Rozenbloom S, et al. 105. Grundy SM. Metabolic syndrome pandemic.
Adiposity, gut microbiota and faecal short chain Arterioscler Thromb Vasc Biol. 2008;28:629–36.
fatty acids are linked in adult humans. Nutr 106. Festi D, Schiumerini R, Eusebi LH, et al. Gut
Diabetes. 2014;4:e121. https://doi.org/10.1038/ microbiota and metabolic syndrome. World J
nutd.2014.23. Gastroenterol. 2014;20(43):16079–94.
91. Blaut M. Gut microbiota and energy balance: role 107. Martinez-Gonzalez MA, Martin-Calvo N. The major
in obesity. Proc Nutr Soc. 2015;74:227–34; doi: European dietary pattern and metabolic syndrome.
10.1017/S0029665114001700. Rev Endocr Metab Disord. 2013;14(3):265–71.
108. Vetrani C, Costabile G, Luongo D, et al. Effects of mended intakes of dietary fiber in Canada. Front
whole-grain cereal foods on plasma short chain fatty Pharmacol. 2015;6:167. https://doi.org/10.3389/
acid concentrations in individuals with the metabolic fphar.2015.00167.
syndrome. Nutrition. 2016;32:217–21. 124. Liu L, Wang S, Liu J. Fiber consumption and all-
109. Chambers ES, Viardot A, Psichas A, et al. Effects cause, cardiovascular, and cancer mortalities: A sys-
of targeted delivery of propionate to the human tematic review and meta-analysis of cohort studies.
colon on appetite regulation, body weight main- Mol Nutr Food Res. 2015;59:139–46.
tenance and adiposity in overweight adults. Gut. 125. Park Y, Subar AF, Hollenbeck A, et al. Dietary
2015;64(11):1744–54. fiber intake and mortality in the NIH-AARP
110. Yoon NR, Yoon S, Lee S-M. Rice cakes containing Diet and Health Study. Arch Intern Med.
dietary fiber supplemented with or without Artemisia 2011;171(12):1061–8.
annua and Gynura procumbens Merr. alleviated the 126. Chuang S-C, Norat T, Murphy N, et al. Fiber
risk factors of metabolic syndrome. Clin Nutr Res. intake and total and cause-specific mortality in the
2016;5:79–88. European Prospective Investigation into Cancer and
111. Brahe LK, Le Chatelier E, Prifti E. Specific Nutrition cohort. Am J Clin Nutr. 2012;96:164–74.
gut microbiota features and metabolic mark- 127. Wei H, Gao Z, Liang R, et al. Whole-grain con-
ers in postmenopausal women with obesity. Nutr sumption and the risk of all-cause, CVD and can-
Diabetes. 2015; e159. doi: https://doi.org/10.1038/ cer mortality: a meta-analysis of prospective cohort
nutd.2015.9. studies. Br J Nutr. 2016;116(3):514–25. https://doi.
112. Galisteo M, Duarte J, Zarzuelo A. Effects of dietary org/10.1017/S0007114516001975.
fibers on disturbances clustered in the metabolic 128. Nguyen B, Bauman A, Gale J, et al. Fruit and vege-
syndrome. J Nutr Biol. 2008;19:71–84. table consumption and all-cause mortality: evidence
113. Rehman T. Role of the gut microbiota in age-related from a large Australian cohort study. Int J Behav
chronic inflammation. Endocr Metab Immune Nutr Phys Act. 2016;13:9. https://doi.org/10.1186/
Disord Drug Targets. 2012;12:361–7. s12966-016-0334-5.
114. Brüssow H. Microbiota and healthy ageing: observa- 129. Claesson MJ, Jeffery IB, Conde S. Gut microbiota
tional and nutritional intervention studies. Microbial composition correlates with diet and health in the
Biotechnol. 2013;6:326–34. elderly. Nature. 2012;488:178–85.
115. O’Toole PWO, Jeffery IB. Gut microbiota and aging. 130. Claesson MJ, Cusack S, O’Sullivan O, et al.
Science. 2015;350(6265):1214–5. Composition, variability, and temporal stability of
116. Zapata HJ, Quagliarello VJ. The microbiota and the intestinal microbiota of the elderly. Proc Natl
microbiome in aging: potential implications in Acad Sci U S A. 2011;108(suppl. 1):4586–91.
health and age-related diseases. J Am Geriatr Soc. 131. Jeffery IB, Lynch DB, O’Toole PW. Composition
2015;63(4):776–81. and temporal stability of the gut microbiota in older
117. Rowe JW, Kahn RL. Human aging: usual and suc- persons. ISME J. 2016;10:170–82.
cessful. Science. 1987;237:143–9. https://doi. 132. Biagi E, Franceschi C. Rampelli S, et al. Curr Bio:
org/10.1126/science.3299702. Gut microbiota and extreme longevity; 2016. https://
118. Stenman LK, Burcelin R, Lahtinen S. Establishing doi.org/10.1016/j.cub.2016.04.016
a causal link between gut microbes, body weight 133. Wang F, Yu T, Huang G, Cai D. Gut microbiota com-
gain and glucose metabolism in humans -towards munity and its assembly associated with age and diet
treatment with probiotics. Benefic Microbes. in Chinese centenarians. J Microbiol Biotechnol.
2015;7(1):11–22. 2015;25(8):1195–204.
119. Cuervo A, Salazar N, Ruas-Madiedob P, et al. Fiber 134. Biagi E, Nylund L, Candela M, et al. Through age-
from a regular diet is directly associated with fecal ing, and beyond: gut microbiota and inflamma-
short-chain fatty acid concentrations in the elderly. tory status in seniors and centenarians. PLoS One.
Nutr Res. 2013;33:811–6. 2010;5(5):e10667. https://doi.org/10.1371/journal.
120. Jiao J, J-Y X, Zhang W, et al. Effect of dietary fiber pone.0010667.
on circulating C-reactive protein in overweight and 135. Mariat D, Firmesse O, Levenez F, et al. The
obese adults: a meta-analysis of randomized con- Firmicutes/Bacteroidetes ratio of the human micro-
trolled trials. Int J Food Sci Nutr. 2015;66(1):114–9. biota changes with age. BMC Microbiol. 2009;9:123.
121. Grooms KN, Ommerborn MJ, Quyen D, et al. https://doi.org/10.1186/1471-2180-9-123.
Dietary fiber intake and cardiometabolic risk among 136. van Tongeren SP, Slaets JP, Harmsen HJ, Welling
US adults, NHANES 1999-2010. Am J Med. GW. Fecal microbiota composition and frailty. Appl
2013;126(12):1059–67. Environ Microbiol. 2005;71:6438–42.
122. Cassidy A, De Vivo I, Liu Y, et al. Associations 137. Bartosch S, Fite A, Macfarlane GT, McMurdo
between diet, lifestyle factors, and telomere length ME. Characterization of bacterial communi-
in women. Am J Clin Nutr. 2010;91:1273–83. ties in feces from healthy elderly volunteers
123. Abdullah MM, Gyles CL, Marinangeli CP, and hospitalized elderly patients by using real-
et al. Cost-of-illness analysis reveals potential time PCR and effects of antibiotic treatment on
healthcare savings with reductions in type 2 dia- the fecal microbiota. Appl Environ Microbiol.
betes and cardiovascular disease following recom- 2004;70:3575–81.
Fiber-Rich Dietary Patterns
and Foods in Laxation 5
and Constipation
Keywords
Fiber-rich foods • Laxation • Bowel movement • Functional constipation
• Chronic constipation • Colon health • Wheat bran • Oat bran • Prunes
• Kiwi fruit • Polydextrose • Psyllium • Chicory inulin • Prebiotics
• Symbiotics
Key Points the most widely studied fiber; when baseline

• The consumption of healthy dietary patterns transit time was >48 h, each extra g/day of
with adequate dietary fiber (>25 g/day), rec- wheat bran significantly increased total stool
ommended fluid intake, and regular physical weight by 3.7 g and reduced transit time by
activity, are especially beneficial in preventing 45 min.
and alleviating constipation. • Increased fiber intake did not change transit
• Fiber mechanisms associated with improved time in individuals with an initial time of
laxation and alleviated constipation include: <48 h. However, in people with an initial
increasing stool weight and bulk volume transit time ≥48 h, transit time was reduced
(through fiber and microbiota physical vol- by approximately 30 min per gram of cereal,
ume and water holding capacity), and gas vol- fruit or vegetable fiber, regardless of
ume trapped in the stool to increase bowel fermentability.
movement frequency and quality, especially in • Several RCTs suggest that daily intake of
constipated individuals. prunes (dried plums) and/or kiwi fruit can
• Adequate intake of fiber from cereal, fruits, help in relieving constipation symptoms simi-
vegetables and common fiber-rich food ingre- larly to psyllium.
dients including polydextrose, psyllium, chic-
ory inulin and prebiotics or symbiotics have
the potential to increase population-wide levels 5.1 Introduction
of regularity and provide constipation relief.
• In general, less fermentable dietary fiber tends Constipation symptoms vary from person to per-
to increase fecal weight to a greater amount son but are usually described as infrequent bowel
than more fermentable fibers. Wheat bran is movements, straining, passage of hard stools,

146 5 Fiber-Rich Dietary Patterns and Foods in Laxation and Constipation
and/or difficulty in passing stools, which effects 5.2 verview of Fiber, Laxation
O
most people at some time in their lives [1–3]. In and Constipation
Western countries, about 30% of the general
adult population experiences problems with con- There are a number of health professional orga-
stipation during their life time, with elderly peo- nizations recommending the intake of adequate
ple and women being most affected. However, dietary fiber (>25g/day or 14 g/1,000 kcal) to
many people experience constipation that is prevent constipation [9, 10, 13–16]. The
more than a minor annoyance, and can be American Medical Association recommends the
chronic, sometimes severe, and have significant consumption of adequate fiber intake of at least
and often debilitating effects on their quality of 25 g/day as a first step that may improve or
life. Physicians frequently define constipation as eliminate constipation by increasing the intake
fewer than three bowel movements/week and of fiber containing fruit, vegetables, whole-
most people have bowel movement frequency grains (e.g., wheat bran breakfast cereals),
between 3 per day and 3 per week [4]. Chronic legumes, and nuts and seeds or appropriate fiber
constipation is more commonly diagnosed in supplement products such as psyllium [9, 10].
female patients at 2–3 times the rate of males. The American College of Gastroenterology sug-
Individuals with constipation rarely report it to gests that fiber supplements, especially from
anyone or seek medical treatments [5, 6]. By soluble fibers such as psyllium, may be effective
80 years of age, the prevalence of chronic consti- in the management of chronic constipation in
pation is about 34% in women and 26% in men adults [13]. The Academy of Nutrition and
[1]. The prevalence of childhood constipation is Dietetics recommends the adequate amounts of
challenging to estimate but worldwide about fiber from a variety of plant foods to support
16% of mothers of toddlers report some degree laxation by increasing fecal biomass, increasing
of constipation in their children [7, 8]. stool frequency, and reducing intestinal transit
Constipation can be acute lasting for a day or time [14]. The European Food Safety Authority
two, or chronic lasting for weeks, months, or (EFSA) panel recommended that the consump-
years, as a result of a variety of potential factors tion of at least 25 g fiber/day from food was
including low dietary fiber (fiber) diets, inade- adequate for normal laxation in adults [15]. The
quate fluid intake, inactivity or certain medicines American Academy of Pediatrics recommends a
such as opiates used to control pain [1–3, 9, 10]. fiber intake of 0.5 g/kg/day for all children, or
Diet and lifestyle modifications, including ade- alternatively for children older than age 2 years,
quate dietary fiber (>25 g/day; especially from fiber amount equivalent to their age in years
whole plant foods) plus water intake and regular plus 5 g/day (to a maximum of 35 g/day for
physical activity may especially represent an older children or adolescents) as low fiber intake
effective, inexpensive, and feasible therapeutic may be a risk factor for chronic constipation
way to prevent and alleviate constipation [11, [16]. These organizations generally recommend
12]. Fiber increases colonic stool volume and introducing dietary fiber gradually to minimize
water content directly, or stimulates motility bloating, distension, flatulence, and cramping,
leading to shortened colonic transit and decreased which could limit compliance and effectiveness
water absorption [1–3]. Also, fiber supplements, [9, 10, 13].
or other types of laxatives plus fluids are fre- Adequate intake of healthy fiber-rich dietary
quently considered to effectively manage consti- patterns, especially whole foods, minimally pro-
pation. The objective of this chapter is to cessed foods, or foods enriched with fiber-rich
comprehensively review the role of fiber-rich ingredients, such as with wheat bran or psyl-
diets and foods in promoting laxation and allevi- lium, along with adequate fluid intake and phys-
ating constipation. ical activity are the primary lifestyle factors for
5.2 Overview of Fiber, Laxation and Constipation 147
regularity in laxation [11, 12, 17–21]. Low fiber increasing stool frequency and decreasing laxa-
intake is often associated with constipation in tive use than consuming only 1 L of fluid/day
epidemiologic studies [11, 12, 21]. The Nurses’ (Fig. 5.2) [19]. For physical activity and fiber,
Health Study (62,036 women; age range 36 to the Nurses’ Health Study (62,036 women; aged
61 years; 3327 reported bowel movement fre- 36–61 years; 5.4% were classified as consti-
quency every 3rd day or less) observed that pated defined as ≤2 bowel movements weekly;
women with a median intake of about 20 g fiber median daily fiber intake 20 g vs. 7 g) found that
daily had a 36% lower prevalence of constipa- women in the highest quintile of physical activ-
tion compared with women who consumed ity (2–3 times/week) and fiber intake had a mul-
about 7 g of fiber daily (Fig. 5.1) [11]. For physi- tivariate lower prevalence of constipation by
cal activity and fiber, this study found that 68% compared with those in the lowest quintile
women in the highest quintile of physical activ- of physical activity (<1 time weekly) and fiber
ity (2–3 times/week) and fiber intake had a mul- intake [11]. In these women, higher physical
tivariate lower prevalence of constipation by activity reduced constipation risk by 44% and
68% compared with those in the lowest quintile higher fiber intake reduced risk by 36%. A 2015
of physical activity (<1 time per week) and fiber study of Canadian adults found that each 1 g/day
intake; higher physical activity reduced consti- increase in dietary fiber from foods was pro-
pation risk by 44% and higher fiber intake jected to reduce constipation rates by about 2%
reduced risk by 36%. Adequate water intake is [22]. Similarly, in the US, it was estimated that
important for fiber laxation mechanisms to work if fiber intake was increased by 9 g/day from
optimally and low water intake may increase the bran (equivalent to one serving of high fiber
prevalence of constipation [17, 18]. An inter- breakfast cereal/day), there could be a billion
vention trial with chronically constipated indi- dollars in annual savings in medical costs due to
viduals (117 subjects; mean age 39 years; decreased constipation [23]. A meta-analysis (5
64% women; 10 g fiber/1000 kcals vs. 7 g randomized controlled trials [RCTs]) found that
fiber/1000 kcal; 2 months) showed that just increased fiber intake to the adequate intake
increasing fiber intake by 3 g/1000 kcal plus 2 L range significantly improved stool frequency by
of liquids/day was significantly more effective at 19% (p < .05) compared to the placebo [24].
1.1
1
Constipation Risk in Women
0.9
0.8
0.7
Fig. 5.1 Association
0.6
between level of fiber
intake and risk of
constipation in women 0.5
(p < .0001; multivariate 7.1 10.2 12.5 18.1 20.2
adjusted) (adapted
from [11]) Median Total Fiber Intake (g/day)
Fig. 5.2 Effect of level 1.1 liters water/day 2.1 liters water/day
of water intake with a
3
standard diet providing
25 g fiber/day on stool
frequency and laxative 2
usage in 117 patients
Change per Week

with chronic functional 1
constipation (p < .001)
(adapted from [19]) 0
Stool frequency (#) Laxative use (doses)
-1
-2
-3
However, there was no significant difference in fermentation capacity. There is a fecal bulking
stool consistency and painful defecation between index for standardized measurement of the rela-
the two groups. tive colonic bulking efficacy of foods relative to a
typical edible serving size of wheat bran [26–28].
Table 5.1 provides an estimation of the fecal bulk-
5.3 iber Related Laxation
F ing capacity of whole and processed plant foods
Mechanisms relative to wheat bran from a validated model
system [28]. The fecal bulking index values for
The way in which fiber affects bowel habit can- whole-grains breads or wheat bran enriched
not be explained on the basis of one simple breakfast cereals range from about 12 to 50.
hypothesis [25]. There are several ways by which Wheat bran and other fibers are fairly resistant to
fiber increases laxation. First, plant cell walls fermentation in the large bowel so the retained
especially rich in lignin with moderate water 3-dimensional fiber structure has a water binding
holding capacity and low to moderate ferment- capacity of 5 to 6 g water/g fiber in the distal
ability such as wheat bran; or soluble fibers with colon. More fermentable fibers provide some
relatively low fermentability, such as psyllium, bulk mainly due to increased bacterial mass and
which have a high water holding capacity leading trapped gas [27]. Some soluble fiber, fermenta-
to increased colonic volume, are among the most tion-resistant ingredients or supplements, can
effective options for increasing fecal bulk and have polysaccharides with high water binding or
stimulating colonic laxation. Second, prebiotics, gelling with fecal bulking capacities in excess of
from especially highly fermentable soluble fibers 100x, with psyllium having a value of 500x rela-
such as inulin, can stimulate increased microbi- tive to wheat bran. Generally, the greater the wet
ota numbers and add volume to the colonic fecal bulk weight of the stool from fiber and the food
mass. Third, fiber fermentation to hydrogen, matrix and the more rapid the rate of passage
methane and carbon dioxide gases, which can be through the colon and the better the laxative
trapped within colonic contents, add volume to effect. In addition to dietary effects, laxation is
increase fecal bulk. All of these mechanisms can also influenced by personality type, and level of
additively increase bulk in the colon, often speed- stress and physical activity with introverted,
ing up the rate of passage through the bowel. relaxed or sedentary individuals having the slow-
Each fiber source has a different bulking and est laxation rates [29].
5.4 Laxation Effects of Fiber-Rich Diets, and Specific Foods and Supplements 149
Table 5.1 Fecal bulking index and total fiber content per and specific representative RCTs on the laxative
100 g of plant foods [28]. and constipation alleviating effects of fiber-rich
Fecal bulking Total fiber diets and foods are summarized in Table 5.2 [24,
Food index (%) content (g/100 g) 27, 31–57]. Appendix A provides a list of the top
Ingredients 50 fiber rich whole (minimally processed) plant
Wheat bran 100 44 foods.
Wheat germ 37 16
Rye flour 21 12
Pea flour 11 16
5.4.1 Systematic Reviews
Soy flour 9 18
and Meta-Analyses
Oat bran 8 11
Corn meal 2 7
Eight systematic reviews and meta-analyses pro-
Breakfast Cereals
vide an important overview of the best fiber-rich
All-bran 51 30
foods for laxation and alleviation of constipation
Bran flakes 26 19
[24, 31–34, 39, 41, 51]. These systematic reviews
Muesli 17 7
suggest that adequate intake of fiber from cereal,
Rolled oats 17 9
Puffed wheat 8 8
fruits, vegetables and common fiber-rich food
Special K 8 4
ingredients including polydextrose, psyllium and
Wheat Chex 3 2 chicory inulin have the potential to increase
Puffed rice −0.4 6 population- wide levels of regularity and may
Cornflakes −2 3 play a role in providing constipation relief.
Bakery Products
Ryvita crisp bread 23 14 5.4.1.1 Cereal, Fruit and Vegetable
Whole wheat bread 12 6 Sources
Multi-grain bread 4 6 A 2016 systematic review of intervention trials on
White bread 1 3 cereal, fruit and vegetable fibers (136 experimen-
Vegetables/Pulses tal studies; healthy subjects) found that cereal and
Lentils, boiled 9 2 vegetable fiber resulted in similarly increased fecal
Green peas, boiled 7 6 weight with fruit fiber being less effective [32].
Spinach, cooked 6 2 Less fermentable food fibers increased fecal
Cabbage, boiled 4 2 weight to a greater amount than more fermentable
Carrots, cooked 3 3 fibers. Fiber did not change transit time in indi-
Fruit viduals with an initial time of <48 h. In those with
Pear, dried 27 7 an initial transit time ≥48 h, transit time was
Apricot, dried 3 3 reduced by approximately 30 min per gram of
cereal, fruit or vegetable fiber, regardless of fer-
mentability. This analysis indicates that slow tran-
5.4 axation Effects of Fiber-
L sit time (≥48 h) may be normalized by increasing
Rich Diets, and Specific fiber, regardless of the type consumed.
Foods and Supplements
5.4.1.2 Cereal Sources
Although the beneficial effects of whole-grain and Four specific systematic reviews and/or meta-
wheat bran on laxation were known since analyses have evaluated the effects of cereal
Hippocrates in 370 BC, the advancement of the fiber-rich food sources on bowel function [24, 33,
dietary fiber hypothesis refocused interest on the 34, 39]. A 2015 systematic review of cereal fiber
effects of dietary fiber and digestive health in a intervention in studies with healthy subjects (65
wider range of diets and foods [30]. Meta-analyses trials; 90% wheat bran) found that wheat bran
Table 5.2 Summary of fiber-rich foods, food ingredient, dietary supplement intervention trials in laxation and
constipation.
Systematic Reviews and Meta-analyses of Intervention Trials
Yu et al. (2017) Prebiotics:5 RCTs; 199 Prebiotics increased weekly stool frequency by 1
Determine the effects of patients. Symbiotics: bowel movement/week and improved stool
prebiotics and symbiotics on 8 RCTs; 825 patients. consistency. Subgroup analysis showed specific
adults with functional effects for galacto-oligosaccharides on stool
constipation [31]. frequency, consistency, ease of defecation and
abdominal pain. Symbiotics significantly
improved stool frequency by 1.15 bowel
movements/ week, improved consistency and
reduced whole-gut transit time by 13.5 h in
patients with functional constipation. Subgroup
analysis showed specific effects for fructo-
oligosaccharides and probiotic combinations on
stool frequency, consistency, straining defecation
and bloating.
DeVries et al. (2016) 136 experimental studies; in Cereal and vegetable fiber had similar effects on
Summarize the effects of healthy subjects. fecal weight whereas fruit fibers were less
cereal, vegetable and fruit effective. Lower fermentable fibers increased fecal
fiber on fecal weight and weight to a greater degree than more fermentable
transit time [32]. fibers. Fiber did not change transit time in those
with an initial time of <48 h. In those with an
initial transit time ≥48 h, transit time was reduced
by 30 min per gram of cereal, fruit or vegetable
fibers, regardless of fermentability.
DeVries et al. (2015) 65 intervention studies; Each extra g/day of wheat fiber increased total
Review and quantitatively among generally healthy stool weight by 3.7 g (p < .0001), dry stool weight
examine the effects of cereal populations. by 0.75 g (p < .0001), and stool frequency by
fiber-rich foods and 0.004 times (p = .0346). Transit time decreased by
ingredients on bowel 0.78 h per additional g/day (p < .0001) of wheat
function [33]. fiber among those with initial transit times >48 h.
Thies et al. (2014) 14 intervention trials. Trials in healthy subjects suggest that oats or oat
Systematically review bran can significantly increase stool weight and
intervention studies on the decrease constipation. Oat consumption
effects of oats or oat bran on significantly increased wet and dry stool weight in
bowel function [39]. 6 out of 9 studies (from 15 to 88% increase) and 5
out of 6 studies (from 15 to 101% increase),
respectively. Stool frequency did not change
significantly in 5 studies, improved in 2 studies
and reduced in 1 study relative to wheat-bran and
rice-bran interventions. Transit time decreased
significantly by 17% in only 1 out of 4 studies.
Lever et al. (2014) 4 RCTs; one in constipated In one trial with constipated subjects, 3 weeks of
Assess the effect of prunes on subjects and three in prune consumption (100 g/day) improved stool
stool frequency and non-constipated subjects. frequency (3.5 vs. 2.8 per week (p = .006) and
consistency [43]. stool consistency (3.2 vs. 2.8 on Bristol stool form
scale, p = .02) compared with psyllium (22 g/day)
as a positive control. In the 3 trials with non-
constipated subjects, prunes softened stool
consistency and increased stool weight (628 g
vs. 514 g/72-h wet weight, p = .001) compared
with control.
Yang et al. (2012) 5 placebo controlled, Increasing fiber resulted in significantly increased
Investigate the effects of double-blinded RCTs; mean stool frequency by 19% vs. placebo (p < .05)
dietary fiber on stool weight limited to bran and but there was no significant difference in stool
and transit time [24]. glucomannan primarily. consistency, laxative use and painful defecation
between the two groups. Improved stool frequency
was reported by all 5 RCTs, with either a trend or
a significant improvement for the fiber group vs.
control.
Suares and Ford (2011) 6 RCTs (3 psyllium, 1 bran, Compared with placebo, psyllium improved global
Assess in systematic review 1 rye bread and 1 inulin). symptoms (86% vs. 47%), straining (56% vs. 29%),
the efficacy of soluble and pain on defecation, stool consistency, increase in
insoluble fiber the mean number of stools per week (3.8 stools
supplementation in the per week after therapy compared with 2.9 stools
management of chronic per week at baseline), and a reduction in the
idiopathic constipation [51]. number of days between stools. Evidence for any
benefit of insoluble fiber was inconsistent. Formal
meta-analysis was not undertaken due to concern
about methodological quality of identified studies.
Muller-Lissner et al. (1988) 20 RCTs. Bran increased the stool weight and decreased the
Investigate the effects of transit time in each study in healthy controls and
wheat bran on stool weight in patients with irritable bowel syndrome, with
and transit [34]. diverticula, and with chronic constipation.
However, bran was shown to be only partially
effective in restoring normal stool weight and
transit time in constipated subjects.
Mixtures of Fiber-Rich Plant Foods
Wisten and Messner (2005) Parallel RCT: Patients in the porridge group had a daily
Study the effects of daily 20 patients in secondary defecation without laxatives on average 76% of
consumption of a fruit- and geriatric hospitals; porridge the time compared with 23% of the time in the
fiber-rich porridge on stool (flaxseed, chopped prunes non-porridge group (p = .003). The discomfort
frequency, perceived and apricots, raisins, rolled was less in the porridge group (2.5 vs. 6.5 on a
wellbeing and laxative usage, oats and oat bran) vs. control 10-degree visual analogue scale (p = .008) when
when compared with standard diet without compared with the control group. The cost for
traditional treatment with porridge; 1-week run-in and laxatives was lower in the porridge group by 93%.
laxatives, in geriatric patients 2-week study.
(Sweden) [41].
Haack et al. (1998) Crossover RCT: Fiber provided by a mixed-food diet: (1) increases
Determine the responses of 9 healthy, young men; stool weight as effectively as cereal bran; (2) even
healthy adult men to increased consumed constant diets high amounts of fiber do not change transit time or
intakes of fiber-rich foods, with 3 amounts of fiber defecation frequency if subjects already have
<15, 30, and 45 g/day provided by a mixture of between 1 to 3 daily bowel movements; (3) food
(US) [35]. fruit, vegetables, and grains patterns containing legumes and whole grains are
which contained 16, 30, and necessary to achieve recommended fiber intakes of
42 g total fiber/day, of which 14 g/1000 kcal; and (4) mixed-food fiber has little
2.9, 4.8, and 7.7 g were effect on calcium balance when calcium intakes
soluble; 1 month each diet. are high (≥1.5 g/day).
(continued)
Cereal Brans
Lawton et al. (2013) Open Label Trial: The inclusion of 1 bowel of bran cereal /daily over
Investigate the effect of wheat 153 low fiber consumers two weeks, significantly improved subjective
bran on subjective perception (baseline <15 g/day); one perception of: (1) bowel function (e.g., ease of
of bowel function and bowl of ready-to-eat defecation) and digestive feelings (less bloating,
digestion, feelings and general breakfast cereal containing constipation, feeling sluggish and digestive
well-being (UK) [36]. at least 5.4 g fiber (3.5 g discomfort) and (2) general wellbeing (feeling less
from wheat bran); 14 days; fat, more mentally alert, slim, happy and energetic
completed a daily symptom whilst experiencing less stress, mental and
diary. physical tiredness, difficulty concentrating and
fewer headaches).
Sturtzel et al. (2009) Single-blind Parallel RCT: Laxatives were successfully discontinued by 59%
Determine effects of adding 30 frail inhabitants of a (p < .001) in the fiber-group; in the control-group
oat-bran to a low fiber geriatric hospital; aged there was an increase of 8% (p = 0.218). Body
habitual diet on use of 57–100 years with laxative weight remained constant in the fiber-group and
laxatives, well-being and body use; 15 subjects received decreased in the control-group (p = .002). The
weight of the inhabitants of a 7-8 g oat-bran/day (fiber oat-fiber supplementation was well tolerated.
long-term-care facility group) mixed up in the daily
(Austria) [40]. common diet of the ward
and 15 received no oat bran
(control group); 12 weeks.
Vuksan et al. (2008) Crossover RCT: Compared to the low fiber control, all study
Assess the effects of 23 free-living participants; cereals induced significant (p < .05) increases in
increasing fiber intake on baseline diet 35% fat and fecal bulk from the control diet; less intestinal
bowel habits and 12 g fiber/day; received 25 transit time; and significantly (p < .05) greater
gastrointestinal tolerance in to 29 g added fiber/day from bowel movement frequency, while maintaining a
healthy persons consuming a each of 5 breakfast cereals: good level of tolerance. Bran buds with psyllium
typical Canadian or US diet All-bran (AB), bran buds was more effective than other cereals in terms of
(Canada) [37]. with corn (BBC), bran buds increasing fecal wet weight (p < .05).
with psyllium (BBP), BBC
with viscous fiber blend
(VFB), or a low-fiber
control; 3 weeks, with each
study arm separated by a
washout of ≥1 week; 7-day
stool collections and a
symptom diary were
obtained during the last
week of each study arm.
Hongisto et al. (2006) 2-by-2 RCT: The rye bread group had shortened total intestinal
Investigate the effects of 59 healthy women with transit time by 0.7 days (p = .007), increased fecal
fiber-rich rye bread and self-reported constipation: 4 frequency by 0.3 per day (<.001), softened feces
yogurt containing diet groups: (1) rye bread by 0.3 units (p < .001) had easier defecation by
lactobacillus GG (LGG) on plus LGG yogurt, (2) rye 0.4 units (p < .001) and gastrointestinal symptoms
intestinal transit time and bread, (3) LGG yogurt, and score was increased by 1.6 units (p < .001)
bowel function (Finland) [38]. (4) control; 3 weeks. compared to the low-fiber toast consumed in the
LGG and control groups. There were fewer
gastrointestinal symptoms in the rye bread plus
LGG group compared to the rye bread group by
1.3 units (p = .027).
Jenkins et al. (1999) Crossover RCT: In both studies, wheat bran supplements
Test the effects of wheat bran In healthy subjects, 2 significantly increased fecal bulk compared to the
particle size on colonic studies, each with 3 phases: control (p < .004), with no significant differences
function (Canada) [27]. (1) 23 subjects; 19 g/day between brans of different particle size and no
fiber from wheat bran with differences in fecal water content. However, higher
mean particle size (MPS) fecal butyrate concentrations (p < .007), and
50 mm or 758 mm in bread breath CH4 levels (p = .025) were seen on the
or a control low fiber bread smaller MPS wheat bran compared to the other
and (2) 24 healthy subjects; two treatments, suggesting increased bacterial
breakfast cereal (ad libitum) fermentation. Fine MPS wheat bran is an effective
with wheat bran MPS fecal bulking agent and may have added
692 mm and 1158 mm and advantages in promoting colonic microbiota
the control was low fiber. health.
1 month metabolic ward;
fecal collections last week of
each diet.
Whole Fruits and Vegetables vs. Fruit and Vegetable Juices
Kelsay et al. (1978) Crossover RCT: High fiber fruits and vegetables vs. low fiber fruit
Assess the effect of high fiber 12 men; age range and vegetable juice intake significantly reduced
vs. low fiber fruits and 37–58 years; weight range fecal transit time (38 vs. 52 h), increased the
vegetables on bowel function 68–95 kg; high fiber fruit number of daily bowel movements (1.4 vs. 1), and
(US) [42]. and vegetable diet vs. a low increased daily wet fecal weight (208 g vs. 90 g).
fiber diet containing fruit
and vegetable juices;
26 days; no washout.
Prunes (Dried Plums)
Attaluri et al. (2011) Single-blind, Crossover The number of complete spontaneous bowel
Assess and compare the RCT: movements per week and stool consistency scores
effects of dried plums and 40 constipated subjects; improved significantly (p < 0.05) with dried plums
psyllium in patients with 37- females; mean age when compared to psyllium (Fig. 5.3). Straining
chronic constipation (US) 38 years; 50 g dried plums and global constipation symptoms did not differ
[44]. or 11 g psyllium twice daily significantly between treatments (p > .05). Dried
(6 g fiber/day); 3 weeks with plums and psyllium were rated as equally
a 1 week washout period. palatable and both were safe and well tolerated.
Kiwi Fruit
Chang et al. (2010) Parallel RCT: The intake of kiwi fruit significantly improved
Examine the impact of kiwi 54 patients with IBS-C and weekly defecation frequency (p < .05) and
fruit intake on bowel function 16 healthy adults; 41 IBS-C decreased colon transit time (p = .026) in the
in patients diagnosed with patients and 16 healthy IBS-C group. This study indicates that kiwi fruit
constipated irritable bowel subjects consumed 2 improves bowel function in adults diagnosed with
syndrome (IBS-C) patients Hayward green kiwi fruits IBS-C.
(Taiwan) [47]. and 13 IBS-C patients in the
control group took two
placebo capsules/day;
4 weeks.
Chan et al. (2007) Open Label Trial: Responder rate was 54.5% in the constipated
Investigate the effect of 33 constipated patients and group. The mean complete spontaneous bowel
increased kiwi fruit intake on 20 healthy volunteers; kiwi movements per week increased after treatment
Chinese constipated patients fruit twice daily; 4 weeks. from 2.2 to 4.4 (p = .013) and transit time
(China) [46]. improved (p = .003). There was also significant
improvement in the scores for bothersomeness of
constipation, and satisfaction of bowel habit, and
decrease in days of laxative used.
(continued)
Rush et al. (2002) Crossover RCT: Kiwi fruit consumption was associated with a
Evaluate the effect of regular 38 healthy adults of significant increase in frequency of defecation
kiwi fruit intake on laxation inage > 60 years consumed (p = .012), stool bulk produced (p = .002) and
elderly people (New Zealand) their normal diet, one kiwi softness of stools (p < .0001).
[45]. fruit per 30 kg bodyweight
vs. no kiwi fruit; 3 weeks,
followed by a 3-week
crossover period; daily
records were taken on
frequency of defecation and
characteristics of the stools.
Polydextrose and Soluble Corn Fiber
Shimada et al. (2015) Triple-blind RCT: The polydextrose group showed significant
Evaluate the effects of 50 constipated dialysis improvement in stool frequency from 3.0 to 7.5
polydextrose on constipated patients; 51–79 years of age; times weekly; there were no laxation problems
dialysis patients (Japan) [50]. laxative for >3 months and such as abdominal distension, cramps or diarrhea
dialysis >6 months; food (p < .001) (Fig. 5.5).
products containing 10 g
polydextrose vs. 0
polydextrose control;
8 weeks.
Timm et al. (2013) Double-blind, Crossover 5-day fecal wet weight was higher after the
Compare the laxative effects RCT: polydextrose and SCF treatments vs. . LF control
of polydextrose and soluble 36 healthy men and women; (p ≤ .0007). The number of stools per day and
corn fiber (SCF) compared to 10-g polydextrose/day and daily fecal output also were significantly greater
a low fiber control eaten daily 10-g SCF/ day vs. low fiber during the polydextrose period compared with the
as a muffin and cereal (US) (LF) control (about 14 g LF control (Fig. 5.4). The whole gut transit time
[48]. fiber/day) control diet; did not differ among treatments. The polydextrose
10-day treatment with a treatment resulted in a softer stool (p = .002) than
2-week washout period; the SCF and LF control. Fecal pH was lowered by
collected fecal samples the polydextrose treatment (p = .02), whereas SCF
during the last 5 days of tended to lower it compared with the LF control
each treatment and (p = .07). Polydextrose and SCF subjects reported
completed food diaries and significantly more flatulence compared with when
gastrointestinal tolerance they consumed the LF control.
questionnaires on day 1, 2,
and 10.
Vester Boler et al. (2011) Crossover RCT: Fecal wet weight was highest (p = .03) when
Evaluate digestive effects of 21 healthy adult men; subjects consumed SCF compared with
polydextrose and soluble 21 g/day polydextrose or NFC. Fecal dry weight tended to be greater
maize fiber in healthy adults soluble maize fiber (SCF) (p = .07) when subjects consumed polydextrose
(US) [49]. vs. no supplemental fiber compared with NFC. Bifidobacterium spp.
(NFC) in a snack bar; concentrations were greater (p < .05) when
21 days with fecal collection subjects consumed SCF compared with NFC. All
during the last 5 days. tolerance scores were low (<2.5), indicating only
slight discomfort; although flatulence (p = .001)
and distention (p = .07) were increased by
polydextrose and SCF vs. NFC. Faecal pH was
lower (p < .01) when subjects consumed SCF
compared with NFC, while polydextrose was
intermediate. These functional fibers appear to be
beneficial to gut health while leading to minimal
GI-upset.
Psyllium
Nunes et al. (2005) Double-blind RCT: 87% of individuals receiving psyllium vs. only
Evaluate the effects of 60 adults; 65% women; 10 g 30% of those in the placebo group had normal
psyllium laxatives in adults psyllium daily; 2 weeks. bowel movement frequency (p < 0.001).
with chronic constipation Psyllium was shown to be effective in relieving
(Brazil) [52]. chronic constipation.
McRorie et al. (1998) Double-blind RCT: Psyllium was superior to docusate sodium for
Compare the effects of 170 adults; mean age softening stools by increasing water content and
psyllium and docusate sodium 37 years; 90% women; 5.1 g improving overall laxative efficacy.
on chronic constipation (US) psyllium twice daily;
[53]. 2 weeks.
Ashraf et al. (1995) Double-blind RCT: In individuals with chronic constipation, psyllium
Evaluate the effects of 22 adults; 14 females; 5 g increased stool frequency compared with placebo
psyllium therapy on stool psyllium twice daily; and improved stool consistency, reduced pain on
characteristics and colon 8 weeks. defecation, lessened straining and increased sense
transit in chronic constipation of complete evacuation compared to baseline.
(US) [54].
Stevens et al. (1988) Parallel RCT: Both fiber sources decreased transit time, and
Compare the effects of 12 subjects; psyllium, wheat increased the daily number of defecations and wet
psyllium and wheat bran on bran or low fiber diet; and dry weight of stools. Bran increased transit
colonic transit time and stool 2 weeks. time greater than psyllium and psyllium had a
characteristics (US) [55]. greater effect on stool weight and % bound water.
The fiber sources reduced the subjective ratings of
hard stools by 40% compared to the control low
fiber group.
Fenn (1986) Single-blind RCT: 87% of subjects allocated to psyllium trial
Assess the effect of psyllium 201 subjects; 150 females; reported an improvement in global symptoms
on chronic constipation (UK) primary outcome compared with 47% of subjects receiving placebo
[56]. improvement in global (p < 0.001). Also, psyllium significantly reduced
symptoms; 2-weeks. abdominal pain and straining on defecation.
Psyllium was twice as effective in reducing
chronic constipation symptoms compared to the
placebo.
Chicory Inulin
Micka et al. (2017) Double blind, Crossover Consumption of chicory inulin significantly
Determine the effect of RCT: increased stool frequency compared to placebo
chicory inulin on stool 44 healthy constipated (median 4.0 vs. 3.0 stools/week; p = .038); stools
frequency in healthy subjects; 75% women; mean were softer and there was a trend toward higher
constipated subjects (German) age 47 years; 12 g/day inulin satisfaction vs. placebo (p = .059).
[57]. from chicory or 12 g/day
maltodextrin; 4 weeks.
improved measures of bowel function [33]. constipation but in constipated individuals their
Specifically, when baseline transit time was stool weight and transit time were not completely
>48 h, each extra g/day of wheat bran signifi- restored to normal [34]. A systematic review of
cantly increased total stool weight by 3.7 g and oats products intervention studies (14 trials) sug-
reduced transit time by 45 min. A meta-analysis gests that in healthy subjects oats or oat bran can
of wheat bran trials (20 RCTs; healthy and con- significantly increase stool weight and decrease
stipated subjects) showed that bran increased the constipation [39]. Oat consumption significantly
stool weight and decreased the transit time in increased wet stool weight in six out of nine stud-
healthy controls and in individuals with chronic ies (from 15 to 88% increase) and dry stool
weight in five out of six studies (from 15 to 101% improved stool consistency, ease of defecation
increase). Stool frequency did not change and abdominal pain and (2) symbiotics, espe-
significantly in five studies, improved in two
cially fructo-oligosaccharides and probiotic com-
studies and reduced in one study relative to binations, significantly improved stool frequency
wheat-bran and rice-bran interventions. Transit by 1.15 bowel movements/week and consistency,
time decreased significantly by 17% in only one and reduced gut transit time by 13.5 hours, strain-
out of four studies. A meta-analysis (five double ing defecation and bloating in patients with func-
blind RCTs; primarily bran and glucomannan) tional constipation [31].
demonstrated that increased fiber intake signifi-
cantly improved stool frequency by 19% (p < .05)
but there was no significant improvement in stool 5.4.2 Specific Intervention Trials
consistency or painful defecations [24].
5.4.2.1 Mixed Fiber-Rich Diets
5.4.1.3 Prunes (Dried Plums) Two RCTs provide important insights on the ben-
A systematic review of prunes and gastrointesti- eficial effects of fiber-rich diets on bowel func-
nal function (4 RCTs;165 participants; mean age tion in both regular and low caloric diets [35, 41].
36–54 years; 73% women; three studies in
healthy and one in constipated subjects; duration 5.4.2.2 Dose-Response
2 weeks to 3 months; 84–100 g prunes/day; con- A US crossover, dose response RCT (9 healthy,
trols included grape juice, dried apples, cookies, young male students; 16, 30, and 42 g fiber/day
psyllium plus water) found that in constipated from a mixture of fruits, vegetables and cereal
subjects prunes were similar to psyllium in grains; duration 1 month for each dose with a
increasing stool frequency and improving stool 15-day washout) found that mean daily stool
consistency and in non-constipated subjects weights increased with the amount of fiber intake;
prunes softened stool consistency and increased wet fecal mass was 109, 156 and 195 g for 16, 30,
stool weight [43]. and 42 g fiber intake, respectively [35]. Increasing
dietary fiber intake from 16 to 30 g/day increased
5.4.1.4 Psyllium mean stool frequency from 0.7 to about 1 per day
A systematic review of the efficacy of soluble but increasing fiber intake from 30 to 42 g/day
and insoluble fiber supplementation in the man- did not further increase stool frequency.
agement of chronic idiopathic constipation (6 Increasing fiber intake from a mixture of plant
RCTs including 3 psyllium, 1 bran, 1 rye bread sources tended to be as effective as consuming
and 1 inulin trials) determined that psyllium cereal bran but there was no improvement in
was the most effective fiber in promoting bowel stool frequency if the baseline stool frequency
function [51]. Compared with placebo, psyl- rate was already between 1 and 3 times per day.
lium reduced global symptoms (86% vs. 47%),
straining (56% vs. 29%), pain on defecation, 5.4.2.3 Pajala Porridge
improved stool consistency, and increased Pajala porridge containing rolled oats, oat bran,
mean number of stools per week (3.8 stools per flax seeds, chopped prunes, apricots and raisins
week after therapy compared with 2.9 stools has been shown to be a well-tolerated fiber food
per week at baseline). option for elderly residents in long-term care
facilities to aid in the alleviation of constipation
5.4.1.5 Prebiotics and Symbiotics [41]. A parallel RCT (20 adults age >65 years;
A 2017 meta-analysis (prebiotics: 5 RCTs; breakfast porridge with 7.5 g fiber vs. breakfast
199 patients; symbiotics: 8 RCTs; 825 patients) without porridge; 1-week run-in; 2 weeks) dem-
found that: (1) prebiotics, especially galacto- onstrated that fiber-rich porridge was effective,
oligosaccharides, increased weekly stool fre- well-liked and tolerated and reduced the need
quency by one bowel movement/week and for laxatives in geriatric patients. Specifically,
the porridge group significantly improved the 36 years; breakfast cereal with 19 g fiber from
number of bowel movements without laxatives wheat bran/day made from medium and coarse
compared to the control group (76% vs. 23% of particle size bran vs. a low fiber cereals control;
the time) and bowel movement discomfort was 1 month metabolic ward) showed that both
significantly 40% lower in the porridge vs. the medium and coarse wheat bran breakfast cereals
control group. similarly increased daily wet stool bulk com-
pared to the low fiber breakfast cereal [27].
5.4.2.4 Cereal Bran-Rich Foods Smaller particle size bran fiber did not adversely
Six intervention trials assessed the effects of affect stool bulking or frequency of bowel
cereal bran-rich foods in five publications [27, movements.
36–38, 40].
Breads
Breakfast Cereals Two RCTs evaluated the effects of bran enriched
Three intervention trials evaluated the effect of bread on bowel function [27, 38]. A 3-phase met-
increased wheat bran in breakfast cereal [27, 36, abolic, crossover RCT (23 healthy subjects; 12
37]. An open label trial (153 subjects; 81 women and 11 men; mean age 58 years; bread
females and 72 males; mean age 34 years; mean with 19 g fiber/day from very fine or medium
baseline total fiber intake 10.5 g/day; one bowel wheat bran vs. a low fiber control bread; 2 weeks)
of wheat bran-containing ready-to-eat-breakfast found that both fine and medium wheat bran par-
cereal with 5.4 g/day fiber (3.5 g from wheat ticle size enriched bread significantly increased
bran); 14 days) found significant improvements fecal bulk by 58 to 68 g/day compared to the low
in subjective perception of bowel function (e.g., fiber bread [27]. There was a small but signifi-
ease of defecation), digestive feelings (less cantly increased bowel movement frequency for
bloating, constipation, feeling sluggish and the medium bran fiber bread (1.5/day) compared
digestive discomfort) and general wellbeing to the fine bran fiber bread (1.4/day) and to low
(feeling less fat, more mentally alert, slim, fiber bread (1.3/day). This study confirms the
happy and energetic; while experiencing less effectiveness of bread with fine wheat bran in
stress, mental and physical tiredness, difficulty improving fecal bulk and laxation and also found
concentrating and fewer headaches) [36]. A par- that the fine ground wheat bran was significantly
allel RCT (23 subjects; mean age 35 years; 12 fermented to produce a significantly higher fecal
women and 11 men; 4 different fiber rich break- butyrate concentration, a contributor to colonic
fast cereals including All Bran or Bran Buds microbiota health, compared to the medium
with fiber blends of corn and psyllium at 2.5 ground bran. A parallel RCT (59 women with
servings/day to provide about 25 g fiber/day constipation; mean age 41 years; 4 diets: (1)
added to the habitual Western diet with 12 g whole rye bread (30 g fiber/day), (2) whole rye
fiber/day; 3 weeks; 1 week washout) showed bread plus Lactobacillus rhamnosus GG (LGG)
that all the fiber enriched breakfast cereals sig- enriched yogurt, (3) LGG enriched yogurt, and
nificantly improved fecal wet bulk (199 vs. (4) a control low fiber bread; 3 weeks) found that
128 g/day), reduced transit times (29 hrs vs. rye bread shortened total intestinal transit time by
41 hrs), and increased bowel movement fre- 17 hours, increased fecal frequency by 0.3 per
quency (1.2/day vs. 0.97/day) compared to the day, softened feces by 0.3 units and made defeca-
low fiber control diet, while maintaining a good tion easier by 0.4 units, but also increased gastro-
level of tolerance [37]. Bran cereal with psyl- intestinal symptoms score (higher bloating and
lium was more effective than the other breakfast flatulence) by 1.6 units compared to the low-fiber
cereals in increasing stool wet weight for softer toast consumed in the LGG and control groups
consistency. A 3-phase crossover RCT evaluat- [38]. There were fewer gastrointestinal symp-
ing the effect of bran particle size (24 healthy toms in the rye bread plus LGG group compared
subjects; 12 females and 12 males; mean age to the rye bread group by 1.3 units.
5.4.2.5 Oat Bran in Soups and Desserts prunes significantly improved constipation symp-
A blinded parallel RCT (30 assisted living sub- toms as reflected by a significant increase in the
jects; mean age 85 years; 5.1 g oat bran fiber/day number of complete and spontaneous bowel
vs. 0 fiber control; 12 weeks) found that oat bran movements/week (Fig. 5.3) and improved stool
blended into the daily lunch soup or dessert consistency (softer stools) compared with a psyl-
served in a standard diet, or incorporated into the lium fiber supplement [44]. This study showed
afternoon cake significantly reduced laxative that psyllium was also useful in improving bowel
usage by 59% whereas the control group slightly symptoms in individuals with mild to moderate
increased laxative use by 8% [40]. The oat bran constipation and affirmed prior studies on psyl-
was well tolerated. lium in chronic constipation. The laxative effects
of prunes (dried plums) are most likely due to a
5.4.2.6 Fruits and Vegetables combination of sorbitol (14.7 g per 100 g) and
dietary fiber (6 g per 100 g).
ruit and Vegetables: Whole vs. Juice
F
A crossover RCT (12 men; age range 37–58 years; Kiwi Fruit
weight range 68–95 kg; high fiber fruit and veg- Three RCTs assess the effects of kiwi fruit
etable diet vs. a low fiber diet containing fruit and intake on bowel laxative function. Kiwi fruit cell
vegetable juices; 26 days; no washout) showed walls have unique viscous polysaccharides with
that the higher fiber intake from fruits and vege- exceptionally high swell or water binding capac-
tables vs. low fiber juice intake significantly ity and have fecal bulking and stool softening
reduced fecal transit time (38 vs. 52 h), increased properties similar to that of psyllium [45–47]. A
the number of daily bowel movements (1.4 vs. 1), 2002 crossover RCT with elderly adults (38
and increased daily wet fecal weight (208 g vs. healthy, overweight subjects; mean age 73 years;
90 g) [42]. 25 females and 13 males; 2 kiwi fruit/day vs. no
kiwi fruit; 3-week duration with a 3-week wash-
runes (Dried Plums)
P out) showed that kiwi fruit significantly
A crossover RCT comparing prunes vs. psyllium enhanced laxation, including bulkier and softer
(40 constipated subjects; 92% women; mean age stools, increased ease of defecation, and more
38 years; 50 g prunes or 11 g psyllium twice frequent bowel movements [45]. A second RCT
daily; 3 weeks; 1-week washout) found that in constipated Chinese subjects (33 constipated
Baseline Treatment (3 weeks) Follow-up (no treatment for 6 weeks)

8
7
Weekly Bowel Movements
2
Fig. 5.3 Effect of dried
1
prunes vs. psyllium
supplement on bowel 0
movements/week 50 g prunes (dried plums) 11 g psyllium
(p = .002) (adapted
from [44]) Twice Daily
subjects; mean age 50 years; 24 females; 20 muffins and cereal vs. low fiber control; 10 days
healthy subjects with regular bowel movements; with 2 weeks of washout) showed that polydex-
mean age 51 years; 16 females; kiwi fruit twice trose enriched foods significantly improved laxa-
daily; 4 weeks) found that kiwi fruit signifi- tion activity compared with low fiber control
cantly doubled complete spontaneous bowel foods (Fig. 5.4) [48]. This study shows that the
movements from 2 to 4 times per week along addition of 20 g polydextrose in foods is well tol-
with significantly improving transit time and erated and has moderate laxative effects. Similar
satisfaction with bowel habits [46]. However, in findings were observed for consuming 21 g/day
the subjects with normal regularity, kiwi fruit of polydextrose in a snack bar compared to a no
resulted in no significant changes in normal fiber control snack bar [49]. A triple-blind, paral-
bowel function. In a third RCT, subjects with a lel RCT (50 constipated Japanese hemodialysis
combination of irritable bowel syndrome and patients; mean age 65 years; 60% with diabetes;
constipation (IBS-C) (54 subjects; 49 females; 2 34 men and 16 women; 10 g polydextrose/day in
kiwi fruit/day; 4 weeks) reported that kiwi fruit foods vs. control; 4 weeks) demonstrated that
consumption significantly shortened colon tran- polydextrose significantly improved stool fre-
sit time, increased defecation frequency and quency, softened the stool, and improved ease of
improved overall bowel function [47]. This defecation, without inducing adverse gastrointes-
study suggests that kiwi fruit (taken as a routine tinal effects (Fig. 5.5) [50].
dietary constituent) appears to be a safe and
effective natural laxative for individuals with Psyllium
IBS-C. Five RCTs support psyllium’s effectiveness as a
relatively low fermentable, stool bulking gel in
5.4.2.7 Common Fiber-Rich Food and promoting laxation and alleviating chronic con-
Supplement Ingredients stipation [52–56]. A double-blind placebo con-
trolled RCT (60 adults with chronic constipation;
Polydextrose 65% women; 10 g psyllium daily; 2 weeks)
Three RCTs evaluated the effect of polydextrose, found that 87% of individuals receiving psyl-
a common synthetic low energy, low-moderate lium vs. only 30% of those in the placebo group
fermentability fiber ingredient, on bowel function had normal bowel movement frequency
[48–50]. A double-blind, placebo, crossover (p < 0.001) [52]. The other psyllium RCTs all
RCT (36 healthy adults; mean age 26 years; 18 showed similar effects in relieving chronic con-
females and 18 males; 20 g polydextrose/day in stipation [53–56]. These trials consistently
Polydextrose muffins/cereal Low fiber muffins/cereal

6
5
Bowel Function
Fig. 5.4 Effect of 3

adding 20 g/day
polydextrose in the diet 2
vs. low fiber diet
(p < .05 for all) (adapted 1
from [48]). * 1 (separate
hard lumps) and 7 0
(entirely liquid) and ** 0 # stools over Bristol stool Flatulence**
(none) and 10 (extreme) 5 days consistency*
Fig. 5.5 Laxative 10 g Polydextrose/day Placebo

effects of polydextrose
9
based jelly intake on 50
constipated Japanese 8
hemodialysis outpatients
Weekly Defecation Frequency

(51–79 years of age) 7
(p < 0.05) (adapted
from [50]) 6
0
Baseline Week 1 Week 2 Week 3 Week 4 2 week
follow-up
show that psyllium improved the stool fre- Conclusions

quency and softness as well as improving global Globally constipation is a common com-
symptoms including stool consistency, and plaint, especially among elderly adults,
reducing pain and straining on defecation in which affects most people at some time in
individuals with constipation. In a comparison their lives. The consumption of adequate
study of psyllium and wheat bran, psyllium was fiber (>25 g/day or 14 g/1,000 kcals), rec-
shown to be more effective than wheat bran at ommended fluid intake, and regular physical
increasing stool water (softening) and overall activity, are especially beneficial in
stool weight but wheat bran was more effective preventing and alleviating constipation.

in speeding up fecal transit time verses a low Fiber m echanisms are associated with
fiber control [54]. improved laxation by increasing: stool
weight and bulk (through fiber’s physical
Chicory Inulin volume and water holding capacity), micro-
A 2017 German double blind, placebo controlled, biota numbers and volume, and gas volume
cross-over RCT (44 healthy constipated subjects; trapped in the stool to increase bowel move-
75% women; mean age 47 years; 12 g/day inulin ment frequency and quality, especially in
from chicory or 12 g/day maltodextrin; 4 weeks) constipated individuals. Adequate intake of
found that chicory inulin significantly increased fiber from cereal, fruits, vegetables and
stool frequency compared to placebo (median 4.0 common fiber-rich food ingredients includ-
vs. 3.0 stools/week (p = .038), increased softening polydextrose, psyllium, chicory inulin
ing of stools and subjects trended toward higher and prebiotics or symbiotics have the poten-
satisfaction vs. placebo (p = .059) [57]. The level tial to increase population-wide levels of
of flatulence was rated as 1.1 with intake of pla- regularity and provide constipation relief. In
cebo vs. 1.9 with chicory inulin consumption, general, less fermentable dietary fiber tends
which were both relatively low on a 5-point scale to increase fecal weight to a greater amount
(0–4) but the difference was statistically signifi- than more fermentable fibers. For wheat
cant (p < 0.001). bran, the most widely studied fiber, when
Appendix A 161
baseline transit time was >48 h, each extra time was reduced by approximately 30 min
g/day of wheat bran significantly increased per gram of cereal, fruit or vegetable fiber,
total stool weight by 3.7 g and reduced tran- regardless of fermentability. Several RCTs
sit time by 45 min. Increased fiber intake did suggest that daily intake of prunes (dried
not change transit time in individuals with plums) and/or kiwi fruit can help in reliev-
an initial time of <48 h. However, in people ing constipation symptoms similarly to
with an initial transit time ≥48 h, transit psyllium.
ppendix A: Fifty High Fiber Whole or Minimally Processed Plant Foods

A
Ranked by Amount of Fiber per Standard Food Portion Size
Shredded wheat ready-to-eat cereal 1–1 1/4 cup (50-60 g) 5.0–9.0 155–220 3.2–3.7
Green peas, cooked (fresh, frozen, canned) 1/2 cup (80 g) 3.5–4.4 59–67 0.7–0.8

Orange 1 medium (130 g) 3.1 69 0.5
Banana 1 medium (118 g) 3.1 105 0.9
Dates 1/4 cup (about 38 g) 2.9 104 2.8
Nutr Diet. 2015;115:1861–1870.
trends. 2015;97, 98;Table D1.8.
Slavin, J.L. Position of the American Dietetic Association: Health implications of dietary fiber. J. Am. Diet. Assoc.
2008;108:1716–1731.
gov/ accessed 8.22.2015.
position statement on constipation. Gastroenterology.

References 2013;144:211–7.
6. Bharucha AE, Pemberton JH, Locke GR III. American
1. De Giorgio R, Ruggeri E, Stanghellini V, et al. Gastroenterological Association technical review on
Chronic constipation in the elderly: a primer for the constipation. Gastroenterology. 2013;144:218–38.
gastroenterologist. BMC Gastroenterol. 2015;15:130. 7. Tabbers MM, Benninga MA. Constipation in chil-
https://doi.org/10.1186/s12876-015-0366-3. dren: fibre and probiotics. BMJ Clin Evid. 2015; pii
2. Tack J, Müller-Lissner S, Stanghellini V. Diagnosis 0303. PMID:25758093.
and treatment of chronic constipation–a European per- 8. Borowitz SM, Cox DJ, Tam A, et al. Precipitants of
spective. Neurogastroenterol Motil. 2011;23:697–710. constipation during early childhood. J Am Board Fam
3. Higgins PDR, Johanson JF. Epidemiology of consti- Pract. 2003;16:213–8.
pation in North America: a systematic review. Am J 9. Wald A. Constipation. JAMA. 2016;315(2):214.
Gastroenterol. 2004;99(4):750–9. 10. Wald A. Constipation: pathophysiology and manage-
4. Connell AM, Hilton C, Irvine G, et al. Variation of ment. Curr Opin Gastroenterol. 2015;31:45–9.
bowel habit in two population samples. Br Med J. 11. Dukas L, Willett WC, Giovannucci EL. Association
1965;5470:1095–9. between physical activity, fiber intake, and other life-
5. Bharucha AE, Dorn SD, Lembo A, Pressman style variables and constipation in a study of women.
A. American Gastroenterological Association medical Am J Gastroenterol. 2003;98(8):1790–6.
References 163
12. Markland AD, Palsson O, Goode PS. Association of 27. Jenkins DJ, Kendall CW, Vuksan V, et al. The

low dietary intake of fiber and liquids with constipa- effect of wheat bran particle size on laxation and
tion: evidence from the National Health and Nutrition colonic fermentation. J Am Coll Nutr. 1999;18(4):
Examination Survey (NHANES). Am J Gastroenterol. 339–45.
2013;108(5):796–803. 28. Monro JA. Adequate intake values for dietary fibre
13. Ford AC, Moayyedi P, Lacy BE, et al. American
based on faecal bulking indexes of 66 foods. Eur J
College of Gastroenterology Monograph on the Clin Nutr. 2004;58:32–9.
management of irritable bowel syndrome and 29. Tucker DM, Sandstead HH, Logan GM, et al. Dietary
chronic idiopathic constipation. Am J Gastroenterol. fiber and personality factors as determinants of stool
2014;109:S2–S26. output. Gastroenterology. 1981;81:879–83.
14. Dahl WJ, Stewart ML. Position of the Academy of 30. Burkitt DP, Walker AR, Painter NS. Effect of dietary
Nutrition and Dietetics: health implications of dietary fibre on stools and the transit-times, and its role in the
fiber. J Acad Nutr Diet. 2015;115:1861–70. causation of disease. Lancet. 1972;2:1408–12.
15. European Food Safety Authority Panel on Dietetic 31. Yu T, Zheng YP, Tan JC, et al. Effects of prebiotics
Products, Nutrition, and Allergies. Scientific opinion and synbiotics on functional constipation. Am J Med
on dietary reference values for carbohydrates and Sci. 2017;353(3):282–92. https://doi.org/10.1016/j.
dietary fibre. EFSA J. 2010;8(3):1462. amjms.2016.09.014.
16. Committee on Nutrition, American Academy of
32. de Vries J, Birkett A, Hulshof T, et al. Effects of cereal,
Pediatrics. Carbohydrate and Dietary Fiber. Kleinman fruit and vegetable fibers on human fecal weight and
RE (ed). Pediatric Nutrition Handbook. 6th edition, transit time: a comprehensive review of intervention
Community on Nutrition, American Academy of trials. Nutrients. 2016;8:130. https://doi.org/10.3390/
Pediatrics, Illinois USA. 2009; 104. nu8030130.
17. Murakami K, Sasaki S, Okubo H, et al. Association 33. de Vries J, Miller PE, Verbeke K. Effects of cereal
between dietary fiber, water and magnesium intake fiber on bowel function: A systematic review of
and functional constipation among young Japanese intervention trials. World J Gastroenterol. 2015;
women. Eur J Clin Nutr. 2007;61:616–22. 21(29):8952–63.
18. Wrick KL, Robertson JB, Van Soest PJ, et al. The 34. Muller-Lissner SA. Effect of wheat bran on weight
influence of dietary fiber source on human intestinal of stool and gastrointestinal transit time: a meta-
transit and stool output. J Nutr. 1983;113:1464–79. analysis. BMJ. 1988;26:615–7.
19. Anti M, Pignataro G, Armuzzi A, et al. Water supple- 35. Haack VS, Chesters JG, Vollendorf NW, et al.

mentation enhances the effect of high fiber diet on stool Increasing amounts of dietary fiber provided by foods
frequency and laxative consumption in adult patients normalizes physiologic response of the large bowel
with functional constipation. Hepatogastroenterology. without altering calcium balance or fecal steroid
1998;45(21):727–32. excretion. Am J Clin Nutr. 1998;68:615–22.
20. Cummings JH. The effect of dietary fiber on fecal 36. Lawton CL, Walton J, Hoyland A. Short term (14
weight and composition. In:3rd ed. Spiller GA, days) consumption of insoluble wheat bran fibre-
ed. CRC Handbook of Dietary Fiber in Human containing breakfast cereals improves subjective
Nutrition. 3rd ed. Boca Raton, FL: CRC Press; 2001. digestive feelings, general wellbeing and bowel
p. 183–241. function in a dose dependent manner. Nutrients.
21. Roma E, Adamidis D, Nikolara R, et al. Diet and 2013;5:1436–55.
chronic constipation in children: role of dietary fiber. 37. Vuksan V, Jenkins AL, Jenkins DJA, et al. Using

J Pediatr Gastroenterol Nutr. 1999;28(2):160–74. cereal to increase dietary fiber intake to the recom-
22. Abdullah MM, Gules CL, Marinangeli CP, et al.
mended level and the effect of fiber on bowel function
Dietary fibre intake and reduction in functional con- in healthy persons consuming North American diets.
stipation rates among Canadian adults: a cost-of ill- Am J Clin Nutr. 2008;88:1256–62.
ness analysis. Food Nutr Res. 2015;59:28646. https:// 38. Hongisto S-M, Paajanen L, Saxelin M, Korpela R. A
doi.org/10.3402/fnr.v59.28646. combination of fibre-rich rye bread and yoghurt con-
23. Schmier JK, Miller PE, Levine JA, et al. Cost savings taining Lactobacillus GG improves bowel function
of reduced constipation rates attributed to increased in women with self-reported constipation. Eur J Clin
dietary fiber intake: a decision-analytical model. Nutr. 2006;60:319–24.
BMC Public Health. 2014;14:374. 39. Thies F, Masson LF, Boffetta P, Kris-Etherton P. Oats
24. Yang J, Wang H-P, Zhou L, Xu C-F. Effect of dietary and bowel disease: a systematic literature review. Br J
fiber on constipation: A meta-analysis. World J Nutr. 2014;112:S31–43.
Gastroenterol. 2012;18(48):7378–83. 40. Sturtzel B, Elmadfa I. Intervention with dietary fiber to
25. Cummings JH. Constipation, dietary fibre and the
treat constipation and reduce laxative use in residents
control of large bowel function. Postgrad Med J. of nursing homes. Ann Nutr Metab. 2008;52(Suppl
1984;60:811–9. 1):54–6.
26. Monro JA. Faecal bulking index: A physiological
41. Wisten A, Messner T. Fruit and fibre (Pajala porridge)
basis for dietary management of bulk in the distal in the prevention of constipation. Scand J Caring Sci.
colon. Asia Pacific. J Clin Nutr. 2000;9(2):74–81. 2005;19:71–6.
42. Kelsay JL, Behall KM, Prather ES. Effect of fiber 51.

Suares NC, Ford AC. Systematic review: the
from fruits and vegetables on metabolic responses of effects of fibre in the management of chronic idio-
human subjects. Am J Clin Nutr. 1978;31:1149–53. pathic constipation. Aliment Pharmacol Ther.
43. Lever E, Cole J, Scott SM, et al. Systematic review: 2011;33:895–901.
the effect of prunes on gastrointestinal function. 52. Nunes FP, Nunes CP, Levis E, et al. A double-blind
Aliment Pharmacol Ther. 2014;40:750–8. trial of a celandin, Aloe vera and psyllium laxative
44. Attaluri A, Donahoe R, Valestin J, et al. Randomised preparation in adult patients with constipation. Rev
clinical trial: dried plums (prunes) vs. psyllium for con- Bras Med. 2005;62:352–7.
stipation. Aliment Pharmacol Ther. 2011;33:822–8. 53. McRorie JW, Daggy BP, Morel JG, et al. Psyllium
45. Rush EC, Patel M, Plank LD, et al. Kiwifruit pro- is superior to docusate sodium for treatment of
motes laxation in the elderly. Asia Pacific. J Clin Nutr. chronic constipation. Aliment Pharmacol Ther.
2002;11(2):164–8. 1998;12:491–7.
46. Chan AO, Leung G, Tong T, Wong NYH. Increasing 54. Ashraf W, Park F, Lof J, Quigley EM. Effects of psyl-
dietary fiber intake in terms of kiwifruit improves con- lium therapy on stool characteristics, colon transit and
stipation in Chinese patients. World J Gastroenterol. anorectal function in chronic idiopathic constipation.
2007;13(35):4771–5. Aliment Pharmacol Ther. 1995;9:639–47.
47. Chang C-C, Lin Y-T, Lu Y-T, et al. Kiwifruit improves 55. Steven J, Van Soest PJ, Robertson JB, Levitsky

bowel function in patients with irritable bowel syn- DA. Comparison of the effects of psyllium and wheat
drome with constipation. Asia Pacific. J Clin Nutr. bran on gastrointestinal transit time and stool charac-
2010;19(4):451–7. teristics. J Am Diet Assoc. 1988;88(3):323–6.
48. Timm DA, Thomas W, Boileau TW, et al. Polydextrose 56. Fenn GC, Wilkinson PD, Lee CE, Akbar FAA.
and soluble corn fiber increase five-day fecal wet weight general practice study of the efficacy of regu-
in healthy men and women. J Nutr. 2013;143:473–8. lan in functional constipation. Br J Gen Pract.
49. Vester Boler BMV, Rossoni Serao MC, Bauer LL, 1986;40:192–7.
et al. Digestive physiological outcomes related to 57. Micka A, Siepelmeyer A, Holz A, et al. Effect of
polydextrose and soluble maize fibre consumption by consumption of chicory inulin on bowel function in
healthy adult men. Br J Nutr. 2011;106:1864–71. healthy subjects with constipation: a randomized,
50. Shimada M, Nagano N, Goto S, et al. Effect of poly- double-blind, placebo-controlled trial. Int J Food Sci
dextrose intake on constipation in Japanese dialysis Nutr. 2017;68(1):82–9. https://doi.org/10.1080/09637
patients: a triple-blind, randomized, controlled trial. 486.2016.1212819.
J Nutr Sci Vitaminol. 2015;61:345–53.
Dietary Patterns, Foods and Fiber
in Irritable Bowel Syndrome 6
and Diverticular Disease
Keywords
Irritable bowel syndrome • Diverticular disease • Colon microbiota
• Dietary fiber • Western dietary pattern • Healthy dietary pattern • Low
FODMAP diets • Butyrate • Psyllium • Wheat bran • Celiac disease
Key Points develop complications such as severe bouts

of diverticulitis or bleeding that may lead to
• Irritable bowel syndrome (IBS) is the most sepsis and death.
common gastrointestinal disorder occurring in • Healthy dietary patterns and low intake of fer-
people <45 years. Diverticular disease is mentable oligosaccharides, disaccharides,
among the most clinically and economically monosaccharides and polyols (FODMAPs)
significant gastroenterological conditions in may help to lower the risk and alleviate symp-
people ≥65 years of age. Having a history of toms associated with IBS and diverticular dis-
IBS appears to increase the risk of diverticular ease. For IBS, psyllium is the most consistent
disease in older age. fiber source found to help provide moderate
• IBS, previously called colitis, does not gener- relief of symptoms. For uncomplicated diver-
ally show visible structural or anatomic abnor- ticular disease, fiber-rich healthy diets and low
malities, but is characterized by abdominal red or processed meat consumption decreases
pain, bloating, distension, and changes in the risk, and fiber-rich diets, and foods or sup-
bowel habits. Celiac disease may be con- plements containing wheat bran, psyllium or
founding and difficult to distinguish from IBS methylcellulose may help to alleviate diver-
symptoms. ticular disease symptoms and/or improve
• Diverticular disease may evolve from bowel function.
colonic diverticulae (herniate pouches) • Fiber related mechanisms that may help
potentially caused by high colonic intralu- reduce risk or manage symptoms of IBS or
minal pressure which occurs in most people uncomplicated diverticular disease are related
with aging but only approximately 20% to: (1) improved colonic health by promoting
of individuals with diverticulae develop better laxation and stool bulk, and a healthier
abdominal symptoms (symptomatic uncom- microbiota ecosystem with higher fecal ratio
plicated diverticular disease). A smaller per- of probiotic to pathogenic bacteria and higher
centage of older individuals eventually butyrate concentrations associated with lower

166 6 Dietary Patterns, Foods and Fiber in Irritable Bowel Syndrome and Diverticular Disease
colonic inflammation and improved colono- bowel. IBS affects 10–15% of the global popula-
cyte structure and function; and (2) reduced tion with peak prevalence in people from 20 to
risk or rate of annual body weight and central 39 years of age and it is twice as common in
abdominal fat gains (or promotion of a grad- females as males [5]. Studies estimate that the
ual lowering of body weight and waist cir- IBS rate in North American children is 14% of
cumference in overweight or obese high school students and 6% of middle school
individuals). students [1, 4]. IBS accounts for as much as 12%
of total visits to primary care providers with
between 2.4 and 3.5 million annual physician
6.1 Introduction visits for IBS in the United States alone [3]. IBS
is generally diagnosed when a person has had
Irritable bowel syndrome (IBS) and diverticular abdominal pain or discomfort at ≥3 times a
disease are both chronic and relapsing functional month for the previous 3 months without other
colonic disorders that are among the most clini- disease or injury that could explain the pain [1].
cally and economically significant global gastro- The pathogenesis of IBS is multifactorial and not
enterological conditions [1–11]. People with IBS completely understood but potential dysfunctions
(and celiac disease) experience pain and disor- that have been reported in patients with IBS
dered bowel habits that develop in childhood and include altered gastrointestinal motility, increased
young adulthood that are similar to diverticular bloating, abnormal flatulence, colonic hypersen-
disease symptoms which usually has onset after sitivity, abdominal pain, and microbiota dysbio-
age 65 years. Dietary fiber (fiber) rich diets and sis [1, 2, 5–8]. Abdominal pain is the most
fiber supplements have been extensively evalu- common symptom and often is described as a
ated for their ability to help in both IBS and cramping sensation. Among patients about 40%
diverticular disease prevention and management. of people have mild IBS, 35% moderate IBS, and
Fifty of the highest whole (minimally processed) 25% severe IBS [3]. IBS has four different sub-
plant food sources of fiber are summarized in types: IBS with constipation, IBS with diarrhea,
Appendix A. The objective of this chapter is to mixed IBS alternating constipation and diarrhea,
review the effects of dietary patterns, low and unsubtyped with a milder degree of abnormal
FODMAP diets and foods, especially those rich stool consistency [1].
in fiber, on IBS and diverticular disease risk and
symptoms.
6.1.2 Diverticular Disease
6.1.1 Irritable Bowel Syndrome Diverticular disease is among the most clinically
and economically significant gastroenterological
Irritable bowel syndrome (IBS) is the most com- conditions in older people. It was virtually
mon functional gastrointestinal disorder occur- unknown before the expansion of low fiber diets
ring in people <45 years [1–4]. It is a chronic and and highly processed food in the early to mid-
relapsing functional colonic disorder that does twentieth century [9–11]. Diverticula (diverticu-
not generally show visible structural or anatomic losis) are colonic submucosal herniated pouches,
abnormalities, but is characterized by abdominal which increase with age and occur in 5–10% in
pain, bloating, distension, and changes in bowel adults under 40 years, 30% by age 50 years, and
habits. IBS patients often have colonic micro- 70% by the age of 85 years, but in most individu-
scopic and molecular abnormalities including als the diverticulae remain asymptomatic show-
lowgrade inflammation and associated neuronal ing no or few complications over one’s lifetime
hyperexcitability, and microbiota dysbiosis. [11–16]. Although over half of adults over
Previously, IBS was called colitis, mucous coli- 65 years old will have diverticulosis, 80% of this
tis, spastic colon, nervous colon, and spastic population remain asymptomatic or only
6.2 Irritable Bowel Syndrome 167
10
Odds Ratio for Diverticular Disease

8
0
< 65 years with IBS >= 65 years no IBS >= 65 years with IBS
Fig. 6.1 Relation between diarrhea predominate irritable bowel syndrome (IBS) and diverticular disease risk with
aging (adapted from [31])
experience infrequent, relatively minor colonic increased odds of colorectal cancer by 36% [17].
symptoms, but approximately 20% may develop In the US, complications associated with diver-
abdominal symptoms (symptomatic uncompli- ticular disease account for >300,000 hospital
cated diverticular disease) and, eventually, com- admissions, 1.5 million inpatient care days, and
plications such as bouts of diverticulitis or ≥$2.5 billion in direct costs [9–11]. The combi-
bleeding [14–16]. Symptomatic uncomplicated nation of an aging population, low dietary
diverticular disease is characterized by recurrent fiber (fiber) Western diets and higher intake
abdominal symptoms several times a year attrib- of fermentable oligosaccharides, disaccharides,
uted to irritated diverticula, with symptoms such monosaccharides and polyols (FODMAPs)
as abdominal pain and bloating similar to those in which are often added to processed foods, are
irritable bowel syndrome. The impact of these factors that may increase the risk of diverticulosis
complaints is variable, and the severity and fre- in individuals with pre-existing colonic structural
quency of symptoms may range from mild and defects which potentially manifest themselves
rare episodes, to a severe, chronic, recurrent dis- during bouts of high colonic intraluminal pres-
order, impacting daily activities and the quality sure [9, 14, 18–30]. Also, diarrhea-predominate
of life of patients. Some 15% of the patients with IBS has been associated with an increased risk of
acute diverticulitis may have resultant complica- developing diverticular disease, especially among
tions with the development of varying levels of people >65 years of age (Fig. 6.1) [31].
abscesses, perforation, fistula, peritonitis,
spasms, and/or bleeding which can be associated
with weakness, dizziness or light-headedness and 6.2 Irritable Bowel Syndrome
abdominal cramping [9, 12–16]. Some of these
cases may lead to emergency room visits and can 6.2.1 O
verview of Diets and Specific
be associated with sepsis or death. A meta- Foods
analysis (11 cross-sectional, 1 case-control and 1
cohort studies) found that diverticular disease can Food can be a major trigger of IBS symptoms
significantly increase the odds of developing and diet management has a potentially important
colonic adenomas by 68% with a trend for role in alleviating symptoms. However, there is
an incomplete understanding of how food affects wheat products [38, 39]. A meta-analysis found
IBS symptoms since there are few rigorous, that the pooled prevalence of IBS symptoms in
blinded RCTs [6–8, 32–36], so it is not uncom- patients with treated celiac disease on a gluten
mon for IBS individuals to generate their own free diet was up to 40%, in part because of the
theories to explain this phenomenon or seek fructans and galactans present in gluten free
guidance from other, usually unsupported, dietary diets [40]. Baked goods such as breads generally
remedies [32]. Certain foods and drinks that are: contain low to moderate in quantities of fructans
most commonly linked to IBS symptoms in some (0.61–1.94 g/100 g), with rye bread being the
people are: beans, cabbage, and other foods that richest source (1.94 g/100 g) [41]. Surprisingly,
may cause gas, such as foods containing oligo- gluten-free bread contains similar quantities of
saccharides or other highly fermentable fibers, fructan (0.36–1.79 g/100 g) to other breads.
foods high in fat, some higher lactose milk prod- Consequently, the widespread consumption of
ucts, or foods and beverages with large amounts bread products including gluten free products
of low calorie sweeteners such as sugar alcohols may make a significant contribution to fructan
[1]. Between 60 and 80% of patients with IBS intakes and contribute to IBS symptoms. Despite
report postprandial worsening symptoms and adhering to a gluten free diet, patients with celiac
adverse reactions to one or more foods, and many disease exhibit a 5-fold higher odds of IBS
patients avoid specific foods to reduce symptoms symptoms compared to healthy individuals as
[32–35]. These symptoms tend to occur or IBS may coexist with celiac disease in some
worsen within 3 h after meal consumption in patients [33, 38–40].
patients with IBS [33]. The relationship between
fiber intake and IBS is complex and dependent on
the subtype of IBS. In people with IBS- 6.2.3 Pathophysiology
constipation, fiber can improve constipation
symptoms and may help with reducing colonic IBS is characterized by increased susceptibility
pain as fiber softens stool so that it moves to bloating and bowel distension [42, 43]. In a
smoothly through the colon [1]. Fiber intake study (20 patients with IBS; 20 healthy volun-
should be slowly increased with recommended teers; 75% women) 90% of patients with IBS
levels of water to reduce the risk of increased gas developed colonic gas retention compared to
and bloating and medications should be con- only 20% of the control subjects (p < 0.01).
sumed at least 2 h after consuming fiber-rich The IBS patients had excessive gas retention
meals or supplements to avoid any potential drug and impaired gas clearance from the proximal
interactions. FODMAPs appear to increase colon, as opposed to the distal colon [43].
symptoms and low FODMAP diets may improve Increased susceptibility to gas production and
bloating and abdominal pain or discomfort but bloating occurs in nearly all patients with IBS,
they do not consistently improve bowel diarrhea especially after the consumption of ferment-
or constipation [6, 32, 37]. able carbohydrates [42–44]. Although IBS
colons generally lack v isible structural or ana-
tomic abnormalities, emerging research shows
6.2.2 Celiac Disease that there are colonic microscopic and molecu-
lar abnormalities from low-grade colonic
It is difficult to clinically distinguish IBS from inflammation and neuronal hyperexcitability,
adult-onset celiac disease [38–40]. Both IBS and and microbiota dysbiosis. Multifactorial low
celiac disease patients can have abdominal grade colonic inflammation is involved in the
symptoms triggered by the ingestion of wheat pathogenesis of IBS with studies showing
products. In celiac disease patients, this is due to colonic microscopic and molecular abnormali-
wheat gluten intolerance, while in IBS, the effect ties mainly characterized by an increased infil-
is attributed to fructans and galactans in the tration of mast cells [45–49]. Mast cells are
innate immune cells involved in food allergies, decreased butyrate producing bacteria such as
wound healing, and protection against patho- F. prausnitzii activates mucosal innate immune
gens. Increase in the numbers of colonic mast responses which increase colonic epithelial
cells in IBS patients has been related to permeability, activate nociceptive sensory
increased colonic permeability [45]. The diges- pathways and dysregulate the enteric neuromo-
tive tract contains an extensive enteric neuron torsensory function and brain-gut axis leading
network to control mucosal transport and motil- to IBS symptoms. It has been suggested that
ity and in response to persistent colonic inflam- dysbiotic bowel syndrome could be another
mation incoming mast cells communicate with name for IBS [57].
the central nervous system by release mediators
such as histamine or cytokines, which can
evoke neuronal hyperexcitability, a major fac- 6.2.4 Dietary Fiber
tor for IBS pain [46–49]. Their release of vari-
ous compounds, such as histamine, tryptase, Empirical thinking suggests that increased fiber
and chymase can evoke neuronal hyperexcit- may help to promote long-term alleviation of
ability, a major factor for IBS pain [46–49]. IBS symptoms because of fiber’s known ability
Abnormalities in the colonic enteric nervous to promote digestive health by: promoting regu-
system may alter digestion, gastrointestinal lar bowel movements; increasing stool bulk;
motility, and cause hypersensitivity which lowering colonic pH to protect against patho-
appear to have a pivotal role in the pathogene- gens; supporting healthier microbiota; and con-
sis of IBS in susceptible individuals [45]. Foods trolling colonic permeability and inflammation
containing FODMAPs increase the levels of [58, 59]. However, according to the American
inflammatory metabolome metabolites in the College of Gastroenterology monograph on IBS,
urine associated with the pathophysiology of the effectiveness of fiber-rich diets and supple-
IBS [50, 51]. A single blinded, parallel RCT ments in relieving IBS symptoms is inconsistent
(40 IBS patients, 83% IBS mixed or diarrhea, as insoluble fibers such as wheat bran provide
35 females, mean age 51 years; 3 weeks) found minimal relief, while some soluble fibers, espe-
that low FODMAP diets can reduce urinary cially psyllium, provide moderate relief to IBS
histamine 8-fold compared to high FODMAP symptoms [32]. A 2014 meta-analysis (14
diets [51]. RCTs; 940 subjects; 6 bran trials including 441
Emerging research supports the link subjects; 7 psyllium trials including 499 sub-
between colonic microbiota dysbiosis and the jects) found significant benefits for fiber in
development and prolongation of IBS symp- reducing the pooled mean IBS risk by 14% com-
toms [52]. One of the key features of IBS is the pared to a placebo, with no significant heteroge-
erratic pattern of stool form, with both hard neity between studies [60]. A stratified analysis
and loose stool within a time period as short as showed that bran had an insignificant effect on
24 h, suggesting that stool microbiota might the treatment of IBS by lowering risk by 10%,
also be unstable in IBS. It has been hypothe- whereas psyllium resulted in a significant 17%
sized that IBS may develop in predisposed reduction in IBS occurance. In a 2015 system-
individuals following an acute bout of infec- atic review and meta-analysis (22 RCTs, 1299
tious gastroenteritis, which has been linked to participants; 4–40 g fiber/day; 3–16 weeks), it
disturbance of the colonic microbiota with was shown that fiber, especially soluble fiber,
overgrowth of pathogens such as Escherichia appears to have a role in improving the symp-
coli, Salmonella, Shigella and Pseudomonas, toms of IBS with a low risk of harm [41]. There
2-fold increase in the ratio of Firmicutes to was a significant improvement in global assess-
Bacteroidetes and a marked reduction in diver- ment of symptoms among those randomized to
sity [52–56]. A dysbiotic colonic microbiota soluble fiber by 49% or any fiber by 27% (Fig.
including increased pathogenic bacteria and 6.2). Soluble fiber also reduced mean abdominal
1.6
Odds of Global Improvement in IBS Symptoms

1.5
1.4
1.3
1.2
1.1
1
All Fiber Soluble Fiber Insoluble Fiber
Fig. 6.2 Association between dietary fiber sub-type and management of irritable bowel syndrome (IBS) symptoms
(adapted from [61])
pain scores by 1.84 units, whereas insoluble symptoms compared to the placebo (Fig. 6.3)
fiber did not show improvement in any outcome. [62]. Other soluble fibers with potential IBS pro-
The analysis concludes that soluble fiber appears tective effects similar to psyllium include par-
to improve symptoms of IBS, whereas there is tially hydrolyzed guar gum [50–52] and pectin
no evidence for recommending insoluble fiber [53]. The RCTs summarized in Table 6.1 provide
for IBS. These meta-analyses identify soluble evidence that psyllium is moderately effective in
fiber, especially psyllium, if consumed with ade- alleviating IBS symptoms [62–69] whereas
quate water, has been shown in several RCTs to wheat bran fiber supplementation did not improve
help provide varying degrees of support for alle- IBS symptoms [62, 74–78]. Based on 2015 IBS
viating IBS symptoms and promoting regularity global perspective guidelines and other evidence
[60, 61]. some IBS relief may be provided by the gradual
Table 6.1 provides a comprehensive summary introduction of a low FODMAP fiber-rich diet or
of RCTs on the effects of fiber on IBS symptoms with a psyllium supplement because of its solu-
[62–78]. Overall, this analysis found that there is ble, stool bulking, low to moderate fermentabil-
limited evidence that certain fibers have modest ity, along with sufficient fluids intake [79]. These
effectiveness in alleviating IBS symptoms with- dietary options have some scientific support and
out a statistically significant increase in overall they are relatively inexpensive and safe, espe-
adverse events compared to placebo. In a double cially compared with the available drugs
blind RCT (275 patients in primary care; 164 approved for IBS [60, 69].
completers; mean age 34 years; 78% women;
10 g psyllium or 10 g ground wheat bran added to
yogurt and ingested twice daily vs. placebo 6.2.5 Low FODMAPS Diets
yogurt with rice flour; 3 months) found that psyl-
lium, but not wheat bran, added to yogurt was Consumption of moderate amounts of FODMAPs
effective in the clinical management of IBS by healthy individuals generally has very limited
Table 6.1 Summary of randomized controlled trials (RCTs) with specific fiber food ingredients or supplements studies
and irritable bowel syndrome (IBS) symptoms
Psyllium
Bijkerk et al. (2009) Double-blind, Parallel RCT: Psyllium significantly improved relief of
To evaluate the effects of 275 patients in primary care with 164 IBS symptoms compared to wheat bran or
psyllium supplement for completers; mean age 34 years; 78% placebo in primary care patients (Fig. 6.3).
the treatment of IBS female; 10 g psyllium added to yogurt After 3-months symptom severity was
(Netherlands) [62] twice daily vs. placebo yogurt with reduced with psyllium by 34% compared
rice flour; 12 weeks with 18% for the placebo (p = 0.03)
Jalihal and Kurian (1990) Double-blind, Parallel RCT: Psyllium significantly improved bowel
Assess the effects of 22 patients in secondary care with 9% satisfaction by reducing global symptoms
psyllium on IBS symptoms loss to follow-up; 20% female; 75% in the majority of IBS-diarrhea subjects
(India) [63] constipation; 30 g psyllium vs. placebo compared to the placebo but produced no
daily; 4 weeks change in abdominal pain or flatulence
Prior and Whorwell Double-blind, Parallel RCT: Psyllium significantly improved global
(1987) 80 patients in tertiary center with 29% IBS symptoms compared to the placebo
Evaluate the effects of loss to follow-up; 90% female; 49% (82% vs. 53% improvement). Also,
psyllium on managing IBS constipation; 3.6 g psyllium vs. psyllium significantly improved
symptoms (UK) [64] placebo 3× daily; 12 weeks constipation but not abdominal pain or
bloating significantly
Kumar et al. (1987) Dose Response, Crossover RCT: Psyllium significantly improved 3 major
Determine the optimal dose 14 female/19 male patients; psyllium IBS symptoms; constipation, abdominal
of psyllium for IBS 10, 20, and 30 g/day; 14-day study; pain and diarrhea. The 20 and 30 g doses
management (India) [65] 3 days of stool collection for each dose were more effective than a 10-g dose but
no washout period/randomized dosing compliance was reduced with the 30-g
of all 3 doses for 14 days with a dose. The optimum dose of psyllium in the
1-week washout treatment of IBS was 20 g/day
Nigam et al. (1984) Double blind, Parallel RCT: Psyllium significantly reduced risk of
Determine the effect of 42 patients in secondary care with no global IBS symptoms by 38%
psyllium on alleviating IBS loss to follow-up; 45% female;
symptoms (India) [66] psyllium vs. placebo; 12 weeks
Arthurs and Fielding Double-blind, Parallel RCT: Psyllium significantly reduced global IBS
(1983) 80 patients in secondary care with symptoms by 25%
Evaluate the effects of 2.5% loss to follow-up; 78% female; 2
psyllium on controlling IBS psyllium sachets vs. placebo; 4 weeks
symptoms (Ireland) [67]
Longstreth et al. (1981) Double blind RCT: Both psyllium and placebo significantly
Assess the effect of 77 patients in secondary care with 60 improved subjective global IBS symptoms
psyllium on IBS alleviation completers; 83% female; psyllium vs. by 70%. A strong placebo effect occurs in
(US) [68] placebo; 8 weeks patients with painful IBS
Ritchie and Truelove Double-blind, Parallel RCT: Psyllium significantly alleviated IBS
(1979) 100 patients in tertiary care with 4% symptoms by 42%
Determine the effectiveness loss to follow-up; 77 women and 33
of psyllium on treating IBS men; 2 sachets psyllium/day; 12 weeks
(UK) [69]
Partially Hydrolyzed Guar Gum (PHGG)
Niv et al. (2016) Double-blind, Parallel RCT: After 12 weeks, PHGG significantly
Study the effects of PHGG 121 IBS patients with 108 completers lowered bloating score by 2.9 units and
on symptoms of IBS 59% mixed, 25% diarrhea and 16% bloating and gas score by 3.2 units vs.
patients (Israel) [70] constipation IBS; 66% female; mean placebo. The effect lasted for at least
age 43 years; 6 g PHGG or placebo; 4 weeks after the last PHGG dose. PHGG
12 weeks; 4 weeks of follow-up had no effect on other IBS symptoms or
quality of life scores. There was a
significantly higher rate of dropouts in the
placebo compared with the PHGG group
(49% vs. 22%)
(continued)
Russo et al. (2015) Prospective open label trial: PHGG was significantly associated with
Investigate the effect of 86 constipated IBS subjects; mean age improved symptom scores, stool form/
PHGG on constipation 37 years; mean BMI 24; 82% female; consistency, colonic transit time, and
-predominant IBS (Italy) 2-week run-in baseline evaluation; reduced use of laxatives
[71] 4-week intervention with 5 g PHGG
consumed daily with water after
breakfast
Parisi et al. (2002) Multicenter Open RCT: 188 patients Per protocol analysis showed that both
Compare the effects of with 59% IBS-constipation; mean age wheat bran and PHGG were effective in
wheat bran and PHGG on 40 years; 74% female; wheat bran diet improving pain and bowel habits
IBS symptoms (Italy) [72] 30 g/day vs. PHGG 5 g/day in a (p > 0.05). Intention-to-treat analysis of
beverage; 12 weeks duration; after core IBS symptoms (abdominal pain and
4 weeks patients were allowed to bowel habits) showed a significantly
change groups depending on greater success in the PHGG group by
symptoms 60% than in the wheat bran fiber group by
40%. PHGG was more effective and better
tolerated than the wheat bran diet in
improving core IBS symptoms
Pectin
Xu et al. (2015) Parallel RCT: In IBS-diarrhea patients, pectin
Evaluate the efficacy of 87 patients with IBS-diarrhea; 24 g significantly reduced global symptom
pectin on diarrhea pectin/day vs. placebo; 6-week scores, Bristol stool scale scores, and
predominate IBS (China) intervention improved quality of life scores compared
[73] to placebo scores. The pectin acted as a
prebiotic and no significant adverse effects
were observed. Pectin appears to help
alleviate symptoms of IBS- diarrhea
Wheat Bran
Bijkerk et al. (2009) Double-blind, Parallel RCT: Wheat bran was less effective at relieving
Assess the effects of wheat 275 patients in primary care with 164 IBS symptoms than psyllium or placebo
bran vs. psyllium completers; mean age 34 years; 78% (Fig. 6.3). After 3 months, symptom
supplement for the female; 10 g ground wheat bran added severity was reduced for the wheat bran by
treatment of IBS to yogurt twice daily vs. psyllium and 22% (p = 0.61) compared with 18% for the
(Netherlands) [62] placebo yogurt with rice flour for placebo. Wheat bran had insignificant
12 weeks benefits in IBS patients in primary care
Rees et al. (2005) Single-blinded, Parallel RCT: Wheat bran modestly but significantly
Evaluate the effect of 28 patients from tertiary center with increased fecal wet weight by 28 g in 24 h
coarse wheat bran on IBS 21% lost to follow-up; 86% female; compared with the placebo group.
symptom management mean age 36 years; 100% constipation However, bran was ineffective in
(UK) [74] predominant; 10–20 g/day of coarse alleviating other bowel function measures
wheat bran supplement added to the and IBS symptoms
normal diet vs. a low fiber placebo;
8–12 weeks
Lucey et al. (1987) Double-blind, Crossover RCT: Wheat bran was ineffective in alleviating
Study the effects of wheat 44 patients from tertiary center with IBS symptoms compared to the placebo
bran on IBS symptoms 36% lost to follow-up; 79% female;
(UK) [75] mean age 32 years; wheat bran 15.6 g
fiber/day vs. placebo <0.5 g fiber/day
in biscuits; 12 weeks
Kruis et al. (1986) Parallel RCT: Wheat bran significantly improved IBS
Assess the effects of wheat 80 patients from tertiary center with symptoms vs. placebo after 12 weeks, but
bran on alleviating IBS 17.5% lost to follow-up; 62.5% not after 16 weeks. The long-term effect of
symptoms (German) [76] female; wheat bran 15 g fiber/day vs. wheat bran vs. placebo on IBS symptoms
placebo; 16 weeks was not confirmed
Manning et al. (1977) Parallel RCT: Wheat bran significantly improved IBS
Determine the effect of 26 patients from tertiary center with symptoms and resulted in an objective
wheat bran on IBS 8% lost to follow-up; 46% female; change in colonic motor activity vs. a
symptoms (UK) [77] 20 g wheat bran/day from bran and low-fiber diet. Patients with IBS may have
whole wheat bread vs. low fiber diet; beneficial effects on pain symptoms with
6 weeks increased daily intake of wheat bran
Soltoft et al. (1977) Double-blind, Parallel RCT: Miller’s wheat bran improved subjective
Evaluate the effect of 59 patients from tertiary center with reported IBS symptoms by 52% compared
Miller’s wheat bran on IBS 12% lost to follow-up; 64% women; with 65% improvement reported in the low
symptoms (Denmark) [78] bran 30 g/day in biscuits vs. low fiber wheat fiber control group. Miller’s bran
wheat biscuits; 6 weeks was less effective in reducing IBS
symptoms than a low fiber diet
Psyllium Wheat bran Placebo (rice flour)

60
Patients with Adequate Symptom Relief (%)
50
40
30
20
10
0
1 2 3 4 5 6 7 8 9 10 11 12
Study Duration (weeks)
Fig. 6.3 Effect of psyllium vs. wheat bran (10 g/day each) on irritable bowel syndrome (IBS) symptom relief (p <0.05)
(adapted from [62])
adverse effects, but for patients with IBS they of 6–8 weeks followed by the reintroduction of
often cause IBS symptoms because they are all individual FODMAPs to assess an individual’s
rapidly fermented, poorly absorbed, osmotically tolerance of each via a series of food challenges
active and rapidly increase gas production, with [80]. There are potential long-term challenges
additive effects contributing to IBS symptoms [6, associated with restricting the intake of FODMAPs
33]. Sources of low and high FODMAP foods are because it excludes a wide variety of foods from
listed in Table 6.2. The validation of the effective- the diet, which may potentially adversely affect
ness of a low FODMAP diet may involve the dietary nutrient quality and lead to colonic micro-
elimination of all known or suspected types of biota dysbiosis and other colonic related health
food with high content of FODMAPs for a period concerns and require a consult with a dietitian.
Table 6.2 Potential food sources of FODMAPs (fermentable oligosaccharides, disaccharides, monosaccharides and
polyols) [37, 41, 81]
Component High FODMAP food source Low-FODMAP food source
Fructose Fruit: apple, pear, peach, mango, watermelon Fruit: banana, blueberry, durian, grapefruit,
Other: honey or other sweeteners with fructose grape, kiwifruit, lemon, lime, mandarin,
orange, passion fruit, raspberry, and strawberry
Other: honey substitutes (maple syrup, golden
syrup)
Lactose Dairy: milk (cow, goat, sheep), ice cream, soft Dairy: lactose free milk, hard and camembert
cheeses, regular yogurt cheese, Greek yogurt, butter
Dairy substitutes: Ice cream
Substitutes: sorbet, rice and almond milk
Polyols (e.g., Vegetable: artichoke, asparagus, beetroot, Vegetable: bamboo shoots, bok choy, carrot,
sorbitol, Brussels sprout, broccoli, cabbage, cauliflower, celery, corn, eggplant, green beans, lettuce,
mannitol, fennel, garlic, leeks, okra, onion, peas, chives, parsnip, pumpkin, spring onion, tomato
maltitol, mushrooms, shallots Cereal: gluten-free and spelt bread/cereal
xylitol, Legume: chickpeas, lentils, red kidney beans, products
erythritol, baked beans
polydextrose, Fruit: watermelon, apple, pear, white peach,
and isomalt) persimmon, avocado.
Cereal: wheat and rye when eaten in large
amounts
Chewing gum/hard candies
Fructans and/ Fruit: nectarine, white peach, grapefruit, Fruit: banana, blueberry, durian, grapefruit,
or galactans watermelon, longon, persimmon, cantaloupe grape, honeydew melon, kiwifruit, lemon,
Vegetable: artichoke, leeks, onions, Brussel lime, mandarin, orange, passion fruit,
sprouts, garlic, beet root raspberry
Grain products: both gluten containing and Vegetable: most vegetables
gluten-free Sweeteners: sucrose, glucose
There are a number of RCTs and observa- were significantly reduced with a low FODMAPs
tional studies that generally support the benefits diet compared to a high FODMAPs diet (Fig.
of low FODMAP diets in managing IBS symp- 6.4) [51]. A 2015 multi-center, single blind, par-
toms but more rigorous trials are needed to estab- allel RCT (75 IBS patients, mean age 43 years;
lish long-term efficacy and safety. Several mean BMI 24; 82% female; 4 weeks) found that
systematic reviews of RCTs and observational low FODMAPs diets had similar effectiveness to
studies show that low FODMAP diets may be traditional IBS guidance, with regular meal pat-
effective in the management IBS symptoms, tern; avoidance of large meals; and reduced
especially with IBS-diarrhea [33, 81]. A 2015 intake of fat, insoluble fibers, caffeine, and gas-
systematic review of 6 RCTs on the effect of low producing foods, such as beans, cabbage, and
FODMAP vs. control diets showed significantly onions (Fig. 6.5) [83]. A New Zealand prospec-
reduced IBS symptom severity scores by 66%, tive study (192 IBS patients; average age
abdominal pain by 81%, bloating by 75%, over- 47 years; low FODMAP diet; 84% female; 47%
all symptoms by 81% and increased quality of completers; average follow- up 15.7 months)
life by 84% [82]. The effects of seven RCTs on found that IBS symptoms were significantly
the effect of low FODMAP diets and IBS symp- improved at follow-up with 72% of completers
toms are summarized in Table 6.3 [44, 51, 83– [88]. A 2017 meta-analysis (7 RCTs) found high
87]. A single blinded, parallel RCT (40 IBS grade evidence of an improved general symptom
patients, 83% mixed or diarrhea predominate, score in IBS patients who consumed a low
87% female, mean age 51 years; 3 weeks) FODMAP diet compared to those consuming a
showed that the IBS-symptom severity scores traditional IBS diet [89].
Table 6.3 Summary of low FODMAPs diet randomized controlled trials (RCTs) and the management of irritable
bowel syndrome (IBS) symptoms
McIntosh et al. (2016) Single-blinded, Parallel RCT: Low FODMAP diet significantly reduced
Evaluate effects of low 40 IBS patients, 83% mixed or diarrhea IBS-symptom severity scores vs. high FODMAP
and high FODMAP predominate, with 93% completers, 87% diet (Fig. 6.4). Low FODMAP diets significantly
diets on IBS symptoms female, mean age 51 years; low vs. high reduced urinary histamine levels by 8-fold but
[51] FODMAP diets; 3 weeks appeared to have potential adverse effects on the
microbiota ecosystem in the long-term
Böhn et al. (2015) Multi-center, Single-blind, Parallel About 50% of patients in both the low-
Compare the effects of RCT: FODMAP and traditional IBS diet groups had
a diet low in 75 IBS patients with 46% mixed/ reductions in IBS severity scores by ≥50
FODMAPs with unsubtyped IBS; 67 completers; mean compared with baseline (p = 0.72; Fig. 6.5).
traditional dietary age 43 years; mean BMI 24; 82% Food diaries demonstrated good adherence to
advice in patients with female; low FODMAP diet vs. both diets. Low FODMAP and traditional IBS
IBS (Sweden) [83] traditional IBS dietary guidance for dietary advice were equally effective in reducing
4 weeks IBS symptoms. Combining elements from these
2 diet strategies might further reduce symptoms
of IBS
Yoon et al. (2015) Multi-center, Double-blind, Parallel Diarrhea in patients receiving low-FODMAPs
Examine the dose RCT: was significantly improved compared with those
effects of FODMAP 100 IBS-diarrhea hospital patients; 84 receiving moderate- or high-FODMAPs. These
level on IBS symptoms completers; mean age 60 years; mean results support the hypothesis that a low-
(Korea) [84] BMI 20; 70% male; 3 enteral diets with FODMAP formula may reduce diarrhea leading
1 g (low), 2.2 g (moderate), and 3.7 g to an improvement in nutritional status and IBS
(high) of FODMAPs; 14 days recovery
Halmos et al. (2014) Single-blind, Crossover RCT: IBS subjects had significantly lower overall
Investigate the effects 45 out-patients with IBS with 30 gastrointestinal symptom scores while on a diet
on IBS symptoms of a completers; 43% IBS-constipation; low in FODMAPs, compared with the usual
diet low in FODMAPs secondary care; 70% female; median Australian diet. Bloating, pain, and gas also
compared with an age 28 years; mean BMI 24; low <0.5 were reduced while IBS patients were on the
Australian diet serving FODMAPs per meal diets vs. low-FODMAP diet. A diet low in FODMAPs
(Australian) [85] Australian diet (high in FODMAPs) and and supplemented with psyllium and resistant
supplemented with psyllium and starch were effective in helping to manage IBS
resistant starch; 3 weeks symptoms
Pedersen et al. (2014) Unblinded, Parallel RCT: The low FODMAPs diet significantly decreased
Investigate the effects 123 out-patients; secondary care; 108 the overall IBS severity scores vs. the Western
of low FODMAP diets completers; 85% IBS- diarrhea or diet. LGG probiotic significantly lowered IBS
and probiotics on IBS mixed; median age 37 years; 73% symptoms but to a lesser extent than the low
symptoms (Denmark) female; low FODMAPs diet, probiotic FODMAPs diet. Significant improvements were
[86] Lactobacillus rhamnosus GG (LGG) observed for the IBS-diarrhea and IBS-mixed
supplement, and Western diet; 6 weeks subtypes only.
Low FODMAPs diet and probiotic LGG are
effective in controlling IBS symptoms in the
IBS-diarrhea or mixed subtypes but not
IBS-constipation
Biesiekierski et al. Double-blind, Crossover RCT: IBS symptoms were significantly improved on
(2013) 40 IBS patients with non-celiac gluten low FODMAP diets and significantly worsened
Investigate effects of sensitivity; primary/secondary care to a similar degree on regular diets including
gluten and a FODMAP outpatient setting; 37 completers; 43% gluten or whey protein. Gluten-specific IBS
diet on IBS symptoms IBS-diarrhea, 35% IBS-constipation; effects were observed in 8% of participants.
(Australia) [87] median age 45 years; 84% female; There were no specific or dose-dependent effects
low FODMAP diets, high-gluten [16 g of gluten in patients with non-celiac gluten
gluten/day], low-gluten [2 g gluten/day], sensitivity placed on diets low in FODMAPs
or control [16 g whey protein/day] diets;
1-week trial; 2 weeks of washout
(continued)
Ong et al. (2010) Single-blind, Crossover RCT: Patients with IBS produced significantly more
Compare patterns of 15 IBS patients in secondary care, hydrogen gas than healthy controls while on the
breath hydrogen and median age 41 years, 87% female; 15 high FODMAP diet vs. the FODMAP-restricted
methane and healthy subjects, median age 23 years, diet. For IBS patients, all symptoms were
symptoms produced in 60% women; FODMAP-restricted diet significantly lower while on the FODMAP-
response to diets that (9 g/day) or a high FODMAP diet (50 g/ restricted diet, including abdominal pain,
differed only in day); 2-day trial; 7-day washout; diets bloating, passage of gas, nausea, heart burn and
FODMAP content were matched for total energy, starch, lethargy. The passage of gas was also
(Australia) [44] protein, fat and resistant starch and fiber; significantly lower in healthy subjects while they
all food was provided to the subjects were on the FODMAP-restricted diet.
FODMAP restricted diet reduced intestinal gas
and other symptoms in IBS patients compared to
healthy individuals
Abdominal pain intensity Abdominal pain freq. Abdominal distension

Bowel habit dissatisfaction General life interference Total IBS severity score
40
30
20
% Change from Baseline
10
-10
-20
-30
-40
-50
-60
Low FODMAPs Diet High FODMAPs Diet
Fig. 6.4 Effect of low vs. high FODMAP diets on a range of irritable bowel syndrome (IBS) symptoms (adapted
from [51])
6.3 Diverticular Disease 6.3.2.1 Nuts and Seeds

Previously there was the notion that undigested
6.3.1 Dietary Factors
particles from nuts, seeds, and popcorn might
lodge in portions of the diverticulum and
A list of specific dietary patterns and foods, and
hypothetically lead to diverticular disease
their association with diverticular disease risk are
complications, and patients were often advised in
summarized in Table 6.4 [21, 90, 92].
the past to avoid these foods [91, 92]. However,
the Health Professionals Follow-up Study
(47,228 men; mean age ranging from 51 to
6.3.2 Specific Foods and Beverages 60 years at baseline; mean BMI 25; 18-year
follow-up; 801 incident cases of diverticulitis)
Some foods and beverages historically have been observed that nuts, corn and popcorn consump-
linked to diverticular risk. tion did not significantly increase the risk of
6.3 Diverticular Disease 177
Low FODMAPs Diet (n=33) Traditional IBS Diet (n=34)

325
IBS Symptom Severity

300
275
250
225
0 14 29
Duration (days)
Fig. 6.5 Effect of low FODMAPs diet compared to traditional irritable bowel syndrome (IBS) dietary guidance* on
IBS patient symptom severity score (p = 0.72) (adapted from [83]). *Traditional IBS dietary guidance to avoid large
meals and reduce intake of fat, high lactose milk products, caffeine, and gas producing foods such as beans, cabbage
and onions
Table 6.4 Examples of dietary patterns and foods 6.3.2.2 Coffee

associated with symptomatic diverticular disease risk Coffee consumption has not been observed to
[21, 30, 31] have any effect on diverticular disease [93, 94].
Dietary patterns/specific foods Decrease Increased The Health Professionals Follow-up Study analy-
(high intake) risk risk sis (47,678 men; 40–75 years old; 4 years of fol-
Western dietary pattern √ low-up) observed no association between
Healthy dietary pattern (with √ caffeine, specific caffeinated beverages, and
adequate fiber)
decaffeinated coffee and the risk of symptomatic
Beef, pork or lamb (main dish) √
diverticular disease [94].
Processed meat √
Bacon √
Hot dogs √
6.3.2.3 Alcoholic Beverages
Green leafy vegetables √ Among people with diverticular disease, alcohol-
Whole peaches, apricots, or √ ics appear to have a 3 times greater risk of being
plums hospitalized for diverticulitis than occasional
Whole oranges √ drinkers [93]. The Health Professionals Follow-up
Whole apples √ Study analysis (47,678 men; 40–75 years old;
Blueberries √ 4 years of follow-up) observed after adjustments
Large cookie √ for age, physical activity, and energy intake of
Sugar sweetened beverages √ total dietary fiber and fat, that alcohol intake
French fried potatoes √ (comparing three standard drinks of alcohol/day
White bread, cookies, donuts √ to nondrinkers) was only weakly and insignifi-
and so on cantly associated with risk of symptomatic diver-
ticular disease [94]. When the prevalence of
diverticulitis (Fig. 6.6) [92]. This suggests that diverticulosis from 18 countries was analyzed
the recommendation to avoid these foods to pre- against alcohol use, there was a strong correlation
vent diverticular complications should be recon- between diverticulosis risk and national per-cap-
sidered as these foods may actually lower risk of ita alcohol consumption rates (r = 0.68; p = 0.002)
diverticular disease. [95]. The European Prospective Investigation into
Nuts (p-trend =.04) Corn (p-trend =.44) Popcorn (p-trend =.007)
1.2
Relative Risk of Diverticulitis 1.1
0.9
0.8
0.7
0.6
< 1/month 1-3/month 1/week >=2/week
Frequency of Consumption (servings)
Fig. 6.6 Effect of nuts, corn and popcorn intake frequency on diverticulitis risk (adapted from [92])
Cancer and Nutrition (EPIC)-Oxford study, a supportive evidence, on the basis of low risk and
cohort of mainly health conscious participants theoretical benefit, a high fiber diet and/or fiber
recruited from around the UK (47,033 men and supplementation should be considered in asymp-
women living in England or Scotland; 5459 (33%) tomatic diverticulosis to reduce the likelihood of
vegetarians; mean 11.6 years of follow-up; 812 disease progression and in symptomatic divertic-
cases of diverticular disease) found that there was ulosis to reduce symptom episodes and prevent
no significant association between the consump- acute diverticulitis” [99]. FODMAPs highly fer-
tion of alcohol and risk of diverticular disease mentable fiber sources should be limited because
after adjusting for smoking [96]. of their risk of increasing colonic pressure due to
flatulence and osmotic load, which may lead to
colonic defects or diverticular disease symptoms
6.3.3 Dietary Fiber or complications [20]. Fifty of the highest whole
(minimally processed) plant food sources of
The National Institute of Diabetes and Digestive dietary fiber are summarized in Appendix A.
and Kidney Diseases (NIDDK) publication on
diverticular disease indicates that the symptoms 6.3.3.1 Fiber-Rich Dietary Patterns
of diverticular disease may be treated with an
appropriate combination of high-fiber diet or Observational Studies
fiber supplements, medications, and possibly Eight observational studies on the effect of
probiotics [97]. NIDDK suggests a (1) slow pro- healthy fiber-rich and lower red or processed
gressive increase in fiber-rich foods to minimize meat diets or vegetarian diets provide insights on
gas and abdominal discomfort, (2) fiber supple- the risk of diverticulosis and management diver-
ments methylcellulose or psyllium 1–3 times a ticular disease and are summarized in Table 6.5
day, along with the consumption of adequate [21, 96, 100–105].
water. A 2002 review concluded that diets high in Diverticulosis. Three observational studies
fiber and low in total fat and red meat and a life- assess the effects of fiber-rich dietary patterns on
style with more physical activity might help pre- diverticulosis risk [100–102]. A UK case-control
vent diverticular disease [98]. A 2011 review study (56 vegetarians for >10 years and 264
concluded that “despite the lack of high-quality non-vegetarians; age >45 years) found that
Table 6.5 Summary of fiber-rich or healthy dietary pattern observational studies on diverticulosis and diverticular
disease risk
Diverticulosis
Case-controlled study
Gear et al. (1979) 56 vegetarians (members of Vegetarians had a significantly higher mean
Determine the effect of fiber in the UK Vegetarian Society fiber intake (41.5 g/day) than non-vegetarians
vegetarian and non-vegetarian diets for ≥10 years; (21.4 g/day). Diverticulosis was significantly
on diverticulosis prevalence (UK) age > 45 years; 60% higher in non-vegetarians (33%) than in
[100] female); 264 non- vegetarians (12%). Low intake of cereal fiber
vegetarians (≥ 45 years; was associated with the presence of
55% female); barium diverticulosis, especially for women
enema; food frequency
questionnaire
Cross-sectional studies
Peery et al. (2013) 539 individuals with colonic No association was observed between total
Examine the link between low fiber diverticula; mean age of fiber intake and diverticulosis in comparing the
intake and the risk of asymptomatic 60 years; 1569 controls highest quartile to the lowest (mean intake 25
diverticulosis (US) [101] without diverticula; mean versus 8 g/day)
age 57 years; 60% males;
mean BMI 29; mean total
fiber intake 15 g/day
Peery et al. (2012) 878 cases of diverticulosis; Higher fiber diets were not protective against
Study the association between high mean age 59 years; 1226 asymptomatic diverticulosis
fiber intake and the risk of controls without diverticula;
asymptomatic diverticulosis (US) mean age 54 years; mean
[102] total fiber intake 19 g/day
Diverticulitis/diverticular disease
Prospective studies
Strate et al. (2017) 46,295 men; mean baseline After adjustment for other risk factors, men in
Examine the effect of major dietary age 53 years; 26 years of highest quintile of Western dietary pattern
patterns on risk of diverticulitis follow-up; 1063 incident scores had an increased multivariate risk of
(Health Professionals Follow-up cases of diverticulitis diverticulitis by 55% vs. men in the lowest
Study; US) [21] quintile. In contrast, men with higher prudent/
healthy scores were associated with decreased
risk of diverticulitis by 26–33% (Fig. 6.7). The
association between dietary patterns and
diverticulitis was predominantly attributable to
intake of fiber (lowering risk) and red meat
(raising risk)
Crowe et al. (2014) 690,075 women; mean age Fiber significantly reduced risk of diverticular
Characterize the effect of different 60 years; 17,325 were disease related hospital admissions or death
fiber sources on diverticular disease admitted to hospital or died with total, cereal and fruit fiber having the
risk (The Million Women Study, with diverticular disease; strongest effects (Fig. 6.8)
UK) [103] stable diet for last 5 years;
mean total fiber intake 14 g/
day; 6 years of follow-up
Crowe et al. (2011) 47,033 health conscious Vegetarian diets and a high intake of fiber were
Assess the effect of vegetarian diets adults; 1/3 reported both associated with a reduction in diverticular
and fiber intake on risk of consuming a vegetarian disease related hospital admission or death risk.
diverticular disease risk (European diet; 76% female; median There was a significant inverse association with
Prospective Investigation into BMI 23; mean follow-up total fiber intake and diverticular disease risk
Cancer and Nutrition [EPIC]- time of 11.6 years; 812 (≥26 g/day vs. <14 g/day), after multivariate
Oxford, UK) [96] cases of diverticular disease adjustments (Figs. 6.9 and 6.10)
(continued)
Aldoori et al. (1998) 43,881 male health Insoluble fiber was significantly associated
Evaluate specific fiber types and professionals; 40–75 years with a decreased risk of diverticular disease by
diverticular disease risk in men of age at baseline; 362 cases 37%, and this inverse association was
(Health Professionals Follow-up of symptomatic diverticular particularly strong for cellulose which reduced
Study; US) [104] disease; 4 years of risk by 48%
follow-up
Aldoori et al. (1994) 47,888 male health Total fiber intake was inversely associated with
Examine the association between professionals; 40–75 years the risk of diverticular disease, after adjustment
fiber and sources of fiber with the of age; 385 cases of for age, energy-adjusted total fat intake, and
diagnosis of symptomatic symptomatic diverticular physical activity, with a significant 42% lower
diverticular disease in men (Health disease; 4 years of risk at the extremes of fiber intake; fruit and
Professionals Follow-up Study; US) follow-up vegetable fiber were the most effective fiber
[105] sources. A high-red-meat, low-fiber diet
increased risk over 2-fold compared with those
on a low-red-meat, high-fiber diet
diverticulosis was significantly higher in non-veg- red meat consumption. The UK Million Women
etarians (33%) than in vegetarians (12%); vegetar- Study (690,075 women; mean age 60 years;
ians had a significantly higher mean fiber intake 17,325 were admitted to the hospital or died with
(41.5 g/day) than non-vegetarians (21.4 g/day) diverticular disease; 6 years of follow- up)
[100]. Two US cross-sectional studies (mean age observed that fiber significantly reduced diver-
of the subjects was in the mid-50s; mean fiber ticular disease risk with cereal and fruit fiber hav-
intake of 15–19 g/day) observed no association ing the strongest effects (Fig. 6.8) [103]. An
between total fiber intake and diverticulosis inci- EPIC UK/Oxford cohort (47,033 adults; 76%
dence [101, 102]. Scientifically, the association female; median BMI 23; 33% vegetarians; mean
between diet and diverticulosis is difficult to prove 11.6 years of follow-up; 812 diverticular disease
because of the long latency of diverticula forma- hospital admissions or deaths) showed that con-
tion, the often lack of symptoms of diverticulosis, suming a vegetarian diet significantly lowered
and the challenges of obtaining accurate dietary the multivariate adjusted risk of diverticular dis-
fiber intake. ease by 31% vs. meat eaters (Fig. 6.9) [96]. There
Diverticular Disease (Diverticulitis). Five was also an inverse association between fiber
prospective studies examine the effects of dietary intake and diverticular disease with a significant
patterns higher in fiber and lower in red or pro- 41% lower risk (≥26 g/day vs. <14 g/day) (Fig.
cessed meat consumption or vegetarian diets on 6.10) [96]. Two US Health Professionals
diverticular disease risk [21, 96, 103–105]. A Follow-up studies from the 1990s also showed
2017 Health Professionals Follow-up Study similar inverse relationships between fiber intake
(46,295 men; mean baseline age 53 years; and diverticular disease risk [104, 105].
26 years of follow-up; 1063 incident cases of
diverticulitis) found that after adjustment for Intervention Trials
other risk factors, men with highest prudent/ Table 6.6 summarizes 11 intervention trials (6
healthy scores were associated with decreased RCTs and 5 open label trials) on the effects of a
risk of diverticulitis by 26–33% whereas men variety of fiber-rich dietary patterns and foods/
with the highest Western dietary pattern scores supplement sources with adjunctive antibiotic
had an increased multivariate risk of diverticulitis usage on diverticular disease symptoms [106–
by 55% (Fig. 6.7) [21]. The association between 116]. All 6 RCTs showed beneficial effects on
dietary patterns and diverticulitis was predomi- symptoms and/or bowel function with a number
nantly attributable to low intake of fiber and high of different fiber sources including fiber-rich
Western Diet (p-trend <.0001) Prudent Diet (p-trend =.004)

Healthy Eating Index (p-trend <.0001)
1.8
1.6
Hazard Ratio for Diverticulitis
1.4
1.2
0.8
0.6
0.4
1 2 3 4 5
Dietary Pattern Score (Quintiles)
Fig. 6.7 Dietary pattern quality score and risk of diverticulitis in men (adapted from [21])
25
20
Risk of Hospital Admission or Death from Diverticular Disease (%)
15
10
0
Total Fiber Cereal Fiber Fruit Fiber Vegetable fiber Potato Fiber
(p <.0001) (p <.0001) (p <.0001) (p =.634) (p <.0001)
-5
-10
-15
-20
-25
Fig. 6.8 Association between fiber intake and risk of diverticular disease related hospital admission or death risk per
5 g fiber/day intake from the UK Million Women Study (mean baseline age 60 years; 6 years of follow-up) (adapted
from [103])
Relative Risk of Diverticular Disease

0.9
0.8
0.7
0.6
0.5
0.4
Meat Meat Meat Vegetarian or
>= 100 g/day 50-99 g/day < 50 g/day vegan
Fig. 6.9 Effect of non-vegetarian vs. vegetarian diets on uncomplicated diverticular disease risk (adapted from [96])
1.1
1
Relative Risk of Diverticular Disease
0.9
0.8
0.7
0.6
0.5
0.4
<14 17.5 21 26 >26
Total Fiber Intake (g/day)
Fig. 6.10 Effect of total fiber intake on uncomplicated diverticular disease risk (p < 0.001) (adapted from [96])
diets, bran, bran crisps, psyllium, methylcellu- and wheat bran on alleviating symptoms [112–
lose [106–111]. Three RCTs found that high fiber 116]. However, presently three systematic
diets can improve symptoms and/or bowel func- reviews conclude that quality evidence on the
tion [106, 107, 111]. Three RCTs suggest that efficacy of fiber treatment for the reduction of
wheat bran, psyllium or methylcellulose supple- symptoms associated with uncomplicated diver-
mented diets can improve symptoms and bowel ticular disease and for the prevention of acute
function [108–110]. Five open-label trials all diverticulitis, is limited because of the low num-
support the beneficial effects of fiber-rich diets ber of high quality RCTs [117–119].
Table 6.6 Summary of fiber-rich diet and fiber intervention trials in uncomplicated diverticular disease symptoms
Lahner et al. (2012) Multicenter, Parallel RCT: A high-fiber diet was effective in relieving
Evaluate the effects of adding a 45 patients; mean age 66 years; 66% abdominal symptoms. The combination of
symbiotic supplement to a high female; base diet ≥30 g daily fiber high-fiber diet and symbiotic can relieve
fiber diet in the treatment of plus 7 g Flortec© symbiotic abdominal bloating as well as abdominal
symptomatic diverticular disease formulation containing 5 × 109 CFU pain
(Italy) [106] viable L. paracasei B12060 plus a
mixture of xylo-oligosaccharides
(700 mg) and arabinogalactone
(1243 mg) or no supplement; ≥
1.5 L of water/daily; 6 months
Smits et al. (1990) Parallel RCT: Bowel frequency and stool consistency
Compare the efficacy and 43 patients; high fiber diet (30–40 g improved similarly with both treatments.
tolerance of lactulose and a fiber daily) vs. lactulose (30 mL Pain on bowel movement and abdominal
high-fiber diet in the treatment of daily); 12 weeks pain improved with both treatments in
symptomatic diverticular disease respect to frequency and severity
(UK) [107]
Ornstein et al. (1981) Double-blind, Crossover RCT: The bran crisp bread and psyllium drink
Compare the effects of bran and 58 patients; median age 64 years; significantly improved symptoms of
psyllium on symptomatic 62% female; bran crisp bread (7 g constipation when compared to the initial
diverticular disease (UK) [108] fiber), psyllium beverage (9 g fiber)score. No significant differences in pain,
and placebo (2.3 g fiber) added to a lower bowel symptoms and total symptom
daily habitual 15 g fiber diet; scores were reported since there was only
16 weeks a 5–7 g difference between the test fibers
and placebo
Brodribb (1977) Double-blind RCT: Daily wheat bran crisp bread significantly
Evaluate the effects of wheat 18 patients; 6.7 g fiber wheat bran decreased mean overall symptom scores
bran on symptomatic diverticular crisp bread daily vs. 0.6 g fiber vs. placebo. Although wheat bran crisp
disease [109] placebo crisp bread; 3 months bread significantly lowered overall pain
score, there were no significant differences
in bowel function scores. No adverse
effects were recorded
Hodgson (1977) Double-blind RCT: Patients in the methylcellulose group had
Assess the effect of methyl 30 patients; 2 tablets methylcellulose significantly greater symptom decrease
cellulose on symptomatic vs. 2 tablets placebo; 3 months than those in the placebo group
diverticular disease (UK) [110]
Taylor and Duthie (1976) Crossover RCT: Bran proved to be the most effective
Determine the effect of bran 20 patients; high-fiber diet, treatment, not only in improving the
tablets on symptomatic Normacol plus, and bran tablets; symptoms but also in returning to normal
diverticular disease [111] 1 month the abnormal pathophysiological changes.
Bran tablets were both convenient and
acceptable as well as effective
Open Label Trials
Leahy et al. (1985) 31 patients on high fiber diets; 25 High fiber diets significantly reduced
Compare the effects of higher vs. patients; typical Western fiber diets; symptoms recurrence (19% vs. 44%),
lower fiber diets on diverticular average follow-up 57 months complications (6% vs. 20%) and required
disease symptoms (UK) [112] less surgery (6% vs. 32%) compared to
the low fiber control group
(continued)
Hyland and Taylor (1980) 100 patients; 75% consumed high Of the patients consuming high fiber diets,
Evaluate the long-term effects of fiber diets; 5–7 years 90% remained symptom-free
a high fiber diet on symptomatic
diverticular disease (UK) [113]
Eastwood et al. (1978) 31 patients; mean age 60 years; 20 g All supplements equally alleviated
Assess the effects of different coarse wheat bran, 2 sachets of symptoms
types of fiber on symptomatic psyllium, and 20–40 mL/day
diverticular disease (UK) [114] lactulose; 4-week duration
Brodribb and Humphreys 40 patients; 24 g wheat bran daily; Wheat bran decreased all symptoms by
(1976) 6 months 60%, accelerated transit times in patients
Examine the effects of wheat with >60 h, reduced intracolonic pressure.
bran on symptomatic diverticular Barium enema studies showed less spasm
disease (UK) [115]. in 8 patients and no diverticula in 3
patients after taking bran
Painter et al. (1972) 70 patients; 86% completers; 89% of patients had relieved symptoms
Evaluate the effect of high fiber high-fiber, low-sugar diet including and none of the completers required
on symptomatic diverticular unprocessed bran; average surgery
disease (UK) [116] 22 months
6.3.3.2 Fiber Mechanisms weights and structure, compared to low fiber

Fiber is known for its effects on promoting controls [125]. Wheat bran was more effective at
colonic health and weight control, which may reducing transit time and psyllium was more
contribute to reducing diverticular disease risk effective at increasing stool water content (softer
through a number of biological mechanisms [58, stools) and weight. A systematic review of 65
120–125]. intervention studies found that wheat bran,
improves bowel function by significantly increas-
Colonic Health ing total wet stool weight by 3.7 g/gram intact
Fiber may improve colonic health to reduce wheat fiber and reduces transit time by 45 min/g
diverticular disease risk by two primary mecha- when baseline transit time is greater than 48 h
nisms: (1) promoting stool bulk and regular laxa- [126]. Furthermore, alterations in colonic micro-
tion and (2) maintaining a healthy colonic biota composition related to low fiber Western
microbiota ecosystem [126–132]. Consistent diets can have an adverse effect on colonic health
with the original fiber hypothesis on diverticular especially with aging, leading to increased inci-
disease, fiber promotes laxation by increasing dence of colonic dysbiosis whereas the consump-
fecal bulk and stool frequency, and reducing tion of adequate fiber can lower the colonic
intestinal transit time by increasing fecal water- lumen pH, increase the balance of healthy metab-
holding capacity, and improves the quantity and olites and inhibit the growth of pathogenic bacte-
quality of microbiota for overall colonic health. ria and colonic inflammation [127–132].
Fiber sources that combine low fermentability Bacterial fermentation of fiber to maintain an
and high water binding capacity from fiber-rich adequate colonic butyrate concentration is criti-
diets containing a variety of whole or minimally cal for maintaining distal colonic health. Butyrate
processed plant foods or from wheat bran, psyl- exerts potent effects on a variety of colonic
lium fiber, and methylcellulose containing foods mucosal functions such as inhibition of inflam-
or supplements are particularly effective in pro- mation by reinforcing various components of the
moting laxation [58, 120–128]. In a 1988 RCT colonic defense barrier and the inhibition of
both wheat bran and psyllium husk fiber were nuclear factor kappa B (NF-κβ) activation and
shown to decrease transit time and increase daily histone deacetylation, and the activation of
stool regularity as well as promote healthier stool G-coupled receptors.
ody Weight Regulation

B Conclusions
Obesity increases the risk of diverticular disease Irritable bowel syndrome (IBS) is the most
and its complications [133–136]. An increased common gastrointestinal disorder occurring
BMI has been linked to an elevated risk for symp- in people <45 years. Diverticular disease is
tomatic diverticular disease and its complications among the most clinically and economically
with central obesity as an independent risk factor significant gastroenterological conditions in
due to the release of proinflammatory cytokines people ≥65 years of age. Having a history of
from the visceral fat [133]. The Health IBS appears to increase the risk of diverticu-
Professionals Follow-up Study (47,228 men; lar disease in older age. IBS, previously
baseline age range 40–75; 18 years of follow-up; called colitis, does not generally show visible
801 cases of diverticulitis) found that men with a structural or anatomic abnormalities, but is
BMI > 30 had increased risk of diverticulitis by characterized by abdominal pain, bloating,
78% and >300% higher risk for diverticular hem- distension, and changes in bowel habits.
orrhage compared to men with a normal BMI Celiac disease may be confounding and dif-
[135]. A Swedish prospective cohort (36,592 ficult to distinguish from IBS symptoms.
women; 12 years of follow-up) showed that Diverticular disease may evolve from colonic
women with an overweight or obese BMI had a diverticulae (herniate pouches) potentially
higher risk of diverticular disease by 30% and a caused by high colonic intraluminal pressure
200% higher risk of a colonic abscess or perfora- which occurs in most people with aging but
tion compared to women with a normal BMI only approximately 20% of individuals with
(20–24.9) [136]. Populations with fiber-rich diets diverticulae develop abdominal symptoms
tend to be leaner than those with low fiber diets (symptomatic uncomplicated diverticular
[137–143]. In the Nurses’ Health Study, women disease). A smaller percentage of older indi-
in the highest quintile of fiber intake had a sig- viduals eventually develop complications
nificant 49% lower risk of major weight gain than such as severe bouts of diverticulitis or bleed-
women in the lowest quintile. Weight gain was ing that may lead to sepsis and death. Healthy
inversely associated with the intake of high-fiber dietary patterns and low intake of FODMAPs
and whole-grain foods [139]. A systematic may help to lower the risk and alleviate
review of 43 prospective cohort, case-control and symptoms associated with IBS and diverticu-
randomized trials found probable evidence that lar disease. For IBS, psyllium is the fiber
increased fiber intake was predictive of less source most consistently found to help pro-
weight gain and higher intake of refined grains, vide moderate relief of symptoms. For
sweets and desserts, and high energy diets were uncomplicated diverticular disease, fiber-rich
predictive of elevated weight gain and waist size healthy diets and low red or processed meat
[140]. A Finnish Trial of overweight middle-aged consumption decrease the risk, and fiber-rich
men and women showed that lower dietary fat diets, and foods or supplements containing
and higher fiber intake are significant predictors wheat bran, psyllium or methylcellulose may
of sustained weight reduction, even after help to alleviate diverticular disease symp-
adjustment for other risk factors [141]. A long- toms and/or improve bowel function. Fiber
term RCT suggests that consuming >30 g fiber/ related mechanisms that may help reduce
day can effectively promote weight loss similarly risk or manage symptoms of IBS or uncom-
to that of reduced energy diet regimens [142]. plicated diverticular disease are related to:
After weight loss is achieved, healthy fiber rich (1) improved colonic health by promoting
dietary patterns, can slow weight regain to main- better laxation and stool bulk, and a healthier
tain a 4–10 kg weight loss after 1 year and 3–4 kg microbiota ecosystem with higher fecal ratio
after 2 years [143]. of probiotic to pathogenic bacteria, and
higher butyrate concentration associated weight and central abdominal fat gains (or
with lower colonic inflammation and promotion of a gradual lowering of body
improved colonocyte structure and function, weight and waist size in overweight or obese
and (2) reduced risk or rate of annual body individuals).
ppendix A: Fifty high fiber whole or minimally processed plant foods ranked
A
by amount of fiber per standard food portion size
Dietary Calories Energy density
Food Standard portion size fiber (g) (kcal) (calories/g)
Shredded wheat ready-to-eat cereal 1–1 1/4 cup (50–60 g) 5.0–9.0 155–220 3.2–3.7
References 187

Orange 1 medium (130 g) 3.1 69 0.5
Banana 1 medium (118 g) 3.1 105 0.9
Dates 1/4 cup (about 38 g) 2.9 104 2.8
Nutr Diet. 2015;115:1861–70
trends. 2015;97, 98; Table D1.8
Slavin JL. Position of the American Dietetic Association: health implications of dietary fiber. J Am Diet Assoc.
2008;108:1716–31
2010. 7th ed. Washington, DC: U.S. Government Printing Office. 2010; Table B2.4; http://www.choosemyplate.gov/.
Accessed 22 Aug 2015
References 6. Staudacher HM, Irving PM, Lomer MCE, Whelan

K. Mechanisms and efficacy of dietary FODMAP
restriction in IBS. Nat Rev Gastroenterol Hepatol.
1. National Institute of Diabetes and Digestive and
2014;1(4):256–66.
Kidney Diseases. Definition and facts for irritable
bowel syndrome. 2015. http://www.niddk.nih.gov/ 7. El-Salhy M, Ostgaardi H, Gundersen D. The role
health-information/health-topics/digestive-diseases/ of diet in the pathogenesis and management of
irritable-bowel-syndrome/Pages/definition-facts. irritable bowel syndrome (review). Int J Mol Med.
aspx www.digestive.niddk.nih.gov. Accessed 6 Jul 2012;29:723–31.
2016. 8. Barbara G, Cremon C, Carini G, et al. The
2. Wilkins T, Pepitone C, Alex B, Schade RR. immune system in irritable bowel syndrome. J
Diagnosis and management of IBS in adults. Am Neurogastroenterol Motil. 2011;17:349–59.
Fam Physician. 2012;86(5):419–26. 9. Tursi A. Diverticulosis today: unfashionable and
3. International Foundation for Functional still under-researched. Ther Adv Gastroenterol.
Gastrointestinal Disorders. Statistics. 2016. 2016;9(2):213–28.
http://aboutibs.org/facts-about-ibs/statistics.html. 10. Templeton AW, Strate LL. Updates in diverticular
Accessed 6 Jul 2016. disease. Curr Gastroenterol Rep. 2013;15(8):339–46.
4. Lovell RM, Ford AC. Global prevalence of and risk 11. Jacobs DO. Diverticulitis. N Engl J Med.
factors for irritable bowel syndrome: a meta-analysis. 2007;357:2057–66.
Clin Gastroenterol Hepatol. 2012;10:712–21. 12. Tursi A, Papagrigoriadis S. Review article: the cur-
5. Lovell RM, Ford AC. Effect of gender on preva- rent and evolving treatment of colonic diverticular
lence of irritable bowel syndrome in the c ommunity: disease. Aliment Pharmacol Ther. 2009;30:532–46.
systematic review and meta-analysis. Am J 13. Tursi A. Diverticular disease: a therapeutic overview.
Gastroenterol. 2012;107:991–1000. World J Gastrointest Pharmacol Ther. 2010;1(1):27–35.
14. Tursi A, Papa A, Danese S. Review article: the 29. Krokowicz L, Stojcev Z, Kaczmarek FB, et al.
pathophysiology and medical management of Microencapsulated sodium butyrate administered to
diverticulosis and diverticular disease of the colon. patients with diverticulosis decreases incidence of
Aliment Pharmacol Ther. 2015;42:664–84. diverticulitis—a prospective randomized study. Int J
15. Muhammad A, Lamendola O, Daas A, et al. Color Dis. 2014;29:387–93.
Association between colonic diverticulosis and 30. Daniels L, Budding AE, de Korte K, et al. Fecal
prevalence of colorectal polyps. Int J Color Dis. microbiome analysis as a diagnostic test for
2014;29(8):947–51. diverticulitis. Eur J Clin Microbiol Infect Dis.
16. Cuomo R, Barbara G, Andreozzi P, et al. Symptom 2014;33(11):1927–36.
patterns can distinguish diverticular disease from 31. Jung H-K, Choung RS, Locke GR, et al.
irritable bowel syndrome. Eur J Clin Investig. Diarrhea- p redominant irritable bowel syn-
2013;43(11):1147–55. drome is associated with diverticular disease:
17. Jaruvongvanich V, Sanguankeo A, Wijarnpreecha a population-based study. Am J Gastroenterol.
K, Upala S. Risk of colorectal adenomas, advanced 2010;105(3):652–61.
adenomas and cancer in patients with colonic 32. Ford AC, Moayyedi P, Lacy BE, et al. American
diverticular disease: Systematic review and meta- College of Gastroenterology Monograph on the
analysis. Dig Endosc. 2017;29:71–82. https://doi. management of irritable bowel syndrome and
org/10.1111/den12701. chronic idiopathic constipation. Am J Gastroenterol.
18. Painter NS, Burkitt DP. Diverticular disease of the 2014;109:S2–S26.
colon: a deficiency disease of western civilization. 33. Rao SSC, Yu S, Fedewa A. Systematic review:
BMJ. 1971;2:450–4. dietary fibre and FODMAP-restricted diet in the
19. Reinhard T. Diverticular disease—a reexamina- management of constipation and irritable bowel
tion of the fiber hypothesis. Today’s Dietitian. syndrome. Aliment Pharmacol Ther. 2015;41(12):
2014;16(3):46. 1256–70.
20. Uno Y, Velkinburgh JC. Logical hypothesis: low 34. Böhn L, Störsrud S, Törnblom H, et al. Self-reported
FODMAP diet to prevent diverticulitis. World J food-related gastrointestinal symptoms in IBS are
Gastrointest Pharmacol Ther. 2016;7(4):503–12. common and associated with more severe symptoms
21. Strate LL, Keeley BR, Cao Y, et al. Western and reduced quality of life. Am J Gastroenterol.
dietary pattern increases and prudent dietary pat- 2013;108:634–41.
tern decreases, risk of incident diverticulitis in 35. Simrén M, Månsson A, Langkilde AM, et al. Food-
a prospective cohort study. Gastroenterology. related gastrointestinal symptoms in the irritable
2017;152(5):1023–30. https://doi.org/10.1053/ bowel syndrome. Digestion. 2001;63:108–15.
jgastro.2016.12.038. 36. Monsbakken KW, Vandvik PO, Farup PG. Perceived
22. Commane DM, Arasaradnam RP, Mills S, et al. food intolerance in subjects with irritable bowel
Diet, ageing and genetic factors in the pathogen- syndrome- etiology, prevalence and consequences.
esis of diverticular disease. World J Gastroenterol. Eur J Clin Nutr. 2006;60:667–72.
2009;15(20):2479–88. 37. Cuomo R, Andreozzi P, Zito FP, et al. Irritable
23. Wilkins T, Embey K, George R. Diagnosis and man- bowel syndrome and food interaction. World J
agement of acute diverticulitis. Am Fam Physician. Gastroenterol. 2014;20(27):8837–45.
2013;87(9):612–20. 38. De Giorgio R, Volta U, Gibson PR. Sensitivity to
24. Wick JY. Diverticular disease: eat your fiber? wheat, gluten and FODMAPs in IBS: facts or fic-
Consult Pharm. 2012;27(9):613–8. tion? Gut. 2016;65:169–78.
25. Aldoori W, Ryan-Harshman M. Preventing divertic- 39. El-Salhy M, Hatlebakk JG, Gilja OH, Hausken
ular disease. Review of recent evidence on high-fibre T. The relation between celiac disease, non-celiac
diets. Can Fam Physician. 2002;48:1632–7. gluten sensitivity and irritable bowel syndrome.
26. Sheth AA, Longo W, Floch MH. Diverticular Nutr J. 2015;14:92. https://doi.org/10.1186/
disease and diverticulitis. Am J Gastroenterol. s12937-015-0080-6.
2008;103:1550–6. 40. Sainsbury A, Sanders DS, Ford AC. Prevalence
27. Tursi A, Elisei W, Picchio M, et al. Moderate to of irritable bowel syndrome–type symptoms in
severe and prolonged left lower-abdominal pain patients with celiac disease: a meta-analysis. Clin
is the best symptom characterizing symptomatic Gastroenterol Hepatol. 2013;11:359–65.
uncomplicated diverticular disease of the colon: a 41. Whelan K, Abrahmsohn O, David GJP, et al.
comparison with fecal calprotectin in clinical set- Fructan content of commonly consumed wheat,
ting. J Clin Gastroenterol. 2015;49(3):218–21. rye and gluten-free breads. Int J Food Sci Nutr.
28. Jeyarajah S, Akbar N, Moorhead J, et al. A clini- 2011;62(5):498–503.
copathological study of serotonin of the sigmoid 42. Lacy BE, Gabbard SL, Crowell MD. Pathophysiology,
colon mucosa in association with chronic symptoms evaluation, and treatment of bloating: hope, hype,
in uncomplicated diverticulosis. Int J Color Dis. or hot air? Gastroenterol Hepatol. 2011;7(11):
2012;27(12):1597–605. 729–39.
References 189
43. Serra J, Azpiroz F, Malagelada JR. Impaired transit High Impact Case Rep. 2016;4(2):2324709616648458.
and tolerance of intestinal gas in the irritable bowel https://doi.org/10.1177/2324709616648458.
syndrome. Gut. 2001;48:14–9. 58. Dahl WJ, Stewart ML. Position of the Academy
44. Ong DK, Mitchell SB, Barrett JS, et al. Manipulation of Nutrition and Dietetics: health implications of
of dietary short chain carbohydrates alters the pat- dietary fiber. J Acad Nutr Diet. 2015;115:1861–70.
tern of gas production and genesis of symptoms in 59. EFSA Panel on Dietetic Products, Nutrition, and
irritable bowel syndrome. J Gastroenterol Hepatol. Allergies. Scientific opinion on dietary reference
2010;25:1366–73. values for carbohydrates and dietary fibre. EFSA J.
45. Sinagra E, Pompei G, Tomasello G, et al. 2010;8(3):1462.
Inflammation in irritable bowel syndrome: myth 60. Moayyedi P, Quigley EMM, Lacy BE, et al. The
or new treatment target? World J Gastroenterol. effect of fiber supplementation on irritable bowel
2016;22(7):2242–55. syndrome: a systematic review and meta-analysis.
46. Zhang L, Song J, Hou X. Mast cell and irritable Am J Gastroenterol. 2014;109:1367–74.
bowel syndrome: from the bench to the bedside. 61. Nagarajana N, Mordena A, Bischof D, et al. The
J Neurogastroenterol Motil. 2016;22(2):181–92. role of fiber supplementation in the treatment of
https://doi.org/10.5056/jnm15137. irritable bowel syndrome: a systematic review
47. Reed DE, Barajas-Lopez C, Cottrell G, et al. Mast and meta-analysis. Eur J Gastroenterol Hepatol.
cell tryptase and proteinase-activated receptor 2 2015;27(9):1002–10. https://doi.org/10.1097/
induce hyperexcitability of guinea-pig submucosal MEG.0000000000000425.
neurons. J Physiol. 2003;547(2):531–42. 62. Bijkerk CJ, de Wit NJ, Muris JW, et al. Soluble or
48. Barbara G, Stanghellini V, De Giorgio R, et al. insoluble fibre in irritable bowel syndrome in pri-
Activated mast cells in proximity to colonic mary care? Randomised placebo controlled trial.
nerves correlate with abdominal pain in irritable Br Med J. 2009;339:b3154. https://doi.org/10.1136/
bowel syndrome. Gastroenterology. 2004;126: bmj.b3154.
693–702. 63. Jalihal A, Kurian G. Ispaghula therapy in irritable
49. Guilarte M, Santos J, de Torres I, et al. Diarrhoea- bowel syndrome: improvement in overall well-being
predominant IBS patients show mast cell activa- is related to reduction in bowel dissatisfaction. J
tion and hyperplasia in the jejunum. Gut. 2007;56: Gastroenterol Hepatol. 1990;5:507–13.
203–39. 64. Prior A, Whorwell PJ. Double blind study of ispa-
50. Scalbert A, Brennan L, Manach C, et al. The food ghula in irritable bowel syndrome. Gut. 1987;28:
metabolome: a window over dietary exposure. Am J 1510–3.
Clin Nutr. 2014;99:1286–308. 65. Kumar A, Kumar N, Vij JC, et al. Optimum dosage of
51. McIntosh K, Reed DE, Schneider T, et al. ispaghula husk in patients with irritable bowel syn-
FODMAPs alter symptoms and the metabolome drome: correlation of symptom relief with whole gut
of patients with IBS: a randomised controlled trial. transit time and stool weight. Gut. 1987;28:150–5.
Gut. 2016;66(7):1241–51. https://doi.org/10.1136/ 66. Nigam P, Kapoor KK, Rastog CK, et al. Different
gutjnl-2015-311339. therapeutic regimens in irritable bowel syndrome. J
52. Distrutti E, Monaldi L, Ricci P, Fiorucci S. Gut Assoc Physicians India. 1984;32:1041–4.
microbiota role in irritable bowel syndrome: new 67. Arthurs Y, Fielding JF. Double blind trial of ispa-
therapeutic strategies. World J Gastroenterol. ghula/poloxamer in the irritable bowel syndrome. Ir
2016;22(7):2219–41. Med J. 1983;76:253.
53. Simrén M, Barbara G, Flint HJ, et al. Intestinal 68. Longstreth GF, Fox DD, Youkeles L, et al. Psyllium
microbiota in functional bowel disorders: a Rome therapy in the irritable bowel syndrome. Ann Intern
foundation report. Gut. 2013;62(1):159–76. Med. 1981;95:53–6.
54. Jalanka-Tuovinen J, Salojarvi J, Salonen A, 69. Ritchie JA, Truelove SC. Treatment of irritable
et al. Faecal microbiota composition and host- bowel syndrome with lorazepam, hyoscine butylbro-
microbe cross-talk following gastroenteritis and mide, and ispaghula husk. Br Med J. 1979;1:376–8.
in post-infectious irritable bowel syndrome. Gut. 70. Niv E, Halak A, Tiommny E, et al. Randomized clin-
2014;63(11):1737–45. ical study: partially hydrolyzed guar gum (PHGG)
55. Ponnusamy K, Choi JN, Kim J, et al. Microbial versus placebo in the treatment of patients with
community and metabolomic comparison of irri- irritable bowel syndrome. Nutr Metab. 2016;13:10.
table bowel syndrome faeces. J Med Microbiol. https://doi.org/10.1186/s12986-016-0070-5.
2011;60(6):817–27. 71. Russo L, Andreozzi P, Zito FP, et al. Partially hydro-
56. Bonfrate L, Tack J, Grattagliano I, et al. Microbiota lyzed guar gum in the treatment of irritable bowel
in health and irritable bowel syndrome: current syndrome with constipation: effects of gender,
knowledge, perspectives and therapeutic options. age and body mass index. Saudi J Gastroenterol.
Scand J Gastroenterol. 2013;48:995–1009. 2015;21(2):104–10.
57. Benno P, Dahlgren A-L, Befrits R, et al. From IBS to 72. Parisi GC, Zilli M, Miani MP, et al. High-fiber diet
DBS: the dysbiotic bowel syndrome. J Investig Med supplementation in patients with irritable bowel
syndrome (IBS) a multicenter, randomized, open 87. Biesiekierski JR, Peters SL, Newnham ED, et al. No
trial comparison between wheat bran diet and par- effects of gluten in patients with self-reported non-
tially hydrolyzed guar gum (PHGG). Dig Dis Sci. celiac gluten sensitivity after dietary reduction of
2002;47(8):1697–704. fermentable, poorly absorbed, short chain carbohy-
73. Xu L, Yu W, Jiang J, et al. Efficacy of pectin in the drates. Gastroenterology. 2013;145:320–8.
treatment of diarrhea predominant irritable bowel 88. de Roest RH, Dobbs BR, Chapman BA, et al. The
syndrome. Zhonghua Wei Chang Wai Ke Za Zhi. low FODMAP diet improves gastrointestinal symp-
2015;18(3):267–71. toms in patients with irritable bowel syndrome:
74. Rees G, Davies J, Thompson R, et al. Randomised- a prospective study. Int J Clin Pract. 2013;67:
controlled trial of a fibre supplement on the symp- 895–903.
toms of irritable bowel syndrome. J R Soc Prom 89. Varju P, Farkas N, Hegyi P et al. Low fermentable
Health. 2005;125:30–4. oligosaccharides, disaccharides, monosaccharides
75. Lucey MR, Clark ML, Lowndes JO, et al. Is bran and polyols (FODMAP) diet improves symptoms in
efficacious in irritable bowel syndrome? A double- adults suffering from irritable bowel syndrome (IBS)
blind placebo controlled crossover study. Gut. compared to standard IBS diet: a meta-analysis of
1987;28:221–5. clinical studies. PLOS ONE. 2017;12(8):e0182942.
76. Kruis W, Weinzierl P, Schussler P, et al. Comparison 90. Strate LL. Lifestyle factors and the course of diver-
of the therapeutic effects of wheat bran and pla- ticular disease. Dig Dis. 2012;30:35–45.
cebo in patients with the irritable bowel syndrome. 91. Horner JL. Natural history of diverticulosis of the
Digestion. 1986;34:196–201. colon. Am J Dig Dis. 1958;3(5):343–50.
77. Manning AP, Heaton KW, Harvey RF. Wheat fibre and 92. Strate LL, Liu YL, Syngal S, et al. Nut, corn and
irritable bowel syndrome. Lancet. 1977;2(8035):417–8. popcorn consumption and the incidence of diverticu-
78. Soltoft J, Gudmand-Hoyer E, Krag B, et al. A lar disease. JAMA. 2008;300(8):907–14.
double-blind trial of the effect of wheat bran on 93. Böhm SK. Risk factors for diverticulosis, diverticu-
symptoms of irritable bowel syndrome. Lancet. litis, diverticular perforation, and bleeding: a plea
1977;8034:270–2. for more subtle history taking. Viszeralmedizin.
79. Quigley EMM, Fried M, Gwee K-A, et al. Irritable 2015;31:84–94.
bowel syndrome: a global perspective. World 94. Aldoori WH, Giovannucci EL, Rimm EB, et al. A
Gastroenterol Org Guidelines 2015;19–20. prospective study of alcohol, smoking, caffeine, and
80. Gibson PR, Shepherd SJ. Evidence-based dietary the risk of symptomatic diverticular disease in men.
management of functional gastrointestinal symp- Ann Epidemiol. 1995;5:221–8.
toms: The FODMAP approach. J Gastroenterol 95. Sharara AI, El-Halabi MM, Mansour NM, et al.
Hepatol. 2010;25:252–8. Alcohol consumption is a risk factor for colonic
81. Muir JG, Shepherd SJ, Rosella O, et al. Fructan and diverticulosis. J Clin Gastroenterol. 2013;47(5):
free fructose content of common Australian vegeta- 420–5.
bles and fruit. J Agric Food Chem. 2007;55:6619–27. 96. Crowe FL, Appleby PN, Allen NE, Key TJ. Diet
82. Marsh A, Eslick EM, Eslick GD. Does a diet low and risk of diverticular disease in Oxford cohort of
in FODMAPs reduce symptoms associated with European Prospective Investigation into Cancer and
functional gastrointestinal disorders? A compre- Nutrition (EPIC): prospective study of British vege-
hensive systematic review and meta- analysis. Eur J tarians and non-vegetarians. BMJ. 2011;343:d4131.
Nutr. 2015;55(3):897–906. https://doi.org/10.1007/ https://doi.org/10.1136/bmj. d4131.
s00394-015-0922-1. 97. The National Institute of Diabetes and Digestive and
83. Böhn L, Störsrud S, Liljebo T, et al. Diet low in Kidney Diseases (NIDDK). Diverticular disease.
FODMAPs reduces symptoms of irritable bowel 2013. NIH Publication No. 13–1163.
syndrome as well as traditional dietary advice: 98. Aldoori W, Ryan-Harshman M. Preventing divertic-
a randomized controlled trial. Gastroenterology. ular disease. Review of recent evidence on high-fibre
2015;149:1399–407. diets. Can Fam Physician. 2002;48:1632–5.
84. Yoon SR, Lee JH, Lee JH, et al. Low-FODMAP 99. Tarleton S, DiBaise JK. Low-residue diet in diver-
formula improves diarrhea and nutritional status ticular disease: putting an end to a myth. Nutr Clin
in hospitalized patients receiving enteral nutrition: Pract. 2011;26:137–42.
a randomized, multicenter, double-blind clinical 100. Gear JS, Ware A, Fursdon P, et al. Symptomless
trial. Nutr J. 2015;14:116. https://doi.org/10.1186/ diverticular disease and intake of dietary fibre.
s12937-015-0106-0. Lancet. 1979;1(8115):511–4.
85. Halmos EP, Power VA, Shepherd SJ, et al. A diet low 101. Peery AF, Sandler RS, Ahnen DJ, et al.
in FODMAPs reduces symptoms of irritable bowel Constipation and a low-fiber diet are not associ-
syndrome. Gastroenterology. 2014;146:67–75. ated with diverticulosis. Clin Gastroenterol Hepatol.
86. Pedersen N, Andersen NN, Vegh Z, et al. Ehealth: 2013;11(12):1622–7.
low FODMAP diet vs lactobacillus rhamnosus GG 102. Peery AF, Barrett PR, Park D, et al. A high-fiber diet
in irritable bowel syndrome. World J Gastroenterol. does not protect against asymptomatic diverticulo-
2014;20:16215–26. sis. Gastroenterology. 2012;142(2):266–72.
References 191
103. Crowe FL, Balkwill A, Cairns BJ, et al. Source of Products, Nutrition, and Allergies (NDA), Parma,
dietary fibre and diverticular disease incidence: a pro- Italy. EFSA J. 2010;8(3):1462.
spective study of UK women. Gut. 2014;63:1450–6. 121. Slavin JL. Position of the American Dietetic
104. Aldoori WH, Giovannucci EL, Rockett HRH, Association: health implications of dietary fiber. J
et al. Prospective study of dietary fiber types and Am Diet Assoc. 2008;108:1716–31.
symptomatic diverticular disease in men. J Nutr. 122. Slavin J. Fiber and prebiotics: mechanisms and
1998;128:714–9. health benefits. Forum Nutr. 2013;5:1417–35.
105. Aldoori WH, Giovannucci EL, Rimm EB, et al. A 123. Eswaran S, Muir J, Chey WD. Fiber and functional
prospective study of diet and the risk of symptomatic gastrointestinal disorders. Am J Gastroenterol.
diverticular disease in men. Am J Clin Nutr. 2013;108(5):718–27.
1994;60:757–64. 124. Cummings JH. The effect of dietary fiber on fecal
106. Lahner E, Esposito G, Zullo A, et al. High fibre weight and composition. In: Spiller GA, editor. CRC
diet and Lactobacillus paracasei B21060 in symp- handbook of dietary fiber in human nutrition. 3rd ed.
tomatic uncomplicated diverticular disease. World J Boca Raton: CRC; 2001. p. 183–241.
Gastroenterol. 2012;18(41):5918–24. 125. Stevens J, VanSoest PJ, Robertson JB, Levitsky
107. Smits BJ, Whitehead AM, Prescott P. Lactulose in DA. Comparisons of the effects of psyllium
the treatment of symptomatic diverticular disease: and wheat bran on gastrointestinal transit time
a comparative study with high-fibre diet. Br J Clin and stool characteristics. J Am Diet Assoc.
Pract. 1990;44:314–8. 1988;88(3):323–6.
108. Ornstein MH, Littlewood ER, Baird IM, et al. Are 126. de Vries J, Miller PE, Verbeke K. Effects of cereal
fibre supplements really necessary in diverticular fiber on bowel function: a systematic review
disease of the colon? A controlled clinical trial. of intervention trials. World J Gastroenterol.
BMJ. 1981;282:1353–6. 2015;21(29):8952–63.
109. Brodribb AJ. Treatment of symptomatic diverticular 127. Koh A, De Vadder F, Kovatcheva-Datchary P,
disease with a high fibre diet. Lancet. 1977;1:664–6. Backhed F. From dietary fiber to host physiology:
110. Hodgson WJ. The placebo effect. It is impor- short-chain fatty acids as key bacterial metabolites.
tant in diverticular disease? Am J Gastroenterol. Cell. 2016;165:1332–45.
1977;67:157–62. 128. Pituch-Zdanowska A, Banaszkiewicz A, Albrecht
111. Taylor I, Duthie HL. Bran tablets and diverticular P. The role of dietary fibre in inflammatory bowel
disease. Br Med J. 1976;1(6016):988–90. disease. Prz Gastroenterol. 2015;10(3):135–41.
112. Leahy AL, Ellis RM, Quill DS, Peel ALG. High 129. Conlon MA, Bird AR. The impact of diet and life-
fibre diet in symptomatic diverticular disease of the style on gut microbiota and human health. Forum
colon. Ann R Coll Surg Engl. 1985;67(3):173–4. Nutr. 2015;7:17–44.
113. Hyland JM, Taylor I. Does a high fibre diet prevent 130. Lynch DB, Jeffery IB, Cusack S, et al. Diet micro-
the complications of diverticular disease? Br J Surg. biota health interactions in older subjects: implica-
1980;67:77–9. tions for healthy aging. Interdiscip Top Gerontol.
114. Eastwood MA, Smith AN, Brydon WG, et al. 2015;40:141–54.
Comparison of bran, ispaghula, and lactulose on colon 131. Claesson MJ, Jeffery IB, Conde S. Gut microbiota
function in diverticular disease. Gut. 1978;19:1144–7. composition correlates with diet and health in the
115. Brodribb AJ, Humphreys DM. Diverticular dis- elderly. Nature. 2012;488:178–85.
ease: three studies. III. Metabolic effect of bran 132. Hamer HM, Jonkers D, Venema K, et al. Review
in patients, with diverticular disease. Br Med J. article: the role of butyrate on colonic function.
1976;1(6007):428–30. Aliment Pharmacol Ther. 2008;27:104–19.
116. Painter NS, Almeida AZ, Colebourne KW. 133. Kopylov U, Ben-Horin S, Lahat A, et al. Obesity,
Unprocessed bran in treatment of diverticular dis- metabolic syndrome and the risk of development of
ease of the colon. Br Med J. 1972;2(5806):137–40. colonic diverticulosis. Digestion. 2012;86:201–5.
117. Carabotti M, Annibale B, Severi C, Lahner E. Role of 134. Nagata N, Sakamoto K, Arai T, et al. Visceral fat
fiber in symptomatic uncomplicated diverticular dis- accumulation affects risk of colonic diverticular
ease: a systematic review. Forum Nutr. 2017;9:161. hemorrhage. Int J Color Dis. 2015;30(10):1399–406.
https://doi.org/10.3390/nu9020161. 135. Strate LL, Liu YL, Aldoori WH, et al. Obesity
118. Unlu C, Daniles L, Vrouenraets BC, et al. A system- increases the risks of diverticulitis and diverticular
atic review of high-fibre dietary therapy in diverticu- bleeding. Gastroenterology. 2009;136:115–22.
lar disease. Int J Colorectal Dis. 2012;27:419–27. 136. Hjern F, Wolk A, Hakansson N. Obesity, physical
119. Maconi G, Barbara G, Bosetti C, et al. Treatment of inactivity, and colonic diverticular disease r equiring
diverticular disease of the colon and prevention of hospitalization in women: a prospective cohort
acute diverticulitis: a systematic review. Dis Colon study. Am J Gastroenterol. 2012;107:296–302.
Rectum. 2011;54(10):1326–38. 137. Slavin JL. Dietary fiber and body weight. Nutrition.
120. European Food Safety Authority (EFSA). Scientific 2005;21:411–8.
opinion on dietary reference values for carbohy- 138. Davis JN, Hodges VA, Gillham B. Normal-weight
drates and dietary fibre. EFSA Panel on Dietetic adults consume more fiber and fruit than their
age- and height-matched overweight/obese counter- 141. Lindstrom J, Peltonen M, Eriksson JG, et al. High-
parts. J Am Diet Assoc. 2006;106:833–40. fibre, low-fat diet predicts long-term weight loss and
139. Liu S, Willett WC, Manson JE, et al. Relation decreased type 2 diabetes risk: the Finnish Diabetes
between changes in intakes of dietary fiber and grain Prevention Study. Diabetologia. 2006;49:912–20.
products and changes in weight and development of 142. Ma Y, Olendzki BC, Wang J, et al. Single-component
obesity among middle-aged women. Am J Clin Nutr. versus multi-component dietary goals for the meta-
2003;78:920–7. bolic syndrome: a randomized trial. Ann Intern Med.
140. Fogelholm M, Anderssen S, Gunnarsdottir I, Lahti- 2015;162:248–57.
Koski M. Dietary macronutrients and food consump- 143. Dietary Guidelines Advisory Committee (DGAC).
tion as determinants of long-term weight change in Scientific Report. Advisory report to the secretary of
adult populations: a systematic literature review. health and human services and the secretary of agri-
Food Nutr Res. 2012:56. https://doi.org/10.3402/fnr. culture. Part D. Chapter 2: Dietary patterns, foods
v56i0.19103. and nutrients and health outcomes. 2015;1–35.
Part III
Weight Management and Related Diseases
Dietary Patterns and Fiber in Body
Weight and Composition 7
Regulation
Keywords
Dietary patterns • Mediterranean diet • DASH diet • New Nordic diet
Vegetarian diet • Western diet • Dietary fiber • Weight loss • Energy
density • Obesity • Overweight • Body weight • Waist circumference Body
mass index • Visceral fat
Key Points Western diet pattern; (2) lowering available

metabolizable energy; and (3) increasing post-
• The human gastrointestinal and energy metab- prandial satiety by affecting both the upper
olism regulatory systems evolved with pre- digestive tract and colonic microbiota.
agricultural high fiber diets (>50 g fiber/day). • Fiber intake is inversely associated with obe-
• Prospective cohort studies and randomized sity risk and populations with higher fiber
controlled trials (RCTs) show that high adher- diets tend to be leaner than those with low
ence to healthy fiber-rich dietary patterns fiber diets.
such as the Mediterranean (MedDiet), Dietary • Prospective cohort studies suggest that
Approaches to Stop Hypertension (DASH), increased total fiber intake by ≥12 g/day to a
New Nordic, and vegetarian diets may at a total daily fiber intake of >25 g, especially
minimum help to prevent weight gain and can compared to refined low fiber diets, can pre-
support weight loss and lower waist circum- vent weight gain by 3.5–5.5 kg each decade.
ference compared to low-fat or Western diets • RCTs show that adequate fiber intake ≥28 g
in overweight or obese individuals. fiber/day from fiber-rich diets can reduce
• Mechanisms associated with healthy fiber- body weight and waist circumference com-
rich dietary pattern effects on managing body pared to low fiber Western diets (≤20 g fiber/
weight and central obesity include: (1) reduc- day). Fiber-rich diets are generally more
ing dietary energy density directly or displac- effective at promoting weight loss than fiber
ing higher energy foods associated with the supplements.

196 7 Dietary Patterns and Fiber in Body Weight and Composition Regulation
7.1 Introduction dressings, and sugars and syrups were positively

associated with obesity. A meta-analysis (117
The worldwide overweight and obesity pan- trials; 4815 participants) found that hypocaloric
demic is among the greatest public health chal- diets were more effective for weight loss and
lenges of our time [1, 2]. Since approximately increased exercise was more effective at reduc-
the 1980s, the global prevalence of overweight ing visceral adiposity, reducing visceral fat by a
and obesity increased by more than 28% for mean of 6% versus only 1% for hypocaloric
adults and 47% for children, resulting in an diets [16]. Elevated BMI or excessive adiposity
increase from about 850 million overweight or in adulthood and increasingly in childhood is a
obese people in 1980 to over two billion in 2013 growing risk factor for major chronic diseases
[1]. Obesity is a complex multifactorial condi- such as diabetes, cardiovascular disease, non-
tion resulting from the chronic disruption alcoholic fatty liver disease, chronic kidney dis-
between energy intake and energy expenditure, ease, and a number of obesity related cancers
involving genetic, environmental, lifestyle, and [17–19].
emotional factors [3–5]. During the last several For overweight or obese individuals who
decades there has been an increased exposure to successfully lose weight, as many as 80% typi-
higher energy dense and lower dietary fiber cally drift back to their original weight or more
(fiber) containing foods and increasingly seden- [20]. This is because after weight loss there are
tary lifestyles, which have led to net habitual an array of metabolic regulatory processes at
positive energy balances and weight gain in work to promote weight regain, so it is difficult
western populations [1, 4–14]. A small positive to maintain weight loss [21–24]. After body fat
energy balance of 50 kcals/day, by increased loss, thermogenesis reduces and leads to fat
energy intake and/or reduced activity, can lead loss resistance and drops in hormone levels,
to an annual weight gain of 0.4–0.9 kg/year [5– such as leptin and thyroid hormones, which can
7]. Further, a higher habitual intake of 200 kcal/ cause the risk of increased energy intake above
day above energy balance in overweight or energy expenditure [22, 23]. Weight loss trig-
obese women may increase weight gain by as gers strong overeating signals sent to the brain’s
much as 9 kg/year [8]. Energy dense diets, com- hypothalamus to increase appetite. Also, in this
mon in the Western-style diet, are positively period, adipocytes face cellular stress associ-
associated with higher body mass index (BMI) ated with the physical forces that arise within
and risk of obesity [8–14]. Moreover, since peo- the shrinking cells, causing them to actively
ple tend to eat approximately the same amount promote renewed fat storage. The determinants
or volume of food on a day-to-day basis, regard- of weight maintenance are genetics, behavior,
less of the food energy density, the common and environment with diet behavior thought to
advice of just eating less of all foods may not be be the most important factor that influences
the optimal approach for weight management weight regain. A cross- sectional study of
[7]. The 2017 nationally representative Canadian weight loss maintainers who lost >10% of their
Community Health Survey (11,748 adults; 18 body weight and maintained that loss for
years of age or older) found that individuals ≥5 years reported that they consumed a diet
consuming diets higher in energy density (1.2 with lower energy density (1.4 kcal/g) than the
vs. 0.7 kcal/g) and lower in dietay fiber (5.9 vs. weight re-gain individuals (1.8 kcal/g) [21].
12.2 g/1,000 kcals) significantly elevated their These weight loss maintainers consumed more
risk for obesity by 257% [15]. This study also fiber-rich foods such as vegetables (4.9
found that whole fruits, dark green, orange, and servings/day) and whole-grain products (2.2
non-starchy vegetables and 100% juices and servings/day) compared to less than 1 daily
yogurt were inversely associated with obesity serving of vegetables and whole grains for the
whereas fast foods, carbonated beverages, weight regainers [21]. In addition to eating a
refined grains, processed meats, sauces and low energy dense and high fiber diet, successful
7.2 Healthy Dietary Patterns 197
long-term weight loss maintenance is associ- 7.2 Healthy Dietary Patterns

ated with five additional strategies to help
counteract weight regain metabolic processes: 7.2.1 Overview
(1) engaging in physical activity; (2) eating
breakfast; (3) self-monitoring weight on a regu- Compared with the usual Western diet, the con-
lar basis; (4) limiting consumption of higher sumption of healthy dietary patterns, including
energy dense foods; and (5) catching dietary the US dietary guidelines diet, Mediterranean
“miss-steps” before they turn into a habit [20]. diet (MedDiet), Dietary Approaches to Stop
Two common dietary approaches for weight Hypertension (DASH) diet, and healthy vegetar-
loss and reducing the risk of central obesity ian (lacto-ovo) diets, by overweight and obese
include: (1) reducing daily energy intake by individuals can result in weight loss or at least
20–35% for a negative energy balance or (2) prevent weight gain depending on the degree of
eating lower energy dense and healthy fiber- adherence and fiber level [14]. Appendix A sum-
rich dietary patterns vs. Western dietary pattern marizes the food and nutrient composition of
[22–25]. A 2017 meta-analysis (12 observa- some major healthy dietary patterns vs. the
tional studies) found that highest vs. lowest American Western dietary pattern. The 2015 US
adherence to healthy dietary patterns reduced Advisory Guidelines Advisory Committee scien-
central obesity risk by 19% whereas a Western tific report concluded that there was strong evi-
dietary pattern increased risk by 16% [24]. In a dence showing that overweight and obese adults,
randomized controlled trial (RCT) with obese preferably as part of a comprehensive lifestyle
adults with metabolic syndrome, it was shown change intervention, can achieve clinically mean-
that those who simply consumed a high fiber ingful weight loss ranging from 4 to 12 kg after
dietary pattern had similar weight loss to those 6-months through a variety of healthy dietary
on a more complex multi-component, hypoca- patterns that achieve an energy deficit [14].
loric diet plan after one year (Fig. 7.1) [25]. Thereafter, slow weight regain is observed, with
The objective of this chapter is to comprehen- total weight loss at 1 year of 4–10 kg and at
sively review the effects of healthy dietary pat- 2 years of 3–4 kg. All these healthy dietary pat-
terns and fiber on body weight and composition terns double the fiber content from about 16 g/
regulation. day in the usual Western diet to >30 g/day and
Weight (kg) BMI (kg/m2) Waist (cm)

0.5
0.0
High Fiber Diet Only AHA Multicomponent Weight
-0.5 Loss Program
Change over 1 year
-1.0
-1.5
-2.0
-2.5
-3.0
Fig. 7.1 Effect of a high fiber based weight loss diet (≥30 g fiber/day) vs. the American Heart Association (AHA)
multi-component weight loss program in 240 metabolic syndrome adults after 1 year (p > 0.05) (adapted from [25])
decrease added sugars intake by more than half regulation from observational studies are sum-
by emphasizing the increased consumption of marized studies in Table 7.1 [26–46].
plant-based foods, such as whole grains, fruits,
vegetables, pulses, and nuts. 7.2.2.1 Overall Diet Quality
Observational studies consistently show that
habitual intake of higher overall diet quality, espe-
7.2.2 Observational Studies cially with adequate fiber intake from whole plant
foods and lower intake of red meat and high
The effect of dietary quality and specific healthy energy dense processed foods and beverages, is
dietary patterns on body weight and composition inversely associated with weight gain and central
Table 7.1 Summary of diet quality score and specific healthy dietary pattern observational studies on body weight and
composition regulation
Overall dietary pattern quality
Systematic review
Asghari et al. (2017). 34 studies (10 prospective studies and Diet quality indices based on dietary
Review observational studies, 24 cross-sectional studies) guidelines (Healthy Eating Index)
focusing on the association of were inversely associated with
diet quality indices with general parameters of weight status in most
obesity or abdominal obesity in studies. Scoring on the basis of
adults [26] dietary diversity was directly
associated with weight gain
Observational studies
Shah et al. (2016). 5079 subjects; mean age 61 years; Those with a higher diet-quality score
Investigate the relationship 47% males (multivariate adjusted) in AHA goals and the MedDiet were
between dietary quality and generally older and female, with a
regional adiposity in a cross- lower BMI, CRP, and markers of
sectional analysis (Multi-Ethnic insulin resistance. After adjustment, a
Study of Atherosclerosis higher diet-quality score (highest vs.
[MESA]; US) [27] lowest dietary score quartile) was
associated with lower visceral fat:
461 cm2/m vs. 524 cm2/m (p-trend
<0.01; Fig. 7.2), less pericardial fat
(41 vs. 48 cm3/m (p-trend <0.01), and
lower hepatic steatosis (by hepatic
attenuation; 59 vs. 61 Hounsfield
units (p-trend <0.01). Greater intake
of fiber containing fruits, vegetables,
whole-grains and seeds/nuts, and
yogurt was associated with decreased
adiposity, while red/processed meats
were associated with greater
adiposity
Hu et al. (2016). 2656 adolescents recruited in middle/ The mean weight increased from 61
Examine the previously validated high school; mean age 15 years; to 76 kg. Independent of lifestyle
diet quality score, and weight 10 years of follow-up to mean age factors and energy intake, a 15-point
change among adolescents 25 years (multivariate adjusted). higher diet quality score at age
transitioning into young A higher diet quality score had 15 years was associated with 1.5 kg
adulthood (US) [28] increasing levels of beans, whole grain less weight gain and lower BMI by
and nuts, white meat (fish, poultry), 0.5 over 10 years (p < 0.001).
fruits and vegetables, and low-fat Establishment of high-quality dietary
dairy, while decreasing the intake of patterns in adolescence may help
processed foods, red meat, and sweet reduce excess weight gain by young
and salty foods (eg, salty snacks, soft adulthood
drinks, sweet breads, grain desserts)
Feliciano et al. (2016). 67,175 postmenopausal women; mean A 10% improvement in any dietary
Examine whether changes in diet baseline age was 63 years, waist size pattern quality score was associated
quality predicts changes in 83 cm (34.6 in.), and BMI 27; 3-year with 0.07–0.43 cm smaller increase
central adiposity among follow-up; completed FFQs, and waist in waist size over 3 years (all
postmenopausal women size was measured; trunk fat was P < 0.05). After adjusting for weight
(Women’s Health Initiative measured in 4254 women via change, associations attenuated to
Observational Study; US) [29]. dual-energy X-ray absorptiometry 0.02–0.10 cm but remained
Dietary patterns (Healthy Eating (DXA); 3-year changes (multivariate statistically significant for all patterns
Index-2010, Alternate Healthy adjusted) except Alternate MedDiet. Results
Eating Index-2010, Alternate were similar for DXA trunk fat.
MedDiet, and Dietary Improvements in diet quality are
Approaches to Stop modestly protective against gain in
Hypertension (DASH) waist size, which is partially due to
lesser weight gain. Achieving and
maintaining a higher quality diet after
menopause may help protect against
gains in central adiposity
Fung et al. (2015). 123,098 women and 22,973 men; There was significantly less weight
Evaluate the association between mean baseline age 49 years for NHS I gain over a 4-year cycle with each
change of diet quality indexes and 36 years for NHS II, 48 years for standard deviation increase of diet
and concurrent weight change HPFS; mean BMIs ranged from quality score in both men and
over 20 years (Nurses’ Health 23–24.7; 20 years of follow-up; weight women. Improvement of diet quality
Study [NHS) I and II, and Health measures every 4 years (multivariate was associated with less weight gain,
Professionals Follow-up Study adjusted) especially in younger women and
[HPFS]; US) [30] overweight individuals
Lassale et al. (2012). 3151 participants; 1680 men and 1471 This study suggests that baseline diet
Assess and compare the women; mean baseline age 52 years; quality, measured by different dietary
predictive value of six different 13 years of follow-up (multivariate scores, is a good predictor of weight
dietary scores on risk of weight adjusted) gain across genders. Dietary quality
gain and obesity score appears to be especially
(Supplementation en Vitamines predictive of obesity risk in
et Mine’raux Antioxydants; middle-aged men. These findings
France) [31] support the broader use of dietary
scores for weight gain prevention at
the population level
Wolongevicz et al. (2010). 590 normal-weight women; BMI < 25, Women with lower diet quality were
Determine how diet quality aged 25–71 years; 16 years of significantly 76% more likely to
effects risk of being overweight follow-up (multivariate adjusted) become overweight or obese
or obese in women (Framingham compared with those with higher diet
Offspring and Spouse Study; quality (p-trend = 0.009)
US) [32]
Esmaillzadeh and Azadhakht 486 women, mean age 50 years; usual Women in the upper category of the
(2008). dietary intakes were evaluated by FFQ; healthy pattern score were less likely
Cross-sectional evaluation of with the use of factor analysis three to be obese by 59% and centrally
major dietary patterns and the major dietary patterns were extracted: obese by 52% (p < 0.05), whereas
prevalence of general obesity and healthy (9.5 g fiber/1000 kcal), those in the upper quintile of the
central adiposity among women Western (3 g fiber/1000 kcals, and Western pattern had greater risk for
(Iranian) [33] Iranian 8.5 g fiber/1000 kcals) general obesity by: 250% and for
(multivariate adjusted) central obesity by 533% (p < 0.01).
The Iranian dietary pattern was not
significantly associated with general
or central obesity
(continued)
Schulz et al. (2005). 24,958 participants; mean baseline age Mean annual weight gain gradually
Identify a dietary pattern 50 years, mean BMI 26; 4.4 years of decreased with increasing diet quality
predictive of subsequent annual follow-up (multivariate adjusted) score (p-trend <0.0001). A diet
weight change by using diet pattern characterized by low-fat,
composition information high-fiber (13.5 g fiber vs. 8.8 g
(European Prospective fiber/1000 kcal) foods such as
Investigation into Cancer and whole-grain bread, fruits, vegetables,
Nutrition EPIC - Potsdam and cereals was associated with body
cohort; Germany) [34] weight maintenance or prevention of
excess body weight gain in non-
obese subjects at baseline. This study
supports the importance of adequate
fiber intake for weight control
Newby et al. (2003). 459 healthy men and women; 52% The mean annual gain in BMI was
Assess the effect of dietary men; 95% whites; mean age for 0.30 units for subjects in the white
pattern on BMI and waist size women 57 years and men 61 years; bread and meat and potato pattern
(Baltimore Longitudinal Study healthy diet pattern high fruit, compared no change for those in the
of Aging; US) [35] vegetables, whole-grains, low in red healthy pattern (p < 0.01) (Fig. 7.3).
meat, fast foods, and soda (27 g fiber/ The mean annual gain in waist size
day) vs. white bread and meat and was 3 times as great for subjects in
potato pattern (multivariate adjusted) the white bread and meat and potato
pattern vs. those in the healthy
pattern (1.3 cm vs. 0.4 cm) (p < 0.05)
Mediterranean diet (MedDiet)
Li et al. (2015). 27,544 women; mean age 40; mean Among Swedish women, higher
Study long-term changes in BMI 22; MedDiet score 0–9; 12 years adherence to the MedDiet was not
anthropometric measures in a of follow-up (multivariate adjusted) associated with increased body
generally healthy population weight and waist size compared to the
(Swedish women) [36] average median gain in body weight
by 5 kg and waist size by 7.0 cm
Funtikova et al. (2014). 3058 subjects; 51% women, mean High adherence to the MedDiet was
Evaluate the association of baseline age 49 years; 10 years of inversely associated with waist size
adherence to the MedDiet and follow-up (multivariate adjusted) by 1.5 cm (p = 0.024). The 10-year
changes in waist size and 10-year risk of abdominal obesity was
incidence of abdominal obesity reduced in the highest tertile score by
(Spain) [37] up to 21%
May et al. (2012). 325,537 participants; 94,445 men and Men and women who reported to be
Investigate the combined effect 231,092 women, mean age physically active, never-smoking and
of physical activity, dietary 51–58 years; mean BMI 25–27; adherent to the MedDiet gained less
pattern, and smoking status on median 5 years of follow-up weight over a 5-year period for men
prospective gain in body weight (multivariate adjusted) by 537 g and women by 200 g and
and waist size (EPIC- about 1 cm less waist size compared
PANACEA) [38] to participants with no or poor
adherence to healthy behaviors
Beunza et al. (2010). 10,376 men and women; university Subjects with the lowest MedDiet
Investigate the risk of weight graduates; mean age 38 years; mean adherence had an average 0.3 kg
gain (≥ 5 kg) or the risk of 5.7 years of follow-up (multivariate annual weight gain, whereas those
developing overweight or obesity adjusted) with highest adherence had a loss of
(The Seguimiento Universidad 0.059 kg/year and a 24% lower risk
de Navarra (SUN) Cohort; of gaining ≥5 kg over the first 4 years
Spain) [39] of follow-up
Romaguera et al. (2010). 325,537 participants; 94,445 men and This study found that eating a
Assess associations between 231,092 women, mean age MedDiet may help to prevent weight
adherence to the MedDiet, 51–58 years; mean BMI 25–27; gain and the development of
weight change, and incidence of median 5 years of follow-up overweight and obesity. High
overweight or obesity (EPIC- (multivariate adjusted) adherence to the MedDiet reduced
Physical Activity, Nutrition, mean weight by 0.16 kg and risk of
Alcohol, Cessation of Smoking, becoming obese by 10% compared to
Eating out of home and obesity low adherence. A similar association
project PANACEA; EU) [40] between adherence to the MedDiet
and weight change was observed in
men and women
(p-interaction = 0.823). The
protective effect of MedDiet against
weight gain was stronger in younger
people (<40 years of age), and in
nonobese (BMI <30) individuals at
baseline (p -interactions <0.0001)
Sanchez-Villegas et al. (2006). 6319 participants; mean age In young, normal weight adults, those
Evaluate the potential relation 34–40 years; mean BMI 23; 28-months in the lowest quartile of MedDiet
between compliance with follow-up; 7.9 g vs. 14.9 g/ score gained 0.73 kg compared to
traditional MedDiet score and fiber/1000 kcal (multivariate adjusted) those in the top quartile who gained
subsequent weight maintenance 0.45 kg. Although there was an initial
and changes (SUN Cohort; inverse dose-response relationship
Spain) [41] (p-trend = 0.016), the results were
not statistically significant
(p-trend = 0.291)
Mendez et al. (2006). 17,238 women, mean baseline age Higher adherence to the MedDiet
Examine whether a MedDiet 47 years; 10,589 men; mean baseline was associated with a 30% lower risk
pattern is associated with age 50 years; mean of 3.3 years of of becoming obese. Associations
reduced 3-years incidence of follow-up (multivariate adjusted) were similar in women and men.
obesity (EPIC-Spain) [42] MedDiet adherence was not
associated with incidence of
overweight or obesity in initially
normal-weight subjects
Dietary Approaches to Stop Hypertension (DASH) diet
Barak et al. (2014). 293 female nurses aged >30 years; Increased adherence to the DASH
Investigate adherence to DASH general and abdominal obesity were diet was associated with older age
diet and general and central defined as BMI ≥25 and waist size (p < 0.01) and lower waist size
obesity in female nurses ≥88 cm; usual dietary intakes were (p = 0.04). Initially, there was no
(cross-sectional study; Iran) [43] assessed using a validated FFQ; DASH statistically significant difference in
diet score was based on foods and the prevalence of general obesity
nutrients emphasized or minimized in between extreme quartiles of the
the DASH diet (multivariate adjusted) DASH diet score but after fully
adjusting for dietary factors, those in
the highest quartile of DASH diet
score were 71% less likely to have
general obesity. A marginally
significant trend towards decreasing
prevalence of central obesity was
seen with increasing quartile of the
DASH diet score with a 63% lower
waist size (p = 0.09)
(continued)
Berz et al. (2011). 2327 girls with ten annual visits Adolescent girls with higher
Study the effects of the DASH starting at age 9 years; follow-up of adherence to the DASH eating
eating pattern on BMI 10 years (multivariate adjusted) pattern had smaller gains in
throughout adolescence BMI. Girls in the highest vs. lowest
(National Growth and Health quintile of the DASH score had
Study; US) [44] significantly lower adjusted mean
BMI of 24.4 vs. 26.3. The strongest
individual food group predictors of
BMI were total fruit with a mean
BMI of 26.0 vs. 23.6 for <1 vs. ≥2
servings/day (p < 0.001) and low-fat
dairy with a mean BMI of 25.7 vs.
23.2 for <1 vs. ≥2 servings/day
(p < 0.001). Whole grain
consumption was more weakly but
beneficially associated with BMI
Vegetarian diet
Tonstad et al. (2009). 22,434 men and 38,469 women; mean This study indicates that
Assess the effects of different age 58 years; data from Seventh-Day vegetarianism is protective against
types of vegetarian diets on body Adventist church members across obesity and diabetes. Mean BMI was
weight and diabetes risk North America; type of vegetarian diet 23.6 for vegans, 25.7 for lacto-ovo
compared with non-vegetarians was categorized based on a FFQ vegetarians, 26.3 for pesco-
(The Adventist Health Study-2 (multivariate adjusted) vegetarians, and 27.3 for semi-
cohort; US) [45] vegetarians, and 28.8 for
non-vegetarians. Prevalence of type 2
diabetes increased from 2.9% in
vegans to 7.6% in non-vegetarians
Berkow and Barnard (2006). 40 studies reporting the weight status 29 of 40 observational studies
Review of published of vegetarians and non-vegetarians reported that vegetarians weighed
observational studies on the significantly less than non-
associations between vegetarian vegetarians as measured by BMI or
diets and reduced body body weight. These studies found
weight [46] that the weight and BMI of both male
and female vegetarians were 3–20%
lower than those of non-vegetarians.
Obesity prevalence ranges were from
0 to 6% in vegetarians and from
about 5 to 45% in non-vegetarians
obesity in both men and women, especially in nuts, plus yogurt was associated with decreased
non-obese subjects at baseline (Table 7.1) [26– adiposity, while red and processed meats were
35]. A systematic review (10 cohort studies and associated with greater adiposity (Fig. 7.2) [27].
24 cross-sectional studies) found that high adher- A study of US adolescents (2656 middle/high
ence to dietary quality based on the dietary guide- school youth; mean age 15 years; 10-year follow-
lines (or Healthy Eating Indices) is inversely up to mean age 25 years) showed that a 15-point
associated with obesity and other body weight higher diet quality score (on a 100 point scale) at
parameters [26]. A cross-sectional analysis of the age 15 years was associated with 1.5 kg less
Multi-Ethnic Study of Atherosclerosis [MESA] weight gain and lower BMI by 0.5 kg/m2 over
(5079 adults; mean age 61 years; 47% men) 10 years (p < 0.001), independent of other life-
demonstrated that higher intake of fiber contain- style factors and energy intake [28]. The Women’s
ing fruits, vegetables, whole-grains and seeds/ Health Initiative Observational Study (67,175
530
520
Visceral fat area (cm2/m) 510
500
490
480
470
460
450
1 2 3 4
Quartiles of Diet-Quality Score
Fig. 7.2 Association between diet-quality score and visceral fat area from a cross-sectional analysis of the US Multi-
Ethnic Study of Atherosclerosis (p < 0.01; multivariate adjusted) (adapted from [27])
postmenopausal women with waist size measure- quality, especially low-fat, high-fiber (13.5 g fiber
ments; 3 years) found that a 10% improvement in vs. 8.8 g fiber/1000 kcal) diets (in non-obese sub-
diet quality can significantly reduce multivariate jects at baseline) was inversely associated with
adjusted waist size by up to 0.1 cm over 3 years weight gain [34]. The 2003 Baltimore
[29]. A pooled analysis of the large Nurses’ Longitudinal Study of Aging (459 men and
Health Study I and II, and Health Professionals women; mean age about 60 years; healthy diet
Follow-Up study (123,098 women and 22,973 [27 g fiber/day; high in fruit, vegetables, and
men; women mean baseline age 36–48 years; men whole-grains, and low in red meat, fast foods, and
mean baseline age 58; normal BMI; 20 years of sugary soda] vs. other types of Western diets
follow-up) showed that higher adherence to all ≤20 g/fiber/day) observed significantly less
types of high quality diet patterns was signifi- annual BMI and waist size gain for the healthy vs.
cantly associated with less weight gain over each Western dietary patterns (Fig. 7.3) [35].
4-year weight assessment period in both men and
women, especially in younger women or over- 7.2.2.2 Mediterranean Dietary Pattern
weight individuals [30]. Several prospective stud- (MedDiet)
ies report that higher dietary quality is a good Of the healthy dietary patterns, the MedDiet has
predictor of lower weight gain and obesity risk in been the most studied for weight control
women [31, 33]. The Framingham Offspring and (Table 7.1) [36–42]. Higher adherence to the
Spouse Study (590 normal weight women at base- MedDiet has consistently been shown to be
line; 16-year follow-up) found that subjects with inversely associated with weight gain or risk of
lower quality diets were associated with a 76% general or central obesity in both men and women.
increased risk of becoming overweight or obese A prospective cohort study among Swedish
compared to subjects with high quality diets [32]. women (27,544 women; mean baseline age 40;
The European Prospective Investigation into 12-years of follow-up) showed that women with
Cancer and Nutrition (EPIC) - Potsdam, German higher adherence to the MedDiet were not associ-
cohort (24,958 participants; mean baseline age ated with increased body weight and waist size
50 years; 4.4 years of follow-up) showed that diet compared to the average median gain in body
Healthy pattern White bread pattern * Meat and potatoes pattern *

1.4
1.2
1
Mean Annual Change
0.8
0.6
0.4
0.2
0
BMI Waist size (cm)
Fig. 7.3 Effect of healthy vs. Western type derived food dietary pattern on BMI and waist size in men and women from
the Baltimore Longitudinal Aging Study (*p < 0.05) (adapted from [35]). Healthy pattern: contained relatively greater
contributions from “healthy” foods, including fruit, high-fiber cereal, and reduced fat dairy, and relatively smaller con-
tributions from fast food, nondiet soda, and salty snacks. White bread pattern: higher levels of refined carbohydrates
and baked goods. Meat and potatoes pattern: higher levels of meats and higher energy dense vegetables such as potato
products
weight by 5 kg and waist size by 7.0 cm in the actions <0.0001) [40]. Also, several Spanish
overall cohort [36]. A Spanish cohort study (3058 cohort studies show that higher adherence to the
subjects; 51% women; mean baseline age MedDiet significantly reduced risk of obesity by
49 years; 10 years of follow-up) found that high 10–30% with mean weight reductions by 0.16 kg
adherence to the MedDiet lowered waist size by over 3.3–5.0 years [40–42].
1.5 cm (p = 0.024; fully adjusted models) [37].
The EPIC-Spanish cohort (325,537 participants; 7.2.2.3 Dietary Approaches to Stop
mean baseline age 51–58 years; 5 years of follow- Hypertension (DASH) Dietary
up) showed that non-smoking men and women Pattern
with higher physical activity and high adherence Two studies indicate that high adherence to
to the MedDiet gained less weight and waist size DASH diets helps to prevent weight gain in ado-
compared to subjects adhering to the Western lescent girls and women (Table 7.1) [43, 44]. A
lifestyle [38]. A 2010 Spanish university study 2014 cross-sectional study (293 female nurses;
(10,376 men and women; mean baseline age > 30 years) showed that women with highest
38 years; 5.7 years of follow-up) found that those adherence to the DASH diet were 71% less
subjects with the lowest MedDiet adherence had likely to have general obesity and had a margin-
an average 0.3 kg annual weight gain, whereas ally significant trend towards a lower prevalence
those with highest MedDiet adherence lost of central obesity as measured by waist size by
0.06 kg/year and had 24% lower risk of gaining 63% (p = 0.09) [43]. A 2011 US cohort study in
≥5 kg over the first 4 years of follow-up [39]. adolescent girls (2327 girls; annual visits start-
High adherence to the MedDiet was shown to ing at baseline age 9 years; 10-year follow-up)
protect against weight gain with better effects in found that higher adherence to the DASH eating
younger people (<40 years of age) and in non- pattern was associated with a significantly lower
obese (BMI <30) individuals at baseline (p-inter- BMI by 1.9 units [44].
7.2.2.4 Vegetarian Dietary Pattern the comparator diets. A meta-analysis in people

Observational studies generally show that veg- with type 2 diabetes (9 RCTs; 1178 adults; age
etarian diets protect against weight gain and range 26–77 years; 1 month to 4 years of fol-
may promote weight loss depending on the low-up) found a small but significant mean loss
level of strictness or adherence [45, 46]. The of weight by 0.3 kg and BMI by 0.3 kg/m2 com-
Adventist Health Study-2 prospective study pared to control diets along with significant
(60,903 subjects; mean age 58 years; 60% improvements in glycemic control and reduc-
female; 5 years of follow-up) found that vege- tions in cardiovascular disease risk factors [48].
tarian diet strictness was inversely associated Another meta-analysis (16 RCTs; 3436 partici-
with BMI with vegans having a 5-unit lower pants; 1 month to 2 years) demonstrated that
BMI than nonvegetarians [45]. Twenty-nine of subjects on the MedDiet significantly reduced
40 observational studies reported significantly weight by 1.75 kg and BMI by 0.6 kg/m2 com-
reduced BMI or body weight by 3–20% in veg- pared to the control diet [49]. The effect of the
etarians compared with non-vegetarians [46]. MedDiet on weight loss was further improved
The incidence of obesity ranged from 0 to 6% with an energy restricted diet by 3.9 kg, with
in vegetarians and from about 5 to 45% in increased physical activity by 4.0 kg, or with
non-vegetarians. trial durations >6 months by 2.7 kg. This analy-
sis also showed that unrestricted intake of
MedDiets does not promote weight gain, which
7.2.3 Randomized Controlled Trials helps to lessen weight gain concerns about the
(RCTs) MedDiet’s liberal use of olive oil and nuts. A
meta-analysis of metabolic syndrome subjects
Table 7.2 summarizes RCTs on the effects of (33 RCTs, and 2-cohort and 13-cross-sectional
healthy dietary patterns on body weight and com- studies; 534,906 subjects) demonstrated that the
position [47–63]. MedDiet significantly reduced waist size, blood
pressure, fasting blood glucose and the preva-
7.2.3.1 Mediterranean Diet (MedDiet) lence of metabolic syndrome, and increased
HDL-C [50].
Four systematic reviews and meta-analyses of Specific RCTs
RCTs consistently show that ab libitum intake Six RCTs describe the various aspects of the
of MedDiets do not result in weight gain and effects of MedDiets, including the importance of
high adherence to MedDiets supports weight adequate fiber intake on reducing and managing
loss and lower waist size compared to control body weight and composition [51–56]. Three
diets such as low-fat and Western diets espe- Spanish PREDIMED trials show impressive
cially in overweight, obese or type 2 diabetic long-term effects of the MedDiet on maintaining
individuals, and with longer trial duration (> and moderately lowering body weight and WC
6 months), or in conjunction with restricted [51–53]. The PREDIMED trial (7447 subjects
energy intake or increased physical activity with type 2 diabetes or high cardiovascular risk;
[47–50]. A meta-analysis of long-term MedDiet mean baseline age 67 years and BMI 30; MedDiet
intake (5 RCTs; 998 subjects; comparator diets plus extra virgin olive oil or tree nuts and reduced-
low-fat, low carbohydrate and American fat control diet; 5-year duration) showed in a
Diabetes Association (ADA); ≥ 12 months) long-term intervention that the unrestricted
showed that adherence to MedDiets resulted in MedDiets were not associated with weight gain
greater weight loss than a low-fat diet, but pro- [51]. MedDiet plus extra virgin olive oil signifi-
duced similar weight loss to low carbohydrate cantly reduced mean body weight by 0.43 kg
or ADA diets [47]. Also, the MedDiet was more compared to the lower fat control diet. Both
effective in lowering BMI and waist size than all MedDiets lowered waist size vs. the control diet
Table 7.2 Summary of dietary pattern randomized controlled trials (RCTs) in body weight and composition
regulation
Mancini et al. (2016). 5 RCTs, 998 subjects; trials compared The MedDiet resulted in greater mean weight
Systematic review of the MedDiet to low-fat diet, a loss by −4.1 to −10.1 kg vs. a low-fat diet by
the effect of the low-carbohydrate diet, and the 2.9 to −5.0 kg, but produced similar mean
MedDiet on weight loss American Diabetes Association weight loss by −4.1 to −10.1 kg vs. low
and waist size [47] (ADA) diet; ≥ 12 months carbohydrate and ADA diets by −4.7 to
(12–48 months) −7.7 kg. Also, the MedDiet lowered mean BMI
by −1.0 to −3.3 kg/m2 vs. the other diets by 1.4
to −1.8 kg/m2 and mean waist size vs. low-fat
diets by −3.5 to −9.3 cm vs. low-fat diet by 2.6
to −3.5 cm)
Huo et al. (2014). 9 RCTs; 1178 diabetic subjects; age In diabetic subjects, the MedDiet had greater
Meta-analysis of the range at baseline 26–77 years; mean reductions in BMI by 0.29 units, body
effects of MedDiet on 4 weeks to 4 years weight by 0.29 kg, hemoglobin A1c by 0.30
glycemic control, unit, fasting plasma glucose by 0.72 mmol/L,
weight loss and fasting insulin by 0.55 μU/mL, total cholesterol
cardiovascular risk by 0.14 mmol/L, triglycerides by 0.29 mmol/L,
factors in type 2 and both systolic and diastolic BP by
diabetes patients [48] 1.4 mm Hg. Also, HDL-C was increased by
0.06 mmol/L vs. control diets
Esposito et al. (2011). 16 RCTs; 3436 participants; mean MedDiet can be effective in lowering body
Evaluate the effect of age 35–69 years and BMI 26–35; weight, especially with energy restriction,
MedDiets on body 1 month to 2 years duration increased physical activity, and >6 months in
weight using meta- duration. Overall, the MedDiet significantly
analysis [49] reduced weight by 1.75 kg and BMI by
0.57 units. In studies lasting longer than
6 months mean weight loss was 3.9 kg. Also,
MedDiet accompanied with either a restricted
energy diet or increased physical activity
reduced weight by approximately 4 kg
Kastorini et al. (2011). 50 original research studies (35 Adherence to the MedDiet
Meta-analysis of the RCTs, 2 prospective and 13 was associated with significantly reduced MetS
effect of a MedDiet on cross-sectional studies) through April prevalence by 31%, waist size by 0.42 cm,
metabolic syndrome 30, 2010; 534,906 participants triglycerides by 6.1 mg/dL, systolic blood
(MetS) and its pressure (BP) by 2.4 mmHg, diastolic BP by
components [50] 1.6 mmHg and fasting glucose by 3.9 mg/dL
and increased HDL-C by 1.2 mg/dL
Specific RCTs
Estruch et al. (2016). Multicenter, Parallel RCT: This study showed that the long-term intake of
Assess the long-term 7447 adults with type 2 diabetes or plant-based, unrestricted-calorie, high-fat diets,
effects of ad libitum, ≤3 CVD risk factors; mean age such as the traditional MedDiet does not
high-vegetable-fat 67 years and BMI 30; three different promote weight gain. The adjusted difference in
MedDiets on ad libitum diets: MedDiet plus extra 5-year loss in body weight in the MedDiet plus
bodyweight and waist virgin olive oil; MedDiet plus tree extra virgin olive oil group was 0.43 kg
size in older people at nuts (total dietary fat approx. 100 g/ (p = 0.044) and in the plus nut group was
risk of cardiovascular day for both; or advice to reduced fat 0.08 kg (p = 0.730), compared with the reduced
disease, most of whom control diet (96 g fat/day). Advice to fat control group. The adjusted difference in
were overweight or restrict dietary energy or promote 5-year reduction in waist size was 0.55 cm
obese (PREDIMED; physical activity was not advised; (p = 0.048) in the MedDiet with extra virgin
Spain) [51] 5 years olive oil group and 0.94 cm (p = 0.006) in the
nut group, compared with the reduced fat
control group (Fig. 7.4)
Alvarez-Perez et al. Parallel RCT: This study found that unrestricted MedDiets that
(2016). 351 free-living subjects, mean age contain approximately 40% total fat can be alternative
Assess effect of the 67 years and BMI 31; 64% women; options to reduced-fat diets for weight maintenance.
MedDiet on with type 2 diabetes or ≤3 CVD risk Significant reductions in body weight by 1 kg, BMI
anthropometric and factors; ad libitum diets: MedDiet plus by 0.5 units and waist size by 1.1 cm (p < 0.05; all)
body composition extra virgin olive oil, MedDiet plus were observed for the MedDiet plus extra virgin olive
parameters mixed tree nuts, or a control low-fat diet oil vs. the control group. The MedDiet plus nuts
(PREDIMED trial; guidance; study assessed changes in group exhibited a significant reduction in waist size
Spain) [52] anthropometric measures of body by 2.3 cm (p < 0.001). The control group showed a
weight, BMI, waist size, total body fat significant increase in total body fat by 1% (p = 0.02)
%; 1 year
Damasceno et al. Parallel RCT: Compared to the MedDiet plus extra virgin olive
(2013). 169 subjects with type 2 diabetes or oil and reduced-fat control, participants in the
Investigate effect of ≤3 CVD risk factors; mean age tree nut-enriched MedDiet showed significantly
MedDiets on changes in 67 years; 75% women; BMI 29.5; reduced waist size by 5 cm (p = 0.006 for both)
adiposity and lipoprotein subclasses (particle and increased LDL size with a net increase by
lipoprotein subfractions concentrations and size) were 0.2 nmol/L (p < 0.05 for both). Also, there were
vs. a reduced fat control determined by NMR spectroscopy; increased large HDL concentrations in both the
diet (PREDIMED diets; three different ad libitum diets: olive oil and nut supplemented MedDiets
Spain) [53] MedDiet plus extra-virgin olive oil,
MedDiet plus mixed tree nuts (30 g
walnuts, almonds, and hazelnuts/day),
or a control reduced-fat diet; 1-year
Shai et al. (2008). Parallel RCT: Compared to other diet groups, the MedDiet
Compare the 322 moderately obese subjects; mean group consumed the largest amounts of dietary
effectiveness of age 52 years; mean BMI 31; males fiber and the low-carbohydrate group consumed
weight-loss diets (US) 86%; 3 restricted-calorie diets: the smallest amount of carbohydrates
[54] low-fat, MedDiet, or low- and the largest amounts of fat, protein, and
carbohydrate; 2 years cholesterol (P < 0.05 for all). The mean weight
loss was 4.4 kg for the MedDiet group, 4.7 kg for
the low-carbohydrate group and 2.9 kg for the
low-fat group (P < 0.001) (Fig. 7.5). MedDiet and
low-carbohydrate diets appear to be effective
alternatives to low-fat diets for weight loss and
there are more favorable effects on glycemic
control with the MedDiet and on lipids with the
low-carbohydrate diet. The rate of adherence to
these diets was 95% at 1 year and 85% at 2 years
Esposito et al. (2004). Parallel RCT: Compared to the control diet, subjects on the
Assess the effect of a 180 metabolic syndrome patients (99 MedDiet had a significantly greater mean
MedDiet on weight and men and 81 women); 2 diets: MedDiet decrease in body weight by 11 kg and BMI by
cardiometabolic advice about how to increase daily 4.2 units (p < 0.001) (Fig. 7.6). Also, compared
markers associated with consumption of whole grains, fruits, to the control group, the MedDiet group had
metabolic syndrome vegetables, nuts, and extra virgin olive significantly reduced serum concentrations of
(Italy) [55] oil (32 g fiber/day); control group hs-CRP (p = 0.01), lower insulin resistance
followed a prudent diet (carbohydrates, (p < 0.001), and 50% fewer patients classified
50–60%; proteins, 15–20%; total fat, still continued to be with metabolic syndrome
30%; 17 g fiber/day); 2 years (p < 0.001)
Esposito et al. (2003). Parallel RCT: The intervention group consumed 9 g fiber/day
Determine the effect of 120 premenopausal women; mean more and 310 kcals less than the usual diet
an energy restricted age 35 years and BMI 35; control group (p < 0.001). Changes in body
MedDiet and physical intervention group received detailed weight and BMI are shown in Fig. 7.7. The
activity on body weight, advice to reduce weight by ≥10% intervention was also associated with reduction in
systemic inflammation, with reduced energy MedDiet and CRP by 0.8 mg/L and HOMA- insulin resistance
and insulin resistance increased physical activity vs. control by 0.9 units vs. control (p = 0.008; both)
(Italy) [56] group given general info on healthy
food choices and exercise; 2 years
(continued)

DASH diet
Systematic review and meta-analysis
Soltani et al. (2016). 13 RCTs; 1291 participants; 10 on Subjects on the DASH diet lost more weight by
Assess the effect of a body weight, 6 on BMI and 2 on 1.42 kg in 24 weeks, reduced BMI by 0.42 units
DASH dietary pattern waist size; 8–52 weeks in 8–52 weeks and waist size by 1.05 cm in
on body weight and 24 weeks compared with controls. The effect
composition in adults was greater in overweight/obese participants
[57] and when compared with typical Western diet
or a population’s usual diets. Lower caloric
DASH led to more weight reduction when
compared with other low-energy diets. The
DASH diet is effective for weight management,
especially for weight reduction in overweight
and obese participants
Healthy Nordic diets
Nordic weight loss diet for lactating women
Bertz et al. (2012, Parallel RCT with Follow-up This dietary treatment was sufficient to
2015). Analysis: significantly and clinically meaningfully
Assess the effect of an 68 women; BMI 25–35; mean age promote weight and total fat loss, and lower
energy restricted diet on 33 years and BMI 30; intervention BMI in lactating women, and to sustain weight
weight loss among weight loss diet based on Nordic loss at 9-month follow-up after the intervention
overweight/obese Nutrition Recommendation: subjects ended. The intervention diet was lower in
lactating women were instructed to restrict energy energy by approx. 400–500 kcals and increased
(Lifestyle Weight Loss intake by 500 kcals, limit sweets fiber intake by 3 g/1000 kcal. Changes in body
During Lactation Trial; and snacks to 100 kcal/week, weight are shown in Fig. 7.8. BMI was reduced
Sweden) [58, 59] substitute lower fat and sugar by ≥3 units and total weight was reduced by
alternatives for usual foods, cover ½ 5.5–6.7 kg (p < 0.001)
the lunch and dinner plate with
vegetables, and reduce portion size
vs. usual diet; 12 weeks plus
9-month follow-up
New Nordic diet
Paulsen et al. (2014). Parallel RCT: Free-living intake of the NND reduced mean
Evaluate health effects 181 centrally obese men and women; body weight by 3.2 kg (Fig. 7.9) and mean
of the New Nordic Diet, 71% women; mean age of 42 years waist size by 2.9 cm (p < 0.001; both)
which is a food-based (20–66 years), mean BMI 30.2, and compared to the average Danish diet. Also, the
dietary concept recently waist size 100 cm (39 in.); received New Nordic Diet produced greater reductions
developed in the Nordic either the New Nordic Diet (high in in systolic blood pressure by 5.1 mmHg
countries in fruit, vegetables, dairy products whole compared to the average Danish diet
collaboration with grains, and fish, and low in sugar, (p < 0.001). The weight loss was shown despite
Copenhagen’s gourmet cakes, pastries, and biscuits) or an the fact that the diet was developed as highly
restaurant NOMA. The average Danish diet; New Nordic Diet palatable and offered ad libitum, and the study
NND is based on contained 19 g/day more total fiber and was not specifically designed as a weight-loss
regional foods in 21 kcals (87.5 kJ) less energy/100 g study
season, with a strong than average Danish diet; 26 weeks
emphasis on
palatability, healthiness,
and sustainability,
while aligning with
regional food culture
and dietary preferences
(Denmark) [60]
Vegetarian diet
Huang et al. (2016). 12 RCTs; 1151 subjects; median Overall, individuals assigned to the vegetarian
Investigate the effects duration of 18 weeks diet groups lost significantly 2 kg more weight
of lacto-ovo-vegetarian than those assigned to the non-vegetarian diet
and vegan diets on groups. Subgroup analysis detected significant
weight reduction [61] weight reduction in subjects consuming a vegan
diet by 2.5 kg and, to a lesser extent, in those
given lacto-ovo-vegetarian diets by 1.5 kg.
Trials on subjects consuming vegetarian diets
with energy restriction revealed a significantly
greater weight reduction of 2.2 kg than those
without energy restriction of 1.7 kg
Barnard et al. (2015). 15 RCTs; 755 participants (197 Vegetarian diets were associated with a mean
Estimate the effect of ovo-lactovegetarian and 558 vegan); weight loss of 3.4 kg (p < 0.001) in an
vegetarian diets on ≥ 4 weeks without energy intake intention-to-treat analysis vs. control diet.
body weight [62] limitations Greater weight loss was shown in studies with
higher baseline weights, smaller proportions of
female participants, older participants, or longer
durations, and in studies in which weight loss
was a goal
Specific comparative RCT of different types of vegetarian diets
Turner-McGrievy Parallel RCT: After 6 months, weight was significantly
et al. (2015). 63 overweight and obese adults; 19% reduced in the vegan group by 7.5% and
Determine the effect of non-white; 27% men; mean baseline ovo-lacto vegetarian by 6.3% compared to the
plant-based diets on age 48 years and BMI 35; omnivorous, semi-vegetarian and pesco-
weight loss (US) [63] randomized into 5 diets: a low-fat, vegetarian groups by approximately 3%
low-glycemic index diet: vegan, (p = 0.03) (Fig. 7.10). Vegan diets may result in
vegetarian, pesco-vegetarian, greater weight loss than more modest
semi-vegetarian, or omnivorous; recommendations
6 months
after 5 years (Fig. 7.4). Two sub-cohort effective as low-fat or low carbohydrate diets in
PREDIMED trials found that: (1) MedDiets (351 promoting significant weight loss over 2 years
free-living subjects with type 2 diabetes or ≤3 (p < 0.001) (Fig. 7.5) [54]. An Italian trial in 180
CVD risk factors, mean age 67 years and BMI adults with metabolic syndrome found significant
31; 64% women) resulted in significantly lower body weight, BMI and waist size lowering effects
body weight, BMI and waist size compared to the for an ab libitum 32 g fiber/day MedDiet (includ-
lower-fat control diet (which also significantly ing about 500 g of whole-grains, vegetables,
increased total body fat by 1%) [52]; and (2) Tree fruit, legumes, and nuts) compared to a 17 g fiber/
nut-enriched MedDiet (169 subjects with type 2 day prudent diet (including about 200 g of whole-
diabetes or ≤3 CVD risk factors; mean baseline grains, vegetables, fruit, legumes, and nuts) after
age 67 years and BMI 29.5; 75% women; 1 year) 2 years (Fig. 7.6) [55]. Also, an Italian weight
significantly reduced waist size by 5 cm and loss trial (120 premenopausal women; mean
increased LDL size by 0.2 nmol/L compared to baseline age 35 years and BMI 35; 2 years)
the MedDiet plus extra virgin olive oil, and the showed significant effects of an energy restricted
lower-fat control diet [53]. In a US-based com- MedDiet compared to the usual diet along with
parative analysis of non-energy restricted diets increased physical activity on lowering body
(322 obese subjects; 86% men; mean baseline weight, BMI, CRP and insulin resistance
age 52 years and BMI 31), MedDiets were as (HOMA-IR) (Fig. 7.7) [56].
Lower fat control MedDiet plus extra MedDiet plus tree nuts
virgin olive oil
101.5
100.5
Waist Size Change (cm)
99.5
98.5
97.5
0 1 2 3 4 5
Years of Follow-up
Fig. 7.4 Effect of unrestricted MedDiet and lower fat control diets on mean waist size from the PREDIMED trial over
5 years (multivariate adjusted values) (adapted from [51])
0.0
Low-fat diet MedDiet Low-carbohydrate diet
-0.5
Change in Body Weight (kg) from Baseline
-1.0
-1.5
-2.0
-2.5
-3.0
-3.5
-4.0
-4.5
-5.0
Fig. 7.5 Effect of the non-energy restricted MedDiet vs. low fat and low carbohydrate non-energy restricted diets in
322 obese adults (about 90% men) over 2 years (adapted from [54])
7.2.3.2 Other Diets by 1 cm in 24 weeks compared with control

diets [57].
DASH Diet
A meta-analysis (13 RCTs; 1291 participants; Nordic Diets
8–52 weeks) showed that adults on a DASH Nordic Weight Loss Diet for Lactating
diet lost more weight by 1.42 kg and BMI Women. Childbearing is associated with weight
by 0.42 kg m2 in 8–24 weeks and waist size gain because of gestational weight gain and
Weight (kg) BMI (kg/m2) Waist circumference (cm)

0.0
MedDiet (32 g fiber/day) Prudent diet (17 g fiber/day)
-0.5
-1.0
Change from Baseline
-1.5
-2.0
-2.5
-3.0
-3.5
-4.0
-4.5
Fig. 7.6 Effect of a high fiber-rich MedDiet compared to a moderate fiber prudent diet in 180 adults with metabolic
syndrome over 2 years (p < 0.001; all) (adapted from [55])
Weight (kg) BMI (units)

0.0
Hypocaloric MedDiet Usual Diet
(25 g fiber/day)* (16 g fiber/day)
-4.0
Change from baseline
-8.0
-12.0
-16.0
Fig. 7.7 Effect of a hypocaloric Mediterranean diet (MedDiet) plus increased physical activity vs. usual diet and exer-
cise advice in 120 obese women over 2 years (p < 0.001; both) (adapted from [56])
postpartum weight retention, which can make it promotion of weight loss, lowering of BMI, and
difficult for women to return to pre-pregnancy total body fat loss in lactating women, and sus-
weight [58, 59]. The Swedish Lifestyle Weight tained weight loss at 9-month follow-up after
Loss During Lactation Trial (68 women; pre- the intervention ended compared to the usual
pregnancy BMI 25–35; mean age 33 years and diet [58, 59]. The primary guidelines for the
BMI 30; intervention weight loss diet based on Nordic diet was to target restricted energy intake
Nordic Nutrition Recommendation: vs. usual by 500 kcals by limiting sweets and snacks to
diet; 12 weeks duration plus 9-month follow- 100 kcal/week, substituting lower fat and sugar
up) found significant and clinically meaningful alternatives for usual foods, designating half of
the lunch and dinner plate for vegetables, and Vegetarian Diets
reducing portion size, which reduces energy Two meta-analyses of vegetarian diets show that
intake by approx 400–500 kcals and increase all vegetarian diets protect against weight gain
fiber intake by 3 g/1000 kcal compared to the [61, 62]. A 2016 systematic review and meta-
usual diet. Changes in body weight are shown in analysis (12 RCTs; 1151 subjects; 18-week mean
Fig. 7.8. BMI was reduced by ≥3 units and total duration) found that individuals on vegetarian
weight loss was reduced by 5.5–6.7 kg diets lost significantly 2 kg more weight than
(p < 0.001). those assigned to the non-vegetarian diets [61].
New Nordic Diet (NND). This food-based Subgroup analysis detected significant weight
dietary concept was developed in the Nordic reduction in subjects consuming a vegan diet by
countries in collaboration with the world-leading 2.5 kg and, to a lesser extent, in those given lacto-
Copenhagen gourmet restaurant NOMA [60]. ovo-vegetarian diets by 1.5 kg. Trials on subjects
This diet is based on regional foods in season, consuming energy restricted vegetarian diets
with a strong emphasis on palatability, healthi- found a significantly greater weight reduction by
ness, and sustainability, which are aligned with 2.2 kg than those without energy restriction by
regional food culture and dietary habits. The 1.7 kg. The weight loss for subjects with follow-
basic food components of the NND include: fruit up of <1 year was greater than those with follow-
and vegetables (especially berries, cabbages, root up of ≥1 year (−2.05 kg vs. −1.13 kg). A 2015
vegetables, and legumes), potatoes, fresh herbs, meta-analysis (15 RCTs; 755 adults; 197 lacto-
mushrooms, nuts, whole grain, meats from live- ovo vegetarians and 558 vegans; 75% females;
stock and game, fish and shellfish, and seaweed, no energy restricted diets; ≥ 4 weeks) showed
which provide 19 g fiber/day more than the that combined lacto-ovo vegetarian and vegan
average Danish diet. A Danish trial (181 adults; diets significantly reduced weight by 3.4 kg,
71% women; mean age 42 years and BMI 30; despite the absence of specific guidance on
26-week duration) found that the unrestricted energy intake or exercise [62]. Greater weight
NND significantly reduced body weight (Fig. 7.9) loss was found in studies with higher baseline
and waist size by 2.9 cm compared to the average weights, older participants, or longer durations.
Danish diet [60]. In a 5-arm plant-based weight loss RCT
Usual Diet Nordic Weight loss Diet

88
86
84
82
Weight Loss (kg)
80
78
76
74
72
70
Baseline 3 months Intervention 1 year (9 months
follow-up)
Fig. 7.8 Effects of Nordic energy restricted and higher fiber diet intervention on body weight in lactating overweight
and obese women (p < 0.001) (adapted from [58, 59])
7.3 Effect of Fiber on Body Weight and Composition 213
New Nordic Diet Average Danish Diet

0.5
0
w0 w2 w4 w8 w12 w16 w20 w24 w26
-0.5
Body Weight Change (kg)
-1
-1.5
-2
-2.5
-3
-3.5
-4
-4.5
Fig. 7.9 Change in body weight (intention-to-treat) for the New Nordic Diet compared to the average Danish diet over
26 weeks (p < 0.001) (adapted from [60])
Vegan Lacto-ovo -veg Pesco - veg

Semi - veg Omnivore
0
0 2 6
-1 Months
-2
Weight Loss (kg)
-3
-4
-5
-6
-7
-8
Fig. 7.10 Effect of type of vegetarian (veg) diets on weight loss in adults after 6 months (p-trend =0.01 for vegan and
lacto-ovo-vegetarian diets) (adapted from [63])
(63 subjects; mean age 48 years; mean BMI 35; ovo-vegetarian diets have similar greater effects
73% female; vegetarian vs. omnivorous diets; on weight loss compared to omnivorous, semi-,
6 months) reported that vegan and lacto- and pesco-vegetarian diets (Fig. 7.10) [63].
7.3 ffect of Fiber on Body

E lations do not consume an adequate level of fiber
Weight and Composition daily (14 g fiber/1000 kcals or 25 g/day for
women and 38 g/day for men) [68, 69]. A Spanish
7.3.1 Fiber-Rich Diets cross-sectional study (1655 adults; age
18–64 years) found that fiber intake (adjusted by
The human gastrointestinal and energy regula- energy intake) was significantly higher in sub-
tory systems evolved over most of the last jects with normal weight and without excess
40,000 years on diets with ≥50 g fiber/day [64]. abdominal obesity compared to those who were
These pre-agricultural high fiber dietary patterns overweight or obese [70]. Total fiber intake was
are in sharp contrast to the present low fiber, high inversely associated with obesity risk among US
energy dense Western diets, a relatively recent adults in an analysis of NHANES 1999–2010
occurrence in human evolution (Table 7.3) [65]. (Fig. 7.11) [71]. Populations with higher fiber
Ancestral fiber-rich whole foods diets stimulated diets tend to be leaner than those with low fiber
the evolution of the important colon microbiota diets [72–74]. A systematic review of 43 pro-
ecosystem, which is equivalent to a symbiotic spective cohort, case-control and randomized tri-
‘organ’ that supports optimal energy metabolic als found moderately strong evidence that
and cardiometabolic health processes, increased fiber-rich foods have a protective role against
excretion of fecal metabolizable energy and fiber weight gain and increased waist size [72]. In
fermentation to short chain fatty acids (SFCAs) 2010, the European Food Safety Authority
associated with a range of biological activities (EFSA) recommended that adults should con-
associated with better weight control [66, 67]. sume >25 g fiber per day (from whole-grain cere-
About 95% of Americans or other Western popu- als, fruit, and vegetables) to improve weight
maintenance and sustain weight reduction in
Table 7.3 Daily nutritional intake of pre-agricultural vs. overweight and obese individuals [75].
present day Western dietary pattern [65]
Nutritional Current Western 7.3.1.1 Observational Studies
components Pre-agricultural diet diet Observational studies generally support an
Diet energy Low High inverse association between total fiber intake
density from diets rich in whole (minimally processed)
(kcal/g) plant foods and lower body weight, waist size,
Dietary bulk More Less and body and visceral fat (VAT) (Table 7.4) [76–
(satiating)
86]. Adequate fiber intake of >25 g fiber/day or
Sugar and Limited amount of 17%
sweeteners honey 14 g fiber/1000 kcals from whole plant-based
(% energy) diets is a suggested target to reduce risk of weight
Grain Low High gain, prevent risk of obesity, and promote modest
products (all whole grain) (mostly refined) weight loss [68, 75]. The US Women’s Health
Fruit, 65% 8% Study (18,146 women; baseline age ≥ 45 years;
vegetables,
normal baseline BMI; 15.9-years of follow-up)
and nuts
(% energy) found that women with higher fiber intake did not
Fiber Intake 50–100 <15–17 have significant changes in body weight or BMI
(g/day) [76]. A study in 252 women (mean baseline age
Protein 37% from lean 15% from meat, 40 years; 20-months of follow-up) found that
Intake game, eggs, fish, poultry, dairy, each 1 g/1000 kcal increase in total fiber signifi-
(% energy) shellfish, or nuts fish, eggs,
legumes, or nuts
cantly reduced body weight by 0.25 kg and body
Fat Intake 22 32 fat by 0.25%, by reducing total metabolizable
(% energy) energy intake [82]. Specific studies suggest that
Physical Active >1000 kcal/ Sedentary increased total fiber intake by approximately
activity day (<150–490 kcal/ ≥12 g/day, above the typical Western diet fiber
(kcal/day) day) levels, especially as a replacement for refined low
1.1
0.9
Obesity Risk
0.8
0.7
0.6
0.5
<8.1 8.1-12.0 12.0-16.2 16.2-22.4 >22.5
Total fiber intake (g/day)
Fig. 7.11 Relationship between increasing fiber intake and adult obesity risk from the US National Health and Nutrition
Examination (NHANES) Survey 1999–2010) (adapted from [71])
fiber food, can significantly prevent long-term low-fat dietary pattern reduced body weight and
(8–12 years) weight gain by 3.5–5.5 kg in both improved body composition compared to the low
men and women [83, 85, 86]. For WC, studies fiber Western control diets (Table 7.5) [25, 88–
show an inverse association with increased total 93]. Low energy dense diets derived from higher
fiber intake [78–80, 84]. The European consumption of fruits, vegetables, and fiber were
Prospective Investigation into Cancer and shown to limit weight regain 0.3% in subjects
Nutrition (EPIC) study (89,432 participants; with a history of recent weight reduction com-
mean baseline age 53 years; 6.5 years of follow- pared to an increase of 1.3% for those on the
up) found that 10 g/day increase in total and usual diet after 7 months (p = 0.002) [88]. A ran-
cereal fiber reduced waist size by approximately domized trial (240 metabolic syndrome subjects;
1 cm/year [79]. For visceral adipose tissue (VAT), mean baseline age 52 years and BMI 35; 1 year)
several studies show an inverse association with found that a high fiber diet (goal to consume
increased fiber intake, with children appearing to >30 g fiber/day) was as effective as a reduced
be especially responsive to the effects of low energy multicomponent AHA weight loss pro-
fiber, energy dense diets on visceral fat gain [81]. gram after one year [25]. An Australian RCT (72
subjects; mean age 43 years and BMI 34;
7.3.1.2 Randomized Controlled 12-week duration) demonstrated that the intake
Trials (RCT) of 31 g fiber or various combinations of diets
A systematic review of clinical studies found with psyllium significantly reduced body weight,
that increasing fiber intake by 14 g fiber/day in BMI and % body fat compared to a 20-g fiber/
overweight or obese individuals, with ad libitum day control (Fig. 7.12) [91]. The Finnish
energy intake, was associated with a mean 10% Diabetes Prevention Study (522 prediabetic sub-
decrease in energy intake and a reduction of jects; 67% females; mean baseline age 55 years;
weight by 1.9 kg after 4 months [87]. Seven 4-year duration) showed total fiber intake
RCTs found the consumption of adequate fiber (>15.5 g/1000 kcal vs. <11 g/1000 kcal) signifi-
intake ≥28 g fiber/day or ≥14 g fiber/1000 kcal cantly reduced body weight by 2.6 kg (p-trend =
from fiber-rich diets combined with or without a 0.001) and waist size by 1.3 cm (p-trend = 0.033)
Table 7.4 Summary of observational studies on fiber intake in body weight and composition regulation
Rautiainen et al. (2015). 18,146 women, baseline This study found no significant association
Investigate the effect of fiber age ≥ 45 years; BMI 18.5 to <25; with total fiber intake and weight gain or
intake on weight change and mean 15.9 years of follow-up; FFQ having an overweight or obese BMI
the risk of becoming and self-reported body weight on (p-trend = 0.13)
overweight or obese annual questionnaire (multivariate
(Women’s Health Study; adjusted)
US) [76]
Fischer et al. (2015). 583 adults; mean age 61 years; VAT VAT was positively associated with
Cross-sectional assessment volumes from MRI; nutrient intake nutrients characteristic of animal products
of how usual patterns of estimated by a 112-item food- (except for dairy), (β: 0.25; p < 0.0001), but
nutrient intake are frequency questionnaire linked to the negatively with total fiber (β: −0.17;
associated with visceral German Food Code and Nutrient p < 0.0001), and nutrients found in milk.
adipose tissue (VAT), Database; foods, nutrients, or total Subcutaneous abdominal and trunk adipose
subcutaneous, abdominal energy intake associations with tissue were mainly associated with total
and trunk adipose tissue adipose tissue compartments via energy intake
(German adults) [77] multiple linear regression
(multivariate adjusted)
Lin et al. (2011). 3083 individuals (1546 men and 1537 Waist size was inversely related to total
Assess the effect of total women); baseline age ≥ 15 years. fiber intakes (β = −0.118, p < 0.001) and
fiber and sources on BMI 42% of women and 29% of men were positively related to fruit-fiber intakes
and waist size (Belgian) [78] abdominally obese. The main (β = −0.731 (p = 0.001). Intake of cereals
contributors to total fiber intake were and cereal products fiber were significantly
cereals and cereal products (34%), associated with lower BMI (β = −0.045,
potatoes and other tubers (18.6%), p = 0.025), but the association was
fruits (14.7%) and vegetables (14.4%) attenuated by energy intake adjustments
Du et al. (2010). 89,432 participants; mean baseline Higher intake of total fiber, especially cereal
Investigate the association age 53 years; average 6.5 years of fiber, helps to prevent body weight and
of total dietary fiber, cereal follow-up (multivariate adjusted) waist size gain. For a 10 g/day higher total
fiber, and fruit and vegetable fiber intake, the mean weight loss was 39 g/
fiber with changes in body year and waist size loss was 0.08 cm/year.
weight and waist size (EU A 10 g/day higher cereal fiber intake was
EPIC) [79] associated with lower body weight/year by
77 g and lower waist size/year by 0.10 cm.
Fruit and vegetable fiber were not
associated with weight change but had a
similar association with waist size change
when compared with intake of total fiber
and cereal fiber
Romaguera et al. (2010). 48,631 participants; mean baseline In women, an increased fiber intake by 10 g
Assess the association age 50 years; mean BMI 26; median fiber/day significantly reduced waist size by
between dietary factors and 5.5 years of follow-up (multivariate 0.06 cm. Waist size was also significantly
prospective changes in waist adjusted) increased for every 1 kcal/g higher energy
circumference (WC) and density by up to 0.15 cm and for every 10
visceral adiposity (European glycemic index units by up to 0.06 cm
Prospective Investigation
into Cancer and Nutrition
[EPIC] Study) [80]
Davis et al. (2009). 85 overweight Latino youth; aged Reduced fiber intake by 3 g/1000 kcals
Assess the relation between 11–17 years; body composition by significantly increased visceral fat by 21%
changes in dietary intake, dual-energy X-ray absorptiometry vs. an increase in fiber intake of 3 g
and specifically sugar and and magnetic resonance imaging; fiber/1000 kcals, which reduced visceral fat
fiber intakes, with changes 2 years of follow-up (multivariate by 4%
in adiposity and risk factors adjusted)
for type 2 diabetes in a
longitudinal analysis of
overweight Latino youth
(US) [81]

Tucker et al. (2009). 252 women; mean baseline age For each 1 g/1000 kcal increase in fiber
Evaluate the effects of total 40 years; mean weight 65.6 kg; intake there was a significant decrease in
fiber intake on risk of 20 months of follow-up; 7-day food body weight by 0.25 kg and fat by 0.25%.
gaining weight and body fat records (multivariate adjusted) After adjustment for energy intake, there
in women over time (US) was a reduction of about 33% but the values
[82] still retained significance. Fiber’s influence
occurs primarily through reducing energy
intake over time
Koh-Banerjee et al. (2004). 27,000 men; mean baseline age Total fiber intake was inversely related to
Evaluate the associations 52 years; 8 years of follow-up weight gain independent of whole grains
between changes in cereal (multivariate adjusted) (p-trend <0.0001). The men consuming 17 g
fiber intake and weight fiber/day gained 1.40 kg, whereas the men
change (HPFS; US) [83] with 26 g fiber/day gained 0.39 kg. After
adjusting for measurement error, there was
reduced weight gain by 5.5 kg (12 lbs) for
each 20-g/day increment in total fiber intake
Koh-Banerjee et al. (2003). 16,587 men; mean baseline age An increase of 12 g total fiber/day was
Determine the effects of 44–65 years; 9 years of follow-up associated with a 0.63 cm decrease in waist
changes in diet and physical (multivariate adjusted) size (p < 0.001), whereas smoking cessation
activity, on waist size among and a 20-h/week increase in television
men (Health Professionals’ watching were associated with a 1.98 cm
Follow-up Study [HPFS]; and 0.59 cm waist gain, respectively
US) [84] (p < 0.001). Increases of 25 metabolic
equivalent hrs/week in vigorous physical
activity and in weight training by ≥30 min/
week were associated with 0.38-cm and
0.91-cm decreases in waist size,
respectively (p < 0.001 for each
comparison)
Liu et al. (2003). 74,000 female nurses; mean baseline Women consuming a mean total fiber intake
Investigate the associations age 50 years; 12 years of follow-up of 20 g vs. 13 g total fiber/day gained an
between the intakes of fiber (multivariate adjusted) average of 1.52 kg (3.4 lbs) less weight
and whole- or refined-grain (p-trend <0.0001) independent of body
products and weight gain weight at baseline, age, and changes in
over time (Nurses’ Health covariate status; over 2–4 years women
Study; US) [85] gained less weight by 0.76 kg and BMI by
0.28 units. An increase in total fiber intake
by 12 g/day is estimated to reduce weight
gain by 3.5 kg (8 lb) in 12 years. Women in
the highest quintile of total fiber intake had
a 49% lower risk of major weight gain than
women in the lowest quintile (p-trend
<0.0001)
Ludwig et al. (1999). 2909 healthy adults; mean baseline Total fiber intake was significantly inversely
Examine the role of fiber age 26 years; >10.5 g fiber vs. <5.9 g associated with body weight (p = 0.001),
intake on weight gain, fiber/1000 kcal; 10 years of waist-to-hip ratio and fasting insulin;
insulin status and follow-up (multivariate adjusted) adjusted for BMI. Increased total fiber
cardiovascular disease reduced weight gain by 8 lbs., waist-hip
(CVD) risk factors (The ratio by 0.1, and fasting insulin by 0.8–1.4
Coronary Artery Risk μU/mL in young adults. CVD risk factor
Development in Young was also significantly lowered
Adults [CARDIA]; US) [86]
Table 7.5 Summary of randomized controlled trials (RCTs) on fiber-rich diets in body weight and composition
regulation
Karimi et al. (2016). Parallel RCT: Subjects on the low energy dense diets
Assess effects of low 70 subjects with recent history of reduced weight by 0.3% vs. subjects on the
energy density diet vs. a weight reduction; mean age 55 years; usual diet control who gained 1.3% more
usual diet on weight 50% male; low energy dense diet weight (p = 0.002). The results were similar
maintenance, lipid contained 30% fat, 15% protein, and for waist size with a loss of 0.4 cm on the low
profiles, and glycemic 55% carbohydrate (20 g fiber, fruit energy dense diet vs. a gain of 0.3 cm on the
control (Iran) [88] 3.7 servings and vegetables 5.5 usual diet (p = 0.004). Also, the low energy
servings) vs. usual diet including 35% dense diet group decreased fasting blood
fat, 15% protein, and 50% glucose by 9.5% vs. an increase by 0.4% on
carbohydrate (14 g fiber; fruit 2.5 the usual diet (p = 0.0001). These findings
servings and vegetables 3.3 servings); support the beneficial effects of a low energy
dietary intake was assessed by using dense diet derived from higher consumption
3-day food records; 7 months of fruits, vegetables, and fiber on attenuating
weight regain
Ma et al. (2015). Parallel RCT: At 12 months: (1) mean body weight was
Evaluate the effects of a 240 metabolic syndrome subjects; reduced by 2.1 kg in the high-fiber diet group
simple high fiber diet mean baseline age 52 years; mean vs. 2.7 kg in the AHA weight loss program
compared to a BMI 35; goal ≥30 g fiber/day diet or (mean group difference 0.6 kg); (2) mean
multicomponent an AHA weight loss program diet waist size was increased by 0.1 in. for the
American Heart plan including caloric reduction of high fiber diet vs. a loss of 0.4 in. for the
Association (AHA) 500–1000 kcal/day; 1 year AHA weight loss program (mean group
weight loss plan on body difference, 0.5 in.); and (3) mean BMI was
weight, waist size and reduced by 0.8 units for the high fiber diet
BMI (US) [25] and 1.0 units for the AHA weight loss
program (Fig. 7.1). There were no significant
differences in weight loss, BMI, or waist size
between the groups. This study suggests that
simply consuming a high fiber diet may be a
reasonable alternative to a traditional,
challenging, hypocaloric weight loss diet plan
Turner et al. (2013). Parallel RCT: Both fiber-rich food groups increased fiber
Examine the effect of two 20 subjects; mean age 47 years, 18 intake from about 17 g/day to about 29 g
high-fiber hypocaloric females and 2 males; mean BMI 31; fiber/day and lowered energy density by 38%
diets on weight loss (US) high fiber, reduced energy by for the bean group and 29% for the variety of
[89] 300–400 kcal/day diets with either fiber foods group. Both diets significantly
1.5 cups beans/day or a variety of reduced body weight, with the bean diet by
fruits, vegetables, and whole grains; 1.6 kg and the variety of fiber food diet by
25–35 g fiber; 4 weeks 1.1 kg. Combined mean weight loss was
1.4 kg (p < 0.001)
Mecca et al. (2012). Parallel RCT: Subjects on high fiber diet lost 4% more
Investigate the 50 subjects; 11 males and 39 females; weight and BMI by 4%, and waist size by 7%
effectiveness of a high mean age 50 years; mean BMI 33.0; vs. the lower fiber control diet (p < 0.05; all)
fiber lifestyle intervention high-fiber diet group (daily 32 g fiber;
on overweight-obese 540 g fruits and vegetables) vs.
adults (Brazil) [90] control group receiving general
nutrition education (17 g fiber/day);
10 weeks
Pal et al. (2011). Parallel RCT: All fiber rich diets compared to the control
Assess the effects of 72 participants; mean age 43 years; Western diet significantly reduced body
increased fiber intake mean BMI about 34; diets: control weight, BMI, and % body fat after 12 weeks
from a healthy diet, diet plus placebo (20 g fiber/day); (Fig. 7.12)
psyllium or their control diet plus psyllium (55 g fiber/
combinations on body day); healthy fiber-rich food diet plus
weight and composition placebo (31 g fiber/day); or healthy
(Australia) [91] fiber-rich food diet plus psyllium
(59 g fiber/day); 12 weeks
Ferdowsian et al. (2010). Parallel RCT: The higher fiber diet group lost significantly
Study the effects of a high 113 adults; BMI >25; randomized more weight by 5.2 kg and waist size by
fiber, low fat vegan diet on into a low-fat, vegan diet group at 5.5 cm compared to the lower fiber Western
body weight and 29 g fiber/day vs. Western habitual diet control group (p < 0.0001). Weight loss
composition in overweight diet at 15 g fiber/day; 22 weeks of 5% of body weight was more frequently
subjects (US GEICO found for subjects in the high fiber group by
Corporate Site) [92] 49% vs. control group by 11% (p < 0.0001)
Lindstrom et al. (2006). Parallel RCT: Participants consuming the low-fat, high-fiber
Investigate the effect of 522 participants with impaired diet lost significantly 2.4 kg more weight than
total dietary fiber, fat, and glucose tolerance; mean age 55 years; those on the high-fat, low-fiber diet. The fiber
energy density on body 67% female; mean BMI 31; standard density of the diet was inversely associated
weight and WC (Finnish lifestyle vs. high fiber, low-fat diets with weight and waist size (Figs. 7.13 and
Diabetes Prevention and exercise counseling; 15 g fiber vs. 7.14)
Study) [93] 11 g fiber/1000 kcals; 4 years
Weight (kg) BMI (kg/m2) Body fat (%)

0.0
Usual diet Healthy diet Usual diet + psyllium Healthy diet +
(20 g fiber/d) (31 g fiber/d) (57 g fiber/d) psyllium
-0.5 (57 g fiber/d)
-1.0
-1.5
-2.0
-2.5
-3.0
-3.5
Fig. 7.12 Effect of usual diet and healthy diet with and without added psyllium (12 g 3×/day) in 72 obese adults (mean
age 43 years; BMI 34) after 12 weeks (p < 0.05) (adapted from [91])
(Fig. 7.13) after multivariate adjustments [93]. weight loss may include: (1) replacing a low
Also, in this study, the adjusted 3-year weight fiber, high glycemic breakfast cereal with a fiber-
reduction among those whose diets were both rich bran breakfast cereal; (2) eating an apple
low in fat and high in fiber was 3.1 kg compared instead of a cookie at lunch; (3) adding artichokes
to 0.7 kg for subjects on the high fat and low fiber or chickpeas to a salad; and (4) snacking on nuts,
diet (Fig. 7.14). Four RCTs generally show that sunflower seeds, or popcorn instead of potato
the consumption of 29–32 g vs. 15–20 g total chips. A list of 50 of the top dietary fiber contain-
fiber/day significantly reduced body weight, BMI ing foods are listed in Appendix B.
and/or waist size over 4–22 weeks [89–92].
The daily substitution of a fiber-rich food for a
lower fiber, energy dense food item at each meal 7.3.2 Isolated Fiber Ingredients
and one snack is one approach to changing from
a Western diet (15–17 g of daily fiber) to a healthy Three systematic reviews and 4 specific RCTs on
weight controlling diet with ≥30 g fiber/day. the effects of isolated fiber ingredients on weight
Examples of potential food switches to achieve regulation are summarized in Table 7.6 [91, 94–
≥30 g fiber/day and lower energy density needed 99]. These RCTs showed that isolated fiber ingre-
to prevent weight gain or to promote or maintain dients were very heterogeneous and generally less
Body weight (kg) p-trend =.001 Waist size (cm) p-trend =.033
0
< 11 11 to 13 13 to 15.5 > 15.5
Total fiber density (g/1,000 kcal)
-0.5
-1
-1.5
-2
-2.5
-3
-3.5
Fig. 7.13 Effect of dietary fiber density on body weight and waist size in 522 overweight/obese pre-diabetic adults over
3 years from the Finnish Diabetes Prevention Study (adapted from [93])
Lower Energy Density Higher Energy Density

0
Low-fat/high fiber Low fat/low fiber High fat/high fiber High fat/low fiber
-0.5
-1
Weight Change (kg)
-1.5
-2
-2.5
-3
-3.5
Fig. 7.14 Effect of dietary energy density on weight loss in 522 overweight/obese pre-diabetic adults over 3 years from
the Finnish Diabetes Prevention Study (adapted from [93])
effective than whole foods and fiber-rich diets. In with 250 mL water, 3 times daily 5–10 min before
a systematic review of 66 randomized trials that breakfast, lunch, and dinner to a Western diet
examined the effects of isolated fibers, the overall (20 g fiber/day) significantly lowered body
weight reduction was a modest 0.1 kg per 10 g weight, and % body fat (p < 0.05) with an effect
fiber after 4 weeks with a high degree of variabil- similar to consuming a plant-based healthy diet
ity [95]. Two systematic reviews and clinical with 31 g fiber/day (Fig. 7.12) [91]. During the
studies of guar gum and inulin-type fructans consumption of ad libitum diets, the addition of
showed insignificant or inconsistent weight loss 27 g/day of fermentable, soluble gel-forming/
effects [94, 96]. Adding 12 g psyllium fiber mixed viscous fibers (pectin and β-glucan) and
Table 7.6 Summary of randomized controlled trials (RCTs) on fiber supplements and ingredients in body weight
regulation
Systematic reviews and meta-analysis
Liber et al. (2013). For the pediatric population; 4 RCTs; Very inconsistent effects on weight loss with
Systematically evaluate 232 children. For the adult limited data suggest that long-term use of
the effects of inulin-type population; 15 RCTs; 545 subjects inulin type fructans may contribute to weight
fructan supplementation reduction
on appetite, energy
intake, and body weight
(BW) in children and
adults [94]
Wanders et al. (2011). 58 RCTs appetite control; 26 RCTs Overall, effects on energy intake and body
Systematically investigate on acute energy intake, 38 RCTs on weight were relatively small, and distinct
fiber types on appetite, long-term energy intake; and 66 dose-response relationships were not
energy intake and body RCTs on body weight observed. Short- and long-term effects of fiber
weight (systematic appear to differ and have multiple
review) [95] mechanisms relating to their different
physicochemical properties. Fibers
characterized as being more viscous (e.g.
pectins, β-glucans and guar gum) reduced
appetite or energy intake more often than less
viscous fibers
Pittler and Ernst (2001). 11 RCTs; 203 subjects; 7.5–20 g guar There was a non-significant mean body
Determine the efficacy of gum/day; 4 weeks to 6 months weight difference in subjects receiving guar
guar gum as a therapeutic gum compared with those receiving placebo
option for reducing body by 0.04 kg
weight (meta-analysis)
[96]
Specific RCTs
Hu et al. Parallel RCT: Soy fiber supplemented breakfast biscuits
Examine the effects of 39 college students; mean age significantly reduced body weight by 0.7 kg
soy fiber on body weight 23 years; mean BMI 26; biscuits and BMI 0.44 vs. a control biscuit (p < 0.05)
and body composition in supplemented with 27.5 g soy fiber/
overweight and obese day for breakfast vs. control low fiber
participants (China) [97] biscuits; 12 weeks
Pal et al. (2011). Parallel RCT: Adding 12 g psyllium fiber 3 times/day to a
Compare the effects of 72 participants; mean age 43 years; Western diet (20 g fiber/day) significantly
fiber intake from a mean BMI 34; diets: control diet plus lowers body weight, and % body fat
healthy diet vs. a control placebo (20 g fiber/day); control diet (p < 0.05) with a similar effect to a 31 g fiber/
diet plus psyllium or a plus 12 g psyllium fiber 3 times/day; day healthy diet (Fig. 7.12)
healthy diet plus psyllium healthy fiber-rich food diet (31 g
on body composition fiber/day), or healthy fiber-rich food
(Australia) [91] diet plus 12 g psyllium fiber 3 times/
day); 12 weeks
Salas-Salvado et al. Double-blind, Parallel RCT: Weight loss was higher
(2008). 200 overweight or obese subjects; after both doses of fiber by about 4.6 kg vs.
Compare the effect of the mean age 48 years; mean BMI 31; 3.8 kg for placebo (p = 0.43). Also, post-
administration of a 78% female; reduced calorie diet prandial satiety increased in both fiber groups
mixture of fibers on body with 3 g psyllium and 1 g vs. placebo
weight-loss, satiety glucomannan either twice or three
(Spain) [98] times daily vs. placebo; 16 weeks
(continued)
Howarth et al. (2003). Crossover RCT: This study showed a limited role for
Evaluate the short-term 11 subjects; age 23–46 years; BMI short-term use of high fermentable fiber and
effects of high vs. low 20–34; 27 g/day of fermentable fiber low fermentable fiber supplements in
fermentable fiber (pectin and β-glucan) and non- promoting weight loss in humans consuming
supplements on hunger, fermentable fiber (methylcellulose); an ad libitum diet. Over 3 weeks, fiber
energy intake, and weight ad libitum diets; 3 weeks; 4 weeks of supplements insignificantly reduced energy
loss (US) [99] washout; daily fiber supplements intake for high fermentable fiber by 7%
were divided into approximately (p = 0.31) and low fermentable fiber by 9.5%
three 10-g portions to be eaten (p = 0.11); body weight by 0.13 kg (p > 0.05),
30 min before each meal with and % body fat for low fermentable fiber by
355 mL of a non-caloric liquid, to 0.3% (p = 0.56); high fermentable fiber by
achieve a maximum effect as a 0.1% (p = 0.66)
preload
non-fermentable fiber (methylcellulose) con- 7.4 Fiber Biological Mechanisms

sumed in a non-caloric beverage 30 min before
each meal was shown to insignificantly reduce Postulated biological mechanisms associated with
body weight by 0.13 kg for both fermentable and adequate fiber intake (> 25 g/day; 14 g
nonfermentable fibers and reduce body fat for the fiber/1000 kcal) and healthy dietary patterns for
fermentable fibers by 0.1% and for the non-fer- the prevention of weight gain and the promotion of
mentable fiber by 0.3% after 3 weeks [99]. A reduction in body weight, waist size, body and vis-
Chinese study (39 college students; mean age ceral fat are summarized in Fig. 7.15 [100–129].
23 years and BMI 26; 12 weeks) found that bis-
cuits supplemented with 27.5 g isolated soy fiber
significantly reduced body weight by 0.7 kg and 7.4.1 Energy Density
BMI by 0.44 vs. a control biscuit (p < 0.05) [97].
A Spanish double-blind, placebo controlled trial Lower energy dense fiber-rich dietary patterns, as
with fiber supplemented hypocaloric diets (200 replacements for higher energy dense Western
overweight or obese subjects; mean age 48 years patterns, help to promote balanced energy intake
and BMI 31; 78% women; 16 weeks) showed that to prevent weight gain or provide negative energy
a relatively low supplementation blend of 4 g balance to help promote weight loss depending
psyllium and glucomannan 2 or 3 times/day insig- on the level of dietary fiber intake [8, 33, 42, 43,
nificantly reduced body weight 4.6 kg compared 49, 50, 98]. This is because: (1) fiber is generally
to 3.8 kg for the placebo (p = 0.43), suggesting considered to be 2 kcals/g or less as compared to
that higher levels of fiber supplementation may be 4 kcals/g for digestible carbohydrates such as
required for significant additional weight loss sugar and starch as fiber is not digested in the
[98]. These trials suggest that: (1) in ab libitum small bowel and (2) lower energy dense fiber-rich
diets most fiber supplements can be effective in foods displace higher density foods [100–103].
weight control/loss if used in high amounts when The lower fiber energy density results from most
accompanied by water or noncaloric beverage fiber sources being fermented by colonic bacteria
prior to each meal or added to common foods as to varying degrees into SCFA metabolites and
replacements for high energy dense, low fiber gases (carbon dioxide, hydrogen, and methane)
foods to stimulate enough physical bulking and and/or the excretion of undigested fiber in the
other mechanisms to be as effective as fiber-rich stool [100–105]. A 2015 longitudinal analysis
whole foods [91, 97, 99]; and (2) in hypocaloric (2037 obese participants; mean age 47 years; 10
diets the addition of multiple doses of >4 g vis- years follow-up) found that the intake of energy
cous soluble fiber in noncaloric beverages at each dense and low fiber diets (1.8 kcal/g and 8 g
meal may boost the weight loss effects [98]. fiber/1,000 kcal for men and 1.7 kcal/g and
7.4 Fiber Biological Mechanisms 223
Adequate Fiber Intake
Energy Density:
Lowers energy density; fiber (2 kcals/g) vs refined
carbohydrates (4 kcals/g)
Post-prandial Satiety Signaling:

Increases food volume, bulk, or viscosity
Prolongs chewing time to slow eating rate
Slows gastric emptying and reduces hunger
Circulatory System:
Attenuates blood glucose, insulin and C-reactive protein
Promotes insulin sensitivity
Promotes satiety hormones such as cholecystokinin(CCK),

glucagon-like peptide-1(GLP-1) and peptide YY
Colon Fermentation and Microbiotia:

Fosters healthier colonic microbiota and higher colon short
chain fatty acids levels to promote satiety and leaner energy
metabolism associated with a leaner phenotype.
Net Metabolizable Energy:

Higher macronutrient fecal excretion (e.g. dietary fat) for
lower net metabolizable energy
Lowers Risk of Weight Gain and Obesity

Reduces Risk of Abdominal and Visceral Fat
Promotes Weight Loss
Fig. 7.15 Fiber and healthy dietary pattern mechanisms associated with body weight and composition regulatory
control [100–129]
9 g fiber/1,000 kcal for women) was associated increase intraluminal concentration or viscosity,
with significant increased weight gain by 1.7 kg, slow gastric emptying, and create a mechanical
waist size by 1.5 cm and BMI by 0.6 kg/m2 along barrier to enzymatic digestion of macronutrients
with increased cardiometabolic risk factors [103]. such as starch in the small intestine.
7.4.2 Eating and Digestion Rates 7.4.3 Postprandial Satiety Signaling
Fiber-rich meals tend to be more mouth filling and Increased fiber intake has been shown to trigger a
harder to swallow because of their higher bulk, number of hormonal satiety inducing activities
physical density, volume, or viscosity compared [109, 110]. High fiber meals or β-glucan and
with energy-matched, low-fiber meals, and more psyllium supplements compared to energy
rapidly reduce hunger after ingestion [100, 105]. matched low fiber control diets can: (1) decrease
Fiber-rich foods or clinically proven fiber supple- plasma ghrelin, a stomach hunger promoting hor-
ments, especially bulky, viscous soluble fibers, mone, and slow the rate of postprandial increases
in glucose and insulin blood levels to prevent fat obesity because of increased viscosity due to
reactive hypoglycemia known to promote hunger high β-glucan content slowing gastric emptying,
[100, 105–110]; (2) trigger the increased secretion digestion, and absorption which is associated with
of the hormone cholecystokinin (CCK), a brain postprandial low-glycemic index and low-insulin
neuropeptide known to decrease food intake, from responses that promote favorable effects on vari-
the proximal small intestine to slow gastric empty- ous actions of insulin such as facilitating the
ing and increase satiety [106, 111, 112]; and/or (3) induction of glycogen synthesis, fatty acid synthe-
delay the absorption of nutrients long enough to sis, and the esterification of fatty acids to actively
deliver a portion of them to the distal ileum, where reduce elevated VFA [116].
they are not normally present, to stimulate the
release of a cascade of metabolic responses called
the “ileal brake” phenomenon including the 7.4.4 Colonic Effects
release of satiety hormones glucagon-like pep-
tide-1 (GLP-1), known to control appetite, which 7.4.4.1 Microbiota
slows gastric emptying and small bowel transit, The effects of a healthy diet with adequate fiber
decreases glucagon secretion, increases pancreatic from whole plant foods or prebiotics are increas-
β-cell growth, and improves insulin sensitivity ingly being shown to have important roles in mod-
[113, 114] and increased peptide YY (PYY) ulating the composition and metabolic function of
known to reduce appetite by further slowing gas- the colonic microbial communities to help
tric emptying [106]. Soluble viscose fiber has improve weight regulation and prevent obesity
been associated with an accelerated reduction in relative to a lower fiber, Western diet. A 2017
elevated visceral fat tissue (VAT) [115, 116]. The cross-sectional and longitudinal analysis of the
Insulin Resistance and Atherosclerosis Family Twins UK Study (1632 healthy females; mean
Study (339 African Americans and 775 Hispanic baseline age 50 years; 9 years of follow-up) found
Americans; mean age approx. 43 years; 5-year that high colonic microbiome bacterial diversity
follow-up) found that soluble fiber intake and par- and high-fiber intake are correlated with lower
ticipation in vigorous activity were inversely weight gain in women independently of calorie
related to change in VAT, independent of change intake and other confounders [117]. Women who
in BMI [115]. For each 10-g increase in soluble gained weight had a significantly lower colonic
fiber, rate of VAT accumulation decreased by bacterial diversity and lower fiber intake as well as
3.7% (p = 0.01). Soluble fiber was not associated a higher relative concentration of Bacteroides,
with change in subcutaneous fat (0.2%, p = 0.82). which was strongly and negatively correlated with
Active participants had a 7.4% decrease in rate of lower microbiome diversity. Among the bacteria
VAT accumulation versus less active participants associated with lower risk of weight gain are
(p = 0.003). A double-blind RCT (100 subjects; Clostridiales, especially those in the
age range 30-70 years; mean BMI approx 28; Ruminococcaceae family. An analysis of herita-
mean visceral fat area 125 cm2; rice control vs. bility of colonic microbiome diversity estimates
rice with high β-glucan barley (4.4 g/d) or that 60% of the variation is dependent on lifestyle
β-glucan–free barley; 12 weeks) found that VFA and other environmental variables and not deter-
was significantly reduced in both the high β-glucan mined by genetic make-up. A 2015 USA analysis
barley by 10.7 cm2 and the placebo by groups 6.8 of a double-blind RCT (21 healthy men; mean age
cm2, which was insignificantly between the groups 27.5 years; mean BMI 27; 21 g/day polydextrose
[116]. However, a subgroup analysis of subjects or soluble corn fiber; 21 days) showed that
with high baseline VFA (> 100 cm2) showed a sig- increased fiber intake induced changes in the
nificant decrease in the β-glucan barley group colonic microbiome of healthy adults [18]. A shift
VFA by 10 cm2 more than those in the placebo in the Bacteroidetes: Firmicutes ratio was observed
group. This finding indicated that daily consump- when participants consumed soluble fiber, and
tion of an amount of barley providing 4.4 g of changes in bacterial populations were associated
β-glucan reduced VFA in individuals with visceral with shifts in the bacterial metagenome. A 2017
7.4 Fiber Biological Mechanisms 225
Canadian double-blind RCT (42 children; mean Conclusions

age 10 years; mean BMI 26.5; 8 g/d oligofructose- Weight regulation is a complex, multifactorial
enriched inulin vs. maltodextrin/d; 16 weeks) process involving energy intake and energy
found that oligofructose-enriched inulin signifi- expenditure, which are affected by genetics,
cantly decreased body weight by 3.1%, body fat dietary pattern, and other lifestyle and emo-
by 2.4%, and trunk fat by 3.8% compared with tional factors. The current global obesity pan-
children in the placebo group who had a slight demic is in large part a result of increased
increase in these parameters [119]. Additionally, exposure to higher energy dense and lower
children consuming oligofructose-enriched inulin fiber diets and increasingly sedentary Western
had a significantly 35% lower interleukin (IL)-6 lifestyles over the last several decades, which
reduction from baseline and 16S rRNA sequenc- have led to net habitual positive energy bal-
ing revealed significant increases in species of the ances and weight gain. Even a small daily
genus Bifidobacterium and decreases in positive energy balance of 50 kcals/day, by
Bacteroides vulgatus vs. the placebo group. increased energy intake and/or reduced activ-
ity, can lead to an annual weight gain of 0.4–
7.4.4.2 Metabolizable Energy 0.9 kg/year. The human gastrointestinal and
Compared to low fiber foods, fiber-rich foods energy metabolism regulatory systems
tend to decrease the efficiency of macronutrient evolved with pre-agricultural high fiber diets.
bioavailability, especially that of dietary fat, lead- Prospective cohort studies and RCTs show
ing to higher fecal macronutrient excretion [117]. that high adherence to healthy fiber-rich
The consumption of >25 g fiber/day can lead to dietary patterns such as the MedDiet, DASH,
the excretion of 3–4% of macronutrient energy in New Nordic, and vegetarian diets may at a
the feces, which is equivalent to 80 kcals in a minimum help to prevent weight gain and can
2000-kcal diet [120–122]. support weight loss and lower waist size com-
pared to low-fat or Western diets in over-
7.4.4.3 Satiety and Energy Metabolism weight or obese individuals. Mechanisms
SCFAs are involved in the crosstalk existing associated with healthy fiber-rich dietary pat-
between microbes, human appetite and energy tern effects on managing body weight and
regulation [123–129]. Fiber fermentation pro- central obesity include: (1) reducing dietary
duces SCFAs, 95% of which consist of acetate, energy density directly or displacing higher
propionate, and butyrate in a molar ratio of energy foods associated with the Western diet
60:20:20. It has been estimated that as much as 50 pattern; (2) lowering available metabolizable
to 70% of the fiber from mixed diets is ferment- energy; and (3) increasing postprandial satiety
able depending on physical properties [105, 106]. by affecting both the upper digestive tract and
SCFAs can contribute to energy homeostasis and colonic microbiota. Fiber intake is inversely
satiety by affecting multiple cellular metabolic associated with obesity risk and populations
pathways and receptor-mediated mechanisms with higher fiber diets tend to be leaner than
[123–127]. Butyrate reduces systemic inflamma- those with low fiber diets. Prospective cohort
tion, improves insulin sensitivity, and possibly studies suggest that increased total fiber intake
increases energy expenditure [128]. In obese sub- by ≥12 g/day to a total daily fiber intake of
jects, propionate appears to increase the release of >25 g, compared to refined low fiber diets, can
postprandial plasma PYY and GLP-1 from prevent weight gain by 3.5–5.5 kg each
colonic cells to help reduce energy intake [129]. decade. RCTs show that adequate fiber intake
A 24-week study indicated that colonic generated ≥28 g fiber/day from fiber-rich diets can
propionate entering the circulatory system helped reduce body weight and waist circumference
to reduce body weight gain and significantly compared to low fiber Western diets (≤20 g
reduce intra-abdominal fat accretion and intrahe- fiber/day). Fiber-rich diets are generally more
patocellular lipid content in overweight adults effective at promoting weight loss than fiber
with non-alcoholic fatty liver disease [129]. supplements.
Appendix A
Comparison of Common Dietary Patterns per 2000 kcals (Approximated Values)

Healthy Healthy
Western dietary USDA base DASH diet Mediterranean vegetarian pattern
Components pattern (US) pattern pattern pattern (Lact-ovo based) Vegan pattern
Emphasizes Refined grains, Vegetables, Potassium rich Whole grains, Vegetables, fruit, Plant foods:
low fiber foods, fruit, whole- vegetables, fruits,vegetables, whole-grains, vegetables,
red meats, sweets grain, and and low fat milk fruit, dairy legumes, nuts, fruits, whole
and solid fats low-fat milk products products, olive seeds, milk grains, nuts,
oil, and products, and soy seeds, and soy
moderate wine foods foods
sugar sweetened lean meat, fish, poultry, fish, nuts, seeds, and milk alternatives, milk
beverages, and nuts, seeds, and and seeds pulses and vegetable oils alternatives
fast foods vegetable oils
vegetables, and added sugars sweets and refined grains, meats, or fish; products
whole-grains sugar-sweetened and sweets limited sweets
beverages
(% Total kcal)
Protein (% 16 17 18 16 14 13
Total kcal)
Total fat (% 33 32 27 34 32 30
Total kcal)
Saturated fat 11 8 6 8 8 7
(% Total kcal)
Unsat. fat (% 22 25 21 24 26 25
Total kcal)
Fiber (g) 16 31 29+ 31 35+ 40+
Potassium (mg) 2800 3350 4400 3350 3300 3650
Vegetable 19 27 25 27 19–27 18–27
oils (g)
Solid fats (g) 31 18 − 17 21 16
Sodium (mg) 3600 1790 1100 1690 1400 1225
Added sugar (g) 79 (20 tsp) 32 (8 tsp) 12 (3 tsp) 32 (8 tsp) 32 (8 tsp) 32 (8 tsp)
Plant food groups
Fruit (cup) ≤1.0 2.0 2.5 2.5 2.0 2.0
Vegetables ≤1.5 2.5 2.1 2.5 2.5 2.5
(cup)
Whole-grains 0.6 3.0 4.0 3.0 3.0 3.0
(oz.)
Legumes (oz.) − 1.5 0.5 1.5 3.0 3.0+
Nuts/Seeds 0.5 0.6 1.0 0.6 1.0 2.0
(oz.)
Soy products 0.0 0.5 − − 1.1 1.5
(oz.)
Svetkey LP, Simons-Morton D, Vollmer WM, et al. Effects of dietary patterns on blood pressure. Arch Intern Med.
1999;159:285–93
(1) Dietary Guidelines Advisory Committee. Scientific Report of the 2010 Advisory Guidelines Advisory Report to the
Secretary of Health and Human Services and the Secretary of Agriculture. Part B. Section 2: Total Diet. 2010; Table
B2.4. (2) Dietary Guidelines Advisory Committee. Scientific Report of the 2015 Advisory Guidelines Advisory Report
to the Secretary of Health and Human Services and the Secretary of Agriculture. Appendix E-3.7: Developing Vegetarian
and Mediterranean-style Food Patterns. 2015;1–9
Appendix B 227
Appendix B
Fifty High Fiber Foods Ranked by Amount of Fiber per Standard Food Portiona
Energy density
Food Standard portion size Dietary fiber (g) Calories (kcal) (calories/g)
High fiber bran 1/3–3/4 cup (30 g) 9.1–14.3 60–80 2.0–2.6
ready-to-eat-cereal
Shredded wheat ready-to-eat 1–1 1/4 cup (50–60 g) 5.0–9.0 155–220 3.2–3.7
cereal
Chickpeas (Garbanzo) beans, 1/2 cup (120 g) 8.1 176 1.4
canned
Artichoke, global or French, 1/2 cup (84 g) 7.2 45 0.5
cooked
Wheat bran flakes ready-to-eat 3/4 cup (30 g) 4.9–5.5 90–98 3.0–3.3
cereal
Green peas, cooked (fresh, 1/2 cup (80 g) 3.5–4.4 59–67 0.7–0.8
frozen, canned)
Mixed vegetables, cooked from 1/2 cup (45 g) 4.0 59 1.3
frozen
Energy density
Food Standard portion size Dietary fiber (g) Calories (kcal) (calories/g)
Sunflower seed kernels, dry 1 ounce (about 28 g) 3.1 165 5.8
roasted
Orange 1 medium (130 g) 3.1 69 0.5
Banana 1 medium (118 g) 3.1 105 0.9
Dates 1/4 cup (about 38 g) 2.9 104 2.8
Dietary Guidelines Advisory Committee. Scientific Report of the 2015 Advisory Guidelines Advisory Report to the
Secretary of Health and Human Services and the Secretary of Agriculture Part D. Chapter 1: Food and Nutrient Intakes,
and Health: Current Status and Trends. 2015; 97, 98. Table D1.8
a
USDA National Nutrient Database for Standard Reference, Release 27. http://www.ars.usda.gov/nutrientdata. Accessed
17 Feb 2015
8. Davis JN, Hodges VA, Gillham MB. Normal-weight

References adults consume more fiber and fruit than their age
and height matched overweight/obese counterparts.
1. World Health Organization. Obesity and overweight. J Am Diet Assoc. 2006;106:835–40.
Geneva. 2014. www.who.int/mediacentre/fact- 9. Mozaffarian D, Hao T, Rimm EB, et al. Changes
sheets/fs311en/. Accessed 18 Jan 2015. in diet and lifestyle and long-term weight gain
2. Swinburn BA, Sacks G, Hall KD, et al. The global in women and men. N Engl J Med. 2011;363:
obesity pandemic: shaped by global drivers and local 2392–404.
environments. Lancet. 2011;378:804–14. 10. Rolls BJ. What is the role of portion in weight man-
3. Moehlecke M, Canani LH, Lucas Oliveira L, et al. agement? Int J Obes. 2014;38:S1–8.
Determinants of body weight regulation in humans. 11. Vernarelli JA, Mitchell DC, Rolls BJ, Hartman
Arch Endocrinol Metab. 2016;60(2):152–62. TJ. Dietary energy density is associated with obe-
4. Centers for Disease Control and Prevention. sity and other biomarkers of chronic disease in US
Overweight and obesity: causes and consequences. adults. Eur J Nutr. 2015;54(1):59–65.
2012. http://www.cdc.gov/obesity/adult/causes/index. 12. Karl JP, Roberts SB. Energy density, energy intake
html. Accessed 21 Feb 2015. and body weight regulations in adults. Adv Nutr.
5. Hill JO. Can a small-changes approach help 2014;5:835–50.
address the obesity epidemic? A report of the Joint 13. Raynor HA, Jeffery RW, Phelan S, et al. Amount
Task Force of the American Society for Nutrition, of food group variety consumed in the diet and
Institute of Food Technologists, and International long-term weight loss maintenance. Obes Res.
Food Information Council. Am J Clin Nutr. 2005;13(5):883–90.
2009;89:477–84. 14. Dietary Guidelines Advisory Committee. Scientific
6. Zhai F, Wang H, Wang Z, et al. Closing the energy Report of the 2015 Advisory Guidelines Advisory
gap to prevent weight gain in China. Obes Rev. Report to the Secretary of Health and Human
2008;9(Suppl 1):107–12. Services and the Secretary of Agriculture. Part D.
7. Centers for Disease Control and Prevention. Low Chapter 2: Dietary patterns, foods and nutrients, and
energy dense foods and weight management: cut- health outcomes. 2015;1–33.
ting calories while controlling hunger. Research to 15. Jessri M, Wolfinger RD, Lou NY, L’Abbe MR.
Practice Series, No 5. 2015. http://www.cdc.gov/ Identification of dietary patterns associated with
nccdphp/dnpa/nutrition/pdf/r2p_energy_density. obesity in a nationally respresentative survey of
pdf. Accessed 21 Feb. Canadian adults: application of a priori, hybrid,
References 229
and simplified pattern techniques. Am J Clin Nutr. 31. Lassale C, Fezeu L, Andreeva VA, et al. Association
2017;105:669–84. between dietary scores and 13-year weight change
16. Verheggen RJ, Maessen MF, Green DJ, et al. A and obesity risk in a French prospective cohort. Int J
systematic review and meta-analysis on the effects Obes. 2012;36(11):1455–62.
of exercise training versus hypocaloric diet: dis- 32. Wolongevicz DM, Zhu L, Pencina MJ, et al. Diet
tinct effects on bodyweight and visceral adipose quality and obesity in women: the Framingham
tissue. Obes Rev. 2016;17(8):664–90. https://doi. Nutrition Studies. Br J Nutr. 2010;103(8):1223–9.
org/10.1111/obr.12406. 33. Esmaillzadeh A, Azadbakht L. Major dietary pat-
17. de Mutsert R, Sun Q, Willett WC, et al. Overweight terns in relation to general obesity and central adipos-
in early adulthood, adult weight change, and risk of ity among Iranian women. J Nutr. 2008;138:358–63.
type 2 diabetes, cardiovascular diseases, and certain 34. Schulz M, Nothlings U, Hoffmann K, et al.
cancers in men: a cohort study. Am J Epidemiol. Identification of a food pattern characterized by
2014;179:1353–65. high-fiber and low-fat food choices associated with
18. Wilson PW, D’Agostino RB, Sullivan L, et al. low prospective weight change in the EPIC-Potsdam
Overweight and obesity as determinants of cardio- cohort. J Nutr. 2005;135:1183–9.
vascular risk: the Framingham experience. Arch 35. Newby PK, Muller D, Hallfrisch J, et al. Dietary
Intern Med. 2002;162:1867–72. patterns and changes in body mass index and
19. Wang YC, McPherson K, Marsh T, et al. Health and waist circumference in adults. Am J Clin Nutr.
economic burden of projected obesity trends in the 2003;77:1417–25.
USA and the UK. Lancet. 2011;378:815–25. 36. Li Y, Roswall N, Ström P, et al. Mediterranean and
20. Wing RR, Phelan S. Long-term weight loss mainte- Nordic diet scores and long-term changes in body
nance. Am J Clin Nutr. 2005;82(Suppl):222S–5S. weight and waist circumference: results from a large
21. Raynor HA, Van Walleghen EL, Bachman cohort study. Br J Nutr. 2015;114:2093–102.
JL. Dietary energy density and successful weight 37. Funtikova AN, Benıtez-Arciniega AA, Gomez SF,
loss maintenance. Eat Behav. 2011;12(2):119–25. et al. Mediterranean diet impact on changes in abdom-
22. MacLean PS, Higgins JA, Giles ED, et al. The role inal fat and 10-year incidence of abdominal obesity
for adipose tissue in weight regain after weight loss. in a Spanish population. Br J Nutr. 2014;111:1481–7.
Obes Rev. 2015;16(Suppl.1):45–54. https://doi.org/10.1017/S0007114513003966.
23. Mariman ECM. An adipobiological model for 38. May AM, Romaguera D, Travier N, et al. Combined
weight regain after weight loss. Adipobiology. 2011; impact of lifestyle factors on prospective change
3:7–13. in body weight and waist circumference in partici-
24. Rezagholizadeh F, Djafarian K, Khosravis S, Shab- pants of the EPIC-PANACEA Study. PLoS One.
Bidar S. A posteriori: healthy dietary patterns may 2012;7(11):e50712. https://doi.org/10.1371/journal.
decrease the risk of central obesity: findings froma pone.0050712.
systematic review and meta-analysis. Nutr Res. 39. Beunza JJ, Toledo E, Hu FB, et al. Adherence to the
2017;41:1-13. Mediterranean diet, long-term weight change, and
25. Ma Y, Olendzki BC, Wang J, et al. Single-component incident overweight or obesity: The Seguimiento
versus multi-component dietary goals for the meta- Universidad de Navarra (SUN) cohort. Am J Clin
bolic syndrome: a randomized trial. Ann Intern Med. Nutr. 2010;92:1484–93.
2015;162:248–57. 40. Romaguera D, Norat T, Vergnaud A-C, et al.
26. Asghari G, Mirmiran P, Yuzbashian E, Azizi F. A Mediterranean dietary patterns and prospective
systematic review of diet quality indices in relation weight change in participants of the EPIC-PANACEA
to obesity. Br J Nutr. 2017;117(8):1055–65. https:// project. Am J Clin Nutr. 2010;92:912–21.
doi.org/10.1017/S0007114517000915. 41. Sanchez-Villegas A, Bes-Rastrollo M, Martinez-
27. Shah RV, Murthy VL, Allison JP, et al. Diet and adi- Gonzalez MA, Serra-Majem L. Adherence to a
pose tissue distributions: The Multi-Ethnic Study Mediterranean dietary pattern and weight gain
of Atherosclerosis. Nutr Metab Cardiovasc Dis. in a follow-up study: the SUN cohort. Int J Obes.
2016;26:185–93. 2006;30:350–8.
28. Hu T, Jacobs DR, Larson NI, et al. Higher diet quality 42. Mendez MA, Popkin BM, Jakszyn P, et al.
in adolescence and dietary improvements are related Adherence to a Mediterranean diet is associated
to less weight gain during the transition from adoles- with reduced 3-year incidence of obesity. J Nutr.
cence to adulthood. J Pediatr. 2016;178:188–93. 2006;136:2934–8.
29. Feliciano CEM, Tinker L, Manson JE, et al. Change 43. Barak F, Falahi E, Keshteli AH, et al. Adherence to the
in dietary patterns and change in waist circumference Dietary Approaches to Stop Hypertension (DASH)
and DXA trunk fat among postmenopausal women. diet in relation to obesity among Iranian female
Obesity. 2016;24:2176–84. https://doi.org/10.1002/ nurses. Public Health Nutr. 2014;18(4):705–12.
oby.21589. 44. Berz JPB, Singer MR, Guo X, et al. Use of a
30. Fung TT, Pan A, Hou T, et al. Long-term change in DASH food group score to predict excess weight
diet quality is associated with body weight change gain in adolescent girls in the National Growth
in men and women. J Nutr. 2015;145(8):1850–6. and Health Study. Arch Pediatr Adolesc Med.
https://doi.org/10.3945/jn.114.208785. 2011;165(6):540–6.
45. Tonstad S, Butler T, Yan R, Fraser GE. Type of veg- 58. Bertz F, Winkvist A, Brekke HK. Sustainable weight
etarian diet, body weight, and prevalence of type 2 loss among overweight and obese lactating women
diabetes. Diabetes Care. 2009;32:791–6. is achieved with an energy-reduced diet in line with
46. Berkow SE, Barnard N. Vegetarian diets and weight dietary recommendations: results from the LEVA
status. Nutr Rev. 2006;64(4):175–88. randomized controlled trial. J Acad Nutr Diet.
47. Mancini JG, Filion KB, Atallah R, Eisenberg 2015;115:78–86.
MJ. Systematic review of the Mediterranean 59. Bertz F, Brekke HK, Ellegard L, et al. Diet and exer-
diet for long-term weight loss. Am J Med. cise weight-loss trial in lactating overweight and
2016;129:407–15. obese women. Am J Clin Nutr. 2012;96:698–05.
48. Huo R, Du T, Xu Y, et al. Effects of Mediterranean- 60. Poulsen SK, Due A, Jordy AB, et al. Health effect of
style diet on glycemic control, weight loss and the New Nordic Diet in adults with increased waist
cardiovascular risk factors among type 2 diabe- circumference: a 6-mo randomized controlled trial.
tes individuals: a meta-analysis. Eur J Clin Nutr. Am J Clin Nutr. 2014;99:35–45.
2014;4(11):e005497. https://doi.org/10.1136/ 61. Huang R-Y, Huang C-C, Hu FB, Chavarro
bmjopen-2014-005497. JE. Vegetarian diets and weight reduction: a meta-
49. Esposito K, Kastorini CM, Panagiotakos DB, analysis of randomized controlled trials. J Gen Intern
Giugliano D. Mediterranean diet and weight loss Med. 2016;31(1):109–16. https://doi.org/10.1007/
diet: meta-analysis of randomized controlled trials. s11606-015-3390-7.
Metab Syndr Relat Disord. 2011;9(1):1–12. 62. Barnard ND, Levin SM, Yokoyama Y. Systematic
50. Kastorini C-M, Milionis HJ, Esposito K, et al. The review and meta-analysis of changes in body weight
effect of Mediterranean diet on metabolic syndrome in clinical trials of vegetarian diets. J Acad Nutr
and its components: a meta-analysis of 50 stud- Diet. 2015;115(6):954–69.
ies and 534,906 individuals. J Am Coll Cardiol. 63. Turner-McGrievy GM, Davidson CR, Wingard EE,
2011;57(11):1299–313. et al. Comparative effectiveness of plant-based diets
51. Estruch R, Martinez-Gonzalez MA, Corella D, et al. for weight loss: a randomized controlled trial of five
Effect of a high-fat Mediterranean diet on body- different diets. Nutrition. 2015;31:350–8.
weight and waist circumference: a prespecified 64. Eaton SB, Konner MJ, Cordain L. Diet-dependent
secondary outcomes analysis of the PREDIMED acid load, Paleolithic nutrition, and evolutionary
randomised controlled trial. Lancet Diabetes health promotion. Am J Clin Nutr. 2010;91:295–7.
Endocrinol. 2016;4:666–76. https://doi.org/10.1016/ 65. Jew S, Abumweis SS, Jones PJH. Evolution of the
S2213-8587(16)30085-7. human diet: linking our ancestral diet to modern
52. Alvarez-Perez J, Sanchez-Villegas A, Diaz-Benitez functional foods as a means of disease prevention. J
EM, et al. Influence of a Mediterranean dietary Med Food. 2009;12(5):925–34.
pattern on body fat distribution: results of the 66. Chambers ES, Morrison DJ, Frost G. Control of
PREDIMED–Canarias intervention randomized appetite and energy intake by SCFA: what are the
trial. J Am Coll Nutr. 2016;35(6):568–80. https:// potential underlying mechanisms? Proc Nutr Soc.
doi.org/10.1080/07315724.2015.1102102. 2015;74(3):328–36.
53. Damasceno NRT, Sala-Vila A, Cofán M, et al. 67. Deehan EC, Walter J. The fiber gap and disappear-
Mediterranean diet supplemented with nuts ing gut microbiome: implications for human health.
reduces waist circumference and shifts lipoprotein Trends Endocrinol Metab. 2016;27(5):239–41.
subfractions to a less atherogenic pattern in sub- 68. Dietary Guidelines Advisory Committee. Scientific
jects at high cardiovascular risk. Atherosclerosis. Report of the 2015. Advisory Guidelines Advisory
2013;230:347–53. Report to the Secretary of Health and Human
54. Shai I, Schwarzfuchs D, Henkin Y, et al. Weight loss Services and the Secretary of Agriculture. Part
with a low-carbohydrate, Mediterranean, or low fat D. Chapter 1: Food and nutrient intakes, and health:
diet. N Engl J Med. 2008;359:229–41. current status and trends. 2015; Figure D1.2:131.
55. Esposito K, Marfella R, Ciotola M, et al. Effect of 69. Dahl WJ, Stewart ML. Position of the Academy
a Mediterranean-style diet on endothelial dysfunc- of Nutrition and Dietetics: health implications of
tion and markers of vascular inflammation in the dietary fiber. J Acad Nutr Diet. 2015;115:1861–70.
metabolic syndrome. a randomized trial. JAMA. 70. González-Rodríguez LG, Sánchez JMP, Aranceta-
2004;292(12):1440–6. Bartrina J, et al. Intake and dietary food sources of
56. Esposito K, Pontillo A, Di Palo C, et al. Effect fibre in Spain: differences with regard to the preva-
of weight loss and lifestyle changes on vascular lence of excess body weight and abdominal obe-
inflammatory markers in obese women: a random- sity in adults of the ANIBES Study. Forum Nutr.
ized trial. JAMA. 2003;289:1799–804. 2017;9:326. https://doi.org/10.3390/nu9040326.
57. Soltani S, Shirani F, Chitsazi MJ, Salehi-Abargouei 71. Grooms KN, Ommerborn MJ, Quyen D, et al.
A. The effect of dietary approaches to stop hyper- Dietary fiber intake and cardiometabolic risk among
tension (DASH) diet on weight and body composi- US adults, NHANES 1999-2010. Am J Med.
tion in adults: a systematic review and meta-analysis 2013;126(12):1059–67.
of randomized controlled clinical trials. Obes Rev. 72. Lairon D. Dietary fiber and control of body weight.
2016;17:442–54. Nutr Metab Cardiovasc Dis. 2007;17:1–5.
References 231
73. Slavin JL. Dietary fiber and body weight. Nutrition. 87. Howarth NC, Saltzman E, Roberts SB. Dietary fiber
2005;21:411–8. and weight regulation. Nutr Rev. 2001;59(5):129–39.
74. Fogelholm M, Anderssen S, Gunnarsdottir I, 88. Karimi G, Azadbakht L, Haghighatdoost F,
Lahti-Koski M. Dietary macronutrients, and food Esmaillzadeh A. Low energy density diet, weight
consumption as determinants of long-term weight loss maintenance, and risk of cardiovascular disease
change in adult populations: a systematic literature following a recent weight reduction program: a ran-
review. Food Nutr Res. 2012;56:19103. domized control trial. J Res Med Sci. 2016;21:32.
75. European Food Safety Authority (EFSA). EFSA https://doi.org/10.4103/1735-1995.181992.
Panel on Dietetic Products, Nutrition, and Allergies. 89. Turner TF, Nance LM, Strickland WD, et al. Dietary
Opinion on dietary reference values for carbohydrates adherence and satisfaction with a bean-based high-
and dietary fibre. EFSA J. 2010;8(3):1462. 27–37. fiber weight loss diet: a pilot study. ISEN Obes.
76. Rautiainen S, Wang L, Lee I-M, et al. Higher intake 2013;2013:5. https://doi.org/10.1155/2013/915415.
of fruit, but not vegetables or fiber, at baseline is 90. Mecca MS, Moreto F, Burini FHP, et al. Ten-week
associated with lower risk of becoming overweight lifestyle changing program reduces several indica-
or obese in middle-aged and older women of normal tors for metabolic syndrome in overweight adults.
BMI at baseline. J Nutr. 2015;145:960–8. Diabetol Metab Syndr. 2012;4:1–7.
77. Fischer K, Moewes D, Koch M, et al. MRI- 91. Pal S, Khossousi A, Binns C, et al. The effect of a
determined total volumes of visceral and subcu- fibre supplement compared to a healthy diet on body
taneous abdominal and trunk adipose tissue are composition, lipids, glucose, insulin and other meta-
differentially and sex-dependently associated bolic syndrome risk factors in overweight and obese
with patterns of estimated usual nutrient intake in individuals. Br J Nutr. 2011;105:90–100.
a northern German population. Am J Clin Nutr. 92. Ferdowsian HR, Barnard ND, Hoover VJ, et al. A
2015;101:794–807. multi-component intervention reduces body weight
78. Lin Y, Huybrechts I, Vandevijvere S, et al. Fibre and cardiovascular risk at a GEICO corporate site.
intake among the Belgian population by sex–age Am J Health Promot. 2010;24(6):384–7.
and sex–education groups and its association 93. Lindstrom J, Peltonen M, Eriksson JG, et al. High-
with BMI and waist circumference. Br J Nutr. fibre, low-fat diet predicts long-term weight loss and
2011;105:1692–703. decreased type 2 diabetes risk: The Finnish Diabetes
79. Du H, van der AD, Boshuizen HC, et al. Dietary Prevention Study. Diabetologia. 2006;49:912–20.
fiber and subsequent changes in body weight and 94. Liber A, Szajewska H. Effects of inulin-type fruc-
waist circumference in European men and women. tans on appetite, energy intake, and body weight in
Am J Clin Nutr. 2010;91:329–36. children and adults: systematic review of
80. Romaguera D, Angquist L, Du H, et al. Dietary randomized controlled trials. Ann Nutr Metab. 2013;
determinants of changes in waist circumference 63:42–54.
adjusted for body mass index—a proxy measure of 95. Wanders AJ, Van de Borne JJ, de Graaf C, et al.
visceral adiposity. PLoS One. 2010;5(7):e11588. Effects of dietary fibre on subjective appetite,
81. Davis JN, Alexander KE, Ventura EE, et al. Inverse energy intake and body weight: a systematic
relation between dietary fiber intake and visceral review of randomized controlled trials. Obes Rev.
adiposity in overweight Latino youth. Am J Clin 2011;12(9):724–39.
Nutr. 2009;90:1160–6. 96. Pittler MH, Ernst E. Guar gum for body weight
82. Tucker LA, Thomas KS. Increasing total fiber intake reduction meta-analysis of randomized trials. Am J
reduces risk of weight and fat gains in women. J Med. 2001;110:724–30.
Nutr. 2009;139:576–81. 97. Hu X, Gao J, Zhang Q, et al. Soy fiber improves
83. Koh-Banerjee P, Franz M, Sampson L, et al. Changes weight loss and lipid profile in overweight and obese
in whole-grain, bran, and cereal fiber consumption adults: a randomized controlled trial. Mol Nutr Food
in relation to 8-yr weight gain among men. Am J Res. 2013;57:2147–54.
Clin Nutr. 2004;80:1237–45. 98. Salas-Salvado J, Farres X, Luque X, et al. Effect
84. Koh-Banerjee P, Chu N-F, Spiegelman DM, et al. of two doses of a mixture of soluble fibres on body
Prospective study of the association of changes in weight and metabolic variables in overweight
dietary intake, physical activity, alcohol consump- or obese patients: a randomised trial. Br J Nutr.
tion, and smoking with 9-y gain in waist circum- 2008;99:1380–7.
ference among 16,587 US men. Am J Clin Nutr. 99. Howarth NC, Saltzman E, McCrory MA, et al.
2003;78:719–27. Fermentable and non-fermentable fiber supplements
85. Liu S, Willett WC, Manson JE, et al. Relation did not alter hunger, satiety or body weight in a pilot
between changes in intakes of dietary fiber and grain study of men and women consuming self-selected
products and changes in weight and development of diets. J Nutr. 2003;133:3141–4.
obesity among middle-aged women. Am J Clin Nutr. 100. Pereira MA, Ludwig DS. Dietary fiber and body
2003;78:920–7. weight regulation observations and mechanism.
86. Ludwig DS, Pereira MA, Kroenke CH, et al. Dietary Childhood Obes. 2001;48(4):969–79.
fiber, weight gain, and cardiovascular disease risk 101. Food and Agriculture Organization of the United
factors in young adults. JAMA. 1999;282:1539–46. Nations. Food energy-methods of analysis and
conversion factors. FAO Food and Nutrition Paper. 116. Aoe S, Ichinose Y, Kohyama N, et al. Effects of high
2003;77:59. b-glucan barley on visceral fat obesity in Japanese
102. Livesey G. Energy values of unavailable carbohy- individuals: a randomized, doubleblind study. Nutr.
drate and diets: an inquiry and analysis. Am J Clin 2017;42:1–6.
Nutr. 1990;51(4):617–37. 117. Menni C, Jackson MA, Pallister T, et al. Gut micro-
103. Johns DJ, Lindroos A-K, Jebb SA, et al. Dietary pat- biome and high fibre intake are related to lower long-
terns, cardiometabolic risk factors, and the incidence term weight gain. Int J Obes. 2017;41:1099–1105.
of cardiovascular disease in severe obesity. Obesity. 118. Holscher HD, Caporaso JG, Hooda S, et al. Fiber
2015;23(5):1063–70. supplementation influences phylogenetic structure
104. Oku T, Nakamura S. Evaluation of the relative avail- and functional capacity of the human intestinal
able energy of several dietary fiber preparations microbiome: follow-up of a randomized controlled
using breath hydrogen evolution in healthy humans. trial. Am J Clin Nutr. 2015;10(1):55–64.
J Nutr Sci Vitaminol. 2014;60:246–54. 119. Nicolucci AC, Hume MP, Martinez I, et al.

105. McRorie JW. Evidence-based approach to fiber
Prebiotic reduce body fat and alter intestinal
supplements and clinically meaningful health microbiota in children who are overweight or with
benefits, part 1. What to look for and how to rec- obesity. Gastrenterology. 2017;153:711–22.
ommend an effective fiber therapy. Nutr Today. 120. Miles CW. The metabolizable energy of diets differ-
2015;50(2):82–9. ing in dietary fat and fiber measured in humans. J
106. Sanchez D, Miguel M, Aleixandre A. Dietary fiber, Nutr. 1992;122:306–11.
gut peptides, and adipocytokines. J Med Food. 121. Miles CW, Kelsay JL, Wong NP. Effect of dietary
2012;15(3):223–30. fiber on the metabolizable energy of human diets. J
107. Holt SH, Miller JB. Particle size, satiety and the Nutr. 1988;118:107–1081.
glycaemic response. Eur J Clin Nutr. 1994;48: 122. Baer DJ, Rumpler WV, Miles CW, Fahey GC. Dietary
496–502. fiber decreases the metabolizable energy content and
108. Rebello CJ, Chu Y-F, Johnson WD, et al. The role nutrient digestibility of mixed diets fed to humans. J
of meal viscosity and oat β-glucan characteristics Nutr. 1997;127:579–86.
in human appetite control: a randomized crossover 123. Cani PD, Lecourt E, Dewulf EM. Gut microbiota
trial. Nutr J. 2014;13:49. fermentation of prebiotics increases satietogenic and
109. Vitaglione P, Lumaga RB, Stanzione A, et al. incretin gut peptide production with consequences
β-Glucan-enriched bread reduces energy intake and for appetite sensation and glucose response after a
modifies plasma ghrelin and peptide YY concentra- meal. Am J Clin Nutr. 2009;90:1236–43.
tions in the short term. Appetite. 2009;53:338–44. 124. Everard A, Cani PD. Gut microbiota and GLP-1.
110. Karhunen LJ, Juvonen KR, Flander SM, et al. A Rev Endocr Metab Disord. 2014;15:189–96.
psyllium fiber enriched meal strongly attenuates 125. Kaji I, Karaki S, Kuwahara A. Short-chain fatty acid
postprandial gastrointestinal peptide release in receptor and its contribution to glucagon-like pep-
healthy young adults. J Nutr. 2010;140:737–44. tide-1 release. Digestion. 2014;89:31–6.
111. Bourdon I, Olson B, Backus R, et al. Beans, as a 126. Tarini J, Wolever TM. The fermentable fibre inulin
source of dietary fiber, increase cholecystokinin and increases postprandial serum short-chain fatty acids
apolipoprotein B48 response to test meals in men. J and reduces free-fatty acids and ghrelin in healthy
Nutr. 2001;131:1485–90. subjects. Appl Physiol Nutr Metab. 2010;35(1):9–16.
112. Beck EJ, Tosh SM, Batterham MJ, et al. Oat beta- 127. Kasubuchi M, Hasegawa S, Hiramatsu T, et al.
glucan increases postprandial cholecystokinin levels, Dietary gut microbial metabolites, short-chain fatty
decreases insulin response and extends subjective acids, and host metabolic regulation. Forum Nutr.
satiety in overweight subjects. Mol Nutr Food Res. 2015;7:2839–49.
2009;53:1343–51. 128. Conterno L, Fava F, Viola R, Tuohy KM. Obesity
113. Martinez-Rodriguez R, Gil A. Nutrient-mediated and the gut microbiota: does up-regulating colonic
modulation of incretin gene expression: a systematic fermentation protect against obesity and metabolic
review. Nutr Hosp. 2012;27:46–53. disease? Genes Nutr. 2011;6:241–60.
114. Hussain SS, Bloom SR. The regulation of food 129. Chambers ES, Viardot A, Psichas A, et al. Effects
intake by the gut-brain axis: implications for obesity. of targeted delivery of propionate to the human
Int J Obes (Lond). 2013;37:625–33. colon on appetite regulation, body weight main-
115. Hairston KG, Vitolins MZ, Anderson AM, et al. tenance and adiposity in overweight adults. Gut.
Lifestyle factors and 5-year abdominal fat accumu- 2015;64(11):1744–54.
lation in a minority cohort: the IRAS family study.
Obesity. 2011;20:421–27.
Whole Plant Foods in Body Weight
and Composition Regulation 8
Keywords
Whole plant foods • Weight maintenance • Weight loss • Body fat • Central
obesity • Waist circumference • Whole grains • Fruit • Vegetables • Dietary
pulses • Nuts
Key Points size or body fat gain and obesity, especially

for healthier varieties. However, higher energy
• Foods commonly associated with weight gain
dense, lower fiber fruits and vegetables may
are the high intake of French fries, sugar-
promote weight gain.
sweetened beverages, and red and processed
• RCTs indicate that lower energy dense, higher
meats, and the foods that tend to be inversely
fiber and flavonoid rich fruits and vegetables
associated with weight gain are non-starchy
can support lower risk of weight gain or mod-
vegetables, high fiber and flavonoid rich fruits,
est weight loss and promote additional weight
whole grains, nuts, and plain yogurt.
loss in a hypocaloric diet or help to support
• Healthy lower energy dense dietary patterns
weight maintenance after weight loss.
rich in whole or minimally processed plant
• RCTs show that the daily consumption of
foods (whole plant foods) tend to be associ-
dietary pluses and nuts do not promote weight
ated with a lower risk of weight gain and obe-
gain, and may support modest weight loss.
sity compared to the more common Western
Nuts consumed as a snack or legumes as a
diets high in processed foods.
meal protein source in weight loss diets do not
• Prospective cohort studies show >3 daily
tend to interfere with weight loss or weight
whole-grain servings (especially with total
maintenance after weight loss.
cereal fiber at approximately 10 g/day), can
reduce body weight and waist size compared
to < one half serving/day. Randomized control
trials (RCTs) indicate that whole-grains are 8.1 Introduction
more effective in reducing body fat and waist
size than body weight or BMI. Overweight and obesity status in the human pop-
• For fruit and vegetables, cohort studies find an ulation largely remained an exception until the
association with a lower risk of weight, waist 1970s, when increasing urbanization, sedentary

234 8 Whole Plant Foods in Body Weight and Composition Regulation
jobs and the availability of processed foods fiber diet; (3) consuming breakfast; (4) self-
produced a sharp rise in overweight and obesity monitoring weight on a regular basis; (5) limiting
in both children and adults [1]. The worldwide consumption of higher energy dense foods; and
overweight and obesity pandemic is among the (6) catching dietary miss-steps before they
greatest public health challenges of our time with become a habit [15, 29–31]. A pooled prospec-
over two billion people now overweight or obese tive investigation of three cohorts from the
globally [2–4]. Obesity or excessive abdominal Nurses’ Health Studies and Health Professionals
adiposity in adulthood and childhood is a grow- Follow-up (120,877 U S women and men; over
ing risk factor for major chronic diseases [5–9]. 4 years) found that specific foods are indepen-
These conditions are associated with increased dently associated with weight change (Fig. 8.1);
health care costs and reduced workforce produc- the foods positively associated with increased
tivity and an estimated >300,000 premature adult weight were French fried potatoes, potato chips,
deaths each year in the US [10, 11]. A small daily sugar-sweetened beverages, red and processed
positive energy balance of 50 kcals/day, by meats and the foods inversely associated with
increased energy intake, lower fiber diets, and/or weight gain were vegetables, whole grains, fruits,
reduced activity, can lead to an annual weight nuts, and yogurt (p ≤ 0.005 for each comparison)
gain of 0.4–0.9 kg/year [12–16]. Further, a higher [32]. This analysis and future research in this area
habitual intake of 200 kcal/day above energy bal- have important implications for obesity preven-
ance in overweight or obese women may increase tion strategies [32]. The objective of this chapter
weight gain by as much as 9 kg/year [17]. People is to review the effects of specific whole or mini-
tend to eat similar amounts or volumes of food on mally processed plant foods (whole plant foods)
a day-to-day basis regardless of the food energy on weight and body composition regulation.
density, so the common advice of just eating less
of all foods may not be the optimal approach for
weight management [18–22]. A systematic 8.2 Whole Plant Foods
review found that higher energy dense, lower
fiber dietary patterns may pre-dispose children to Although whole plant foods are more generally
later increased risk of being overweight or obese associated with lower energy density, reduced
as adults [23]. For overweight or obese individu- risk of weight gain or obesity than highly pro-
als who successfully lose weight, as many as cessed plant foods [33], whole plant foods can
80% typically drift back to their original weight vary widely in nutrient composition, energy den-
or more [24]. This is because after weight loss sity, and physical properties (Appendix A). Thus,
there are an array of metabolic regulatory pro- not all whole plant foods are equally effective in
cesses causing a cascade of dietary energy den- weight management. Nine intervention trials and
sity signals designed for weight regain [25–27]. review articles consistently show that unrestricted
One study showed that weight loss maintainers whole food plant-based diets are significantly
for >5 years reported consuming a diet with a sig- more effective in lowering body weight and other
nificantly lower energy density (1.4 kcal/g) than anthropometric measures that the usual or
the weight regain individuals (1.8 kcal/g) [28]. Western diets [34–42]. A 2017 New Zealand
The primary diet difference was that the weight based randomized controlled trial (RCT; 65
maintainers consumed more fiber-rich foods such subjects; mean age 56 years; mean BMI 34;
as vegetables (4.9 servings/day) and whole-grain non-energy restricted whole food plant-based
products (2.2 servings/day) compared to less than diet vs. usual Western type diet; regular exercise
one daily serving of vegetables and whole grains was not mandated; 6 and 12 months) showed that
for the weight regainers. Successful long-term the whole plant food diet reduced BMI 3.9 kg/m2
weight loss maintenance is associated with six after 6 months and 4.2 kg/m2 after 12 months
key strategies to help counteract weight regain compared to the usual diet [34]. A worksite 2016
metabolic processes: (1) engaging in physical US open label trial (35 subjects; mean age 43 years;
activity; (2) eating a low energy dense and high 91% female; nutrient-dense, plant rich diet with
8.2 Whole Plant Foods 235
Sweet desserts
Potato chips
Baked, boiled, mashed potatoes
French fried potatoes
Sugar sweetened beverages
100% fruit juice
Refined grains
Diet soda
Whole-grains
Nuts
Vegetables
Fruit
–1 –0.5 0 0.5 1 1.5 2 2.5 3 3.5

Multivariate-adjusted 4-year weight change (lbs)
Fig. 8.1 Association between specific “whole or processed plant food” choices and 4-year weight change in US men
and women from pooled Nurses’ Health Studies and Health Professionals Follow-up Study data (adapted from [32])
daily greens, beans, legumes, vegetables, fresh or Diets also reduced systolic blood pressure by 5.1
frozen whole fruits, nuts, seeds and whole grains mm Hg and diastolic blood pressure by 3.2 mm
and minimal intake of refined grains, vegetables Hg compared to the average Danish diet. Ab libi-
oils, processed foods and meats or full-fat dairy; tum consumption of approximately 75% whole
6 weeks) found significant reduction in body plant based diets and plus dairy products, fish and
weight, BMI, waist and hip measurements com- eggs with limited sweets, highly processed foods,
pared to baseline measurements [35]. A 2016 red or processed meats is a viable option for the
Australian base Mobile Health Lifestyle Program treatment and prevention of overweight and obe-
for the Prevention of Weight Gain in Young sity [37–42]. The MyPlate visual educational tool
Adults RCT (250 subjects; age range 18 to 35 was developed to encourage Americans to
years; 60% women; mean BMI 27; 9 months) increase their intake of whole or minimally pro-
showed greater odds of reducing weight gain by cessed fruits, vegetables and grains to promote
meeting recommendations for fruit and vegetable better health and weight control [43].
intake by 383% and 242%, respectively, limiting
intake of sugar sweetened beverages by 311%,
and limiting take-out meals by 188% [36]. The 8.2.1 Whole-Grain Foods
2014 Danish RCT (181 subjects; 71% women;
mean age 42 years; mean BMI 30; ab libitum Whole-grain products (brown rice, oatmeal, pop-
New Nordic Diet high in fruit, vegetables, whole- corn, whole wheat, or rye bread and crackers, and
grains and fish vs. average Danish Western type whole-grain/fiber-rich breakfast cereals) contain
diet; 26 weeks) found that the subjects on the the whole intact grain kernel with natural levels of
New Nordic Diets lost 3 kg more body weight fiber, vitamins, minerals and phytochemicals [44–
than those on the average Danish diet for the 48]. In contrast, refined grain products (white
intention-to-treat analysis [42]. The New Nordic rice and white bread, pastry, donuts and low fiber
breakfast cereals) are mainly comprised of the whole-grain [55]. In a Nurses Health Study
starchy endosperm with most of the fiber, vitamins, (74,091 women; mean baseline age 50 years;
minerals and phytonutrients removed during pro- 12 years of follow-up) women consuming 2.3
cessing. The US dietary guidelines recommend >3 daily servings of whole grains weighed 0.9 kg less
servings of whole-grains/day and <3 servings of than those consuming 0.7 servings, whereas
refined grains/day to promote health and wellness women with similar refined-grain intake gained
associated with reduced risk of various chronic dis- 1.2 kg [56]. A Minneapolis Public School Study
eases [44, 49]. However, only about 1% of (240 students; mean baseline age 13 years; 2 years)
Americans follow the recommendation for whole- showed that students who consumed >1.5 whole-
grain intake as the average intake is <1 ounce whole grain servings daily reduced BMI by 7% com-
grains/day and 70% of Americans exceed the rec- pared to students consuming <0.5 servings [57].
ommended intake for refined grains [44, 49, 50].
8.2.1.2 Randomized Controlled
8.2.1.1 Observational Studies Trials (RCTs)
Observational studies consistently show that RCTs of whole grain intake on weight and body
higher intake of whole-grains, but not refined composition are summarized in Table 8.2
grains, is associated with lower BMI and/or [58–68]. A comprehensive meta-analysis (26
reduced obesity risk [50, 51]. A systematic review RCTs; 2060 participants; 18–150 g whole-grains;
and five prospective cohort studies on weight and 2–16 weeks) showed that overall increased
body composition are summarized in Table 8.1 whole-grain intake had insignificant increased
[52–57]. The systematic review of cohort and body weight by 0.06 kg and reduced waist size by
cross-sectional studies (15 studies; 119,829 par- 0.10 cm, and a significant reduction of body fat
ticipants) found that whole grain intake resulted by 0.48% compared with control diets [58]. A
in a mean reduced BMI by 0.6 units, waist size by subgroup analysis found that brown and black
2.7 cm and waist- to-hip ratio by 0.023 per three rice significantly decreased body weight by
servings/day compared to <0.5 serving/day [52]. 1.1 kg and body fat by 1.2%, oats significantly
This analysis also showed that the consumption of decreased waist size by 1.2 cm and whole wheat
three servings of whole grains increased total cereal lowered body fat by 0.71% (p = 0.08) com-
fiber by 9 g/day and reduced total fat by 11 g/day pared to control diets.
in the diet. Prospective studies consistently show Several RCTs on increased whole-grain in ad
an inverse association between whole grain intake libitum diets show mixed outcomes on weight
(rich in cereal fiber) and body weight [53–57]. In and body composition [59–61]. A crossover RCT
a European Prospective Investigation into Cancer (33 adults, mean age 48 years; mean BMI 28;
and Nutrition (EPIC) Study (89,432 participants; mean 168 g vs. 28 g whole grains; 6 weeks)
mean baseline age 53 years; 6.5 years of follow- detected a slight trend toward lower body weight
up) individuals with the highest intake of whole for whole grains [59] whereas a parallel RCT
grains providing 10 g daily cereal fiber signifi- (316 subjects; mean baseline age 46 years; mean
cantly reduced both weight and waist size [53]. BMI 30; <30 g, 60 g, and 120 g whole grains;
The Physician Health Study (17,881 men; mean 8–16 weeks) did not find a significant effect on
baseline age 53 years; follow-up at 8 and 13 years) body weight or fat (%) [61]. However, a Japanese
showed an inverse association between breakfast crossover RCT (27 males; mean BMI 26; brown
cereal intake and weight gain with a significant vs. white rice; 8 weeks for each phase) found that
22% lower risk of >10 kg weight gain, compared brown rice decreased body weight, BMI and
with the lowest consumers [54]. The Health waist size, but these weight changes returned to
Professionals Follow-up Study (27,082 men; baseline values by the end of the white rice diet
mean baseline age 52 years; 8 years of follow-up) period [60]. Also, the intra-abdominal visceral
reported that each daily 40 g increase in whole fat (%) was significantly lower after 8 weeks of
grains reduced weight gain by 0.5 kg with brown rice consumption than after a comparable
bran being approximately twice as effective as period of white rice (Fig. 8.2).
Table 8.1 Summary of whole-grain prospective cohort studies on body weight and body composition regulation
Systematic review
Harland and Garton (2008). 15 cohort and cross-sectional Whole-grain significantly reduced BMI by
Review evidence relating to the studies; 119,829 primarily 0.6 units, waist size by 2.7 cm, and waist:
intake of whole-grain and European and American adults hip ratio by 0.023 in individuals consuming
healthy body weight [52] 3 servings/day compared to <0.5 servings/
day. Higher intake of whole-grain led to
increased fiber intake by 9 g/day and
lowered total fat by 11 g/day and saturated
fat by 3.9 g/day
Du et al. (2013). 89,432 participants, mean A 10-g/day higher cereal fiber intake was
Investigate the association of baseline age 53 years; average associated with annual mean weight
total fiber, cereal fiber, and fruit 6.5 years of follow-up reduction by 0.77 g and lower waist size by
and vegetable fiber with changes (multivariate adjusted) 0.10 cm. Fruit and vegetable fiber was not
in weight and waist size. associated with weight change but had a
(European Prospective similar association with waist size
Investigation into Cancer and reduction as cereal fiber
Nutrition [EPIC]) [53]
Bazzano et al. (2005). 17,881 men; mean baseline age Men who consumed any type of breakfast
Assess the association between 53 years; mean BMI 24; >1 cereal consistently weighed less than those
whole-grain and refined grain whole-grain servings/day vs. who rarely consumed breakfast cereals
breakfast cereal intakes and risk rarely consume; 8 and 13 years of (p-trend = 0.01). Those who consumed >1
of overweight and weight gain follow-up (multivariate adjusted) serving/day of breakfast cereals were 22%
(The Physician’s Health Study; less likely after 8 years and 12% less likely
US) [54] after 13 years to become overweight
compared with men who rarely or never
consumed breakfast cereals
Koh-Banerjee et al. (2004). 27,082 men; mean baseline age Whole-grain intake was significantly
Ascertain the associations 52 years; mean BMI 25; 27 g inversely associated with long-term weight
between changes in quantitative whole-grains/day vs. 11 g WG/ gain. For every 40 g/day increased intake
estimates of whole-grain intake day; 8 years of follow-up of whole-grain foods, a dose-response
and 8-year weight gain among (multivariate adjusted) relationship was observed for a
men (Health Professionals significantly reduced weight gain by
Follow-up Study; US) [55] 0.5 kg. Bran that was added to the diet or
obtained from fortified-grain foods further
reduced the risk of weight gain for every
20 g/day by 0.36 kg
Liu et al. (2003). 74,091 women; mean baseline Women in the highest quintile of whole-
Examine the associations age 50 years; mean BMI 25; grain intake weighed 0.9 kg less than
between the intakes of dietary median intake of 2.3 whole-grain women in the lowest quintile of intake,
fiber and whole-grain or servings/1000 kcals vs. 0.07 whereas women in the highest quintile of
refined-grain products and whole-grain servings/1000 kcals; refined-grain intake weighed 1.2 kg more
weight gain over time (Nurses’ 12 years of follow-up than women in the lowest quintile of
Health Study; US) [56] (multivariate adjusted) intake. Women in the highest quintile of
fiber intake had a 49% lower risk of major
weight gain than women in the lowest
quintile
Steffen et al. (2003). 240 students; mean baseline age The students consuming higher whole-
Investigate the association 13 years; >1.5 whole-grain vs. grain had significantly lower BMI by 1 unit
between whole-grain intake and <0.5 whole-grain servings/day; compared to the students with lower
BMI in adolescents 2 years of follow-up (multivariate whole-grain intake, after multivariate
(Minneapolis Public School adjusted) adjustments. Also, the students with higher
Students Study; US) [57] whole-grain-intake had significantly
greater insulin sensitivity
Table 8.2 Summary of whole-grain randomized controlled trials (RCTs) on body weight and composition regulation
Pol et al. (2013). 26 RCTs; 2060 participants; Whole-grain intake had insignificant effects on
Assess the effects of daily whole-grain dose ranged increased body weight by 0.06 kg and decreased
whole-grain foods from 18 to 150 g; duration waist size by 0.10 cm, and a small but significant
compared with non-whole- ranged from 2 to 16 weeks lowering effect on percentage of body fat by 0.48%
grain foods on changes in with the majority of studies compared with the non-whole-grain control. Whole
body weight, percentage of lasting 4–6 weeks wheat cereal lowered body fat more than a control
body fat, and waist size [58] by 0.71% (p = 0.08). A subgroup analysis for
individual grains showed that only whole-grain rice
decreased body weight by 1.1 kg and percentage of
body fat by 1.2% compared with the white rice
control. Whole-grain oats decreased waist size by
1.2 cm more than the control
RCTs
Ad libitum energy intake
Ampatzoglou et al. (2015). Crossover RCT During the whole-grain intervention, there was a
Assess the impact of 33 subjects; 12 males and 21 significant increase in plasma alkylresorcinols and
increasing whole-grain females; mean age 48 years; total fiber intake, without any effect on energy or
intake on body weight, mean BMI 28; mean 28 g vs. other macronutrients. Although there were no
blood pressure, blood lipids, 168 g whole-grain/day; effects on studied variables, there were trends
blood glucose, microbiota, 6 weeks; 4 weeks of washout; toward increased 24-h fecal weight (p = 0.08) and
and gastrointestinal adherence was achieved by reduction in body weight (p = 0.10) and BMI
symptoms in healthy, specific dietary advice and (p = 0.08) during the high whole-grain intervention
middle-aged adults (UK) provision of a range of cereal compared with the low whole-grain period
[59] food products
Shimabukuro et al. (2014). Crossover RCT In the group that ate brown rice for the first
Evaluate the effects of 27 male subjects with 8-weeks test period body weight, BMI and waist
brown rice and white rice metabolic syndrome; mean age size were decreased but returned to baseline values
on abdominal fat 41 years; mean BMI 28; by the end of the white rice diet period. In the
distribution and metabolic switch from brown to white group that are white rice in the first period body
parameters (BRAVO study; rice and white to brown rice; weight, BMI and waist size were comparable with
Japan) [60] subjects 8 weeks on each rice; the baseline values, but waist size was lower after
no washout switching to brown rice diet. Also, intra-abdominal
visceral fat (%) was significantly lower after brown
rice than after white rice consumption (Fig. 8.2)
Brownlee et al. (2011). Parallel RCT The consumption of whole-grain from 30–120 g/
Evaluate the effect of 316 participants; mean aged day (added fiber 6–11 g/day) for 16 weeks did not
substituting whole-grain 46 years; mean BMI 30; 3 significantly change weight or body fat (%)
foods in the diet of habitual diets: (1) 60 g whole-grain/day
refined-grain consumers on for 16 weeks; (2) 60 g
markers of CVD risk and whole-grains/day for 8 weeks
weight measures (UK) [61] followed by 120 g whole
grains/day for 8 weeks; and
(3) < 30 g whole-grain/day for
16 weeks (control)
Restricted energy intake
Harris Jackson et al. Parallel RCT Replacing refined grains with whole-grains within
(2014). 50 subjects;mean age 46 years; a weight-loss diet did not significantly improve
Investigate the effect of mean BMI 33; controlled weight, BMI or abdominal visceral adipose tissue
consuming whole-grains to weight-loss diet containing loss. However, the whole-grain diet significantly
replace refined grains in the 163–301 g whole-grain/day vs. reduced the prevalence of prediabetes by 90%
diets of individuals with 0 g whole-grain/day; 12 weeks compared with 13% for the refined grain diets.
metabolic syndrome or at Whole-grain diets were more effective at
risk for metabolic syndrome normalizing blood glucose levels and reducing the
(US) [62] risk of individuals with prediabetes from
progressing to type 2 diabetes
Kristensen et al. (2012). Parallel RCT Body weight decreased significantly from baseline
Study the effect of replacing 79 postmenopausal women; in refined wheat group by 2.7 kg and whole-grain
refined wheat with mean age 68 years; mean BMI wheat group by 3.6 kg with no significance
whole-grain wheat on body 30; energy-restricted diet (by between the groups (p = 0.11). The reduction in
weight and composition 300 kcal/day) with 105 g total body fat % was significantly greater in the
(EU) [63] whole wheat grains daily or whole-grain group (Fig. 8.3). Serum total and LDL
refined wheat foods; 12 weeks cholesterol significantly increased by 5% in the
refined wheat group but did not change in the
whole-grain group (p = 0.02)
Maki et al. (2010). Parallel RCT Both groups lost weight, in the whole-grain oat
Investigate the effect of a 144 subjects; mean age cereal group by 2.2 kg and the control by 1.7 kg
whole-grain, ready to-eat 49 years; mean BMI 32; 78% (p = 0.325). Waist size decreased significantly
(RTE) oat cereal containing female; 2 portions/day of more with whole-grain oat cereal by 3.3 cm
viscous fiber, as part of a whole-grain RTE oat whole compared with 1.9 cm for the control (p = 0.012)
dietary program for weight grain cereal (3 g/day oat (Fig. 8.4)
loss (US) [64] β-glucan) or energy-matched
low-fiber foods (control),
reduced energy diet by
500 kcal/day; 12 weeks
Katcher et al. (2008). Parallel RCT Body weight, waist size, and body fat (%)
Determine whether 50 metabolic syndrome adults; decreased significantly in both groups, but there
including whole-grain foods 25 males and 25 females; was a significantly greater decrease in body fat (%)
in a hypocaloric diet mean age 46 years; mean BMI in the abdominal region in the whole-grain group
(reduced by 500 kcal/day) 36; 5 whole-grain servings/day than in the refined-grain group. C-reactive protein
effects weight loss and vs. <0.25 servings in the (CRP) decreased 38% in the whole-grain group
improves CVD risk factors refined grain group; all independent of weight loss compared to no change
(US) [65] participants were given the in the refined-grain group. Total and LDL
same dietary advice in other cholesterol decreased in both diet groups. Total
respects for weight loss; fiber and magnesium intakes increased in the
12 weeks whole-grain compared to the refined-grain group
Kim et al. (2008). Parallel RCT The subjects on the brown and black rice
Assess the effect of type of 40 overweight Korean women; supplemented diets showed a significantly greater
rice consumed on weight 20–35 years of age; energy reduction in weight by 1.4 kg and body fat by
control when consumed restricted diets containing 1.2% compared to the white rice group
with an energy restricted either white rice or mixture of
diet (Korea) [66] brown rice and black rice;
6 weeks
Melanson et al. (2006). Parallel RCT In the hypocaloric diet, consumption of whole-
Investigate the effects of 134 adults; mean age 42 years; grain breakfast cereals insignificantly improved
exercise plus a hypocaloric mean BMI 31; hypocaloric weight loss by 0.3 kg compared to the refined
whole-grain diet on weight diet with and without cereal diet
loss (US) [67] whole-grain breakfast cereals;
23 vs. 17 g fiber/day; 24 weeks
Saltzman et al. (2001). Parallel RCT In the hypocaloric diets, there was no difference in
Evaluate the effects of oats 43 subjects; mean BMI 26; weight loss by adding 45 g/day oats vs. the control
on weight loss and body mean age 45 years; 2 diet as both groups lost about 4 kg (p = 0.8). The
composition (US) [68] hypocaloric diets (maintenance oat supplemented diet significantly lowered mean
energy minus 1000 kcals/day): systolic blood pressure and total and LDL
(1) diet containing oats 45 cholesterol
g/day and (2) control diet (no
added oats); 6 weeks
Switch from White Rice to Brown Rice Switch from Brown Rice to White Rice
–30 –25 –20 –15 –10 –5 0 5 10 15 20 25

Change in intra-abdominal visceral fat area (%)
Fig. 8.2 Effect of brown vs. white rice on mean intra-abdominal visceral fat in Japanese men with metabolic syndrome
after 8 weeks (p < 0.018) (adapted from [60])
Seven RCTs show increased whole-grain in to eat (RTE) oat whole grain cereal (3 g/day oat
energy restricted diets has the capacity to signifi- β-glucan) or energy- matched low-fiber foods
cantly affect abdominal or total body fat, waist (control), as part of diet energy reduced by
circumference, CRP levels and diabetes risk if 500 kcal/day; 12 weeks) found that waist size
adequate whole grains are consumed [62–68]. decreased significantly more with oat cereal by
Two RCTs suggest that whole grain from b reakfast 3.3 cm compared with 1.9 cm for the control
cereal or oats has no additive effect in hypocaloric (p = 0.012) (Fig. 8.4) [64]. A RCT in individuals
diets on weight or body composition but the level with metabolic syndrome (50 adults; mean age
of overall whole grain intake in the RCTs appears 46; mean BMI 36; 5 whole grain servings vs.
to have been relatively modest based on the fiber <0.25 servings; 12 weeks) reported a significant
intake range from 13 to 23 g/day [67, 68]. Another reduction in abdominal fat (%) and CRP levels in
RCT (50 subjects; mean age 46 years; mean BMI the whole grain vs. refined grain group [65]. In a
33; controlled weight-loss diet containing 163– rice based hypocaloric diet (40 Korean women;
301 g whole-grains/day vs. 0 g whole-grains/day; age 20–35 years; 6 weeks) consuming brown and
12 weeks) showed that whole grains had no effect black rice was shown to significantly reduce body
on weight or body composition but they did sig- weight and body fat (%) vs. white rice [66].
nificantly reduce the risk of prediabetes by 90%
compared to 13% lower risk in the refined grain
group [62]. A RCT in postmenopausal women 8.2.2 Fruits and Vegetables
(79 women; mean age 68 years; mean BMI 30;
energy-restricted diet (by 300 kcal/day) with 8.2.2.1 Background
105 g whole wheat grain daily or refined wheat The Dietary Guidelines for Americans (Myplate.
foods; 12 weeks) found that body weight gov) recommends that whole or minimally pro-
decreased significantly from baseline in the cessed fruits and vegetables make-up one-half of
refined wheat group by 2.7 kg and in the whole a meal’s plate [43, 44]. Fruits and vegetables
grain wheat group by 3.6 kg with no significance include a diverse group of plant foods that vary
between the groups (p = 0.11) [63]. However, the widely in their health effects due to a range of
whole grain group had a significantly lower body energy, fiber, glycemic index, nutrients, and
fat (%) (Fig. 8.3). An oat breakfast cereal RCT phytonutrients contents, and physical properties.
(144 subjects; mean age 49 years; mean BMI 32; Adequate intake of fruits and vegetables (>400 g/
78% female; 2 portions/day of whole-grain ready day) makes important contributions to health
Refined wheat Whole-grain wheat

0
Body weight Total fat mass Central fat mass
(p =.11) (p =.07) (p = .04)
–1
–2
Percentage Change
–3
–4
–5
–6
Fig. 8.3 Effect of whole-grain in an energy-restricted dietary intervention on change in body weight, total fat mass, and
central fat mass in postmenopausal women after 12 weeks (adapted from [63])
Control Oat RTE cereal

0
Baseline 2 4 6 8 10 12
–0.5
–1
Waist Circumference (cm)
–1.5
–2
–2.5
–3
Dietary weight loss program
–3.5
–4
Fig. 8.4 Effect of whole-grain oat ready-to-eat (RTE) on waist size in obese adults in a dietary weight loss program
with 3 g/day oat β-glucan twice daily for 12 weeks (p = 0.012) (adapted from [64])
because of their unique concentrations of: anti- of fruits and less than two cups of vegetables
oxidant vitamins and phytochemicals, especially daily, with the primary contributors consisting of
vitamins C and A, flavonoids and carotenoids, juice and processed potatoes, compared to the
minerals (especially electrolytes potassium and current recommendations of 2 cups of fruit and
magnesium, and low sodium), and fiber [69–73]. 2.5 cups of vegetables per day. More than 85% of
Globally, fruits and vegetables consumption is at the US and likely other Western population fall
only a small fraction of the recommended levels short of meeting the daily recommendations [43,
[74]. Most Americans consume less than one cup 44]. A potential benefit of a diet rich in fruit and
non-starchy vegetables is their low energy den- 8.2.2.2 Systematic Reviews

sity which may help in preventing weight gain and Meta-Analyses
due to their water, fiber and bulk volume, which Overall, systematic reviews and meta-and pooled
contribute to satiation compared to the typical analyses of prospective studies and RCTs show
low fruit and vegetable diets associated with the that whole or minimally processed fruits and veg-
Western lifestyle. Also, a 2016 pooled analysis etables have an important role in reducing risk of
of 124,086 US men and women suggests that weight gain and obesity and central obesity with
consuming high flavonoid fruits and vegetables, some exceptions (Table 8.3) [76–80]. A systematic
such as apples, pears, berries, and peppers, may review of potatoes on weight gain (five prospective
be especially helpful in preventing weight gain studies; 170,413 subjects; follow-up between 2
and obesity compared to other types of low fla- and 20 years) found that total potato intake
vonoid fruits and vegetables [75]. (excluding French fries), were inconsistently
Table 8.3 Summary of fruit and vegetable prospective cohort studies and randomized controlled trials (RCTs) studies
on body weight and composition regulation
Systematic review and meta-pooled analyses
Borch et al. (2016). 5 prospective studies; 170,413 For potatoes (excluding French fries), two
Evaluate the relationship subjects; follow-up between 2 studies showed a positive association with
between potatoes and obesity and 20 years adiposity and two studies showed no
risk [76] association with adiposity. In the three studies
that investigated French fries separately, a
positive association between intake of French
fries and measures of adiposity was shown. In
two of these studies, it was shown that intake
of French fries had a stronger positive
association with both BMI and weight gain
than did boiled, baked, or mashed potatoes
Schwingshackl et al. (2015). 17 cohort studies (from 20 Higher intake of fruits was associated with
Perform a systematic review reports); 563,277 participants; reduced annual weight by 13.7 g/100 g intake
and meta-analysis of 9 months to 20 years (slightly over one serving). No significant
prospective cohort studies on changes were observed for combined fruits
fruits and vegetables and vegetables or vegetable intake. Increased
consumption in relation to intake of fruits was associated with a reduction
changes in anthropometric in waist circumference by 0.04 cm/year.
measures [77] Comparing the highest vs. lowest intake,
reduced risk of adiposity was observed for
combined fruits and vegetables by 9%, fruit by
17%, and vegetables by17%
Kaiser et al. (2014). 7 RCTs; 1103 participants; This analysis demonstrated that increased
Synthesize the best available primary or secondary outcome fruits and vegetables intake including 100%
evidence on the effectiveness of of body weight; including juice insignificantly increased body weight by
the general recommendation to 100% juices; >8 weeks 0.04%
eat more fruits and vegetables
for weight loss or the prevention
of weight gain [78]
Mytton et al. (2014). 8 RCTs; 1026 participants; High fruits and vegetables intake significantly
Quantify the relationship excluding 100% juice; mean reduced body weight by 0.68 kg vs. lower
between changes in fruits and 14.7 weeks fruits and vegetables intake, despite no
vegetables (excluding juices difference in daily energy intake (p = 0.07).
intake), energy intake and body Increased fruits and vegetables intake, in the
weight [79] absence of specific advice to decrease
consumption of other foods, appears unlikely
to lead to weight gain in the short-term and
may subsequently have a role in weight
maintenance or loss
Ledoux et al. (2011). 12 intervention trials (11 on In energy restricted intervention trials, higher
Assess the relationship between adults and one on children) intake of fruits and vegetables was weakly
fruits and vegetables intake and and 11 longitudinal studies associated with weight loss among overweight
adiposity [80] (seven on adults and four on or obese adults, but not children. In
children) longitudinal studies, high fruits and vegetables
intake was associated with less or slower
weight gain over lengthy time intervals among
adults, but to a lesser degree among children
Prospective or longitudinal cohort studies
Shefferly et al. (2016). 8950 children; examined at Regular 100% juice consumption between
Evaluate the relationship ages 2, 4 and 5 years ages 2 and 4 years increased the odds of
between 100% juice intake and (multivariate adjusted) becoming overweight by 30%. However,
weight status in pre-school significant increase in weight was not
children (Early Childhood observed in children at age 5 years
Longitudinal Study-Birth
Cohort; US) [81]
Bertoia et al. (2015). 133,468 men and women; in Increased intake of fruits was inversely
Examine the effect of increased women, the average baseline associated with 4-year weight loss per daily
fruits and vegetables intake on age was 36 and 49 years; in serving for total fruits by 0.53 lb., berries by
weight change over time, men, the average baseline age 1.11 lbs., and apples and pears by 1.24 lbs.
including subtypes and was 47 years; multiple 4-year (Fig. 8.5). Increased intake of several
individual fruits and vegetables weight measurement cycles vegetables was also inversely associated with
(Nurses’ Health Study I and II, over 24 years of follow-up weight loss per daily serving for total
and Health Professionals (multivariate adjusted) vegetables by 0.25 lb., tofu/soy by 2.47 lbs.
Follow-up Study; US) [82] and cauliflower by 1.37 lbs. (Fig. 8.6). In
contrast, increased intake of starchy
vegetables, including corn, peas, and potatoes,
was associated with weight gain (Fig. 8.7).
Vegetables having both higher fiber and lower
glycemic load were more strongly associated
with weight loss compared with lower-fiber,
higher-glycemic-load vegetables
Rautiainen et al. (2015). 18,146 women; mean baseline Vegetable intake was associated with greater
Investigate whether intake of age 54 years; mean baseline weight gain (p-trend = 0.02) and subjects with
fruits and vegetables, and total BMI 22; mean follow-up of higher fruit intake had a 13% lower risk of
fiber is associated with weight 15.9 years; 8125 women becoming overweight or obese (higher vs.
change and the risk of becoming became overweight or obese lower intake). Overall, greater intake of fruit,
overweight and obese (Women’s (multivariate adjusted) but not vegetables, by middle-aged and older
Health Study; US) [83] women with a normal BMI is associated with
lower risk of becoming overweight or obese
Vergnaud et al. (2012). 373,803 participants; mean Baseline fruits and vegetables intake was
Assess the association between baseline age 52 years; mean associated with weight loss in men and women
the baseline consumption of BMI 26; country-specific who quit smoking during follow-up. There
fruits and vegetables and weight validated questionnaires; per was a weak association between vegetable
change (European Prospective 100 g fruits and vegetables/ intake and weight loss in women who were
Investigation into Cancer and day and weight change overweight, were former smokers and weak
Nutrition (EPIC) study) [84] (g/year); mean follow-up of associations between fruit intake and weight
5 years (multivariate adjusted) loss in women who were >50 years of age,
were of normal weight, or were never smokers
(continued)
Representative RCTs
Tapsell et al. (2014). Single Blind, Parallel RCT Both groups significantly reduced intake of
Assess the effects of higher 120 adults; mean BMI 30; high energy dense vegetables and increased
vegetable consumption on mean age 49 years; two 20% portions of low energy dense vegetable as
weight loss (Australia) [85] energy deficit groups with instructed. The higher percentage energy from
healthy diet advice to consume vegetables was positively associated with
vegetables each day. The test weight loss and sustainability (Fig. 8.8).
group was asked to consume Weight loss was sustained for 12 months by
>5 servings of low energy both groups, but the higher vegetable group
dense vegetables each day, but reported significantly greater hunger
the control vegetable group satisfaction
consumed half the portions
(0.5 vs. 1.0 cup cooked or 1
vs. 2 cups of raw,
respectively); 12 months
Christensen et al. (2013). Parallel RCT Higher fruit intake lowered mean body weight
Investigate the effects of fruit 63 subjects with type 2 by 0.8 kg more than the lower fruit diet
intake on HbA1c, body weight, diabetes; mean age 58 years; (p = 0.19; 2.5 kg vs. 1.7 kg) and mean waist
and waist size in people with mean BMI 32; 78% male; diet size was lowered by 1.3 cm more for the
type 2 diabetes (Denmark) [86] >2 servings vs. <2 servings higher vs. lower fruit intake (p = 0.36; 4.3 cm
fruit daily, difference in fruit vs. 3.0 cm)
intake 172 g; 12 weeks
Dow et al. (2012). Parallel RCT Red grapefruit was associated with modest
Evaluate the effect of red 74 adults; mean BMI 32; 1/2 weight loss by 0.6 kg, significantly reduced
grapefruit on body weight, red grapefruit consumed 3 waist size by 2.45 cm and significantly
blood pressure and blood lipids times daily vs. control diet; improved systolic blood pressure by 3.2 mm
(US) [87] 6 weeks Hg and reduced LDL-C by 18.7 mg/dL
Peterson et al. (2011). Crossover RCT Blood lipids and lipoproteins remained
Evaluate the effect of the 102 adults; mean age 55 years; unchanged with the addition of figs. Body
consumption of dried California 69% females; dried California weight insignificantly increased by 0.4 kg
Mission figs (Ficus carica Mission Figs (120 g/day) (p = 0.08)
‘Mission’) on serum lipid levels added to their usual diet vs.
and body weight (US) [88] their usual diet; 5 weeks
Basu et al. (2010). Parallel RCT Blueberry supplementation did not
Examine the effects of 48 subjects; mean age significantly affect body weight or waist size
blueberries on metabolic 50 years; mean BMI 38; freeze (p > 0.05)
syndrome parameters and body dried blueberries equivalent to
weight (US) [89] 350 g blueberries/day vs.
control; 8 weeks
Whybrow et al. (2007). Parallel RCT Increased fruits and vegetables intake reduced
Examine the effects of 34 males and 28 females; body weight from 0.14 to 0.29 kg compared to
incorporating fruits and mean age 43 years; mean BMI an increase in body weight for the no fruit and
vegetables into diets on body 24; 300 g or 600 g fruits and vegetable control by 0.48 kg (p = 0.242)
weight (Scotland) [90] vegetables daily vs. no fruits
and vegetables control;
isocaloric diets; 8 weeks
De Oliveira et al. (2003). Parallel RCT Compared to baseline, women consuming
Investigate effect of fruit intake 49 women, mean age 44 years; apples and pears lost 0.34 kg more body
on body weight change (Brazil) mean weight 79 kg; dietary weight than women consuming oatmeal
[91] supplements: apples or pears cookies (p = 0.004; 1.22 kg vs. 0.88 kg)
vs. oat cookies three times
daily; 12 weeks
associated with increases in adiposity [76]. men and women; mean age baseline about
However, three studies that investigated French 48 years, mean BMI approximately 25; follow-up
fries separately showed a positive association of 4 year cycles) observed highly variable effects
between intake of French fries and measures of of specific fruits and vegetables on weight change,
adiposity and two of these studies indicated that after adjustment for covariates [82]. Increased
intake of French fries had a stronger positive asso- intake of fruits was associated with 4-year weight
ciation with both BMI and weight gain than boiled, loss per one daily serving for total fruits by
baked, or mashed potatoes. Increasing fruits and 0.53 lb, berries by 1.11 lb, and apples and pears
vegetables intake as a weight loss strategy has pro- by 1.24 lbs (Fig. 8.5). Increased intake of several
duced conflicting results mainly associated with vegetables was also associated with weight loss
the inclusion or exclusion of 100% fruit juice, rais- per one daily serving for total vegetables by
ing the possibility that these divergent findings 0.25 lb, tofu/soy by 2.47 lbs and cauliflower
may be the result of the lower satiation signals by1.37 lbs (Fig. 8.6). In contrast, increased intake
when consuming fruits and vegetables as juice of starchy vegetables, including corn, peas, and
compared to their whole or minimally processed potatoes, was associated with weight gain
forms [43, 78, 79]. A meta-analysis of prospective (Fig. 8.7). Vegetables having higher fiber and fla-
studies (17 cohorts; 563,277 participants; follow- vonoids, and lower energy density and glycemic
up from 9 months to 20 years) found that each load were more strongly associated with weight
daily 100 g fruit had a modest lowering effect on loss compared with lower-fiber, and higher energy
body weight by 14 g per year and that fruit intake dense and higher-glycemic-load vegetables. The
was inversely associated with waist size and a Women’s Health Study (18,146 women; mean
higher intake significantly lowered the risk of adi- baseline age 54 years; mean 15.9 years of follow-
posity by 9% [77]. Although this weight reduction up) found that greater intake of fruit, but not veg-
is modest it represents a composite of all fruits and etables by middle-aged and older women with a
vegetables, and does not rule out the potential for normal BMI was associated with lower risk of
specific healthy low energy dense, fiber and flavo- becoming overweight or obese [83]. The EPIC
noids rich fruits and vegetables to have much Study (373,803 participants; mean baseline age
larger beneficial lowering effects on body weight 52 years; mean follow-up of 5 years) found that
and central obesity. A meta-analysis of whole or fruits and vegetables intake was associated with
minimally processed fruits and vegetables, exclud- weight loss in men and in women who quit smok-
ing 100% fruit juice, (8 RCTs; 1026 participants; ing during the follow-up period [84]. There was a
mean 14.7 weeks) showed a significant reduction weak association between fruit intake and weight
in body weight by 0.68 kg compared to isocaloric loss in women >50 years, normal weight or never
lower fruits and vegetables intake in diets [79]. smokers. A longitudinal study in preschool chil-
Additionally, a systemic review of energy- dren (8950 children; examined at ages 2, 4, and
restricted diets (12 intervention trials and 11 longi- 5 years) reported that regular consumption of
tudinal studies) demonstrated in intervention trials 100% juice between ages 2 and 4 years was asso-
that higher intake of fruits and vegetables was ciated with a significant 30% increased risk of
associated with modest weight loss in overweight being overweight but not at age 5 years [81].
and obese adults but not in children [80].
8.2.2.4 Randomized Trials
8.2.2.3 Prospective Cohort Studies Seven representative RCTs related to increased
Table 8.3 provides a summary of fruits and vege- fruits and vegetables intake and change in body
tables in weight regulation [81–84]. Several pro- weight and waist circumference are summarized
spective studies showed that specific types of in Table 8.3 [85–91]. An Australian energy weight
fruits and vegetables vary in their effect on body loss RCT (120 subjects; mean age 49 years; mean
weight. Pooled data from three US prospective BMI 30; 12 months) showed that increased vege-
studies including the Nurses’ Health Studies and table intake significantly lowered body weight,
Health Professionals Follow-up Study (133,468 especially with low energy dense vegetables
Oranges
Bananas
Avocado
Raisins & grapes
Strawberries
Apples & pears
Blueberries
Total fruit
–1.5 –1.3 –1.1 –0.9 –0.7 –0.5 –0.3 –0.1

Weight loss (lbs) for each daily serving over 4 years
Fig. 8.5 Association between one daily serving of various fruits and weight loss in US men and women over 4 years
from the Nurses’ Health Studies and Health Professionals Follow-up Study (pooled data; multivariate adjusted) (adapted
from [82])
Tomatoes
Celery
Carrots
Brussels sprouts
Broccoli
Peppers
Cauliflower
Soy & tofu
Total vegetables
–3 –2.8 –2.6 –2.4 –2.2 –2 –1.8 –1.6 –1.4 –1.2 –1 –0.8 –0.6 –0.4 –0.2 0
Weight loss (lbs) for each daily serving over 4 year
Fig. 8.6 Association between each daily serving of various vegetables and weight loss in US men and women over
4 years from the Nurses’ Health Studies and Health Professionals Follow-up Study (pooled data; multivariate adjusted)
(adapted from [82])
Corn
Peas
Potatoes*
Cabbage
Onions
Winter squash
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4
Weight gain (lbs) for each daily serving over 4 years
Fig. 8.7 Association between each daily serving of commonly consumed vegetables and weight gain in US men and
women over 4 years from the Nurses’ Health Studies and Health Professionals Follow-up Study (pooled data; multivari-
ate adjusted) (adapted from [82]). *Includes baked, boiled, mashed white potatoes, sweet potatoes and yams and
excludes French fries and potato chips
5 Daily Half Vegetable Portions 5 Daily Full Vegetable Portions

0.0
0 3 6 9 12
Study Duration (months)
–2.0
Weight Change (kg)
–4.0
–6.0
–8.0
Fig. 8.8 Effect of amount of vegetables added to a diet on weight loss diet plan (20% reduced energy) in 120 obese
adults over 12 months (p = 0.024) (adapted from [85])
(Fig. 8.8) [85]. A Brazilian RCT (49 women; half red grapefruit three times per day modestly
mean age 44 years; 12 weeks) found that snacking but significantly lowered body weight by 0.6 kg
on apples or pears significantly reduced body and waist size by 2.45 cm after 6 weeks [87],
weight by 0.34 kg more than women snacking on whereas blueberries and figs were associated with
oatmeal cookies [91]. Three other RCTs showed an insignificant increase in body weight [88, 89].
inconsistent effects of specific fruit intake on A Danish RCT (63 subjects with type 2 diabetes;
weight and waist size: (1) increased intake of a mean age 58 years; >2 fruit servings compared to
<2 fruit servings; 12 weeks) reported that subjects Study (120,784 men and women; protein intake
with higher fruit intake (by a mean intake of and weight changes in 4 year cycles; 16–24 years
172 g) lost more body weight by 0.8 kg and had a of follow-up) showed that increased protein
lower waist size by 1.3 cm compared to lower foods per serving/day had different relations with
fruit intake but these reductions were insignificant 4 year weight gain, with significant weight gain
[86]. A Scottish dose-response RCT found that for red meats, chicken with skin, and regular
increasing fruit and vegetables intake to 300 g and cheese by 0.13–1.17 kg; no association for sugar
600 g reduced body weight by up to 0.29 kg sweetened yogurt, legumes, or eggs; and signifi-
whereas the diet with 0.0 g fruits and vegetables cant weight loss for plain yogurt, peanut butter,
increased body weight by 0.48 kg (p = 0.24) [90]. nuts, chicken without skin, low-fat cheese, and
seafood by −0.14 to −0.71 kg (Fig. 8.9) [93].
8.2.3 Protein Foods

8.2.4 Legumes
Protein rich whole or minimally processed plant
foods have been considered potentially protective 8.2.4.1 Background
against long-term weight gain and obesity by Legumes including dietary pulses (e.g., pinto
increasing satiety and increasing energy metabo- beans, split peas, lentils, chickpeas) and soybeans,
lism [92]. A pooled analysis of three large US are rich in fiber and protein with relatively low
prospective cohorts including the Nurses’ Health glycemic response properties [94]. A serving of
Studies and the Health Professionals Follow-up legumes is half a cup or 90–100 g cooked legumes,
Cheese
Plain Yogurt
Sweetened Flavored Yogurt
Peanut Butter
Nuts
Legumes
Daily Serving
Eggs
Seafood
Chicken without Skin
Chicken with Skin
Hot Dog
Beef as Main Dish
Lean Hamburger
Regular Hamburger
–1 –0.5 0 0.5 1 1.5

4 Year Weight Change (kg)
Fig. 8.9 Association between each daily serving of protein foods and weight change in US men and women over
4 years from the Nurses’ Health Studies and Health Professionals Follow-up Study (pooled data; multivariate adjusted)
(adapted from [93])
which contains 5–10 g of fiber, 7–8 g of protein, primarily to a high unsaturated fat content and
and <5% of energy as fat, with the exception of low water content, human studies have shown
chickpeas and soybeans which have 15% and 47% that nuts have a lower metabolizable energy than
energy from fat, respectively. Pulses promote sati- predicted from the Atwater energy tables because
ety by adding bulk, and high levels of fiber (e.g., of the incomplete absorption of nut fat and other
resistant starch) and protein, especially as a replace- macronutrients [99, 100]. Human studies consis-
ment for meat products [38]. An NHANES cross- tently report that the regular consumption of tree
sectional study found that bean consumers had nuts, as a replacement for less healthful foods,
significantly lower body weight and a 22% lower can help prevent weight gain [101, 102].
risk of being obese than non-consumers [95]. Mechanistic studies indicate that nuts’ weight
However, pulse consumption has been in decline control effects are largely attributable to their
with the global shift to Western-style diets [96]. high satiety and low metabolizable energy (poor
For example, between the 1960s and 1990s, legume bioaccessibility leading to inefficient energy
intake decreased by 40% in India and by 24% in absorption) properties [103].
Mexico. Legumes are infrequently consumed by
North Americans and northern Europeans, with 8.2.5.2 Prospective Cohort Studies
<8% of Americans consuming pulses daily. Prospective studies consistently show that
increased nut consumption is protective against
8.2.4.2 Randomized Trials weight gain and obesity [104, 105]. The Spanish
Dietary pulses promote modest weight-loss even in Seguimiento Universidad de Navarra (SUN)
non-energy restricted diets [97]. A meta-analysis project (8865 men and women; mean baseline
(21 RCTs; median intake 1 serving (132 g)/day; age 38 years; mean BMI 23; 28 months) indicates
940 participants; median 6 weeks) showed a that people who consumed nuts >2 times/week
significant weight reduction by 0.34 kg for diets had a 40% lower multivariate risk for weight gain
containing dietary pulses compared with control compared with non-nut consumers who gained
diets without them [97]. This modest weight loss an average of 0.4 kg more over the 28 months of
with dietary pulse intake was demonstrated in both follow-up [104]. The Nurses’ Health Study II
energy-restricted diets and in diets intended for (51,188 women in the Nurses’ Health Study II;
weight maintenance. Six trials (509 participants) mean baseline age 36 years; mean BMI 24;
reported that dietary pulse consumption insignifi- 8 years of follow-up) found that women consum-
cantly reduced waist size by 0.37 cm. A trend was ing nuts >2 times/week had a significant 33%
shown in six trials (340 participants) that supported lower risk of obesity and gained 0.51 kg less
lower body fat by 0.34% (p = 0.07) [97]. These weight compared to non-nut consumers over
findings are generalizable to overweight and obese 8 years of follow-up [105]. The different effects
populations suggesting that one daily serving of of total nuts, peanuts and tree nuts intake on obe-
dietary pulses does not lead to weight gain and sity risk are illustrated in Fig. 8.10. A 2017 EPIC
may support modest weight loss. study (373,293 adults; age range 25–70 years; 5
years) showed that higher nut intake significantly
reduced weight gain by 0.07 kg compared to the
8.2.5 Total and Specific Nuts average weight gain of 2.1 kg for the overall pop-
ulation [106]. Also, this study showed that the
8.2.5.1 Background highest nuts consumers had a significantly 5%
Nuts (e.g., almonds, pistachios, walnuts, hazel- lower risk of becoming overweight or obese.
nuts, pecans, peanuts) are nutrient dense sources
of fiber, protein, unsaturated fat, vitamins (e.g., 8.2.5.3 Randomized Controlled Trials
B-vitamins and vitamin E), minerals (e.g., potas- RCTs consistently show that diets supplemented
sium and magnesium), and phytosterols, and with nuts do not increase body weight, body mass
polyphenols [35, 98]. Although nuts have a rela- index, or waist circumference compared with
tively high energy density (about 6 kcal/g) due control diets [107–116].
Total nuts p-trend =.003 Peanuts p-trend = .094

1.1
1.05
Hazard Ratio for Obesity 1
0.95
0.9
0.85
0.8
0.75
0.7
0.65
0.6
Never/almost never 1-3 times/month 1 time/week >= 2 times/week
Nut Intake Frequency
Fig. 8.10 Association between nut intake frequency and obesity risk in healthy middle-aged women over 8 years
(adapted from [105])
Meta-Analysis weight gain [110]. A long-term weight loss RCT

A systematic review and meta-analysis (33 pub- (123 subjects; mean baseline age 47 years; mean
lished RCT; 75% almonds and walnuts; 1866 BMI 34; about 90% women; hypocaloric diet with
subjects; 2–152 weeks of duration) found that 56 g almonds as snacks daily, vs. no added snack,
increased nut intake resulted in an insignificant nut free diet; 18 months) demonstrated that there
decreased mean body weight by 0.47 kg, BMI by was no significant 18-month weight change
0.40, and waist circumference by 1.25 cm [107]. between the two diets with the almond group los-
Although the decreases in weight and body com- ing 3.7 kg vs. the unsupplemented group losing
position are relatively small, the results reduce 5.9 kg [111]. Two shorter-term weight loss RCTs
any concerns that eating nuts may promote obe- (65–108 subjects; 50–84 almonds/daily vs. nut
sity in general or when eaten as a cardioprotec- free diets or diets rich in complex carbohydrates;
tive food. Almonds and walnuts represent 75% of 3–6 months) found significantly lower body
the weight management research. weight and BMI with almond consumption com-
pared to control diets [112, 113].
Almonds
Almond RCTs have consistently shown that Walnuts
almonds added to the habitual diet non-significantly Walnuts have similar types and numbers of
increase body weight or when incorporated into a weight loss studies as almonds and show that
hypocaloric diet promote significant additional adding walnuts to the habitual diet does not sig-
weight loss [108–113]. Two crossover RCTs dem- nificantly cause weight gain. A crossover RCT
onstrate that habitual diets supplemented with two (90 subjects; mean age 54 years; mean BMI 26;
servings or 320–344 kcals of almonds daily, com- habitual diet plus added walnuts to 12% of energy
pared to unsupplemented control diets for or 28–56 g, or no added walnuts; 6 months) dem-
10 weeks to 6 months did not increase body weight onstrated that the walnut group consumed a net
(Fig. 8.11) [108, 109]. A parallel RCT in subjects mean increase of 133 kcals/day compared to the
at increased risk for type 2 diabetes (137 subjects; control group, resulting in an insignificant weight
48 males and 89 females; mean age approx. gain of 0.4 kg vs. the 3.1 kg theoretically calcu-
30 years; almonds (43 g/day) with breakfast or lated weight gain that had been projected for that
lunch, alone as a morning or afternoon snack or no time period and intake level [114]. Another
almonds control; 4 weeks) showed that almonds crossover RCT (46 subjects; mean age 57 years;
reduced hunger and did not increase the risk of mean BMI 33; 28 women and 18 men; 56 g
Control (no added almonds) 344 kcals almonds added daily

75
Body Weight Change (kg)

72.5
70
67.5
65
Baseline 1 10
Weeks
Fig. 8.11 Effect of adding two servings (344 kcals) of almonds to the habitual diet of 24 women on change in body
weight over 10 weeks (p > 0.05) (adapted from [109])
walnuts or 350 kcals/day added to the habitual rich in whole or minimally processed plant
diet vs. no walnuts added to the habitual diets; foods (whole plant foods) tend to be associ-
8 weeks) found that the walnut free diet resulted ated with a lower risk of weight gain and obe-
in small but significant mean reductions in BMI sity compared to the more common Western
by 0.4 units and body weight by 1 kg compared diets high in processed foods. Prospective
to the walnut added diet. The BMI and weight cohort studies show >3 daily whole-grain
increases in the walnut group were much less servings (especially with total cereal fiber at
than expected based on the added energy intake approximately10 g/day), can reduce body
[115]. A weight loss RCT (245 women; mean weight and waist size compared to < one half
age 50 years; mean BMI 33; 42 g walnuts added; serving/day. RCTs indicate that whole-grains
6 months) found that women consuming walnuts are more effective in reducing body fat and
(35% energy from fat), lower fat (20% energy) waist size than body weight or BMI. For fruits
and low carbohydrate (45% energy from fat) and vegetables, cohort studies find an associa-
weight loss diets had similar weight loss by tion with a lower risk of weight, waist size or
approx 6 kg and reduction in BMI by 3 units body fat gain and obesity, especially for
(p < 0.05) but the walnut diet resulted in the most healthier varieties. However, higher energy
favorable changes in blood lipids [116]. dense, lower fiber fruits and vegetables may
promote weight gain. RCTs indicate that
Conclusions lower energy dense, higher fiber and flavo-
The worldwide overweight and obesity pan- noid rich fruits and vegetables can support
demic is among the greatest public health lower risk of weight gain or modest weight
challenges of our time with over two billion loss and promote additional weight loss in a
people overweight or obese globally now. hypocaloric diet or help to support weight
Foods commonly associated with weight gain maintenance after weight loss. RCTs show
are the high intake of French fries, sugar- that the daily consumption of dietary pluses
sweetened beverages, and red and processed and nuts do not promote weight gain, and may
meats, and the foods that tend to be inversely support modest weight loss. Nuts consumed
associated with weight gain are non-starchy as a snack or legumes as a meal protein source
vegetables, high fiber and flavonoid rich in weight loss diets do not appear to interfere
fruits, whole grains, nuts, and plain yogurt. with weight loss or weight maintenance after
Healthy lower energy dense dietary patterns weight loss.
ppendix A: Estimated range of energy, fiber, nutrients and phytochemicals

A
composition of whole plant foods/100 g edible portiona, b
Nutrients/ Wheat, oats, Apples, Dates, dried Potatoes, Lentils, Almonds,
phytochemicals brown rice, pears, figs, spinach, chickpeas, Brazil nuts,
whole grain bananas, apricots, carrots, split peas, cashews,
bread, grapes, cranberries, peppers, black beans, hazelnuts,
cereal, pasta, oranges, raisins, and lettuce, green pinto beans, macadamias,
rolls, and blueberries, prunes beans, and soy pecans,
crackers strawberries, cabbage, beans walnuts,
and onions, peanuts,
avocados cucumber, sunflower
cauliflower, seeds, and
mushrooms, flaxseed
and broccoli
Energy (kcals) 110–350 30–170 240–310 10–115 85–170 520–700
Protein (g) 2.5–16 0.5–2.0 0.1–3.4 0.2–5.0 5.0–17 7.8–24
Available 23–77 1.0–25 64–82 0.2–25 10–27 12–33
Carbohydrate (g)
Fiber (g) 3.5–18 2.0–7.0 5.7–10 1.2–9.5 5.0–11 3.0–27
Total fat (g) 0.9–6.5 0.0–15 0.4–1.4 0.2–1.5 0.2–9.0 46–76
SFAa (g) 0.2–1.0 0.0–2.1 0.0 0.0–0.1 0.1–1.3 4.0–12
MUFAa (g) 0.2–2.0 0.0–9.8 0.0–0.2 0.1–1.0 0.1–2.0 9.0–60
PUFAa (g) 0.3–2.5 0.0–1.8 0.0–0.7 0.0.0.4 0.1–5.0 1.5–47
Folate (ug) 4.0–44 <5.0–61 2–20 8.0–160 50–210 10–230
Tocopherols (mg) 0.1–3.0 0.1–1.0 0.1–4.5 0.0–1.7 0.0–1.0 1.0–35
Potassium (mg) 40–720 60–500 40–1160 100–680 200–520 360–1050
Calcium (mg) 7.0–50 3.0–25 10–160 5.0–200 20–100 20–265
Magnesium (mg) 40–160 3.0–30 5.0–70 3.0–80 40–90 120–400
Phytosterols (mg) 30–90 1.0–83 – 1.0–54 110–120 70–215
Polyphenols (mg) 70–100 50–800 – 24–1250 120–6500 130–1820
Carotenoids (ug) – 25–6600 0.6–2160 10–20,000 50–600 0.0–1200
Circulation. 2013;128:553–65
USDA. What We Eat in America, NHANES 2011–2012, individuals 2 years and over (excluding breast-fed children).
Rodriguez-Casado A. The health potential of fruits and vegetables phytochemicals: notable examples. Crit Rev Food
Sci Nutr. 2016;56(7):1097–107
An. 1999;12:169–96
tent during maturation and from different growing areas. J Agric Food Chem. 2009;57(21):10408–13
a
b
Nutrient Database for Standard Reference, Release 27. http://www.ars.usda.gov/nutrientdata. Accessed 17 Feb 2015
References 253
References 17. Zhai F, Wang H, Wang Z, et al. Closing the energy

gap to prevent weight gain in China. Obes Rev.
2008;9(Suppl 1):107–12.
1. Haslam D. Weight management in obesity—past
18. Davis JN, Hodges VA, Gillham MB. Normal-weight
and present. Int J Clin Pract. 2016;70(3):206–17.
adults consume more fiber and fruit than their age
2. World Health Organization. Obesity and overweight.
and height matched overweight/obese counterparts.
Geneva (Switzerland). 2014. http://who.int/media-
J Am Diet Assoc. 2006;106:835–40.
centre/factsheets/fs311en/. Accessed 18 Jan 2015.
3. Trust for America’s Health and the Robert Wood 19. Centers for Disease Control and Prevention. Low
Johnson Foundation. F as in fat: how obesity threat- energy dense foods and weight management: cut-
ens America’s future. 2012. http://healthyameri- ting calories while controlling hunger. Research to
cans.org/assets/files/TFAH2012FasInFat18.pdf. Practice Series, No 5. http://www.cdc.gov/nccd-
Accessed 27 Feb 2015. php/dnpa/nutrition/pdf/r2p_energy_density.pdf.
4. Swinburn BA, Sacks G, Hall KD, et al. The global Accessed 21 Feb 2015.
obesity pandemic: shaped by global drivers and local 20. Rolls BJ. What is the role of portion in weight man-
environments. Lancet. 2011;378:804–14. agement? Int J Obes. 2014;38:S1–8.
5. de Mutsert R, Sun Q, Willett WC, et al. Overweight 21. Vernarelli JA, Mitchell DC, Rolls BJ, Hartman
in early adulthood, adult weight change, and risk of TJ. Dietary energy density is associated with obe-
type 2 diabetes, cardiovascular diseases, and certain sity and other biomarkers of chronic disease in US
cancers in men: a cohort study. Am J Epidemiol. adults. Eur J Nutr. 2015;54(1):59–65.
2014;179:1353–65. 22. Karl JP, Roberts SB. Energy density, energy intake
6. Stevens J, Erber E, Truesdale KP, et al. Long- and body weight regulations in adults. Adv Nutr.
and short-term weight change and incident cor- 2014;5:835–50.
onary heart disease and ischemic stroke: the 23. Ambrosini GL. Childhood dietary patterns and later
Atherosclerosis Risk in Communities Study. Am J obesity: a review of the evidence. Proc Nutr Soc.
Epidemiol. 2013;178:239–48. 2014;73:137–46.
7. Faris MAE, Attlee A. Obesity and cancer: what’s the 24. Wing RR, Phelan S. Long-term weight loss mainte-
interconnection? Adv Obes Weight Manag Control. nance. Am J Clin Nutr. 2005;82(Suppl):222S–5S.
2015;2(4):1–6. 25. Mariman ECM. An adipobiological model for weight
8. Ogden CL, Carroll MD, Kit BK, et al. Prevalence regain after weight loss. Adipobiology. 2011;3:7–13.
of childhood and adult obesity in the United States, 26. MacLean PS, Higgins JA, Giles ED, et al. The role
2011–2012. JAMA. 2014;311:806–14. for adipose tissue in weight regain after weight loss.
9. Rinella ME. Nonalcoholic fatty liver disease. A sys- Obes Rev. 2015;16(Suppl. 1):45–54.
tematic review. JAMA. 2015;313(22):2263–73. 27. Bell ES, Roll BJ. Energy density of foods affects energy
10. Flegal KM, Graubard BI, William DF, et al. Excess intakes across multiple levels of fat content in lean and
deaths associated with underweight, overweight and obese women. Am J Clin Nutr. 2001;73:1010–8.
obesity. JAMA. 2005;293:1861–7. 28. Raynor HA, Van Walleghen EL, Bachman JL, et al.
11. Allison DB, Fontaine KR, Manson JE, et al. Annual Dietary energy and successful weight loss mainte-
deaths attributable to obesity in the United States. nance. Eat Behav. 2011;12(2):119–25.
JAMA. 1999;282:1530–8. 29. Davey GK, Spencer EA, Appleby PN, et al. EPIC–
12. Hill JO. Can a small-changes approach help Oxford: lifestyle characteristics and nutrient intakes in
address the obesity epidemic? A report of the Joint a cohort of 33,883 meat-eaters and 31,546 non meat-
Task Force of the American Society for Nutrition, eaters in the UK. Public Health Nutr. 2003;6:259–68.
Institute of Food Technologists, and International 30. Rolls BJ, Drewnowski A, Ledikwe JH. Changing the
Food Information Council. Am J Clin Nutr. energy density of the diet as a strategy for weight
2009;89:477–84. management. J Am Diet Assoc. 2005;5(Suppl 1):
13. Williamson DF, Kahn HS, Remington PL, et al. The S98–S103.
10-year incidence of overweight and major weight 31. Ledikwe JH, Blanck HM, Khan LK, et al. Dietary
gain in US adults. Arch Intern Med. 1990;150: energy density is associated with energy intake
665–72. and weight status in US adults. Am J Clin Nutr.
14. Mozaffarian D, Hao T, Rimm EB, et al. Changes 2006;83:1362–8.
in diet and lifestyle and long-term weight gain in 32. Mozaffarian D, Hao T, Rimm EB, et al. Changes in
women and men. N Engl J Med. 2011;363:2392–404. diet and lifestyle and long-term weight gain in women
15. Peeters A, Magliano DJ, Backholer K, et al. Changes and men. Engl J Med. 2011;364(25):2392–404.
in the rates of weight and waist circumference gain 33. Tuso PJ, Ismail MH, Ha BP, Bartolotto C. Nutritional
in Australian adults over time: a longitudinal cohort update for physicians: plant-based diets. Perm J.
study. BMJ Open. 2014;4(1):e003667. https://doi. 2013;17(2):61–6.
org/10.1136/bmjopen-2013-003667. 34. Wright N, Wilson L, Smith M, et al. The BROAD
16. Heitmann BI, Garby L. Patterns of long-term weight study: a randomised controlled trial using a whole
changes in overweight developing Danish men and plant-based diet in the community for obesity,
women aged between 30 and 60 years. Int J Obes ischemic heart disease or diabetes. Nutr Diabetes.
Relat Metab Disord. 1999;23:1074–8. 2017;7:e256.
35. Sutliffe JT, Fuhrman J, Carnot MJ, et al. Nutrient- 51. Giacco R, Pella-Pepo G, Luongo D, et al. Whole
dense, plant-rich dietary intervention effective grain intake in relation to body weight: from epi-
at reducing cardiovascular risk factors for work- demiological evidence to clinical trials. Nutr Metab
sites: a pilot study. Altern Ther Health Med. Cardiovasc Dis. 2011;21:901–8.
2016;2295:24–9. 52. Harland JI, Garton LE. Whole-grain intake as a
36. Allman-Farinelli M, Partridge SR, McGeechan K, marker of healthy body weight and adiposity. Public
et al. A mobile health lifestyle program for preven- Health Nutr. 2008;11:554–63.
tion of weight gain in young adults (TxT2Bfit): nine- 53. Du H, van der AD, Boshuizen HC, et al. Dietary
month outcomes of a randomized control trial. JMIR fiber and subsequent changes in body weight and
Mhealth Uhealth. 2016;4(2):e78. waist circumference in European men and women.
37. Turner-McGrievy G, Mandes T, Crimarco A. A plant- Am J Clin Nutr. 2010;91:329–36.
based diet for overweight and obesity prevention and 54. Bazzano LA, Song Y, Bubes V, et al. Dietary intake
treatment. J Geriatr Cardiol. 2017;14:369–74. of whole and refined grain breakfast cereals and
38. Rebello CJ, Greenway FL, Finley JW. A review of weight gain in men. Obes Res. 2005;13:1952–60.
the nutritional value of legumes and their effects on 55. Koh-Banerjee P, Franz M, Sampson L, et al. Changes
obesity and its related co-morbidities. Obes Rev. in whole-grain, bran, and cereal fiber consumption
2014;15:392–407. in relation to 8-y weight gain among men. Am J Clin
39. Hever J, Cronise RJ. Plant-based nutrition for health- Nutr. 2004;80:1237–45.
care professionals: implementing diets as a primary 56. Liu S, Willett WC, Manson JE, et al. Relation
modality in the prevention and treatment of chronic between changes in intakes of dietary fiber and grain
disease. J Geriatr Cardiol. 2017;14:355–68. products and changes in weight and development of
40. McMacken M, Shah S. A plant-based diet for the obesity among middle-aged women. Am J Clin Nutr.
prevention and treatment of type 2 diabetes. J Geriatr 2003;78:920–7.
Cardiol. 2017;14:342–54. 57. Steffen LM, Jacobs DR Jr, Murtaugh MA, et al.
41. de Freitas Junior LM, de Almeida Jr EB. Medicinal Whole grain intake is associated with lower body
plants for the treatment of obesity: ethnopharmaco- mass and greater insulin sensitivity among adoles-
logical approach and chemical and biological stud- cents. Am J Epidemiol. 2003;158:243–50.
ies. Am J Transl Res. 2017;9(5):2050–64. 58. Pol K, Christensen R, Bartels EM, et al. Whole
42. Poulsen SK, Due A, Jordy AB, et al. Health effects grain and body weight changes in apparently healthy
of the New Nordic diet in adults with increased waist adults: a systematic review and meta-analysis of
circumference: a 6-mo randomized controlled trial. randomized controlled studies. Am J Clin Nutr.
Am J Clin Nutr. 2014;99:35–45. 2013;98:872–84.
43. Cavallo DN, Horino M, McCarthy WJ. Adult intake 59. Ampatzoglou A, Atwal KK, Maidens CM, et al.
of minimally processed fruits and vegetables: asso- Increased whole grain consumption does not affect
ciations with cardiometabolic disease risk factors. J blood biochemistry, body composition, or gut micro-
Acad Nutr Diet. 2016;116(9):1387–94. https://doi. biology in healthy, low-habitual whole grain con-
org/10.1016/j.jand.2016.03.019. sumers. J Nutr. 2015;145:215–21.
44. Dietary Guidelines Advisory Committee. Scientific 60. Shimabukuro M, Higa M, Kinjo R, et al. Effects
report. Advisory report to the Secretary of Health and of the brown rice diet on visceral obesity and
Human Services and the Secretary of Agriculture. endothelial function: the BRAVO study. Br J Nutr.
Part D. Chapter 1: Food and nutrient intakes, and 2014;111:310–20.
health: current status and trends. 2015; p. 1–78. 61. Brownlee IA, Moore C, Chatfield M, et al. Markers
45. Slavin J. Why whole grains are protective: biologi- of cardiovascular risk are not changed by increased
cal mechanisms. Proc Nutr Soc. 2003;62:129–34. whole-grain intake: The WHOLEheart Study, a ran-
46. Seal CJ, Brownlee IA. Whole-grain foods and domised, controlled dietary intervention. Br J Nutr.
chronic disease: evidence from epidemiologi- 2010;104:125–34.
cal and intervention studies. Proc Nutr Soc. 62. Harris Jackson K, West SG, Heuvel JPV, et al.
2015;74:313–9. Effects of whole and refined grains in a weight-loss
47. http://wholegrainscouncil.org/whole-grains-101/ diet on markers of metabolic syndrome in indi-
What-counts-as-a-serving. Accessed 26 Dec 2015. viduals with increased waist circumference: a ran-
48. Karl JP, Saltzman E. The role of whole grains in domized controlled-feeding trial. Am J Clin Nutr.
body weight regulation. Adv Nutr. 2012;3:697–707. 2014;100:577–86.
49. http://health.gov/dietary guidelines/2015/guide- 63. Kristensen M, Toubro S, Jensen MG, et al. Whole
lines/. Accessed 26 Jan 2016. grain compared with refined wheat decreases the
50. McGill CR, Fulgoni VL III, Devareddy L. Ten-year percentage of body fat following a 12-week, energy-
trends in fiber and whole grain intakes and food restricted dietary intervention in postmenopausal
sources for the United States population: National women. J Nutr. 2012;142:710–6.
Health and Nutrition Examination Survey 2001– 64. Maki KC, Beiseigel JM, Jonnalagadda SS, et al.
2010. Forum Nutr. 2015;7:1119–30. Whole-grain ready-to-eat oat cereal, as part of a
References 255
dietary program for weight loss, reduces low-density 77. Schwingshackl L, Hoffmann G, Kalle-Uhlmann T,
lipoprotein cholesterol in adults with overweight et al. Fruit and vegetable consumption and changes
and obesity more than a dietary program includ- in anthropometric variables in adult populations: a
ing low-fiber control foods. J Am Diet Assoc. systematic review and meta-analysis of prospective
2010;110:205–14. cohort studies. PLoS One. 2015;10(10):e0140846.
65. Katcher HI, Legro RS, Kunselman AR, et al. The https://doi.org/10.1371/journal.pone.0140846.
effects of a whole grain–enriched hypocaloric diet 78. Kaiser KA, Brown AW, Bohan Brown MM, et al.
on cardiovascular disease risk factors in men and Increased fruit and vegetable intake has no discern-
women with metabolic syndrome. Am J Clin Nutr. ible effect on weight loss: a systematic review and
2008;87:79–90. meta-analysis. Am J Clin Nutr. 2014;100(2):567–76.
66. Kim JY, Kim JH, Lee DH, et al. Meal replacement 79. Mytton OT, Nnoaham K, Eyles H, et al. Systematic
with mixed rice is more effective than white rice in review and meta-analysis of the effect of increased
weight control, while improving antioxidant enzyme vegetable and fruit consumption on body weight and
activity in obese women. Nutr Res. 2008;28:66–71. energy intake. BMC Public Health. 2014;14:886.
67. Melanson KJ, Angelopoulos TJ, Nguyen VT, et al. https://doi.org/10.1186/1471-2458-14-886.
Consumption of whole-grain cereals during weight 80. Ledoux TA, Hingle MD, Baranowski T. Relationship
loss: effects on dietary quality, dietary fiber, magne- of fruit and vegetable intake with adiposity: a sys-
sium, vitamin B-6, and obesity. J Am Diet Assoc. tematic review. Obes Rev. 2011;12:e143–50.
2006;106:1380–8. 81. Shefferly A, Scharf RJ, DeBoer MD. Longitudinal eval-
68. Saltzman E, Moriguti JC, Das SK, et al. Effects of a uation of 100% juice consumption on BMI status in
cereal rich in soluble fiber on body composition and 2–5-year-old children. Pediatr Obes. 2016;11(3):221–7.
dietary compliance during consumption of a hypoca- 82. Bertoia ML, Mukamal KJ, Cahill LE, et al. Changes
loric diet. J Am Coll Nutr. 2001;20:50–7. in intake of fruits and vegetables and weight change
69. Slavin JL, Lloyd B. Health benefits of fruits and veg- in United States men and women followed for up
etables. Adv Nutr. 2012;3:506–16. to 24 years: analysis from three prospective cohort
70. United States Department of Agriculture. 2013. studies. PLoS Med. 2015;12(9):e1001878. https://
Choose my plate. http://www.choosemyplate.gov/. doi.org/10.1371/journal.pmed.1001878.
Accessed 17 Feb 2015. 83. Rautiainen S, Wang L, Lee IM, et al. Higher intake
71. Boeing H, Bechthold A, Bub A, et al. Critical of fruit, but not vegetables or fiber, at baseline is
review: vegetables and fruit in the prevention of associated with lower risk of becoming overweight
chronic diseases. Eur J Nutr. 2012;51:637–63. or obese in middle-aged and older women of normal
72. WHO/FAO. Diet, nutrition and prevention of BMI at baseline. J Nutr. 2015;145(5):960–8. https://
chronic disease: report of a Joint WHO/FAO Expert doi.org/10.3945/jn.114.199158.
Consultation. Geneva, Switzerland: World Health 84. Vergnaud A-C, Norat T, Romaguera D, et al. Fruit
Organization. 2003/2004. http://whqlibdoc.who.int/ and vegetable consumption and prospective weight
trs/WHO_TRS_916.pdf. Accessed 17 Feb 2015. change in participants of the European Prospective
73. World Health Association. Global strategy on Investigation into Cancer and Nutrition–Physical
diet, physical activity and health—promoting fruit Activity, Nutrition, Alcohol, Cessation of Smoking,
and vegetable consumption around the world. Eating Out of Home, and Obesity study. Am J Clin
2013. http://who.int/dietphysicalactivity/fruit/en/. Nutr. 2012;95:184–93.
Accessed 17 Feb 2015. 85. Tapsell LC, Batterham RL, Thorne RL, et al. Weight loss
74. Micha R, Khatibzadeh S, Shi P, et al. Global, regional effects from vegetable intake: a 12-month randomised
and national consumption of major food groups in controlled trial. Eur J Clin Nutr. 2014;68:778–85.
1990 and 2010: a systematic analysis including 266 86. Christensen AS, Viggers L, Hasselström K, Gregersen
country-specific nutrition surveys worldwide. BMJ S. Effect of fruit restriction on glycemic control in
Open. 2015;5(9):e008705. https://doi.org/10.1136/ patients with type 2 diabetes-a randomized trial. Nutr J.
bmjopen-2015-008705. 2013;12:29. https://doi.org/10.1186/1475-2891-12-29.
75. Bertoia ML, Rimm EB, Mukamal KJ, et al. Dietary 87. Dow CA, Going SB, Chow H-H, et al. The effects of
flavonoid intake and weight maintenance: three pro- daily consumption of grapefruit on body weight, lip-
spective cohorts of 124,086 US men and women fol- ids, and blood pressure in healthy, overweight adults.
lowed for up to 24 years. BMJ. 2016;352:i17. https:// Metabolism. 2012;61:1026–35.
doi.org/10.1136/bmj.i17. 88. Peterson JM, Montgomery S, Haddad E, et al.
76. Borch D, Juul-Hindsgaul N, Veller M, et al. Potatoes Effect of consumption of dried California mis-
and risk of obesity, type 2 diabetes, and cardiovas- sion figs on lipid concentrations. Ann Nutr Metab.
cular disease in apparently healthy adults: a system- 2011;58:232–8.
atic review of clinical intervention and observational 89. Basu A, Du M, Leyva MJ, et al. Blueberries decrease
studies. Am J Clin Nutr. 2016;104:489–98. https:// cardiovascular risk factors in obese men and women
doi.org/10.3945/ajcn.116.132332. with metabolic syndrome. J Nutr. 2010;140:1582–7.
90. Whybrow S, Harrison CLS, Mayer C, Stubbs 104. Bes-Rastrollo M, Sabate J, Gomez-Gracia E, et al.
RJ. Effects of added fruits and vegetables on dietary Nut consumption and weight gain in a Mediterranean
intakes and body weight in Scottish adults. Br J Nutr. cohort: the SUN Study. Obesity (Silver Spring).
2007;95:496–503. 2007;15:107–16.
91. de Oliveira MC, Sichieri R, Moura AS. Weight 105. Bes-Rastrollo M, Wedick NM, Martinez-Gonzalez
loss associated with a daily intake of three apples MA, et al. Prospective study of nut consumption,
or three pears among overweight women. Nutrition. long-term weight change, and obesity risk in women.
2003;19:253–6. Am J Clin Nutr. 2009;89:1913–9.
92. Westerterp-Plantenga MS, Lemmens SG, Westerterp 106. Freisling H, Noh H, Slimani N, V. Nut intake and
KR. Dietary protein-its role in satiety, energetics, 5-year changes in body weight and obesity risk in
weight loss and health. Br J Nutr. 2012;108(Suppl adults: results from the EPIC-PANACEA study. Eur
2):S105–12. J Nutr. 2017; doi:10.1007/s0094-017-1513-0.
93. Smith JD, Hou T, Ludwig DS, et al. Changes in intake 107. Flores-Mateo G, Rojas-Rueda D, Basora J, et al. Nut
of protein foods, carbohydrate amount and quality, and intake and adiposity: meta-analysis of clinical trials.
long-term weight change: results from 3 prospective Am J Clin Nutr. 2013;97:1346–55.
cohorts. Am J Clin Nutr. 2015;101:1216–24. 108. Fraser GE, Bennett HW, Jaceldo KB, Sabate J.
94. McCrory MA, Hamaker BR, Lovejoy JC, Effect on body weight of a free 76 kilojoules (320
Eichelsdoerfer PE. Pulse consumption, satiety, and calorie) daily supplement of almonds for six months.
weight management. Adv Nutr. 2010;1:17–30. J Am Coll Nutr. 2002;21(3):275–83.
95. Papanikolaou Y, Fulgoni VL III. Bean consumption 109. Hollis J, Mattes R. Effect of chronic consumption
is associated with greater nutrient intake, reduced of almonds on body weight in healthy humans. Br J
systolic blood pressure, lower body weight, and a Nutr. 2007;98:651–56.
smaller waist circumference in adults: results from 110. Tan SY, Matter RD. Appetitive, dietary and health
the National Health and Nutrition Examination effects of almonds consumed with meals or as
Survey 1999–2002. J Am Coll Nutr. 2008;27:569–76. snacks: a randomized, controlled trial. Eur J Clin
96. Messina V. Nutritional and health benefits of dried Nutr. 2013;67:1205–14.
beans. Am J Clin Nutr. 2014;100(suppl):437S–42S. 111. Foster GD, Shantz KL, Veur SSV, et al. A random-
97. Kim SJ, de Souza RJ, Choo VL, et al. Effects of ized trial of the effects of an almond-enriched, hypo-
dietary pulse consumption on body weight: a sys- caloric diet in the treatment of obesity. Am J Clin
tematic review and meta-analysis of randomized Nutr. 2012;96:249–54.
controlled trials. Am J Clin Nutr. 2016;103:1213–23. 112. Abazarfard Z, Salehi M, Keshavarzi S. The effect of
98. Mattes RD, Kris-Etherton PM, Foster almonds on anthropometric measurements and lipid
GD. Impact of peanuts and tree nuts on body profile in overweight and obese females in a weight
weight and healthy weight loss in adults. J Nutr. reduction program. A randomized controlled clinical
2008;138(suppl):1741S–5S. trial. J Res Med Sci. 2014;19(5):457–64.
99. Alper CM, Mattes RD. The effects of chronic peanut 113. Wien MA, Sabate JM, Ikle DN, et al, Almonds vs
consumption on energy balance and hedonics. Int J complex carbohydrates in a weight reduction pro-
Obes. 2002;26:1129–37. gram. Int J Obesity. 2003;27:1365–72.
100. Novotny JA, Gebauer SK, Baer DJ. Discrepancy 114. Abazarfard Z, Eslamian G, Salehi M, Keshavarzi
between the Atwater factor predicted and empiri- S. A randomized controlled trial of the effects of an
cally measured energy values of almonds in human almond-enriched, hypocaloric diet on liver function
diets. Am J Clin Nutr. 2012;96:296–301. tests in overweight/obese women. Iran Red Crescent
101. Tan SY, Dhillon J, Mattes RD. Review of the effects Med J. 2016;18(3):e23628.
of nuts on appetite, food intake, metabolism, and body 115. Sabate J, MacIntyre C, Siapco G, et al. Does regular
weight. Am J Clin Nutr. 2014;100(suppl):412S–22S. walnut consumption lead to weight gain? Br J Nutr.
102. Jackson CL, Hu FB. Long-term associations of nut 2005;94:859–64.
consumption with body weight and obesity. Am J 116 Katz DL, Davidhi A, Ma Y, et al. Effects of walnuts
Clin Nutr. 2014;100(suppl):408S–11S. on endothelial function in overweight adults with
103. Mattes RD, Dreher ML. Nuts and healthy body visceral obesity: a randomized, controlled, crossover
weight maintenance mechanisms. Asia Pac J Clin trial. J Am Coll Nutr. 2012;31(6):415–23.
Nutr. 2010;19(1):137–41.
Dietary Patterns and Whole Plant
Foods in Type 2 Diabetes 9
Prevention and Management
Keywords
Type 2 diabetes • Dietary quality • Prediabetes • Dietary patterns
Mediterranean diet • DASH diet • Vegan diet • Vegetarian diet • Nordic
food index • Western diet • Whole-grains • Fruits • Vegetables • Legumes
Nuts
Key Points health are the Dietary Approaches to Stop

Hypertension (DASH), vegan and the healthy
• A healthy lifestyle including habitual intake of Nordic food index diets.
a high quality dietary pattern, regular physical • Prospective cohort studies show that whole
activity, and weight control are key compo- (minimally processed) plant foods including
nents of type 2 diabetes (diabetes) prevention whole-grains, fruits, vegetables, dietary
and management. pulses, and nuts and flaxseed are significantly
• Prospective cohort studies show that high associated with lower risk of diabetes. For
quality dietary patterns including the whole grains, 3 servings/day reduced diabetes
Alternative Healthy Eating Index score have a risk by 23% and of the whole-grains oats and
significant inverse association with diabetes oat bran are the most effective in managing
risk, and Western dietary patterns have a posi- glycemic control in people with diabetes. For
tive association with risk. fruits and vegetables, higher intake of fruits,
• Higher adherence to the Mediterranean diet especially berries, and green leafy vegetables,
(MedDiet) is associated with a 19–23% yellow vegetables, non-starchy root and cruci-
reduced risk of developing diabetes, while the ferous vegetables are particularly effective in
results of randomized controlled trials (RCTs) lowering diabetes risk. Three weekly servings
show that the MedDiet can reduce risk of dia- of French fries significantly increase diabetes
betes by 30% and can reduce glycosylated risk by 41% compared to only 5% for other
hemoglobin (HbA1c) levels by 0.30–0.47% in forms of potatoes (baked, boiled or mashed).
people with diabetes. Other healthy dietary Higher intake of sugar sweetened fruit juice is
patterns which are effective in reducing diabe- significantly associated with increased diabe-
tes risk and in management of diabetics’ tes risk by 28%, while higher intake of 100%

258 9 Dietary Patterns and Whole Plant Foods in Type 2 Diabetes Prevention and Management
fruit juice is not associated with diabetes risk. which collectively substantially increase health
Higher intake of dietary pulses, peanuts, tree and economic burdens [7–10]. A 2017 meta-
nuts and flaxseed are also associated with analysis (48 cohort studies) found that high
lower diabetes risk. adherence to Western dietary patterns (charac-
• Healthy dietary patterns and specific whole terized by red and processed meats, refined
foods beneficially affect glycemic and car- grains, fried foods, high fat dairy and eggs)
diometabolic risk factors, which are impor- increased diabetes risk by 44% whereas healthy
tant for preventing and managing diabetes, diets (characterized by fruits, vegetables, legumes,
by helping to control body weight, visceral poultry and fish) reduced diabetes risk by 16%
fat, glucose- insulin homeostasis, oxidative [7]. In the USA population, a nationally represen-
stress, inflammation, and endothelial health, tative comparative risk assessment model esti-
lipoprotein concentrations, and blood mated that 45% of all cardiometabolic deaths
pressure. (due to heart disease, stroke, and diabetes) were
associated with low intakes of fruits, vegetables,
nuts and seeds, whole grains, healthy vegetable
9.1 Introduction oils and seafood and excessive intake of sodium,
red and processed meats and sugar sweetened
Type 2 diabetes (diabetes) is a rapidly increasing beverages [10]. In USA population subgroups,
major chronic disease (characterized by insulin suboptimal (unhealthy) diets are associated with
dysfunction in production and utilization) and is higher mortality rates in men than in women, at
a leading cause of premature disability and death younger vs older ages, among blacks and
with a global impact [1]. The prevalence of pre- Hispanics vs whites, and among individuals with
diabetes and diabetes has increased globally in low and medium education vs higher education.
parallel with the rising levels of obesity in adults Unhealthy diets and inactivity associated with
and children, a phenomenon sometimes called increased diabetes risk may cause epigenetic
diabesity [1–5]. Diabesity is on track to be the changes that could lead to perpetuation of inter-
largest epidemic in human history [2]. As of 2017, generational increases in obesity and diabetes
there are > 415 million people in the world with risk [2]. The objective of this chapter is to review
diabetes, which is more than the population of the effects of dietary patterns and healthy whole
the USA [2]. If this global trend continues, by and minimally processed plant foods (whole
2030 about one billion people are expected to be plant foods) in diabetes prevention and
prediabetic and diabetic. Improperly managed, management.
diabetes leads to a number of health issues,
including heart diseases, stroke, blindness, nerve
damage, leg and foot amputations, and death [1– 9.2 Dietary Patterns
5]. Adults (> 35 years of age) with newly diag-
nosed diabetes have a 38% higher risk of all-cause 9.2.1 Overview
mortality compared to people without diabetes
[3]. Adult-onset diabetes usually begins when a A lifestyle that includes a healthy dietary pattern,
person is in his or her mid-50s, but 90% of diabe- regular physical activity, and weight control are
tes risk can be reduced by following a healthy life- key components of diabetes prevention and
style, especially with healthy diets and adequate management (along with appropriate use of phar-
physical activity [1, 4–8]. In the US and also other macotherapy) while the Western lifestyle is asso-
Western populations it is estimated that <10% of ciated with a substantial increase in diabetes risk,
the populations follow a diabetes preventive life- especially in individuals with a high genetic pre-
style [3–5, 9]. Dietary patterns, especially are a disposition [1–4, 11, 12]. Although dietary pat-
major influence on diabetes and related cardio- tern varies across different populations, general
metabolic premature mortality risk, conditions guidance to increase whole plant foods
9.2 Dietary Patterns 259
(e.g., whole-grains, fruit, vegetables, legumes, for achieving metabolic goals; including carbo-
nuts and seeds) and to reduce intake of refined hydrate counting, The MyPlate Method, individ-
carbohydrates, meat, processed meat, and fried ualized meal plans based on percentages of
foods may help reduce the growing global public macronutrients, exchange lists for meal planning,
health burden of diabetes. The Alternate Healthy glycemic index or load, and common healthy
Eating Index (AHEI), which is a measure of dietary patterns including the Mediterranean
adherence to dietary guidelines (higher intake of diet (MedDiet), Dietary Approaches to Stop
vegetables, fruits, nuts and legumes, long-chain Hypertension (DASH), vegetarian, vegan, low
fats, and whole grains; lower intake of sugar carbohydrate, and low fat diets [4, 12–14]. A
sweetened beverages, red and processed meat, summary of common healthy dietary patterns is
trans fat, and sodium; and moderate alcohol con- provided in Appendix A.
sumption) is strongly associated with a lower dia-
betes risk [13]. A review of the findings from the
Nurses’ Health Study on diet and diabetes is 9.2.2 D
iet Nutrient Quality
summarized in Table 9.1 [13]. For diabetes man- and Diabetes Risk
agement, there are a wide range of approaches
for meal planning and dietary patterns that have Table 9.2 summarizes systematic review and
been shown to be clinically effective, with many meta-analyses, individual cohort studies, and a
including a reduced energy intake or controlling randomized controlled trial (RCT) on diet quality
available carbohydrate intake [12]. One diet or a related to adherence to a healthy or Western
person with diabetes should work with their dietary pattern and diabetes risk [15–31].
health care team on eating patterns, preferences,
and metabolic goals. Multiple meal planning 9.2.2.1 Systematic Reviews and Pooled
approaches and eating patterns can be effective or Meta-Analyses
Eight systematic reviews and/or meta-or pooled
analyses of prospective cohort studies show that
diabetes risk has a significant inverse association
Table 9.1 Summary of the findings from the Nurses’ with healthy dietary patterns with a diabetes risk
Health Studies related to diet and type 2 diabetes risk [13]
reduction of 14–39% (high vs. low adherence) and
Increase type 2 Western patterns have a positive association with
diabetes risk Decrease type 2 diabetes risk
an increased risk by 31–300% (high vs. low
Diets low in fiber, Diets rich in fiber, especially
especially cereal from cereal products, and with a adherence), after multivariate adjustments includ-
fiber, and with a low glycemic index or load ing BMI [15–22]. A 2017 systematic review and
high glycemic meta-analysis of the association between intake
index or load of 12 major foods and risk of diabetes (88 cohort
Diets rich in Diets rich in healthy vegetable
studies) found that optimal daily consumption of
animal fats oils
risk-decreasing foods (whole grains, vegetables,
High intake of Higher intake of whole grains;
refined grains 50 g per day of cooked brown fruits, and dairy) results in a 42% reduction in
rice instead of the same amount diabetes risk and high daily intake of risk-increas-
of white rice, was associated with ing foods (red and processed meats, sugar sweet-
a 16% lower diabetes risk
ened beverages and eggs) is associated with a
High intake of Greater intake of green leafy
threefold increased diabetes risk compared to
french fries vegetables and specific whole fruits
rich in flavonoids, such as non-consumption of these foods (Fig. 9.1) [15]. A
blueberries, grapes, apples, and 2016 pooled analysis of the Nurses’ Health
pears Studies and the Health Professionals Follow-up
High alcohol intake Moderate alcohol consumption Study (124,607 men and women; >20 years of
Higher intake of plain yogurt follow-up; 9361 diabetes cases) showed that diet
Regular coffee consumption quality, based on AHEI score assessed over
Table 9.2 Summary of prospective cohort studies on diet quality and type 2 diabetes (diabetes) and prediabetes risk
Systematic reviews and meta-or pooled analyses
Schwingshackl et al. 88 prospective cohort studies Optimal daily consumption of risk-decreasing foods
(2017). (2 servings whole grains; 2–3 servings vegetables;
Systematic review 2–3 servings fruits; 3 servings dairy) results in a 42%
and meta-analysis of reduction in diabetes risk compared to non-
the relation between consumption of these foods. High daily intake of
intake of 12 major risk-increasing foods of 2 servings red meat (170 g/
food groups and risk day), 4 servings processed meat (105 g/day), 3
of diabetes [15] servings sugar sweetened beverages (750 mL/day),
and one egg (55 g/day) is associated with a threefold
increased diabetes risk compared to non-consumption
Ley et al. (2016). 124,607 participants; >20 years of A >10% decrease in AHEI score over 4 years was
Evaluate diet quality follow-up; 9361 cases of diabetes associated with a significantly higher subsequent
changes during a documented; diet quality, based on adjusted diabetes risk of 34%, whereas a >10%
4-year period on Alternate Healthy Eating Index increase in AHEI score was associated with a lower
diabetes incidence (AHEI) score; assessed every risk by 16%. Greater improvement in diet quality
(Nurses’ Health 4 years was associated with lower diabetes risk across
Studies and Health baseline diet quality status for low, medium, or high
Professionals initial diet quality (p-trend <0.001) (Fig. 9.2).
Follow-up Study; Changes in body weight explained 32% of the
US) [16] association between AHEI changes (per 10%
increase) and diabetes risk
Satija et al. (2016). 160,188 women and 40,539 men; This analysis found an inverse association between
Examine the dietary data collected every an overall plant-based diet and diabetes incidence,
association of an 2–4 years using a semi-quantitative which was stronger for a healthier version of the
overall plant-based food frequency questionnaire. plant-based diet (Fig. 9.3). Overall plant based diet
diet and healthy vs. follow-up of >20 years; 16,162 emphasizing plant foods and low in animal foods
less healthy versions diabetes cases (multivariate was associated with about 20% reduction in diabetes
of a plant based diet adjusted including BMI). risk. Healthy plant-based diets had a larger decrease
with diabetes Plant-based diet index: (1) plant in diabetes risk by 34%. Less healthy plant foods
incidence in three foods received positive scores, were associated with a 16% increased diabetes risk
cohort studies while animal foods (animal fats,
(Nurses’ Health dairy, eggs, fish/seafood, poultry/
Studies and Health red meat, miscellaneous animal-
Professionals based foods) received reverse
Follow-up Study; scores
US) [17] Healthful-plant based diet index:
(1) healthy plant foods (whole
grains, fruits, vegetables, nuts,
legumes, vegetable oils, tea/coffee)
received positive scores, while less
healthy plant foods (fruit juices,
sweetened beverages, refined
grains, potatoes, sweets/desserts)
and animal foods received reverse
scores
Maghsoudi et al. Ten cohort studies; 404,528 Adherence to ‘healthy’ dietary patterns (vegetables,
(2015). participants; age range fruits and whole grains) significantly reduced the
Systematically review 27–84 years; 5–23 years of risk of diabetes by 14%, whereas ‘unhealthy’ dietary
prospective cohort follow-up; 18,584 diabetes cases patterns (red and processed meats, high-fat dairy and
studies for the refined grains) increased diabetes risk by 30%.
association between Subgroup analysis showed that unhealthy dietary
dietary patterns and patterns in which foods with high phytochemical
diabetes incidence, content were consumed attenuated diabetes risk to a
and quantify the 6% increase (Fig. 9.4)
effects using a
meta-analysis [18]
McEvoy et al. (2014). Nine cohort studies; 309,430 Pooled results indicated a significant 15% lower
Examine the participants; 4–23 years of diabetes risk for those in the highest category of
association between follow-up; 16,644 diabetes cases healthy dietary pattern (greater fruit, vegetable, and
posteriori derived complex carbohydrates) compared with those in the
dietary patterns and lowest category. Those in the highest category of the
risk of diabetes [19] Western pattern (greater refined carbohydrates,
processed meat, and fried foods) had a significant
41% increased risk of diabetes compared to those in
the lowest category
Alhazmi et al. (2014). 15 cohort studies; 378,525 Healthy dietary patterns significantly reduced mean
Examine the effect of participants; 4–23 years of diabetes risk by 21% and unhealthy patterns
dietary patterns and follow-up; 23,372 diabetes cases increased mean risk by 44% (highest vs. lowest
the risk of diabetes adherence)
[20]
Esposito et al. (2014). 20 cohort studies; 446,213 Leading healthy diets are equally and consistently
Assess the participants; 3.2–23 years of associated with a 20% reduced risk of future diabetes.
association between follow-up; 21,372 diabetes cases The diabetes risk did not differ by geography (USA,
different diets and Europe, and Asia), the duration of follow-up (<10 and
prevention of >10 years), and type of healthy diets (Mediterranean
diabetes [21] and DASH diets, Alternate Healthy Eating Index;
dietary scores) with common components, including
whole grains, fruit, vegetables, nuts, legumes, healthy
table oils (i.e., extra virgin olive oil), protein sources
such as white meat and seafood, little or moderate
alcohol, and reduced intake of red and processed
meats and sugar sweetened beverages
Esposito et al. (2010). Ten cohort studies; 190,000 Healthy dietary patterns (high intake of fruit,
Investigate the role of participants; 2–23 years of vegetables, whole-grains, fish, poultry and low
dietary patterns in follow-up; 8932 diabetes cases intake of red meat, processed foods, sugar-
preventing diabetes sweetened beverages, and starchy foods) were
[22] associated with a 39% reduction in diabetes risk
compared to Western diets
Doostvandi et al. 904 Iranian men and women; mean There was a positive association between Western
(2016). baseline age 39 years; 45% were and traditional scores with change in 2 h-serum
Investigate the men; measurements included: major glucose levels, while the healthy pattern was
association of major dietary patterns were Western, negatively related to changes in fasting serum
dietary patterns with traditional and healthy; fasting serum glucose, 2-h serum glucose, fasting serum insulin and
the risk of impaired insulin, fasting serum glucose, 2-h HOMA-IR. Highest compared with the lowest tertile
glucose and insulin serum glucose, impaired glucose of the Western dietary pattern was accompanied by a
homeostasis during a tolerance and homeostasis model higher risk for development of impaired glucose
3-year follow-up (Iran; assessment of insulin resistance tolerance by 209%; a higher score on the healthy
Tehran Lipid and (HOMA-IR); 3 years of follow-up dietary pattern was associated with a significantly
Glucose Study) [23] (multivariate adjusted) reduced risk of fasting serum insulin by 47%
Cespedes et al. (2016). 101,504 participants included For each diet quality index, a 1-standard-deviation
Compare associations postmenopausal women without higher score was associated with 10–14% lower
of four diet quality diabetes; diets: Alternate MedDiet diabetes risk (p < 0.001). Adjusting for overweight
index scores with Index, Healthy Eating Index 2010, and obesity at baseline attenuated but did not
diabetes risk, by race/ Alternate Healthy Eating Index eliminate associations to 5–10% lower risk per
ethnicity, and with/ 2010, and DASH diet indices; 1-standard-deviation higher score (p < 0.001). For
without adjustment median 15 years of follow-up; all four dietary indices examined, higher scores were
for overweight/ 10,815 diabetes cases (multivariate inversely associated with diabetes risk overall and
obesity at enrollment adjusted) across racial/ethnic groups. All the healthful diets
(US; Women’s were inversely associated with diabetes
Health Initiative) [24]
(continued)
Hong et al. (2016). 2900 adults; mean baseline age A 1-unit increase in the ‘healthy traditional’ pattern
Examine the 51 years; mean BMI 23.5; 3 years score was associated with a decrease of 0.97 mg/dL
associations between of follow-up (multivariate adjusted) in fasting plasma glucose (p = 0.017), while a 1-unit
dietary patterns and increase in the ‘fruits, eggs and juice’ pattern score
diabetes risk (China) was associated with an increase of 0.90 mg/dL in
[25] fasting plasma glucose (p = 0.023). For men, higher
‘fruits, eggs and juice’ pattern was associated with
an 88% greater risk of hyperglycaemia vs. lower. In
women, higher intake of a “healthy traditional”
pattern tertile 3 was associated with a 41% lower
risk compared to lower intake
Kröger et al. (2014). 12,403 diabetes cases and 16,154 Higher adherence to specific dietary patterns,
Assess the controls; mean baseline age commonly characterized by high intake of fruits or
association between 53 years; mean BMI 26; vegetables and low intake of processed meat,
pre-defined dietary 12–20 years of follow-up sugar-sweetened beverages and refined grains
patterns and diabetes (multivariate adjusted) showed a 9–13% lower diabetes risk per standard
risk in European deviation increase
populations (EU;
European Prospective
Investigation into
Cancer and Nutrition
study (EPIC)—
InterAct Study) [26]
Alhazmi et al. (2014). 8370 women; mean baseline age A high Dietary Guideline Index score was associated
Determine the ability 53 years; Dietary Guideline Index; with significantly reduced diabetes risk by 49%
of two diet quality 6 years of follow-up, 311 incident
scores to predict the cases of diabetes (multivariate
incidence of diabetes adjusted)
in women (The
Australian
Longitudinal Study
on Women’s Health)
[27]
Gopinath et al. (2013). 2564 participants; mean age Compared to the highest vs. lowest tertile of total
Assess the association 64 years; mean BMI 26; 10-years diet score, there was a significant 75% decrease in
between diet quality of follow-up (multivariate adjusted) risk of impaired fasting glucose in men. Also, in
and both impaired men, each 2-point increase in total diet score was
fasting glucose and associated with a significant 52% reduction in the
diabetes among older 10-year incidence of impaired fasting glucose. No
adults (The Blue significant associations were observed among
Mountains Eye women or with the 10-year incidence of diabetes
Study; Australia)
[28]
de Koning et al. 41,615 men; questionnaires were Dietary patterns characterized by high intakes of
(2011). mailed to participants every plant-based foods such as whole grains; moderate
Evaluate the effect of 2–4 years; follow-up of 20 years; alcohol; and low intakes of red and processed meat,
diet-quality scores on 2795 cases of diabetes sodium, sugar-sweetened beverages, and trans-fat
diabetes risk in men (multivariate adjusted) reduce diabetes risk by 9–13% per standard
(US—Health deviation. High-quality diets yielded the greatest
Professional reduction in diabetes cases when followed by those
Follow-up Study) with a high BMI
[29]
Fung et al. (2007). 80,029 women; aged 38–63 years; Women with high total diet quality had a significant
Assess the association follow-up of 18 years (multivariate 36% lower diabetes risk. Women whose total diet
between total diet adjusted) improved during follow-up had a lower risk of
quality (Alternate diabetes than women whose low-quality score did
Healthy Eating Index) not change
and risk of diabetes in
women (Nurses’
Health Study; US)
[30]
Randomized controlled trial (RCT)
Kim et al. (2017). Crossover RCT: 49 subjects The Western diets reduced insulin sensitivity
Evaluate the effect of without diabetes; 15 men and 34 compared to a healthy diets in relatively insulin-
Healthy vs. Western women; mean age 36 years; mean resistant individuals. Dramatic improvement in diet
diets on diabetes risk BMI 27; two 4-week weight-stable quality increased insulin sensitivity by about 50% in
factors (Australia) dietary interventions: Western only 4 weeks in people at risk of developing
[31] diet—high in red and processed diabetes, whereas no effect on insulin sensitivity was
meat and refined grains vs. healthy seen in people with a lower risk of developing
diet—high in whole grains, nuts, diabetes
dairy and legumes with no red meat
350
300
250
Type 2 Diabetes Risk (%)
200
150
100
50
0
Optimal daily consumption of High daily intake of
–50 risk-decreasing foods * risk increasing foods **
–100
Fig. 9.1 Association between risk decreasing and risk increasing foods on type 2 diabetes risk (adapted from [15]).
*Optimal daily consumption of risk-decreasing foods (2 servings whole grains; 2–3 servings vegetables; 2–3 servings
fruits; 3 servings dairy) results in a 42% reduction in diabetes risk compared to non-consumption of these foods. **High
daily intake of risk-increasing foods: 2 servings red meat (170 g/day), 4 servings processed meat (105 g/day), 3 servings
sugar sweetened beverages (750 mL/day), and one egg (55 g/day) is associated with a threefold increased diabetes risk
compared to non-consumption of these foods
4 years, showed that >10% decreased score was impact on diabetes risk. Another 2016 pooled data
associated with a significantly 34% higher diabe- from the US Nurses’ Health Studies and the Health
tes risk, whereas a >10% increased score was Professionals Follow-up Study (160,188 women
associated with a significantly 16% lower risk and 40,539 men; >20 years of follow-up) found
(Fig. 9.2) [16]. Change in AHEI score either posi- that specific healthy plant-based diets emphasiz-
tively or negative from baseline had a significant ing foods with higher fiber, lower glycemic index
Decreasing Score Increasing Score

40
30
Diabetes Risk (%)
20
10
0
No Change 3 to 10% > 10%
–10
–20
4-Year Change in AHEI
Fig. 9.2 Association between 4-year changes in Alternate Healthy Eating Index (AHEI) scores and type 2 diabetes
(diabetes) risk in US men and women from pooled data from the Nurses’ Health Studies and Health Professionals
Follow-up Study [16]
Overall plant-based diet index Healthful plant based diet Index

Unhealthful plant-based diet index
1.2
1.1
1
Harzard Ratio for Diabetes
0.9
0.8
0.7
0.6
0.5
0.4
1 2 3 4 5 6 7 8 9 10
Plant-based Diet Quality (Deciles)
Fig. 9.3 Asssociation between type of plant-based dietary index and type 2 diabetes (diabetes) risk based on the pool
analysis of men and women from the Nurses’ Health Studies and Health Professionals Follow-up Study (p-trend <0.001
for all indices) (adapted from [21])
and higher amounts of phytochemicals were more increased diabetes risk by 30% and a generally
effective in lowering diabetes risk than just a gen- healthy pattern significantly reduced risk by 14%.
eral recommendation to increase intake of whole However, in this analysis the regular inclusion of
plant foods (Fig. 9.3) [17]. A 2015 meta-analysis phytochemically rich fruit and vegetables, and
of high quality healthy vs. low quality Western whole-grains to the Western diet reduced risk from
diets (ten cohort studies; 404,528 participants) 30% to an insignificant 6% (Fig. 9.4) [18]. A 2014
showed that a Western pattern significantly meta-analysis of posteriori dietary patterns (nine
50
40
30
Diabetes Risk (%)
20
10
0
Healthy diet Unhealthy diet Unhealthy diet
(high adherence) (moderate/high (low phytochemical-
–10 phytochemical-rich rich foods)
foods)
–20
Fig. 9.4 Association between a healthy dietary pattern and two unhealthy dietary patterns with different levels of
phytochemical-rich foods an d type 2 diabetes risk from a meta-analysis of ten cohort studies (adapted from [18])
cohort studies; 309,430 participants) reported that that high adherence to a range of higher quality
intake of fruit, vegetables and complex carbohy- diets, including the MedDiet, HEI 2010, AHEI,
drates (highest vs. lowest) reduced risk by 15% and DASH diet, all lowered diabetes risk by
and the highest intake of refined carbohydrates, 5–10% with each one standard deviation increase
processed meats, and fried foods (highest to low- in diet quality score [24]. Three cohort studies
est) increased risk by 41% [19]. Another 2014 from Australia, China and Iran found that high
meta-analysis of the adherence to a number of adherence to a Western diet is associated with
healthy diets including the Mediterranean impaired glucose tolerance, high fasting glucose
(MedDiet), DASH, and Alternate Healthy Eating and insulin and impaired insulin sensitivity and
Index diets (20 cohort studies; 446,213 partici- higher adherence to a healthy pattern is associated
pants) found that these healthy diets are basically with healthier glycemic control [23, 25, 28]. The
equally associated with a 20% lower diabetes risk European Prospective Investigation into Cancer
and risk rate was not appreciably different in the and Nutrition study (EPIC)—InterAct Study
US, Europe, or Asia [21]. Two other meta-analy- (12,403 diabetes cases and 16,403 controls; mean
ses demonstrated that high adherence to healthy baseline age 53 years; 12–20 years of follow-up)
dietary patterns reduced diabetes risk by between showed that a healthy dietary pattern was associ-
39 and 65% compared to high adherence to ated with a 9–13% lower diabetes risk per stan-
Western diets [20, 22]. dard deviation increase [26]. The Australian
Longitudinal on Women’s Health (8,370 women;
9.2.2.2 Prospective Cohort Studies baseline age 45–50 years; 6 years of follow-up)
Table 9.2 summarizes eight global cohort studies found that the women with the highest diet qual-
which consistently show that healthy dietary pat- ity score had a 49% lower diabetes risk than
terns protect against diabetes risk and Western those with the lowest scores [27]. The Health
patterns promote diabetes risk [23–31]. The US Professionals Follow-up Study showed that high
Women’s Health Initiative (101,504 postmeno- nutrient quality diets are associated with the
pausal women; 15 years of follow-up) showed greatest reduction in diabetes risk in men with
elevated BMIs compared to the lowest quality A 2014 meta-analysis (nine cohort studies and
diets [29]. The Nurses’ Health Study (80,029 US one large, long-term RCT) found that higher
women; baseline age 39–63; 18 years of follow- adherence to the MedDiet significantly reduced
up) observed that women with low diet quality risk of diabetes by 23% [35]. Another 2014 meta-
who improved their diet health quality even for a analysis (17 RCTs; 2300 overweight, obese, pre-
short time had a lower risk of diabetes by as much diabetic, or high CVD risk subjects; >12 weeks)
as 36% compared to women who did not change showed that higher MedDiet adherence lowered
their low diet quality score [30]. insulin resistance risk by significantly reducing
hs CRP, IL-6 and intra cellular adhesion mole-
9.2.2.3 Randomized Controlled cule 1 and increasing flow mediated dilation [37].
Trial (RCT)
An Australian crossover RCT (49 subjects with- Diabetes Management
out diabetes; 15 men and 34 women; mean age Four systematic reviews and meta-analyses of
36 years; BMI 27; Western diet vs. Healthy diet; RCTs consistently show that high adherence to
two 4-week weight-stable dietary trials) found the MedDiet is an effective dietary option for
that a Western diet reduced insulin sensitivity diabetes management [32, 33, 37, 38]. A 2015
compared to a healthy diet only in relatively systematic review of MedDiets and diabetes
insulin-
resistant individuals [32]. Dramatic management (five meta-analyses of RCTs includ-
improvement in healthy diet quality increased ing a ‘de novo’ meta-analyses of three RCTs with
insulin sensitivity by about 50% in only 4 weeks >6 months of duration) found mean HbA1c low-
in people at risk of developing diabetes. ering by 0.3%–0.47% compared to usual care or
a low-fat diet [32]. A 2015 meta-analysis (nine
RCTs; 1178 diabetic patients; 4 weeks to 4 years)
9.2.3 Mediterranean Diet (MedDiet) showed that high adherence to the MedDiet sig-
nificantly lowered mean HbA1c by 0.30%, fast-
The effect of the MedDiet in preventing and man- ing blood glucose by 13 mg/dL, fasting insulin
aging diabetes has been the most extensively by 0.55 units, BMI by 0.29 units, weight by
studied of the healthy dietary patterns in both 0.29 kg, lowered LDL-cholesterol, triglycerides
prospective cohort studies and RCTs (Table 9.3) and blood pressure, and increased HDL-
[32–42]. cholesterol [33]. This analysis demonstrates the
multi-beneficial effects of the MedDiet on man-
9.2.3.1 Meta-Analyses aging both diabetes and cardiovascular health. A
2013 meta-analysis (eight RCTs; 3–48 months)
Diabetes Prevention found that the MedDiet significantly improved
Four meta-analyses of prospective cohort studies HbA1c compared to usual care or a low-fat diet
and RCTs consistently show that the high adher- [37]. Another 2013 meta- analysis (20 RCTs,
ence to MedDiets significantly reduced diabetes 3073 diabetic patients; >6 months) demonstrated
risk [32, 34–36]. A 2015 review (seven meta- that the MedDiet was more effective than low
analyses) found that higher adherence to the carbohydrate, low glycemic index, and high pro-
MedDiet significantly reduced the risk of diabe- tein diets by significantly reducing HbA1c by
tes by 19–23% with a primary mechanism asso- 0.47% (Fig. 9.5) and promoting weight loss by
ciated with reduced body weight [32]. Another 1.84 kg (Fig. 9.6) [38].
2015 meta-analysis (eight cohort studies and one
large, long-term RCT; 122,810 subjects) showed 9.2.3.2 Representative Studies
that greater adherence to a MedDiet was associ-
ated with the primary prevention of diabetes with Diabetes Prevention
a higher adherence to the MedDiet significantly A RCT and cohort study are representative of
associated with a 19% lower risk of diabetes [34]. the MedDiet effects on diabetes risk [39, 40].
Table 9.3 Summary of Mediterranean diets (MedDiet) studies in type 2 diabetes (diabetes) risk and management
Esposito et al. (2015). Eight meta-analyses; diabetes (1) Diabetes Prevention: Two meta-analyses
Assess the evidence of patients and prediabetes subjects; found that higher adherence to the MedDiet
previous meta-analyses five RCTs; ‘de novo’ meta- significantly reduced the risk of diabetes by
on efficacy of a analysis of three long-term 19–23% and five meta-analyses showed that the
MedDiet on the risk (> 6 months) RCTs MedDiet reduced body weight
and management of (2) Diabetes Management: A ‘de novo’ meta-
diabetes [32] analysis of three RCTs >6 months found in diabetic
patients that the MedDiet lowered mean HbA1c
levels by 0.47% compared with the usual care or a
low-fat diet. Also, four meta-analyses showed that
adherence to the MedDiet reduced HbA1c in
diabetic patients from 0.3% to 0.47%, compared
with a low-fat control diet
Huo et al. (2015). Nine RCTs; 1178 diabetic Compared with control diets, the MedDiet
Conducted a meta- patients; ages varied from 26 to significantly reduced mean HbA1c by 0.30%; mean
analysis of RCTs to 77 years; 4 weeks to 4 years fasting plasma glucose by 13 mg/dL, mean fasting
explore the effects of insulin by 0.55 μU/mL, mean BMI by 0.29 units,
MedDiet on glycemic and mean body weight by 0.29 kg. Additionally,
control, weight loss and blood lipid and lipoprotein profiles were
cardiovascular risk significantly improved with mean reductions in
factors in diabetic total cholesterol by 13 mg/dL and triglycerides by,
patients [33] 25.7 mg/dL and increases in HDL-C by 2.3 mg/dL;
plus, a decline in systolic BP by 1.45 mm Hg and
diastolic BP by 1.41 mm Hg
Schwingshackl et al. Eight cohort studies and one RCT; Greater adherence to a MedDiet was associated
(2015). 122,810 subjects free of diabetes; with a significant reduction in the risk of diabetes
Assess the effect of follow-up ranged between 3.2 and by 19% which is clinically relevant for the use of
MedDiet adherence on 20 years MedDiet in primary prevention of diabetes
the risk of diabetes [34]
Koloverou et al. Nine cohort studies and one RCT; Higher adherence to the MedDiet was associated
(2014). 136,846 subjects; follow up with 23% reduced risk of developing diabetes with
Assess prospective ranged from 3.5 to 14 years no risk attenuation by region or health status
studies that have
evaluated the effect of a
MedDiet on the
development of
diabetes [35]
Schwingshackl et al. 17 RCTs; 2300 overweight obese, MedDiet significantly increased flow mediated
(2014). prediabetic, or high cardiovascular dilation by 1.9% and adiponectin by 1.7 μg/mL,
Assess the effects of risk subjects; >12 weeks and significantly reduced hs CRP by 0.98 mg/l,
adherence to the interleukin (IL)-6 by 0.42 pg/mL, and intracellular
MedDiet on endothelial adhesion molecule-1 by 23.7 ng/mL. The MedDiet
function and improved endothelial function and reduced
inflammation [36] inflammation, which is associated with reduced
risk of insulin resistance and other complications
Carter et al. (2013). Eight RCTs; diabetes patients and The MedDiet significantly improved HbA1c levels
Determine if the effect prediabetes subjects; MedDiet, compared to the low-fat diet and usual care but not
of the MedDiet, low fat diet, Palaeolithic diet, or compared to the Palaeolithic diet
irrespective of weight usual care; 3–48 months
loss, can aid in glucose
control in diabetic or
prediabetic persons
[37]
(continued)
Ajala et al. (2013). 20 RCTs; 3073 diabetic patients; The low carbohydrate, low-GI, MedDiet, and
Assess the effect of interventions that lasted high-protein diets all led to significantly reduced
various diets on >6 months that compared low HbA1c by 0.12%, 0.14%, 0.47%, and 0.28%,
glycemic control, carbohydrate, vegetarian, vegan, respectively, compared with their respective control
lipids, and weight loss low-glycemic index (GI), high diets with the MedDiet having the largest effect
in the management of fiber, MedDiet, and high-protein size (Fig. 9.5). Also, the MedDiet led to the most
diabetic patients [38] diets with control diets including weight loss by 1.8 kg (p < 0.00001) (Fig. 9.6)
low-fat, high-GI, American
Diabetes Association, and
low-protein diets
Representative studies
Diabetes prevention
Salas-Salvado et al. Multi-center Parallel RCT: 3541 Multivariate-adjusted diabetes risk was reduced by
(2014). men and women without diabetes 40% for the MedDiet supplemented with extra
Assess the efficacy of but with high cardiovascular virgin olive oil and 18% for the MedDiet
MedDiets for the disease risk at primary care supplemented with nuts compared with the low fat
primary prevention of centers; mean age 67 years; 1 of 3 control diet. There was a 30% diabetes risk
diabetes (Spain— diets: MedDiet plus extra-virgin reduction with both MedDiets combined when
Prevencion con Dieta olive oil, MedDiet supplemented compared to the control diet
Mediterranea trial with nuts, or a low-fat control diet;
(PREDIMED) [39] no increase in physical activity or
weight loss; 4.1 years of duration
Martinez-Gonzalez 13,380 Spanish university A 2-point increase in the score was associated with
et al. (2008). graduates without diabetes; mean a 35% relative reduction in the risk of diabetes,
Assess the relation age 39 years; mean BMI 23; with a significant inverse linear trend (p = 0.04) in
between adherence to followed for a median of 4.4 years the multivariate analysis. (Fig. 9.7)
the MedDiet and
incidence of diabetes
among healthy
participants (Spain;
Seguimiento
Universidad de Navarra
follow-up [SUN
Project]) [40]
Diabetes management
Maiorino et al. (2016). Parallel RCT: 215 men and The MedDiet lowered CRP by 37% and increased
Assess the MedDiet women with newly diagnosed adiponectin by 43% compared to the low-fat diet
influence on both CRP diabetes; MedDiet (54 males and group which remained unchanged in both
and adiponectin in 54 females) vs. a low-fat diet (52 outcomes. Diabetic patients with the highest
newly diagnosed males and 55 females); 1 year MedDiet scores (6–9 points) had lower circulating
diabetics, and whether CRP levels and higher circulating total adiponectin
adherence to MedDiet levels than the diabetic patients who scored <3
affects their circulating points on the scale (p = 0.001)
levels (Italy RCT [41]
Esposito et al. (2014). Parallel RCT: 215 participants; Lower-carbohydrate MedDiet resulted in a
Long-term effects of mean age 52 years; mean BMI substantial long-term reduction of HbA1c levels,
dietary interventions on 29.6; 50% females; newly higher rate of diabetes remission, and delayed need
glycemic control, need diagnosed diabetes; low- for diabetes medication in patients with newly
for diabetes carbohydrate MedDiet vs. low-fat diagnosed diabetes vs. low fat diets (Figs. 9.8
medications, and diet; 6.1–8.1 years and 9.9)
remission of diabetes
(Italy) [42]
0
Low Low glycemic MedDiet High protein
–0.05 carbohydrate index diet diet
diet
–0.1
–0.15
–0.2
HbA1c (%)
–0.25
–0.3
–0.35
–0.4
–0.45
–0.5
Fig. 9.5 Effect of the MedDiet vs. other dietary patterns and diabetic subject HbA1c levels from a meta-analysis of
RCTs lasting >6 months (adapted from [38])
1.5
0.5
Weight Change (kg)
0
Low carb diet Low glycemic MedDiet High protein
–0.5 (p =.21) index diet (p <.00001) diet (p =.54)
(p =.36)
–1
–1.5
–2
–2.5
Fig. 9.6 Effect of the MedDiet vs. other dietary patterns and diabetic subject weight change from a meta-analysis of
RCTs lasting >6 months (adapted from [38])
The Spanish Prevencion con Dieta Mediterranea compared to the low-fat control diet [39]. The
trial [PREDIMED] (3541 men and women with- Spanish Seguimiento Universidad de Navarra
out diabetes but with high cardiovascular dis- (SUN) follow-up (13,380 university graduates
ease risk; at primary care centers; mean age free of diabetes; mean baseline age 39 years;
67 years; MedDiet supplemented with extra- median 4.4 years of follow-up) found that a
virgin olive oil or with nuts, and a low-fat con- 2-point increase in MedDiet score was associ-
trol diet; 4.1 years) found that the combined ated with a 35% lower risk of diabetes (Fig. 9.7)
MedDiets lowered diabetes risk by 30%, when [40].
1.2
1
Relative Risk for Diabetes
0.8
0.6
0.4
0.2
0
Low Moderate High
MedDiet Score
Fig. 9.7 Association between MedDiet adherence and risk for type 2 diabetes (diabetes) in 13,380 Spanish university
graduates without diabetes with a median follow-up of 4.4 years (p = 0.04; multivariate adjusted) (adapted from [40])
Low carbohydrate MedDiet Low fat diet

7.8
7.6
7.4
7.2
HbA1c (%)
7
6.8
6.6
6.4
6.2
6
5.8
0 1 2 3 4 5 6 7
Follow-up (Years)
Fig. 9.8 Effect of a low carbohydrate MedDiet and HbA1c levels in newly diagnosed type 2 diabetic patients vs. a
low-fat diet over 7 years (p < 0.001) (adapted from [42]
Diabetes Management nosed diabetic patients (215 participants; mean

Two Italian RCTs are representative of MedDiet baseline age 52 years; mean BMI 29.6; 50%
effects on diabetes management [41, 42]. The females; low-carbohydrate MedDiet vs. low-fat
MEDITA trial (215 men and women; newly diag- diet; 6.1–8.1 years) reported that subjects on the
nosed type 2 diabetics; MedDiet vs. a low-fat lower-carbohydrate MedDiet had substantial
diet; 1 year) found that CRP was lowered by 37% long-term reduction of HbA1c levels (Fig. 9.8), a
and adiponectin rose by 43% in the MedDiet higher rate of diabetes remission (Fig. 9.9) and
group, but remained unchanged in the low-fat delayed need for diabetes medication compared
diet group [41]. The other RCT in newly diag- to patients on the low-fat diets [42].
Low carbohydrate MedDiet Low fat diet

16
14
12
% Remission Rate
10
0
1 2 3 4 5 6
Follow-up (Years)
Fig. 9.9 Effect of low carbohydrate MedDiet on remission rates in newly diagnosed type 2 diabetic patients vs. low-fat
diet over 6 years (p < 0.001) (adapted from [42])
9.2.4 O
ther Healthy Dietary 9.2.4.3 Vegetarian Diets
Patterns
Diabetes Management
Prospective cohort studies and RCTs on the effect There are a number of RCTs confirming vegetarian
of other healthy dietary patterns in preventing diets, especially low-fat vegan diets, as an effective
and managing diabetes are summarized in dietary option for the management of diabetes [45–
Table 9.4 [43–51]. 48]. A 2014 meta-analysis (six RCTs; 255 diabetic
patients; 4–74 weeks) found that vegetarian diets
9.2.4.1 Lower Carbohydrate Diets significantly reduced HbA1c by 0.39% and direc-
A 2017 meta-analysis (ten RCTs; average age tionally lowered fasting blood glucose by 7 mg/
58 years; 1376 subjects) found that over 1 year, dL. This analysis showed that healthy, lower glyce-
moderate carbohydrate diets (energy percent- mic index and load vegetarian diets are an espe-
age below 45%) lowered HbA1c 0.34% com- cially effective option for diabetes management
pared with high carbohydrate diets and the [45]. A 2016 Korean RCT (93 diabetic patients;
greater the carbohydrate restriction, the better pesco-vegan diet vs. Korean Diabetes Association
the glucose- lowering effect (p < 0.01) and dietary guidance; 12 weeks) showed that the pesco-
beneficial effects on triglycerides and HDL- vegan diet significantly reduced HbA1C levels by
cholesterol levels [43]. 0.5% compared to the conventional diabetes diet,
which lowered HbA1c levels by 0.2% after adjust-
9.2.4.2 Dietary Approaches to ing for total energy intake and waist circumference
Stop Hypertension changes [46]. Two US RCTs in individuals with
(DASH) Diet diabetes (99 subjects each; mean age approx
A meta-analysis (20 RCTs; 2890 subjects) found 55 years; 22–74 weeks) found that low-fat vegan
that the DASH diet significantly reduced mean diets improved HbA1c more effectively than
fasting insulin levels by 0.16 units with a modest American Diabetes Association (ADA) dietary
improvement in insulin sensitivity when pre- guidelines but the effect is significantly correlated
scribed for more than 16 weeks [44]. with greater weight loss [47, 48].
Table 9.4 Summary of other healthy dietary patterns studies in type 2 diabetes (diabetes) risk and management
Moderate vs. high carbohydrate diets
Snorgaard et al. (2017). 10 RCTs; average age 58 years; Over 1 year, moderate carbohydrate diets (energy
Meta-analysis 1376 subjects percentage below 45%) lowered HbA1c 0.34%
comparing diets compared with high carbohydrate diets. The
containing low to greater the carbohydrate restriction, the better the
moderate amounts glucose-lowering effect (R = − 0.85; p < 0.01).
of carbohydrate to diets The effect of the moderate and high carbohydrate
containing high amounts diets had similar effects on BMI, body weight,
of carbohydrate in LDL cholesterol, quality of life and attrition rates
subjects with diabetes
[43]
Dietary approaches to stop hypertension (DASH) diet
Meta-analysis
Shirani et al. (2013). 20 RCTs; fasting blood glucose The DASH diet significantly reduced mean
Examine the effects of (nine RCTs; 974 subjects), fasting fasting insulin levels by 0.16 units suggesting a
DASH diet consumption insulin (seven RCTs; 787 subjects), modest improvement in insulin sensitivity when
on the indices of HOMA-IR (four RCTs; 677 prescribed for more than 16 weeks. There were
glycemic control such as subjects); age range 21–69 years; no significant beneficial effects on fasting blood
fasting blood glucose, 3–24 weeks glucose or HOMA-IR
serum fasting insulin
level, and Homeostatic
Model Assessment
insulin resistance
(HOMA-IR) [44]
Vegetarian diet
Diabetes management
Meta-analysis
Yokoyama et al. (2014). Six RCTs; 255 diabetic patients; Consumption of vegetarian diets was associated
Examine the association mean age 42.5 years; control diets: with a significantly reduced HbA1c by 0.39%
between vegetarian diets omnivorous, American Diabetes and a non-significantly reduced fasting blood
and glycemic control in Association guidelines, low fat diet; glucose level by 7.0 mg/dL vs. control diets
diabetes [45] BMI 26–35; 4–74 weeks
Representative RCTs
Lee et al. (2016). 93 diabetic patients; pesco-vegan The pesco-vegan diet significantly reduced HbA1c
Compare the effects of a diet vs. conventional diet levels by 0.5% vs. the conventional diabetes diet
pesco-vegan diet and recommended by the Korean which lowered HbA1c levels by 0.2% (p = 0.017).
conventional diabetic Diabetes Association 2011; primary The participants with high compliance had larger
diet on glycemic control endpoint change in HbA1c level; reduction in HbA1c levels for pesco-vegan diet by
(Korea) [46] 12 weeks 0.9% and control by 0.3%. The beneficial effect of
pesco-vegan diets was noted even after adjusting
for changes in total energy intake or waist
circumference over the 12 weeks
Barnard et al. (2009). 99 diabetic patients with Weight loss was significant within each diet group
Compare the effects of a medications; mean age 55 years; but not significantly different between groups.
low-fat vegan diet and 61% female; low-fat vegan diet or a HbA1c changes from baseline to 74 weeks was
conventional diabetes diet following 2003 American reduced for vegan diet by 0.34 units and ADA diet
diet recommendations Diabetes Association (ADA) by 0.14 (p = 0.43). HbA1c reductions from
on glycemia, weight, guidelines; primary endpoints: baseline to last value before any medication
and plasma lipids (US) HbA1c and plasma lipids; 74 weeks adjustment for vegan diet was 0.40 and ADA diet
[47] 0.01 (p = 0.03). In analyses before alterations in
lipid-lowering medications, total cholesterol
decreased by 6.8 mg/dL in the vegan diet and
0.4 mg/dL in ADA diet (p = 0.01); LDL
cholesterol decreased by 13.5 mg/dL in the vegan
diet and 3.4 mg/dL in ADA diet (p = 0.03)
Barnard et al. (2006). 99 diabetes patients, mean age 43% of the vegan group and 26% of the ADA
Evaluate the effect of a 56 years; 40% males; low-fat vegan group participants reduced diabetes medications.
low-fat vegan diet on diet vs. the ADA dietary guidelines Including all participants, HbA1c decreased for
glycemic control in diet; 22 weeks the vegan group by 0.96 points and for the ADA
diabetic patients (US) group by 0.56 points (p = 0.089). HbA1C
[48] decreased by 1.23 points in the vegan group
compared with 0.38 points in the ADA group
(p = 0.01). Body weight decreased in the vegan
group by 6.5 kg and in the ADA group by 3.1 kg
(p < 0.001). Body weight change was
significantly correlated with HbA1C change
(r = 0.51). Among subjects not on lipid-lowering
medications, vegan group LDL cholesterol was
significantly lowered by 10.5% more than the
ADA group
Diabetes prevention
RCT
Barnard et al. (2005). 64 postmenopausal women; mean Mean body weight decreased in vegan group by
Investigate the effect of age 56 years; mean BMI 33; 5.8 kg, compared with 3.8 kg in the control
a low-fat, vegan diet on low-fat, vegan diet or a control diet group (p = 0.012). In the vegan group the mean
body weight, based on National Cholesterol insulin sensitivity index increased from baseline
metabolism, and insulin Education Program guidelines, to 14 weeks by 1.1 units (p = 0.017), the
sensitivity, while without energy intake limits; difference between vegan and healthy control
controlling for exercise 14 weeks groups was 0.8 (p = 0.19)
in free-living individuals
(US) [49]
Prospective study
Tonstad et al. (2009). 22,434 men and 38,469 women; Mean BMI was lowest in vegans (23.6) and
Assess the prevalence of mean baseline age 62.5 years; mean incrementally higher in lacto-ovo vegetarians
diabetes in people BMI 32; 62% female; 4 years of (25.7), pesco-vegetarians (26.3), semi-
following different types follow-up vegetarians (27.3), and nonvegetarians (28.8).
of vegetarian diets Prevalence of diabetes increased from 2.9% in
compared with that in vegans to 7.6% in nonvegetarians; the prevalence
nonvegetarians was intermediate in participants consuming
(Adventist Health Study lacto-ovo (3.2%), pesco (4.8%), or semi-
2; US) [50] vegetarian (6.1%) diets. After multivariate
adjustment diabetes risk increased progressively
from vegans to semi-vegetarians to
nonvegetarians (Fig. 9.10)
Healthy Nordic Diet
Prospective cohort study
Lacoppidan et al. (2015). 57,053 Danish men and women; Greater adherence to the healthy Nordic food
Investigate the mean baseline age 56 years; mean index was significantly associated with lower
association between BMI 25; 7366 diabetes cases; diabetes risk after adjusting for potential
adherence to a healthy median follow-up of 15.3 years confounders. A high adherence (5–6 points) was
Nordic food index and associated with a statistically significant lower
the risk of diabetes (The diabetes risk in women by 25% and in men by
Danish Diet, Cancer and 38% compared to those with an index score of 0
Health Cohort Study) points (poor adherence) (Fig. 9.11). These results
[51] suggest that regional healthy diets other than the
MedDiet may also be recommended for diabetes
prevention. Also, the intake of > median of
specific foods such as oatmeal, root vegetables
and rye bread were significantly associated with
lower diabetes risk (Fig. 9.12)
1.2
1
Odds Ratio for Type 2 Diabetes
0.8
0.6
0.4
0.2
0
Vegan Lacto-ovo Pesco-vegetarian Semi-vegetarian Non-vegetarian
vegetarian
Fig. 9.10 Association between type of vegetarian diet compared to non-vegetarian diet and type 2 diabetes risk
(multivariate adjusted) from the Adventist Health Study 2 (adapted from [50])
Diabetes Prevention 38% (Fig. 9.11) [51]. Also, the intake of >

A representative RCT and a cohort study demon- median of specific foods such as oatmeal, root
strated the effectiveness of vegetarian diets in vegetables (carrots, radishes, beetroot, turnips,
protecting against diabetes risk [49, 50]. A RCT and other related varieties) and rye bread were
in postmenopausal women (64 women; mean age significantly associated with lower diabetes risk
56 years; mean BMI 33; low-fat vegan diet vs. (Fig. 9.12).
National Cholesterol Education Program guide-
lines; 14 weeks) demonstrated that the vegan diet
was more effective than a nationally recom- 9.3 Whole Plant Foods
mended cardiovascular healthy diet in lowering
body weight and improving insulin sensitivity 9.3.1 Overview
[49]. The 2009 Adventist Health Study 2 (22,434
men and 38,469 women; mean baseline age Whole plant foods collectively are an important
62.5 years; mean BMI 32; 62% female; 4 years) source of healthy nutrients (e.g., vitamins, miner-
found that all forms of vegetarian diets were als), phytochemicals (e.g., polyphenols, carot-
more effective at lowering diabetes risk com- enoids), fiber, and some are goods sources of
pared to non-vegetarian diets (Fig. 9.10) [50]. protein (e.g., legumes, nuts and seeds) compared
to more refined plant foods but their composition
9.2.4.4 Healthy Nordic Food Index can vary widely (Appendix B). The US dietary
The Nordic food index focuses on the intake of guidelines recommend at least 3 servings/whole-
fish, cabbage, rye bread, oatmeal, apples and grains/day and less than or equal to 3 servings of
pears, and root vegetables [51]. The Danish refined grains/day to promote health and wellness
Diet, Cancer and Health Cohort Study (57,053 associated with reduced risk of diabetes and
Danish men and women; mean baseline age chronic diseases [52]. However, only about 1% of
56 years; mean BMI 25; 7366 diabetes cases; Americans follow the recommendation for
median follow- up of 15.3 years) found that whole-grain intake as the average American’s
greater adherence to the healthy Nordic food intake is <1 ounce whole grains/day and 70%
index was significantly associated with lower exceed the recommended intake for refined grains
diabetes risk in women by 25% and in men by [52, 53]. Adequate intake of fruits and vegetables
Women (p <.0001) Men (p <.0001)

1.1
Hazard Ratio for Diabetes 1
0.9
0.8
0.7
0.6
0.5
0 1 2 3 4 5 to 6
Healthy Nordic Food Index Adherence
Fig. 9.11 Adherence to the Healthy Nordic Food Index and type 2 diabetes (diabetes) risk, after lifestyle adjustment
(adapted from [51])
Above median intake Below median intake
Rye bread (Men)
Rye bread (Women)
Root vegetables (Men)
Root vegetables (Women)
Oatmeal (Men)
Oatmeal (Women)
0.4 0.5 0.6 0.7 0.8 0.9 1

Hazard Ratio for Diabetes
Fig. 9.12 Association between specific Healthy Nordic Food Index foods and type 2 diabetes (diabetes) risk (adapted
from [51])
(> 400 g/day) makes important contributions to [57]. More than 85% of the population fall short
human health because they provide: antioxidant of meeting the daily recommendation as most
vitamins and phytochemicals, especially vita- Americans consume less than one cup of fruits
mins C and A, flavonoids and carotenoids, elec- and less than two cups of vegetables daily, with
trolytes including potassium and magnesium, the primary contributors consisting of juice and
with low sodium), and dietary fiber [54–56]. processed potatoes, compared to the current rec-
Globally, fruits and vegetables consumption is ommendations of two cups of fruit and 2.5 cups
only a small fraction of the recommended levels of vegetables per day [52]. Legumes including
dietary pulses (e.g., pinto beans, split peas, len- in whole plant foods, especially those lower in
tils, chickpeas) and soybeans, are rich in fiber and energy density, may help in preventing weight
protein with relatively low glycemic response gain and diabetes risk, and help to manage longer-
properties [58, 59]. A serving of legumes is half a term health risk in people compared the minimal
cup or 90–100 g cooked legumes, which contains intake of whole plant foods in diets associated
5–10 g of fiber, 7–8 g of protein, and <5% of with the Western lifestyle.
energy as fat, with the exception of chickpeas and
soybeans which have 15% and 47% energy from
fat, respectively. However, pulse consumption 9.3.2 Whole-Grains
has been in decline with the global shift to
Western-style diets [60]. For example, between 9.3.2.1 Prospective Studies
the 1960s and 1990s, legume intake decreased by
40% in India and by 24% in Mexico. Legumes Systematic Reviews and Meta-Analyses
are infrequently consumed by North Americans Three systematic reviews and meta-analyses con-
and northern Europeans, with <8% of Americans sistently concluded that >3 whole-grain servings/
consuming pulses daily. Nuts (e.g., almonds, pis- day, especially from breakfast cereals, breads, or
tachios, walnuts, hazelnuts, pecans, peanuts) are brown rice, lowered diabetes risk by 18–40% com-
nutrient dense sources of fiber, protein, unsatu- pared to refined grains [62–64]. A 2016 meta-anal-
rated fat, vitamins (e.g., B-vitamins and vitamin ysis (16 cohort studies; 643,946 participants;
E), minerals (e.g., potassium and magnesium), 29,413 diabetes cases) showed that three whole-
phytosterols, and polyphenols [59]. Although grain servings/day reduced diabetes risk by 23%
nuts have a relatively high energy density (about (based on ten studies) [62]. Inverse associations
6 kcal/g) due primarily to a high unsaturated fat with risk of diabetes were observed for specific
content and low water content, human studies whole grains including whole grain bread, whole
have shown that nuts have a lower metabolizable grain cereals, wheat bran and brown rice compared
energy than predicted from the Atwater energy to an increased diabetes risk for white rice
tables because of the incomplete absorption of (Fig. 9.13). A 2015 meta-analysis (seven cohort
nut fat and other macronutrients [61]. Diets rich studies and one cross-sectional study; 316,051
1 serving of white rice
10 g of wheat bran
1/2 serving of brown rice
1 serving of whole grain breakfast cereal
3 servings of whole grain bread
–35 –25 –15 –5 5 15 25

Diabetes Risk (%)
Fig. 9.13 Dose response analysis of the association of daily intake of specific grains on type 2 diabetes (diabetes) risk
from a meta-analysis of 16 cohort studies (adapted from [62])
participants; follow-up ranging from 6 to 22 years; diabetes risk markers [69–78]. Five RCTs found
15,573 diabetes cases) found a significant linear that diets rich in cereal fiber and whole-grains
inverse relationship between whole grain intake (approximately 28–40 g cereal fiber or >6 whole-
and diabetes risk (p < 0.0001) with an overall abso- grain servings/day) supported a significant
lute multivariate reduction of 0.3% in the diabetes 10–25% improvement in insulin sensitivity and/or
risk rate for each additional 10 g of whole-grain reduced insulin resistance (HOMA-IR) scores
consumed daily [63]. A 2012 systematic review compared to diets with <18 g fiber/day [69–73].
and meta-analysis (45 cohort studies and 21 RCTs) Five RCTs report that diets rich in cereal fiber and
showed that increased intake of whole grain and whole-grains (approximately 23–32 g cereal fiber
cereal fiber lowers the risk of diabetes, CVD, and or >6 whole-grain servings/day) showed insignifi-
weight gain when consumed at >48 g whole grains cant effects on insulin sensitivity [74–78].
(approximately 3 servings)/day [64].
Diabetes Management
9.3.2.2 Cohort Studies Oat and barley β-glucan are the most effective
Representative prospective studies show that type of whole grains for improving glycemic
whole-grains are very effective at reducing dia- control and glycosylated hemoglobin (HbA1c) in
betes risk [65–68]. The Women’s Health managing diabetes health [79–82]. A 2016 meta-
Initiative Observational Study (72,215 post- analysis (four RCTs) found that compared to
menopausal women; mean baseline age 64 years, control, type 2 diabetes patients administrated oat
mean BMI 27; 7.9 years of follow-up; 3465 dia- β-glucan from 2.5 to 3.5 g/day for 3–8 weeks pre-
betes cases) found that women consuming >2 sented significantly lowered concentrations in
servings/day of whole-grains reduced multivari- fasting plasma glucose by 0.52 mmol/L (p = 0.01)
ate diabetes risk by 37% compared to those con- and HbA1c by 0.21% (p = 0.03) [79]. Another
suming very little or no intake [65]. A Swedish 2016 meta-analysis (18 RCTs; 1024 subjects; oat
prospective study (3180 women and 2297 men; products [20–136 g/day] vs. oat and barley
mean baseline age 47 years; 8–10 years of fol- β-glucan extract [3–10 g/day] found that higher
low-up) showed that higher intake of whole grain whole oats and oat bran but not oat and barley
(59 g/day compared with 31 g/day) was associ- β-glucan extracts were significantly associated
ated with a 34% lower risk for deterioration in with lower HbA1c, fasting glucose and fasting
glucose tolerance from prediabetes to diabetes insulin in diabetic individuals [80].
[66]. The Physicians’ Health Study reported that
men consuming whole-grain breakfast cereals
had a significant 33% lower diabetes risk com- 9.3.3 Fruits and Vegetables
pared to those consuming refined breakfast cere-
als [67]. The pooled analysis of the Health 9.3.3.1 Systematic Reviews
Professionals Follow-up Study and the Nurses’ and Meta-Analyses
Health Study (39,765 men and 157,463 women; Eight systematic reviews and meta-analyses of
follow-up of 14–22 years) estimated that replac- prospective cohort studies find high variability in
ing 50 g white rice/day with the same amount of the effects of fruit and vegetables on diabetes
brown rice was associated with a 16% lower risk risk [83–89]. A 2016 meta-analysis (23 cohort
of type 2 diabetes [68]. studies) found reduced diabetes risk for higher
vs. lower intake of total fruit by 9%, for blueber-
9.3.2.3 Randomized Controlled ries by 25%, for green leafy vegetables by 13%,
Trial (RCTs) for yellow vegetables by 28%, for cruciferous
vegetables by 18% and fruit fiber lowered risk by
Diabetes Prevention 7% and vegetable fiber by 13% [83]. Another
Whole-grain RCTs report inconsistent effects on 2016 systematic review (six cohort studies and
insulin sensitivity, HOMA-IR scores and other one nested cohort study; 326,675 subjects;
4–20 years of follow-up) showed that of potato 9.3.3.2 Prospective Cohort Studies

products only French fries are consistently posi- Individual fruit and vegetables differ depending on
tively associated with diabetes risk in men and fiber and caloric content, and glycemic index and
women in all three large cohort studies [84]. A load values [90–95]. A 2017 pooled analysis from
2015 meta-analysis of fruit (nine cohort studies; the US and EU cohort studies (NIH-AARP
403,259 participants) found a non-linear associ- 401,909 participants and EPIC 20,629 partici-
ation of fruit intake and diabetes risk (p for non- pants; mean baseline age approx. 64 years; mean
linearity <0.001) with a threshold of 200 g/day follow-up approx. 12 years) found that the
total fruit intake to reduce risk of diabetes by NIHAARP participants consuming higher fruit
13% [85]. A 2012 EU InterAct meta-analysis and green leafy vegetable intake were associated
(five cohort studies) reported that specific groups with a reduced diabetes risk by 5% and 13%,
of vegetables, especially green leafy vegetables respectively but no significant diabetes reduc-
(spinach, chard, endive, lettuce, watercress) and tions were observed in the total pooled or EPIC
root vegetables (carrots, radishes, beetroot, and analyses [90]. A 2014 Finnish prospective study
turnips) appear to be the most beneficial in pro- found that individuals in the highest quartile for
tecting against diabetes (Fig. 9.14) [87]. A meta- intake of fruits, berries, and vegetables (with the
analysis of fruit juice (four sugar-sweetened fruit exclusion of potatoes and fruit juices) had a sig-
juice cohort studies, 191,686 participants; and nificant 24% reduction in diabetes risk compared
four 100% fruit juice cohort studies, 137,663 to those in the lowest quartile [91]. A pooled anal-
participants) indicated that a higher intake of ysis of the Nurses’ Health Studies and the Health
sugar sweetened fruit juice was significantly Professionals Follow-up Study observed a high
associated with increased diabetes risk by 28%, degree of heterogeneity in the effect of whole
while higher intake of 100% fruit juice was not fruits on diabetes risk per three weekly servings
associated with diabetes risk (Fig. 9.15) [89]. (Fig. 9.16) [92]. For vegetables, a greater quantity
Other meta- analyses generally report similar (>3 servings vs. <1 serving/day) or a greater vari-
findings [86, 88]. ety (11 vs. 5 varieties/week) was found to reduce
Low Intake Medium Intake High Intake

1.05
0.95
Relative Risk for Diabetes
0.9
0.85
0.8
0.75
0.7
0.65
0.6
Total Vegetable Green Leafy Vegetables* Root Vegetables*
Fig. 9.14 Association between total vegetable and specific vegetable intake and type 2 diabetes (diabetes ) risk from
European Prospective Investigation into Cancer-InterAct study and meta-analysis (adapted from [87]). *Green leafy veg-
etables (spinach, chard, endive, lettuce, watercress) and root vegetables (carrots, radishes, beetroots, turnips)
30
p =.02
25
Diabetes Risk (%)
20
15
10
5
p =.62
0
Sweetened Fruit Juice 100% Fruit Juice
Fig. 9.15 Association between sweetened and 100% juice (highest vs. lowest intake) on type 2 diabetes (diabetes) risk
from a meta-analysis of eight cohort studies (adapted from [89])
Cantaloupe
Strawberries
Oranges
Peaches, plums, apricots
Grapefruit
Bananas
Apples and pears
Prunes
Grapes and rasins
Blueberries
–30 –25 –20 –15 –10 –5 0 5 10 15

Diabetes Risk per 3 Weekly Servings
Fig. 9.16 Association between 3 servings per week of fruit varieties and type 2 diabetes (diabetes) risk from pooled
data from US men and women in the Nurses’ Health Studies and the Health Professionals Follow-up Study (adapted
from [92])
45
p <.001
40
35
Diabetes Risk (%)
30
25
20
15
10
p =.01
5
0
Baked, Boiled or Mashed Potatoes French Fries
3 Servings/Week
Fig. 9.17 Association between 3 servings per week of potato products and type 2 diabetes (diabetes) risk from pooled
data from US men and women in the Nurses’ Health Studies and the Health Professionals Follow-up Study (adapted
from [94])
diabetes risk by about 24% each [93]. A pooled can serve as a protein replacement for animal
analysis of the Nurses’ Health Studies and the products in a healthy plant based dietary pattern
Health Professionals Follow-up Study found three [102]. Prospective studies indicate that dietary
weekly servings of French fries significant pulses are more effective in lowering diabetes
increased diabetes by 41% vs. 5% for other forms risk than are soy products [102–107]. A 2017
of potatoes (Fig. 9.17) [94]. Also, the Nurses’ prospective assessment of the PREDIMED study
Health Study showed in women that two weekly (3349 participants with a baseline age of 67
servings of French fries were associated with an years; 62% women; 4.3 years of follow-up; 266
increased risk of diabetes by 16% [95]. diabetes cases) found that individuals with the
highest quartile of total legume intake (35g/day)
9.3.3.3 Randomized Controlled had a 355 lower diabetes risk compared to those
Trials (RCTs) in the lowest quartile (p-trend = 0.04) [102].
For whole fruit and vegetables, several RCTs Also, participants with the highest intake of len-
indicate that the consumption of 5–7 portions/ tils (10g/day) reduced diabetes risk by 33%
day has inconsistent effects on insulin sensitivity (p-trend = 0.05) and chickpeas reduced risk by
unless there is specific guidance to consume 32% (p-trend = 0.06). The substitution of 30g/
higher fiber and lower glycemic varieties [96–98]. day of legumes for a half serving of eggs, fish,
Snacking on raisins significantly improved gly- meat, whole grain or white bread, rice, pasta or a
cemic and insulinemic control, and lowered baked potato reduce diabetes risk by 40 to 50%.
HbA1c compared to common high glycemic This study concluded that the routine consump-
snacks [99–101]. tion of legumes in a healthy plant-based diet
such as the MedDiet may reduce the risk of dia-
betes in older adults with elevated CVD risk
9.3.4 Legumes [102]. In India’s Third National Family Health
Survey, non-oil seed pulses, such as lentils, were
9.3.4.1 Prospective Studies associated with a significant 30% reduced preva-
Legumes (e.g., chickpeas, lentils, beans, peas, lence of diabetes among women but not men
soy and peanuts) are a good source of protein [103]. The Nurses’ Health Study found that pea-
and fiber and have a low glycemic index, which nut butter was inversely associated with risk of
9.4 Mechanisms 281
diabetes. The consumption of >140 g (5 ounces) Nurses’ Health Study found that >2 servings/
peanut butter/week significantly decreased dia- week of walnuts significantly lowered diabetes
betes risk by 21% [104]. In a study of over risk in women by 24% (p-trend = 0.002; after
43,000 Chinese Singaporeans, the consumption BMI adjustments) whereas the consumption of
of tofu at least two times per week significantly total nuts and other tree nuts was also inversely
reduced risk of diabetes after adjusting for BMI associated with diabetes risk but these associa-
[105]. Pooled data from the Nurses’ Health tions were insignificant after adjusting for BMI
Studies and the Health Professionals Follow-up (p-trend = 0.49) [113]. Further, a 2013 Nurses’
Study observed that consumption of soy foods Health Study and Health Professionals Follow-up
(tofu and soy milk) was not significantly associ- Study analysis of nut intake and cause-specific
ated with a lower diabetes risk with a 7% diabe- mortality in men and women showed the con-
tes risk reduction for >1 serving per week sumption of >5 nuts servings/week lowered dia-
compared to non-consumers (p-trend = 0.14) betes specific mortality by 16% [114].
[106]. Two representative cohort studies from
Hawaii and Japan did not show significant diabe- 9.3.5.2 Randomized Controlled Trials
tes protective benefits for soy products in either (RCTs)
men or women [107, 108]. Four representative RCTs, in adults with prediabe-
tes or metabolic syndrome, indicate that the con-
9.3.4.2 Randomized Controlled Trials sumption 30 to 60 g/day of almonds, pistachios,
(RCTs) walnuts or 30 g/day mixed nuts (50% walnuts,
In a systematic review and meta-analysis of 41 25% hazelnuts and 25% almonds) significantly
randomized trials, non-oil-seed pulses were improved fasting insulin, insulin sensitivity, insu-
shown to modestly improve medium to longer lin resistance (HOMA-IR), and/or β-cell function
term glycemic control through possible insulin- after 12–16 weeks [115–118]. A 2015 walnut
sparing mechanisms [109]. Two representative RCT (112 prediabetic subjects; added 56 g (366
RCTs with overweight adults demonstrated that kcals) walnuts to the habitual diet vs. no added
the daily consumption of 100 g cooked chickpeas walnuts; 6 months) showed significantly improved
or 50 g whole pea flour in muffins significantly diet quality, endothelial function, total and LDL
lowered fasting insulin and insulin resistance cholesterol, but no effects on anthropometric mea-
within 4–12 weeks [110, 111]. sures, blood glucose level, and blood pressure
[115]. A 2010 almond RCT (65 prediabetic adults;
hypocaloric American Diabetes Association
9.3.5 Tree Nuts and Flaxseed (ADA) diet including 56 g whole almonds or a
nut-free ADA control diet; 16 weeks) showed that
The effects of increased intake of tree nuts and the ADA diet supplemented with almonds signifi-
flaxseed and diabetes risk has been evaluated in a cantly improved fasting insulin and HOMA-IR
range of prospective studies and RCTs [104, and β-cell function (HOMA-B) compared to the
112–120]. control ADA diet [117]. Also, flaxseed (13–40 g/
day) has been shown to effectively improve insu-
9.3.5.1 Prospective Cohort Studies lin sensitivity in obese and prediabetic adults
A 2014 systematic review and meta-analysis [119, 120].
(five cohort studies and the PREDIMED trial)
found that four weekly nut servings were signifi-
cantly associated with a lower diabetes risk by 9.4 Mechanisms
12% [112]. A 2002 Nurses’ Health Study
observed that the consumption of >5 nut serv- Healthy dietary patterns and whole plant foods
ings/week significantly lowered diabetes risk by influence various diabetes and cardiometabolic
27% compared to never/almost never nut con- risk factors by improving microbiota health,
sumption [104]. However, a 2013 updated body weight, visceral fat, glucose- insulin
homeostasis, oxidative stress, inflammation, and fruits and vegetables, such as apples, pears, ber-
endothelial health, lipoprotein concentrations, ries, and peppers, may be especially helpful in
and blood pressure for lower diabetes risk and preventing weight gain and obesity compared to
better diabetes management outcomes [4, 13, other types of low flavonoid fruits and vegeta-
121]. Clinical trials implicate low adiponectin, bles [129]. About 95% of Americans or other
common in overweight and obese individuals Western populations do not consume an ade-
with elevated visceral fat, in the pathogenesis of quate level of fiber daily (14 g fiber/1000 kcals
diabetes, coronary artery disease and hyperten- or 25 g/day for women and 38 g/day for men)
sion [122, 123]. Healthy dietary patterns and [130]. A 2017 analysis of the Diabetes Prevention
weight loss appear to increase adiponectin con- Program (DPP) RCT data found a dietary shift
centrations, which are important for reducing toward greater intake of dietary fiber, fruits, and
diabetes risk and aiding in diabetes management vegetables and lower fat diets promoted weight
by improving insulin sensitivity in the muscle loss in individuals at high risk of developing dia-
and liver and through its anti-inflammatory betes, with each 5 g/day increase in fiber reduc-
effects. Anti- inflammatory and anti-oxidant ing body weight by 1.26 kg over year [131]. A
dietary patterns are inversely related to CRP lev- 2017 Brazilian crossover RCT (19 subjects with
els and positively to blood levels of carotenoids diabetes; 53% women; mean age 66 years; iso-
[124]. High intake of the MedDiet was associ- caloric 370 kcal breakfasts; 5.4 g soluble fiber
ated with lower odds of diabetes by 83% (multi- from whole fruit or guar gum supplement vs 0.8
variate adjusted; highest compared with lowest g soluble fiber in the usual breakfast control)
quintile; p-trend = 0.013). High levels of anti- found that higher soluble fiber intake from a
oxidant nutrients, phenolics and carotenoids in a papaya and orange or guar gum at breakfast was
diet abundant in fruits, berries and vegetables significantly associated with similar lower post-
could potentially reduce the risk of diabetes and prandial glucose over 180 minutes in patients
the associated increased risk of microvascular with diabetes compared to the usual low soluble
and macrovascular complications [124–127]. A fiber breakfast [132]. In contrast, polydextrose, a
2017 PREDIMED substudy (1139 high CVD or non-digestible prebiotic oligosaccharide used
diabetic risk subjects; mean baseline age 68 widely across most sectors of the food industry
years; 55% women; low-fat control diet or one with an energy value of 1 kcal/g, was not shown
of two Mediterranean diets, supplemented with in a 2017 crossover RCT to have significant
either extra virgin olive oil or nuts; vs lower fat effects on lowering postprandial glucose or insu-
diets; 1 year) found an association between lin concentrations when added to beverage or
increased total urinary polyphenol excretion (in low moisture bar high in carbohydrates food
spot urine samples) with decreased levels of products because of relative low sustain intesti-
inflammatory biomarkers and an improvement nal viscosity or gelling properties [133]. A 2008
in cardiovascular risk factors such as LDL- Canadian parallel RCT (210 participants with
cholesterol, HDL-cholesterol and systolic and diabetes treated with antihyperglycemic medica-
diastolic blood pressure [126]. Also, a meta- tions; 61% men; mean age 61 years; high-cereal
analysis of RCTs found that the addition of >8 g fiber or low-glycemic index diets; 6 months)
of dietary fiber from food naturally rich in fiber showed that a low-glycemic index diet resulted
has significant lowering effects on circulating in moderately lower HbA1c levels by 0.50%
CRP levels in overweight/obese adults [128]. A compared with a 0.18% lower HbA1c levels for
2016 pooled analysis (124,086 US men and high-cereal fiber diet [134]. A 2017 Japanese
women) showed that consuming high flavonoid parallel RCT (28 participants with diabetes;
9.4 Mechanisms 283
35% men; mean age 66 years; 250 kcals of in people with diabetes. Other healthy dietary
brown vs white rice; 8 weeks) found that brown patterns which are effective in reducing diabe-
rice intake (which provided 5.6 g/day increase in tes risk and in management of diabetics’
fiber vs white rice) improved endothelial func- health are the DASH, vegan and the healthy
tion by 20.4% vs. -5.8% for white rice (p = Nordic food index diets. Prospective cohort
0.004) [135]. The hs-CRP level tended to studies show that whole (minimally pro-
improve in the brown rice diet group compared cessed) plant foods including whole-grains,
with the white rice diet group (0.01 μg/L vs. fruits, vegetables, dietary pulses, and nuts and
−0.04 μg/L, p = 0.063). The area under the curve flaxseed are significantly associated with
for postprandial glucose was modestly but con- lower risk of diabetes. For whole grains, 3
sistently significantly lower in the brown rice servings/day reduced diabetes risk by 23%
diet group. and of the whole-grains oats and oat bran are
the most effective in managing glycemic con-
Conclusions trol in people with diabetes. For fruits and
Diabetes and related obesity (diabesity) is on vegetables, higher intake of fruits, especially
track to be the largest pandemic in human his- berries, and green leafy vegetables, yellow
tory, with the rate of increase in diabetes vegetables, non-starchy root and cruciferous
increasing at about twice as fast as predicted vegetables are particularly effective in lower-
in 2000. Improperly managed, diabetes leads ing diabetes risk. Three weekly servings of
to a number of health issues, including heart French fries significantly increase diabetes
diseases, stroke, kidney disease, blindness, risk by 41% compared to only 5% for other
nerve damage, leg and foot amputations, and forms of potatoes (baked, boiled or mashed).
premature death. A healthy lifestyle including Higher intake of sugar sweetened fruit juice is
habitual intake of a high quality dietary pat- significantly associated with increased diabe-
tern, regular physical activity, and weight con- tes risk by 28%, while higher intake of 100%
trol are key components of diabetes prevention fruit juice is not associated with diabetes risk.
and management. Prospective cohort studies Higher intake of dietary pulses, peanuts, tree
show that high quality dietary patterns includ- nuts and flaxseed are also associated with
ing the Alternative Healthy Eating Index score lower diabetes risk. Healthy dietary patterns
have a significant inverse association with dia- and specific whole foods beneficially affect
betes risk, and Western dietary patterns have a glycemic and cardiometabolic risk factors,
positive association with risk. Higher adher- which are important for preventing and man-
ence to the MedDiet is associated with a aging diabetes, by helping to control body
19–23% reduced risk of developing diabetes, weight, visceral fat, glucose-insulin homeo-
while the results of RCTs show that the stasis, oxidative stress, inflammation, and
MedDiet can reduce risk of diabetes by 30% endothelial health, lipoprotein concentrations,
and can reduce HbA1c levels by 0.30–0.47% and blood pressure.
Appendix A
Comparison of Western and Healthy Dietary Patterns per 2000 kcal (Approximated Values)
Healthy
Healthy vegetarian
Western dietary USDA base DASH diet Mediterranean pattern
Components pattern (US) pattern pattern pattern (lact-ovo based) Vegan pattern
Emphasizes Refined grains, Vegetables, fruit, Potassium rich Whole grains, Vegetables, Plant foods:
low fiber foods, whole-grain, and vegetables, vegetables, fruit, vegetables,
red meats, low-fat milk fruits, and low fruit, dairy whole-grains, fruits, whole
sweets, and solid fat milk products, olive legumes, nuts, grains, nuts,
fats products oil, and seeds, milk seeds, and soy
moderate wine products, and foods
soy foods
sugar sweetened lean meat, fish, poultry, fish, seeds, and milk milk
beverages, and nuts, seeds, and nuts, and seeds pulses alternatives, alternatives
fast foods vegetable oils and vegetable
oils
vegetables, added sugars sweets and refined grains, meats, or fish; products
whole-grains sugar- and sweets limited sweets allowed
sweetened
beverages
(% Total kcal)
Protein (% Total 16 18 18 18 14–15 13–14
kcal)
Total fat (% Total 33 33 27 32 34 33
kcal)
(% Total kcal)
Unsat. fat 22 25 21 24 26 25
(% Total kcal)
Fiber (g) 16 31 29+ 31 35+ 40+
Potassium (mg) 2800 3350 4400 3350 3300 3650
Vegetable oils (g) 19 27 25 27 19–27 18–27
Sodium (mg) 3600 1790 1100 1690 1400 1225
Plant food groups
Fruit (cup) ≤1.0 2.0 2.5 2.5 2.0 2.0
Vegetables (cup) ≤1.5 2.5 2.1 2.5 2.5 2.5
Whole-grains (oz.) 0.6 3.0 4.0 3.0 3.0 3.0
Legumes (oz.) − 1.5 0.5 1.5 3.0 3.0+
Nuts/Seeds (oz.) 0.5 0.6 1.0 0.6 1.0 2.0
Soy products (oz.) 0.0 0.5 − − 1.1 1.5
terns. 2015;1–9
Appendix B 285
Appendix B
Estimated Range of Energy, Fiber, Nutrients and Phytochemicals Composition of Whole or Minimally
Processed Foods/100 g Edible Portion
Nutrients and Wheat, oats, Apples, pears, Dates, dried figs, Potatoes, Lentils, Almonds, Brazil
phytochemicals barley, brown bananas, grapes, apricots, spinach, carrots, chickpeas, nuts, cashews,
rice, whole oranges, cranberries, peppers, lettuce, split peas, hazelnuts,
grain bread, blueberries, raisins, and green beans, black beans, macadamias,
cereal, pasta, strawberries, and prunes cabbage, onions, pinto beans, pecans, walnuts,
rolls and avocados cucumber, and soy beans peanuts,
crackers cauliflower, sunflower seeds,
mushrooms, and and flaxseed
broccoli
Energy (kcal) 110–350 30–170 240–310 10–115 85–170 520–700
Protein (g) 2.5–16 0.5–2.0 0.1–3.4 0.2–5.0 5.0–17 7.8–24
Available 23–77 1.0–25 64–82 0.2–25 10–27 12–33
Carbohydrate (g)
Fiber (g) 3.5–18 2.0–7.0 5.7–10 1.2–9.5 5.0–11 3.0–27
Total fat (g) 0.9–6.5 0.0–15 0.4–1.4 0.2–1.5 0.2–9.0 46–76
SFAa (g) 0.2–1.0 0.0–2.1 0.0 0.0–0.1 0.1–1.3 4.0–12
MUFAa (g) 0.2–2.0 0.0–9.8 0.0–0.2 0.1–1.0 0.1–2.0 9.0–60
PUFAa (g) 0.3–2.5 0.0–1.8 0.0–0.7 0.0.0.4 0.1–5.0 1.5–47
Folate (ug) 4.0–44 <5.0–61 2–20 8.0–160 50–210 10–230
Tocopherols (mg) 0.1–3.0 0.1–1.0 0.1–4.5 0.0–1.7 0.0–1.0 1.0–35
Potassium (mg) 40–720 60–500 40–1160 100–680 200–520 360–1050
Calcium (mg) 7.0–50 3.0–25 10–160 5.0–200 20–100 20–265
Magnesium (mg) 40–160 3.0–30 5.0–70 3.0–80 40–90 120–400
Phytosterols (mg) 30–90 1.0–83 – 1.0–54 110–120 70–215
Polyphenols (mg) 70–100 50–800 – 24–1250 120–6500 130–1820
Carotenoids (ug) – 25–6600 0.6–2160 10–20,000 50–600 0.0–1200
Circulation. 2013;128:553–65
Rodriguez-Casado A. The health potential of fruits and vegetables phytochemicals: notable examples. Crit Rev Food
Sci Nutr. 2016;56(7):1097–107
Holden JM, Eldridge AL, Beecher GR, et al. Carotenoid content of U.S. Foods: an update of the database. J Food Comp
An. 1999;12:169–96
SFA saturated fat, MUFA monounsaturated fat, PUFA polyunsaturated fat
References 15. Schwingshackl L, Hoffmann G, Lampousi A-M,

et al. Food groups and risk of type 2 diabetes mellitus:
a systematic review and meta-analysis of prospec-
1. Asif M. The prevention and control of type-2
tive studies. Eur J Epidemiol. 2017;32(5):363–75.
diabetes by changing lifestyle and dietary pat-
https://doi.org/10.1007/s10654-0246-y.
tern. J Educ Health Promot. 2014;3:1. https://doi.
16. Ley SH, Pan A, Li Y, et al. Changes in overall diet
org/10.4103/2277-9531.127541.
quality and subsequent type 2 diabetes risk: three US
2. Zimmer PZ. Diabetes and its driver: the larg-
prospective cohorts. Diabetes Care. 2016;39:2011–
est epidemic in human history? Clin Diabetes
8. https://doi.org/10.2337/dc16-0574.
Endrocrinology. 2017;3:1.
17. Satija A, Bhupathiraju SN, Rimm EB, et al. Plant-
3. Zucker I, Shohat T, Dankner R, Chodick G. New
based dietary patterns and incidence of type 2
onset in adulthood is associated with substantial
diabetes in US men and women: results from
risk for mortality at all ages: a population based his-
three prospective cohort studies. PLoS Med.
torical cohort study with a decade-long follow-up.
2016;13(6):e1002039. https://doi.org/10.1371/jour-
Cardiovasc Diabetol. 2017;16:105.
nal.pmed.1002039.
4. Ley SH, Hamdy O, Mahan V, Hu FB. Prevention
18. Maghsoudi Z, Ghiasvand R, Salehi-Abargouei
and management of type 2 diabetes: dietary
A. Empirically derived dietary patterns and incident
components and nutritional strategies. Lancet.
type 2 diabetes mellitus: a systematic review and
2014;383:1999–2007.
meta-analysis on prospective observational studies.
5. Tabák AG, Herder C, Rathmann W, et al. Prediabetes: Public Health Nutr. 2015;19(2):230–41.
a high-risk state for developing diabetes. Lancet. 19. McEvoy CT, Cardwell CR, Woodside JV, et al.
2012;379(9833):2279–90. Posteriori dietary patterns are related to risk of type
6. Salas-Salvado J, Martinez-Gonzalez MA, Bullo M, 2 diabetes: findings from a systematic review and
Ros E. The role of diet in the prevention of type 2 dia- meta-analysis. J Acad Nutr Diet. 2014;114:1759–75.
betes. Nutr Metab Cardiovasc Dis. 2011;21:32–48. 20. Alhazmi A, Stojanovski E, McEvoy M, Garg ML.
7. Jannasch F, Kroger J, Schulze MB. Dietary patterns The association between dietary patterns and type
and type 2 diabetes: a systematic literature review 2 diabetes: a systematic review and meta-anal-
and meta-analysis of prospective studies. J Nutr. ysis of cohort studies. J Human Nutr Dietetics.
2017;147:1174–82. 2014;27:251–60.
8. Tuomilehto J, Linstrom J, Eriksson JG, et al. 21. Exposito K, Chiodini P, Maiorino MI, et al. Which
Prevention of type 2 diabetes mellitus by changes in diet for the prevention of type 2 diabetes? A
lifestyle among subjects with impaired glucose tol- meta-analysis of prospective studies. Endocrine.
erance. N Engl J Med. 2001;344:1343–50. 2014;47(1):107–16.
9. Nield L, Summerbell CD, Hooper L, et al. Dietary 22. Esposito K, Kastorini CM, Panagiotakos DB,
advice for the prevention of type 2 diabetes mellitus Giugliano D. Prevention of type 2 diabetes by
in adults. Cochrane Database Syst Rev. 2008;(3). Art. dietary patterns: a systematic review of prospec-
No.: CD005102. https://doi.org/10.1002/14651858. tive studies and meta-analysis. Metab Syndr Relat
CD005102.pub2. Disord. 2010;8(6):471–6.
10. Micha R, Pehalvo JL, Cudhea F, et al. Association 23. Doostvandi T, Bahadoran Z, Mozaffari-Khosravi H,
between dietary factors and mortality from heart et al. Food intake patterns are associated with the risk
disease, stroke and type 2 diabetes in the United of impaired glucose and insulin homeostasis: a pro-
States. JAMA. 2017;317(9):912–24. https://doi. spective approach in the Tehran Lipid and Glucose
org/10.1001/jama.2017.0947. Study. Public Health Nutr. 2016;19(13):2467–74.
11. Dietary Guidelines Advisory Committee (DGAC). https://doi.org/10.1017/S1368980016000616.
Scientific report. Advisory report to the Secretary 24. Cespedes FM, Hu FB, Tinker L, et al. Multiple
of Health and Human Services and the Secretary healthful dietary patterns and type 2 diabetes in
of Agriculture. Part D. Chapter 2: Dietary patterns, the Women’s Health Initiative. Am J Epidemiol.
foods and nutrients and health outcomes. 2015;1–35. 2016;183(7):622–33.
12. Evert AB, Boucher JL, Cypress M, et al. Nutrition 25. Hong X, Xu F, Wang Z, et al. Dietary patterns and
therapy recommendations for the management of the incidence of hyperglyacemia in China. Public
adults with diabetes. Diabetes Care. 2014;37(suppl Health Nutr. 2015;19(1):131–41.
1):S120–47. 26. Kröger J, Schulze MB, Romaguera D, et al.
13. Ley SH, Korat A, Sun Q, et al. Contribution of the Adherence to predefined dietary patterns and inci-
Nurses’ Health Studies to the uncovering risk factors dent type 2 diabetes in European populations: EPIC-
for type 2 diabetes: diet, Lifestyle, biomarkers, and InterAct Study. Diabetologia. 2014;57:321–33.
genetics. Am J Public Health. 2016;106(9):1624–30. 27. Alhazmi A, Stojanovski E, McEvoy M, et al. Diet
https://doi.org/10.2105/AJPH.2016.303314. quality score is a predictor of type 2 diabetes risk
14. Maghsoudi Z, Azadbakht L. How dietary patterns in women: The Australian Longitudinal Study on
could have a role in prevention, progression, or man- Women’s Health. Br J Nutr. 2014; 112:945–51.
agement of diabetes mellitus? Review on the current 28. Gopinath B, Rochtchina E, Flood VM, Mitchell
evidence. J Res Med Sci. 2012;17(7):694–709. P. Diet quality is prospectively associated with
References 287
incident impaired fasting glucose in older adults. 42. Esposito K, Maiorino MI, Petrizzo M, et al. The
Diabet Med. 2013;30(5):557–62. effects of a Mediterranean diet on the need for dia-
29. de Koning L, Chiuve SE, Fung TT, et al. Diet- betes drugs and remission of newly diagnosed type
quality scores and the risk of type 2 diabetes in men. 2 diabetes: follow-up of a randomized trial. Diabetes
Diabetes Care. 2011;34:1150–6. Care. 2014;37:1824–30.
30. Fung TT, McCullough M, van Dam RM, Hu 43. Snorgaard O, Poulsen GM, Andersen HK, et al.
FB. A prospective study of overall diet quality and Systematic review and meta-analysis of dietary car-
risk of type 2 diabetes in women. Diabetes Care. bohydrate restriction in patients with type 2 diabe-
2007;30(7):1753–7. tes. BMJ Open Diabetes Res Care. 2017;5:e000354.
31. Kim Y, Keogh JB, Clifton PM. Consumption of red and https://doi.org/10.1136/bmjdrc-2016-000354.
processed meat and refined grains for 4 weeks decreases 44. Shirani F, Salehi-Abargouei A, Azadbakht
insulin sensitivity in insulin-resistant adults: a random- L. Effects of DASH diet on some risk for develop-
ized crossover study. Metabolism. 2017;68:173–83. ing type 2 diabetes: a systematic review and meta-
https://doi.org/10.1016/j.metabol.2016.12.011. analysis on controlled clinical trials. Nutrition.
32. Esposito K, Maiorino MI, Bellastella G, et al. A 2013;29(7):939–47.
journey into a Mediterranean diet and type 2 diabe- 45. Yokoyama Y, Barnard ND, Levin SM, Watanabe
tes: a systematic review with meta-analyses. BMJ M. Vegetarian diets and glycemic control in diabetes:
Open. 2015;5(8):e008222. https://doi.org/10.1136/ a systematic review and meta-analysis. Cardiovasc
bmjopen-2015-008222. Diagn Ther. 2014;4(5):373–82.
33. Huo R, Du T, Xu Y, et al. Effects of Mediterranean- 46. Lee Y-M, Kim S-A, Lee I-K, et al. Effect of a brown
style diet on glycemic control, weight loss and rice based vegan diet and conventional diabetic diet
cardiovascular risk factors among type 2 diabe- on glycemic control of patients with type 2 diabe-
tes individuals: a meta-analysis. Eur J Clin Nutr. tes: a 12-week randomized clinical trial. PLoS One.
2015;69:1200–8. 2016;11(6):e0155918. https://doi.org/10.1371/jour-
34. Schwingshackl L, Missbach B, König J, Hoffmann nal.pone.0155918.
G. Adherence to a Mediterranean diet and risk of 47. Barnard ND, Cohen J, Jenkins DJA, et al. A low-
diabetes: a systematic review and meta-analysis. fat vegan diet and a conventional diabetes diet in
Public Health Nutr. 2015;18(7):1292–9. the treatment of type 2 diabetes: a randomized,
35. Koloverou E, Esposito K, Giugliano D, Panagiotakos controlled, 74-wk clinical trial. Am J Clin Nutr.
D. The effect of Mediterranean diet on the develop- 2009;89(suppl):1S–9S.
ment of type 2 diabetes mellitus: a meta-analysis 48. Barnard ND, Cohen J, Jenkins DJ, et al. A low-fat
of 10 prospective studies and 136,846 participants. vegan diet improves glycemic control and cardio-
Metabolism. 2014;63:903–11. vascular risk factors in a randomized clinical trial
36. Schwingshacki L, Hoffmann G. Mediterranean in individuals with type 2 diabetes. Diabetes Care.
dietary pattern, inflammation and endothelial func- 2006;29:1777–83.
tion: a systematic review and meta-analysis of 49. Barnard ND, Scialli AR, Turner-McGrievy G, et al.
intervention trials. Nutr Metab Cardiovasc Dis. The effects of a low-fat, plant-based dietary inter-
2014;24(9):923–39. vention on body weight, metabolism, and insulin
37. Carter P, Achana F, Troughton J, et al. A sensitivity. Am J Med. 2005;118:991–7.
Mediterranean diet improves HbA1c but not fasting 50. Tonstad S, Butler T, Yan R, Fraser GE. Type of veg-
blood glucose compared to alternative dietary strat- etarian diet, body weight, and prevalence of type 2
egies: a network meta-analysis. J Hum Nutr Diet. diabetes. Diabetes Care. 2009;32:791–6.
2014;27:280–97. 51. Lacoppidan SA, Kyrø C, Loft S, et al. Adherence
38. Ajala O, English P, Pinkney J. Systematic review to a Healthy Nordic Food Index is associated with
and meta-analysis of different dietary approaches to a lower risk of type-2 diabetes-The Danish Diet,
the management of type 2 diabetes. Am J Clin Nutr. Cancer and Health Cohort Study. Forum Nutr.
2013;97:505–16. 2015;7:8633–44.
39. Salas-Salvado J, Bullo M, Estruch R, et al. 52. Dietary Guidelines Advisory Committee. Scientific
Prevention of diabetes with Mediterranean diets. report. Advisory report to the Secretary of Health and
Ann Inter Med. 2014;160(1):1–10. Human Services and the Secretary of Agriculture.
40. Martínez-González MA, de la Fuente-Arrillaga Part D. Chapter 1: Food and nutrient intakes, and
C, Nunez-Cordoba JM, et al. Adherence to health: current status and trends. 2015;1–78.
Mediterranean diet and risk of developing diabetes: 53. McGill CR, Fulgoni VL III, Devareddy L. Ten-year
prospective cohort study. BMJ. 2008;336:1348–51. trends in fiber and whole grain intakes and food
41. Maiorino ML, Bellastella G, Petrizzo M, et al. sources for the United States population: National
Mediterranean diet cools down the inflammatory Health and Nutrition Examination Survey 2001–
milieu in type 2 diabetes: the MEDITA random- 2010. Forum Nutr. 2015;7:1119–30.
ized controlled trial. Endocrine. 2016;54(3):634–41. 54. Slavin JL, Lloyd B. Health benefits of fruits and
https://doi.org/10.1007/s12020-016-0881-1. vegetables. Adv Nutr. 2012;3:506–16.
55. WHO/FAO. Diet, nutrition and prevention of 69. Weickert MO, Roden M, Isken F, et al. Effects of sup-
chronic disease: report of a Joint WHO/FAO Expert plemented isoenergetic diets differing in cereal fiber
Consultation. Geneva, Switzerland: World Health and protein content on insulin sensitivity in over-
Organization. 2003/2004. http://whqlibdoc.who.int/ weight humans. Am J Clin Nutr. 2011;94:459–71.
trs/WHO_TRS_916.pdf. Accessed 17 Feb 2015. 70. Pereira MA, Jacobs DR, Pins JJ, et al. Effect of whole
56. World Health Association. Global strategy on diet, grains on insulin sensitivity in overweight hyperin-
physical activity and health—promoting fruit and sulinemic adults. Am J Clin Nutr. 2002;75:848–55.
vegetable consumption around the world. 2013. 71. Weickert MO, Mohlig M, Schofl C, et al. Cereal
http://who.int/dietphysical%20activity/fruit/en/. fiber improves whole-body insulin sensitivity in
Accessed 17 Feb 2015. overweight and obese women. Diabetes Care.
57. Micha R, Khatibzadeh S, Shi P, et al. Global, regional 2006;29:775–80.
and national consumption of major food groups in 72. Robertson MD, Bickerton AS, Dennis AL, et al.
1990 and 2010: a systematic analysis including 266 Insulin-sensitizing effects of dietary resistant starch
country-specific nutrition surveys worldwide. BMJ and effects on skeletal muscle and adipose tissue
Open. 2015;5(9):e008705. https://doi.org/10.1136/ metabolism. Am J Clin Nutr. 2005;82:559–67.
bmjopen-2015-008705. 73. Landberg R, Andersson SO, Zhang JX, et al. Rye
58. Rebello CJ, Greenway FL, Finley JW. A review of whole grain and bran intake compared with refined
the nutritional value of legumes and their effects on wheat decreases urinary C peptide, plasma insulin,
obesity and its related co-morbidities. Obes Rev. and prostate specific antigen in men with prostate
2014;15:392–407. cancer. J Nutr. 2010;140:2180–6.
59. Ros E, Hu FB. Consumption of plant seeds and car- 74. Giacco R, Lappi J, Costabile G, et al. Effects of
diovascular health epidemiological and clinical trial rye and whole wheat versus refined cereal foods on
evidence. Circulation. 2013;128:553–65. metabolic risk factors: a randomised controlled two-
60. Messina V. Nutritional and health benefits of dried centre intervention study. Clin Nutr. 2013;32:941–9.
beans. Am J Clin Nutr. 2014;100(suppl):437S–42S. 75. Juntunen KS, Laaksonen DE, Poutanen KS, et al.
61. Mattes RD, Kris-Etherton PM, Foster GD. High fiber rye bread and insulin secretion and sensi-
Impact of peanuts and tree nuts on body weight tivity in healthy postmenopausal women. Am J Clin
and healthy weight loss in adults. J Nutr. Nutr. 2003;77:385–91.
2008;138(suppl):1741S–5S. 76. Giacco R, Clemente G, Cipriano D, et al. Effects of
62. Aune D, Keum N, Giovannucci E, et al. Whole grain the regular consumption of wholemeal wheat foods
consumption and risk of cardiovascular disease, can- on cardiovascular risk factors in healthy people. Nutr
cer, and all cause and cause specific mortality: sys- Metab Cardiovasc Dis. 2010;20:186–94.
tematic review and dose-response meta-analysis of 77. Andersson A, Tengblad S, Karlström B, et al. Whole
prospective studies. BMJ. 2016;353:i2716. https:// grain foods do not affect insulin sensitivity or markers
doi.org/10.1136/bmj.i2716. of lipid peroxidation and inflammation in healthy, mod-
63. Chanson-Rolle A, Meynier A, Aubin F, et al. erately overweight subjects. J Nutr. 2007;137:1401–7.
Systematic review and meta-analysis of human 78. Brownlee IA, Moore C, Chatfield M, et al. Markers
studies to support a quantitative recommendation of cardiovascular risk are not changed by increased
for whole grain intake in relation to type 2 diabe- whole grain intake: the WHOLE heart study, a ran-
tes. PLoS One. 2015;10(6):e0131377. https://doi. domised, controlled dietary intervention. Br J Nutr.
org/10.1371/journal.pone.0131377. 2010;104:125–34.
64. Ye EQ, Chacko SA, Chou EL, et al. Greater whole- 79. Shen XL, Zhao T, Zhou Y, et al. Effect of oat β-glucan
grain intake is associated with lower risk of type 2 intake on glycaemic control and insulin sensitivity
diabetes, cardiovascular disease, and weight gain. J of diabetic patients: a meta-analysis of randomized
Nutr. 2012;142:1306–13. controlled trials. Forum Nutr. 2016;8:39. https://doi.
65. Parker ED, Liu S, Van Horn L, et al. The association org/10.3390/nu8010039.
of whole grain consumption with incident type 2 dia- 80. He LX, Zhao J, Huang YS, Li Y. The difference
betes: The Women’s Health Initiative Observational between oats and beta-glucan extract intake in the
Study. Ann Epidemiol. 2013;23(6):321–7. management of HbA1c, fasting glucose and insulin
66. Wirstrom T, Hilding A, Gu HF, et al. Consumption sensitivity: a meta-analysis of randomized controlled
of whole grain reduces risk of deteriorating glucose trials. Food Funct. 2016;7(3):1413–28. https://doi.
tolerance, including progression to prediabetes. Am org/10.1039/c5fo1364.
J Clin Nutr. 2013;97:179–87. 81. Thies F, Masson LF, Boffetta P, Kris-Etherton
67. Kochar J, Djousse L, Gaziano JM. Breakfast cereals P. Oats and CVD risk markers: a systematic litera-
and risk of type 2 diabetes in the Physicians’ Health ture review. Br J Nutr. 2014;112:S19–30.
Study I. Obesity. 2007;15(12):3039–44. 82. Bao L, Cai X, Xu M, Li Y. Effect of oat intake on
68. Sun Q, Spiegelman D, van Dam RM, et al. White rice, glycaemic control and insulin sensitivity: a meta-
brown rice, and the risk of type 2 diabetes in US men analysis of randomised controlled trials. Br J Nutr.
and women. Arch Intern Med. 2010;170(11):961–9. 2014;112:457–66.
References 289
83. Wang P-Y, Fang J-C, Gao Z-H, et al. Higher intake 97. Taniguchi A, Yamanaka-Okumura H, Nishida Y,
of fruits, vegetables or their fiber reduces the risk of et al. Natto and viscous vegetables in a Japanese
type 2 diabetes: a meta-analysis. J Diabetes Investig. style meal suppress postrandial glucose and
2016;7:56–69. insulin responses. Asia Pac J Clin Nutr.
84. Borch D, Juul-Hindsgaul N, Veller M, et al. Potatoes 2008;17(4):663–8.
and risk of obesity, type 2 diabetes, and cardiovas- 98. Flood A, Mai V, Pfeiffer R, et al. The effect of
cular disease in apparently healthy adults: a system- high-fruit and -vegetable, high-fiber, low fat dietary
atic review of clinical intervention and observational intervention on serum concentrations of insulin,
studies. Am J Clin Nutr. 2016;104:489–98. https:// glucose, IGF-1 and IGFBP-3. Eur J Clin Nutr.
doi.org/10.3945/ajcn.116.132332. 2008;62(2):186–96.
85. Li S, Miao S, Huang Y, et al. Fruit intake decreases 99. Anderson JW, Waters AR. Raisin consump-
risk of incident type 2 diabetes: an updated meta- tion by humans: effects on glycemia and insu-
analysis. Endocrine. 2015;48(2):454–60. linemia and cardiovascular risk factors. J Food Sci.
86. Li M, Fan Y, Zhang X, et al. Fruit and vegetable intake 2013;78(S1):A11–7.
and risk of type 2 diabetes mellitus: meta-analysis of 100. Anderson JW, Weiter KM, Christian AL, et al.
prospective cohort studies. BMJ Open. 2014;4:10. Raisins compared with other snack effects on gly-
https://doi.org/10.1136/bmjopen-2014-005497. cemia and blood pressure: a randomized, controlled
87. Cooper AJ, Forouhi NG, Ye Z, et al. Fruit and vege- trial. Postgrad Med. 2014;126(1):37–43.
table intake and type 2 diabetes: EPIC-InterAct pro- 101. Esfahani A, Lam J, Kendal CWC. Acute effects of
spective study and meta-analysis. Eur J Clin Nutr. raisin consumption on glucose and insulin response
2012;66(10):1082–92. in healthy individuals. J Nutr Sci. 2014;3(c1). https://
88. Carter P, Gray LJ, Troughton J, et al. Fruit and veg- doi.org/10.1017/jns.2013.33.
etable intake and incidence of type 2 diabetes mel- 102. Becerra-Tomas N, Diaz-Lopez A, Rosique-Esteban
litus: systematic review and meta-analysis. BMJ. N, et al. Legume consumption is inversely associated
2010;341:c4229. https://doi.org/10.1136/bmj.c4229. wirh type 2 diabetes incidence in adults: a prospec-
89. Xi B, Li S, Liu Z, et al. Intake of fruit juice and inci- tive assessment from the PREDIMED study. Clin
dence of type 2 diabetes: a systematic review and Nutr. 2017; doi: 1016/j.clnu.2017.03.015.
meta-analysis. PLoS One. 9(3):e93471. https://doi. 103. Agrawal S, Ebrahim S. Association between legume
org/10.1371/journal.pone.0093471. intake and self-reported diabetes among adult men and
90. Mamluk L, O’Doherty MG, Orfanos P, et al. Fruit women in India. BMC Public Health. 2013;13(1):706.
and vegetable intake and risk of incident of type 2 https://doi.org/10.1186/1471-2458-13-706.
diabetes: results from the consortium on health and 104. Jiang R, Manson JE, Stampfer MJ, et al. Nut and
ageing network of cohorts in Europe and the United peanut butter consumption and risk of type 2 diabe-
States (CHANCES). Eur J Clin Nutr. 2017;71(1):83– tes in women. JAMA. 2002;288(20):2554–60.
91. https://doi.org/10.1038/ejcn. 2016.143. 105. Mueller NT, Odegaard AO, Gross MD, et al. Soy
91. Mursu J, Virtanen JK, Tuomainen T-P, et al. Intake intake and risk of type 2 diabetes mellitus in Chinese
of fruit, berries, and vegetables and risk of type 2 Singaporeans: soy intake and risk of type 2 diabetes.
diabetes in Finnish men: The Kuopio Ischaemic Eur J Nutr. 2012;51(8):1033–40.
Heart Disease Risk Factor Study. Am J Clin Nutr. 106. Ding M, Pan A, Manson JE, et al. Consumption of
2014;99:328–33. soy foods and isoflavones and risk of type 2 diabe-
92. Muraki I, Imamura F, Manson JE, et al. Fruit con- tes: a pooled analysis of three US cohorts. Eur J Clin
sumption and risk of type 2 diabetes: results from Nutr. 2016;70(12):1381. https://doi.org/10.1038/
three prospective longitudinal cohort studies. BMJ. ejcn.2016. 117.
2013;347:f5001. https://doi.org/10.1136/bmj.f5001. 107. Nanri A, Mizoue T, Takahashi Y, et al. Soy prod-
93. Cooper AJ, Sharp SJ, Lentjes MAH, et al. prospec- uct and isoflavone intakes are associated with a
tive study of the association between quantity and lower risk of type 2 diabetes in overweight Japanese
variety of fruit and vegetable intake and incident women. J Nutr. 2010;140:580–6.
type 2 diabetes. Diabetes Care. 2012;35:1293–300. 108. Morimoto Y, Steinbrecher A, Kolonel LN, et al.
94. Muraki I, Rimm EB, Willett WC, et al. Potato con- Soy consumption is protective against diabetes in
sumption and risk of type 2 diabetes: results from Hawaii: The Multiethnic Cohort. Eur J Clin Nutr.
three prospective cohort studies. Diabetes Care. 2011;65(2):279–82.
2016;39:376–84. 109. Sievenpiper JL, Kendall CWC, Esfahani A, et al.
95. Halton TL, Willett WC, Liu S, et al. Potato and Effect of non-oil-seed pulses on glycaemic con-
french fry consumption and risk of type 2 diabetes trol: a systematic review and meta-analysis of
in women. Am J Clin Nutr. 2006;83:284–90. randomised controlled experimental trials in
96. Wallace IR, McEvoy CT, Hunter SJ, et al. Dose- people with and without diabetes. Diabetologia.
response effect of fruit and vegetables on insulin 2009;52:1479–95.
resistance in people at high risk of cardiovascular 110. Pittaway JK, Robertson IK, Ball MJ. Chickpeas may
disease. A randomized controlled trial. Diabetes influence fatty acid and fiber intake in an ad libitum
Care. 2013;36(12):3888–96. diet, leading to small improvements in serum lipid
profile and glycemic control. J Am Diet Assoc. anti-oxidant dietary pattern and diabetes in British
2008;108:1009–13. adults: results from the national diet and nutrition
111. Marinangeli CPF, Jones PJH. Whole and fractionated survey rolling programme years 1-4. Int J Food Sci
yellow pea flours reduce fasting insulin and insulin Nutr. 2015;67(5):553–61. https://doi.org/10.1080/09
resistance in hypercholesterolaemic and overweight 637486.2016.1179268.
human subjects. Br J Nutr. 2011;105:110–7. 125. Wood AD, Strachan AA, Thies F, et al. Patterns of
112. Afshin A, Micha R, Khatibzadeh S, Mozaffarian dietary intake and serum carotenoid and tocopherol
D. Consumption of nuts and legumes and risk of status are associated with biomarkers of chronic
incident ischemic heart disease, stroke, and diabetes: low-grade systemic inflammation and cardiovascu-
a systematic review and meta-analysis. Am J Clin lar risk. Br J Nutr. 2014;112:1341–52. https://doi.
Nutr. 2014;100:278–88. org/10.1017/S0007114514001962.
113. Pan A, Sun Q, Mason JE, et al. Walnut consumption 126. Medina-Remón A, Rosa Casas R, Anna Tressserra-
is associated with lower risk of type 2 diabetes in Rimbau A, et al. Polyphenol intake from a
women. J Nutr. 2013;143:512–8. Mediterranean diet decreases inflammatory bio-
114. Bao Y, Han J, Hu FB, et al. Association of nut con- markers related to atherosclerosis: a substudy
sumption with total and cause-specific mortality. N of the PREDIMED trial. Br J Clin Pharmacol.
Engl J Med. 2013;369:2001–11. 2017;83:114–28.
115. Njike VY, Ayettey R, Petraro P, et al. Walnut
127. Lin D, Xiao M, Zhan J, et al. An overview of plant
ingestion in adults at risk for diabetes: effects phenolic compounds and their importance in human
on body composition, diet quality, and car- nutrition and management of type 2 diabetes.
diac risk measures. BMJ Open Diabetes Res Molecules. 2016;21:1374. https://doi.org/10.3390/
Care. 2015;3:e000115. https://doi.org/10.1136/ molecules21101374.
bmjdrc-2015-000115. 128. Jiao J, Xu J-Y, Zhang W, et al. Effect of dietary fiber
116. Hernandez-Alonso P, Salas-Salvado J, Baldrich- on circulating C-reactive protein in overweight and
Mora M, et al. Beneficial effects of pistachio con- obese adults: a meta-analysis of randomized con-
sumption on glucose metabolism, insulin resistance, trolled trials. Int J Food Sci Nutr. 2015;66(1):114–9.
inflammation, and related metabolic risk mark- https://doi.org/10.3109/09637486.2014.959898.
ers: a randomized clinical trial. Diabetes Care. 129. Bertoia ML, Rimm EB, Mukamal KJ, et al. Dietary
2014;37(11):3098–105. flavonoid intake and weight maintenance: three pro-
117. Wien M, Bleich D, Raghuwanshi M, et al. spective cohorts of 124,086 US men and women fol-
Almond consumption and cardiovascular risk fac- lowed for up to 24 years. BMJ. 2016;352:i17. https://
tors in adults with prediabetes. J Am Coll Nutr. doi.org/10.1136/bmj.i17.
2010;29(3):189–97. 130. Dahl WJ, Stewart ML. Position of the Academy
118. Casas-Agustench P, Lopez-Uriarte P, Bullo M, of Nutrition and Dietetics: health implications
et al. Effects of one serving of mixed nuts on serum of dietary fiber. J Acad Nutr Diet. 2015;115:
lipids, insulin resistance and inflammatory mark- 1861–70.
ers in patients with the metabolic syndrome. Nutr 131. Sylvetsky AC, Edelstein SL, Walford G, et al. A
Cardiovasc Dis. 2011;21(2):126–35. high-carbohydrate, high-fiber, low-fat diet results in
119. Rhee Y, Brunt A. Flaxseed supplementation improved weight loss among adults at high risk of type 2 dia-
insulin resistance in obese glucose intolerant people: a betes. J Nutr. 2017; doi: 10.3945/ jn.117.252395.
randomized crossover design. Nutr J. 2011;10(1):44. 132. de Carvalho CM, de Paula TP, Viana LV, et al. Plasma
https://doi.org/10.1186/1475-2891-10-44. glucose and insulin responses after consumption of
120. Hutchins AM, Brown BD, Cunnane SC, et al. Daily breakfasts with different sources of soluble fiber in
flaxseed consumption improves glycemic control in type 2 diabetes patients: a randomized crossover
obese men and women with pre-diabetes: a random- clinical trial. Am J Clin Nutr. 2017; doi: 10.3945/
ized study. Nutr Res. 2013;33(5):367–75. ajcn.117.157263.
121. McMacken M, Shah S. A plant-based diet for the 133. Rahman S, Zhao A, Xiao D, et al. A randomized,
prevention and treatment of type 2 diabetes. J controlled trial evaluating polydextrose as a fiber in a
Geriatric Cardio. 2017;14:342–54. wet and dry matrix on glycemic control. J Food Sci.
122. Lopez-Jaramillo P. The role of adiponectin in car- 2017; doi: 10.1111/1750-3841.13855.
diometabolic diseases: effects of nutritional inter- 134. Jenkins DJA, Kendall CWC, McKeown-Eyssen G,
ventions. J Nutr. 2016;146(Suppl):422S–6S. et al. Effect of a low-glycemic index or a high-cereal
123. Fisman EZ, Tenenbaum A. Adiponectin: a manifold fiber diet on type 2 diabetes a randomized trial.
therapeutic target for metabolic syndrome, diabetes, and JAMA. 2008;300(23):2742–53.
coronary disease? Cardiovasc Diabetol. 2014;13(1):103. 135. Kondo K, Morino K, Nishio Y, et al. Fiber-rich diet
https://doi.org/10.1186/1475-2840-13-103. with brown rice improves endothelial function in
124. McGeoghegan L, Muirhead CR, Almoosawi S. type 2 diabetes mellitus: A randomized controlled
Association between an anti-inflammatory and trial. PLOS ONE. 2017;12(6):e0179869
Dietary Patterns, Foods,
Nutrients and Phytochemicals 10
in Non-Alcoholic Fatty Liver Disease
Keywords
Non-alcoholic fatty liver disease • Non-alcholic steatohepatitis • Western
lifestyle • Dietary fiber • Omega-3 fatty acids • Monounsaturated fat
Carotenoids • Flavonoids • Soy • Coffee • Obesity • Diet quality • Weight
loss • Mediterranean diet • DASH diet • Low carbohydrate diet
Key Points fatty acids, dietary fiber, vitamin E, carot-

enoids, flavonoids and caffeine, and foods
• Non-alcoholic fatty liver disease (NAFLD) is and beverages including oily fish, extra vir-
the most common cause of chronic liver dis- gin olive oil, oatmeal, coffee, and soy are
ease in the world and its prevalence is increas- associated with lower risk of NAFLD or its
ing concurrently with the obesity pandemic. complications.
• The high prevalence of NAFLD is generally • Higher quality diets including moderate
due to unhealthy high energy dietary patterns energy intake, higher intake of whole (mini-
and sedentary lifestyles leading to obesity, mally processed) plant foods, and low-fat
insulin resistance and metabolic syndrome, dairy, and lower intake of red and processed
which are strongly associated with elevated meat and added sugar and salt, and adequate
hepatic steatosis and increased diabetes risk. physical activity and sleep are associated
• Excessive caloric intake especially from high with prevention and management NAFLD.
intake of refined carbohydrates and saturated • The Western lifestyle is associated with higher
fat promotes increased fatty liver. High intake NAFLD risk and progression to nonalcoholic
of added sugar such as sugar sweetened bever- steatohepatitis (NASH).
ages tends to be a stronger promoter of • Higher adherence to Mediterranean or Dietary
enzymes involved in hepatic de novo lipogen- Approaches to Stop Hypertension (DASH)
esis and NAFLD than higher-fat diets. diets, especially if energy controlled, may be
• Certain nutrients and phytochemicals such effective in managing NAFLD risk and
as omega-3 fatty acids, monounsaturated complications.

292 10 Dietary Patterns, Foods, Nutrients and Phytochemicals in Non-Alcoholic Fatty Liver Disease
10.1 Introduction effects of dietary patterns, specific foods, nutri-

ents and phytochemicals on NAFLD risk and
Nonalcoholic fatty liver disease (NAFLD) is management.
defined by excessive fat accumulation in the liver
(>5% of liver weight) and represents varying
degrees of liver dysfunction and damage that is 10.2 D
iet and Nonalcoholic Fatty
generally associated with a high adherence to the Liver Disease (NAFLD)
Western lifestyle, independently of excessive Progression
alcohol intake or other causes of liver disease
[1–2]. The pandemic of NAFLD, which is the Fatty liver accumulation results from an imbal-
most common cause of chronic liver disease in ance between lipid deposition and removal, asso-
the world, has emerged concurrently with the ciated with excessive uptake of various blood
global obesity pandemic [1–5]. In the USA, over lipids, increased de novo lipogenesis of fatty
64 million individuals are estimated to have some acids and triglycerides, and lower rates of very
form of NAFLD with an annual direct medical low-density lipoproteins (VLDL) release [1, 2,
cost of more than $100 billion [3]. Also, in four 4]. The progression of NAFLD to NASH involves
European countries (Germany, France, UK and a complex interaction of excessive hepatocellular
Italy) over 52 million people have NAFLD with triglyceride accumulation, reduced release of
an annual cost of 35 billion euros. Nonalcoholic VLDL, dysfunctional mitochondrial oxidation,
steatohepatitis (NASH) is a progression of and epigenetic dysfunctional changes via liver
NAFLD characterized by hepatocellular injury or specific DNA methylation. The habitual diet,
hepatic steatosis and associated hepatocyte especially with higher energy and added sugar
injury, inflammation, and fibrosis [1, 4]. intake, plays a relevant role in the pathogenesis
Approximately two-thirds of obese adults have of NAFLD, and both high risk and protective
NAFLD and 20% have NASH. NASH occurs foods have been identified, but the contribution
more frequently in: (1) women than in men, (2) of excess calories remains critical [7–13]. The
obese Hispanics, (3) adults with morbid obesity high prevalence of NAFLD is generally due to
(BMI of ≥40), and (4) older people, especially unhealthy (Western) dietary patterns and seden-
those with sarcopenia [3, 5, 6]. Higher skeletal tary lifestyles leading to obesity and metabolic
muscle mass appears to have a beneficial effect in syndrome, which are strongly associated with
the prevention of NAFLD [6]. The primary clini- hepatic steatosis [14]. Therefore, the first line of
cal risk factors for NAFLD are excess body treatment is lifestyle modification for the preven-
weight (e.g., abdominal fatness), insulin resis- tion and management of NAFLD involving
tance (e.g., prediabetes or type 2 diabetes), and healthy dietary patterns, weight reduction and
cardiovascular disease (e.g., elevated triglycer- increased physical activity leading to an improve-
ides or low HDL-cholesterol), which are gener- ment in serum liver enzymes, reduced hepatic
ally associated with the Western lifestyle, fatty infiltration, and, in advanced conditions,
including high energy dense, low fiber and nutri- reduced degree of hepatic inflammation and
ent dense diets and inactivity [7–13]. Two pri- fibrosis. Key dietary tips for preventing and man-
mary metabolic dysfunctions associated with the aging NAFLD include to: (1) avoid excessive
progression of NAFLD are: (1) insulin resistance energy intake by limiting added sugar, including
has an essential role in the early stages of steato- sugar sweetened beverages and increasing the
sis and (2) hepatic oxidative stress plays an intake of fiber-rich whole foods and (2) for heavy
important role in the progression from simple meat eaters, eating less red and processed meat
steatosis to NASH [7–9]. An overview of the and increasing fish intake. A 2016 Iranian case
pathogenesis of NAFLD is summarized in control study (159 cases and 158 controls; mean
Fig. 10.1 [1, 2, 9, 10]. The primary objective of age 48 years; 57% women) found that: (1) waist
this chapter is to comprehensively review the size and BMI for the NAFLD participants were
10.2 Diet and Nonalcoholic Fatty Liver Disease (NAFLD) Progression 293
Western lifestyle (e.g.,

Body fatness (central), Increased hepatic sedentary, energy
pre-diabetes/diabetes, visceral/ectopic dense diets, high
dyslipidemia & adipose tissue glycemic load,
metabolic syndrome
saturated fat & low
fiber diets
Pro-inflammatory Insulin resistance & Increased free fatty acids

altered glucose & altered lipid metabolism
Hepatic steatosis
Lipotoxicity & inflammation Portal endotoxin from intestinal dysbiosis
Steatohepatitis
(Hepatocyte injury, ballooning,
cell death...)
Risk of cirrhosis or
hepatocellular cancer
Fig. 10.1 Potential factors associated with the pathogenesis of nonalcoholic fatty liver disease (NAFLD) (adapted from
[1, 2, 9, 10])
higher than the healthy participants (p < 0.05); lipogenesis. Dietary fructose, especially from
(2) physical activity level in healthy individuals high added sugar intake, is a stronger promoter of
was more than in patients with NAFLD; (3) enzymes involved in hepatic de novo lipogenesis
dietary intake of saturated fatty acids and sugar in than higher-fat diets. Several properties of fruc-
patients with NAFLD was more than in healthy tose metabolism make it particularly lipogenic
individuals (p < 0.05); and (4) intake of fiber, including its absorption via the portal vein and
folic acid, vitamin D, zinc, and potassium in delivery primarily to the liver to stimulate de
healthy individuals was more than in patients novo lipogenesis enzymes for its conversion into
with NAFLD (p < 0.05) [15]. Weight loss is the triglycerides. Also, fructose supports lipogenesis
most effective way to promote liver fat removal especially conjunction with insulin resistance, as
as 7–10% weight loss is associated with reduced fructose does not require insulin for its metabo-
liver fat, NASH remission, and reduction of lism, and it directly stimulates sterol regulatory
fibrosis [16]. element-binding transcription factor 1 (SREBF1),
Over-consumption of processed foods and a major transcriptional regulator of de novo lipo-
beverages high added sugar are major risk factors genesis. Also, excessive fructose intake increases
for development of NAFLD [17]. Hepatic lipids liver endoplasmic reticulum stress, activates
are derived from dietary intake, esterification of stress-related kinase caused mitochondrial dys-
plasma free fatty acids or hepatic de novo lipo- function leading to increased hepatic glucose
genesis. A central abnormality in NAFLD is pri- transporter type-5 (Glut5) (fructose transporter)
marily associated with increased hepatic de novo gene expression and hepatic lipid peroxidation
14
12
10
Liver Fat (%)
0
Baseline Carbohydrate Overfeeding Weight Loss after
(3 weeks)* Carbohydrate Overfeeding
(6 months)*
Fig. 10.2 Mean liver fat (%) before and after carbohydrate overfeeding and weight loss (p < 0.05*) (adapted from [18])
and inflammation. The lipogenic and proinflam- NAFLD (Fig. 10.3) [19]. Medical students con-
matory effects of fructose appear to be due to suming >2 sugar sweetened soft drinks/day had
transient ATP depletion by its rapid phosphoryla- significantly higher BMI, waist size and diastolic
tion within the cell and its effect of increasing blood pressure than those consuming <1 soft
intracellular and serum uric acid levels. It is pro- drink/day.
jected that population reduction of added sugar
by 20% and 50% could reduce annual direct
medical costs in the USA by more than 10 and 21 10.3 B
ody Weight and Central
billion, respectively, by 2035 [12]. A clinical trial Adiposity
(16 obese subjects; mean BMI 31; hypercaloric
diet >1000 kcal simple carbohydrates/day; Obesity is associated with increased NAFLD risk
3 weeks; hypocaloric weight loss diet for [20]. A 2015 meta-analysis of the effect of adi-
6 months) found that the overfeeding of refined posity on NAFLD risk (5 cohort studies; 6394
carbohydrates significantly increased hepatic fat men and 4246 women; 20–88 years; 1–7 years of
level by 27% compared to baseline levels in follow-up; diagnostic from ultrasonography and
3 weeks and it required 6 months on a weight loss computed tomography) found a positive associa-
diet to restore hepatic fat back to baseline levels tion with NAFLD risk for increased BMI by
(Fig. 10.2) [18]. This trial also found that the 19–43%, waist size by 4–115% and weight gain
increased rate of liver fat increase was tenfold by 21–57% [21]. A 2016 meta-analysis (20 stud-
greater relative to the increase in total body ies) investigating the independent relationship
weight during the high refined carbohydrate between central and general obesity found higher
3-week over-feeding period. Additionally, liver NAFLD risk for higher waist size by 34%, waist
transaminases were significantly increased with to hip ratio by 206% and BMI by 85%, which
refined carbohydrate overfeeding. An Indian indicates that central obesity has a greater impact
cross-sectional study (242 medical students) on NAFLD than general obesity [21]. The high-
found that the consumption of sugar sweetened est vs. lowest adiposity measures increase
soft drinks was positively correlated with NAFLD for BMI by 43%, waist size by 115%
increased incidence of metabolic syndrome and and weight gain by 57%. The 2015 Copenhagen
10.3 Body Weight and Central Adiposity 295
Sugar Sweetened Soft Drink Intake

> or = 2/day 1/day < 1/day
80
70
60
50
Incidence (%)
40
30
20
10
0
Metabolic Syndrome NAFLD
Fig. 10.3 Association between frequency of sugar sweetened soft drink intake and incidences of metabolic syndrome
and nonalcohol fatty liver disease (NAFLD) in 242 Indian medical students (adapted from [19])
School Health Records Register prospective resolution with a weight loss of <3%. In a weight
study (244,464 boys and girls, born between loss study of Danish children RCT (117 over-
1930 and 1989; heights and weights obtained weight/obese children; average age 12 years;
from mandatory school health examinations at 10-week weight- loss camp; exercised moder-
ages 7–13 years; NAFLD cases were observed in ately for 1 h per day and restricted in their caloric
1264 men and 1106 women) showed that higher intake to induce weight loss) found significant
BMIs between 7 and 13 years of age were posi- improvements in all of the parameters of liver fat,
tively associated with NAFLD in both sexes [22]. transaminases, BMI, insulin sensitivity, and
At age 7 years each 1 unit BMI increase was parameters of the metabolic syndrome [28].
associated with a higher adult risk of NAFLD in Twelve months after return to their habitual life-
men by 15% and women by 12%. These results style, these improvements were maintained in
suggest that BMI gain in childhood is an impor- only 24% of the children and 76% returned to
tant indicator of the risk of development of adult their baseline BMI and insulin sensitivity levels.
NAFLD. For overweight or obese individuals, This study illustrates the need for the implemen-
weight loss of ≥7% can help to normalize liver tation of long-term lifestyle approaches to pre-
fat and liver enzyme levels [23]. The most effec- vent and manage NAFLD. In a weight loss RCT
tive method of treating NAFLD is the combina- of NASH patients (31 subjects; mean age
tion of weight loss and exercise [14, 24, 25]. 48 years; mean BMI 33; biopsy proven NASH;
Randomized controlled trials (RCTs) on weight intensive lifestyle intervention with combination
loss show improved NAFLD and NASH, with of diet, physical activity and behavior modifica-
the larger benefit seen with >5% weight loss [26]. tion with a goal of 7–10% weight loss vs. limited
A Hong Kong based RCT (154 NAFLD patients; structured education control; 48 weeks) showed
mean baseline age 51 years; dietitian-led lifestyle that the intervention group lost an average of
modification vs. standard of care; 12 months) 9.3% of their weight vs. 0.2% in the control
found that weight loss can lead to NAFLD remis- group (p = 0.003) [29]. NASH histologic activity
sion (Fig. 10.4) [27]. In NAFLD patients with scores improved significantly in the intervention
>10% weight loss 97% achieved NAFLD remis- group by 55% compared to 29% in the control
sion, compared to 13% who achieved NAFLD group (p = 0.05).
10.4 S
pecific Foods and Dietary 10.4.2 Monounsaturated Fatty Acids
Components (MUFAs)
In addition to reduced energy diets and increased MUFAs are a class of fatty acids that are found in
physical activity to promote weight loss, specific olive oil, nuts, and avocados [35]. The beneficial
foods or beverages and nutrients may modify the effects of MUFAs on cardiovascular disease risk
course of NAFLD and NASH [30]. and blood lipid profiles include decreasing oxidized
LDL, LDL cholesterol, total cholesterol (TC), and
triglycerides concentrations, without decreasing
10.4.1 Healthy Oils and Fatty Acids HDL-C, which is typically seen with low-fat, high
carbohydrate diets. Foods rich in MUFAs displace
The type of dietary fat has mixed effects on the carbohydrates and saturated fat which leads to
pathogenesis of NAFLD and NASH, with healthy reductions in fasting glucose and blood pressure
vegetable oils and fatty acids having a potentially and to an increase in HDL-cholesterol in individu-
protective effect against hepatic steatosis [31]. als with diabetes or metabolic syndrome.
Consequently, an increase in the intake of MUFAs
10.4.1.1 Omega 3 Fatty Acids as a replacement for high carbohydrate and satu-
Omega-3 fatty acids are present in oily fish, rated fat foods, may be protective against NAFLD.
krill oil, algae, walnuts, pine nuts, and flax- A RCT (37 men and 8 women with diabetes; mean
seeds. Several meta-analyses of RCTs suggest age 58 years; mean BMI 30; mean HbA1c 6.6; 8
that increased intake of omega-3 PUFAs can weeks) found that an isocaloric diet enriched in
effectively reduce NAFLD and NASH severity MUFAs significantly reduced hepatic fat content by
[32, 33]. One 2016 meta-analysis (10 RCTs; 8 25% more than a high in carbohydrate diet, inde-
trials placebo controlled; 577 NAFLD/NASH pendent of exercise level [36]. Extra virgin olive oil
patients; median duration 12 months; median is rich in both oleic acid and in phenolic compounds
dose of omega-3 PUFAs 2.9 g/day) found [37]. Several experimental studies on the biological
omega-3 PUFAs improved liver fat, liver activities of olive oil phenolic compounds suggest a
enzymes, triglycerides, and HDL -cholesterol potential for their regulative effect on hepatic lipid
in patients with NAFLD/NASH [32]. Another metabolism by reducing the lipogenic pathway, and
2016 meta- analysis (7 RCTs with 3 double thus providing protection from liver steatosis by
blind trials; 442 subjects [227 NAFLD or exerting protective effects on hepatic mitochondria,
NASH patients and 215 healthy controls]; a key regulator of fatty acid oxidation that counter-
median age 49 years; duration 6–18 months; acts excessive fat storage in hepatocytes and the
omega-3 PUFAs intake 0.83–5 g/day) showed pathogenesis of NAFLD.
that omega-3 PUFAs effectively lowered
NAFLD and NASH severity by decreasing tri-
glycerides, total cholesterol, alanine transami- 10.4.3 Fiber-Rich Whole (Minimally
nase (ALT), and increasing HDL-cholesterol Processed) Plant Foods
and there was a trend toward improving liver
enzymes and LDL-C [33]. Omega-3 PUFAs Increased intake of fiber-rich whole plant foods to
potentially beneficial mechanisms associated achieve the recommended adequate fiber intake
with reduced NAFLD risk include: regulating (14 g/1000 kcals) has the potential to help prevent
gene transcription factors related to both lipid or reduce NAFLD and NASH severity. Fiber-rich
metabolism and insulin sensitivity; stimulating plant foods also tend to have higher antioxidant
hepatic beta oxidation; decreasing endogenous nutrients and phytochemicals (e.g., vitamin E,
lipid production; and down regulating the phenolic acids, carotenoids) density. There are
expression of pro-inflammatory molecules and several mechanisms associated with the increased
of oxygen reactive species [34]. intake of fiber rich diets and whole plant foods.
10.4 Specific Foods and Dietary Components 297
100
90
80
70
NAFLD Remission (%)
60
50
40
30
20
10
0
< 3% 3.0 to 4.9% 5.0 to 6.9% 7.0 to 9.9% >10%
Weight Loss from Baseline Weight
Fig. 10.4 Effect of weight loss on NAFLD remission after 12 months (p-trend p <0.001) (adapted from [27])
10.4.3.1 Body Weight and Central BMI 35; 12 months) it was shown that a single-
Obesity Control goal dietary recommendation to increase fiber
The substitution of high-fiber, lower energy intake (≥30 g/day) vs. a complex, multicompo-
dense foods for lower fiber, higher energy dense nent AHA dietary guidelines weight loss diet
foods typical in the Western diet has been resulted in similar mean weight loss of 2.1 kg and
associated with reduced weight gain and obesity 2.7 kg, respectively [47]. Also, there were no
risk [38–43]. Populations with higher fiber diets between-group differences in NAFLD factors
tend to be leaner than those with low fiber diets. including fasting plasma insulin level, HOMA-IR
A systematic review of 43 prospective and clini- score, HbA1c level, total, LDL, and HDL choles-
cal studies concluded that fiber intake was terol, triglyceride, hs-CRP, or IL-6 levels. This
inversely associated with the risk of gaining body study suggests that a simplified approach to
weight and waist size [42]. Fiber-rich diets have weight reduction emphasizing only increased
been shown to help prevent and/or reverse vis- fiber intake may be a reasonable alternative for
ceral fat accumulation and related central body persons with difficulty adhering to more compli-
fat measures [43–45]. There are several specific cated diet weight loss regimens to help reverse or
RCTs that support the effects of increased fiber reduce risk of NALFD or metabolic syndrome.
rich diets on preventing or attenuating NAFLD
and NASH. In a double-blind RCT (40 subjects; 10.4.3.2 Systemic Inflammation
mean age 38 years; mean BMI 29; 12 weeks), and Insulin Resistance
one oat cereal pack (145 kcal, with carbohydrate Attenuation
25 g; protein 4.0 g; lipid 2.5 g; fiber 3.7 g and Elevated CRP is a biomarker of NAFLD severity
β-glucan 1.5 g) mixed with 250 mL hot water and or progression to NASH and increased fibrosis
replacing a staple food twice daily in meals was [48–49]. A 2017 Italian case-control study found
found to reduce weight and body fat, especially that participants with NASH compared to healthy
abdominal fat and to have liver-protective effects controls consumed significantly more liquid carbo-
compared with the placebo (Fig. 10.5) [46]. In hydrates (approx 145 g vs 90 g/day) and less fiber
obese metabolic syndrome individuals (240 sub- (approx 15 g vs 28 g/day) [50]. A meta-analysis
jects; mean age 52 years; 72% women; mean (14 RCTs) showed that intervention with dietary
fiber or fiber-rich foods, compared with control, nate and butyrate derived from the fermentation
produced a slight, but significant reduction of of fiber by the colonic microbiota bacteria have
0.4 mg/L in circulating CRP levels in overweight been demonstrated to up-regulate autophagy flux
and obese subjects. Subgroup analyses showed that in hepatic cells, which may help to detoxify, repair
such a significant reduction was only observed or regenerate cellular damage associated with
after combining studies where the total fiber intake Western lifestyles including excessive added
was 8 g/d higher in the intervention group than in sugar/fructose intake [56].
the control group [51].
NAFLD and NASH is highly prevalent in peo-
ple with pre-diabetes and type 2 diabetes reflecting 10.4.4 Coffee
the frequent incidence of overweight/obesity and
insulin resistance [52]. High fiber intake has been Coffee contains caffeine, phenols, chlorogenic
shown to be associated with lower pre-diabetes and acids, sugars, organic acids, polysaccharides and
type 2 diabetes risk in both observational and ran- aromatics, triacylglycerols, tocopherols, diterphe-
domized trials. A dose-response meta-analysis of noic alcohols and fatty acids (e.g., cafestol and
17 prospective studies found a non-linear inverse kahweol), which can vary according to the variety,
association between fiber intake and diabetes risk farming practices and the method of preparation
[53]. In the EPIC-InterAct Study, higher total fiber [30]. Epidemiological data suggests that coffee
and cereal fiber intake was associated with about may have a protective effect against the develop-
an 18% lower risk of diabetes after adjustment for ment and in the management of NAFLD and
lifestyle and dietary factors [54]. The Finnish NASH. A systematic review and meta-analysis
Diabetes Prevention Study (522 subjects; 31% men; (5 observational studies) found a significantly
mean BMI 31; mean baseline age 55 years; 4.1 years decreased risk of NAFLD among coffee drinkers
of follow-up) showed that increasing the intake of by 29% and significantly decreased risk by 30% of
fiber-rich whole-grain cereals and fruits and vegeta- liver fibrosis among patients with NAFLD who
bles, is important, not only in terms of overall health drank coffee on a regular basis [57]. Caffeine, the
but also for sustained weight reduction and the pre- primary bioactive phytochemical in coffee, may
vention of type 2 diabetes [41]. provide a hepatoprotective effect via its strong
antioxidant properties helping to reduce oxidative
10.4.3.3 Colonic Microbiota-Liver Axis and inflammation stress in the liver. Also, several
Adequate fiber intake is a potential important con- non-caffeine compounds, such as chlorogenic
tributor to promoting colonic microbiota health acid, are strong antioxidants that have been shown
and protecting against NAFLD, and r educing the to inhibit the accumulation of lipids in hepato-
adverse effects of excessive foods and beverages cytes, promote insulin sensitivity, and reduce
with added sugars [55, 56]. The health effects of inflammatory response.
dietary fiber and prebiotics against NAFLD
pathology potentially include: (1) improving
colonic microbiota health by increasing probiotic 10.4.5 Vitamin E
bacteria levels, short chain fatty acids (SCFAs)
such butyrate, colonic barrier integrity and blood Vitamin E is often used in the treatment of NAFLD
lipid profiles, and decreasing pathogenic bacteria and NASH [58, 59]. A 2015 meta-analysis on both
levels, insulin resistance and fatty liver severity; NAFLD and NASH (5 RCTs) found that vitamin
and (2) decreased hepatic lipid content, oxidative E significantly improved liver function and histo-
and inflammatory stress, circulating levels of logic changes in patients with NAFLD and NASH
transaminases, pro-inflammatory cytokines, and compared to controls [58]. Specifically, vitamin E
endotoxaemia [55]. The SCFAs such as propio- reduced steatosis in NAFLD by 0.54 U/L and in
NASH by 0.67 U/L plus lowered hepatic inflam- 10.4.7 Flavonoids and Soy

mation and enzymes and fibrosis in NASH
patients. Also, another 2015 meta-analysis of Flavonoids, commonly found in fruits, vegetables,
NASH patients (3 RCTs) showed vitamin E sup- nuts, soy, cocoa, and red wine, have been widely
plementation had a significant and positive effect investigated in animal models of NAFLD in stud-
in the improvement of steatosis, ballooning degen- ies including soy isoflavones, green tea flavonoids,
eration, lobular inflammation and fibrosis in quercetin, and rutin [63]. Experimental studies
patients with NASH [59]. As oxidative stress plays have shown the protective biochemical effects of
a vital role in the progression of steatosis to NASH, flavonoids on lipid metabolism, insulin resistance,
vitamin E can act as a lipid-soluble antioxidant. It oxidative stress, and inflammation and beneficial
stabilizes free radical compounds by complexing therapeutic effects on steatosis and liver. A RCT
with unpaired electrons and protecting against (45 patients with NAFLD; mean age 48 years;
lipid peroxidation by acting directly with a variety diets: a soy enriched low-calorie, low-carbohy-
of oxygen radicals [59]. Vitamin E may attenuate drate vs. non-soy low-calorie, low-carbohydrate
NASH via multiple mechanisms, including up- control; 8 weeks) found that both diets promoted
regulating superoxide dismutase activity and similar weight loss but the soy based diet was most
inhibiting genes related to inflammation, fibrosis effective in reducing levels of aspartate amino-
and hepatocellular necrosis. Almonds are a good transferase (AST), serum fibrinogen and malondi-
food source of the α-tocopherol form of vitamin E. aldehyde, which suggest beneficial effects on liver
function in patients with NAFLD and NASH [64].
Currently, more clinical trials are needed to con-
10.4.6 Carotenoids firm the efficacy of flavonoids in the treatment of
NAFLD and NASH patients.
Carotenoids are fat-soluble pigments that give the
yellow, red and orange color to fruits and veg-
etables with β-carotene, lycopene, α-carotene, 10.5 Dietary Patterns
β-cryptoxanthin, lutein and zeaxanthin among the
most studied carotenoids [60]. There are multiple Table 10.1 summarizes the effects of dietary qual-
protective mechanisms for the effects of carotenoids ity and specific dietary patterns studies on the pre-
in NAFLD and NASH, including antioxidant and vention and management of NAFLD and NASH
anti-inflammatory activity. A longitudinal cohort [65–76, 78–81, 83, 84]. An overview of the food
study among middle-aged and older Japanese components and nutrients in specific healthy
subjects showed that the risk of developing ele- dietary patterns are summarized in Appendix A.
vated serum liver enzymes was inversely associ-
ated with baseline serum α- and β-carotene and
β-cryptoxanthin concentrations. This supports the 10.5.1 Low-Carbohydrate Diets
hypothesis that antioxidant carotenoids, especially
provitamin A carotenoids such as α- and β-carotene Low carbohydrate diets have been the focus of
and β-cryptoxanthin might help prevent the devel- potential diet therapies for NAFLD in recent years
opment of NAFLD [61]. An inverse dose-response but there are considerable inconsistencies among
association between serum levels of carotenoids studies on these diets [65]. A 2016 systematic
and the prevalence of NAFLD was observed in a review and meta-analysis (10 RCTs; 230 subjects;
large, community-based study of a middle-aged and mean age approx 47 years; 2–26 weeks) found that
elderly Chinese population [62]. Specifically, serum the consumption of low-carbohydrate diets (with
levels of α -carotene, β -carotene, lutein, zeaxanthin <50% of calories from carbohydrates) in NAFLD
and total serum carotenoids were associated with patients did not significantly reduce the serum con-
significantly lower risk or progression of NAFLD. centration of liver enzymes, but reduced the liver
Table 10.1 Summaries of dietary quality and specific dietary patterns studies on nonalcoholic fatty liver disease
(NAFLD) and nonalcoholic steatohepatitis (NASH)
Low-carbohydrate diets
Haghighatdoost et al. Systematic review and meta-analysis: The consumption of a low-carbohydrate diet
(2016). 10 RCTs; 230 subjects; mean age in NAFLD patients did not reduce the serum
Assess the effect of approx 47 years; <50% of calories from concentration of liver enzymes, but reduced
low-carbohydrate diets carbohydrate, 2–26 weeks the liver fat content. Low carbohydrate diets
on managing NAFLD significantly decreased mean liver fat by
[65] 11.5%
Energy-restricted diets/weight loss
Arefhosseini et al. Double-blind RCT: Both energy-restricted diets decreased liver
(2011). 44 ultrasonography-proven overweight enzymes regardless of their composition,
Examine the effects of NAFLD patients; mean age 41 years; whereas the energy restricted 40% fat energy
two different approx 50% men; diets reduced energy diet was more effective in reduction of weight
compositions of low by 500 kcal; 25 or 40% energy from and triglyceride levels than the diet with 25%
energy diets on NAFLD fat; 6 weeks fat energy
patients (Iran) [66]
Rodriguez-Hernandez Intervention trial: After 6 months, women on both low
et al. (2011). 59 women; mean age 46 years; mean carbohydrate and low-fat diets lost approx.
Evaluate the effect of BMI 38; low carbohydrate diet vs. a 5.6% body weight along with having
weight loss with a low low-fat diet; weight loss; 6 months decreased aminotransferase levels by 22–41%
carbohydrate diet vs. (no significant difference between the diets)
low-fat diet on
aminotransferase levels
in obese women with
NAFLD (Mexico) [67]
Browning et al. (2011). Intervention trial Mean weight loss was similar between the
Determine the 18 NAFLD subjects; 5 men and 13 groups by 4.0 kg in the calorie-restricted
effectiveness of dietary women; mean age 45 years; BMI 35; group and 4.6 kg in the carbohydrate-
carbohydrate and calorie carbohydrate-restricted to <20 g/day or restricted group (p = 0.363). Liver
restriction in reducing calorie restricted to 1200–1500 kcal/ triglycerides decreased significantly with
hepatic triglycerides in day; 2 weeks; hepatic triglycerides weight loss but decreased more in
subjects with NAFLD were measured before and after carbohydrate-restricted subjects by 55% than
(US) [68] intervention by magnetic resonance in calorie-restricted subjects by 28%
spectroscopy
Dietary quality
Katsagoni et al. (2017). Cross-sectional study: A healthy dietary pattern including high
Identify dietary and 136 patients (mean age 47 years; 70% consumption of low-fat dairy products,
lifestyle patterns effects men; mean BMI 31) with ultrasound- vegetables, fish, and optimal sleep duration
on clinical characteristics proven NAFLD; diet and physical was negatively associated with insulin
of individuals with activity level were assessed through resistance (p = 0.008) and liver stiffness
NAFLD (Athens, appropriate questionnaires. Habitual (p = 0.05) after controlling for age, sex, BMI,
Greece) [69] night sleep hours and duration of energy intake, smoking habits, adiponectin,
midday naps were recorded and tumor necrosis factor-α. A Western
dietary pattern including high consumption of
full-fat dairy products, refined cereals,
potatoes, red meat, as well as high television
viewing time was positively associated with
insulin resistance (p = 0.005), although this
association was weakened after adjusting for
adiponectin and tumor necrosis factor-α
Chan et al. (2017). Cross-sectional study: A 10-unit decrease in DQI-I was associated
Examine the association 797 ethnic Chinese subjects; mean age with 24% increase in the odds of having
of diet-quality scores on 48 years (range 19–72 years), 42% NAFLD in the age and sex adjusted model
NAFLD prevalence were male; Diet Quality Index- (p = 0.009) and the association remained
(Hong Kong) [70] International (DQI-I) and significant after further adjusting for other
Mediterranean Diet Score (MDS); 28% lifestyle factors, metabolic and genetic factors
of subjects were diagnosed with (p = 0.027). Multivariate regression analyses
NAFLD showed an inverse association with the intake
of vegetables and legumes, fruits and dried
fruits with NAFLD prevalence (p < 0.05)
Katsagoni et al. (2017). Case-control study: NAFLD patients compared to controls
Explore potential 100 patients with ultrasound-proven consumed less vegetables and nuts, more
associations between NAFLD and 55 healthy controls; sweets, drank less coffee and alcohol (all
dietary intake, physical matched for age, sex, and BMI p < 0.05), and exhibited a lower level of
activity, and sleeping physical activity (p = 0.006). High sweets
habits, and the presence consumption increased NAFLD risk by 113%
of NAFLD (Athens, (p = 0.008) after adjusting for multiple
Greece) [71] confounders, including body weight status.
Optimal sleep duration was associated with
lower NAFLD risk by 62% (p = 0.05)
Wehmeyer et al. (2016). Case-control study: NAFLD patients consumed more daily
Assess the association 55 patients diagnosed with NAFLD calories compared with healthy controls
between dietary patterns were compared to an age and gender- (p < 0.001) and per 1000 kcals a significantly
on NAFLD risk and the matched cohort of 88 healthy higher intake of glucose and protein but a
efficacy in a real-life individuals. lower intake of fiber and minerals than
setting at a tertiary Longitudinal sub-group analysis: 24 healthy controls. In the longitudinal analysis,
medical center NAFLD patients (16 males and 8 patients who significantly reduced their
(Germany) [72] females); mean age 46 years; 6 months caloric intake had lower liver enzyme alanine
after receiving dietary advice aminotransferase (ALT) levels after 6 months
(p < 0.001)
Adriano et al. (2016). Cross-sectional study: NAFLD was inversely associated with higher
Assess the association of 229 older adults; mean age 68 years; adherence to a healthy dietary pattern
dietary patterns with 103 (45%) elderly with NAFLD; (included fruits, vegetables/legumes, white
NAFLD in an elderly diagnosis by ultrasound examination meat, olive oil, margarine and bread/ toast,
population (Brazil) [73] disclosed hepatic steatosis at any stage, and low in red meat) with a reduced risk by
in the absence of excess intake of 30% (p-trend = 0.037) and was directly
alcoholic beverages; 4 dietary patterns associated with the regional snacks pattern
were identified: Traditional, regional with an increased risk of 42%
snacks, energy dense and healthy (p-trend = 0.035) after adjustment for
confounders. None of the other patterns were
associated with NAFLD risk
Yang et al. (2015). Cross-sectional study: After adjusting for potential confounders,
Investigate the 999 Chinese adults; mean age 51 years; higher intake of the animal food pattern
associations between 34.5% were classified as having scores had greater prevalence of NAFLD by
dietary patterns and the NAFLD; 4 major dietary patterns were 35% (p < 0.05) than those in the lowest intake
risk of NAFLD in a identified: Traditional Chinese (coarse and in contrast higher grains-vegetables
middle-aged Chinese grains, fruits, eggs, fish and shrimp, pattern, had a lower prevalence of NAFLD by
population (Hefei milk, and tea), animal food (kelp/ 22% (p < 0.05). There were no significant
Nutrition and Health seaweed and mushroom, pork, beef, effects on NAFLD for the other diets
Study; China) [74] mutton, poultry, cooked meat, eggs,
fish and shrimp, beans and grease),
grains-vegetables (coarse grains,
tubers, vegetables, mushrooms and
kelp/seaweed, beans and fish) and
high-salt diets (rice, pickled vegetables,
processed meat, bacon, salted duck
egg, salted fish and tea)
(continued)
Goletzke et al. (2013). Population-based cohort study: Lower fiber intake was related to significantly
Investigate longitudinal 866 participants not confirmed with higher GGT and fasting triglycerides levels,
associations between NAFLD; median age 67 years with fruit fiber being the most relevant fiber
carbohydrate quality (≥49 years); 63% women; 5 years of source for significantly lowering GGT and
(including dietary follow-up; multi-level mixed regression TG levels
glycaemic index (GI) and analysis was used to relate dietary GI
intakes of sugar, starch and sugar, starch and fiber intake to the
and fiber) and markers of liver enzymes alanine aminotransferase
liver function in an older (ALT), g-glutamyltransferase (GGT),
Australian population and fasting trigylcerides (multivariate
(Blue Mountains Eye adjusted)
Study; Australia) [75]
Oddy et al. (2013). Prospective cohort study: NAFLD was present in 15% of adolescents. A
Examine prospective 995 adolescents; 54.1% overweight or higher Western dietary pattern score at
associations between obese at age 14 years; follow-up at age 14 years was associated with a greater risk of
dietary patterns and 17 years NAFLD at 17 years by 59% (p < 0.005),
NAFLD in a population- although those associations were no longer
based cohort of significant after adjusting for BMI at 14 years.
adolescents (Western A healthy dietary pattern at 14 years appeared
Australian Pregnancy protective against NAFLD at 17 years in
Cohort) [76] centrally obese adolescents by 37%
(p = 0.033)
Trovato et al. (2016). Case-control study: Poorer adherence to a MedDiet profile,
Investigate the effect of 532 NAFLD and 667 non-NAFLD sedentary habits, less sun exposure and use of
Mediterranean diet healthy subjects, mean age 48 years Western diet foods were greater in subjects
(MedDiet) lifestyle on (21–60 years). with NAFLD. Multiple linear regression
NAFLD, with The adherence to MedDiet score was analysis, weighted by years of age, BMI, and
ultrasound-detected fatty assessed on the basis of a 1-wk. recall insulin resistance (HOMA-IR) were the most
liver, compared to computerized questionnaire which powerful independent predictors of fatty liver
non-alcoholic healthy included detailed physical activity severity
subjects (Italy) [78] reports. The Western dietary profile
score, plus a questionnaire quantifying
sun exposure score and sleep habits
Aller et al. (2015). Cross-sectional study: One unit increase in MedDiet score was
Explore potential 82 patients; adherence to the MedDiet associated with a lower likehood of having
associations between pattern by 14-item assessment; 43% steatohepatitis by 57% and steatosis by 58%.
adherence to the had a low grade of steatosis (grade 1 of Secondly, one unit of HOMA-IR was
MedDiet and histological classification) and 57% had a high associated with higher odds of having
characteristics of patients grade of steatosis (grade 2 and 3) steatosis by 101% and liver fibrosis by 38%
with NAFLD (Spain)
[79]
Kontogianni et al. Case-control study: MedDiet score was negatively correlated to
(2014). 58 NAFLD patients and 58 healthy patients’ serum ALT and insulin levels, insulin
Explore the impact of controls matched by age, sex and body resistance index and severity of steatosis and
adherence to the mass index positively to serum adiponectin levels.
MedDiet on the presence Patients with NASH had lower adherence to
and severity of NAFLD the MedDiet compared to those with simple
(Greece) [80] fatty liver. One unit increase in the
MedDietScore was associated with 36%
lower likelihood of having NASH, after
adjusting for sex and abdominal fat level
Ryan et al. (2013). Cross-over RCT: Mean weight loss was not different between
Examine effects of the 12 non-diabetic subjects; 6 females/6 the two diets (p = 0.22). There was a
MedDiet on steatosis and males; mean age 55 years; mean BMI significant reduction in hepatic steatosis after
insulin sensitivity 32; biopsy-proven NAFLD; MedDiet the MedDiet by 32% compared with the
(Australia) [81] and a low fat-high carbohydrate diet; control diet (p = 0.012). Insulin sensitivity
6 weeks duration on each arm; 6 weeks improved with the MedDiet vs. no change for
of wash-out the control diet (p = 0.03) (Fig. 10.6)
Dietary Approaches to Stop Hypertension (DASH)
Zade et al. (2016). Randomized controlled trial (RCT): Adherence to the energy restricted DASH
Determine the effects of 60 NAFLD patients; mean age pattern, compared to the calorie restricted
the Dietary Approaches 41 years; mean BMI 28; 50% women; control diet, significantly lowered weight,
to Stop Hypertension BMI, alanine aminotransferase, alkaline
randomly allocated to calorie-restricted
(DASH) diet on weight control or DASH dietary pattern; phosphatase, insulin levels, insulin resistance
loss and metabolic status 8 weeks; 350–700 kcal/day restricted (HOMA-IR) and increased insulin sensitivity
in overweight patients diets consisted of 52–55% check index (QUICKI). Also, compared with
with NAFLD (Iran) [84] carbohydrates, 16–18% proteins and the control diet, the DASH diet resulted in
30% total fats; however, the DASH diet significant reductions in serum triglycerides
was designed to be rich in fruits, and total-/HDL-cholesterol ratio, hs-CRP,
vegetables, whole grains, and low-fat malondialdehyde, increased levels of nitric
dairy products and low in saturated oxide and glutathione. Also, significantly
fats, cholesterol and refined grains greater percentage of patients in the DASH
group had greater decrease in the severity of
NAFLD than those in the usual control diet
(80% vs. 43%) and greater reduction in both
waist and hip circumference (Fig. 10.7)
Hekmatdoost et al. Case-control study: Participants in the top quartile of a DASH diet
(2016). 102 patients with newly diagnosed score were 30% less likely to have NAFLD;
Examine the association NAFLD and 204 controls. Adherence however, after more adjustment for
between adherence to the to DASH-style diet was assessed using dyslipidemia and BMI the risk was reduced to
DASH diet and risk of a validated food frequency an insignificant 8%
NAFLD (Iran) [85] questionnaire, and a DASH diet score
based on food and nutrients
emphasized or minimized in the DASH
diet
fat content [65]. Low carbohydrate diets signifi-reduced energy diet with 40% energy from fat
cantly decreased mean liver fat by 11.5%. was more effective than the 25% fat energy diet
in reduction of body weight and triglycerides
levels [66]. A Mexican RCT (9 women; mean
10.5.2 Energy Restricted/Weight age 46 years; mean BMI 38; energy controlled
Loss Diets diets low carbohydrate vs. a low-fat; 6 months)
showed that there was no difference in weight
Three RCTs assessed effects of energy restricted loss and aminotransferase levels between the
diets on the management of NAFLD [66–68]. An two diets [67]. A US RCT (18 NAFLD subjects;
Iranian double blind RCT (44 ultrasonography- 5 men and 13 women; mean age 45 years; BMI
proven overweight NAFLD patients; mean 35; diets: energy and carbohydrate-restricted
age 41 years; approx 50% men; two diets with (<20 g/day) vs. calorie restricted; 2 weeks)
reduced energy by 500 kcal with 25% vs. 40% found similar weight loss in both diets but liver
energy from fat; 6 weeks) found that both triglycerides decreased significantly more in
restricted energy diets decreased liver enzymes carbohydrate-restricted subjects by 55% than in
regardless of their composition, while the calorie-restricted subjects by 28% [68].
10.5.3 Dietary Pattern Quality inversely associated with higher adherence to the
healthy pattern (included fruits, vegetables/legumes,
Eight observational studies provide insights regard- white meat, olive oil, margarine and bread/ toast,
ing the effects of dietary pattern quality on NAFLD and low in red meat) with a reduced risk by 30%
or NASH [69–76]. A 2017 Greek cross-sectional whereas a snacking pattern was directly associated
analysis (136 NAFLD patients) observed that a with an increased multivariate risk of 42% [73]. A
healthy dietary pattern including high consumption 2015 Chinese cross-sectional study (999 subjects;
of low-fat dairy products, vegetables, fish, and opti- mean age 51 years; 34.5% confirmed NAFLD
mal sleep duration were associated with lower insu- cases) found that subjects in the highest quartile of
lin resistance and liver stiffness whereas a Western the animal food pattern scores had greater preva-
dietary pattern including high consumption of full- lence of NAFLD by 35% than those in the lowest
fat dairy products, refined cereals, p otatoes, and red quartile whereas high intake of a whole grains and
meat was associated with high insulin resistance vegetable pattern had a lower prevalence of NAFLD
independent of BMI and energy intake [69]. Also, a by 22%, after adjusting for BMI [74]. Two 2013
2017 Greek case-control study (100 patients with Australian prospective studies demonstrated the
ultrasound-proven NAFLD and 55 healthy con- importance of a healthy diet in preventing NAFLD
trols; matched for age, sex, and BMI) found that in both adults and children [75, 76]. A population-
NAFLD patients compared to controls consumed based cohort from the Blue Mountains Eye Study
less vegetables and nuts, more sweets, drank less (866 participants; median age 67 years; 63%
coffee, ate more sweets, and had a lower level of women; 5-year follow-up) found that lower fiber
physical activity [71]. High sweets consumption intake was associated with higher gamma-glutam-
increased NAFLD risk by 113%, after adjusting for yltransferase (GGT) and fasting triglycerides levels
multiple confounders, including body weight status. [75]. An evaluation of adolescents associated with
Optimal sleep duration was associated with lower the Western Australian Pregnancy Cohort (95 ado-
NAFLD risk by 62%. A 2017 Hong Kong cross- lescents; 54.1% overweight or obese at age 14 years;
sectional study (797 ethnic Chinese subjects; mean follow-up at age 17 years) found that 15% of the
age 48 years, 42% were male; 28% diagnosed with adolescents had NAFLD [76]. Also, this study
NAFLD) demonstrated that a 10-unit (100 point showed that a higher Western dietary pattern score
scale) decrease in Diet Quality Index-International at age 14 years was associated with a greater risk of
was associated with 24% increase in the odds of NAFLD at 17 years by 59%, whereas a healthy
having NAFLD whereas an inverse association dietary pattern at 14 years appeared protective
with NAFLD prevalence was observed with higher against NAFLD at 17 years in centrally obese ado-
intake of vegetables and legumes, fruits and dried lescents by 37%.
fruits, after multivariate adjustment [70]. A 2016
German case-control study (55 patients diagnosed
with NAFLD were compared to an age and gender- 10.5.4 Mediterranean Diet (MedDiet)
matched cohort of 88 healthy individuals) and
Longitudinal subgroup analysis (24 NAFLD Among several specific dietary approaches that
patients; 16 males and 8 females; mean age 46 years; exert positive effects in NAFLD patients, the
6 months after receiving dietary advice) found that MedDiet improves risk factors associated with
NAFLD patients consumed more daily calories NAFLD and NASH, metabolic syndrome and dia-
(higher glucose and lower fiber and minerals/1000 betes [77]. The main dietary features of the MedDiet
kcals) compared with healthy controls [72]. This include extra virgin olive oil, fish, nuts, fruits, veg-
analysis showed that NAFLD patients who signifi- etables, beans, high-fiber breads with limited meat,
cantly reduced their caloric intake had lower liver cheese, and sweets. The liberal use of extra virgin
enzyme levels. A 2016 Brazilian cross-sectional olive oil enhances the bioavailability of fatsoluble
study (229 older adults; mean age 68 years; 45% antioxidant vitamins and phytochemicals from the
with confirmed NAFLD) showed that NAFLD was MedDiet. Specifically, MedDiet mechanisms for
Change in Biometrics Units vs Control

–0.5
–1
–1.5
–2
–2.5
–3
Waist-to-hip
Weight (kg) BMI Body fat (%)
ratio
Biometrics –2.6 –0.96 –1.32 –0.02
Fig. 10.5 Effect of oatmeal* as an isocaloric replacement for other foods on weight biometrics compared to control
(double-blind RCT; p<.05 for all) (adapted from [46]). *Each serving of oatmeal contained 145 kcal with 3.7 g of fiber
including 1.5 g of β-glucan mixed with 250 mL hot water twice daily
NAFLD protection include: fiber-rich foods which aminotransferase, insulin resistance index and
induce low glycemic response are beneficial for severity of steatosis and positively to serum adipo-
improved glucose and insulin metabolism; unsatu- nectin levels [80]. Also, each unit increase in the
rated fatty acids which are associated with better MedDiet score was associated with 36% lower
hepatic lipid metabolism; and antioxidant com- odds of NASH, after adjusting for sex and abdomi-
pounds, such as dietary polyphenols and carot- nal fat level. A 2013 Australian cross-over RCT (12
enoids, which are thought to attenuate oxidation non-diabetic subjects; 6 females/6 males; mean age
and inflammation. All these factors are thought to 55 years; mean BMI 32; with biopsy-proven
help to reduce NALFD risk and slow progression to NAFLD; MedDiet vs. a low fat-high carbohydrate
NASH. Three observational [78–80] and two RCTs diet; 6 weeks each, separated by a 6-week washout
[81, 82] assessed the effects of the MedDiet on period) found that the mean weight loss was not dif-
NAFLD or NASH risk and management. A 2016 ferent between the two diets but subjects on the
Italian case-control study (532 NAFLD and 667 MedDiet had a significant relative reduction in
non-NAFLD healthy subjects; mean age 48 years) hepatic steatosis by 32% and increased insulin sen-
showed that poorer adherence to a MedDiet, seden- sitivity compared with the low-fat, high carbohy-
tary habits, less sun exposure and use of “Western drate diet (Fig. 10.6) [81]. A 2017 Italian double-blind
diet” foods were associated with greater NAFLD RCT (98 adults with NAFLD; low glycemic index
incidence [78]. Also, older age, higher BMI and MedDiet vs usual (control) diet; 6 months) showed
elevated HOMA-IR were the strongest independent that the MedDiet significantly reduced NAFLD
predictors of fatty liver severity. A 2015 Spanish severity compared to the control diet [82].
cross-sectional study (82 NAFLD patients) found
that a one unit increase in MedDiet score was asso-
ciated with a lower likehood of having steatohepati- 10.5.5 Dietary Approaches to Stop
tis by 57% and steatosis by 58% [79]. A 2014 Greek Hypertension (DASH)
case-control study (58 NAFLD patients and 58
healthy controls matched by age, sex and body The DASH diet is rich in fruits, vegetables,
mass index) showed that the MedDiet score was whole grains, and low-fat dairy foods; includes
negatively correlated to patients’ serum alanine meat, fish, poultry, nuts, and beans; and is l imited
MedDiet Low Fat/High Carbohydrate Diet

10
0
% Change from Baseline Fasting insulin HOMA-IR score Intra-hepatic lipid
(p<0.01) (p<0.01) (p<0.05)
–10
–20
–30
–40
–50
Fig. 10.6 Effect of Mediterranean diet (MedDiet) compared to a lower-fat and high carbohydrate diet from randomised,
crossover 6-week trial which included 12 non-diabetic subjects (6 females/6 males) with biopsy-proven NAFLD risk
factors (adapted from [81])
in sugar-sweetened foods and beverages, red vs. 43%) and greater reduction in both waist and
meat, and added fats for a wide range of health hip circumference (Fig. 10.7). A 2016 case-con-
and chronic disease risk reduction benefits [82]. trol study (102 patients with newly diagnosed
A 2017 meta-analysis (7 RCTs) found that adher- NAFLD and 204 controls) found that a higher
ence to the DASH diet is effective in improving DASH diet score was associated with 30% lower
circulating serum inflammatory biomarkers in risk of NAFLD but after adjustment for dyslipid-
adults, compared with a usual diet; therefore, it emia and BMI the risk changed to an insignifi-
could be an effective dietary option to suppress cant 8% reduction [85].
liver inflammation process [83]. A 2016 RCT
(60 NAFLD patients; mean age 41 years; mean Conclusions
BMI 28; 50% women; randomly allocated to NAFLD is the most common cause of chronic
calorie-
restricted DASH diet or control usual liver disease in the world and its prevalence is
diet; 8 weeks) showed that the adherence to the increasing concurrently with the obesity pan-
energy restricted DASH pattern, compared to the demic. The high prevalence of NAFLD is
energy restricted usual diet, significantly generally due to unhealthy high energy
improved weight, BMI, waist size, liver enzymes, dietary patterns and sedentary lifestyles lead-
markers of insulin metabolism, serum triglycer- ing to obesity, insulin resistance and meta-
ides, hs-CRP, plasma malondialdehyde (MDA), bolic syndrome, which are strongly associated
nitric oxide (NO) and glutathione (GSH) levels with elevated hepatic steatosis and increased
[84]. Also, a significantly greater percentage of diabetes risk. Excessive caloric intake, espe-
patients in the DASH group had decreased grade cially from high intake of refined carbohy-
of NAFLD than those in the control group (80% drates and saturated fat, promotes increased
Weight (kg) BMI (kg/m2) Waist circumference (cm) Hip circumference (cm)
0
Energy Restricted DASH Diet Energy-Restricted Usual Diet
–0.5
Change in Weight Units from Baseline
–1
–1.5
–2
–2.5
–3
–3.5
–4
–4.5
–5
Fig. 10.7 Effect of energy restriction Dietary Approaches to Stop Hypertension (DASH) and usual diets by 350-700
kcal on body weight and composition in persons with NAFLD (p <.01 for all) (adapted from [84])
fatty liver. High intake of added sugar such moderate energy intake, higher intake of
sugar sweetened beverages tends to be a whole (minimally processed) plant foods, and
stronger promoter of enzymes involved in low-fat dairy, and lower intake of red and pro-
hepatic de novo lipogenesis and NAFLD than cessed meat and added sugar and salt, and
higher-fat diets. Certain nutrients and phyto- adequate physical activity and sleep are asso-
chemicals such as omega-3 fatty acids, mono- ciated with prevention and management of
unsaturated fatty acids, fiber, vitamin E, NAFLD. The Western lifestyle is associated
carotenoids, flavonoids and caffeine, and with higher NAFLD risk and progression to
foods and beverages including oily fish, extra NASH. Higher adherence to a MedDiet or
virgin olive oil, oatmeal, coffee, and soy are DASH diets, especially if energy controlled,
associated with lower risk of NAFLD or its may be effective in managing NAFLD risk
complications. Higher quality diets including and complications.
Appendix A: Comparison of Western and Healthy Dietary Patterns per 2000 kcal

(Approximated Values)
Healthy Healthy
Western dietary USDA base DASH diet Mediterranean vegetarian pattern Vegan
Components pattern (US) pattern pattern pattern (Lact-ovo based) pattern
Emphasizes Refined grains, Vegetables, Potassium Whole grains, Vegetables, fruit, Plant foods:
low fiber foods, fruit, rich vegetables, whole-grains, vegetables,
red meats, whole-grain, vegetables, fruit, dairy legumes, nuts, fruits, whole
sweets, and solid and low-fat fruits, and products, olive seeds, milk grains, nuts,
fats milk low fat milk oil, and products, and soy seeds, and
products moderate wine foods soy foods
Includes Processed meats, Enriched Whole-grain, Fish, nuts, Eggs, non-dairy Non-dairy
sugar sweetened grains, lean poultry, fish, seeds, and milk alternatives, milk
beverages, and meat, fish, nuts, nuts, and pulses and vegetable oils alternatives
fast foods seeds, and seeds
vegetable oils
vegetables, and added sugars sweets, and refined grains, meats, or fish; products
whole-grains sugar- and sweets limited sweets
sweetened
beverages
(% Total kcal)
Protein (% Total 16 17 18 16 14 13
kcal)
Total fat (% Total 33 32 27 34 32 30
kcal)
Total kcal)
Unsat. fat (% 22 25 21 24 26 25
Total kcal)
Fiber (g) 16 31 29+ 31 35+ 40+
Potassium (mg) 2800 3350 4400 3350 3300 3650
Vegetable oils (g) 19 27 25 27 19–27 18–27
Sodium (mg) 3600 1790 1100 1690 1400 1225
Plant Food Groups
Fruit (cup) ≤1.0 2.0 2.5 2.5 2.0 2.0
Vegetables (cup) ≤1.5 2.5 2.1 2.5 2.5 2.5
Whole-grains 0.6 3.0 4.0 3.0 3.0 3.0
(oz.)
Legumes (oz.) – 1.5 0.5 1.5 3.0 3.0+
Nuts/seeds (oz.) 0.5 0.6 1.0 0.6 1.0 2.0
Soy products 0.0 0.5 – – 1.1 1.5
(oz.)
patterns. 2015;1–9
2010. 7th ed. Washington, DC: U.S. Government Printing Office; 2010. Table B2.4; http://www.choosemyplate.gov/.
References 309
References 16. Marchesini G, Petta S, Dalle Grave R. Diet, weight

loss, and liver health in nonalcoholic fatty liver
disease: pathophysiology, evidence, and prac-
1. Rinella ME. Nonalcoholic fatty liver disease a sys-
tice. Hepatology. 2016;63(6):2032–43. https://doi.
tematic review. JAMA. 2015;313(22):2263–73.
org/10.1002/hep.28392.
2. Tiniakos DG, Vos MB, Brunt EM. Nonalcoholic fatty
17. Softic S, Cohen DE, Kahn CR. Role of dietary fruc-
liver disease: pathology and pathogenesis. Annu Rev
tose and hepatic de novo lipogenesis in fatty liver
Pathol. 2010;5:145–71.
disease. Dig Dis Sci. 2016;61(5):1282–93. https://doi.
3. Younossi ZM, Blissert D, Blissert R, et al. The eco-
org/10.1007/s10620-016-4054-0.
nomic and clinical burden of nonalcoholic fatty liver
18. Sevastianova K, Santos A, Kotronen A, et al. Effect
disease in the United States and Europe. Hepatology.
of short-term carbohydrate overfeeding and long-term
2016;64:1577–86.
weight loss on liver fat in overweight humans. Am J
4. Dongiovanni P, Valenti L. A nutrigenomic approach
Clin Nutr. 2012;96:727–34.
to non-alcoholic fatty liver disease. Int J Mol Sci.
19. Siddiqi Z, Karoli R, Fatima J, et al. Soft drink con-
2017; 18:1534.
sumption and the risk of nonalcoholic fatty liver dis-
5. Williams CD, Stengel J, Asike MI, et al. Prevalence
ease. J Assoc Physicians India. 2017;65(5):28–32.
of nonalcoholic fatty liver disease and nonalcoholic
20. Jiménez-Cruz A, Gómez-Miranda LM, Ramírez GD,
steatohepatitis among a largely middle-aged popula-
et al. Adiposity as a risk factor of nonalcoholic fat dis-
tion utilizing ultrasound and liver biopsy: a prospec-
ease: a systemic review. Nutr Hosp. 2014;29(4):771–5.
tive study. Gastroenterology. 2011;140(1):124–31.
21. Pang Q, Zhang JY, Song SD, et al. Central obesity and non-
6. Tovo CV, Fernandes SA, Buss C, de Mattos AA. alcoholic fatty liver disease risk after adjusting for body
Sarcopenia and non-alcoholic fatty liver disease: is mass index. World J Gastroenterol. 2015;21(5):1650–62.
there a relationship? A systematic review. World J https://doi.org/10.3748/wjg.v21.i5.1650.
Hepatol. 2017;9(6):326–32. https://doi.org/10.4254/ 22. Zimmermann E, Gamborg M, Holst C, et al. Body
wjh.v9.i6.326. mass index in school-aged children and the risk of
7. Schwenger KJP, Allard JP. Clinical approaches to non- routinely diagnosed non-alcoholic fatty liver dis-
alcoholic fatty liver disease. World J Gastroenterol. ease in adulthood: a prospective study based on the
2014;20(7):1712–23. Copenhagen School Health Records Register. BMJ
8. Conlon BA, Beasley JM, Aebersold K, et al. Open. 2015;5:e006998.
Nutritional management of insulin resistance in non- 23. Ratziu V, Bellentani S, Cortez-Pinto H, et al. A posi-
alcoholic fatty liver disease (NAFLD). Forum Nutr. tion statement on NAFLD/NASH based on the EASL
2013;5:4093–114. 2009 special conference. J Hepatol. 2010;53:372–84.
9. Hashiba M, Ono M, Hyogo H, et al. Glycemic vari- 24. Copaci I, Lupescu I, Caceaune E, et al. Noninvasive
ability is an independent predictive factor for devel- markers of improvement of liver steatosis achieved by
opment of hepatic fibrosis in nonalcoholic fatty liver weight reduction in patients with nonalcoholic fatty
disease. PLoS One. 2013;8(11):e76161. liver disease. Rom J Intern Med. 2015;53(1):54–62.
10. Polimeni L, Ben MD, Baratta F, et al. Oxidative stress: 25. Montesi L, Caselli C, Centis E, et al. Physical activ-
new insights on the association of nonalcoholic fatty ity support or weight loss counseling for nonal-
liver disease and atherosclerosis. World J Hepatol. coholic fatty liver disease. World J Gastroenterol.
2015;7(10):1325–36. 2014;20(29):10128–36.
11. Papandreou D, Andreou E. Role of diet on non-
26. Hsu CC, Erik Ness E, Kowdley KV. Nutritional
alcoholic fatty liver disease: an updated narrative approaches to achieve weight loss in nonalcoholic
review. World J Hepatol. 2015;7(3):575–82. fatty liver disease. Adv Nutr. 2017;8:253–65.
12. Vreman RA, Goodell AJ, Rodriquez LA, et al. Health and 27. Wong VW, Chan RS, Wong GL, et al. Community-
economic benefits of reducing sugar intake in the USA, based lifestyle modification programme for non-
including effects via non-alcoholic fatty liver disease: a alcoholic fatty liver disease: a randomized controlled
microsimulation model. BMJ Open. 2017;7:e013543. trial. J Hepatol. 2013;59:536–42.
13. Smits MM, Ioannou GN, Boyko EJ, Utzschneider 28. Grønbaek H, Lange A, Birkebaek NH, et al. Effect of
KM. Non-alcoholic fatty liver disease as an inde- a 10-week weight-loss camp on fatty liver disease and
pendent manifestation of the metabolic syndrome: insulin sensitivity in obese Danish children. J Pediatr
results of a US national survey in three ethnic groups. Gastroenterol Nutr. 2012;54(2):223–8.
J Gastroenterol Hepatol. 2013;28:664–70. 29. Promrat K, Kleiner DE, Niemeier HM, et al.

14. Zelber-Sagi S, Ratziu V, Oren R. Nutrition and
Randomized controlled trial testing the effects of
physical activity in NAFLD: an overview of the weight loss on nonalcoholic steatohepatitis (NASH).
epidemiological evidence. World J Gastroenterol. Hepatology. 2010;51(1):121–9.
2011;17:3377–89. 30. Gupta V, Mah XJ, Garcia MC, et al. Oily fish, cof-
15. Zolfaghari H, Askari G, Siassi F, et al. Intake of nutri- fee and walnuts: dietary treatment for nonalco-
ents, fiber, and sugar in patients with nonalcoholic holic fatty liver disease. World J Gastroenterol.
fatty liver disease in comparison to healthy individu- 2015;21(37):10621–35. https://doi.org/10.3748/wjg.
als. Int J Prev Med. 2016;7:98. v21.i37.10621.
31. Ferramosca A, Zara V. Modulation of hepatic ste- circumference in European men and women. Am J
atosis by dietary fatty acids. World J Gastroenterol. Clin Nutr. 2010;91:329–36.
2014;20(7):1746–55. https://doi.org/10.3748/wjg.v20. 46. Chang H-C, Huang C-N, Yeh D-M, et al. Oat prevents
i7.1746. obesity and abdominal fat distribution and improves liver
32. Lu W, Li S, Li J, et al. Effects of omega-3 fatty acid function in humans. Plant Foods Hum Nutr. 2013;68:18–
in nonalcoholic fatty liver disease: a meta-analysis. 23. https://doi.org/10.1007/ s11130-013-0336-2.
Gastroenterol Res Pract. 2016;1459790. https://doi. 47. Ma Y, Olendzki BC, Wang J, et al. Single-component
org/10.1155/2016/1459790. versus multicomponent dietary goals for the meta-
33. He X-X, Wu X-L, Chen R-P, et al. Effectiveness of bolic syndrome a randomized trial. Ann Intern Med.
omega-3 polyunsaturated fatty acids in non-alcoholic 2015;162:248–57.
fatty liver disease: a meta-analysis of randomized 48. Yoneda M, Mawatari H, Fujita K, et al. High-

controlled trials. PLoS One. 2016;11(10):e0162368. sensitivity C-reactive protein is an independent clini-
https://doi.org/10.1371/journal.pone.0162368. cal feature of nonalcoholic steatohepatitis (NASH)
34. Di Minno MND, Russolillo A, Lupoli R, et al.
and also of the severity of fibrosis in NASH. J
Omega-3 fatty acids for the treatment of non- Gastroenterol. 2007;42(7):573–82.
alcoholic fatty liver disease. World J Gastroenterol. 49. Maleki I, Rastgar A, Hosseini V, et al. High sensi-
2012;18(41):5839–47. https://doi.org/10.3748/wjg. tive CRP and pentraxine 3 as noninvasive biomark-
v18.i41.5839. ers of nonalcoholic fatty liver disease. Eur Rev Med
35. Finelli C, Tarantino G. Is there any consensus as
Pharmacol Sci. 2014;18(11):1583–90.
to what diet or lifestyle approach is the right one 50. Federic A, Dallio M, Caprio GG, et al. Qualitative and
for NAFLD patients? J Gastrointest Liver Dis. quantitative evaluation of dietary intake in patients
2012;21(3):293–302. with non-alcoholic steatohepatitis. Nutrients. 2017;
36. Bozzetto L, Prinster A, Annuzzi G, et al. Liver fat is 9:1074.
reduced by an isoenergetic MUFA diet in a controlled 51. Jiao J, Xu JY, Zhang W, et al. Effect of dietary fiber on
randomized study in type 2 diabetic patients. Diabetes circulating C-reactive protein in overweight and obese
Care. 2012;35:1429–35. adults: a meta-analysis of randomized controlled tri-
37. Priore P, Cavallo A, Gnoni A, et al. Modulation of als. Int J Food Sci Nutr. 2015;66(1):114–9. https://doi.
hepatic lipid metabolism by olive oil and its phenols org/10.3109/09637486.2014.959898.
in nonalcoholic fatty liver disease. IUBMB Life. 52. Ortiz-Lopez C, Lomonaco R, Orsak B, et al.

2015;67(1):9–17. Prevalence of prediabetes and diabetes and metabolic
38. European Food Safety Authority (EFSA). EFSA
profile of patients with nonalcoholic fatty liver disease
panel on dietetic products, nutrition, and allergies. (NAFLD). Diabetes Care. 2012;35:873–8.
Opinion on dietary reference values for carbohydrates 53. Yao B, Fang H, Xu W, et al. Dietary fiber intake and
and dietary fibre. EFSA J. 2010;8(3):1462. risk of type 2 diabetes: a dose-response analysis of
39. Liu S, Willett WC, Manson JE, et al. Relation
prospective studies. Eur J Epidemiol. 2014;29(2):
between changes in intakes of dietary fiber and grain 79–88.
products and changes in weight and development of 54. The InterAct Consortium. Dietary fibre and incidence
obesity among middle-aged women. Am J Clin Nutr. of type 2 diabetes in eight European countries: the
2003;78:920–7. EPIC-InterAct Study and a meta-analysis of prospec-
40. Slavin JL. Dietary fiber and body weight. Nutrition. tive studies. Diabetologia. 2015;58(7):1394–408.
2005;21:411–8. https://doi.org/10.1007/s00125-015-3585-3589.
41. Lindstrom J, Peltonen M, Eriksson JG, et al. High- 55.
Mokhtari Z, Gibson DL, Hekmatdoost A.
fibre, low-fat diet predicts long-term weight loss Nonalcoholic fatty liver disease, the gut microbiome,
and decreased type 2 diabetes risk: the Finnish and diet. Adv Nutr. 2017;8:240–52.
Diabetes Prevention Study. Diabetologia. 2006;49: 56. Karkman A, Lehtimäki J, Ruokolainen L. The ecology
912–20. of human microbiota: dynamics and diversity in health
42. Tucker LA, Thomas KS. Increasing total fiber intake and disease. Ann N Y Acad Sci. 2017;1399(1):78–92.
reduces risk of weight and fat gains in women. J Nutr. https://doi.org/10.1111/nyas.13326.
2009;139:576–81. 57. Wijarnpreechaa K, Thongprayoona C, Ungprasertb
43. Fogelholm M, Anderssen S, Gunnarsdottir I, Lahti- P. Coffee consumption and risk of nonalcoholic fatty
Koski M. Dietary macronutrients and food consump- liver disease: a systematic review and meta-analysis.
tion as determinants of long-term weight change in Eur J Gastroenterol Hepatol. 2017;29(2):e8–e12.
adult populations: a systematic literature review. Food https://doi.org/10.1097/MEG.0000000000000776.
Nutr Res. 2012;56:19103. 58. Sato K, Gosho M, Yamamoto T, et al. Vitamin E has
44. Romaguera D, Angquist L, Du H, et al. Dietary deter- a beneficial effect on nonalcoholic fatty acid disease:
minants of changes in waist circumference adjusted a meta-analysis of randomized controlled trials.
for body mass index—a proxy measure of visceral Nutrition. 2015;31(7–8):923–30.
adiposity. PLoS One. 2010;5(7):e11588. 59. Xu R, Tao A, Zhang S. Association between vitamin
45. Du H, van der AD, Boshuizen HC, et al. Dietary fiber E and non-alcoholic steatohepatitis: a meta-analysis.
and subsequent changes in body weight and waist Int J Clin Exp Med. 2015;8(3):3924–34.
References 311
60. Yilmaz B, Sahin K, Bilen H, et al. Carotenoids and dietary pattern. Medicine. 2016;95:23. https://doi.
non-alcoholic fatty liver disease. Hepatobiliary Surg org/10.1097/MD.0000000000003887.
Nutr. 2015;4(3):161–71. https://doi.org/10.3978/j. 73. Adriano LS, de Carvalho Sampaio HA, Arruda SPM,
issn.2304-3881.2015.01.11. et al. Healthy dietary pattern is inversely associated
61. Sugiura M, Nakamura M, Ogawa K, et al. High
with non-alcoholic fatty liver disease in elderly. Br
serum carotenoids are associated with lower risk for J Nutr. 2016;115:2189–95. https://doi.org/10.1017/
developing elevated serum alanine aminotransferase S0007114516001410.
among Japanese subjects: the Mikkabi cohort study. 74. Yang C-Q, Shu L, Wang S, et al. Dietary patterns
Br J Nutr. 2016;115:1462–9. https://doi.org/10.1017/ modulate the risk of non-alcoholic fatty liver disease
S0007114516000374. in Chinese adults. Forum Nutr. 2015;7:4778–91.
62. Cao Y, Wang C, Liu J, et al. Greater serum carotenoid 75. Goletzke J, Buyken AE, Gopinath B, et al. Carbohydrate
levels associated with lower prevalence of nonalco- quality is not associated with liver enzyme activ-
holic fatty liver disease in Chinese adults. Sci Rep. ity and plasma TAG and HDL concentrations over
2015;5:12951. https://doi.org/10.1038/ srep12951. 5 years in an older population. Br J Nutr. 2013;110:
63. Van De Wier B, Koek GH, Bast A, et al. The potential 918–25.
of flavonoids in the treatment of non-alcoholic fatty 76. Oddy WH, Herbison CE, Jacoby P, et al. The west-
liver disease. Crit Rev Food Sci Nutr. 2017;57(4):834– ern dietary pattern is prospectively associated with
55. https://doi.org/10.1080/10408398.2014.952399. nonalcoholic fatty liver disease in adolescence. Am J
64. Kani AH, Alavian SM, Esmaillzadeh A, et al. Effects Gastroenterol. 2013;108:778–85.
of a novel therapeutic diet on liver enzymes and coag- 77.
Godos J, Federico A, Dallio M, Scazzina
ulating factors in patients with non-alcoholic fatty F. Mediterranean diet and nonalcoholic fatty liver dis-
liver disease: a parallel randomized trial. Nutrition. ease: molecular mechanisms of protection. Int J Food
2014;30:814–21. Sci Nutr. 2017;68(1):18–27. https://doi.org/10.1080/0
65. Haghighatdoost F, Salehi-Abargouei A, Surkan PJ, 9637486.2016.1214239.
Azadbakht L. The effects of low carbohydrate diets on 78. Trovato FM, Martines GF, Brischetto D, Trovato

liver function tests in nonalcoholic fatty liver disease: G, Catalano D. Neglected features of lifestyle:
a systematic review and meta-analysis of clinical tri- their relevance in non-alcoholic fatty liver disease.
als. J Res Med Sci. 2016;21:53. World J Hepatol. 2016;8(33):1459–65. https://doi.
66. Arefhosseini SR, Ebrahimi-Mameghani M, Farsad Naeimi org/10.4254/wjh. v8. i33.1459.
A, et al. Lifestyle modification through dietary interven- 79. Aller R, Izaola O, de la Fuente B, de Luis

tion: health promotion of patients with nonalcoholic fatty D. Mediterranean diet is associated with liver histol-
liver disease. Health Promot Perspect. 2011;1:147–54. ogy in patients with non-alcoholic fatty liver disease.
67.
Rodriguez-Hernandez H, Cervantes-Huerta M, Nutr Hosp. 2015;32(6):2518–24.
Rodriguez-Moran M, Guerrero-Romero F. Decrease 80. Kontogianni MD, Tileli N, Margariti A, et al.

of aminotransferase levels in obese women is related Adherence to the Mediterranean diet is associated
to body weight reduction, irrespective of type of diet. with the severity of non-alcoholic fatty liver disease.
Ann Hepatol. 2011;10:486–92. Clin Nutr. 2014;33(4):678–83.
68. Browning JD, Baker JA, Rogers T, Davis J, et al. Short- 81. Ryan MC, Itsiopoulos C, Thodis T, et al. The

term weight loss and hepatic triglyceride reduction: Mediterranean diet improves hepatic steatosis and
evidence of a metabolic advantage with dietary carbo- insulin sensitivity in individuals with non-alcoholic
hydrate restriction. Am J Clin Nutr. 2011;93:1048–52. fatty liver disease. J Hepatol. 2013;59:138–43.
69. Katsagoni CN, Georgoulis M, Papatheodoridis GVA, 82. Misciagna G, Del Pilar Diaz M, Caramin DV, et al.
et al. Associations between lifestyle characteristics and Effect of s low glycemic index Mediterranean diet on
the presence of nonalcoholic fatty liver disease: a case- non-clacholic fatty liver disease. A randomized con-
control study. Metab Syndr Relat Disord. 2017;15(2):72– trolled clinical trial. J Nutr. 2017;21(4):404–12.
9. https://doi.org/10.1089/met.2016.0105. 83. Soltani S, Chitsazi MJ, Salehi-Abargouei A. The effect
70. Chan R, Wong VW-S, Chu WC-W, et al. Diet-quality of dietary approaches to stop hypertension (DASH)
scores and prevalence of nonalcoholic fatty liver dis- on serum inflammatory markers: a systematic review
ease: a population study using proton-magnetic reso- and meta-analysis of randomized trials. Clin Nutr.
nance spectroscopy. PLoS One. 2017;10(9):e0139310. 2017. https://doi.org/10.1016/j.clnu.2017.02.018.
https://doi.org/10.1371/journal.pone.0139310. 84. Zade MR, Telkabadi MH, Bahmani F, et al. The

71. Katsagoni CN, Papatheodoridis GV, Papageorgiou
effects of DASH diet on weight loss and metabolic
MV, et al. A “healthy diet-optimal sleep” lifestyle pat- status in adults with non-alcoholic fatty liver disease:
tern is inversely associated with liver stiffness and a randomized clinical trial. Liver Int. 2016;36:563–
insulin resistance in patients with nonalcoholic fatty liver 71. https://doi.org/10.1111/liv.12990.
disease. Appl Physiol Nutr Metab. 2017;42(3):250–6. 85. Hekmatdoost A, Shamsipour A, Melbodi M, et al.
https://doi.org/10.1139/apnm-2016-0492. Adherence to the Dietary Approaches to Stop
72. Wehmeyer MH, Zyriax B-C, Jagemann B, et al.
Hypertension (DASH) and risk of nonalcoholic fatty
Nonalcoholic fatty liver disease is associated with liver disease. Int J Food Sci Nutr. 2016;67(8):1024–9.
excessive calorie intake rather than a distinctive https://doi.org/10.1080/09637486.2016.1210101.
Part IV
Cardiovascular and Cerebrovascular
Diseases, and Age-Related
Cognitive Function
Dietary Patterns and Coronary
Heart Disease 11
Keywords
Dietary patterns • Mediterranean diet • DASH diet • Vegetarian diet
• Western diet • Coronary heart disease • Carotid initima-media thickness
• Lipoproteins • C-reactive protein
Key Points electrolytes, and are anti-inflammatory com-

pared with Western diets.
• Globally coronary heart disease (CHD) is the • Meta-analyses estimate that healthy dietary
leading cause of death and morbidity in patterns are associated with a significantly
adults. The type of dietary pattern consumed decreased CHD risk by 20–33%, while
plays an important role in the risk of develop- Western-type patterns are associated with an
ing CHD. increased CHD risk by up to 45%, especially
• Healthy dietary patterns (or higher nutrient qual- in US studies and in individuals over 50 years
ity diets) which are associated with decreased of age.
risk of CHD include higher Healthy Eating • A number of randomized controlled trials
Indices scores, Mediterranean diet (MedDiet), (RCTs) support the role for healthy dietary
Dietary Approaches to Stop Hypertension patterns in reducing CHD risk biomarkers
(DASH) and vegetarian diets and are character- including blood lipids and lipoproteins, sys-
ized by higher consumption of vegetables, fruits, temic inflammatory or oxidative stress factors
whole-grains, low-fat dairy, and seafood, limited such as hs-CRP and oxidized LDL-C, and
(or no) intake of red and processed meat, and carotid atherosclerosis and improving endo-
lower intakes of refined grains, sugar-sweetened thelial health and blood pressure.
foods and beverages compared to an increased • Replacing 5% of energy intake from saturated
CHD risk associated with higher adherence to fats with equivalent energy intake from poly-
Western dietary patterns. unsaturated and monounsaturated fats, or car-
• Healthy dietary patterns have CHD protective bohydrates from whole grains was associated
effects because they are lower in energy den- with a significant 25%, 15%, and 9% lower
sity and higher in fiber, healthier fatty acid risk of CHD, respectively. However, replacing
profiles, essential nutrients, antioxidants, and saturated fat with carbohydrates from refined

316 11 Dietary Patterns and Coronary Heart Disease
starches and/or added sugars was not signifi- Healthy lifestyles play an important role in
cantly associated with lower CHD risk. preventing and managing CHD risk and include
nonsmoking status, BMI <25, physical activity at
goal levels, and a diet consistent with the dietary
11.1 Introduction guidelines [7–11]. However, less than 1% of
Americans meet all four ideal lifestyle criteria for
Globally coronary heart disease (CHD) is the cardiovascular health [6, 12]. Suboptimal diet
leading cause of death and morbidity in adults [1, quality is a leading risk factor for CHD [6] and
2]. While CHD death rates have declined over the less than 5% of the US population meet all the
years because of diagnostic intensity and treat- American Heart Association’s (AHA) definition
ment, the number of deaths remains high and this of an ideal healthy diet including consuming fruit
is expected to increase regardless of ethnicity and vegetables at ≥4.5 cups/day; whole-grains at
with the global aging of populations [1–4]. CHD ≥3 servings/day; fish at ≥ two 3.5 oz. servings/
affects both men and women with prevalence week; sodium at ≤2300 mg/day and sugar sweet-
increasing after the fifth decade of life in men and ened beverages at ≤36 oz./week) [12]. Despite the
the sixth decade of life in women [5]. In the USA, subjects consuming above the the recommended
heart disease effects someone every 43 seconds, energy intake by 22% for women and 10% for
killing over 375,000 people a year [6]. Heart dis- men, high adherence to the AHA healthy dietary
ease is the leading cause of death in women, criteria has been shown to reduce the risk of BMI
causing more deaths than all forms of cancer increase, a major risk factor for CHD (Fig. 11.1)
combined. The lifetime risk of developing heart [12]. The average US Dietary Guidelines based
disease after age 40 is approximately 49% for Healthy Eating Index score for adults is only
men and 32% for women. Between 2015 and about half the ideal healthy diet score needed to
2030, annual US costs related to CHD and related reduce CHD risk (Fig. 11.2) [13]. Thus, Americans
atherosclerotic cardiovascular diseases are fore- are at heightened CHD risk levels with about 43%
casted to increase from $84.8 to $202 billion. of American adults having total cholesterol of
Low AHA diet score (0-1) Higher AHA diet score (2-5)
0.8
0.6
Change in BMI
0.4
0.2
0
0 1 2 3 4
Follow-up (years)
Fig. 11.1 Association between adherence to the American Heart Association’s (AHA) healthy diet* and BMI over
4 years in young adults (p = 0.03) (adapted from [12]). *AHA’s healthy diet: fruit and vegetables ≥4.5 cups/day;
whole-grains ≥3 servings/day; fish ≥ two 3.5 oz. serving/week; sodium ≤2300 mg/day and sugar sweetened beverages
≤36 oz./week)
11.2 Dietary Patterns and CHD Risk 317
100
90
Average HEI Score for US Adults

80
70
60
50
40
30
20
10
0
1999-2000 2001-2002 2003-2004 2005-2006 2007-2008 2009-2010
Fig. 11.2 The average US Dietary Guidelines based Healthy Eating Index (HEI) score for adults from NHANES data
over 10 years (adapted from [13])
200 mg/dL or higher [6]. The French Three City fat, cholesterol, and sodium and increased intake
Study in elderly adults (9,294 participants; 37% of fiber, potassium, and unsaturated fats [8–15].
men; mean baseline age 74 years; median follow- Healthy dietary patterns have CHD protective
up of 10.9 years) found a 67% lower risk of CHD effects because they influence a number of bio-
and stroke with the highest adherence to a healthy logical mechanisms including lowering fasting
lifestyle including a healthy diet [9]. The partici- lipid profiles, circulating inflammation, abdominal
pants healthier lifestyles had lower BMIs, systolic and visceral fat, blood pressure; and improving
blood pressure, total cholesterol, glycemia and insulin sensitivity and microbiota health [6–20].
higher HDL-cholesterol levels. Another study The 2015 Dietary Guidelines Advisory Committee
showed that bowel movement frequency is concluded that strong and consistent evidence
inversely associated with CHD mortality as a dys- demonstrates that healthy dietary patterns, charac-
functional colon from poor quality diets may con- terized by higher consumption of vegetables,
tribute to systemic oxidative and inflammatory fruits, whole-grains, low-fat dairy, and seafood,
stress [11]. The objective of this chapter is to and lower consumption of red and processed meat,
review the effects of dietary patterns on CHD risk and lower intakes of refined grains, and sugar-
and its primary biomarkers. sweetened foods and beverages are associated
with decreased CHD risk relative to high adher-
ence to Western patterns [19]. Healthy dietary pat-
11.2 Dietary Patterns and CHD Risk terns based on the US Dietary Guidelines, Dietary
Approaches to Stop Hypertension (DASH),
The type of dietary pattern consumed has an Mediterranean diet (MedDiet), lacto-ovovegetar-
important role in affecting the risk of developing ian or healthy vegan diets vs. the Western-type
CHD. Prospective cohort studies and randomized pattern are summarized in Appendix A.
controlled trials (RCTs) have consistently demon-
strated that healthy dietary patterns compared to
Western dietary patterns exert clinically meaning- 11.2.1 Healthy vs. Western Dietary
ful lowering of CHD risk and associated risk fac- Patterns
tors, including improving blood lipid profiles and
maintaining normal blood pressure levels based Healthy diets reduce CHD risk by reducing
on reduction of dietary energy density, saturated known biomarker levels and slowing progression
of atherosclerosis compared to Western diets demographics, while Western-type patterns are

known to elevate CHD risk by increasing bio- associated with an increased CHD risk by up to
marker levels and promoting progression of ath- 53% in studies involving US adults, but Asian
erosclerosis [20]. Table 11.1 summarizes populations had 15% increased risk (Fig. 11.3)
prospective cohort studies on healthy and Western [21, 22]. Studies on the effects of specific types of
diets and CHD risk [21–26]. Several meta- Western diets on CHD risk include the following:
analyses suggest that healthy dietary patterns are (1) high adherence to the US Southern diet, rich
associated with a significantly decreased CHD in fried foods, organ and processed meats, and
risk by 20–35% independent of age or other sugar sweetened beverages (17,418 adults;
Table 11.1 Summary of prospective cohort studies on healthy vs. Western dietary patterns and CHD risk
Zhang et al. (2015). 28 cohorts and 7 case-control; Healthy dietary patterns significantly
Evaluate the effects of dietary 983,484 participants decreased risk of CHD by 33% (highest
patterns and the risk of CHD vs. lowest adherence). Western-type
[21] dietary patterns increased CHD risk by
45% (highest vs. lowest adherence) (Fig.
11.3)
Hou et al. (2015). 12 prospective studies; 409,780 Highest adherence to a healthy dietary
Examine potential associations participants; 6298 CHD cases pattern reduced CHD risk by 20% vs.
between dietary patterns and lowest adherences. In the US population,
CHD [22] the Western dietary pattern significantly
increased CHD risk by 45%
Prospective studies
Shikany et al. (2015). 17,418 adults; baseline The highest consumers of the Southern
Evaluate dietary pattern factors age ≥ 45 years; 5 primary dietary pattern (characterized by added fats, fried
related to CHD risk (Reasons for patterns classified as: convenience,food, eggs, organ and processed meats,
Geographic and Racial plant-based, sweets, southern, and and sugar-sweetened beverages)
Differences in Stroke alcohol and salad; median 5.8 years experienced a significant 56% higher risk
[REGARDS]; US) [23] of follow-up; 536 acute CHD cases of CHD (comparing quartile extremes of
(multivariate adjusted) intake). Adding BMI and medical history
variables to the model attenuated the
association somewhat to 37% increased
risk (p = 0.036)
Chiuve et al. (2012). 71,495 women, baseline age The AHEI-2010, which explicitly
Assess the effect of a healthy 30–55 years, follow-up 24 years, emphasizes high intakes of whole grains,
diet and risk of major chronic 4868 CVD cases; 41,029 men, ages PUFA, nuts, and fish and reductions in
disease (Nurses’ Health Study 40–75 years; 22 years of follow-up; red and processed meats, refined grains,
and Health Professionals about 5102 cases of CVD; Alterative and sugar-sweetened beverages, was
Follow-Up Studies; US) [24] Healthy Eating Index (AHEI)-2010 significantly associated with lower risk of
(multivariate adjusted) CHD by 31%
Iqhal et al. (2008). 52 countries; 5761 cases of acute MI, There was an inverse association between
Examine the association between 10,646 controls; mean baseline age the healthy pattern and acute myocardial
dietary patterns and acute 55 years;75% men; sedentary; 3 infarction, with a significantly reduced
myocardial infarction (Canada/ major dietary patterns including risk by 30% (highest vs. lowest
Global INTERHEART study) Asian (high intake of tofu and soy adherence). High adherence to the
[25] and other sources), Western (high in Western dietary pattern increased the
fried foods, salty snacks, eggs, and adjusted acute myocardial infarction risk
meat), and healthy diet (high in fruit by 30%
and vegetables); 4 years of follow-up
Hu et al. (2000). 44,875 men; aged 40–75 years; The healthy pattern significantly reduced
Study whether overall dietary 8 years of follow-up; 2 dietary CHD risk by 30% and the Western
patterns derived from a patterns: Healthy pattern, pattern significantly increased CHD risk
food-frequency questionnaire characterized by higher intake of by 64% (highest vs. lowest adherence)
predict risk of CHD in men vegetables, fruit, legumes, whole
(Health Professionals Follow-up grains, fish, and poultry vs. Western
Study; US) [26] pattern, characterized by higher
intake of red meat, processed meat,
refined grains, sweets and desserts,
French fries, and high-fat dairy
products (multivariate adjusted)
< 50 years > 50 years White Asian and other

60
50
40
30
Risk of CHD (%)
20
10
0
Healthy Dietary Patterns Western Dietary Patterns
–10
–20
–30
–40
Fig. 11.3 Association between dietary patterns and coronary heart disease (CHD) risk in population sub-groups from
a meta-analysis of 35 prospective studies )adapted from [21, 22])
baseline age ≥ 45 years; 5.8 years of follow-up) reduced CHD risk by about 30% [24–26]. A
significantly increased CHD risk by 56% [23]; cross-sectional investigation from the Multi-
(2) diets rich in fried foods, salty snacks and Ethnic Study of Atherosclerosis (MESA) (2407
meats (16,417 participants; mean baseline age men and 2682 women; aged 45–84 years)
55 years; 75% men; 262 centers in 52 countries observed that the inflammatory markers CRP and
representing all geographic regions; 4 years of IL-6 were significantly increased with high
follow-up) significantly increased acute myocar- adherence to Western food patterns (high in pro-
dial infarction (MI) risk by 30% [25]; and (3) cessed meats, fried potatoes, salty snacks, and
higher intake of red and processed meats, refined desserts) and decreased with high adherence to
grains, sweets, French fries and high fat dairy healthy whole food patterns (rich in whole grains,
products (44,875 men; aged 40–75 years; 8 years fruit, nuts, and green leafy vegetables) [27]. A
of follow-up) significantly increased CHD risk by British RCT (165 healthy non-smoking men and
64% [26]. These studies also show that high women; aged 40–70 years; 12 weeks) demon-
adherence to healthy dietary patterns significantly strated that high adherence to the UK Healthy
Dietary Guidelines diet reduced risk of fatal CHD year [32]. Table 11.2 summarizes the prospec-
by 15% and non-fatal CHD by 30% compared to tive studies and RCT trials on MedDiet pat-
the traditional British diet [28]. This RCT also terns, and CHD risk and biomarkers.
showed significant reductions in systolic blood
pressure by 4.2 mmHg, total/HDL-cholesterol 11.2.2.1 Prospective Studies
ratio by 0.13, CRP by 36%, and weight by 1.9 kg Prospective studies consistently demonstrate that
with the UK Dietary Guidelines diet compared to high adherence to the MedDiet significantly
the traditional British diet. reduces CHD risk [33–39]. A meta-analysis (35
cohort studies; 4,172,412 subjects) found that
each two point (18 point scale) increase in
11.2.2 Mediterranean Diet (MedDiet) MedDiet adherence reduced risk of CVD by
10%, which is strongly associated with lower
Increased focus on the MedDiet first started in CHD risk [33]. An Italian prospective study
the 1960s when Ancel Keys, in the Seven evlauating subjects with both low and high
Countries Study, observed that populations liv- Framingham CVD risk scores (1658 participants;
ing near the Mediterranean Sea had the lowest mean age 52 years; 12 years of follow-up)
incidence of CHD, longer healthy life expec- observed that individuals with low CVD risk
tancy and longevity compared to other parts of scores had significantly lower risk of CVD events
the world [15]. Now, the MedDiet is the most by 21% per one unit increase in MedDiet score
extensively investigated of the healthy dietary (9 point scale) [34]. High adherence to the
patterns for CHD benefits [29–31]. Dietary pat- MedDiet in middle-aged individuals showed
terns based on the MedDiet are higher in total reduced risk of CVD events and mortality in
fat (35–40% total energy), lower in saturated older age. An international cohort study assess-
fat (7–10% total energy), higher in fiber (27– ing MedDiet score compared to the Western diet
37 g/day), and rich in monounsaturated fat score (15,482 patients with stable coronary heart
(MUFAs) and adequate in omega-3 s and anti- disease from 39 countries; mean age 67 years;
oxidant and anti-inflammatory plant compo- median 3.7 years of follow-up) found that a one
nents such as polyphenols and carotenoids [31]. unit increase in MedDiet score was associated
The criteria used to assess adherence, included: with 5% lower risk of major adverse cardiovascu-
use of olive oil (or in combination with avocado lar events after adjusting for all covariates
oil) as the main fat; ≥2 servings/day of vegeta- whereas there was no association between one
bles; ≥3 servings/day of fruits; <1 serving/day unit change in Western diet score and major
of red meat, hamburger or meat products (ham, adverse cardiovascular events (18 point scale)
sausage, etc.); <1 serving/day of butter, marga- [35]. This study suggests that an increased intake
rine, or cream; <1 serving/day of sweetened of healthy foods has an important role for sec-
and/or carbonated beverages; ≥1 serving/day ondary prevention of CHD. The Greek ATTICA
of wine; ≥3 servings/week of legumes; ≥3 prospective study (3042 statin using participants;
servings/week of fish or shellfish; (11) <3 serv- age 18–89 years; 50% men; 10 years of follow-
ings/week of commercial sweets or pastries up) observed that high adherence to the MedDiet
(not homemade), such as cakes, cookies, bis- reduced CVD risk by 29% vs. a low MedDiet
cuits or custard; ≥3 servings/week of nuts adherence [36]. This study also found that high
(including peanuts); preferential consumption adherence to MedDiets has adjunctive effects
of chicken or turkey meat instead of veal, pork, with statin use in lowering CHD risk. In the US
hamburger or sausage; and ≥2 servings/week Coronary Artery Risk Development in Young
of sofrito, a sauce made with tomato and onion, Adults (CARDIA) study (5115 subjects; aged
leek or garlic and simmered with olive oil or 18–30 years; 25 years of follow-up), higher
avocado oil [31]. A US study estimates that the adherence to a MedDiet score was shown to
high adherence to a MedDiet can significantly reduce metabolic syndrome risk by 28%, impor-
reduce CHD-related costs by up to $63 billion/ tant as metabolic syndrome is associated with
Table 11.2 Summary of Mediterranean diet (MedDiet) studies on CHD risk and biomarkers
Sofi et al. (2014). 35 cohort studies; 4,172,412 Each 2-point increase in adherence to the
Assess the associations of subjects MedDiet score reduced risk of CVD by 10%
MedDiet on CVD risk [33] (18 point scale)
Prospective studies
Bo et al. (2016). 1658 participants; low CVD risk Low CVD risk individuals per unit of
Investigate the associations of group; mean baseline age 52 years, MedDiet score significantly reduced all-cause
the MedDiet in a population 58% males; high CVD, mean mortality by 17%, CVD mortality by 25%,
with low CVD risk (Italy) [34] baseline age 58 years; 35% males; and CVD events by 21%. High CVD risk
mean 12-years of follow-up; 220 individuals per unit MedDiet score
CVD deaths (multivariate significantly reduced CVD events by 15%
adjusted) while the associations with all-cause and CVD
mortality were not significant (9 point scale)
Stewart et al. (2016). 15,482 patients with stable A one unit increase in MedDiet score was
Determine whether MedDiet coronary heart disease from 39 associated with 5% lower risk of major
score is associated with major countries; mean age 67 years; adverse cardiovascular events after adjusting
adverse cardiovascular events in MedDiet score was calculated for for all covariates (p = 0.002) whereas there
high-risk patients with stable increasing consumption of whole was no association between one unit change
coronary artery disease grains, fruits, vegetables, legumes, in the Western diet score and major adverse
(International) [35] fish, and alcohol, and for less cardiovascular events. This study suggests
meat, and a Western diet score for that an increased intake of healthy foods may
increasing consumption of refined be more important for secondary prevention
grains, sweets and deserts, sugared of CHD than avoidance of less healthy foods
drinks, and fried foods; median typical of Western diets (18 point scale)
3.7 years of follow-up
Panagiotakos et al. (2015). 3042 participants, baseline age MedDiet reduced CVD/CHD risk by 29%
Evaluate the additive protection 18–89 years; 50% men; 10-years (highest vs. lowest adherence), independent
of the MedDiet with statin of follow-up (multivariate of statin use. Patients with hyperlipidemia on
therapy (ATTICA Study; adjusted) a statin with low MedDiet adherence had a
Greece) [36] 75% increase in CVD/CHD risk compared to
normolipidemic subjects with high adherence
to MedDiet
Steffen et al. (2014). 5115 white and African Higher MedDiet scores were associated with
Investigate the effect of Americans; mean baseline age a significant 28% lower metabolic syndrome
MedDiet score on metabolic 25 years; 25 years of follow-up risk by improving risk factors including
syndrome risk (Coronary Artery (multivariate adjusted) abdominal obesity, TG levels, and HDL-C
Risk Development in Young levels. These findings suggest that the risk of
Adults (CARDIA) study; US) developing the metabolic syndrome is lower
[37] when consuming diets rich in fruits,
vegetables, whole grains, nuts and fish
Dilis et al. (2012). 23,929 adult men and women; Each 2-point increase in the MedDiet score
Examine the association of 20–86 years; median follow-up of was associated with lower CHD mortality
MedDiet adherence and CHD 10 years; 636 incident CHD cases risk in women by 25% and in men by 19%
risk (European Prospective and 240 CHD deaths were and reduced CHD risk in women by 15% and
Investigation into Cancer and recorded (multivariate adjusted) in men by 2% (9 point scale). Among men
Nutrition cohort [EPIC]; meat intake was positively associated and in
Greece) [38] women fruit and nut intake were protective
of CHD risk and mortality
(continued)
Guallar-Castillon et al. (2012). 40,757 adults; age 29–69 years; The MedDiet significantly lowered CHD risk
Compare relationship between median follow-up 11 years; 606 by 27% (highest vs. lowest adherence). No
the Western diet and MedDiet CHD events (multivariate association was found between the Western
on CHD risk (EPIC Spanish adjusted) diet and CHD risk
Cohort) [39]
Randomized controlled trials (RCTs)
Systematic review and meta-analyses
Liyanage et al. (2016). 6 RCTs; 10,950 subjects; 6 months This analysis showed that MedDiets lowered
Evaluate the effects of the to 9 years; 477 major vascular risk of major vascular events by 37%,
MedDiet on vascular disease events and 315 vascular deaths; coronary events by 35%, stroke by 35% and
and mortality [40] control diets include low-fat and heart failure by 70% but did not lower CVD
prudent Western diets mortality which was insignificantly lowered
by 10% compared to control diets. The
MedDiet may protect against vascular
disease but both the quantity and quality of
the available evidence is limited and highly
variable
Rees et al. (2013). 11 RCTs on the MedDiet pattern; MedDiets showed modest but significant
Systematically review the clinical 52,044 subjects; 5 trials with lowered TC by 6.2 mg/dL, LDL-C by
effects of the MedDiet pattern on healthy individuals, 6 trials with 2.7 mg/dL, TG by 18.6 mg/dL, and no
CHD prevention (Cochrane subjects having elevated CHD change in HDL-C compared to the control
Systematic Review) [41] risk; 3 months to 8 years diets
Kastorini et al. (2011). 50 studies included 35 RCTs, 2 High adherence to the MedDiet reduced
Examine the effects of MedDiet prospective and 13 cross-sectional metabolic syndrome risk by 31%. The
on metabolic syndrome and its studies; 534,906 participants MedDiet significantly reduced means for
components [42] waist circumference by 0.42 cm, TG by
6 mg/dL, systolic BP by 2.4 mmHg, diastolic
BP by 1.6 mmHg, and glucose by 4.0 mg/dL
and increased HDL-C by 1.2 mg/dL vs.
control diet
Bedard et al. (2015). Parallel RCT: Men on the MedDiet with CRP ≥2 mg/L had
Investigate effects of the 35 men and 27 premenopausal a significantly lowered CRP by 26.5% vs.
MedDiet vs. fast food on CRP women with moderate fast food diets but in women there was no
in men and women (Canada) hyperlipidemia; 24–53 years; significant difference vs. fast food
[43] 4 weeks
Sala-Vita et al. (2014). Parallel RCT: MedDiet plus nuts significantly reduced
Assess the effects of MedDiets 175 high CVD risk adults; mean mean ICA-IMT progression by 0.084 mm vs.
on internal carotid intima-media age 66 years; mean BMI 30 kg/m2; the low-fat control diet (Fig. 11.4). Similar
thickness (ICA-IMT) and 25% male; MedDiet supplemented results were observed for plaque height for
plaque height in subjects at high with extra virgin olive oil or mixed MedDiet plus nuts. The MedDiet plus extra
cardiovascular risk (Prevención tree nuts vs. low fat control; virgin olive oil ICA-IMT or plaque levels
con Dieta Mediterránea 2.4 years were not significantly different from the
[PREDIMED] sub-study; Spain) control diet
[44]
Estruch et al. (2013). Multi-center Parallel RCT: Compared to low-fat diets, the MedDiets
Study the effect of the MedDiet 7447 adults with high CVD; mean reduced risk of CVD/CHD with the extra
on primary CVD/CHD age 67 years; mean BMI 30; 57% virgin olive oil by 30% and with the mixed
prevention (PREDIMED; Spain) women; MedDiets with either tree nuts by 28%. This primary prevention
[45] extra virgin olive oil or mixed tree trial found that an energy-unrestricted
nuts vs. a low-fat diet; 4.8 years MedDiet resulted in a substantial reduction
in the risk of CVD/CHD events among
high-risk persons
Mitjavila et al. (2013). Parallel RCT: Urinary markers of levels of lipid
Investigate effects of the 110 women with metabolic peroxidation associated with increased
MedDiet on systemic lipid syndrome; mean age 69 years; atherosclerosis were reduced by high
oxidative biomarkers in women mean BMI 31; MedDiet adherence to the both MedDiets vs. control
with metabolic syndrome (Spain supplemented with extra virgin diet including lower urinary F2- isoprostane
-PREDIMED Sub-study) [46] olive oil or mixed tree nuts vs. low and 8-oxo-7,8- dihydro-20-deoxyguanosine
fat control; 1 year (DNA damage)
Murie-Fernandez et al. (2011). Parallel RCT: Among participants with baseline
Study the effect of the MedDiet 187 adults; mean age 67 years; C-IMT ≥ 0.9 mm, 1-year C-IMT changes for
on carotid intima-media 51% women; MedDiet both MedDiets were significantly lowered
thickness (C-IMT) supplemented with extra virgin compared to the low-fat control diet. No
(PREDIMED-Navarra; Spain) olive oil or mixed tree nuts vs. low C-IMT regression was shown among
[47] fat control; 1 year participants with lower baseline IMT values
<0.9 mm for any diet.
Thomazella et al. (2011). Parallel RCT: Both diets improved markers of redox
Investigate the effects of high 40 secondary prevention patients; homeostasis and metabolic effects potentially
adherence to MedDiet or low-fat mean age 55 years; MedDiet or related to atheroprotection and promoted
therapeutic lifestyle changes low-fat therapeutic lifestyle similar reduction in BMI and blood pressure.
diets on CHD risk (Brazil) [48] changes diets including plant Compared to the low-fat therapeutic lifestyle
stanols/sterols and viscous fiber; changes diets, the MedDiet promoted
3 months decreases in blood leukocyte count, increases
in HDL-C, and baseline brachial artery
diameter. Compared to the MedDiet, the
low-fat therapeutic lifestyle diets decreased
LDL-C and oxidized LDL-C plasma levels,
although the ratio of oxidized to total LDL-C
remained unaltered
Mena et al. (2009). Parallel RCT: Both MedDiet groups had significantly
Examine the effects of 112 adults with elevated CVD reduced serum interleukin-6 (IL-6) and
Med-Diets vs. a low-fat diet on risk; mean age 68 years; mean soluble intercellular adhesion molecule-1 vs.
inflammatory biomarkers related BMI 28 kg/m2; 43% women; the low-fat diet. MedDiet plus extra virgin
to atherogenesis in subjects MedDiet supplemented with extra olive oil significantly decreased soluble
(PREDIMED sub-study; Spain) virgin olive oil or mixed tree nuts vascular cellular adhesion molecule-1 and
[49] vs. low fat control; 3 months CRP vs. low-fat control diets
Fito et al. (2007). Parallel RCT: Both traditional MedDiets significantly
Investigate the effect of the 372 high CVD risk adults; mean decreased mean oxidized LDL, an
traditional MedDiet for effects age 69 years; 56% women; atherogenic risk factor, compared to the
on LDL-C oxidative stress MedDiet supplemented with extra low-fat diet group
markers (PREDIMED virgin olive oil or mixed tree nuts
Sub-study; Spain) [50] vs. low fat control; 3 months
Estruch et al. (2006). Parallel RCT: Both MedDiets significantly reduced plasma
Evaluate the short-term effects 772 high CVD risk adults; mean glucose levels by 5.4–7.0 mg/dL, systolic
of MedDiets on blood lipids, age 69 years; MedDiet blood pressure by 5.9–7.1 mmHg, and TC/
glycemic control and CRP supplemented with extra virgin HDL-C ratio by 0.26–0.38 compared with
levels (PREDIMED Sub-study; olive oil or mixed tree nuts vs. low the low-fat diet. The MedDiet plus nuts
Spain) [51] fat control; 3 months significantly decreased total cholesterol (TC)
by 6.2 mg/dL and triglycerides (TG)13 mg/
dL and the MedDiet plus extra virgin olive
oil significantly reduced CRP 0.54 mg/L
compared with the low-fat diet
(continued)
Esposito et al. (2004). Parallel RCT: The MedDiet significantly reduced blood
Investigate the long-term effects 180 Italian adults with metabolic lipids, TC, insulin resistance, hs CRP, and
of a MedDiet on syndrome; mean age 44 years; body weight and reduced metabolic
cardiometabolic health in 55% men; MedDiet vs. Western syndrome symptoms by 50% (Fig. 11.5)
subjects with metabolic type diet; 2 years
syndrome (Italy) [52]
Singh et al. (2002). Parallel RCT: The MedDiet significantly reduced total
Evaluate the effect of a MedDiet 1000 adults with angina pectoris, cardiac events vs. the NCEP Step 1 diet by
vs. National Cholesterol MI, or surrogate risk factors for 49%. The MedDiet significantly lowered TC
Education Program (NCEP) coronary artery disease; mean age by 9%, LDL-C by 14%, and TG by 14% and
Step 1 diet on coronary artery 49 years; a fiber-rich MedDiet or a increased HDL-C by 5.2% compared to the
disease risk [53] NCEP step I diet; 2 years NCEP Step 1 diet
de Lorgeril et al. (1999). Parallel RCT: In survivors of a first infarction, the MedDiet
Investigate the effects of the 423 overweight subjects after the significantly lowered myocardial infarction
MedDiet on recurrence rate first MI plus CVD medications; risk by 72% compared to those on the
after a first myocardial MedDiet vs. Western type diet; Western diet
infarction (Lyon Diet Heart 4 years; 275 events recorded
Study; France) [54]
significantly increased CHD risk [37]. The Greek 8 years) concluded that MedDiets have a moder-
cohort of the European Prospective Investigation ate but significant effect on lowering blood lipids
into Cancer and Nutrition (EPIC) study (23,929 and lipoproteins for beneficial effects on lowering
participants; median follow-up of 10 years) CHD risk [41]. A meta-analysis (35 RCTs, 2 pro-
showed that a 2 point increase in MedDiet score spective and 13 cross-sectional studies; 534,906
(0–9 scale) significantly reduced CHD mortality participants) found that greater adherence to the
risk by 25% in women and 19% in men and CHD MedDiet was associated with reduced metabolic
incidence in women by 15% but not in men, syndrome risk by 31% including beneficial effects
which had an insignificant 2% reduction [38]. A on all risk factors including waist circumference,
Spanish EPIC study (40,757 participants; base- TG, blood pressure, blood glucose and HDL-C
line age range 29–69 years; median follow-up of [42]. A 2013 Prevención con Dieta Mediterránea
11 years) found that the highest quintile of the (PREDIMED) RCT (7447 adults with elevated
MedDiet score reduced CHD risk by 27% com- CVD risk; mean age 67 years; mean BMI 30 kg/
pared to the lowest score [39]. m2; 57% women; MedDiets with either extra vir-
gin olive oil or mixed tree nuts vs. control low fat
11.2.2.2 Randomized Controlled diets; 4.8 years) found that both MedDiets reduced
Trials (RCTs) CVD risk by about 30% compared to low-fat diets
RCTs consistently show that MedDiets signifi- [45]. These findings were consistent with a 2006
cantly lower CHD risk and biomarkers compared PREDIMED trial (772 high CVD risk adults;
to low fat or Western diets [40–54]. A 2016 sys- mean age 69 years; MedDiet supplemented with
tematic review and meta-analysis of the effect of extra virgin olive oil or mixed tree nuts vs. low fat
MedDiets on vascular disease and mortality (6 control; 3 months) which showed that the
RCTs; 10,950 subjects; 6 months to 9 years) MedDiets significantly reduced TC, TG, TC/
showed that the MedDiet lowered risk of major HDL-C ratio, glucose levels, CRP, and systolic
vascular events by 37%, coronary events by 35%, blood pressure compared with the low-fat diet
stroke by 35% and heart failure by 70% but not [51]. A Brazilian secondary prevention RCT (40
CVD mortality which was insignificantly lowered medicated patients; mean age 55 years; MedDiet
by 10% compared to control diets including low or low-fat therapeutic lifestyle changes diet
fat and Western diets [40]. A Cochrane systematic including plant stanols/sterols and viscous fiber;
review (11 RCTs; 52,044 subjects; 3 months to 3 months) found that the MedDiet and therapeutic
lifestyle changes diet promoted similar significant CHD because of its beneficial role on systemic
decreases in BMI and blood pressure with cardiovascular risk factors and its possible effect
decreases in blood leukocyte count and increases on reducing the risk of body weight gain, meta-
in HDL-C and brachial artery diameter and the bolic syndrome, and obesity [29, 55, 56].
low-fat therapeutic lifestyle changes decreased
LDL-C and oxidized LD-C plasma levels [48].
Both diets played a role in secondary prevention 11.2.3 Dietary Approaches to Stop
by improving markers of redox homeostasis and Hypertension (DASH) Dietary
metabolic effects potentially related to atheropro- Pattern
tection. Several satellite PREDIMED RCTs sup-
port a potential role in promoting regression of The DASH diet was designed for hypertension
carotid intima-media thickness and plaque with control but it also has effects on reducing several
high adherence to the MedDiet supplemented CHD risk factors, including TC, LDL-C, and
with nuts, especially in individuals with elevated inflammation [15, 19]. This pattern is rich in
plaque (Fig. 11.4) [44, 47]. An Italian RCT (180 fruits, vegetables, and low-fat dairy products,
Italian adults with metabolic syndrome; mean age includes whole grains, poultry, fish, and nuts, and
44 years; 55% men; MedDiet vs. Western diet; limits saturated fat, red meat, sweets, and sugar
2 years) demonstrated that the MedDiet signifi- containing beverages. Compared with the
cantly reduced blood lipids, TC, insulin resis- Western diet, the DASH diet provides lower total
tance, hs-CRP and body weight, and reduced fat, saturated fat, and dietary cholesterol, and
metabolic syndrome risk factors by 50% com- higher potassium, magnesium, calcium, fiber,
pared to Western diets (Fig. 11.5) [52]. Also, a and protein. Table 11.3 summarizes the DASH
Canadian trial showed that the MedDiet was diet prospective studies and RCTs on the on CHD
shown to reduce elevated CRP levels in men by risk and biomarkers [57–67].
26.5% compared to fast food based diets [43].
Several other PREDIMED trials demonstrate that 11.2.3.1 Prospective Cohort Studies
MedDiets lower systemic inflammatory markers Prospective studies consistently show that the
and urinary markers of DNA damage, and lipid DASH dietary pattern helps to reduce CHD and
oxidation compared to low-fat diets [46, 49, 50]. heart failure risk [57–61]. A 2013 meta-analysis (6
The MedDiet protects against the development of studies; 260,000 adults) reports that higher
Internal carotid artery initima media thickness (mm) Plaque height (mm)
0.2
0.15
0.1
Change (mm)
0.05
0
MedDiet + Nuts MedDiet + Virgin Olive Low Fat Diet
Oil
–0.05
–0.1
–0.15
Fig. 11.4 Association between MedDiets and low-fat diets and internal carotid artery mean intima-media thickness
and plaque height over 2.4 years (p = 0.047) (adapted from [44])
5.0
MedDiet vs Western Diet 0.0
–5.0
–10.0
–15.0
–20.0
TC HDL-C TG Insulin HOMA hs-CRP Weight
(mg/dL) (mg/dL) (mg/dL) (uU/ml) score (mg/L) (kg)
Change –9.0 3.0 –19 –3.5 –1.1 –1.0 –2.8
Fig. 11.5 Effect of MedDiet vs. Western diet on cardiometabolic health biomarkers in subjects with metabolic syn-
drome after 2 years (all p = 0.01 to <0.001) (adapted from [52]). *TC total cholesterol, HDL-C high density cholesterol,
TG triglycerides, HOMA measure of insulin resistance, hs-CRP high sensitivity C-reactive protein
adherence to the DASH diet significantly reduced clinical changes reduced the 10-year Framingham
risk of CHD by 21% and heart failure by 29% risk score for CHD risk by 13% [62]. Four other
[57]. The EPIC Dutch cohort (33,671participants; RCTs found that higher adherence to a standard
mean age 49 years at baseline; 74% women; aver- DASH diet significantly reduces the 10-year
age follow-up of 12.2 years) observed that high CHD risk by 12–18% [63–65, 68]. Also, the
adherence to the DASH diet significantly reduced Omni Trial (164 subjects; mean age 54 years;
CHD risk by 18% compared to low adherence 73% women; 6 weeks) found that the level of
[58]. The Nurses’ Health Study (88,517 women; protein or monounsaturated fatty acids (MUFA)
aged 34–59 years; 24 year of follow-up) found a substituted for carbohydrate improves the 10 year
significantly reduced total CHD risk by 24% and CHD risk level (Fig. 11.6) [67]. For CRP, a
fatal CHD risk by 29% for high vs. low DASH diet DASH diet with 12 g added fiber from whole
adherence [61]. For heart failure, higher adherence plant foods or psyllium significantly reduced
to the DASH diet significantly reduced heart CRP by 14 and 18%, respectively, compared to
failure risk in women by 16% and in men by 22% control low fiber diets [65]. In the Omni Trial a
[59, 60]. modification of the standard DASH diet made by
replacing 10% of the energy from carbohydrate
11.2.3.2 Randomized Controlled with protein decreased LDL-C by 3 mg/dL,
Trials HDL-C by 1 mg/dL, and TG by 16 mg/dL or
DASH diet RCTs consistently support improved replacing 10% of energy with unsaturated fat
biomarkers and lower CHD risk versus control (8% MUFA and 2% PUFA) decreased LDL-C by
diets [62–68]. A 2015 systematic review and 3 mg/dL, increased HDL-C by 1 mg/dL and
meta-analysis (20 RCTs; 1917 participants; decreased TG by 10 mg/dL [67]. A report from
2–24 weeks) showed that the DASH diet signifi- the American College of Cardiology/American
cantly lowered systolic BP by 25 mmHg and dia- Heart Association Guide line on Lifestyle
stolic BP by 23 mmHg, TC by 8 mg/dL, and Management to Reduce C ardiovascular Risk
LDL-C by 8 mg/dL vs. control diets; these estimated that adults (TC level < 260 mg/dL,
Table 11.3 Summary of DASH diet studies on CHD risk and related biomarkers
Salehi-Abargouei et al. 6 studies; 260,000 subjects; The DASH-like diet significantly reduced risk
(2013). follow-up of 7–24 years of CVDs by 20%, CHD by 21%, and heart
Investigate the effects of a failure by 29%
DASH-style diet on the
incidence of CHD [57]
Struijk et al. (2014). 33,671 participants; mean baseline Higher adherence to the DASH diet was
Evaluate the effects of DASH age 49 years; 74% women; average associated with significantly reduced CHD
diet on CHD risk (EPIC follow-up of 12.2 years; 1630 risk by 9%
Netherlands cohort) [58] CHD cases (multivariate adjusted)
Levitan et al. (2013). This study included 68,132 Higher DASH diet scores were associated
Evaluate the effect of DASH women. After a median follow-up with significantly lower heart failure
diets on heart failure mortality of 4.6 years, 1385 women died of mortality in women by 16%
in women (Women’s Health heart failure (multivariate adjusted)
Initiative; US) [59]
Levitan et al. (2009). 38,987 men; baseline age 45 to Men with the highest DASH diet adherence
Assess the effect of DASH 79 years; 9 years of follow-up; 710 had a significant 22% lower heart failure
diets on heart failure events in were hospitalized for HF and 97 event risk vs. those with the lowest adherence
men (Sweden) [60] died (multivariate adjusted)
Fung et al. (2008). 88,517 women; baseline age 34 to Women with the highest DASH diet
Examine the effect of the 59 years; 24 years of follow-up; adherence significantly reduced CHD risk by
DASH diet on CHD risk in 976 CHD deaths (multivariate 24% vs. lowest adherence. Similar risk
women (Nurses’ Health Study; adjusted) reductions were observed for nonfatal
US) [61] myocardial infarction and CHD deaths
Meta-analysis
Siervo et al. (2015). 20 RCTs; 1917 participants; The DASH diet significantly reduced systolic
Assess the effects of the DASH 2–24 weeks BP by 25 mmHg and diastolic BP by
diet on cardiovascular risk 23 mmHg, TC by 8 mg/dL and LDL-C by
factors [62] 8 mg/dL vs. control diets. These changes
predicted a reduced 10-year Framingham risk
score for CHD by 13%
Jenkins et al. (2017). Parallel RCT: Overall, at 6 months, similar reductions in all
Assess the effect of dietary 209 men and 710 women; mean groups were found for body weight by
advice and/or food provision age 45 years; mean BMI 32; 3 0.8–1.2 kg, waist size by 1.1 to 1.9 cm, and
for the DASH diet on body diets: (1) subjects encouraged to mean arterial pressure by 0.0–1.1 mmHg.
weight and cardiovascular consume a DASH type diet plus After 18 months, all diets maintained the
disease risk factors (Canada) increased consumption of reductions in body weight, BMI and waist
[63] cholesterol-lowering functional size that were seen at 6 months. Also, HDL-C
foods including soy foods, nuts, rose between 6 and 18 months by
and viscous fiber sources such as 0.05 mmol/L (p < 0.0001) and the total-to-
oats and barley (control), (2) HDL-cholesterol ratio was reduced. Further,
subjects received a weekly food significant reductions in mean arterial
basket reflecting the advice given pressure and Framingham CHD risk scores
to the first treatment group but did were seen at 18 months in all groups
not receive dietary advice, and (3)
subjects received both the weekly
food basket and dietary advice;
6-month intervention; 12-month
follow-up
(continued)
Chen et al. (2010). Parallel RCT: The DASH diet significantly lowered 10-year
Investigate the effects of the 459 prehypertension or stage-1 CHD risk by 18% compared to the low-fat
DASH diet on the 10-year risk hypertension subjects not taking control diet. The high F/V diet insignificantly
of developing CHD (US) [64] anti-hypertensive medication; lowered risk by 7%. Compared with F/V diet,
mean age 45 years; 51% men and the DASH diet significantly reduced
60% African-American; duration estimated 10-year CHD risk by 11%
8 weeks) evaluated 3 diets:
Increased fruits and vegetables
(F/V), DASH diet, and low fat diet;
8 weeks
Maruthur et al. (2009). Parallel RCT: Active healthy lifestyle plus DASH education
Examine the effects of the 810 healthy adults with untreated significantly lowered estimated 10-year CHD
DASH diet on CHD risk pre- or stage I hypertension risk by 14% relative to general advice
(PREMIER Trial; US) [65] randomized; mean age 50 years;
61% women; active education on
healthy lifestyle recommendations,
or active healthy lifestyle plus
DASH education vs. general
advice; 6 months
King et al. (2007). Crossover RCT: Compared to baseline, CRP levels were
Investigate the effect of DASH 35 adults; 18 lean normotensives significantly reduced in the fiber rich foods
diet or fiber supplemented and 17 obese hypertensive DASH diet group by 14% and in the psyllium
DASH diet on CRP (US) [66] individuals; age range fiber-supplemented DASH-type diet group by
18–49 years; 80% women; 18%. There were no differences in weight,
evaluated 2 diets vs. baseline: TG, TC, or insulin resistance status between
High-fiber foods DASH type diet the two DASH diets
(28 g fiber/day) or psyllium
fiber-supplemented DASH type
diet (26 g fiber/day); mean
baseline fiber intake 12 g/day;
3 weeks
Apple et al. (2005). Crossover RCT: Compared with the standard DASH
Evaluate the effects of 3 164 adults with prehypertension or (carbohydrate) diet, the estimated 10-year
variations of the DASH diet on stage 1 hypertension; mean age CHD risk was reduced for the DASH diets
CHD risk (US) [67] 54 years; 73% women; 10% enriched with protein by 5.8% and
energy replacement of standard unsaturated fat (primarily MUFA) by 4.2%
DASH diet carbohydrate with (Fig. 11.6)
protein or monounsaturated fatty
acids (MUFA) vs. standard DASH
diet evaluate; 6 weeks
Obarzanek et al. (2001). Parallel RCT: The 10-year CHD risk was significantly
Investigate the effects of DASH 436 adults; mean age 45 years; decreased for the DASH diet by 12.1%
diet on plasma lipids, focusing 60% African American; baseline compared with a 0.9% increase in risk for the
on subgroups by sex and race TC was ≤260 mg/dL; evaluated 3 Western control diet
(US) [68] diets: Control Western diet, a diet
increased in fruit and vegetables,
or a DASH diet, during which time
subjects remained weight stable;
8 weeks
DASH Diet Extra Protein DASH Diet Extra MUFA

0
Men Women
–2
–4
CHD Risk (%)
–6
–8
–10
–12
–14
Fig. 11.6 Estimated 10-year risk of Framingham coronary heart disease (CHD) with a 10% energy replacement of
standard DASH diet carbohydrate with protein or monounsaturated fatty acids (MUFA) vs. standard DASH diet
(adapted from [67])
LDL-C level < 160 mg/dL and a stable body vegetarian diets significantly reduced ischemic
weight) with high adherence to the basic DASH heart diease mortality by 29% and circulatory dis-
diet would be projected to lower LDL-C by ease risk by 16% compared to non-vegetarian
11 mg/dL, reduce TC:HDL-C ratio, and have no diets [70]. A meta-analysis of RCTs (11 trials;
effect on TG compared with a typical American mean 24 weeks, ranging from 3 weeks to
diet of the 1990s [69]. A 2017 Canadian RCT 18 months) demonstrated that vegetarian diets
(919 overweight men and women; mean age significantly lowered TC by 13.9 mg/dL, LDL-C
45 years; mean BMI 32; 3 groups: (1) encour- by 13.1 mg/dL, and HDL-C by 3.9 mg/dL, but did
aged to consume a DASH type diet plus increase not significantly affect blood TG concentrations
consumption of cholesterol-lowering functional compared to control non-vegetarian diets [71].
foods including soy foods, nuts, and viscous fiber
sources such as oats and barley (control),
(2) weekly food basket reflecting the advice 11.2.5 Elderly Dietary Index
given to the first treatment group but did not
receive dietary advice, and (3) both the weekly The Elderly Dietary Index was constructed using
food basket plus dietary advice; 6-month inter- ten components (ie, questions about the consump-
vention; 12-month follow-up) all groups showed tion frequency of meat, fish, fruits, vegetables,
small significant reductions in body weight by grains, legumes, olive oil, and alcohol as well as
0.8–1.2 kg, waist size by 1.1–1.9 cm, and mean the type of bread and dairy products) according to
arterial pressure by 0.0–1.1 mmHg at 6 months the Modified MyPyramid for older adults and
and Framingham CHD risk scores were signifi- select features of the traditional Mediterranean
cantly lower at 18 months regardless of food diet [72]. Scores from 1 to 4 were assigned to
delivery or guidance approach [63]. each of the ten components of the index with total
Elderly Dietary Index score ranging between 10
and 40. The British Regional Heart Study (3328
11.2.4 Vegetarian Dietary Patterns men; baseline mean age 68 years; 11.3 years of
follow-up; 933 deaths from all causes, 327 CVD
Vegetarian diets reduce CHD risk compared to deaths, 582 CVD events, and 307 CHD events)
non-vegetarian diets [69]. A meta-analysis (7 found a that higher adherence to the Elderly
cohort studies; 124,706 adults) showed that Dietary Index significantly lowered risk of CVD
CHD Events (p-trend =.05) CVD Mortality (p-trend =.03)

1.1
0.9
Hazard Ratios
0.8
0.7
0.6
0.5
0.4
1 2 3 4
Elderly Dietary Index Quartile
Fig. 11.7 Adherence to Elderly Dietary Index on coronary heart disease (CHD) events and cardiovascular disease
(CVD) mortality in men mean baseline age 70 years over 11.3 years of follow-up from the British Regional Heart
Study (multivariate adjusted) (adapted from [73])
mortality by 37% and CHD events by 33%, which with MUFA was associated with a decreased
was independent of sociodemographic, behav- CHD risk. The Alpha-Tocopherol, Beta-
ioral, and cardiovascular risk factors, compared to Carotene Cancer Prevention Study, double-
lower adherence (Fig. 11.7) [73]. blind RCT (21,955 men, baseline age
50–69 years; 19-years of follow-up; 4379 CHD
cases) found that replacing saturated and trans-
11.3 D
ietary Carbohydrates vs. fatty acids with carbohydrates was associated
Dietary Fat with a 3% lowered CHD risk for each 2%
energy intake and replacing MUFA and PUFA
There have been a number of prospective stud- with carbohydrates was associated with an 8%
ies and RCTs on the effects of replacing dietary and a 4.5% increased CHD risk for each 2% of
carbohydrate for fat intake or vice versa on energy intake, respectively (Fig. 11.8) [77]. A
CHD risk [74–78]. A meta-analysis of 8 pro- 2015 pooled analysis of the Nurses’ Health
spective cohort studies concluded that a high Study and Health Professionals Follow-up
dietary glycemic index and glycemic load Study (84,628 women and 42,908 men; mean
increased the risk of CHD in women but not in baseline age 47–54 years; 24–30 years of fol-
men [74]. In contrast, a later cohort study found low-up; 7667 CHD cases) found that replacing
that a higher carbohydrate intake was associ- 5% of energy intake from saturated fats with
ated with an increased CHD risk in men but not equivalent energy intake from either PUFAs,
in women [75]. A pooled analysis (11 cohort MUFAs, or carbohydrates from whole grains
studies) showed that replacing saturated fat was associated with significant 25%, 15%, and
with carbohydrates increased the risk of coro- 9% lower risks of CHD, respectively (Fig. 11.9)
nary events but not of coronary deaths [76]. [78]. However, replacing saturated fat with car-
Also, replacing carbohydrates with trans-fatty bohydrates from refined starches/added sugars
acids was associated with an increased CHD was not significantly associated with lower
risk and replacing carbohydrates with PUFA or CHD risk (p > 0.10).
11.4 Healthful vs. Unhealthful Plant Based Diets 331
Carbohydrates for PUFA
Carbohydrates for MUFA
Carbohydrates for SFA + TFA
–6 –4 –2 0 2 4 6 8 10
Change in CHD Risk (%)
Fig. 11.8 Change in coronary heart disease (CHD) risk with the isocaloric substitution of 2% of carbohydrate energy
for saturated (SFA) and trans-fat (TFA), monounsaturated fat (MUFA), or polyunsaturated fat (PUFA) in men (adapted
from [77])
Whole Grains for SFA
PUFA for SFA
MUFA for SFA
–30 –25 –20 –15 –10 –5 0 5

Change in CHD Risk (%)
Fig. 11.9 Change in coronary heart disease (CHD) risk with the isocaloric substitution of 5% of energy from monoun-
saturated fat (MUFA), polyunsaturated fat (PUFA) and whole grains for saturated fat (SFA) in men and women from
pooled data from the Nurses’ Health Study and Health Professionals Follow-up Study; MUFA and PUFA (p = 0.02) and
whole grains (p = 0.01) (adapted from [78])
11.4 H
ealthful vs. Unhealthful Professionals Follow-up Study (1986 to 2012),
Plant Based Diets without CHD at baseline. An overall plant-based
diet index (PDI) from repeated semiquantitative
Plant-based diets are recommended for CHD food-frequency questionnaire data, by assigning
prevention and management but not all plant positive scores to plant foods and reverse scores
foods are necessarily beneficial for CHD risk to animal foods. A healthy plant-based diet index
reduction [79]. The association of healthy and (hPDI) consisted of whole grains, fruits/vegeta-
less healthy plant based diets were examined in bles, nuts/legumes, oils, tea and coffee, whereas
73,710 women in the Nurses’ Health Study (1984 unhealthy plant foods (uPDI) included sweetened
to 2012), 92,329 women in Nurses’ Health Study juices/beverages, refined grains, French fries,
2 (1991 to 2013), and 43,259 men in Health sweets and some animal foods. Higher adherence
to an overall PDI reduced CHD by 8% (p-trend = processed meat, and lower intakes of refined
.003). This inverse association was greater for grains, sugar-sweetened foods and beverages
hPDI reducing CHD risk by 25% whereas an compared to an increased CHD risk associated
uPDI was associated with an increased CHD risk with higher adherence to Western dietary pat-
by 32% (p-trend <.001 for both). Health profes- terns. Healthy dietary patterns have CHD pro-
sionals should guide their CHD patients to tective effects because they are lower in energy
increase intake of healthy plant foods and reduce density and higher in fiber, healthier fatty acid
intake of less healthy plant foods, sugar sweet- profiles, essential nutrients, antioxidants, and
ened beverages and certain high saturated fat and electrolytes, and are anti-inflammatory com-
energy animal foods. pared with Western diets. Meta-analyses esti-
mate that healthy dietary patterns are associated
Conclusions with a decreased CHD risk by 20–33%, while
Globally, CHD is the leading cause of death Western-type patterns are associated with an
and morbidity in adults. Healthy lifestyles increased CHD risk by up to 45%, especially in
(e.g., BMI <25, physical activity at goal levels, US studies and in individuals over 50 years of
a diet consistent with the dietary guidelines, age. A number of RCTs support the role for
and smoking avoidance) play an important role healthy dietary patterns in reducing CHD risk
in preventing and managing CHD risk. biomarkers including blood lipids and lipopro-
However, less than 1% of Americans meet all teins, systemic inflammatory or oxidative
four ideal criteria for cardiovascular health. stress factors such as hs-CRP and oxidized
The type of dietary pattern consumed plays an LDL- C, and carotid atherosclerosis and
important role in the risk of developing improving endothelial health and blood pres-
CHD. Healthy dietary patterns (or higher nutri- sure. Replacing 5% of energy intake from satu-
ent quality diets) which are associated with rated fats with equivalent energy intake from
decreased risk of CHD include higher Healthy PUFAs, MUFAs, or carbohydrates from whole
Eating Indices scores, the Mediterranean diet grains was associated with a significant 25%,
(MedDiet), Dietary Approaches to Stop 15%, and 9% lower risk of CHD, respectively.
Hypertension (DASH) and vegetarian diets and However, replacing saturated fat with carbohy-
are characterized by higher consumption of drates from refined starches and/or added sug-
vegetables, fruits, whole-grains, low-fat dairy, ars was not significantly associated with lower
and seafood, limited (or no) intake of red and CHD risk.
ppendix A: Comparison of Western and Healthy Dietary Patterns per

A
2000 kcal (Approximated Values)
Healthy
Healthy vegetarian pattern
Western dietary USDA base DASH diet Mediterranean (Lacto-ovo
Diet type pattern (US) pattern pattern pattern based) Vegan pattern
Emphasizes Refined grains, Vegetables, Potassium Whole grains, Vegetables, fruit, Plant foods:
low fiber foods, fruit, rich vegetables, whole-grains, Vegetables,
red meats, whole-grains, vegetables, fruit, dairy legumes, nuts, fruits, whole
sweets, solid fats and low-fat fruits, and low products, olive seeds, milk grains, nuts,
milk fat milk oil, and products, and soy seeds, and
products moderate wine foods soy foods
Includes Processed meats, Enriched Whole-grains, Fish, nuts, Eggs, non-dairy Non-dairy
sugar sweetened grains, lean poultry, fish, seeds, and milk alternatives, milk
beverages, and meat, fish, nuts, and pulses and vegetable oils alternatives
fast foods nuts, and seeds
seeds,
vegetable oils
Appendix A 333
Healthy
Healthy vegetarian pattern
Western dietary USDA base DASH diet Mediterranean (Lacto-ovo
Diet type pattern (US) pattern pattern pattern based) Vegan pattern
Limits Fruits and Solid fats, Red meats, Red meats, No red or white No animal
vegetables, and and added sweets and refined grains, meats, or fish; products
whole-grains sugars sugar- and sweets limited and
sweetened sweets
beverages
(% Total kcal)
Protein (% 16 17 18 16 14 13
Total kcal)
Total fat (% 33 32 27 34 32 30
Total kcal)
(% Total kcal)
Unsat. fat (% 22 25 21 24 26 25
Total kcal)
Fiber (g) 16 31 29+ 31 35+ 40+
Potassium 2800 3350 4400 3350 3300 3650
(mg)
Vegetable oils 19 27 25 27 19–27 18–27
(g)
Sodium (mg) 3600 1790 1100 1690 1400 1225
Added sugar 79 (20 tsp) 32 (8 tsp) 12 (3 tsp) 32 (8 tsp) 32 (8 tsp) 32 (8 tsp)
(g)
Plant food groups
Fruit (cup) ≤1.0 2.0 2.5 2.5 2.0 2.0
Vegetables ≤1.5 2.5 2.1 2.5 2.5 2.5
(cup)
Whole-grains 0.6 3.0 4.0 3.0 3.0 3.0
(oz.)
Legumes (oz.) N/A 1.5 0.5 1.5 3.0 3.0+
Nuts/seeds 0.5 0.6 1.0 0.6 1.0 2.0
(oz.)
Soy products 0.0 0.5 − − 1.1 1.5
(oz.)
terns. 2015; 1–9
gov/ Accessed 22 Aug 2015
References http://www.choosemyplate.gov/. Accessed 22 Aug

2015.
17. U.S. Department of Agriculture, Agriculture Research
1. Mendis S, Puska P, Norrving B. Global atlas on car-
Service, Nutrient Data Laboratory. USDA National
diovascular disease prevention and control. Geneva:
Nutrient Database for Standard Reference, Release 27.
World Health Organization; 2011.
2014. http://www.ars.usda.gov/nutrientdata. Accessed
2. Sayols-Baixeras S, Lluís-Ganella C, Lucas G,
17 Feb 2015.
Elosua R. Pathogenesis of coronary artery disease:
18. Dietary Guidelines Advisory Committee. Scientific
focus on genetic risk factors and identification
Report. Advisory Report to the Secretary of Health
of genetic variants. Appl Clin Genet. 2014;7:
and Human Services and the Secretary of Agriculture.
15–32.
Appendix E-3.7: Developing vegetarian and Mediter
3. Roger VL, Go AS, Lloyd-Jones DM, et al. Heart
ranean-style food patterns; 2015. p. 1–9.
disease and stroke statistics-2011 update: a report
19. Dietary Guidelines Advisory Committee (DGAC).

from the American Heart Association. Circulation.
Scientific Report. Advisory Report to the Secretary
2011;123(4):18–209.
of Health and Human Services and the Secretary of
4. Barquera S, Pedroz-Tobias A, Medina C, et al. Global
Agriculture. Part D. Chapter 2: Dietary patterns, foods
overview of the epidemiology of atherosclerosis car-
and nutrients and health outcomes; 2015. p. 1–35.
diovascular disease. Arch Med Res. 2015;46(5):328–
20. Mente A, de Koning L, Shannon HS, Anand SS. A
38. https://doi.org/10.1016/jarcmed.2015.06.006.
systematic review of the evidence supporting a causal
5. Lloyd-Jones DM, Larson MG, Beiser A, Levy link between dietary factors and coronary heart dis-
D. Lifetime risk of developing coronary heart disease. ease. Arch Intern Med. 2009;169(7):659–69.
Lancet. 1999;353:89–92. 21. Zhang X-Y, Shu L, Si C-J, et al. Dietary patterns, alco-
6. Mozaffarian D, Benjamin EJ, Go AS, et al. Heart hol consumption and risk of coronary heart disease in
disease and stroke statistics-2015 update a report adults: a meta-analysis. Forum Nutr. 2015;7:6582–
from the American Heart Association. Circula 605. https://doi.org/10.3390/nu7085300.
tion. 2015;131:e29–322. https://doi.org/10.1161/ 22. Hou L, Li F, Wang Y, et al. Association between
CIR.0000000000000152. dietary patterns and coronary heart disease: a meta-
7. Fleg JL, Forman DE, Berra K, et al. Secondary preven- analysis of prospective cohort studies. Int J Clin Exp
tion of atherosclerotic cardiovascular disease in older Med. 2015;8(1):781–90.
adults a scientific statement from the American Heart 23. Shikany JM, Safford MM, Newby PK, et al. Southern
Association. Circulation. 2013;128(22):2422–46. dietary pattern is associated with hazard of acute
8. Hu FB. Diet and lifestyle influences on risk of coronary coronary heart disease in the Reasons for Geographic
heart disease. Curr Atheroscler Rep. 2009;11:257–63. and Racial Differences in Stroke (REGARDS)
9. Gage B, Canonico M, Perier M-C, et al. Idea car- Study. Circulation. 2015;132:804–14. https://doi.
diovascular health, mortality and vascular events in org/10.1161/CIRCULATIONAHA.115.020671.
elderly subjects. J Am Coll Cardiol. 2017;69:3015–26. 24. Chiuve SE, Fung TT, Rimm EB, et al. Alternative
10. Lichtenstein AH, Appel LJ, Brands M, et al. Diet and dietary indices both strongly predict risk of chronic
lifestyle recommendations revision 2006: a scien- disease. J Nutr. 2012;142:1009–18.
tific statement from the American Heart Association 25. Iqbal R, Anand S, Ounpuu S, et al. Dietary patterns
Nutrition Committee. Circulation. 2006;114:82–96. and the risk of acute myocardial infarction in 52 coun-
11. Vermorken AJM, Andres E, Cui Y. Bowel movement tries results of the INTERHEART Study. Circulation.
frequency, oxidative stress and disease prevention 2008;118:1929–37.
(Review). Mol Clin Oncol. 2016;5:339–42. https:// 26. Hu FB, Rimm EB, Stampfer MJ, et al. Prospective
doi.org/10.3892/mco.2016.987. study of major dietary patterns and risk of coro-
12. Forget G, Doyon M, Lacerte G, et al. Adoption of nary heart disease in men. Am J Clin Nutr. 2000;72:
American Heart Association 2020 Ideal Healthy Diet 912–21.
Recommendations prevents weight gain in young 27. Nettleton JA, Steffen LM, Mayer-Davis EJ, et al.

adults. J Acad Nutr Diet. 2013;113:1517–22. Dietary patterns are associated with biochemical
13. http://health.gov/dietaryguidelines/2015/guidelines/ markers of inflammation and endothelial activation in
message/. Accessed 17 Jan 2016. the Multi-Ethnic Study of Atherosclerosis (MESA).
14. Tuso P, Stoll SR, Li WW. A plant-based diet, athero- Am J Clin Nutr. 2006;83(6):1369–79.
genesis, and coronary artery disease prevention. Perm 28. Reidlinger DP, Darzi J, Hall WL, et al. How effective
J. 2015;19(1):62–7. are current dietary guidelines for cardiovascular dis-
15. Bhupathiraju SN, Tucker KL. Coronary heart disease ease prevention in healthy middle-aged and older men
prevention: nutrients, foods, and dietary patterns. Clin and women? A randomized controlled trial. Am J Clin
Chim Acta. 2011;412:1493–514. Nutr. 2015;101:922–30.
16. U.S. Department of Agriculture and U.S. Department 29.
Ros E, Martínez-González MA, Estruch R.
of Health and Human Services. Dietary guidelines Mediterranean diet and cardiovascular health: teach-
for Americans, 2010. 7th ed. Washington, DC: ings of the PREDIMED study. Adv Nutr. 2014;5:
U.S. Government Printing Office; 2010. Table B2.4; 330S–6S.
References 335
30. de Lorgeril M. Mediterranean diet and cardiovascu- and its components: a meta-analysis of 50 stud-
lar disease: historical perspective and latest evidence. ies and 534,906 individuals. J Am Coll Cardiol.
Curr Atheroscler Rep. 2013;15(12):370. https://doi. 2011;57:1299–313.
org/10.1007/s11883-013-0370-4. 43. Bédard A, Lamarche B, Corneau L, et al. Sex differ-
31. Martínez-Gonzáleza MA, Salas-Salvadó J, Estruch
ences in the impact of the Mediterranean diet on sys-
R, et al. Benefits of the Mediterranean diet: insights temic inflammation. Nutr J. 2015;14:46. https://doi.
from the PREDIMED study. Prog Cardiovasc Dis. org/10.1186/s12937-015-0035-y.
2015;58(1):50–60. https://doi.org/10.1016/ j.pcad. 44. Sala-Vila A, Romero-Mamani ES, Gilabert R, et al.
2015.04.003. Changes in ultrasound-assessed carotid intima-media
32. Abdullah MMH, Jones JPH, Jones PJH. Economic thickness and plaque with a Mediterranean diet.
benefits of the Mediterranean-style diet consump- A sub-study of the PREDIMED trial. Arterioscler
tion in Canada and the United States. Food Nutr Thromb Vasc Biol. 2014;34:439–45.
Res 2015;59: 27541. https://doi.org/10.3402/fnr.v59. 45. Estruch R, Ros E, Salas-Salvado J, et al. Primary pre-
27541. vention of cardiovascular disease with a Mediterranean
33. Sofi F, Macchi C, Abbate R, et al. Mediterranean diet diet. N Engl J Med. 2013;368:1279–90.
and health status: an updated meta-analysis and a pro- 46. Mitjavila MT, Fandos M, Salas-Salvadó J, et al. The
posal for a literature-based adherence score. Public Mediterranean diet improves the systemic lipid and
Health Nutr. 2014;17(12):2769–82. DNA oxidative damage in metabolic syndrome indi-
34. Bo S, Ponzo V, Goitre I, et al. Predictive role of the viduals. A randomized, controlled, trial. Clin Nutr.
Mediterranean diet on mortality in individuals at low 2013;32:172–8.
cardiovascular risk: a 12-year follow-up population- 47. Murie-Fernández M, Irimia P, Toledo E, et al. Carotid
based cohort study. J Transl Med. 2016;14:91. https:// intima-media thickness changes with Mediterranean
doi.org/10.1186/s12967-016-0851-7. diet: a randomized trial (PREDIMED-Navarra).
35. Stewart RA, Wallentin L, Benatar J, et al. Dietary Atherosclerosis. 2011;219:158–62.
patterns and the risk of major adverse cardiovascular 48. Thomazella MCD, Góes MFS, Andrade CR, et al.
events in a global study of high-risk patients with sta- Effects of high adherence to Mediterranean or low-
ble coronary heart disease. Eur Heart J. 2016;37:1993– fat diets in medicated secondary prevention patients.
2001. https://doi.org/10.1093/eurheartj/ehw125. Am J Cardiol. 2011;108(11):1523–9. https://doi.
36. Panagiotakos DB, Georgousopoulou EN, Georgiopoulos org/10.1016/j.amjcard.2011.07.008.
GA, et al. Adherence to Mediterranean diet offers an 49. Mena MP, Sacanella E, Vazquez-Agell M, et al.

additive protection over the use of statin therapy:results Inhibition of circulating immune cell activa-
from the ATTICA study (2002-2012). Curr Vasc tion: a molecular anti-inflammatory effect of the
Pharmacol. 2015;13(6):778–87. Mediterranean diet. Am J Clin Nutr. 2009;89:248–56.
37. Steffen LM, Horn LV, Daviglus ML, et al. A modified 50. Fito M, Guxens M, Corella D, et al. Effect of a tra-
Mediterranean diet score is associated with a lower ditional Mediterranean diet on lipoprotein oxidation.
risk of incident metabolic syndrome over 25 years A randomized controlled trial. Arch Intern Med.
among young adults: the CARDIA (Coronary Artery 2007;167:1195–203.
Risk Development in Young Adults) study. Br J Nutr. 51. Estruch R, Martinez-Gonzalez MA, Dolores Corella
2014;112:1654–61. D, et al. Effects of a Mediterranean-style diet on car-
38. Dilis V, Katsoulis M, Lagiou P, et al. Mediterranean diovascular risk factors a randomized trial. Ann Intern
diet and CHD: the Greek European Prospective Med. 2006;145:1–11.
Investigation into Cancer and Nutrition cohort. Br J 52. Esposito K, Marfella R, Ciotola M, et al. Effect of
Nutr. 2012;108:699–709. a Mediterranean-style diet on endothelial dysfunc-
39. Guallar-Castillon P, Radriguez F, Tormo MJ, et al. tion and markers of vascular inflammation in the
Major dietary patterns and risk of coronary heart dis- metabolic syndrome a randomized trial. JAMA.
ease in middle-aged persons from a Mediterranean 2004;292:1440–6.
country: the EPIC-Spain cohort study. Nutr Metab 53. Singh RB, Dubnov G, Niaz MA, et al. Effect of an
Cardiovasc Dis. 2012;22(3):192–9. Indo-Mediterranean diet on progression of coro-
40. Liyanage T, Ninomiya T, Wang A, et al. Effects of nary artery disease in high risk patients (Indo-
the Mediterranean diet on cardiovascular outcomes— Mediterranean Diet Heart Study): a randomised
a systematic review and meta-analysis. PLoS One. single-blind trial. Lancet. 2002;360:1455–61.
2016;11(8):e0159252. https://doi.org/10.1371/jour- 54. de Lorgeril M, Salen P, Martin J-L, et al. Mediterranean
nal.pone.0159252. diet, traditional risk factors, and the rate of cardiovas-
41. Rees K, Hartley L, Flowers N, et al. Mediterranean cular complications after myocardial infarction. final
dietary pattern for the primary prevention of car- report of the Lyon Diet Heart Study. Circulation.
diovascular disease. Cochrane Database Syst Rev. 1999;9:779–85.
2013;8:CD009825. https://doi.org/10.1002/14651858. 55.
Kastorini CM, Milionis HJ, Goudevenos JA,
CD009825.pub2. Panagiotakos DB. Mediterranean diet and coronary
42. Kastorini C-M, Milionis HJ, Esposito K, et al. The heart disease: is obesity a link? A systematic review.
effect of Mediterranean diet on metabolic syndrome Nutr Metab Cardiovasc Dis. 2010;20:536–51.
56. Zidi W, Allal-Elasmi M, Zayani Y, et al. Metabolic Dietary Approaches to Stop Hypertension (DASH)
syndrome, independent predictor for coronary artery Trial. Am J Clin Nutr. 2001;74:80–9.
disease. Clin Lab. 2015;61:1545–52. 69. Eckel RH, Jakicic JM, Ard JD, et al. 2013 AHA/
57.
Salehi-Abargouei A, Maghsoudi Z, Shirani F, ACC guideline on lifestyle management to reduce
Azadbakht L. Effects of Dietary Approaches to Stop cardiovascular risk: a report of the American College
Hypertension (DASH)-style diet on fatal or nonfatal of Cardiology/American Heart Association Task
cardiovascular disease incidence: a systematic review Force on Practice Guidelines. J Am Coll Cardiol.
and meta-analysis on observational prospective stud- 2014;63:2960–84.
ies. Nutrition. 2013;29:611–8. 70. Huang T, Yang B, Zheng J, et al. Cardiovascular dis-
58. Struijk EA, May AM, Wezenbeek NL, et al.
ease mortality and cancer incidence in vegetarians:
Adherence to dietary guidelines and cardiovascular a meta-analysis and systematic review. Ann Nutr
disease risk in the EPIC-NL cohort. Int J Cardiol. Metab. 2012;60(4):233–40.
2014;176(2):354–9. 71. Wang F, Zheng J, Yang B, et al. Effects of vegetarian
59. Levitan EB, Lewis CE, Tinker LF, et al. Mediterranean diets on blood lipids: a systematic review and meta-
and DASH diet scores and mortality in women with analysis of randomized controlled trials. J Am Heart
heart failure. The Women’s Health Initiative. Circ Assoc. 2015;4:e002408. https://doi.org/10.1161/
Heart Fail. 2013;6:1116–23. JAHA.115.002408.
60. Levitan EB, Wolk A, Mittleman MA. Relation of consis- 72. Kourlaba G, Polychronopoulos E, Zampelas A,

tency with the dietary approaches to stop h ypertension et al. Development of a diet index for older adults
diet and incidence of heart failure in men aged 45 to 79 and its relation to cardiovascular disease risk fac-
years. Am J Cardiol. 2009;104(10):1416–20. tors: the Elderly Dietary Index. J Am Diet Assoc.
61. Fung TT, Chiuve SE, McCullough ML, et al.
2009;109:1022–30.
Adherence to a DASH-style diet and risk of coronary 73. Atkins JL, Whincup PH, Morris RW, et al. High

heart disease and stroke in women. Arch Intern Med. diet quality is associated with a lower risk of car-
2008;168(7):713–20. diovascular disease and all-cause mortality in
62. Siervo M, Lara J, Chowdhury S, et al. Effects of the older men. J Nutr. 2014;144:673–80. https://doi.
Dietary Approaches to Stop Hypertension (DASH) org/10.3945/n.113.186486.
diet on cardiovascular risk factors: a systematic 74. Dong JY, Zhang YH, Wang P, et al. Meta-analysis
review and meta-analysis. Br J Nutr. 2015;113:1–15. of dietary glycemic load and glycemic index in rela-
63. Jenkins DJA, Boucher BA, Ashbury FD, et al. Effect tion to risk of coronary heart disease. Am J Cardiol.
of current dietary recommendations on weight 2012;109:1608–13.
loss and cardiovascular risk factors. J Am Coll 75. Burger KN, Beulens JW, Boer JM, et al. Dietary gly-
Cardiol. 2017;69:1103–12. https://doi.org/10.1016/j. cemic load and glycemic index and risk of coronary
jacc.2016.10.089. heart disease and stroke in Dutch men and women: the
64. Chen ST, Maruthur NM, Appel LJ. The effect of
EPICMORGEN study. PLoS One. 2011;6:e25955.
dietary patterns on estimated coronary heart disease 76. Jakobsen MU, O’Reilly EJ, Heitmann BL, et al.

risk: results from the Dietary Approaches to Stop Major types of dietary fat and risk of coronary heart
Hypertension (DASH) Trial. Circ Cardiovasc Qual disease: a pooled analysis of 11 cohort studies. Am J
Outcomes. 2010;3(5):484–9. Clin Nutr. 2009;89:1425–32.
65. Maruthur NM, Wang N-Y, Appel LJ. Lifestyle
77. Simila ME, Kontto JP, Satu Mannisto S, et al.

interventions reduce coronary heart disease risk: Glycaemic index, carbohydrate substitution for fat
results from the PREMIER Trial. Circulation. and risk of CHD in men. Br J Nutr. 2013;110:1704–
2009;119(15):2026–31. 11. https://doi.org/10.1017/S0007114513000858.
66. King DE, Egan BM, Woolson RF, et al. Effect of a high- 78. Li Y, Hruby A, Bernstein AM, et al. Saturated fat as
fiber diet vs. a fiber-supplemented diet on C-reactive compared with unsaturated fats and sources of car-
protein level. Arch Intern Med. 2007;167:502–6. bohydrates in relation to risk of coronary heart dis-
67. Appel LJ, Sacks FM, Carey VJ, et al. Effects
ease: a prospective cohort study. J Am Coll Cardiol.
of protein, monounsaturated fat, and carbohy- 2015;66(14):1538–48. https://doi.org/10.1016/j.
drate intake on blood pressure and serum lipids. jacc.2015.07.055.
Results of the OmniHeart randomized trial. JAMA. 79. Satija A, Bhupathiraju SN, Spiegelman D, et al.

2005;294:2455–64. Healthful and unhealthful plant-based diets and the
68. Obarzanek E, Sacks FM, Vollmer WM, et al. Effects risk of coronary heart disease in US adults. J Am Coll
on blood lipids of a blood pressure–lowering diet: The Cardiol. 2017;70(4):411–22.
Whole Plant Foods and Coronary
Heart Disease 12
Keywords
Coronary heart disease • Whole-grains • Fruit • Vegetables • Legumes
Nuts • Seeds • Low density lipoprotein-cholesterol • High density
lipoprotein-cholesterol • Triglycerides • C-reactive protein
Key Points non-soy and/or soy products), and ≥5 serv-

ings/week of nuts and seeds.
• Prospective cohort studies consistently show • Randomized controlled trials (RCTs) gener-
that diets with higher intakes of whole and ally support the beneficial effects of healthy
minimally processed plant foods (whole plant whole plant based foods on CHD risk bio-
foods) including whole-grains, fruits, vegeta- markers including lowering serum lipids and
bles, legumes, and nuts and seeds are associ- blood pressure, improving glucose and insulin
ated with reduced coronary heart disease metabolism, improving endothelial function,
(CHD) risk compared to lower intake. alleviating oxidative stress and inflammation
• Heart healthier versions of whole plant foods and reducing risk of weight gain compared to
are higher in dietary fiber, phytosterols, their refined counterparts.
healthy fatty acids (MUFAs and PUFAs), and • Only a small fraction of the US population
nutrients (e.g., vitamins E and C, potassium meets the recommended intake levels for
and folate), phytochemicals (carotenoids, fla- whole-grains, fruits, vegetables, legumes, and
vonoids and phytosterols) and lower in energy nuts and 70% exceed recommended refined
density, glycemic index and glycemic load. grain intake. Approximately 45% of CHD
• The risk of CHD incidence or mortality is deaths in the US are associated with subopti-
significantly reduced with the intake of ≥3 mal low dietary intake of fruits, vegetables,
servings/day of whole-grains (especially nuts and seeds, whole-grains, and seafood
oats and barley), ≥5 servings (400 g)/day of omega-3 PUFAs, and high intake of red and
fruits and vegetables, ≥4 weekly servings processed meats, sugar sweetened beverages,
(130–150 g cooked) of legumes (both and sodium.

338 12 Whole Plant Foods and Coronary Heart Disease
12.1 Introduction are likely to be similar or better than those due to

drug treatment of elevated blood cholesterol or
Coronary heart disease (CHD) is the leading cause pressure because they contribute to multiple cardio-
of mortality worldwide, especially in countries that metabolic health benefits [8]. Specific foods can
have acculturated to the Western lifestyle [1–6]. have a positive or negative effect on CHD risk
Although there have been some decreasing trends [49, 50]. A comparative analysis of the association
in overall CHD mortality rate over the last several between dietary factors and CHD risk from popula-
decades because of cholesterol lowering drugs and tion dietary habits and demographic data collected
surgical procedures, it is still a leading cause of by the US National Health and Nutrition
death globally and its prevalence is expected to Examination Surveys (NHANES; 1999–2002 and
increase as the global population ages. The role of 2009–2012; 16,620 individuals) found that fruit
elevated lipids, lipoproteins and inflammation as (excluding juice), vegetables including legumes,
modifiable risk factors for CHD is well established nuts and seeds, whole grains, seafood omega-3 fat,
[5–9]. Healthy behaviors, including meeting the and polyunsaturated fat reduce CHD risk and red
recommended intake of whole plant foods, and processed meat and sugar sweetened beverages
increased physical activity, and weight control, increase CHD risk (Fig. 12.1) [49]. Similar findings
have been shown to reduce the risk for CHD [6–9]. were shown in a systematic review of the evidence
A recent study in the US, showed that adults with supporting a causal link between whole plant foods
higher consumption of a plant-based diet rich in and lowering of CHD risk compared to Western
healthier plant foods (whole-grains, fruit, vegeta- diets (Fig. 12.2) [50]. This is exemplified by a case
bles, legumes, nuts, seeds, vegetable oils, tea and study of a 60-year-old man with typical angina.
coffee) had significantly reduced CHD risk by 25% After a positive stress test, he declined both drug
whereas less healthy plant foods (sweetened bever- therapy and invasive treatment and opted instead to
ages, refined grains, French fries, sweets and red adopt a whole food plant-based diet, which con-
and processed meats significantly increased CHD sisted primarily of vegetables, fruits, whole grains,
risk by 32% [10]. Whole and minimally processed legumes, and nuts [51]. After 4 months, his BMI fell
plant foods (whole plant foods) are higher in fiber from 26 kg/m2 to 22 kg/m2, his blood pressure nor-
and nutrient density (e.g., vitamins E and C, and malized, his low-density lipoprotein- cholesterol
trace minerals such as selenium and copper), and (LDL-C) decreased from 158 mg/dL to 69 mg/dL
phytochemicals (carotenoids, flavonoids and phy- and he was able to walk one mile without angina.
tosterols) than refined foods (Appendix A) The objective of this chapter is to comprehensively
[11–19]. Conseqently, healthy whole plant based review the effects of whole plant foods on CHD risk
foods have more favorable effects on CHD risk by and related risk factors.
lowering serum lipids and blood pressure, improv-
ing glucose and insulin metabolism, improving
endothelial function, alleviating oxidative stress 12.2 W
hole Plant Foods
and inflammation and reducing risk of weight gain and Coronary Heart Disease
than refined foods [7–9, 11, 13–47]. However, Risk
whole plant foods or minimally processed plant
foods are consumed at low levels in the US and The US National Center for Health Statistics
other Western diet consuming populations [48]. It reports that about 45% of cardiometabolic deaths
is estimated that most of the US population falls (based on 702,308 cardiometabolic deaths in
short of meeting the recommended minimal intake 2012) are associated with suboptimal intake of
levels for whole-grains by 99%, fruit by 85% and the following dietary factors including low
vegetables by 90%, and 70% of the US population intakes of fruits, vegetables, nuts and seeds,
exceeds the recommended refined grain intake. whole-grains, and seafood omega 3 and other
For primary CHD prevention, the benefits polyunsaturated (PUFA) fats, and excessive
achievable though diet and other lifestyle factors intake of red and processed meats, sugar
12.2 Whole Plant Foods and Coronary Heart Disease Risk 339
Age 70 years Age 50 years
Seafood omega-3 fats (100 mg/day)
PUFA Replace Carbs (5% kcals/day)
Sugar Sweetened Beverages (8 oz/day)
Processed meats (50 g/day)
Whole-grain (50 g/day)
Nuts and seeds (1 oz/week)
Vegetables and legumes (100 g/day)
Fruit (no juice) (100 g/day)
–40 –30 –20 –10 0 10 20 30 40 50

CHD Risk (%)
Fig. 12.1 Comparative analysis of the association between foods and beverages and coronary heart disease (CHD) risk
by adult age from the US National Health and Nutrition Examination Surveys (1999–2002 and 2009–2012; 16,620
individuals) (adapted from [49])
Nuts
Fruit & Vegetables
Whole-grains
0.5 0.6 0.7 0.8 0.9 1

Relative Risk for CHD
Fig. 12.2 Effect of the recommended intake of whole-grains, fruits, vegetables and nuts on a coronary heart disease
(CHD) composite risk score from prospective cohort studies and RCTs (adapted from [50])
s weetened beverages, and sodium [49]. The opti- PUFA as a replacement for carbohydrates or sat-
mal intake of these dietary factors is at least 300 g urated fat, and 250 mg of seafood omega-3 fat,
of non-juice fruit/day, 400 g vegetables including and limiting red meat to 100 g/week and sodium
legumes/day, 5 weekly servings of nuts, 2.5–3 to 2000 mg/day, and not or rarely consuming pro-
servings of whole-grains/day, 11% kcals from cessed meat or sugar sweetened beverages.
12.2.1 Whole-Grains whole grain source studied [64]. In adults aged

≥45 years, the consumption of adequate whole-
Whole-grain foods are an important part of a grain in conjunction with statins use is associated
healthy dietary pattern as they are major con- with healthier lipoprotein profiles compared to
tributors of fiber, minerals, vitamins, antioxi- low whole-grain intake [65]. Table 12.1. summa-
dants, phytosterols and lignans [52–56]. They rizes the findings from whole-grains prospective
consist of the intact, ground, cracked or flaked cohort studies and RCTs on CHD risk and bio-
grain kernels after removal of the inedible parts, markers [26, 28, 66–83].
such as the husk and hull, and include the starchy
endosperm, germ and bran fiber in similar pro- 12.2.1.1 Prospective Cohort Studies
portions that are found in the intact kernels. Prospective studies consistently show that increased
Milling and grinding of whole-grains to refined intake of whole-grains has significant beneficial
grains improves palatability but leaves primarily effects on lowering CHD risk [26, 66–75].
the starchy endosperm, which is lower in fiber,
vitamins, minerals and phytonutrients. Whole Meta-analyses and Systematic Reviews
grains have a wide range of effects on CHD risk Two meta-analyses show that the consumption of
because of their fiber, micronutrient and phyto- ≥3 servings of whole grain/day is significantly
chemical content, and physical properties such associated with lower CHD risk [67, 68]. A 2016
as particle size, solubility and viscosity. Oats and meta-analysis (10 cohorts; 704,317 participants;
barley containing food products are a good 26,243 CVD cases) found that the consumption
source of soluble fiber, the major component of of 3 servings of whole grain breads and breakfast
which is β-glucan, which is known to reduce cereals reduced CHD risk by 19% [67]. There
serum LDL-C and CHD risk [56]. Daily doses of was a clear dose-response relationship between
at least 3 g β-glucan may reduce plasma total and whole-grains and CHD risk up to 210 g/day, with
low-density lipoprotein (LDL) cholesterol levels a slightly steeper reduction in risk up to 3 serv-
by 5% in normocholesterolemic or by 10% in ings/day then a slower reduction above 3 serv-
hypercholesterolemic subjects. Also, there is ings/day. Similarly, a 2015 meta-analysis (15
growing focus on the ancient grain variety, cohort studies and 3 case-control studies; 400,492
Khorasan wheat for its CHD prevention proper- participants) demonstrated that approx. 3 serv-
ties, which are more effective in comparison to ings of whole-grains/day is associated with a
more contemporary wheat varieties [57]. 21% lower CHD risk compared to a low intake
The US dietary guidelines recommend ≥3 (<1 serving) [68]. There was no difference in risk
servings/whole-grains/day and ≤3 servings of reduction between men and women or geograph-
refined grains/day to promote health and well- ical areas observed in this analysis. Several sys-
ness associated with reduced risk of various tematic reviews suggested that increased intake
chronic diseases [48]. However, only about 1% of whole-grains is associated with lowering
of Americans follow the recommendation for C-reactive protein (CRP) as a potential mecha-
whole-grain intake. The average American’s nism for lowering CHD risk [26, 69]. Also, a
intake is <1 ounce whole grains/day and 70% dose-response meta-analysis of whole grains (3
exceed the recommended intake for refined cohort studies; 240,532 participants; 2678 deaths
grains [48, 58]. Adequate intake of whole-grains from CHD) found that each serving (30 g) of
is generally believed to have important health whole grains/day lowered risk of CHD mortality
benefits in reducing CHD risk with recognition by 8%, with significant heterogeneity [66].
of health claims in the US, EU and other coun-
tries [59–63]. Large prospective cohort studies Representative Cohort Studies
consistently associate whole-grain with improved Six prospective cohort studies consistently show
CHD outcomes. RCTs show mixed results for that increased whole-grain intake significantly
CHD biomarkers due to the differing composi- reduced CHD risk and mortality [70–76]. A
tions and physical properties of each specific Danish cohort study (54,871 adults; mean age
Table 12.1 Summaries of whole-grain studies on CHD risk and biomarkers

Systematic Reviews and Meta-analysis
Li et al. (2016). 3 cohort studies; 240,532 participants; Each 1 serving of whole grains/day lowered
Investigate the effect 2678 deaths from CHD risk of CHD mortality by 8%, with
of whole grains on the significant heterogeneity
risk of CHD specific
mortality according to
a dose-response
meta-analysis of
prospective cohort
studies [66]
Aune et al. (2016). 10 cohort studies; 704,317 participants; There was a 19% reduction in CHD risk per
Systematically 26,243 CVD cases 90 g/day increase in whole-grain intake (90 g
quantify the dose- is equivalent to three servings—example, two
response relation slices of bread and one bowl of cereal or one
between consumption and a half pieces of pita bread made from
of whole grain and whole-grains). A dose-response relationship
specific types of grains between whole-grains and CHD risk was
and the risk of shown up to 210 g/day. There was a slightly
cardiovascular disease steeper reduction in risk up to 3 servings/day
(CVD) [67] then a slower reduction above 3 servings/day.
The intake of whole grain bread, whole grain
breakfast cereals, total breakfast cereals, and
bran were inversely associated with CVD risk
Tang et al. (2015). 15 cohort studies and 3 case-control Higher intake of whole-grains (>3 servings)
Systematically studies); 12 studies from the US; 400,492 was significantly associated with a 21%
evaluate the findings participants; follow-up of 5 to 26 years; reduced risk for CHD. In subgroup analyses,
from cohort studies of 14,427 patients with CHD there were no differences in the overall results
the effects of by geography, study duration or gender
whole-grain intake on
CHD risk [68]
Buyken et al. (2014). 7 observational studies; 11,295 Five of 7 studies observed a significant
Systematically evaluate participants; aged 27 to 62 years; BMI of association between a higher whole-grain
the effect of whole- 24 to 31 intake and lower hs-CRP or IL-6
grains on inflammatory concentrations. One study reported only a
markers [69] trend for the relation between whole grain
and hs-CRP. Whole-grain intake was
especially effective in lowering inflammatory
markers among persons with type 2 diabetes
Lefevre et al. (2012). 13 prospective cohort studies; 35,771 Each whole-grain serving significantly
Systematically review participants reduced C-reactive protein (CRP) levels by
the effects of 7%. For highest vs. lowest intake, whole-
whole-grains on grains reduced CRP by 10–29%
systemic inflammation
[26]
Helnaes et al. (2016). 54,871 adults; mean baseline age Total whole-grain intake in the highest
Investigate the 57 years; 72% males and 28% women; quartile lowered risks of myocardial
association between mean 13.6-years of follow-up; 2329 infarction in men by 25% and in women by
whole-grain intake in individuals developed myocardial 27% compared to the lowest quartile. Rye
terms of total intake infarction (multivariate adjusted) bread (in men and women) and oatmeal (in
and intakes of men) were associated with significantly lower
different cereals and risk of myocardial infarction. The effects of
myocardial infarction types of whole grain breads and oatmeal are
(Denmark) [70] summarized in Fig. 12.3
(continued)
Johnsen et al. (2015). 120,000 subjects; median baseline age Higher intake of breakfast cereals lowered
Investigate the effect 51 years for women and 54 years for CHD mortality for women by 9% and for
of whole-grains on men; follow-up of 10–16 years; 298 men by 4%. Doubling intake of total
CHD mortality women and 858 men died from CHD whole-grain products lowered CHD mortality
(Scandinavian (multivariate adjusted) for women by 15% and for men by 8%.
HELGA Norwegian Among women, intake of oats and wheat was
sub-cohort) [71] associated with a borderline statistically
significant lower CHD mortality by 5%
Sonestedt et al. 4535 subjects; mean baseline age Higher whole grain intake significantly
(2015). 60 years; 62% females;14-years of lowered risk for ischemic CVD by 13%
Examine the follow-up (multivariate adjusted) (highest vs. lowest quintile). Higher intake of
association between foods rich in added sugar (sugar and sweets,
the consumption of and sugar-sweetened beverages) were
carbohydrate-rich significantly associated with higher
foods and beverages triglycerides and lower HDL-C
and the risk of incident concentrations
ischemic CVD (The
Malmö Diet and
Cancer Study;
Sweden) [72]
Mellen et al. (2007). 1178 adults; mean baseline age 55 years; Whole-grain intake was inversely associated
Evaluate the effect of 56% female; follow-up of 5 years with carotid common and internal C-IMT
whole-grain intake on (multivariate adjusted) progression
carotid intima-media
thickness (C-IMT)
progression (Insulin
Resistance
Atherosclerosis Study;
US) [73]
Jensen et al. (2004). 42,850 men; baseline age 40–75 years; CHD risk was significantly reduced by 18%
Estimate the 14-years of follow-up; 1818 cases of between extreme quintiles of whole-grain
associations of CHD (1261 nonfatal myocardial intake. Also, CHD risk was significantly
whole-grain, bran, and infarctions and 557 fatal CHD) reduced by 30% with the highest intake of
germ intakes with the (multivariate adjusted) added bran vs. no intake of added bran
incidence of CHD (Fig. 12.4)
(US- The Health
Professionals
Follow-Up Study) [74]
Liu et al. (1999). 75,521 women; baseline age Whole-grain intake significantly reduced
Evaluate whether high 38–63 years; follow-up of 10 years; 761 CHD risk by 25% for 2.7 servings vs. 0.13
whole-grain intake cases of CHD (208 of fatal CHD and 553 servings. In never smokers, whole-grain
affects risk of CHD in of nonfatal myocardial infarction) CHD risk was reduced by 51%
women (US -The (multivariate adjusted)
Nurses’ Health Study)
[75]
Jacobs et al. (1998). 34,492 postmenopausal women; baseline 3.2 servings/day whole-grains was
Determine the effect age 55–69 years; follow-up of 9 years significantly inversely associated with
of whole-grains on the (multivariate adjusted) ischemic heart disease mortality with reduced
risk of ischemic heart risk by 30% vs. 0.2 servings/day. Dark bread
disease death (The intake vs. white bread was shown to
Iowa Women’s Health significantly lower ischemic heart disease
Study; US) [76] risk (Fig. 12.5)
Meta-analyses and Systematic Reviews
Hollaender et al. 24 RCTs; 16 of the studies evaluated Whole-grain intake significantly lowered mean
(2015). hypercholesterolemic subjects; 2275 LDL-C by 3.5 mg/dL and TC by 4.6 mg/dL
Systematically participants; aged 18–75 years; the daily compared with the refined control. Whole-grain
compare the effect of whole-grain dose ranged from 28 to oats was most effective in lowering TC by
whole-grain vs. 213 g/day; 6–8 weeks 6.6 mg/dL. There was no effect of whole-grain
refined-grains on foods on HDL-C but TG was lowered by
fasting lipids [28] 3.5 mg/dL compared with the control
(p = 0.10). A weight-loss diet improved the
efficacy of whole-grains in reducing LDL-C
Zhu et al. (2015). 17 RCTs; 916 hyperlipidemic subjects β-glucan significantly lowered TC by 10 mg/
Systematically including 14 parallel and 3 crossover dL and LDL-C by 8 mg/dL with no
evaluate the clinical studies; 9 oat and 5 barley β-glucan significant differences in HDL-C, and TG vs.
effects of beta-glucan studies compared to appropriate controls. control diets independent of baseline TC or
from oat and barley on The β-glucan dose ranged from 2.8 to LDL-C, age (< or ≥50 years), duration (<or
fasting lipid levels in 10.3 g/day (mean 5.5 g/day); 4–12 weeks ≥8 weeks), barley-derived β-glucan,
subjects [77] oat-derived β-glucan at > or ≤5 g day
Whitehead et al. 28 RCTs with 12 in healthy subjects, Diets containing ≥3 g oat bran β glucan/day
(2015). 13 in hyperlipidemic subjects, and 3 in reduced serum TC and LDL-C relative to
Systematically evaluate diabetic subjects; 2700 subjects; age control by 9.5 and 11.5 mg/dL, respectively,
clinical effects of oat range 25–63 years; average β-glucan with no significant effects on HDL-C or
beta-glucan/day on intake from food and supplements was TG. LDL-C lowering was significantly
fasting lipid levels [78] 5.5 g/day (3–12 g/day); average 6 weeks greater with higher baseline LDL-C
(2–12 weeks)
Lefevre et al. (2012). 7 RCTs; 742 participants; aged Only 1 of 7 RCTs showed significantly
Systematically assess 46–60 years; BMI range of 27–36; reduced CRP for the whole-grain high fiber
the effect of whole 3–16 weeks group as part of a hypocaloric diet compared
grains on CRP [26] to the low fiber control
Representative RCTs
Whittaker et al. Crossover RCT: Ancient Khorasan wheat significantly
(2015). 22 acute coronary syndrome patients; lowered TC by 6.8%, LDL-C by 8.1%,
Examine CHD 9 females and 13 males; median age glucose by 8% and insulin by 24.6%,
protective effects of 61 years; mean BMI 26.9; assigned to independently of age, sex, traditional risk
substituting ancient, consume products (bread, pasta, biscuits factors, medication and diet quality
organic khorasan and crackers) made either from organic (Fig. 12.6). Additionally, there was a
wheat for modern semi-whole khorasan wheat or organic significant reduction in reactive oxygen
conventional grains semi-whole semolina control wheat; species, lipoperoxidation of circulating
(Italy) [79] patients ingested 62 g dry weight monocytes and lymphocytes, and TNF-alpha
khorasan or control/day; 8 weeks; vs. the contemporary semolina wheat
8-weeks washout
Vitaglione et al. Parallel RCT: The whole-wheat group significantly reduced
(2015). 80 healthy overweight and obese TNF-α and increased IL-10 (anti-
To evaluate the effect subjects; mean age about 40 years; 70% inflammatory) levels compared with the
of whole-wheat foods female; sedentary lifestyles; diets: 70 g/ refined wheat group. There was a borderline
on systemic day of whole-wheat foods (>4 servings; significant reduction in IL-6 compared with
inflammation (Italy) 8 g fiber/day) or refined-wheat foods the refined wheat group (p = 0.06)
[80] (2 g fiber/day); 8 weeks
Giacco et al. (2013). Parallel RCT: There were no significant effects from
Evaluate the effect of 146 metabolic syndrome subjects; aged whole- grains on TC, LDL-C, HDL-C, and
whole-grain vs. refined 40–65 years; randomized into wheat and TG or systemic inflammatory markers (e.g.,
cereal foods on rye whole-grain (33 g fiber/day) or CRP, IL-6, and TNF-α) compared to the
cardiometabolic risk refined cereal products (20 g fiber/day); refined grain control. No changes in
factors (Italy and 12 weeks anthropometric parameters were observed
Finland) [81]
(continued)
Brownlee et al. Parallel RCT: Increased whole-grain intake did not
(2010). 316 adult subjects; mean age 46 years; significantly improve any markers of CVD
Study the effect of mean BMI 30; consuming <30 g risk including BMI, % body fat, waist
whole-grains on CHD whole-grains/day; randomized to circumference; fasting lipid profile, glucose
biomarkers in people 3 groups: (1) refined grain control, (2) and insulin, CRP, or endothelial function
who habitually 60 g whole-grains/day for 16 weeks and compared to refined grain
consume refined (3) 60 g whole-grains/day for 8 weeks
-grains (WHOLEheart followed by 120 g whole-grains/day for
study; UK) [82] 8 weeks
Katcher et al. (2008). Parallel RCT: There was no difference in weight loss
To evaluate the effect 50 obese subjects; mean age 46 years between groups, although the whole-grain
of a whole-grain rich BMI > 30; randomized into 500 kcal/day group lost a significantly greater % body fat
hypocaloric diet on deficit groups with dietary advice to in the abdominal region and lowered CRP
cardiometabolic health either avoid all whole-grain foods or levels by 38% vs. the refined grain group
in obese subjects [83] consume only whole-grain foods;
12 weeks
57 years; 13.6 years of follow-up) observed that of follow-up) observed that increased whole-
higher total whole-grain intake lowered myocar- grains intake was inversely associated with
dial infarction risk by about 25% in both men and carotid intima-media thickness progression [73].
women [70]. This was especially true for oatmeal The Malmö Diet and Cancer Study cohort (4535
and crispbread in men and rye bread and oatmeal subjects; mean baseline age 60 years; 62%
in women (Fig. 12.3). The Health Professionals females; 14 years of follow-up) found that higher
Follow-up Study (42,850 men; baseline aged whole grain intake significantly lowered risk for
40–75 years; follow-up of 14 years) observed that ischemic CVD by 13% (highest vs. lowest quin-
higher whole-grain intake ≥50 g/day significantly tile) [72]. Additionally, higher intake of foods rich
reduced CHD risk by18% compared to about 3 g/ in added sugar was significantly associated with
day of whole-grains, and highest intake of added higher triglycerides (TG) and lower HDL-C
bran vs. no intake of added bran significantly concentrations.
reduced risk by 30%, after multivariate adjust-
ments (Fig. 12.4) [74]. The Nurses’ Health Study 12.2.1.2 Randomized Controlled
(75,521; 38–62 years at baseline; 10 years of fol- Trials (RCTs)
low-up) observed that about 3 daily servings vs. Intervention trials show mixed results for CHD
0.1 servings of whole-grains significantly low- biomarkers due to the varying compositions of
ered CHD risk by 25% for all women and 51% for whole-grains and levels of intake [26, 28, 77–
never smoking women [75]. The Iowa Women’s 83]. A 2015 meta-analysis (24 RCTs; 2275 par-
Health Study (34,492 post-menopausal women; ticipants) estimated that increased whole-grain
55–69 years at baseline; 9 years of follow-up) intake significantly lowered mean LDL-C by
observed an inverse association between whole 3.5 mg/dL and total cholesterol (TC) by 4.6 mg/
grain intake and risk of death from ischemic heart dL compared with the refined control [28]. Oat
disease with a 30% lower risk for about 3 serv- whole-grain was twice as effective in lowering
ings/day of whole-grains vs. 0.2 servings, after TC compared to the mean of all the other whole-
multivariate adjustments [76]. This study also grains analyzed. Whole-grains insignificantly
found that whole-grain dark bread significantly lowered TG by 3.5 mg/dL (p = 0.10) and had no
lowered ischemic heart disease risk compared to effect on HDL-C. The two most recent meta-
white bread in postmenopausal women (Fig. 12.5). analyses found that the mean intake of 5.5 g
The US Insulin Resistance Atherosclerosis Study β-glucan from oat or barley whole grains sig-
(1178 adults; mean baseline age 55 years; 5 years nificantly lowered TC and LDL-C by about
Rye bread Whole-grain bread Oatmeal Crispbread

1.1
1.05
1
0.95
Relative Risk of MI
0.9
0.85
0.8
0.75
0.7
0.65
0.6
Men Women
Fig. 12.3 Association between different whole-grain forms and myocardial infarction (MI) risk per 25 g intake (adapted from [70])
Whole-Grains (p =.01) Bran (p <.001)

1.1
Hazard Ratio for CHD Incidence in Men
0.9
0.8
0.7
0.6
3.5/0.0 9.6/0.3 16/1.4 24.7/4.23 42.4/11.1
Whole Grain/Bran Intake (g) per Day
Fig. 12.4 Association between whole-grains and bran and coronary heart disease (CHD) incidence in men (adapted from [74])
10 mg/dL each [77, 78]. A review of whole- compared to modern wheat varieties but the
grains found that only 1 of 7 RCTs showed sig- mechanism is not completely understood
nificantly reduced CRP compared to refined (Fig. 12.6) [57, 79]. Most RCTs show that the
grains and that reduction was from a hypocalo- substitution of whole grain wheat for refined
ric diet RCT [26]. Representative RCTs show wheat does not significantly reduce lipopro-
mixed effects from whole-grain intake on CHD teins, glycemic, systemic inflammatory or endo-
risk biomarkers depending on the study popula- thelial markers associated with increased CHD
tion, whole-grain form and overall diet [79–83]. risk but whole grain wheat may further lower
The ancient grain khorasan wheat has been circulatory inflammatory markers such as CRP
shown in several RCTs to improve lipoprotein in weight loss diets compared to refined grains
profiles, glycemic and inflammatory biomarkers [80–83].
Whole-grain Bread (p-trend =.018) White Bread (p-trend =.13)

1.6
Relative Risk for IHD in Postmenopausal Women 1.4
1.2
0.8
0.6
0.4
1 2 3 4 5
Daily Bread Serving (Quintiles)
Fig. 12.5 Association between type of bread consumed and ischemic heart disease (IHD) risk in 34,492 postmeno-
pausal women, mean age 61 years, 9 years of follow-up (adapted from [76])
Ancient Grain-Khorasan Modern Grain-Semolina

0.2
Change from Baseline (mmol/L)
0.1
–0.1
–0.2
–0.3
–0.4
–0.5
Total Blood
LDL-C HDL-C Triglycerides
cholesterol glucose
(p =.001) (p =.4) (p =.2)
(p =.001) (p =.03)
Ancient Grain-Khorasan –0.3 –0.18 –0.03 –0.06 –0.44
Modern Grain-Semolina 0.13 0.04 –0.04 0.08 0.09
Fig. 12.6 Effect of ancient grain vs. modern grain on major coronary heart disease (CHD) risk biomarkers (adapted
from [79])
12.2.2 Fruits and Vegetables vitamins and phytochemicals, especially pheno-

lics and carotenoids; minerals, especially electro-
Fruit and vegetables are important parts of most lytes; and fiber [84]. The World Health
global dietary guidance recommendations Organization (WHO) Report recommended a
because of their concentrations of: antioxidant minimum daily intake of 400 g of fruits and
vegetables, based on evidence that higher levels fruits and vegetables lowered CHD risk by 24%.
are protective against CHD [85]. This led to the For specific types of fruits and vegetables, higher
launch of various “eat 5 or more” fruit and vege- intake of apples, pears, citrus fruits, 100% fruit
table campaigns in Europe, the US and Australia juices, onions, green leafy vegetables, β-carotene-
[86]. The USDA MyPlate educational concept, rich fruits and vegetables and vitamin C-rich fruits
devotes one-half the plate to fruits and vegetables and vegetables were most effective at lowering
because they are i mportant for a healthy diet and CHD risk (Fig. 12.7). CVD risk was reduced per
also displace from the diet other foods of higher 200 g/day total fruits and vegetables by 8%, for
energy density and lower nutrient and phyto- fruits by 13% and vegetables by 10%. There was a
chemical levels [87]. However, globally, fruit 28% lower CVD risk for intakes of 800 g/day for
and vegetable consumption is only at a small fruits and vegetables and fruits, and 600 g/day of
fraction of the recommended levels [88]. In the vegetables. Of specific types of fruits and vegeta-
US, >85% of the population falls short of meet- bles, higher intake of apples, pears, citrus fruits,
ing the daily fruit and vegetable intake recom- carrots, and green leafy vegetables were most
mendations [58]. For fruit, whole fruit comprises effective in lowering CVD risk whereas a high
only about half the daily fruit intake with the intake of canned fruits increased risk (Fig. 12.8). A
remainder primarily consumed as 100% fruit 2015 dose-response meta-analysis (23 cohort
juice. For vegetables, white potatoes (a good studies; 937,665 participants; 5–37 years of fol-
source of potassium and fiber), are the most com- low-up) found that the CHD risk for intakes of
monly consumed single vegetable, accounting 500 g/day of total fruits and vegetables reduced
for 25% of all vegetable consumption [58]. risk by 12%, 300 g/day of fruits by 16%, and
Potatoes are consumed in a variety of forms, with 400 g/day of vegetables by 18% [90]. In the sub-
about 31% being boiled (including mashed and group analysis, a significant inverse association
in dishes such as potato salad, soups, and stews), was observed in western populations, but not in
22% as chips, sticks, or puffs, 19% as French Asian populations. A 2014 dose-response meta-
fries, 17% as baked, and 12% as home fries or analysis (10 cohorts; 1,147,225 participants)
hash browns. The effects of increased fruit and showed a lower risk of CVD mortality with each
vegetable intake on CHD risk and related bio- serving of fruit by 5% or vegetables by 4% [91].
markers from prospective cohort studies and Two other meta-analyses also found similar pro-
RCTs is summarized in Table 12.2 [30, 31, tective effects of increased fruit and vegetable
89–103]. intake on CHD risk [92, 93].
12.2.2.1 Prospective Cohort Studies Representative Cohort Studies

Six representative cohort studies consistently
Systematic Reviews and Meta-analyses support the protective benefits of fruit and
Five meta-analyses of fruit and vegetable prospec- vegetable consumption on CHD risk [31, 94–98].
tive studies consistently confirm that adequate The effects of fruit and vegetable intake were
fruit and vegetable intake (e.g., 5 servings per day) shown to be consistent across ethnic groups [94].
is significantly associated with CHD risk reduc- In the Shanghai Women’s Health Study and
tion [89–93]. A 2017 systematic review and dose- Shanghai Men’s Health Study (62,211 women,
response meta-analysis (66 CHD cohort studies, > mean baseline age 54 years, 9.8 years of follow-
2 million participants; and 64 CVD cohort studies, up; 55,474 men, mean baseline age 52 years,
> 600,000 participants) found that increased total 5.4 years of follow-up) women had a significant
and specific subtypes of fruits and vegetables sig- 6% reduction in CHD risk per 80 g fruit and veg-
nificantly reduced both CHD and CVD risk [89]. etable intake but there was no significant lower-
CHD risk was reduced per daily 200g for fruits ing of CHD risk in men [31]. The pooled data
and vegetables by 8%, for fruits by 10% and veg- from the Nurses’ Health Study and Health
etables by 16%. The intake of 800 g/day of total Professionals Follow-up Study observed that
Table 12.2 Summaries of fruit and vegetable studies on CHD risk and biomarkers
Meta-Analyses
Aune et al. (2017). CHD: CHD:
Assess the dose- 66 cohort studies; > 2 million Per 200 g/day, CHD risk was reduced for fruits
response relationship participants and vegetables by 8%, for fruits by 10% and
between fruit and CVD: vegetables by 16%. There was a 24% reduction
vegetable intake and 64 cohort studies; > 600,000 in the relative risk at an intake of 800 g/day of
risk of CHD and participants fruits and vegetables. There was a 21% risk
CVD and the effects reduction up to 750–800 g/day for fruits and a
of specific types of 30% risk reduction up to 550–600 g/day for
fruits and vegetables vegetables. For specific types of fruits and
[89] vegetables: higher intake of apples, pears, citrus
fruits,100% fruit juices, β-carotene- or vitamin
C-rich fruits and vegetables and green leafy
vegetables were among the most effective at
lowering CHD risk (Fig. 12.7)
CVD:
Per 200 g/day, CVD risk was reduced for fruits
and vegetables by 8%, for fruits by 13% and
vegetables by 10%. There were approx. 28% risk
reductions for intakes of 800 g/day for fruits and
vegetables combined, and fruits, and 600 g/day
of vegetables. Of specific types of fruits and
vegetables, higher intake of apples and pears,
citrus fruits, carrots, and green leafy vegetables
were among the most effective in lowering CVD
risk (Fig. 12.8)
Gan et al. (2015). 23 cohort studies; 937,665 In the dose-response analysis, the risk of CHD
Systematically participants; over 5 to 37 years of was reduced by 12% per 477 g/day in total fruits
evaluate the follow-up; 18,047 cases of CHD and vegetables, by 16% per 300 g/day in fruit,
relationship of fruit reported and by 18% per 400 g/day in vegetables
and vegetable intake
with CHD risk and
quantify the
dose-response
relationship between
them [90]
Wang et al. (2014). Total fruits and vegetables 4 cohort CVD mortality was significantly reduced per
Systematically studies, 469,551 participants, 6893 daily serving of fruits and vegetables combined
examine and quantify CVD deaths; fruits or vegetables 6 by 4%, for fruit consumption by 5%, and for
a possible dose- cohort studies, 677,674 participants, vegetable consumption by 4%
response relation 9744 CVD deaths; 4.6 to 26 years of
between fruit and follow-up
vegetable intake and
risk of CVD [91]
He et al. (2007). 12 cohort studies; 278,459 Subjects consuming >5 daily fruit and vegetables
Systematically assess individuals; median follow-up of servings significantly reduced CHD risk by 17%
the relationship of 11 years; 9143 CHD events vs. those who ate <3 daily fruit and vegetable
fruit and vegetable servings. The consumption of 3 to 5 daily
intake with CHD risk servings of fruits and vegetables reduced CHD
[92] risk by 7% (p = 0.06)
Dauchet et al. 9 studies; 91,379 men, 129,701 The risk of CHD was significantly reduced for
(2006). women; follow-up duration each daily portion of fruits and vegetables by 4%
Systematically 5–19 years; 5007 CHD events and for each fruit portion by 7%
determine the strength
of the association
between fruit and
vegetable intake and
CHD risk [93]
Sharma et al. (2014). 164,617 adults; mean baseline age In men, there was a significant 27% lower
Assess the effects of 66 years for males and 59 years for ischemic heart disease risk at ≥6.6 daily servings
fruit and vegetable females; 4 years of follow-up; 1140 of vegetables. Inconsistent results were observed
intake on mortality males and 811 females with fatal for women, where the protective association was
rates from ischemic ischemic heart disease cases observed only at mid-level vegetable intake
heart disease among (multivariate adjusted) levels, but not among women with the highest
ethnic groups level of vegetable intake. There was no evidence
(Multi- ethnic Cohort of an association for fruit intake in men or
Study; US) [94] women
Yu et al. (2014). 67,211 women, mean baseline age Women in the highest quartile of total fruit and
Examine the effect of 54 years, mean follow-up of 9.8 years; vegetable intake (median: 814 g/day) had a
fruit and vegetable 148 CHD events; and 55,474 men, significant 38% reduced CHD risk compared
intake on CHD risk mean baseline age 52 years, mean with those in the lowest quartile (median: 274 g/
among Chinese adults follow-up of 5.4 years; 217 CHD day). Each 80 g/day increment of total fruit and
(Shanghai Women’s events (multivariate adjusted) vegetable intake was associated with a 6%
Health Study and the reduction of CHD risk among women. For men,
Shanghai Men’s no significant association was found for fruit and
Health Study; China) vegetable intake when analysed either in
[31] combination or individually
Bhupathiraju et al. 71,141 women, mean baseline age The consumption of 4–5 fruit and vegetable
(2013). 50 years, 24-years of follow-up, 2582 servings/day was associated with lower CHD risk
Measure the roles of cases of CHD; and 42,135 men, mean by 16%, multivariate adjusted. The intake of
quantity and variety of baseline age 53 years, 22-years of citrus fruit, green leafy vegetables, and
fruit and vegetable follow-up, 3607 CHD cases β-carotene- and vitamin C-rich fruit and
intake on the incidence (multivariate adjusted) vegetables were more potent at lowering CHD
of CHD (US Nurses’ risk vs. the general intake of fruits and vegetables
Health Study and (Fig. 12.9)
Health Professionals
Follow-Up studies
[95]
Griep et al. (2011). 20,000 adults; mean baseline age For each daily 25 g increase in the intake of: (1)
Determine the effects 41 years; mean BMI 25; 10 years of mixed fruits and vegetables reduced CHD risk by
of fruit and vegetables follow-up; 245 cases of CHD 2% and (2) orange fruits and vegetables
of different colors (multivariate adjusted) significantly reduced CHD risk by 24%.
(green, orange/yellow, Specifically, carrots, were associated with a
red/purple, and white) significant 32% lower risk of CHD
on 10-year CHD
incidence
(Monitoring Project
on Risk Factors and
Chronic Diseases in
the Netherlands
Study) [96]
(continued)
Griep et al. (2010). 20,069 adults; mean baseline age Higher fruit and vegetable intake >475 g/day was
Evaluate the effects 41 years; 10.5 years of follow-up; 245 associated with a 34% lower CHD risk vs.
of raw vs. processed CHD cases (multivariate adjusted) ≤241 g/day. Both raw or processed fruit and
fruit and vegetable vegetables may protect against CHD risk
intake on CHD risk (Fig. 12.10). A subgroup analysis indicated that
(Monitoring Project increased fruit and vegetable intake was more
on Risk Factors and effective in reducing CHD risk in women and
Chronic Diseases in subjects ≥50 years of age
the Netherlands
Study) [97]
Joshipura et al. 84,251 women, baseline 34 to Each serving/day increase in intake of fruits or
(2001). 59 years of age, followed for 14 years, vegetables was associated with a 4% lower risk
Determine the dose 1127 CHD cases; 42,148 men, 40 to for CHD. Each serving significantly reduced
response of specific 75 years; follow-up of 8 years, 1063 CHD risk for green leafy vegetables by 23% and
fruits and vegetables cases (multivariate adjusted) for vitamin C-rich fruit and vegetables by 6%
on CHD risk (the
Nurses’ Health Study
and the Health
Professionals’
Follow-Up Study;
US) [98]
Systematic Review
Hartley et al. (2013). 4 RCTs; 970 participants; study The pooled data showed that increased fruit and
Systematically duration ranged from 3 months to vegetables intake insignificantly lowered LDL-C
estimate the effect of 1 year by 6.5 mg/dL, HDL-C by 0.4 mg/dL, and TG by
increasing fruit and 9.0 mg/dL compared to low fruit and vegetable
vegetable intake on control diets
fasting lipids in
healthy adults and
those at high risk of
CVD/CHD (Cochrane
Systematic Review)
[30]
Representative RCTs
McEvoy et al. Parallel RCT: There was a borderline significant dose-response
(2015). 93 overweight adults (BMI of effect of increasing fruit and vegetable intake on
Examine the 27–35 kg/m2) with habitually low fruit lowering LDL-C (p-trend =0.06). There were no
dose-response effect and vegetable intake (≤160 g/day) and changes in HDL-C, TG, or CRP levels with
of fruit and vegetable an elevated risk of developing CHD increasing fruit and vegetable intake
intake on CHD risk (estimated ≥20% over 10 years).
factors (Northern Subjects were randomly assigned to
Ireland) [99] consume 2, 4, or 7 portions fruit and
vegetables (equivalent to 160 g, 320 g,
or 560 g of fruits and vegetables
daily); 12 weeks
Wang et al. (2015). Crossover RCT: Compared with baseline, a one avocado/day diet
Evaluate the effect of 45 overweight/obese subjects; 3 diets: significantly reduced LDL-C by 13.5 mg/dL and
a Hass avocado fruit moderate-fat diets including one fresh non-HDL-C by 14.6 mg/dL compared to the
on CHD biomarkers Hass avocado; a moderate-fat moderate MUFA diet reducing LDL-C by
(US) [100] (MUFA) diet avocado free; and a 8.3 mg/dL and non-HDL-C by 8.7 mg/dL, and
lower-fat diet; 5 weeks with 2 week lower fat diet reducing LDL-C by 7.4 mg/dL and
washout non-HDL-C by 4.8 mg/dL. Further, only the
avocado diet significantly reduced LDL, small
dense LDL-C, and the ratio of LDL/HDL from
baseline
Rayn-Haren et al. Crossover RCT: Whole apples lowered LDL-C by 6.7% compared
(2013). 23 healthy subjects evaluated the to 2.2% for apple juice. There were no significant
Compare the effects of fasting lipid and CRP lowering effects changes in HDL-C or CRP. Fiber was identified
whole apples vs. apple of apples and apple juice over 4 weeks as a primary component for whole apple
juice on fasting lipids cholesterol lowering effects
(Denmark) [101]
Choi et al. (2013). Parallel RCT: Triglycerides concentrations were significantly
Investigate whether 100 volunteers; age 23 years; BMI 21; reduced with low kimchi by 6.1 mg/dL and high
serum lipids are 2 dietary groups: 15 g/day vs. 210 g/ kimchi by 7.4 mg/dL. Total cholesterol was
influenced by the day of kimchi intake; 7 days significantly lowered with low kimchi by 6.8 mg/
amount of kimchi dL and high kimchi by 8.9 mg/dL. Serum total
intake (Korea) [102] antioxidant status was significantly increased
dose-dependently in low kimchi by 5.2% and
high kimchi by 7.5%. Fasting blood glucose was
significantly decreased in the high kimchi intake
Watzl et al. (2005). Parallel RCT: Systemic CRP was significantly reduced in the
Investigate the effects 63 non-smoking men; mean age subjects who consumed 8 servings/day of fruits
of low, medium, and 31 years; mean BMI 24 years; run-in and vegetables compared with those who
high intakes of fruits diet with ≤2 servings/day of fruits and consumed 2 servings/day
and vegetables on vegetables for 4 weeks; diet 2, 5, or 8
markers of immune fruits and vegetables servings;
functions, including 4 weeks
nonspecific markers
of inflammation
(Germany) [103]
green leafy vegetables and β-carotene and vita- and TG by 9.0 mg/dL, which were not statisti-
min C rich fruits and vegetables were among the cally significant [30]. A dose-response RCT
most effective in protecting against CHD risk found a borderline significant effect of increasing
(Fig. 12.9) [95, 98]. A prospective Dutch study self-selected fruit and vegetable intake on reduc-
found a 24% reduction in CHD risk for a daily ing LDL-C (p-trend = 0.06) with no change in
25 g increase in orange fruits and vegetables HDL-C, TG, or CRP with increased intake [99].
compared to only 2% CHD reduction for all A similar dose response RCT showed significant
fruits and vegetables [96] and raw forms of fruits reduced CRP with increased fruit and vegetable
and vegetables were found to be more effective portions [103]. RCTs showed avocados, apples
in lowering CHD risk than processed forms and kimchi to significantly reduce blood lipids
(Fig. 12.10) [97]. and lipoproteins to healthy levels [100–102].
12.2.2.2 Randomized Controlled

Trials 12.2.3 Legumes
RCTs on the effects of fruits and vegetables on
CHD biomarkers are inconsistent because of the Legumes, including pulses (e.g., pinto beans,
wide variability in composition, processing, split peas, lentils, chickpeas) and soybeans, are
amounts consumed, subject variability, clinical rich in fiber and protein with relatively low glyce-
designs and compliance variations [30, 99–103]. mic response properties [14, 104]. A serving of
A Cochrane Systematic Review (4 RCTs; 970 legumes is 1⁄2 cup or 90–100 g cooked legumes,
subjects; duration 3–12 months) found limited which contains 5–10 g of fiber and 7–8 g of pro-
evidence for the beneficial effects of fruits and tein. Most legumes contain <5% of energy as fat,
vegetables on fasting lipids with mean lowering with the exception of chickpeas and soybeans
of LDL-C by 6.5 mg/dL, HDL-C by 0.4 mg/dL, which have 15% and 47% energy from fat,
Raw Fruits and Vegetables
Beta-Carotene Rich Vegetables
Tomatoes
Potatoes
Onions
Green Leafy Vegetables
Watermelon
Grapes
100% Fruit Juices
Dried Fruit
Citrus Fruit
Berries
Bananas
Apples and Pears
–30 –25 –20 –15 –10 –5 0

% CHD Risk Reduction (High vs Low Intake)
Fig. 12.7 Associations between specific fruits and vegetables and coronary heart disease (CHD) risk from 2017 meta-
analysis of 66 cohort studies (adapted from [89])
respectively. Legumes contain nutritionally 12.2.3.1 Non-Soy Legumes

important amounts of the B vitamins and miner- The prospective studies on CHD risk and RCTs
als, such as iron, calcium and potassium. They on CHD related biomarkers are summarized for
also contain bioactive phytochemicals such as non-soy legumes in Table 12.3 [106–114].
phenolics, saponins and isoflavones (especially
in soya foods). Legumes are often consumed as a Prospective Cohort Studies
lower energy dense, lower saturated fat, and high A meta-analysis (5 cohort studies; 200,000
fiber meat or milk replacer. Legume consumption adults) and one large prospective study (9632
has been in decline with the global shift to subjects; mean baseline age 48 years at baseline;
Western-style diets [105]. For example, between 19 years of follow-up) support the benefits of
the 1960s and 1990s, legume intake decreased by consuming 100–130 g of non-soy legumes such
40% in India and by 24% in Mexico. Legumes as beans, peas, lentils ≥4 days per week for
are infrequently consumed by North Americans reducing the risk of CHD by 14–22% compared
and northern Europeans, with <8% of Americans to <1 time per week, after adjusting for known
consuming them on any given day. CVD risk factors [106, 107].
Canned Fruits
Vitamin C Rich Fruits & Vegetables
Raw Vegetables
β-Carotene Rich Fruits & Vegetables
Carrots
Tomatoes
Potatoes
Cruciferous Vegetables
Green Leafy Vegetables
Broccoli
Grapes
Dried Fruit
100% Citrus Fruit Juice
Citrus Fruits
Berries
Apples and Pears
–30 –20 –10 0 10 20 30

% CVD Risk Reduction (High vs Low Intake)
Fig. 12.8 Associations between specific fruits and vegetables and cardiovascular disease (CVD) risk from 2017 meta-
analysis of 64 cohort studies (adapted from [89])
Randomized Controlled Trials been shown in several RCTs to significantly

Non-soy legume RCTs find consistent beneficial lower TC and LDL-C compared to usual or
effects on fasting lipid profiles and modest effects wheat-based diets [113, 114].
on CRP levels compared to control diets C-Reactive Protein. A meta-anlysis (8 RCTs;
[108–114]. 480 subjects; duration 4–52 weeks) found that
Fasting Lipid Profiles. A meta-analysis (26 non-soy legumes lowered mean CRP by 0.21 mg/L
RCTs; 1037 subjects; median age 51 years; (p = 0.068) [108]. In obese adults, hypocaloric
median dose 130 g/day; duration ≥3 weeks) indi- legume supplemented diets significantly lowered
cated that 130 g of pulses/day can significantly CRP levels compared to legume free hypocaloric
lower mean LDL-C by 6.6 mg/dL (5%) com- diets, which may be related to the 2.5% weight
pared to control diets [109]. A systematic review loss in the legume group after 8 weeks [112].
showed that daily consumption of 2 servings of
pulses (150 g/day) by overweight men and 12.2.3.2 Soy Products
women with a mean age of 60 years, lowered The prospective studies on CHD risk and RCTs
both TC and LDL-C by about 8% after 2 months on CHD related biomarkers are summarized for
[110]. Chickpeas (100–120 g cooked/day) have soy products in Table 12.4 [34, 115–124].
Vitamin C Rich F/Vs β-Carotene Rich F/Vs Green Leafy Vegetables

1.1
1
Relative Risk of CHD
0.9
0.8
0.7
0.2 0.5 0.7 1 1.1 to 1.5
Daily Servings
Fig. 12.9 Association between specific fruits and vegetables (F/Vs) and coronary heart disease (CHD) risk in men and
women (all p < 0.003 after multivariate adjusted) (adapted from [95])
Raw F/V (p-trend =0.04) Processed F/V (p-trend =0.08)

1.1
Hazard Ratio for CHD Incidence
0.9
0.8
0.7
0.6
1 2 3 4
Fruit and Vegetable Intake (Quartiles)
Fig. 12.10 Association between the level of raw and processed fruits and vegetables (F/V) on coronary heart disease
(CHD) risk (adapted from [97])
Prospective Cohort Studies Randomized Controlled Trials

Two prospective cohort studies show that soy RCTs show that soy products promote improved
product intake is associated with lower CHD risk fasting lipid profiles compared to control soy-
compared to soy free diets [116, 117]. In a Japanese free diets [34, 117–124].
study, women consuming a high intake of soy Systematic Reviews and Meta-analyses.
foods (> 5 times/week) compared to ≤2 times/ There are four meta-analyses and systematic
week, reduced myocardial infarction risk by 45%, reviews on soy and CHD biomarkers [34, 117–
but no risk reduction was observed in men [115]. 119]. A 2015 meta- analysis (35 RCTs; 2670
In the Shanghai Women’s Health Study high soy subjects; 82% women; 4–52 weeks; average
protein intake (>11 g/day) compared to low intake dose 4–18 weeks) showed strong evidence that
(<4.5 g/day) reduced CHD risk by 75% and myo- soy products resulted in significant mean reduc-
cardial infarction by 86% [116]. tions in serum LDL-C by 4.8 mg/dL, TG by
Table 12.3 Summaries of non-soy legumes studies on CHD risk and biomarkers
Afshin et al. (2014). 5 prospective cohort studies; 200,000 Four weekly 100-g servings of non-soy
Systematically estimate the adults; 514 ischemic heart disease legumes was associated with a mean
effect of non-soy legume cases lower ischemic heart disease risk by 14%
intake on ischemic heart
disease risk [106]
Bazzano et al. (2001). 9632 men and women; mean baseline Non-soy legume intake was significantly
Examine the effect of age 49 years; 60% women; follow-up and inversely associated with reduced
non-soy legume intake on of 19 years; 1800 CHD cases and CHD and CVD risk, after adjusting for
CHD and CVD risk 3680 CVD cases (multivariate known CVD risk factors. Non-soy legume
(National Health and adjusted) intake (≥ 4 times/week vs < 1 time/week)
Nutrition Examination reduced CHD risk by 22% and CVD risk
Survey Epidemiological by 11%
Follow-up Study; US) [107]
Salehi-Abargouei et al. 8 RCTs; 480 normal, overweight and Non-soy legume intake showed a small
(2015). obese adults; duration 4–52 weeks but significant trend towards decreasing
Systematically determine the CRP by a mean of 0.2 mg/L
effect of non-soy legume
intake on systemic CRP
levels [108]
Ha et al. (2014). 26 RCTs; 1037 subjects, middle-age, The median intake of 130 g cooked
Systematically evaluate the normo- and hyperlipidemic adults, at pulses/day significantly lowered mean
effect of pulses (beans, moderate risk of coronary artery LDL-C levels 6.6 mg/dL (5%) compared
chickpeas, lentils and peas) disease; ≥ 3 weeks to control non-pulse diets.
on fasting lipids [109]
Bazzano et al. (2011). 10 RCTs; 268 participants; ≥ 3 weeks Legume rich diets significantly mean
Systematically evaluate the lowered TC by 12 mg/dL and LDL-C by
effects of non-soy legumes 8.0 mg/dL compared to control diets
intake on fasting lipids [110]
Representative RCTs
Abeysekara et al. (2012). Crossover RCT: Compared with the regular diet, the
Determine the effects of 108 subjects; mean age 60 years; pulse-based diet significantly decreased
pulse-based diets on the mean weight 76 kg; randomized to TC by 8.3% and LDL-C by 7.9%.
fasting lipids and cardio- consume 2 servings daily of beans, Subjects consumed 36% higher fiber (69%
metabolic health of older chickpeas, peas or lentils (about insoluble and 31% soluble) intake while
adults (Canada) [111] 150 g/day dry weight) or their regular eating the pulse-based diet compared with
diet (pulse-free); 2 months; 1 month the regular diet
washout
Hermsdorf et al. (2011). Parallel RCT: The legume diet significantly reduced TC,
Study the effects of 30 obese subjects; mean BMI 32.5; LDL-C and CRP, and promoted weight
hypocaloric diets with and mean age 36 years; 60% male; loss by 2.5% compared to the control diet
without legumes on CRP randomized into a calorie restricted
responses (Spain) [112] legume free diet or calorie restricted
legume based diet including 4 weekly
different cooked legume servings
(160–235 g chickpeas, lentils, peas
and beans); 8 weeks
(continued)
Pittaway et al. (2008). Sequential RCT: Chickpeas significantly lowered fasting
Evaluate the effects of 45 free-living adults; 13 TC by 7.7 mg/dL and LDL-C by 7.3 mg/
chickpea enriched diets on premenopausal women, 19 dL, compared to the chickpea-free diet.
ad libitum nutrient intake, postmenopausal women, 13 men; There was also a small loss of weight by
body weight, fasting lipids, mean age was 52 years; mean BMI 26; 0.45 kg (p = 0.07) during the chickpea
and other cardiometabolic 120 g cooked chickpeas/day added to phase, which was thought to be related to
changes (Australia) [113] the habitual diet for 12 weeks, the increased fiber intake
followed by 4 weeks of habitual diet
without chickpeas
Pittaway et al. (2006). Crossover RCT: The chickpea group had significantly
Examine the effects of 47 adults; randomized to groups lower TC by 3.9% and LDL-C by 4.6%,
chickpea diets on fasting consuming chickpeas or wheat; compared to the wheat group
lipids (Australia) [114] 5 weeks
Table 12.4 Summaries of soy product studies on CHD risk and biomarkers
Kokubo et al. (2007). 40,462 Japanese adults; mean For women, there was a reduced risk of
Assess the effect of soy and baseline age 50 years; BMI 23; cerebral infarctions by 36% and myocardial
isoflavone intake on risk of average follow-up of 12.5 years; 587 infarctions by 45% for soy intake ≥5 times/
cerebral and myocardial cerebral infarctions cases, 308 week versus ≤2 times/week. No significant
infarctions (The Japan myocardial infarctions cases associations were observed in men
Public Health Center–Based (multivariate adjusted)
Prospective Study) [115]
Zhang et al. (2003). 75,000 Chinese women; mean There was a significant reduction in
Investigate the effect of soy baseline age 51 years; mean BMI adjusted CHD risk by 75% for women
foods on CHD risk in 24; mean of 2.5 years of follow-up; consuming high intake of soy protein foods
Chinese women (The 62 cases of CHD (multivariate (>11 g/day) vs. low intake (<4.5 g/day).
Shanghai Women’s Health adjusted) Also, there was a significant 86% lower risk
Study) [116] of nonfatal MI for the highest vs. the lowest
quartile of soy intake
Meta-Analyses and Reviews
Tokede et al. (2015). 35 RCTs; 2670 adults; 82% women; Intake of soy products significantly reduced
Systematically estimate the average dose 30 g/day (range serum LDL-C by 4.8 mg/dl, TG by 4.9 mg/
clinical effect of soy 14–50 g/day); mean duration dl, and TC by 5.3 mg/dl and significantly
products on fasting lipids (4 weeks to 1 year) increased serum HDL by 1.4 mg/dl. LDL
[117] reductions were more marked in
hypercholesterolemic subjects by 7.5 mg/dl,
than in healthy subjects by 3.0 mg/dl.
LDL-C lowering was stronger for whole
soy products (soy milk, soybeans and nuts)
by 11 mg/dL vs. soy extracts by 3.2 mg/dL
Anderson et al. (2011). 43 RCTs (20 parallel and 23 Soy protein significantly lowered LDL-C by
Systematically evaluate the crossover studies); median intake 5.5% in parallel studies and 4.2% in
clinical effects of soy foods 30 g of soy protein daily; crossover studies vs. control diets. Also,
and/or protein on fasting 4–18 weeks HDL-C and TG levels were 3.2% higher
lipids [34] and 10.7% lower, respectively, with soy vs.
control diets
Yang et al. (2011). 8 RCTs; soy protein (with soy Soy products significantly lowered TC by
Systematically review the isoflavones) or soy products (tofu, 16 mg/dL, LDL-C by 12 mg/dL, and TG by
effects of soy product intake miso or soy nuts); 6 weeks-4 years 19 mg/dL along with a significantly
on fasting lipid profiles and increasing HDL-C by 2 mg/dL. There were
glycemic control in type 2 no significant effects on fasting glucose,
diabetes patients [118] insulin and glycated hemoglobin
Sacks et al. (2006). 22 RCTs; fasting lipids and other The increased intake of soy products by an
Systematically review RCTs CHD risk factors; 4 weeks −1 year average of 50 g/day decreased LDL-C
on the effect of soy protein levels by 3%. No significant effects on
products on fasting lipids HDL-C, TG, or blood pressure were
[119] reported
Representative RCTs
Padhi et al. (2015). Double-blind, Parallel RCT: Consuming 12.5 or 25 g protein from
Determine potential dose 243 hypercholesterolemic adults; defatted soy flour incorporated into muffins
response effects of soy mean age 55 years; mean BMI 28; did not reduce LDL-C or other CHD risk
protein on LDL-C in randomly assigned to muffins with factors in hypercholesterolemic adults
hypercholesterolemic adults 25 g soy protein, 12.5 g soy protein, compared to similar amounts of whey
(Canada) [120] 12.5 g whey protein, or 25 g whey protein
protein daily; 6 weeks while on a
self-selected diet
Azadbakht et al. (2007). Crossover RCT: CRP was reduced in the soy nut diet by
Evaluate the effects of soy 42 postmenopausal, metabolic 8.9% and soy protein diet by1.6% compared
intake on inflammatory syndrome women; randomly to the control DASH diet. The soy nut diet
markers (US) [121] assigned to a control DASH diet, or significantly reduced IL-18 compared with
one of 2 modified DASH diets, where the control diet by 9.2%
a serving of red meat was replaced by
soy protein, or soy nuts; 8 weeks
Jenkins et al. (2002). Parallel RCT: Soy diets resulted in significantly lower TC,
Assess the effects of soy 41 hyper-lipidemic men and estimated coronary artery disease risk, and
foods on both lipid and postmenopausal women; LDL-C to HDL-C ratio compared to the
nonlipid risk factors for randomized to 3 diets: a low-fat low-fat dairy control diet. No significant
coronary artery disease dairy food control diet; 50 g soy differences were seen between the high and
(Canada) [122] protein diet (73 mg isoflavones low-isoflavone soy diets. Soy diets
daily); or 52 g soy protein diet significantly lowered coronary artery
(10 mg isoflavones daily); 1 month disease risk by 10% compared to low fat
diary diets
Gardner et al. (2001). Double-blind, Parallel RCT: Soy protein with isoflavones significantly
Determine the effect of soy 94 postmenopausal, moderately lowered LDL-C compared to soy protein
protein and isoflavones on hypercholesterolemic women, mean without isoflavones, but neither group
fasting lipids (US) [123] age about 60 years; diets significantly differed from the milk group
supplemented with 42 g protein/day)
diets: (1) milk protein, (2) soy
protein (no isoflavones), or (3) soy
protein (80 mg isoflavones);
12 weeks
Ashton et al. (2000). Crossover RCT: Tofu significantly reduced TC by 9 mg/dL,
Investigate the effect on 42 healthy men; age range TG by 13 mg/dL, and HDL-C by 3 mg/dL
fasting lipids of replacing 35–62 years; randomized into compared to the lean meat
lean meat with tofu isocaloric and isoprotein diets with
(Australia) [124] lean meat (150 g/day) vs. tofu
(290 g/day); 1 month
4.9 mg/dL, and total cholesterol (TC) by 5.3 mg/ 12.2.4 Nuts (Tree Nuts and Peanuts)
dL. There was also a significant increase in and Seeds
serum HDL-C by 1.4 mg/dL [117]. Also, LDL-C
reductions were more marked in hypercholes- Nuts and seeds are rich sources of CHD protec-
terolemic subjects by 7.5 mg/dL than in healthy tive macronutrients, micronutrients and phyto-
subjects by 3.0 mg/dL. LDL-C reduction was chemicals [130–132]. The consumption of a
stronger when whole soy products (soy milk, daily handful of nuts (approx. 45–60 g/day) can
soybeans and nuts) were used as the interven- help to prevent or reduce weight gain, cardio-
tion by 11 mg/dL, as compared to soy extracts, metabolic dysfunction and related chronic dis-
which lowered LDL-C by 3.2 mg/dL. Three ear- eases, especially if eaten as a replacement for
lier meta-analyses or systematic reviews report less healthful foods [29, 133, 134]. They are
similar findings [34, 118, 119]. compositionally low in saturated fat and sugar,
Representative RCTs. Five representative rich in unsaturated fats, protein and fiber, and
RCTs illustrate the effects of soy intake on CHD contain a variety of healthy vitamins, minerals
biomarkers [120–124]. Three RCTs in post- and phytonutrients including carotenoids, poly-
menopausal women all find beneficial effects for phenols, and phytosterols [130, 131]. The fatty
soy products in reducing CHD risk and associ- acid composition of nuts is of a healthy type
ated biomarkers including CRP, TC, and LDL-C since the saturated fat levels are low (range
[121–123]. In a study of hypercholesterolemic 4–16%) with 84–96% of the total fat content
men and postmenopausal women, 50 g/day of made up of unsaturated fat, mostly monounsatu-
soy protein reduced coronary artery disease rated fatty acids (MUFAs) and polyunsaturated
(CAD) risk by 10% compared to a low-fat diet fatty acids (PUFAs), predominantly linoleic
without soy protein [122]. The substitution of acid. Walnuts and flaxseed are rich sources of
tofu for lean meat significantly improved the alpha -linolenic acid, the plant n-3 fatty acid.
fasting lipid profile [124]. In contrast, consum- Nuts and seeds are good sources of protein and
ing 12.5 or 25 g protein from defatted soy flour fiber and/or lignans per 28 g serving. Despite
incorporated into muffins did not reduce LDL-C their relatively high Atwater metabolizable
or other any other CHD risk factors in hypercho- energy values of 5–6 kcal/g, they often have a
lesterolemic adults compared to matched levels 10–25% lower net metabolizable energy than the
of whey protein [120]. Health claims were label value [135–137]. Overall, nut consumers
approved by the FDA for soy products and have a significantly higher nutrient quality score
reduced LDL-cholesterol and CHD risk in 1999 than non-nut consumers, which can lead to over-
[125] but not by the European Food Safety all better health and wellness [138]. However, in
Authority (EFSA) [126]. Equol, a bioactive the US slightly over 60% of men and women do
metabolite of the soy isoflavone daidzein was not consume any nuts on a given day and
found to be inversely associated with risk of only14.4% of men and 11.8% of women con-
CHD in women [127]. Although soy foods have sume ≥1.5 ounces of nuts/daily, which is similar
become controversial because of concerns based to global nut intake [139].
primarily on animal studies that their uniquely Table 12.5 summarizes prospective studies
rich phytoestrogen (isoflavone) content may be and RCTs on the effects of nuts and seeds on
contraindicated for breast cancer patients and CHD risk and CHD biomarkers [140–154]. The
women at high risk of developing breast cancer, US FDA approved several qualified health claims
current clinical and epidemiological evidence for tree nuts and peanuts and CHD risk reduction
does not support an increased breast cancer risk with the consumption of at least 42 g (1.5 ounces)/
from soybean isoflavones intake [128, 129]. day [132].
Table 12.5 Summaries of nut and seed studies on CHD risk and biomarkers
Systematic Reviews and Meta- and Pooled Analysis
Mayhew et al. (2016). Total CHD incidents: 3 studies; 123,971 Higher intake of nuts reduced CHD
Systematically review the (87,869 women); followed up of events risk by 32% and reduced the risk
literature and quantify 6–26 years; 4757 CHD events. CHD of CHD deaths by 30%
associations between nut mortality: 7 studies; 278,584 participants
intake and CHD (180,734 women); follow-up of
outcomes and all-cause 5.4–30 years, 8454 deaths
mortality [140]
Ma et al. (2014). 13 prospective studies; 347,477 A linear dose-response inverse
Systematically assess participants; follow-up 5–30 years; 6127 relationship was reported between nuts
dose-response effects coronary artery disease cases and coronary artery disease risk, with a
between the intake of 5% lower risk for each nut serving/week
nuts and coronary artery with a significant lower risk observed for
disease [141] ≥2 nut servings per week. For highest
vs. lowest nut intake, coronary artery
disease risk was reduced by 34%
(Fig. 12.11)
Luu et al. (2015). 3 large prospective studies; 206,029 adults; A higher intake of peanuts and nuts
Examine the effect of primarily peanut consumers; significantly lowered ischemic heart
peanut and tree nut intake 5.4–12.2 years of follow-up; 14,440 deaths disease mortality risk by 30–40%
on CVD and ischemic confirmed
heart disease mortality
risk (Southern
Community Cohort
Study, Shanghai Women’s
Health Study and the
Shanghai Men’s Health
Study) [142]
Kelly and Sabate (2006). 4 large studies; 173,000 adults; followed CHD risk was lowered by 37% for those
Assess the effect of nuts for 6–17 years consuming nuts ≥4 times/week
on CHD risk (US compared to those who never or seldom
Adventist Health, Iowa consumed nuts
Women’s Health, Nurses’
Health, and the
Physicians’ Health
studies [143]
Meta-analyses and Systematic Reviews
Del Gobbo et al. (2015). 61 trials; 2582 subjects; 3–26 weeks Tree nut intake (per serving/day)
Systematically assess the lowered TC by 4.7 mg/dL, LDL-C by
clinical effects of tree 4.8 mg/dL, ApoB by 3.7 mg/dL, and TG
nuts (walnuts, pistachios, by 2.2 mg/dL with no statistically
macadamia nuts, pecans, significant effects on other outcomes.
cashews, almonds, The dose-response between nut intake
hazelnuts, and Brazil and TC and LDL-C was nonlinear with
nuts) on blood lipids stronger effects observed for ≥60 g nuts/
[144] day. The nut dose rather than nut type is
the major determinant of cholesterol
lowering
(continued)
Sabate et al. (2010). 25 trials; 583 normo-and A mean daily intake of 67 g (2.4 ounces)
Systematically evaluate hypercholesterolemic adults from 7 of nuts significantly reduced TC by
the effects of nut countries (not taking lipid-lowering 5.1%, LDL-C by 7.4%, and LDL-C/
consumption on fasting medications); range 3–8 weeks HDL-C by 8.3% compared to control
lipids [145] diets. TG levels were significantly
reduced by 10.2% in subjects with blood
TG levels of ≥150 mg/dL but not in
those with lower levels. Tree nuts
(almonds, walnuts, pistachios, hazelnuts,
pecans, and macadamias) had similar
effects on fasting lipid levels. The
lipid-lowering effects of nuts were
greatest among subjects with high
baseline LDL-C and among those
consuming Western diets
Banel et al. (2009). 13 RCTs; 365 participants; walnuts Compared with control diets, diets with
Systematically assess the providing 10–24% of total calories; walnuts significantly lowered TC by
effect of walnuts on blood duration of 4–24 weeks 10 mg/dL and LDL-C by 9.2 mg/
lipids and CRP [146] dL. HDL-C and TG were not
significantly affected by walnut diets
more than control diets. Results for CRP
were inconsistent
Phung et al. (2009). 5 RCTs; 142 subjects; almond Almonds significantly lowered TC by
Systematically evaluate consumption ranging from 25 to 168 g/ 7.0 mg/dL and borderline significantly
the clinical effects of day; 4 weeks lowered LDL-C by 5.8 mg/dL vs.
almonds on fasting lipids control diets. No significant effects on
[147] HDL-C, TG, or LDL: HDL ratio were
reported
Griel et al. (2006). 22 RCTs on the effects of tree nut Almonds, walnuts, pistachio nuts, hazel
Systematically review the consumption on fasting lipids nuts, pecans, and macadamia nuts
clinical evidence on tree lowered LDL-C by11 mg/dL compared
nuts and fasting lipids to the control diets due to healthy fatty
[148] acid profiles, fiber and phytosterol
content
Representative RCTs
Le et al. (2016). Parallel RCT (Walnuts): The walnut-rich diet resulted in the most
Examine the effects of 245 overweight and obese women; favorable changes in lipid levels while
walnuts on the plasma behavioral weight loss intervention; still associated with a degree of weight
lipid profile in overweight subjects randomly assigned to: (1) lower loss that was comparable to the lower fat
and obese women fat (20% energy), higher carbohydrate diet. TG decreased in all study arms at
participating in a (65% energy) diet; (2) a lower 6 months (p < 0.05). The walnut-rich
behavioral weight loss carbohydrate (45% energy), higher fat diet increased HDL-C more than the
intervention (US) [149] (35% energy) diet; or (3) a walnut-rich, lower fat or lower carbohydrate diet
higher fat (35% energy), lower (P < 0.05). Walnuts also improved
carbohydrate (45% energy) diet; 6 months insulin sensitivity and C-reactive protein
levels
Berryman et al. (2015). Crossover RCT (Almonds): The almond diet significantly decreased
Evaluate the effects of 48 subjects with moderately elevated non-HDL-C by 6.9 mg/dL, LDL-C by
fasting lipid and LDL-C; randomized into a healthy diet 5.3 mg/dL and HDL-C by 1.7 mg/dL
cardiometabolic effects of with 42 g (1.5 oz.) of almonds/day or an compared to the control muffin diet.
almonds in a controlled- identical diet with an isocaloric muffin Also, almond consumption significantly
feeding setting (US) substitution (no almonds/day); diets did reduced abdominal fat by 0.07 kg and
[150] not differ in saturated fat or cholesterol but leg fat by 0.12 kg, despite no differences
the almond group had more unsaturated fat in total body weight
and 3 g/day more fiber vs. control group;
6 weeks
Edel et al. (2015). Double-blind, Parallel RCT (Flaxseed): Flaxseed ingestion resulted in an
Investigate the effects of 100 peripheral artery disease subjects; diet 11–15% reduction of fasting TC and
flaxseed on fasting lipids supplemented with 30 g of milled flaxseed LDL-C compared with placebo after
in patients with statin or 30 g of whole-wheat; plasma lipids were 1 month into the trial (p = 0.05), which
usage and clinically measured at 0, 1, 6, and 12 months was maintained for 6 months. HDL-C
significant CVD (Canada) and TG did not significantly change.
[151] Also, flaxseed promoted additional
LDL-C lowering capabilities when used
in conjunction with statins
Richmond et al. (2013). Crossover RCT (Almonds and TC and LDL-C decreased significantly
Evaluate the effects of Sunflower seeds): on both sunflower seed and almond diets
sunflower seed or almond 22 postmenopausal women with type 2 from baseline, with no difference
consumption on fasting diabetes; consumed diets with the addition between diets. Both diets showed
lipids in post-menopausal of 30 g sunflower kernels or almonds/day; clinically beneficial effects on reducing
women with diabetes 3 weeks; 4-week washout CVD risk
(New Zealand) [152]
Nouran et al. (2009). Parallel RCT (Peanuts): Compared with the habitual diet, adding
Assess the effects of 54 hypercholesterolemic men; randomly peanuts to the habitual diet significantly
peanuts on lipid profiles assigned to: (1) peanuts (about 77 g/day) increased HDL-C by 6.1 mg/dL and
and oxidized LDL-C in added to their habitual diet or (2) their significantly reduced the LDL/HDL-C
men (Iran) [153] habitual diet; 4 weeks ratio by 0.7, but there was an
insignificant effect on TC and LDL-
C. Peanut diets reduced oxidized LDL-C
by 1 mU/L (p = 0.07)
Sabate et al. (1993). Crossover RCT (Walnuts): The walnut enriched NCEP Step 1 diet
Study the effects of 18 men consuming an isocaloric NCEP significantly lowered TC by 12%,
walnuts on serum lipids Step 1 diet with and without 20% of LDL-C by16%, HDL-C by 5%, LDL-C
(US) [154] energy from walnuts replacing common to HDL-C ratio by 12%, and
saturated fat-rich foods triglycerides by 8% compared to the
walnut free NCEP Step 1 diet
12.2.4.1 Prospective Cohort Studies significantly lowered ischemic heart disease

Four systematic reviews and meta- and pooled mortality risk by 30–40% [142]. A 2014 dose-
analyses of prospective cohort studies consis- response meta-analysis (13 prospective studies;
tently support a CHD protective role for frequent 347,477 participants; follow-up 5–30 years)
nuts intake [140–143]. A 2016 systematic review found a linear dose-response inverse relationship
(3 CHD incidents cohort studies, 123,971 sub- between nuts and coronary artery disease risk,
jects, followed up of 6–26 years; and 7 CHD with a 5% lower risk for each nut serving/week
mortality cohort studies; 278,584 participants, with a significant lower risk observed for ≥2 nut
follow-up of 5.4–30 years) found that higher servings per week [141]. Higher nut intake was
intake of nuts reduced CHD events risk by 32% shown to significantly reduce coronary artery
and reduced the risk of CHD deaths by 30% disease risk by 34% (Fig. 12.11).
[140]. A 2015 pooled prospective study (3 large A 2006 pooled analysis (4 large prospective
prospective studies including Southern studies: US Adventist Health, Iowa Women’s
Community Cohort Study, Shanghai Women’s Health, Nurses’ Health, and the Physicians’ Health
Health Study and the Shanghai Men’s Health studies; 173,000 adults; followed for 6 to 17 years)
Study; 206,029 adults; primarily peanut consum- found that CHD risk was lowered by 37% for
ers; 5.4–12.2 years of follow-up) showed that those consuming nuts ≥4 times/week compared to
higher nut intake (primarily from peanuts) those who never or seldom consumed nuts [143].
Relative Risk for CAD 0.9
0.8
0.7
0.6
1 2 3 4 5 6
Weekly Nut Servings
Fig. 12.11 Association between the number of weekly nut servings and coronary artery disease (CAD) risk (p-trend
<0.001) (adapted from [141])
12.2.4.2 Randomized Controlled Conclusions

Trials Although there have been some decreasing
RCTs on tree nuts, seeds and peanuts consis- trends in overall CHD mortality rate over the
tently find similar significant effects on reduc- last several decades because of cholesterol low-
ing lipids and non-HDL lipoproteins but the ering drugs and surgical procedures, CHD is
findings are inconsistent for lowering blood still a leading cause of death globally and its
pressure and CRP [144–154]. A 2015 meta- prevalence is expected to increase as the global
analysis of tree nut RCTs (61 trials; 2582 sub- population ages. Healthy behaviors, including
jects; duration 3–26 weeks) reported that each meeting the recommended intake of whole plant
daily serving of nuts significantly lowered total foods, increased physical activity, and weight
TC by 4.7 mg/dL, LDL-C by 4.8 mg/dL, ApoB control, have been shown to reduce the risk for
by 3.7 mg/dL, and triglycerides by 2.2 mg/dL CHD. Prospective cohort studies consistently
with no statistically significant effects shown show that diets with higher intakes of whole
for HDL-C, blood pressure, or CRP [144]. The plant foods including whole-grains, fruits, veg-
dose-response between nut intake and TC and etables, legumes, and nuts and seeds are associ-
LDL-C was nonlinear with stronger effects ated with reduced CHD risk compared to lower
found for the consumption of ≥60 g nuts/day. intake. Heart healthier versions of whole plant
There were no significant differences between foods are higher in dietary fiber, phytosterols,
nut types on lipids and lipoprotein lowering. healthy fatty acids (MUFAs and PUFAs), and
Similar findings were reported in four previous nutrients (e.g., vitamins E and C, potassium and
nut meta-analyses and systematic reviews of folate), phytochemicals (carotenoids, flavonoids
RCTs [145–148]. RCTs on tree nuts, peanuts, and phytosterols) and lower in energy density,
flaxseed, and sunflower seeds demonstrate simi- glycemic index and glycemic load. The risk of
lar fasting lipid lowering effects [149–154]. CHD incidence or mortality is significantly
Nuts and seeds have been shown to have adjunc- reduced with the intake of ≥3 servings/day of
tive therapeutic effects on non-HDL-C lowering whole-grains (especially oats and barley), ≥5
with statins in hypercholesterolemic individuals servings (400 g)/day of fruits and vegetables,
[151, 155]. ≥4 weekly servings (130–150 g cooked) of
Appendix A 363
legumes (both non-soy and/or soy products), of the US population meets the recommended
and ≥5 servings/week of nuts and seeds. RCTs intake levels for whole-grains, fruits, vegeta-
generally support the beneficial effects of bles, legumes, and nuts and 70% exceed recom-
healthy whole plant based foods on CHD risk mended refined grain intake. Approximately
biomarkers including lowering serum lipids and 45% of CHD deaths in the US are associated
blood pressure, improving glucose and insulin with suboptimal low dietary intake of fruits,
metabolism, improving endothelial function, vegetables, nuts and seeds, whole-grains, and
alleviating oxidative stress and inflammation seafood omega-3 PUFAs, and high intake of red
and reducing risk of weight gain compared to and processed meats, sugar sweetened bever-
their refined c ounterparts. Only a small f raction ages, and sodium.
ppendix A. Estimated Range of Energy, Fiber, Nutrients

A
and Phytochemicals Composition of Whole Plant Foods/100 g Edible Portion

Nutrients/ Wheat, oats, Apples, Dates, dried Potatoes, Lentils, Almonds,
Phytochemicals brown rice, pears, figs, apricots, spinach, carrots, chickpeas, Brazil nuts,
whole grain bananas, cranberries, peppers, lettuce, split peas, cashews,
bread, cereal, grapes, raisins, and green beans, black beans, hazelnuts,
pasta, rolls, oranges, prunes cabbage, pinto beans, macadamias,
and crackers blueberries, onions, and soy beans pecans,
strawberries, cucumber, walnuts,
and avocados cauliflower, peanuts,
mushrooms, sunflower
and broccoli seeds, and
flaxseed
Energy (kcals) 110–350 30–170 240–310 10–115 85–170 520–700
Protein (g) 2.5–16 0.5–2.0 0.1–3.4 0.2–5.0 5.0–17 7.8–24
Available 23–77 1.0–25 64–82 0.2–25 10–27 12–33
Carbohydrate (g)
Fiber (g) 3.5–18 2.0–7.0 5.7–10 1.2–9.5 5.0–11 3.0–27
Total fat (g) 0.9–6.5 0.0–15 0.4–1.4 0.2–1.5 0.2–9.0 46–76
SFAa (g) 0.2–1.0 0.0–2.1 0.0 0.0–0.1 0.1–1.3 4.0–12
MUFAa (g) 0.2–2.0 0.0–9.8 0.0–0.2 0.1–1.0 0.1–2.0 9.0–60
PUFAa (g) 0.3–2.5 0.0–1.8 0.0–0.7 0.0–0.4 0.1–5.0 1.5–47
Folate (μg) 4.0–44 <5.0–61 2–20 8.0–160 50–210 10–230
Tocopherols (mg) 0.1–3.0 0.1–1.0 0.1–4.5 0.0–1.7 0.0–1.0 1.0–35
Potassium (mg) 40–720 60–500 40–1160 100–680 200–520 360–1050
Calcium (mg) 7.0–50 3.0–25 10–160 5.0–200 20–100 20–265
Magnesium (mg) 40–160 3.0–30 5.0–70 3.0–80 40–90 120–400
Phytosterols (mg) 30–90 1.0–83 – 1.0–54 110–120 70–215
Polyphenols (mg) 70–100 50–800 – 24–1250 120–6500 130–1820
Carotenoids (μg) – 25–6600 0.6–2160 10–20,000 50–600 0.0–1200
Circulation. 2013;128:553–565
Rodriguez-Casado A. The health potential of fruits and vegetables phytochemicals: notable examples. Crit Rev. Food
Sci Nutr. 2016;56(7):1097–1107
related co-morbidities. Obes Rev. 2014;15: 392–407
An. 1999;12:169–196
tent during maturation and from different growing areas. J Agric Food Chem. 2009;57(21):10,408–10413
Nutrient Database for Standard Reference, Release 27. http://www.ars.usda.gov/nutrientdata. Accessed 17 February
2015
Association Nutrition Committee. Circulation.

References 2006;114:82–96.
10. Satija A, Bhupathiraju SN, Spiegelman D, et al.
1. Mendis S, Puska P, Norrving B. Global atlas on car- Healthful and unhealthful plant-based diets and risk
diovascular disease prevention and control. Geneva: of coronary heart disease in US adults. J Am Coll
World Health Organization; 2011. Cardiol. 2017;70:411–22.
2. Sayols-Baixeras S, Lluís-Ganella C, Lucas G, 11. Ros E, Hu FB. Consumption of plant seeds and car-
Elosua R. Pathogenesis of coronary artery disease: diovascular health epidemiological and clinical trial
focus on genetic risk factors and identification of evidence. Circulation. 2013;128:553–65.
genetic variants. Appl Clin Genet. 2014;7:15–32. 12. USDA. What We Eat in America, NHANES 2011–
3. Roger VL, Go AS, Lloyd-Jones DM, et al. Heart 2012, individuals 2 years and over (excluding
disease and stroke statistics-2011 update: a report breast-fed children). www.ars.usda.gov/nea/bhnrc/
from the American Heart Association. Circulation. fsrg.
2011;123(4):18–209. 13. Rodriguez-Casado A. The health potential of fruits
4. Enkhmaa B, Surampudi P, Anuurad E, et al. Lifestyle and vegetables phytochemicals: notable examples.
changes: effect of diet, exercise, functional food, Crit Rev Food Sci Nutr. 2016;56(7):1097–107.
and obesity treatment, on lipids and lipoproteins. 14. Rebello CJ, Greenway FL, Finley JW. A review of
[Updated 2015 Jun 8]. De Groot LJ, Beck-Peccoz the nutritional value of legumes and their effects on
P, Chrousos G, et al., editors. Endotext [Internet]. obesity and its related co-morbidities. Obes Rev.
South Dartmouth (MA): MDText.com, Inc.; 2000-. 2014;15:392–407.
PMID:26561697. 15. Gebhardt SE, Thomas RG. Nutritive value of foods,
5. American Heart Association. Coronary artery Home and Garden Bulletin 72: U.S. Department of
disease—coronary heart disease. 2015. http:// Agriculture, Agricultural Research Service; 2002.
www.heart.org/HEARTORG/Conditions/More/ 16. Holden JM, Eldridge AL, Beecher GR, et al.
MyHeartandStrokeNews/Coronary-Artery-Disease- Carotenoid content of U.S. foods: an update of the
Coronary-Heart-Disease_UCM_436416_Article. database. J Food Compos Anal. 1999;12:169–96.
jsp. Accessed 14 Jul 2015. 17. Lu Q-Y, Zhang Y, Wang Y, et al. California Hass
6. Fleg JL, Forman DE, Berra K, et al. Secondary avocado: profiling of carotenoids, tocopherol,
prevention of atherosclerotic cardiovascular dis- fatty acid, and fat content during maturation and
ease in older adults a scientific statement from from different growing areas. J Agric Food Chem.
the American Heart Association. Circulation. 2009;57(21):10408–13.
2013;128(22):2422–46. 18. Wu X, Beecher GR, Holden JM, et al. Lipophilic
7. Hu FB. Diet and lifestyle influences on risk of and hydrophilic antioxidant capacities of common
coronary heart disease. Curr Atheroscler Rep. foods in the United States. J Agric Food Chem.
2009;11:257–63. 2004;52:4026–37.
8. Hu FB, Willett WC. Optimal diets for preven- 19. Bhupathiraju SN, Tucker KL. Coronary heart dis-
tion of coronary heart disease. J Am Med Assoc. ease prevention: nutrients, foods, and dietary pat-
2002;288:2569–78. terns. Clin Chim Acta. 2011;412:1493–514.
9. Lichtenstein AH, Appel LJ, Brands M, et al. Diet 20. Appel LJ, Moore TJ, Obarzanek E, et al. A clinical
and lifestyle recommendations revision to the 2006: trial of the effects of dietary patterns on blood pres-
a scientific statement from the American Heart sure. N Engl J Med. 1997;336:1117–24.
References 365
21. Ornish D, Brown SE, Scherwitz LW, et al. 35. Retelny VS, Neuendorf A, Roth JL. Nutrition pro-
Can lifestyle changes reverse coronary heart tocols for the prevention of cardiovascular disease.
disease? The Lifestyle Heart Trial. Lancet. Nutr Clin Pract. 2008;23:468–76.
1990;336(8708):129–33. 36. National Cholesterol Education Program. Detection,
22. Alissa EM, Ferns GA. Dietary fruits and vegetables evaluation, and treatment of high blood choles-
and cardiovascular diseases risk. Crit Rev Food Sci terol in adults (Adult Treatment Third Report
Nutr. 2017;57(9):1950–62. https://doi.org/10.1080/1 of the National Cholesterol Education Program
0408398.2015.1040487. (NCEP) expert panel III) final report. Circulation.
23. Bao Y, Han J, Hu FB, et al. Association of nut con- 2002;106(25):3143–421.
sumption with total and cause-specific mortality. 37. Liu JF, Liu YH, Chen CM, et al. The effect of
N Engl J Med. 2013;369(21):2001–11. almonds on inflammation and oxidative stress in
24. Gupta SK, Sawhney RC, Rai L, et al. Regression of Chinese patients with type 2 diabetes mellitus: a
coronary atherosclerosis through healthy lifestyle in randomized crossover controlled feeding trial. Eur J
coronary artery disease patients -Mount Abu Open Nutr. 2013;52(3):927–35.
Heart Trial. Indian Heart J. 2011;63(5):461–9. 38. Kerckhoffs DAJM, Brouns F, Hornstra G, Mensink
25. Ye EQ, Chacko SA, Chou EL, et al. Greater whole- RP. Effects on the human serum lipoprotein pro-
grain intake is associated with lower risk of type 2 file of β-glucan, soy protein and isoflavones, plant
diabetes, cardiovascular disease, and weight gain. sterols and stanols, garlic and tocotrienols. J Nutr.
J Nutr. 2012;142:1304–13. 2002;132:2494–505.
26. Lefevre M, Jonnalagadda S. Effect of whole grains 39. Wong JM, de Souza R, Kendall CW, et al. Colonic
on markers of subclinical inflammation. Nutr Rev. health: fermentation and short chain fatty acids.
2012;70(7):387–96. J Clin Gastroenterol. 2006;40:235–43.
27. Bertoia ML, Mukamal KJ, Cahill LE, et al. Changes 40. Jiao J, Xu J-Y, Zhang W, et al. Effect of dietary fiber
in intake of fruits and vegetables and weight change on circulating C-reactive protein in overweight and
in United States men and women followed for up obese adults: a meta-analysis of randomized con-
to 24 years: analysis from three prospective cohort trolled trials. Int J Food Sci Nutr. 2015;66(1):114–9.
studies. PLoS Med. 2015;12(9):e1001878. https:// 41. Ma Y, Olendzki BC, Wang J, et al. Single-component
doi.org/10.1371/journal.pmed.1001878. versus multicomponent dietary goals for the meta-
28. Hollænder PLB, Alastair B, Ross AB, Kristensen bolic syndrome. A randomized trial. Ann Intern
M. Whole-grain and blood lipid changes in appar- Med. 2015;162(4):248–57.
ently healthy adults: a systematic review and meta- 42. Rajaram S, Connell KM, Sabate J. Effect of almond-
analysis of randomized controlled studies. Am J Clin enriched high-monounsaturated fat diet on selected
Nutr. 2015;102(3):556–72. https://doi.org/10.3945/ markers of inflammation: a randomised, controlled,
ajcn.115.109165. crossover study. Br J Nutr. 2010;103:907–12.
29. Kris-Etherton PM, Hu FB, Ros E, Sabate J. The role 43. McKiernan F, Lokko P, Anna Kuevi A, et al. Effects
of tree nuts and peanuts in the prevention of coro- of peanut processing on body weight and fasting
nary heart disease: multiple potential mechanisms. plasma lipids. Br J Nutr. 2010;104:418–26.
J Nutr. 2008;138:1746S–51S. 44. Esposito K, Marfella R, Ciotola M, et al. Effect of
30. Hartley L, Igbinedion E, Holmes J, et al. Increased a Mediterranean-style diet on endothelial dysfunc-
consumption of fruit and vegetables for the primary tion and markers of vascular inflammation in the
prevention of cardiovascular diseases. Cochrane metabolic syndrome a randomized trial. JAMA.
Database Syst Rev. 2013;6:CD009874. https://doi. 2004;292:1440–6.
org/10.1002/14651858.CD009874.pub2. 45. Wang Q, Ellis PR. Oat b-glucan: physico-chemical
31. Yu D, Zhang X, Gao Y-T, et al. Fruit and vegetable characteristics in relation to its blood-glucose
intake and risk of coronary heart disease: results and cholesterol-lowering properties. Br J Nutr.
from prospective cohort studies of Chinese adults in 2014;112:S4–S13.
Shanghai. Br J Nutr. 2014;111(2):353–62. 46. Thies F, Masson LF, Boffetta P, Kris-Etherton
32. Winham DM, Hutchin AM, Johnston CS. Pinto bean P. Oats and CVD risk markers: a systematic litera-
consumption reduces biomarkers for heart disease ture review. Br J Nutr. 2014;112:S19–30.
risk. J Am Coll Nutr. 2007;26(3):243–9. 47. Mozaffarian D, Kumanyika SK, Lemaitre RN, et al.
33. Zaheer K, Akntar MH. An updated review of dietary Cereal, fruit, and vegetable fiber intake and the risk
isoflavones: nutrition, processing, bioavailability of cardiovascular disease in elderly individuals.
and impacts on human health. Crit Rev Food Sci JAMA. 2003;289(13):1659–66.
Nutr. 2017;57(6):1280–93. https://doi.org/10.1080/ 48. Dietary Guidelines Advisory Committee. Scientific
10408398.2014.989958. Report. Advisory Report to the Secretary of
34. Anderson JW, Bush HM. Soy protein effects on Health and Human Services and the Secretary of
serum lipoproteins: a quality assessment and meta- Agriculture. Part D. Chapter 1: Food and nutrient
analysis of randomized, controlled studies. J Am intakes, and health: current status and trends. 2015;
Coll Nutr. 2011;30(2):79–91. 1–78.
49. Micha R, Peñalvo JL, Cudhea F, et al. Association 63. Jonnalagadda SS, Harnack L, Liu RH, et al. Putting
between dietary factors and mortality from heart the whole grain puzzle together: health benefits
disease, stroke, and type 2 diabetes in the United associated with whole grains-summary of American
States. JAMA. 2017;317(9):912–24. https://doi. Society for Nutrition 2010 Satellite Symposium.
org/10.1001/jama.2017.0947. J Nutr. 2011;141:1011S–22S.
50. Mente A, de Koning L, Shannon HS, Anand SS. A 64. Harris KA, Kris-Etherton PM. Effects of whole-
systematic review of the evidence supporting a causal grains on coronary heart disease risk. Curr
link between dietary factors and coronary heart dis- Atheroscler Rep. 2010;12(6):368–76.
ease. Arch Intern Med. 2009;169(7):659–69. 65. Wang H, Lichtenstein AH, Lamon-Fava S, Jacques
51. Massera D, Zaman T, Grace E. Farren GE, et al. PF. Association between statin use and serum cho-
Case report: a whole-food plant-based diet reversed lesterol concentrations is modified by whole-grain
angina without medications or procedures. Case consumption: NHANES 2003–2006. Am J Clin
Rep Cardiol 2015; 2015:978906. doi: https://doi. Nutr. 2014;100:1149–57.
org/10.1155/2015/978906. 66. Li B, Zhang G, Tan M, et al. Consumption of whole
52. U.S. Department of Agriculture and U.S. Department grains in relation to mortality from all causes, car-
of Health and Human Services. Dietary Guidelines diovascular disease, and diabetes. Dose–response
for Americans, 2010. 7th ed. Washington, DC: meta-analysis of prospective cohort studies.
U.S. Government Printing Office; 2010. Table B2.4; Medicine. 2016;95:33. https://doi.org/10.1097/
http://www.choosemyplate.gov/. Accessed 22 Aug MD.0000000000004229.
2015. 67. Aune D, Keum N, Giovannucci E, et al. Whole grain
53. Seal CJ, Brownlee IA. Whole-grain foods and consumption and risk of cardiovascular disease, can-
chronic disease: evidence from epidemiological and cer, and all cause and cause specific mortality: sys-
intervention studies. Proc Nutr Soc. 2015;74:313–9. tematic review and dose-response meta-analysis of
54. Slavin J. Why whole grains are protective: biologi- prospective studies. BMJ. 2016;353:i2716. https://
cal mechanisms. Proc Nutr Soc. 2003;62:129–34. doi.org/10.1136/bmj.i2716.
55. Karl JP, Saltzman E. The role of whole grains in 68. Tang G, Wang D, Long J, et al. Meta-analysis of the
body weight regulation. Adv Nutr. 2012;3:697–707. association between whole grain intake and coronary
56. Othman RA, Moghadasian MH, Jones PJ. heart disease risk. Am J Cardiol. 2015;115:625–9.
Cholesterol-lowering effects of oat β-glucan. Nutr 69. Buyken AE, Goletzke J, Joslowski G, et al.
Rev. 2011;69(6):299–309. https://doi.org/10.111 Association between carbohydrate quality and
1/j.1753-4887.2011.00401. inflammatory markers: systematic review of obser-
57. Sofi F, Whittaker A, Cesari F, et al. Characterization vational and interventional studies. Am J Clin Nutr.
of Khorasan wheat (Kamut) and impact of a replace- 2014;99:813–33.
ment diet on cardiovascular risk factors: cross- 70. Helnæs A, Kyrø C, Andersen I, et al. Intake of whole
over dietary intervention study. Eur J Clin Nutr. grains is associated with lower risk of myocardial
2013;67:190–5. infarction: The Danish Diet, Cancer and Health
58. McGill CR, Fulgoni VL III, Devareddy L. Ten-year Cohort. Am J Clin Nutr. 2016;103:999–1007.
trends in fiber and whole grain intakes and food 71. Johnsen NF, Frederiksen K, Christensen J, et al.
sources for the United States population: National Whole-grain products and whole-grain types are
Health and Nutrition Examination Survey 2001– associated with lower all-cause and cause-specific
2010. Forum Nutr. 2015;7:1119–30. mortality in the Scandinavian HELGA cohort. Br J
59. Seal CJ. Whole grains and CVD risk. Proc Nutr Soc. Nutr. 2015;114(4):608–23. https://doi.org/10.1017/
2006;65:24–34. S0007114515001701.
60. Anderson JW. Whole grains protect against ath- 72. Sonestedt E, Hellstrand S, Schulz C-A, et al. The
erosclerotic cardiovascular disease. Proc Nutr Soc. association between carbohydrate-rich foods and
2003;62:135–42. risk of cardiovascular disease is not modified by
61. US Food and Drug Administration. Health claim genetic susceptibility to dyslipidemia as determined
notification for whole grain foods with moder- by 80 validated variants. PLoS One. 2015;10(4).
ate fat content. 2003. http://www.fda.gov/Food/ https://doi.org/10.1371/journal.pone.0126104.
LabelingNutrition/LabelClaims/FDA Modernization 73. Mellen PB, Liese AD, Tooze JA, et al. Whole-grain
Act FDAMA Claims/ucm073634.htm. Accessed 18 intake and carotid artery atherosclerosis in a multi-
Jul 2015. ethnic cohort: The Insulin Resistance Atherosclerosis
62. EFSA Panel on Dietetic Products, Nutrition and Study. Am J Clin Nutr. 2007;85:1495–502.
Allergies (NDA). Scientific Opinion on the substan- 74. Jensen MK, Koh-Banerjee P, Hu FB, et al. Intakes
tiation of a health claim related to whole grain (ID of whole grains, bran, and germ and the risk of
831, 832, 833, 1126, 1268, 1269, 1270, 1271, 1431) coronary heart disease in men. Am J Clin Nutr.
pursuant to Article 13(1) of Regulation (EC) No 2004;80:1492–9.
1924/2006. EFSA J. 2010;8(10):1766. https://doi. 75. Liu S, Stampfer MJ, Hu FB, et al. Whole-grain con-
org/10.2903/j.efsa.2010.1766. sumption and risk of coronary heart disease: results
References 367
from the Nurses’ Health Study. Am J Clin Nutr. 1990 and 2010: a systematic analysis including 266
1999;70:412–9. country-specific nutrition surveys worldwide. BMJ
76. Jacobs DR, Meyer KA, Kushi LH, Folsom Open. 2015;5:e008705. https://doi.org/10.1136/
AR. Whole-grain intake may reduce the risk of isch- bmjopen-2015-008705.
emic heart disease death in postmenopausal women: 89. Aune D, Edward Giovannucci E, Boffetta P, et al.
The Iowa Women’s Health Study. Am J Clin Nutr. Fruit and vegetable intake and the risk of cardiovas-
1998;68:248–57. cular disease, total cancer and all-cause mortality-a
77. Zhu X, Sun X, Wang M, et al. Quantitative assess- systematic review and dose response meta-analysis
ment of the effects of beta-glucan consumption on of prospective studies. Int J Epidemiol. 2017:1–28.
serum lipid profile and glucose level in hypercho- https://doi.org/10.1093/ije/dyw319.
lesterolemic subjects. Nutr Metab Cardiovasc Dis. 90. Gan Y, Tong X, Li L. Consumption of fruit and veg-
2015;25:714–23. etable and risk of coronary heart disease: A meta-
78. Whitehead A, Beck EJ, Tosh S, Wolever analysis of prospective cohort studies. Int J Cardiol.
TMS. Cholesterol-lowering effects of oat b-glucan: 2015;183:129–37.
a meta-analysis of randomized controlled trials. Am 91. Wang X, Ouyang Y, Jun Liu J, et al. Fruit and vegeta-
J Clin Nutr. 2014;100:1413–21. ble consumption and mortality from all causes, car-
79. Whittaker A, Sofi F, Luisi MLE, et al. An organic diovascular disease, and cancer: systematic review
Khorasan wheat-based replacement diet improves and dose-response meta-analysis of prospective
risk profile of patients with acute coronary syn- cohort studies. BMJ. 2014;349:g4490. https://doi.
drome: a randomized crossover trial. Forum Nutr. org/10.1136/bmj.g4490.
2015;7:3401–15. 92. He FJ, Nowson CA, Lucas M, Macgregor
80. Vitaglione P, Mennella I, Ferracane R, et al. Whole- GA. Increased consumption of fruit and vegetables
grain wheat consumption reduces inflammation in a is related to a reduced risk of coronary heart disease:
randomized controlled trial on overweight and obese meta-analysis of cohort studies. J Hum Hypertens.
subjects with unhealthy dietary and lifestyle behav- 2007;21:717–28.
iors: role of polyphenols bound to cereal dietary 93. Dauchet L, Amouyel P, Hercberg S, Dallongeville
fiber. Am J Clin Nutr. 2015;101:251–61. J. Fruit and vegetable consumption and risk of coro-
81. Giacco R, Lappi J, Costabile G, et al. Effects of nary heart disease: a meta-analysis of cohort studies.
rye and whole wheat versus refined cereal foods J Nutr. 2006;136:2588–93.
on metabolic risk factors: a randomised con- 94. Sharma S, Vik S, Kolonel LN. Fruit and veg-
trolled two- centre intervention study. Clin Nutr. etable consumption, ethnicity and risk of fatal
2013;32(6):941–9. ischemic heart disease. J Nutr Health Aging.
82. Brownlee IA, Moore C, Chatfield M, et al. Markers 2014;18(6):573–8.
of cardiovascular risk are not changed by increased 95. Bhupathiraju SN, Wedick NM, Pan A, et al.
whole-grain intake: the WHOLE heart study, a ran- Quantity and variety in fruit and vegetable intake
domised, controlled dietary intervention. Br J Nutr. and risk of coronary heart disease. Am J Clin Nutr.
2010;104(1):125–34. 2013;98:1514–23.
83. Katcher HI, Legro RS, Kunselman AR, et al. The 96. Griep LMO, Verschuren WMM, Kromhout D, et al.
effects of a whole grain-enriched hypocaloric diet Colours of fruit and vegetables and 10-year inci-
on cardiovascular disease risk factors in men and dence of CHD. Br J Nutr. 2011;106:1562–9.
women with metabolic syndrome. Am J Clin Nutr. 97. Griep LMO, Geleijnse JM, Kromhout D, et al. Raw
2008;87:79–90. and processed fruit and vegetable consumption
84. Slavin JL, Lloyd B. Health benefits of fruits and veg- and 10-year coronary heart disease incidence in a
etables. Adv Nutr. 2012;3:506–16. population-based cohort study in the Netherlands.
85. World Health Organization. Diet, nutrition, and PLoS One. 2010;5(10):e13609. https://doi.
the prevention of chronic diseases. Geneva: World org/10.1371/journal.pone.0013609.
Health Organization. 1990. http://www.who.int/ 98. Joshipura KJ, Hu FB, Manson JE, et al. The effect of
nutrition/publications/obesity/WHO_TRS_797/en/ fruit and vegetable intake on risk for coronary heart
index.html. Accessed 16 Apr 2015. disease. Ann Intern Med. 2001;134:1106–14.
86. Rooney C, McKinley MC, Appleton KM, et al. How 99. McEvoy CT, Wallace IR, Hamill L, et al.
much is ‘5-a-day?’ A qualitative investigation into Increasing fruit and vegetable intake has no
consumer understanding of fruit and vegetable intake dose-response effect on conventional cardiovas-
guidelines. J Hum Nutr Diet. 2016;30(1):105–13. cular risk factors in overweight adults at high
https://doi.org/10.1111/jhn.12393. risk of developing cardiovascular disease. J Nutr.
87. United States Department of Agriculture. Choose 2015;145:1464–71.
my plate. 2013. http://www.choosemyplate.gov/. 100. Wang L, Bordi PL, Fleming JA, et al. Effect of a
Accessed 17 Feb 2015. moderate fat diet with and without avocados on lipo-
88. Micha R, Khatibzadeh S, Shi P, et al. Global, regional protein particle number, size and subclasses in over-
and national consumption of major food groups in weight and obese adults: a randomized, controlled
trial. J Am Heart Assoc. 2015;4:e001355. https://doi. total and low-density lipoprotein cholesterols in adult
org/10.1161/JAHA.114.001355. women and men. Ann Nutr Metab. 2006;50:512–8.
101. Rayn-Haren G, Dragsted LO, Buch-Andersen T, 115. Kokubo Y, Iso H, Ishihara J, et al. Association
et al. Intake of whole apples or clear apple juice has of dietary intake of soy, beans, and isoflavones
contrasting effects on plasma lipids in health volun- with risk of cerebral and myocardial infarctions
teers. Eur J Nutr. 2013;52(8):1875–89. in Japanese populations. The Japan Public Health
102. Choi IH, Sook Noh JS, Han J-S, et al. Kimchi, a Center–Based (JPHC) Study Cohort I. Circulation.
fermented vegetable, improves serum lipid profiles 2007;116:2553–62.
in healthy young adults: randomized clinical trial. 116. Zhang X, Shu XO, Gao Y-T, et al. Soy food consump-
J Med Food. 2013;16(3):223–9. tion is associated with lower risk of coronary heart
103. Watzl B, Kulling SE, Möseneder J, et al. A 4-wk disease in Chinese women. J Nutr. 2003;133:2874–8.
intervention with high intake of carotenoid-rich 117. Tokede OA, Onabanjo TA, Yansane A, et al.
vegetables and fruit reduces plasma C-reactive pro- Soya products and serum lipids: a meta- analysis
tein in healthy, nonsmoking men. Am J Clin Nutr. of randomised controlled trials. Br J Nutr.
2005;82:1052–8. 2015;114:831–43.
104. McCrory MA, Hamaker BR, Lovejoy JC, 118. Yang B, Chen Y, Xu T, et al. Systematic review
Eichelsdoerfer PE. Pulse consumption, satiety, and and meta-analysis of soy products consumption in
weight management. Adv Nutr. 2010;1:17–30. patients with type 2 diabetes mellitus. Asia Pac J
105. Messina V. Nutritional and health benefits of dried Clin Nutr. 2011;20(4):593–602.
beans. Am J Clin Nutr. 2014;100(suppl):437S–42S. 119. Sacks FM, Lichtenstein A, Van Horn L, et al. Soy
106. Afshin A, Micha R, Khatibzadeh S, Mozaffarian protein, isoflavones, and cardiovascular health:
D. Consumption of nuts and legumes and risk of an American Heart Association Science Advisory
incident ischemic heart disease, stroke, and diabetes: for Professionals from the Nutrition Committee.
a systematic review and meta-analysis. Am J Clin Circulation. 2006;113:1034–44.
Nutr. 2014;100(1):278–88. 120. Padhi EMT, Blewett HJ, Duncan AM, et al. Whole
107. Bazzano LA, He J, Ogden LG, et al. Legume con- soy flour incorporated into a muffin and consumed
sumption and risk of coronary heart disease in US at 2 doses of soy protein does not lower LDL
men and women: NHANES I epidemiologic follow- cholesterol in a randomized, double-blind con-
up study. Arch Intern Med. 2001;161:2573–8. trolled trial of hypercholesterolemic adults. J Nutr.
108. Salehi-Abargouei A, Saraf-Bank S, Bellissimo N, 2015;145(12):2665–74. https://doi.org/10.3945/
Azadbakht L. Effects of non-soy legume consump- jn.115.219873.
tion on C-reactive protein: a systematic review and 121. Azadbakht L, Kimiagar M, Mehrabi Y, et al. Soy
meta-analysis. Nutrition. 2015;31:631–9. consumption, markers of inflammation, and endo-
109. Ha V, Sievenpiper JL, de Souza RJ, et al. Effect of thelial function. A cross-over study in postmeno-
dietary pulse intake on established therapeutic lipid pausal women with the metabolic syndrome.
targets for cardiovascular risk reduction: a system- Diabetes Care. 2007;30:967–73.
atic review and meta-analysis of randomized con- 122. Jenkins DJA, Kendall CWC, Jackson C-J, et al.
trolled trials. CMAJ. 2014;186(8):252–62. Effects of high- and low-isoflavone soyfoods on
110. Bazzano LA, Thompson AM, Tees MT, et al. Non- blood lipids, oxidized LDL, homocysteine, and
soy legume consumption lowers cholesterol levels: a blood pressure in hyperlipidemic men and women.
meta-analysis of randomized controlled trials. Nutr Am J Clin Nutr. 2002;76:365–72.
Metab Cardiovasc Dis. 2011;21(2):94–103. 123. Gardner CD, Newell KA, Cherin R, Haskell
111. Abeysekara S, Chilibeck PD, Vatanparast H, Zello WL. The effect of soy protein with or without iso-
GA. A pulse-based diet is effective for reducing flavones relative to milk protein on plasma lipids in
total and LDL-cholesterol in older adults. Br J Nutr. hypercholesterolemic postmenopausal women. Am J
2012;108:S103–10. Clin Nutr. 2001;73:728–35.
112. Hermsdorf HH, Zulet MA, Abete I, Martinez JA. A 124. Ashton E, Ball M. Effect of soy as tofu vs. meat
legume-based diet reduces proinflammatory status on lipoprotein concentrations. Eur J Clin Nutr.
and improves metabolic features in overweight/ 2000;54(10):14–9.
obese subjects. Eur J Nutr. 2011;50(1):61–9. 125. FDA. Food labeling: health claims; soy pro-
113. Pittaway JK, Robertson IK, Ball MJ. Chickpeas may tein and coronary heart disease. Food and Drug
influence fatty acid and fiber intake in an ad libi- Administration, HHS. Final rule. Final Regist.
tum diet, leading to small improvements in serum 1999;64(206):57700–33.
lipid profile and glycemic control. J Am Diet Assoc. 126. EFSA Panel on Dietetic Products, Nutrition and
2008;108:1009–13. Allergies (NDA). Scientific opinion on the sub-
114. Pittaway JK, Ahuja KD, Cehun M, et al. Dietary stantiation of a health claim related to soy protein
supplementation with chickpeas for at least 5 weeks and reduction of blood cholesterol concentrations
results in small but significant reductions in serum
References 369
pursuant to Article 14 of the Regulation (EC) No 143. Kelly JH Jr, Sabate J. Nuts and coronary heart dis-
1924/2006. EFSA J. 2010;8(7):1688. ease: an epidemiological perspective. Br J Nutr.
127. Zhang X, Gao X-T, Yang G, et al. Urinary isofla- 2006;96(2):S61–7.
vonoids and risk of coronary heart disease. Int J 144. Del Gobbo LC, Falk MC, Feldman R, et al. Effects
Epidemiol. 2012;41:1367–75. of tree nuts on blood lipids, apolipoproteins, and
128. Messina M, Messina V, Jenkins DJA. Can breast blood pressure: systematic review, meta-analysis,
cancer patients use soya foods to help reduce risk of and dose-response of 61 controlled intervention tri-
CHD? Br J Nutr. 2012;108:810–9. als. Am J Clin Nutr. 2015;102:1347–56.
129. Caan BJ, Natarajan L, Parker B, et al. Soy food 145. Sabate J, Oda K, Ros E. Nut consumption and blood
consumption and breast cancer prognosis. Cancer lipid levels a pooled analysis of 25 intervention tri-
Epidemiol Biomark Prev. 2011;20:854–8. als. Arch Intern Med. 2010;170(9):821–7.
130. Ros E. Nuts and CVD. Br J Nutr. 2015;113:S111–20. 146. Banel DK, Hu FB. Effects of walnut consumption
131. Prasad K. Flaxseed and cardiovascular disease. on blood lipids and other cardiovascular risk factors:
J Cardiovasc Pharmacol. 2009;54(5):369–77. a meta-analysis and systematic review. Am J Clin
132. US Food and Drug Administration. Qualified Nutr. 2009;90:56–63.
health claims. http://www.fda.gov/Food/ 147. Phung OJ, Makanji SS, White CM, Coleman
IngredientsPackagingLabeling/LabelingNutrition/ CI. Almonds have a neutral effect on serum lipid
ucm073992.htm. Accessed 11 Aug 2015. profiles: a meta-analysis of randomized trials. J Am
133. Jackson CL, Hu FB. Long-term associations of nut Diet Assoc. 2009;109:865–73.
consumption with body weight and obesity. Am J 148. Griel AE, Kris-Etherton PM. Tree nuts and the
Clin Nutr. 2014;100(suppl):408S–11S. lipid profile: a review of clinical studies. Br J Nutr.
134. Mattes RD, Kris-Etherton PM, Foster GD. Impact of 2006;96(2):S68–78.
peanuts and tree nuts on body weight and healthy weight 149. Le T, Flatt SW, Natarajan L, et al. Effects of diet
loss in adults. J Nutr. 2008;138(suppl):1741S–5S. composition and insulin resistance status on plasma
135. Novotny JA, Gebauer SK, Baer DJ. Discrepancy lipid levels in a weight loss intervention in women.
between the Atwater factor predicted and empiri- J Am Heart Assoc. 2016;5:e002771. https://doi.
cally measured energy values of almonds in human org/10.1161/JAHA.115.002771.
diets. Am J Clin Nutr. 2012;96:296–301. 150. Berryman CE, West SG, Fleming JA, et al. Effects
136. Baer DJ, Gebauer SK, Novotny JA. Walnuts con- of daily almond consumption on cardiometabolic
sumed by healthy adults provide less available risk and abdominal adiposity in healthy adults with
energy than predicted by the Atwater factors. J Nutr. elevated LDL-cholesterol: a randomized controlled
2016;146:9–13. trial. J Am Heart Assoc. 2015;4(1):e000993. https://
137. Traoret CJ, Lokko P, Cruz ACRF, et al. Peanut diges- doi.org/10.1161/JAHA.114.000993.
tion and energy balance. Int J Obes. 2008;32:322–8. 151. Edel AL, Rodriguez-Leyva D, Maddaford TG,
138. Brown RC, Tey SL, Gray AR, et al. Nut consump- et al. Dietary flaxseed independently lowers cir-
tion is associated with better nutrient intakes: results culating cholesterol and lowers it beyond the
from the 2008/09 New Zealand Adult Nutrition effects of cholesterol-lowering medications alone
Survey. Br J Nutr. 2016;115:105–12. in patients with peripheral artery disease. J Nutr.
139. Nielsen SJ, Kit BK, Ogden CL. Nut consumption 2015;145:749–57.
among U.S. adults, 2009–2010, NCHS data brief, 152. Richmond K, Williams S, Mann J, et al. Markers of
no 176. Hyattsville: National Center for Health cardiovascular risk in postmenopausal women with
Statistics; 2014. type 2 diabetes are improved by the daily consump-
140. Mayhew AJ, de Souza RJ, Meyre D, et al. A system- tion of almonds or sunflower kernels: a feeding
atic review and meta-analysis of nut consumption study. ISRN Nutr. 2013;2013:626414. https://doi.
and incident risk of CVD and all-cause mortality. Br org/10.5402/2013/626414.
J Nutr. 2016;115:212–25. 153. Nouran MG, Kimiagar M, Abadi A, et al. Peanut
141. Ma L, Wang F, Guo W, et al. Nut consumption and consumption and cardiovascular risk. Public Health
the risk of coronary artery disease: a dose-response Nutr. 2009;13(10):1581–6.
meta-analysis of 13 prospective studies. Thromb 154. Sabate J, Fraser GE, Burke K, et al. Effects of wal-
Res. 2014;134(4):790–4. nuts on serum lipid levels and blood pressure in nor-
142. Luu HN, Blot WJ, Xiang Y-B, et al. Prospective eval- mal men. N Engl J Med. 1993;328:603–7.
uation of the association of nut/peanut consumption 155. Ruisinger JF, Gibson CA, Backes JM, et al. Statins
with total and cause-specific mortality. JAMA Intern and almonds to lower lipoproteins (the STALL
Med. 2015;175(5):755–66. Study). J Clin Lipidol. 2015;9:58–64.
Dietary Patterns and Hypertension
13
Keywords
Hypertension • Blood pressure • Aging • Overweight • Obesity • DASH
diet • Mediterranean diet • Vegetarian diets • Dietary guidelines • Western
diet
Key Points Nordic diet, dietary guidelines- based, and

vegetarian diets are effective in lowering BP,
• Elevated blood pressure (BP), including pre- especially in older, overweight or obese hyper-
hypertension and hypertension, is a common tensive and prehypertensive adults compared
and growing public health problem. Globally, to Western diets.
the overall prevalence of elevated BP is • Healthy dietary patterns lower sodium, exces-
approaching 50% of adults age ≥ 25 years. sive energy and added refined carbohydrate
The adult risk of cardiovascular disease intake, and increase the levels of fiber, plant
(CVD) and renal disease approximately dou- protein, potassium, and other essential nutri-
bles for each 20/10 mm Hg incremental ents and bioactive phytochemicals intake are
increase above 115/75 mm Hg. associated with a lower risk of hypertension
• The major lifestyle factors associated with and elevated BP.
elevated BP and hypertension are aging, espe- • Healthy dietary pattern mechanisms associ-
cially unhealthy aging associated with over- ated with reduced hypertension risk include;
weight and obesity, poor dietary habits, lowering the risk of weight gain, stimulating
inactivity or lack of exercise, and ineffective colon microbiota, improving vascular health
stress management. by normalizing total cholesterol and LDL-C
• There is convincing evidence that high adher- levels, reducing oxidative and inflammatory
ence to healthy dietary patterns, including the stress, improving insulin sensitivity to reduce
Dietary Approaches to Stop Hypertension atherosclerosis risk, and maintaining electro-
(DASH), the Mediterranean (MedDiet), lyte balance.

372 13 Dietary Patterns and Hypertension
13.1 Introduction disease [4, 7]. Also, prehypertension (120/80–

139/89 mm Hg) has been associated with 1.5- to
Elevated blood pressure (BP), including prehy- 2-fold increases in cardiovascular disease events
pertension and hypertension, is a common and after age 55 years and a 43% increased risk of
growing public health problem [1, 2]. Globally, coronary heart disease (CHD) [4].
the overall prevalence of elevated BP is approach- The major factors associated with elevated BP
ing 50% of adults age ≥ 25 years. By 2025, and hypertension are aging, especially unhealthy
because of population growth and an increasing aging associated with overweight and obesity, poor
aging population, it is projected that about dietary habits, inactivity or lack of exercise, and
1.5 billion individuals will have hypertension ineffective stress management [1–9]. Aging is
[2–4]. Over nine million annual deaths world- directly associated with elevated BP risk (Fig. 13.1)
wide are currently attributable to hypertension [4]. Excess body weight is associated with
[1]. The World Health Organization (WHO) has increased activity of the RAAS, insulin resistance,
identified that there is convincing evidence that and reduced kidney function (especially with salt-
obesity and poor dietary patterns are associated sensitive hypertension individuals) [7–9]. Rates of
with increased risk of hypertension [1, 5]. hypertension are twice as likely to occur in obese
Increased BP etiology is linked to the renin- vs. normal weight individuals. Dietary patterns
angiotensin aldosterone system (RAAS), a hor- with low dietary energy density, high fiber density
monal cascade that functions in the homeostatic and containing primarily natural sugar sources
control of BP and extracellular fluid volume [6]. such as whole fruits tend to be associated with
Aldosterone causes the tubules of the kidneys to lower prevalence of overweight or obesity [5].
increase the reabsorption of sodium and water Whereas a dietary pattern with low fiber density
into the blood, while at the same time causing and high in added sugars intake (primarily from
the excretion of potassium, which increases the chocolate and fruit drinks) was associated with
volume of extracellular fluid leading to elevated increased prevalence of overweight or obesity.
BP. Hypertension is a major risk factor for Dietary patterns low in fiber density and high in the
cardiovascular disease (CVD) stroke and renal ratio of sodium to potassium and high in saturated
Men Women
90
80
70
Hypertension Prevalence (%)
60
50
40
30
20
10
0
20-34 35-44 45-54 55-64 65-74 > 75
Age (years)
Fig. 13.1 Prevalence of hypertension in US adults ≥20 years with age (National Health and Nutrition Examination
Survey [NHANES] 2007–2012) (adapted from [4])
0.0
Healthy Diet Aerobic Relaxation Sodium Fish Oil Alcohol
Exercise Restriction Restriction
–1.0
–2.0
–3.0
mm Hg
–4.0
–5.0
–6.0
–7.0
Fig. 13.2 Association between healthy lifestyle factors and blood pressure (BP) from a systematic review of RCTs
(all healthy lifestyle factors p < 0.05) (adapted from [9, 12])
fat to polyunsaturated fat ratios are associated with effective in reducing BP in hypertensive than nor-
increased prevalence of hypertension and prehy- motensive individuals and they provide adjunctive
pertension [5]. A 2017 meta-analysis of cohort BP lowering for hypertensive people already on
studies found an inverse linear dose–response asso- drug therapy. The food components and composi-
ciation between risk of hypertension and physical tions of several dietary patterns are summarized in
activity with a 6% lower risk of hypertension when Appendix A. The objective of this chapter is to
individuals met the minimum recommended guide- comprehensively review the role of dietary patterns
line levels of 150 min/week compared with inac- in hypertension prevention and management.
tive individuals and higher levels of physical
activity reduced hypertension risk by up to 33%
[8]. The Western lifestyle is associated with higher 13.2 Dietary Patterns
BP among both adults and adolescents [2, 9–12].
The effects of various healthy lifestyle factors asso- High adherence to healthy diet patterns are associ-
ciated with lower BP are summarized in Fig. 13.2 ated with significantly lower CVD risks including
[9, 12]. Elevated BP is a major risk factor for stroke, hypertension and CHD compared to high adher-
CHD, damage to retinal blood vessels, and kidney ence to the Western diet [13, 14]. Hypertension
disease/renal failure [1–9]. The adult risk of CVD and CHD are interconnected as among individuals
and renal disease approximately doubles for each aged 40–90 years, each 20/10 mm Hg rise in BP
20/10 mm Hg incremental increase above doubles the risk of fatal coronary events [15, 16].
115/75 mm Hg. Guidelines for prevention, treat- A 2016 meta-analysis (16 cohort and 11 cross-
ment and control of elevated BP focus on lifestyle sectional studies; 295,799 participants) found that
modifications including weight loss and mainte- the highest adherence to a healthy dietary pattern
nance, reduced salt intake, increased fruit and veg- reduced hypertension risk by 19% (p = 0.02) and
etables consumption, routine participation in the highest adherence to the Western diet increased
physical activity, cessation of smoking, limiting hypertension risk by 4% (p > 0.05) compared to
alcohol consumption, and anxiety and stress con- lowest adherence [17]. However, subgroup analy-
trol [1–12]. Healthy dietary patterns are more sis showed that individuals >50 years were more
> 50 years < 50 years

70
60
50
40
30
20
10
0
High Adherence High Adherence Heavy Drinking Light-Moderate
–10 to Healthy Diet to Western Diet Drinking
–20
–30
Fig. 13.3 Association between age, and dietary and drinking patterns on hypertension risk based on a meta-analysis of
27 observational studies (p < 0.0001, for all) (adapted from [17])
sensitive to the effects of the Western diet on such as the DASH, Nordic diet, dietary guidelines
hypertension risk (Fig. 13.3) [17]. Additionally, a and Mediterranean diet (MedDiet); 6 weeks to
heavy drinking pattern increased hypertension risk 2 years) showed that healthy dietary patterns sig-
by 62% (p = 0.004), whereas light-moderate drink- nificantly lowered systolic BP by 4.3 mm Hg and
ing patterns insignificantly increased hypertension diastolic BP by 2.4 mm Hg across a range of
risk by 20% (p = 0.13). A 2015 meta-analysis (35 sodium intakes [21]. A cross-sectional analysis
cohort and case-control studies) found an inverse that used baseline data from the Australian
association between healthy dietary patterns and HealthTrack study (328 adults; 12-month weight
CHD risk with a 33% risk reduction whereas a loss RCT) found that a dietary pattern rich in nuts,
Western diet increased CHD risk by 45% (high vs. seeds, fruit, and fish with a lower sodium to potas-
low adherence) [18]. The high adherence to sium ratio was inversely associated with BP [22].
healthy dietary patterns characterized by high con-
sumption of fruit, vegetables, whole grains,
legumes, seeds, nuts, fish, and low fat dairy and 13.2.1 DASH Dietary Patterns
low consumption of processed meat, sweets, and
alcohol are generally associated with significantly 13.2.1.1 Background
lower hypertension risk [8, 19]. The US 2015 The DASH diet was designed primarily for indi-
Dietary Guidelines Advisory Committee report viduals with hypertension [13]. This pattern is rich
concluded that healthy dietary patterns rich in in fruits and vegetables, low-fat dairy products,
fiber, potassium, carotenoids and other healthy includes whole grains, poultry, fish, and nuts, and
components and low in saturated fat and sodium, limits saturated fat, red meat, sweets, and sugar
especially the Dietary Approaches to Stop containing beverages. Compared with the Western
Hypertension (DASH)-style diets can lower sysdiet, the DASH diet provides lower total fat, satu-
tolic BP by 6 mm Hg and diastolic BP by 3 mm Hg rated fat, and dietary cholesterol, and higher potas-
compared to Western patterns [20]. A meta-analy- sium, magnesium, calcium, fiber, and protein. A
sis (17 RCTs; 4909 prehypertensive or hyperten- DASH adherence score is based on the following
sive participants; 13 trials allowed continuation of primary food and nutrient components: increased
BP-lowering medications; healthy dietary p atterns fruits, vegetables, whole grains, nuts and legumes
and low-fat dairy; and reduced red and processed motensive or leaner individuals [27–29]. A 2016
meats, sweetened beverages, and sodium. meta-analysis (24 RCTs; 23,858 subjects; median
age 45 years; 6–48 months) found that healthy
13.2.1.2 Observational Studies dietary patterns reduced the pooled mean systolic
A number of observational studies support the BP by 3 mm Hg and diastolic BP by 1.8 mm Hg
DASH diet’s benefits in reducing hypertension risk [27]. The DASH diet was the most effective in
factors or hypertension risk directly. In a cross-sec- BP lowering with a reduction for systolic BP by
tional analysis (2047 Irish middle-aged men and 7.6 mm Hg and diastolic BP by 4.2 mm Hg. Low-
women) there was an inverse association between sodium; low-sodium, high-potassium; low-
DASH diet score and systolic BP with significantly sodium, low-calorie; and low-calorie diets also
lower systolic BP in men by 7.5 mm Hg and in led to significant systolic and diastolic BP reduc-
women by 5.1 mm Hg (highest vs. lowest DASH tions, whereas Mediterranean diet (MedDiet)
quintiles) [23]. An Iranian cross-sectional study participants experienced a significant reduction
(293 nurses) found that those in the highest vs. low- in diastolic BP but not systolic BP. A 2015 meta-
est quartile of the DASH diet scores lowered obe- analysis (20 RCTs; 1917 subjects; 2–26 weeks)
sity risk by 71% and central obesity risk by 63%, showed that the DASH diet significantly
both of which are major hypertension risk factors decreased systolic BP by 5.2 mm Hg and dia-
[24]. The Taiwanese CardioVascular Disease Risk stolic BP by 2.6 mm Hg for an estimated 13%
FACtor Twotownship Study (1420 participants; reduction in the 10-year Framingham risk score
mean age 45.5 years; 13-years of follow-up) found for CVD [28]. Changes in both systolic and dia-
that adhering to the DASH diet was beneficial for stolic BP were greater in participants with higher
long term BP control and reduction of stroke risk in baseline BP or BMI. A 2014 meta-analysis (17
the studied Chinese population [25]. Additionally, RCTs; 2561 subjects; 2–26 weeks) found that the
a 12-month WEB-based nutrition education pro- DASH diet significantly reduced pooled mean
gram called ‘DASH for health’ longitudinal obser- systolic BP by 6.7 mm Hg and diastolic BP by
vational study (735 hypertensive or prehypertensive 3.5 mm Hg; RCTs with hypertensive subjects
subjects;12 months on the program) observed a having significantly greater decreases in BP [29].
significantly lower systolic BP by 6.8 mm Hg [26].
BP and body weight were inversely associated with Representative RCTs
the number of visits to the ‘DASH for health’ WEB Eight representative RCTs provide important
site. A 2016 Atherosclerosis Risk Communities insights on the effect of DASH diets on BP and
Study (14,882 participants; age range 45-64 years; hypertension risk [30–37]. A 2016 US crossover
median follow-up of 23 years) found that partici- RCT (36 subjects; mean age 48 years and BMI 27;
pants with the lowest DASH diet score, which is l lower fat and higher fat DASH diets vs. Western
ike the Western dietary pattern, had a significantly diet; 3 weeks) showed DASH diets with full-fat or
increased risk of developing kidney disease by low-fat dairy food had similar effects on BP
16%, after multivariate adjustment [11]. (Fig. 13.4) [30]. The low-fat dairy DASH diet sig-
nificantly reduced LDL-C, HDL-C, apolipopro-
13.2.1.3 Randomized Controlled tein A, intermediate density lipoprotein and large
Trials (RCTs) LDL particles, and LDL peak diameter compared
RCTs have consistently demonstrated that with the Western diet, but not the full-fat dairy
DASH-type diets are the most effective diets for DASH diet, which significantly reduced triglycer-
lowering elevated BP (Table 13.1) [27–37]. ides levels compared with the low-fat dairy DASH
diet. The OmniHeart trial (164 adults; mean age
Meta-Analyses 54 years; 73% women; mean BMI 30; 80% prehy-
Three DASH focused meta-analyses show greater pertensive; stable body weight; 6 weeks) demon-
BP reductions in older hypertensive individuals strated that replacing 10% of carbohydrate calories
with elevated BMIs compared to younger, nor- in the standard DASH diet with either the same
Table 13.1 Summary of Dietary Approaches to Stop Hypertension (DASH) diet RCTs in blood pressure (BP)
management
Objective Study details Study results
Gay et al. (2016). 24 RCTs; 23,858 subjects; duration Overall, all the healthy diets reduced
Quantify the aggregated ranged from 6 to 48 months of follow-up systolic BP by 3 mm Hg and diastolic
BP-lowering effects associated (median: 12 months); age ranged from BP by 1.8 mm Hg. The DASH diet
with dietary patterns and 34 to 67 years (median: 45 years); all was the most effective diet with
special diets interventions on subjects were overweight or obese; reduced systolic BP by 7.6 mm Hg
BP [27] baseline mean systolic BP 136 mm Hg and diastolic BP by 4.2 mm Hg
and diastolic BP 86 mm Hg
Siervo et al. (2015). 20 RCTs; 1917 participants; male and The DASH diet significantly
Assess the effects of the female; age > 18 years; healthy decreased systolic BP by 5.2 mm Hg
DASH diet on BP and CVD individuals with above-optimal BP and and diastolic BP by 2.6 mm Hg.
risk factors [28] stage 1 hypertension; 2–24 weeks Changes in both systolic and diastolic
BP were greater in participants with
higher baseline BP or BMI. These
changes predicted a lower 10-year
Framingham risk score for CVD by
13%
Saneei et al. (2014). 17 RCTs; 2561 participants; 1747 The DASH diet significantly reduced
Examine the effect of the participants with hypertension, 293 systolic BP by 6.7 mm Hg and
DASH diet on BP [29] without hypertension, and 521 with diastolic BP by 3.5 mm Hg. RCTs
undisclosed hypertensive status; with hypertensive subjects had
2–26 weeks significantly greater decrease in BP
Representative RCTs
Chiu et al. (2016). Crossover RCT: DASH diets with low-fat or full fat
Evaluate the effects of full-fat 36 subjects; mean age 48 years; mean dairy had similar BP lowering effects
or low-fat dairy foods in the BMI 27; 40% women; diets: Western compared with the Western control
DASH diet, with a diet control, a standard DASH diet, and diet (Fig. 13.4)
corresponding increase in fat a higher-fat, lower-carbohydrate
and a reduction in sugar modification of the DASH diet
intake, on blood pressure and (HF-DASH diet); 3 weeks each,
plasma lipids and lipoproteins separated by 2-week washout periods
(US) [30]
Al-Solaiman et al. (2010). Crossover RCT: In obese hypertensives, the DASH
Evaluate the effects of the 30 adults; mean age 38 years; 80% diet significantly lowered systolic BP
DASH diet on BP in women; 15 obese, mean BMI 35, mean by 7.6 and diastolic BP by 5.3 mm Hg
abdominally obese baseline BP 136/89 mm Hg; 15 lean, and the usual diet supplemented with
hypertensive vs. lean mean BMI 23; mean BP 110/70 mm Hg; potassium, magnesium and fiber
normotensive adults (US) [31] DASH diet, usual diet, or usual diet lowered systolic BP by 6.2 and
supplemented w/ potassium, magnesium diastolic BP by 3.7 mm Hg compared
and fiber to match DASH diet; 3 weeks to no BP reduction for the usual
with no washout control diet. In lean normotensive
subjects, BP values were not
significantly different among the 3
diets
Blumenthal et al. (2010). Parallel RCT: The DASH plus weight loss
Compare the BP lowering 144 overweight or obese, unmedicated significantly reduced systolic BP by
effects of the DASH diet alone outpatients; pre-hypertension or stage 1 16.1 mm Hg and diastolic BP by
or in combination with a hypertension systolic BP 130–159 mm Hg; 9.9 mm Hg vs. the DASH without
weight management program diastolic BP 85–99 mm Hg; DASH diet weight loss which reduced systolic
(US) [32] alone or combined with a weight BP by 11.2 mm Hg and diastolic BP
management program vs. usual diet by 7.5 mm Hg whereas the usual diet
controls; 4 months lowered systolic BP by 3.4 mm Hg
and diastolic BP by 3.8 mm Hg
Appel et al. (2005). 3-period, crossover multi-center RCT: Compared with the DASH diet, the
Investigate the effects of BP 164 adults; mean age 54 years; 73% protein and unsaturated modified
on partial replacement of women; mean BMI 30; mean baseline DASH diets significantly decreased
carbohydrates in the original BP 131/77 mm Hg; 80% systolic BP and diastolic BP among
DASH diet with protein and prehypertensive; diets: Standard DASH those with hypertension (Table 13.2).
unsaturated fat (US) [33] diet rich in carbohydrates; DASH diet All DASH diets lowered estimated
10% energy replacement by protein, CHD risk by 16–21% compared with
about half from plant sources; and baseline
DASH diet 10% energy replacement by
unsaturated fat, primarily
monounsaturated fat; 6 weeks
Nowson et al. (2005). Parallel RCT: The hypocaloric DASH diet
Evaluate the effect of weight 63 men; mean age 48 years; mean BMI significantly lowered systolic BP by
reduction diets: a low-fat (LF) 30; baseline BP 113/88 mm Hg; diets: 7.6 mm Hg and diastolic BP by
diet and a DASH type diet on Hypocaloric DASH diet or a low-fat 5.4 mm Hg vs. hypocaloric low-fat
BP in men (Australia) [34] diet. Both diet groups engaged in 0.5 h diet (Fig. 13.5). Also, 5% reduction in
of moderate physical activity on most weight decreased systolic BP by
days of the week; 12 weeks 8 mm Hg, and diastolic BP by 5 mm
Hg
Appel et al. (2003). Multicenter parallel RCT: Compared to general advice only, the
Investigate the effects of 810 adults; mean age 50 years; mean established behavioral program
implementing several types of BMI 33; 62% women; mean baseline BP significantly reduced systolic BP by
multi-component, lifestyle 135/85 mm Hg; 5% current smokers; 3.7 mm Hg and the established
programs including the DASH 38% hypertensive no BP medications; behavioral program plus DASH
diet (PREMIER study; US) treatments: Behavioral program (e.g., lowered systolic BP by 4.3 mm Hg.
[35] weight loss, sodium reduction, increased After 6 months, the prevalence of
physical activity, and limited alcohol hypertension in the advice only group,
intake); behavioral plus DASH program; behavioral group and behavioral plus
or advice only as a comparison group DASH diet was 26%, 17% and 12%,
respectively, compared to the 38%
hypertension prevalence at baseline
Miller et al. (2002). Parallel RCT: There was 4.9 kg weight loss in the
Examine the effects of a 44 adults; mean age 54 years; 62% DASH plus exercise group vs. control.
comprehensive DASH lifestyle women; mean BMI 33; mean BP Also, the DASH diet plus exercise
intervention on BP and other 136/84 mm Hg; all subjects on BP group significantly lowered mean
CVD risk factors (US) [36] medications; treatments: a hypocaloric 24-hour ambulatory systolic BP by
DASH diet plus moderate-intensity 9.5 mm Hg and diastolic BP by
exercise 3 days per week vs. a control 5.3 mm Hg
usual diet no exercise; 9 weeks
Apple et al. (1997). Multi-center parallel RCT: The DASH diet significantly reduced
Investigate the effects of a fruit 459 adults; 50% women; 60% African systolic and diastolic BP by 5.5 and
and vegetable rich diet or American; mean age 44 years; mean 3.0 mm Hg more than the control diet
DASH diet vs. a usual BMI 29; mean BP 132/85 mm Hg; diets: (Fig. 13.6). Also, the fruit and
American diet on BP (US) (1) 8–10 portions fruit and vegetable vegetable rich diet significantly
[37] diet, (2) DASH diet with 8–10 portions reduced systolic BP by 2.8 mm Hg
of fruit and vegetables; (3) control and borderline significantly reduced
American low fruit and vegetable diet, diastolic BP by 1.1 mm Hg more than
8 weeks after 3-week run-in on the the control diet. Among the 133
American diet hypertensive subjects, the DASH diet
significantly reduced systolic and
diastolic BP by 11.4 and 5.5 mm Hg
more than the control diet. The DASH
diet was twice as effective in lowering
BP as a high fruit and vegetable diet
alone
Low-Fat Dairy DASH Diet Full-Fat Dairy DASH Diet

0
Systolic BP (mm Hg) Diastolic BP (mm Hg)
–0.5
Change mm Hg vs Western Diet
–1
–1.5
–2
–2.5
–3
–3.5
–4
Fig. 13.4 Effects of full-fat and low-fat dairy foods in the DASH diet compared to the Western control diet on BP
(p ≤0.017 for all BP) [adapted from 30]
Table 13.2 Effect on blood pressure lowering of substituting 10% of carbohydrate energy in the standard DASH diet
with protein or vegetable oil consisting primarily of monounsaturated fatty acids (MUFA) [33]
Mean baseline BP Mean lower BP from baseline by diet
Blood pressure (BP) (mm Hg) Carbohydrate Protein MUFA
Systolic BP (mm Hg)a
All 131.2 8.2 9.5 9.3
Stage 1 hypertension 146.5 12.9 16.1 15.8
Prehypertension 127.5 7.0 8.0 7.7
Diastolic BP (mm Hg)a
All 77.0 4.1 5.2 4.8
Stage 1 hypertension 84.2 6.3 8.6 8.2
Prehypertension 76.3 3.6 4.4 3.9
All partial replacements of protein or MUFA for carbohydrate (p <0.02)
a
calorie content of protein or monounsaturated fat urated and total fat vs. American Western control
significantly lowered systolic and diastolic BP as diet; 8 weeks after 3-week run-in on the American
shown in Table 13.2 [33]. In a DASH weight loss Western diet) found that the DASH diet signifi-
RCT (63 men; mean age 48 years; hypocaloric cantly reduced systolic BP by 5.5 mm Hg and dia-
DASH and low fat diets; mean BMI 30; BP stolic BP by 3.0 mm Hg more than the Western
135/88 mm Hg; 12 weeks), the DASH diet was diet control (Fig. 13.6) [37]. Among the 133 sub-
significantly more effective than low-fat weight jects with hypertension, the DASH diet more than
loss diets in lowering BP at similar weight loss doubled the BP lowering effects by further signifi-
(Fig. 13.5) [34]. A 1999 DASH trial (459 adults; cantly reducing systolic BP by 11.4 mm Hg and
50% women; 60% African American; mean age diastolic BP by 5.5 mm Hg vs. the American
44 years; mean BMI 29; mean BP 132/85 mm Hg; Western diet. For overweight or obese individuals
DASH diet: 8–10 portions of fruit and vegetables, with elevated BP, the addition of exercise and
increased whole-grains, low fat dairy, reduced sat- weight loss to the DASH diet resulted in even
Low Fat Diet DASH Diet

132
130
Systolic BP (mm Hg)
128
126
124
122
120
Baseline 2 4 6 8 10 12
Time (weeks)
Low Fat Diet DASH Diet

85
Diastolic BP (mm Hg)
83
81
79
77
75
Baseline 2 4 6 8 10 12
Time (weeks)
Fig. 13.5 Effect of a hypocaloric DASH-type vs. low fat weight-loss diet on blood pressure (BP) control (p < 0.006)
[adapted from 34]
larger significant BP reductions and improvements 13.2.2.1 Observational Studies

in vascular function [31, 32, 35, 36]. MedDiet prospective studies generally show
moderate beneficial effects on BP or hyperten-
sion risk, with higher adherence to a
13.2.2 Mediterranean Dietary Mediterranean-style diet. Several studies
Patterns (MedDiet) showed a modest decrease in systolic BP by
2.4–3.1 mm Hg and diastolic by 1.3–1.9 mm
The MedDiet incorporates the traditional healthy Hg, which was primarily associated with olive
dietary habits of people from countries bordering oil, fruit and vegetable intake [38–40]. However,
the Mediterranean Sea, which are rich in whole the SUN (Seguimiento Universidad de Navarra)
plant foods such as whole-grains, fresh fruits, prospective cohort (10,800 adults; mean age
vegetables, beans, nuts, and seeds, along with 37 years; mean BMI 23; 70% women; mean
moderate amounts of dairy foods, fish, and poul- 4.6 years of follow-up) observed that higher
try and lower amounts of red and processed meat adherence to the MedDiet did not show a signifi-
meat, and extra virgin olive oil as the major cant association with lowering the incidence of
source of fat and moderate wine is consumed hypertension, most likely because of the relative
moderately with meals [2, 17, 21]. There may be youth and normal BMI of the cohort [41].
some variation in food composition between During pregnancy, low adherence to a MedDiet
regions. The BP lowering evidence for the pattern or high adherence to a Western dietary
MedDiet pattern is more limited and moderate pattern are independently associated with higher
than that for the DASH dietary pattern [7]. BP [42].
American Diet DASH Diet

132.0
130.0
Systolic BP (mm Hg)
128.0
126.0
124.0
122.0
120.0
Baseline 1 2 3 4 5 6 8
Time (weeks)
American Diet DASH Diet

86.0
Diastolic BP (mm Hg)
84.0
82.0
80.0
78.0
Baseline 1 2 3 4 5 6 8
Time (weeks)
Fig. 13.6 Effect of DASH diet vs. Western-type diet on blood pressure (BP) in overweight combined normotensive and
hypertensive individuals (p < 0.001) [adapted from 37]
13.2.2.2 Randomized Controlled 2015 meta-analysis (9 RCTs; 1178 type 2 diabetic

Trials (RCTs) subjects; 4 weeks to 4 years) showed significant
Table 13.3 summarizes the effect of the MedDiet mean systolic BP lowering by 1.5 mm Hg and
RCTs on BP levels in normal and hypertension diastolic BP by 1.4 mm Hg compared to low-fat
subjects [43–49]. The AHA/ACC Lifestyle control diets, along with greater reductions in
Guideline recommended the MedDiet over a hemoglobin A1c, fasting plasma glucose, fasting
low-fat diet for controlling BP [23]. insulin, BMI, body weight, total cholesterol and
triglycerides, and increased HDL-C [44]. A 2011
Systematic Reviews and Meta-analyses meta-analysis (7 RCTs; 3650 overweight or obese
Three meta-analyses support a small but signifi- patients with at least one CVD risk factor or
cant BP lowering effect for MedDiets vs. low-fat patients with established coronary artery disease;
diet controls [43–45]. A 2016 meta-analysis (6 mean age 35–68 years; MedDiet vs. low fat diets;
RCTs; 7000 subjects; ≥1 year) found in normal to ≥6 months) found that the MedDiet significantly
mildly hypertensive individuals that MedDiets reduced mean systolic BP by 1.5 mm Hg and dia-
had a modest but significant BP lowering effect stolic BP by 1.4 mm Hg vs. a low-fat diet; in addi-
on systolic BP by 1.4 mm Hg and diastolic BP tion the MedDiet significantly lowered body
0.7 mm Hg compared to a low-fat diet [43]. A weight by 2.2 kg and BMI by 0.56 units [45].
Table 13.3 Summary of Mediterranean diets (MedDiet) RCTs in blood pressure (BP) management
Nissensohn et al. (2016). 6 RCTs; >7000 normal and mildly MedDiets had modest but significant
Examine MedDiet effects on BP hypertensive individuals; MedDiet vs. mean reductions in both systolic BP
in RCTs ≥1 year duration [43] low fat diets; ≥1 year by 1.4 mm Hg and diastolic BP by
0.7 mm Hg compared to low-fat diets
Huo et al. (2015). 9 RCTs; 1178 type 2 diabetic subjects; MedDiets significantly reduced mean
Investigate the effects of age 26–77 years; MedDiet vs. control systolic BP by 1.5 mm Hg and
MedDiet on cardiovascular risk diets; 4 weeks to 4 years diastolic BP by 1.4 mm Hg vs.
factors in type 2 diabetic low-fat diets. Also, MedDiets led to
subjects [44] greater reductions in HbA1c, fasting
plasma glucose, fasting insulin, BMI,
body weight, total cholesterol and
triglycerides, and increased HDL-C
compared to control diets
Nordmann et al. (2011). 7 RCTs; 3650 overweight/obese MedDiets significantly reduced mean
Evaluate the effects of MedDiet patients with at least one CVD risk systolic BP by 1.7 mm Hg and
on BP and other CVD risk factor or patients with established diastolic BP by 1.5 mm Hg vs. low
biomarkers [45] coronary artery disease; mean age fat diets. Also, the MedDiet
35–68 years; mean BMI ranged from significantly reduced mean body
29 to 35; MedDiet vs. low fat diets; ≥ weight by 2.2 kg, BMI by 0.56 kg/m2,
6 months and total cholesterol by 7.4 mg/dL
Representative RCTs
Storniolo et al. (2017). Parallel RCT: (1) for the MedDiet plus tree nuts,
Evaluate the effect of MedDiets 90 high CVD risk postmenopausal there was a significantly decreased
on BP and endothelial markers women; mean age 68 years; BMI 32; diastolic BP by 5% vs. low fat control
in hypertensive women mean systolic BP 155 and diastolic diet; significantly decreased serum
(PREvención con DIeta 84 mm Hg; diets: Traditional MedDiet endothelin-1 levels by 19% vs.
MEDiterránea [PREDIMED] supplemented with extra virgin olive oil baseline values. (2) For the MedDiet
substudy; Spain) [46] or mixed nuts versus a control low-fat plus extra virgin olive oil, there was
diet; 1 year significantly increased serum stable
nitric oxide metabolites by 64% vs.
baseline values. (3) For the low-fat
diet guidance, there were no
significant changes in BP or
endothelial markers
Domenech et al. (2014). Parallel RCT: MedDiets with both extra virgin olive
Investigate the effects of 235 high CVD risk adults; mean age oil and tree nuts significantly lowered
MedDiets on ambulatory BP in 67 years; 57% women; mean BMI 30; systolic BP by 4.0 mm Hg and
elderly subjects at high CVD mean systolic BP 145 mm Hg, mean diastolic BP by 1.9 mm Hg vs. low
risk over 1 year (PREDIMED diastolic BP 82 mm Hg; diets: fat control diets
substudy; Spain) [47] Traditional MedDiet supplemented with
extra virgin olive oil or mixed tree nuts
versus a control low-fat diet; 1 year
Toledo et al. (2013). Multi-center parallel RCT: The MedDiets significantly reduced
Examine the effects of the 7447 asymptomatic adults with high diastolic BP with extra virgin olive oil
MedDiet on BP after 4 years CVD risk; mean age 67 years; 57% diet by 1.5 mm Hg and with tree nuts
(Primary prevention women; mean BMI 30; mean systolic by 0.65 mm Hg vs. low fat control
PREDIMED study; Spain) [48] BP 143, diastolic BP 83 mm Hg; 70% diet. There were no observed
subjects on anti-hypertensive between-group differences in systolic
medications; diets: MedDiets with BP
either extra virgin olive oil, or mixed
tree nuts vs. low-fat diet; BP measures
at baseline and once yearly; 4 years
(continued)
Estruch et al. (2006). Parallel RCT: MedDiets significantly reduced
Evaluate the effects of the 772 asymptomatic persons at high CVD systolic BP with extra virgin olive oil
MedDiet on BP (PREDIMED risk; mean age 69 years; 59% women; by 5.4 mm Hg and nuts by 7.1 mm
sub-trial; Spain) [49] 42% current smokers; mean BMI 30; Hg, and diastolic BP for extra virgin
all subjects on anti-hypertension olive oil by 1.6 mm Hg and nuts by
medications; diets: MedDiets including 2.7 mm Hg vs. low fat control.
nutritional education and either extra Significant reductions were shown for
virgin olive oil, 1 L/week, or free mixed plasma glucose, cholesterol: HDL-C
nuts, 30 g/day vs. low-fat control diet; ratio and CRP for extra virgin olive
3 months oil MedDiet compared with the
low-fat diet
Table 13.4 Potential healthy dietary pattern mechanisms associated with lower blood pressure (BP) and reduced risk
of hypertension [60–77]
Target Increase Decrease
Food intake Chewing Energy density
Eating time Hunger
Body weight and fat Weight gain
Abdominal fat
Ectopic fat
Stomach Satiety signals Gastric emptying rate
Lipid emulsification
Lipolysis
Liver Lipoprotein uptake Lipogenesis
Bile acid production
Small intestine Satiety signals Dietary fat absorption
Peripheral tissue Insulin sensitivity Insulin resistance
Circulatory system Short-chain fatty acids Fasting lipids (e.g, TG)a
Carotenoids (e.g., lutein) Fasting lipoproteins (e.g., TC, LDL-C)a
Flavonoids and metabolites Inflammatory markers (e.g., CRP)
Nitrate and nitric oxide Oxidized LDL-C
Electrolyte balance Intima-media thickness progression
Large intestine Fermentation Bile acid reabsorption
Short-chain fatty acids (butyrate) Inflammatory activity
Microbiota health Systemic Lipopolysaccharide (LPS)
Satiety signals
Fecal excretion Macronutrients (e.g., dietary fat) Metabolizable energy
TC total cholesterol, LDL-C LDL-cholesterol, TG triglycerides, CRP C-reactive protein
a
Representative RCTs 1 year) found that the MedDiet significantly low-

Four PREvención con DIeta MEDiterránea ered systolic BP with extra virgin olive oil by
(PREDIMED) trials in older adults with high 4.0–5.4 mm Hg and with tree nuts by 4.0–7.1 mm
CVD risk including hypertension clearly show Hg, and lowered diastolic BP with extra virgin
that MedDiets supplemented with either 1 L/ olive oil by 1.6–1.9 mm Hg and with nuts by
week extra virgin olive oil or 30 g/day of mixed 1.9–2.7 mm Hg compared to a low fat diet in
tree nuts (walnuts, almonds, and hazelnuts) sig- older high CVD risk individuals [47, 49]. The
nificantly reduced resting BP and 24-h ambula- large PREDIMED trial (7447 adults with high
tory BP compared with a low-fat control diet CVD risk; mean age 67 years; 57% women;
[46–49]. Two PREDIMED sub-trials (3 months - mean BMI 30; elevated BP; hypertensive meds;
4 years) showed that MedDiets significantly 13.2.4 Dietary Guideline Patterns

reduced diastolic BP by 0.65–1.5 mm Hg. A
2017 mechanism focused RCT (90 high CVD National dietary guideline based patterns, gener-
risk postmenopausal women; mean age 68 years; ally characterized by higher consumption of
mean BMI 32; mean BP 155/84 mm Hg; 1 year) fruit, vegetables, whole grains, legumes, seeds,
found beneficial effects for MedDiets supple- nuts, fish, and low fat-dairy and lower consump-
mented with extra virgin olive oil or nut poly- tion of processed meat, and sweets can lower BP
phenols related to increased serum nitric oxide and hypertension risk [53, 54]. A crossover RCT
and decreased serum endothelin-1 levels, which (31 subjects; mean age 55 years; mean BMI 31;
provide potential mechanistic pathways for BP baseline mean BP 120/74 mm Hg; 4 weeks)
lowering [46]. A possibly related mechanism found that high adherence to the 2010 US
between MedDiets and BP was indicated in a Dietary Guidelines or Healthy Eating Indices
crossover RCT (24 women; mean age 26 years; significantly reduced diastolic BP by 2.1 mm
mean BMI 23; 10 days) which found that the Hg, total cholesterol by 10.2 mg/dL, LDL-C by
MedDiet was associated with significantly ele- 6 mg/dL, HDL-C by 3 mg/dL, and triglycerides
vated contentment, alertness, memory recall and by 5.2 mg/dL compared to the typical Western
calmness compared to the usual diet [50]. (American) diet [53]. A RCT on UK dietary
guidelines (162 adults; mean age 52 years; 60%
women; 26% postmenopausal; mean BMI 25;
13.2.3 Nordic Diet mean BP 120/78 mm Hg; 12 weeks) showed that
high adherence significantly lowered systolic BP
As a Northern European regional alternative to by 3.5 mm Hg, diastolic BP by 2.1 mm Hg, CRP
the MedDiet, the gastronomically driven, envi- by 36%, LDL-C by 11.6 mg/dL, total choles-
ronmentally friendly, new healthy Nordic diet terol to HDL-C ratio by 0.13, triglycerides by
which focuses on fruit and vegetables (espe- 10.6 mg/dL, and waist size by 1.2 cm compared
cially berries, cabbages, root vegetables, and with the traditional British diet (control) [54].
legumes), potatoes, fresh herbs, plants and Overall, high adherence to the UK dietary guide-
mushrooms gathered from the wild, nuts, whole- lines diet reduced the risk of CVD by one-third,
grains, meats from livestock and game, fish, in healthy middle-aged and older men and
shellfish, and seaweed, along with diary and women.
eggs, has been shown to be effective in lowering
BP with and without weight loss [51, 52]. One
RCT (181 adults; mean age 42 years; mean BMI 13.2.5 Vegetarian Dietary Patterns
30; mean baseline BP 122/81 mm Hg; 6 months)
found that adherence to the new Nordic diet Vegetarian diets consist of increased or strict
resulted in significant reductions in systolic BP consumption of plant foods including fruits, veg-
by 5.1 mm Hg and diastolic BP by 3.2 mm Hg in etables, non-fried potatoes, whole grains,
addition to lowering body weight by 3.2 kg, legumes, soy foods, nuts and seeds and reduction
reducing waist circumference by 2.9 cm, and or elimination of consumption of meats, dairy
improved insulin sensitivity compared to the products, and eggs [55]. There are a range of veg-
usual Danish diet [51]. A second RCT (37 adults; etarian diets including the vegan diet (excludes
mean age 55 years; 68% women; mean BMI 31; all animal products), ovo-lacto-vegetarian
12 weeks) found that the new healthy Nordic (excludes all meat and seafood, but contains eggs
diet significantly lowered 24-h ambulatory dia- and dairy products), pesco-vegetarian (excludes
stolic BP by 4.4 mm Hg and mean arterial pres- meat but includes seafood), and semi-vegetarian
sure by 4.2 mm Hg compared to the control diet, (occasional meat allowed) [56]. A 2014 meta-
after 12 weeks, without significant changes in analysis of RCTs (7 RCTs; 311 adults; mean age
body weight [52]. 45 years) and observational studies (32 cohorts;
1.2
Odds Ratio for Hypertension

0.8
0.6
0.4
0.2
0
Vegan Lacto-ovo Partial vegetarian Omnivores
vegetarian
Dietary Patterns
Fig. 13.7 Association between vegetarian dietary type vs. omnivore diet and hypertension risk (p = 0.005 for vegan;
p = 0.02 for lacto-ovo vegetarian) [adapted from 58]
21,604 adults; mean age 47 years) suggest that Conclusions

vegetarian diets lower BP compared to omnivo- Prehypertension and hypertension are com-
rous diets [57]. For the RCTs, vegetarian diets mon and growing public health problems.
significantly lowered the mean systolic BP by Globally, the overall prevalence of elevated
4.8 mm Hg and diastolic BP by 2.2 mm Hg vs. BP is approaching 50% of adults
omnivorous diets. In observational studies, vege- age ≥ 25 years. The adult risk of CVD and
tarian diets were associated with significantly renal disease approximately doubles for each
lower mean systolic BP by 6.9 mm Hg and dia- 20/10 mm Hg incremental increase above
stolic BP by 4.7 mm Hg compared with the 115/75 mm Hg. Prehypertension is associ-
omnivorous diets. In the Adventist Health ated with 1.5- to 2-fold increases in CVD
Study-2 cohort (500 adults; mean age 63 years; events after age 55 years and a 43% increased
BMI 24–29) vegan and l acto-ovo vegetarians had risk of CHD. The major factors associated
significantly lower systolic and diastolic BP and with elevated BP and hypertension are aging,
reduced risk of hypertension compared to omniv- especially unhealthy aging associated with
orous Adventists (Fig. 13.7) [58]. A lifestyle overweight and obesity, poor dietary habits,
comparison RCT reported that Seventh-day inactivity or lack of exercise, and ineffective
Adventist vegetarians had significantly less stress management. High adherence to
hypertension and lower BP compared with healthy dietary patterns, including the
Mormon omnivores, effects which were indepen- DASH, the MedDiet, Nordic Diet, dietary
dent of differences in BMI or sodium intake [59]. guidelines-based, and vegetarian diets, are
effective in lowering BP, especially in older,
overweight or obese hypertensive and prehy-
13.2.6 Potential Mechanisms pertensive adults compared to Western diets.
Healthy dietary patterns reduce the intake
Table 13.4 summarizes potential healthy dietary levels of sodium, excessive energy and added
patterns BP and hypertension protective mecha- refined carbohydrate intake and increase the
nisms which are associated with higher dietary levels of BP lowering nutrients such as fiber,
fiber, lower energy density and are higher in other plant proteins, potassium, and other essential
essential nutrients and bioactive phytochemicals nutrients and bioactive phytochemicals asso-
[60–77]. ciated with a lower risk of hypertension and
Appendix A 385
elevated BP. Healthy dietary pattern mecha- lesterol and LDL-C levels, reducing
nisms associated with reduced hypertension oxidative and inflammatory stress, improv-
risk include; lowering the risk of weight ing insulin sensitivity to reduce atheroscle-
gain, stimulating colon microbiota, improv- rosis risk, and maintaining electrolyte
ing vascular health by normalizing total cho- balance.
ppendix A: Comparison of Western and healthy dietary patterns per

A
2000 kcals (approximated values)
Healthy
vegetarian
Healthy pattern
Western dietary USDA base DASH diet mediterranean (Lact-ovo
Emphasizes Refined grains, Vegetables, Potassium Whole Vegetables, Plant foods:
low fiber foods, fruit, rich grains, fruit, Vegetables,
red meats, whole- vegetables, vegetables, whole- fruits, whole
sweets, and grain, and fruits, and fruit, dairy grains, grains, nuts,
solid fats low-fat low fat milk products, legumes, seeds, and
milk products olive oil, and nuts, seeds, soy foods
moderate milk
wine products,
and soy
foods
Includes Processed Enriched Whole-grain, Fish, nuts, Eggs, Non-dairy
meats, sugar grains, lean poultry, fish, seeds, and non-dairy milk
sweetened meat, fish, nuts, and pulses milk alternatives
beverages, and nuts, seeds, seeds alternatives,
fast foods and and
vegetable vegetable
oils oils
Limits Fruits and Solid fats Red meats, Red meats, No red or No animal
vegetables, and and added sweets, and refined white meats, products
whole-grains sugars sugar- grains, and or fish;
sweetened sweets limited
beverages sweets
(% Total kcal)
Protein (% Total 16 17 18 16 14 13
kcal)
Total fat (% Total 33 32 27 34 32 30
kcal)
Total kcal)
Unsat. fat (% 22 25 21 24 26 25
Total kcal)
Fiber (g) 16 31 29+ 31 35+ 40+
Potassium (mg) 2800 3350 4400 3350 3300 3650
Vegetable oils (g) 19 27 25 27 19–27 18–27
Sodium (mg) 3600 1790 1100 1690 1400 1225
Healthy
vegetarian
Healthy pattern
Western dietary USDA base DASH diet mediterranean (Lact-ovo
Plant food groups
Fruit (cup) ≤ 1.0 2.0 2.5 2.5 2.0 2.0
Vegetables (cup) ≤ 1.5 2.5 2.1 2.5 2.5 2.5
Whole-grains 0.5 3.0 4.0 3.0 3.0 3.0
(oz.)
Legumes (oz.) – 1.5 0.5 1.5 3.0 3.0+
Nuts/seeds (oz.) 0.5 0.6 1.0 0.6 1.0 2.0
Soy products 0.0 0.5 – – 1.1 1.5
(oz.)
terns. 2015;1–9
tension. A systematic review and meta-analysis of

References cohort studies. Hypertension. 2017;69. https://doi.
org/10.1161/HypertensionAHA.116.08994.
1. World Health Organization. A global brief on hyper- 9. Frisoli TM, Schmieder RE, Grodzicki T, Messerli
tension. Silent killer, global public health crisis. FH. Beyond salt: lifestyle modifications and blood
Geneva: WHO; 2013. http://www.who.int/cardiovas- pressure. Eur Heart J. 2011;32:3081–7.
cular_diseases/publications/global brief. hyperten- 10. Steffen PR, Smith TB, Larson M, et al. Acculturation to
sion/en/. Accessed April 2017. western society as a risk factor for high blood pressure: a
2. Castro I, Waclawovsky G, Marcadenti A. Nutrition meta-analytic review. Psychosom Med. 2006;68:386–97.
and physical activity on hypertension: implication of 11. Rebholz CM, Crews DC, Grams ME, et al. DASH
current evidence and guidelines. Curr Hypertens Rev. (Dietary Approaches to Stop Hypertension) diet and
2015;11(2):91–9. risk of subsequent kidney disease. Am J Kidney Dis.
3. Mendis S, Puska P, Norrving B, editors. Global atlas 2016;68(6):853–61.
on cardiovascular disease prevention and control. 12. Dickinson HO, Mason JM, Nicolson DJ, et al.

Geneva: World Health Organization; 2011. Lifestyle interventions to reduce raised blood pres-
4. Mozaffarian D, Benjamin EJ, Go AS, et al. On sure: a systematic review of randomized controlled
behalf of the American Heart Association Statistics trials. J Hypertens. 2006;24(2):215–33.
Committee and Stroke Statistics Subcommittee. Heart 13. Bhupathiraju SN, Tucker KL. Coronary heart disease
disease and stroke statistics-2016 update: a report prevention: nutrients, foods, and dietary patterns. Clin
from the American Heart Association. Circulation. Chim Acta. 2011;412:1493–514.
2016;133:e38–e360. 14. Sotos-Prieto M, Bhupathiraju SN, Mattei J, et al.

5. Livingstone KM, McNaughton SA. Dietary pat- Changes in diet quality and risk of cardiovascular
terns by reduced rank regression are associated with disease among US men and women. Circulation.
obesity and hypertension in Australian adults. Br 2015;32(23):2212–9. https://doi.org/10.1161/
J Nutr. 2017;117:248–59. https://doi.org/10.1017/ circulationAHA.115.017158.
S0007114516004505. 15. Wright JD, Hughes JP, Ostchega Y, et al. Mean sys-
6. Atlas SA. The renin-angiotensin aldosterone system: tolic and diastolic blood pressure in adults aged 18
pathophysiological role and pharmacologic inhibition. and over in the United States, 2001-2008. Natl Health
J Manag Care Pharm. 2007;13(8):S9–S20. Stat Rep. 2011;35:1–24.
7. Appel LJ. The effects of dietary factors on blood pres- 16. Maraj I, Makaryus JN, Ashkar A, et al. Hypertension
sure. Cardiol Clin. 2017;35:197–212. management in the high cardiovascular risk popu-
8. Liu X, Zhang D, Liu Y, et al. Dose-response asso- lation. Int J Hypertens. 2013;2013:7. https://doi.
ciation between physical activity and incident hyper- org/10.1155/2013/382802.
References 387
17. Wang C-J, Shen Y-X, Liu Y. Empirically derived

ized controlled trials. Nutr Metab Cardiovasc Dis.
dietary patterns and hypertension likelihood: a meta- 2014;24:1253–61.
analysis. Kidney Blood Press Res. 2016;41:570–81. 30. Chiu S, Bergeron N, Williams PT, et al. Comparison of
https://doi.org/10.1159/ 000443456. the DASH (Dietary Approaches to Stop Hypertension)
18. Zhang X-Y, Shu L, Si C-J, et al. Dietary patterns, alco- diet and a higher-fat DASH diet on blood pressure and
hol consumption and risk of coronary heart disease in lipids and lipoproteins: a randomized controlled trial.
adults: a meta-analysis. Forum Nutr. 2015;7:6582– Am J Clin Nutr. 2016;103:341–7.
605. https://doi.org/10.3390/nu7085300. 31. Al-Solaiman Y, Jesri A, Mountford WK, et al. DASH
19. Eckel RH, Jakicic JM, Ard JD, et al. 2013 AHA/ lowers blood pressure in obese hypertensives beyond
ACC guideline on lifestyle management to reduce potassium, magnesium and fiber. J Hum Hypertens.
cardiovascular risk: a report of the American College 2010;24(4):237–46.
of Cardiology/American Heart Association Task 32. Blumenthal JA, Babyak MA, Hinderliter A, et al.

Force on Practice Guidelines. J Am Coll Cardiol. Effects of the DASH diet alone and in combination
2014;63:2960–84. with exercise and weight loss on blood pressure and
20. Dietary Guidelines Advisory Committee. Advisory cardiovascular biomarkers in men and women with
report to the Secretary of Health and Human Services high blood pressure: the ENCORE Study. Arch Intern
and the Secretary of Agriculture. Part D. Chapter 2: Med. 2010;170(2):126–35.
Dietary patterns, foods and nutrients and health out- 33. Appel LJ, Sacks FM, Carey VJ, et al. Effects of

comes; 2015. p. 1–35. protein, monounsaturated fat, and carbohydrate
21. Ndanuko RN, Tapsell LC, Charlton KE, et al. Dietary intake on blood pressure and serum lipids results
patterns and blood pressure in adults: a systematic of the OmniHeart Randomized Trial. JAMA.
review and meta-analysis of randomized controlled 2005;294:2455–64.
trials. Adv Nutr. 2016;7:76–89. 34. Nowson CA, Worsley A, Margerison C, et al. Blood
22. Ndanuko RN, Tapsell LC, Charlton KE, et al.
pressure change with weight loss is affected by diet
Associations between dietary patterns and blood type in men. Am J Clin Nutr. 2005;81:983–9.
pressure in a clinical sample of overweight adults. 35. Appel LJ, Champagne CM, Harsha DW, et al. Effects
J Acad Nutr Diet. 2017;117(2):228–39. https://doi. of comprehensive lifestyle modification on blood
org/10.1016/j.and.2016.07.019. pressure control: main results of the PREMIER clini-
23. Harrington JM, Fitzgerald AP, Kearney PM, et al. cal trial. JAMA. 2003;289(16):2083–93.
DASH diet score and distribution of blood pressure 36. Miller ER, Erlinger TP, Young DR, et al. Results of
in middle-aged men and women. Am J Hypertens. the diet, exercise, and weight loss intervention trial
2013;26(11):1311–20. (DEW-IT). Hypertension. 2002;40:612–8.
24. Barak F, Falahi E, Keshteli AH, et al. Adherence to the 37. Appel LJ, Moore TJ, Obarzanek E, et al. A clinical
Dietary Approaches to Stop Hypertension (DASH) trial of the effects of dietary patterns on blood pres-
diet in relation to obesity among Iranian female sure. N Engl J Med. 1997;336:1117–24.
nurses. Public Health Nutr. 2014;18(4):705–12. 38. Nunez-Cordoba JM, Valencia-Serrano F, Toledo E,
25. Lin P-H, Yeh W-T, Svetkey LP, et al. Dietary intakes et al. The Mediterranean diets and incidence of hyper-
consistent with the DASH dietary pattern reduce tension. Am J Epidemiol. 2009;169:339–46.
blood pressure increase with age and risk for stroke 39. Nunez-Cordoba JM, Alonso A, Beunza JJ, et al. Role
in a Chinese population. Asia Pac J Clin Nutr. of vegetables and fruits in Mediterranean diets to pre-
2013;22(3):482–91. vent hypertension. Eur J Clin Nutr. 2009;63:605–12.
26. Moore TJ, Alsabeeh N, Apovian CM, et al. Weight, 40. Psaltopoulou T, Naska A, Orfanos P, et al. Olive oil,
blood pressure, and dietary benefits after 12 months the Mediterranean diet, and arterial blood pressure:
of a Web-based Nutrition Education Program (DASH the Greek European Prospective Investigation into
for Health): longitudinal observational study. J Med Cancer and Nutrition (EPIC) study. Am J Clin Nutr.
Internet Res. 2008;10(4):e52. https://doi.org/10.2196/ 2004;80:1012–8.
jmir.1114. 41. Toledo E, Carmona-Torre FD, Alonso A, et al.

27. Gay HC, Rao SG, Vaccarino V, Ali MK. Effects of Hypothesis-oriented food patterns and incidence
different dietary interventions on blood pressure sys- of hypertension: 6-year follow-up of the SUN
tematic review and meta-analysis of randomized con- (Seguimiento Universidad de Navarra) prospective
trolled trials. Hypertension. 2016;67:733–9. https:// cohort. Public Health Nutr. 2010;13:338–49.
doi.org/10.1161/HypertensionAHA.115.06853. 42. Timmermans S, Steegers-Theunissen RPM, Vujkovic
28. Siervo M, Lara J, Chowdhury S, et al. Effects of the M, et al. Major dietary patterns and blood pressure
Dietary Approach to Stop Hypertension (DASH) diet patterns during pregnancy: The Generation R Study.
on cardiovascular risk factors: a systematic review Am J Obstet Gynecol. 2011;205:337.e1–e12. https://
and meta-analysis. Br J Nutr. 2015;113:1–15. doi.org/10.1016/j.ajog.2011.05.013.
29. Saneei P, Salehi-Abargouei A, Esmaillzadeh A,
43. Nissensohn M, Roman-Vinas B, Sanchez-Villegas A,
Azadbakht L. Influence of Dietary Approaches to et al. The effect of the Mediterranean diet on hyper-
Stop Hypertension (DASH) diet on blood pressure: tension: a systematic review and meta-analysis. J Nutr
a systematic review and meta-analysis on random- Educ Behav. 2016;48:42–53.
44. Huo R, Du T, Xu Y, et al. Effects of Mediterranean-style 58. Pettersen BJ, Anousheh R, Fan J, et al. Vegetarian
diet on glycemic control, weight loss and cardiovas- diets and blood pressure among white subjects: results
cular risk factors among type 2 diabetes individuals: a from the Adventist Health Study-2. Public Health
meta-analysis. Eur J Clin Nutr. 2015;69(11):1200–8. Nutr. 2012;15(10):1909–16.
https://doi.org/10.1038/ejcn.2014.243. 59. Beilin LJ, Rouse IL, Armstrong BK, et al. Vegetarian
45. Nordmann AJ, Suter-Zimmermann K, Bucher HC, diet and blood pressure levels: incidental or causal
et al. Meta-analysis comparing Mediterranean to low association? Am J Clin Nutr. 1988;48:806–10.
fat diets for the modification of cardiovascular risk 60. Sanchez-Muniz FJ. Dietary fibre and cardiovascular
factors. Am J Med. 2011;124(9):841–51. health. Nutr Hosp. 2012;27(1):31–45.
46. Storniolo CE, Casillas R, Bulló M, et al. A
61. Evans CEL, Greenwood DC, Threapleton DE, et al.
Mediterranean diet supplemented with extra vir- Effects of dietary fibre type on blood pressure: a
gin olive oil or nuts improves endothelial markers systematic review and meta-analysis of randomized
involved in blood pressure control in hypertensive controlled trials of health individuals. J Hypertens.
women. Eur J Nutr. 2017;56(1):89–97. https://doi. 2015;33(5):897–911.
org/10.1007/s00394-015-1060-5. 62. Czernichow S, Blacher J, Hercberg S. Antioxidant
47. Doménech M, Roman P, Lapetra J, et al. Mediterranean vitamins and blood pressure. Curr Hypertens Rep.
diet reduces 24-hour ambulatory blood pressure, 2004;6(1):27–30.
blood glucose, and lipids one-year randomized, clini- 63. Gammone MA, Riccioni G, D’Orazio N. Carotenoids:
cal trial. Hypertension. 2014;64:69–76. potential allies of cardiovascular health? Food Nutr
48. Toledo E, Hu FB, Estruch R, et al. Effect of
Res. 2015;59. https://doi.org/10.3402/fnr.v59.26762.
the Mediterranean diet on blood pressure in the 64. Hodgson JM, Croft KD, Woodman RJ, et al. Effects
PREDIMED trial: results from a randomized con- of vitamin E, vitamin C and polyphenols on the rate
trolled trial. BMC Med. 2013;11:207. https://doi. of blood pressure variation: results of two randomised
org/10.1186/1741-7015-11-207. controlled trials. Br J Nutr. 2014;112(9):1551–61.
49. Estruch R, Martínez-González MA, Corella D, et al. https://doi.org/10.1017/S0007114514002542.
Effects of a Mediterranean-style diet on cardiovascu- 65. Koliaki C, Katsilambros N. Dietary sodium, potas-
lar risk factors: a randomized trial. Ann Intern Med. sium and alcohol: key players in the pathophysiology,
2006;145(1):1–11. prevention, and treatment of human hypertension.
50. Lee J, Pase M, Pipingas A, et al. Switching to a 10-day Nutr Rev. 2013;71(6):402–11.
Mediterranean-style diet improves mood and cardio- 66. Lee YP, Puddey IB, Hodgson JM. Protein, fibre,

vascular function in a controlled crossover study. blood pressure: potential benefit of legumes. Clin Exp
Nutrition. 2015;31:647–52. Pharmacol Physiol. 2008;35:473–6.
51. Poulsen SK, Due A, Jordy AB, et al. Health effect of 67. Rangel-Huerta OD, Pastor-Villaescusa B, Aguilera

the New Nordic Diet in adults with increased waist CM, Gil A. A systematic review of the efficacy of bio-
circumference: a 6-mo randomized controlled trial. active compounds in cardiovascular disease: phenolic
Am J Clin Nutr. 2014;99:35–45. compounds. Forum Nutr. 2015;7:5177–216.
52. Brader L, Uusitupa M, Dragsted LO, Hermansen
68. Ried K, Fakler P. Protective effect of lycopene on
K. Effects of an isocaloric healthy Nordic diet on serum cholesterol and blood pressure: meta-analyses
ambulatory blood pressure in metabolic syndrome: of intervention trials. Maturitas. 2011;68:299–310.
a randomized SYSDIET sub-study. Eur J Clin Nutr. 69. Ros E, Hu FB. Consumption of plant seeds and car-
2014;68:57–63. diovascular health epidemiological and clinical trial
53. Schroeder N, Park Y-H, Kang M-S, et al. A random- evidence. Circulation. 2013;128:553–65.
ized trial on the effects of 2010 Dietary Guidelines for 70. Jayalath VH, de Souza RJ, Sievenpiper JL, et al.

Americans and Korean Diet Patterns on cardiovascu- Effect of dietary pulses on blood pressure: a system-
lar risk factors in overweight and obese adults. J Acad atic review and meta-analysis of controlled feeding
Nutr Diet. 2015;115:1083–92. trials. Am J Hypertens. 2014;27(1):56–64.
54. Reidlinger DP, Darzi J, Hall WL, et al. How effective 71. Clark JL, Zahradka P, Taylor CG. Efficacy of flavo-
are current dietary guidelines for cardiovascular dis- noids in the management of high blood pressure. Nutr
ease prevention in healthy middle-aged and older men Rev. 2015;73(12):799–822.
and women? A randomized controlled trial. Am J Clin 72. Hobbs DA, George TW, Lovegrove JA. The effects of
Nutr. 2015;101:922–30. dietary nitrate on blood pressure and endothelial func-
55. Orlich MJ, Jaceldo-Siegl K, Sabaté J, et al. Patterns tion: a review of human intervention studies. Nutr Res
of food consumption among vegetarians and non- Rev. 2013;26:210–22.
vegetarians. Br J Nutr. 2014;112(10):1644–53. 73. Mohammadifard N, Salehi-Abargouei A, Salas-

56.
Rizzo NS, Jaceldo-Siegl K, Sabate J, Fraser Salvadó J, et al. The effect of tree nut, peanut, and
GE. Nutrient profiles of vegetarian and non-vegetarian soy nut consumption on blood pressure: a systematic
dietary patterns. J Acad Nutr Diet. 2013;113:1610–9. review and meta-analysis of randomized controlled
57. Yokoyama Y, Nishimura K, Barnard ND, et al.
clinical trials. Am J Clin Nutr. 2015;101:966–82.
Vegetarian diets and blood pressure. A meta-analysis. 74. Pins JJ, Geleva D, Keenan JM. Do whole-grain oat
JAMA Intern Med. 2014;174(4):577–87. cereals reduce the need for antihypertensive medi-
References 389
cations and improve blood pressure control? J Fam 76. Yang T, Santisteban MM, Rodriguez V, et al. Gut
Pract. 2002;14:353–9. Dysbiosis is linked to hypertension. Hypertension.
75. Li B, Li F, Wang L, Zhang D. Fruit and vegetables con- 2015;65:1331–40.
sumption and risk of hypertension: a meta-analysis. 77. Pluznick JL. A novel SCFA receptor, the micro-

J Clin Hypertens (Greenwich). 2016;18(5):468–76. biota, and blood pressure regulation. Gut Microbes.
https://doi.org/10.1111/jch.12777. 2014;5(2):202–7.
Whole Plant Foods
and Hypertension 14
Keywords
Blood pressure • Hypertension • Aging • Overweight • Obesity • Microbiota
• Electrolytes • Whole grains • Fruit • Vegetables • Legumes • Soy • Nuts
• Flaxseed
Key Points kcals or those rich in β-glucan; fruits and

• Whole (and minimally processed) plant foods vegetables rich in polyphenols or nitrates and
usually contain some mixture of blood pres- their 100% juices have been found to lower
sure (BP) lowering bioactive nutrients and systolic BP; and two daily servings of dietary
phytochemicals such as dietary fiber, potas- pulses or 40 g soy protein are effective in
sium, magnesium, polyphenols, unsaturated lowering BP. Flaxseeds and sesame seeds
fat, and plant protein and are lower in sodium tend to be more effective than nuts in lower-
and sugar compared to highly processed plant ing BP.
foods. • Tea and coffee have different effects on BP.
• Whole plant foods are more effective at reduc- Both black and green tea (> 2 cups/d) mod-
ing BP in adults who are ≥45 years, hyperten- estly lower BP in hypertensive individuals.
sive and obese than adults <45 years, Coffee (>3 cups/d) does not increase hyper-
normotensive, or lean. tension risk in normotensive people but hyper-
• Prospective studies show that the consump- tensive individuals may be more sensitive to
tion of healthy diets with ≥3 daily servings of acute increases in BP after coffee
whole grains, especially oats and barley rich consumption.
in β-glucan, and ≥5 daily servings of fruits • The potential mechanisms by which whole
and vegetables, especially when including ≥4 plant foods may reduce blood pressure and
weekly servings of broccoli, carrots, tofu or hypertension risk are; reducing the risk of
soybeans, raisins, grapes and apples, are weight gain, enhancing insulin sensitivity,
associated with lower hypertension risk com- improving vascular endothelial function,
pared to Western diets. slowing the rate of arterial plaque build-up,
• RCTs support the effectiveness of whole maintaining electrolyte balance, and stimulat-
grains in lowering BP, especially at 50 g/1000 ing a healthier microbiota ecosystem.

392 14 Whole Plant Foods and Hypertension
14.1 Introduction (hypertension, unhealthy lipoprotein profile, and

regular smoking) the faster the rate of carotid
Elevated blood pressure (BP) and hypertension intima-media thickening or atherosclerosis pro-
are major risk factors for stroke, coronary heart gression (Fig. 14.1) [10]. Dairy products, espe-
disease (CHD), damage to retinal blood vessels, cially lower fat products such as plain yogurt,
and kidney disease/renal failure [1–4]. The adult and healthy whole plant foods have been shown
risk of cardiovascular disease (CVD) doubles for to reduce the risk of elevated BP and hyperten-
each 20/10 mm Hg incremental increase above sion [14]. An overview of the effects of specific
115/75 mm Hg. Elevated BP, including prehy- foods on BP are summarized in Table 14.1 [14–
pertension (systolic BP >120–139 mm Hg or 30]. The primary objective of this chapter is to
diastolic BP >80–89 mm Hg), or hypertension provide a review of the role of whole (minimally
(systolic BP ≥140 mm Hg or diastolic processed) plant foods in helping to prevent and
BP > 90 mm Hg), is a common and growing pub- manage elevated BP and hypertension.
lic health problem [5, 6]. By 2025, because of
population growth and an aging population, it is
projected that about 1.5 billion individuals glob-
ally will have hypertension [1–4]. Aging, over- 14.2 Whole Plant Foods
weight or obesity, and lifestyle factors such as
poor nutritional quality dietary patterns, excess Whole and minimally processed plant foods
energy intake and salt intake, sedentary life- (whole plant foods) are generally higher in fiber
styles, alcohol intake, smoking, and anxiety and and major dietary sources of hypertension pro-
stress are important underlying controllable fac- tective nutrients (e.g., vitamins E and C, and trace
tors associated with hypertension risk, because minerals such as selenium and copper), and phy-
of increased activity of the renin-angiotensin- tochemicals (carotenoids, flavonoids and phytos-
aldosterone system, insulin resistance, and terols), which are often lost in the processing of
reduced kidney function [1–12]. Foods and diets highly refined foods [31–42]. Whole plant food
with low dietary energy density, high fiber den- composition is highly variable and is summa-
sity and that contain primarily natural sugar rized in (Appendix A). Potential whole plant
sources such as whole fruits, tend to be associ- foods BP related mechanisms are: (1) reducing
ated with lower prevalence of overweight or obe- dietary energy density and increasing satiety and
sity, whereas foods with low fiber density and satiation, which reduce the risk of weight gain or
high in added sugars intake (primarily from obesity; (2) enhancing insulin sensitivity, which
chocolate and fruit drinks) are associated with may improve vascular and endothelial function;
increased prevalence of overweight or obesity (3) promoting healthier lipid and lipoprotein pro-
[5]. Foods low in fiber density and high in the files, attenuating elevated systemic inflammation
ratio of sodium to potassium and high in satu- and LDL-oxidation for improved endothelial
rated fat to polyunsaturated fat ratios are associ- health and a slower rate of arterial plaque build-
ated with increased prevalence of hypertension up; (4) maintaining electrolyte balance to

and prehypertension [5]. Guidelines for preven- improve RAAS homeostatic extracellular fluid
tion, treatment and control of elevated BP/hyper- volume; and (5) improving colon health via the
tension focus on lifestyle modifications including stimulation of a healthier microbiota and
weight loss and maintenance, reduced salt increased fermentation of fiber to short chain
intake, increased fruit and vegetable consump- fatty acids (SCFAs), leading to potentially
tion, routine participation in physical activity, improved cardiometabolic control [4, 9–11, 17,
cessation of smoking, limiting of alcohol con- 19, 39–52]. However, whole plant foods are con-
sumption, and anxiety and stress control [3–19]. sumed at low levels in the typical Western diet
The higher the number of risk factors with an estimated >90% of US adults not meeting
2.5
Odds Ratio for High Carotid IMT Progression Rates 2
1.5
0.5
0
0 1 2 3
# of Risk Factors
Fig. 14.1 Association between the number of risk factors (hypertension, unhealthy lipoprotein profile, and regular
smoking) and carotid intima-media thickness (IMT) progression rate from The Malmo Diet and Cancer Study (adapted
from [10])
Table 14.1 Effects of specific foods on risk of elevated 14.2.1 Whole-Grains

blood pressure [14–30]
14.2.1.1 Background Strength of
Dietary factors Postulated effects evidence
Whole-grains contain basically the intact kernel
Low-fat dairy and milk Decreased +/+
with its full complement of fiber, vitamins, min-
High-fat dairy and cheese Uncertain/neutral +/−
erals and phytochemicals, whereas refined grain
Processed meat Increased +
products are mainly comprised of the endosperm
Lean red meat Uncertain/neutral +/−
with most of the fiber, vitamins, minerals and
Whole-grains (soluble Decreased +
fiber, oats/psyllium)
phytonutrients removed during processing [54,
Whole-grains (insoluble Uncertain/neutral +/−
55]. The US dietary guidelines recommend ≥3
fiber, wheat/rice) whole-grain servings/day and ≤3 refined grains
Fruits and vegetables Decreased +/+ servings/day to promote health and reduce risk of
Legumes Decreased + various chronic diseases [41]. However, only
Nuts and seeds Decreased + about 1% of Americans follow the recommenda-
Key to evidence: +⁄− = limited or equivocal evidence; tion for whole-grain intake as the average
+ = suggestive evidence from observational studies and American’s intake is <1 ounce whole grains/day
clinical trials; +/+ = persuasive evidence, primarily from and 70% exceed the recommended intake for
clinical trials
refined grains [41, 53]. A 2017 RCT in healthy
men and postmenopausal women found that sub-
the recommended minimal levels of whole- stituting whole-grains for refined grains in the
grains, fruit, vegetables, legumes, or nuts/seeds diet and increasing dietary fiber intake led to
to maintain optimal health and weight control favorable effects that resulted in a 100-kcal/day
outcomes [41, 53]. energy deficit compared with the effects of a
typical American diet that is low in whole grains (28,926 women; mean age 54 years; 10-year fol-
and fiber [56]. This study provides mechanistic low-up; 8722 incidents of hypertension) found a
insights that support an inverse association significantly reduced risk of hypertension by
between whole-grain intake and BMI and adipos- 11% for whole grains, whereas refined grain did
ity, which are also documented in observational not reduce hypertension risk in women (Fig. 14.3)
studies. The weight control effects of consuming [22]. In this study women consuming >4 daily
adequate whole-grain intake may also help to whole grain servings had a 23% lower risk of
protect against hypertension [2–5]. hypertension. The Physicians’ Health Study I
(13,368 men; average age 52 years; 16 years of
follow-up; 7267 cases of hypertension) observed
14.2.2 Prospective Cohort Studies a significantly lower hypertension risk for daily
intake of whole-grain breakfast cereal by 20%
Prospective studies consistently support the and for refined breakfast cereal by 14% com-
inverse association between whole-grain intake pared to no breakfast cereal intake in lean, over-
and risk of hypertension [22, 57–61]. Several weight and obese men [61].
meta-analyses of prospective cohort studies with
quantitative measures of whole-grain intake and
clinical cardiovascular outcomes found that 14.2.3 Randomized Controlled
greater whole grains intake (≥48 g or three serv- Trials (RCTs)
ings/day vs. ≤0.2 servings/day) was associated
with a 21% lower risk of cardiovascular disease In general, the effects of whole-grain products
(CVD) and coronary heart disease (CHD) [57– on BP depends on their composition and physi-
59]. The Health Professionals Follow-Up Study cal properties with whole-grains rich in β-glucans
(31,684 men; mean age 52 years; 18 years of (e.g., oat or barley) more effective in lowering
follow-up; 9227 cases of incident hypertension) BP than insoluble fiber (e.g., whole wheat breads
showed that men with a median intake of about and breakfast cereals) (Table 14.2) [24, 62–73].
three daily whole grain servings (46 g/day) had a A Swedish double blind crossover RCT (40
significantly lower risk of hypertension by19% overweight and obese men and women; age
(Fig. 14.2) [60]. The Women’s Health Study 40 years; macronutrient composition was
Whole-grains (p-trend <.0001) Bran (p-trend =.002) Germ (p-trend =.11)

1.1
Relative Risk of Hypertension
1.0
0.9
0.8
0.7
1 2 3 4 5
Quintile of Daily Intake *
Fig. 14.2 Association between increasing whole-grains, bran and germ intake and hypertension incidence risk in men,
multivariate adjusted (adapted from [59]). *Whole-grains: Q-5 (46 g); Bran: Q-5 (12 g); Germ: Q-5 (2.4 g)
Whole-grains (p-trend=.001) Refined-grains (p-trend =.46)

1.1
Relative Risk of Hypertension 1.05
0.95
0.9
0.85
0.8
1 2 3 4 5
Quintile of Daily Intake *
Fig. 14.3 Association between whole- and refined-grain intake and hypertension risk in women (multivariate adjusted)
(adapted from [60]).*Whole-grains: Q-5 (3.1 servings); refined-grains: Q-5 (4.1 servings)
Table 14.2 Summaries of whole-grain RCTs on blood pressure (BP)

Systematic Review and Meta-analyses
Evans et al. (2015). Assess 5 RCTs on oats; 400 subjects; median Diets rich in β-glucans with a median
the effects of oat and barley increase of 4 g β-glucans; ≥6 weeks increase of 4 g β-glucans from whole oats
whole-grain foods on BP or oat bran enriched foods or oat-based
(UK) [61] breakfast cereals compared with similar
wheat-based test foods significantly lowered
systolic BP by 2.9 mm Hg and diastolic BP
by 1.5 mm Hg
Thies et al. (2014). 25 RCTs; primarily oat bran or Only three trials found oat consumption to
Systematically review the breakfast cereals vs. low fiber or significantly reduce BP vs. control products
literature describing long-term wheat bran cereals; β-glucan level,
intervention studies that and design details not provided
investigated the effects of oats
or oat bran on CVD risk
factors including BP (US) [62]
Representative RCTs
Kirwan et al. (2016). Double-blind, Crossover RCT: 40 The whole-grain diet significantly reduced
Evaluate the efficacy of whole overweight or obese men and women; diastolic BP by 5.8 mm Hg compared to a
grains compared with refined mean age 40 years; complete reduction of 1.6 mm Hg after the refined
grains on body composition, whole-grain and refined-grain diets grain control diet (p = .01). Whole-grain
hypertension, and related were provided for two 8-week intake preserved circulating total
mediators of CVD in periods, with a 10-week washout adiponectin concentrations compared with a
overweight and obese adults between diets. Macronutrient 1.4 μg/mL decline after refined-grain intake
(Sweden) [63] composition was matched, except for (p = .05). Decreases in diastolic blood
the inclusion of either whole grains or pressure correlated with the circulating
refined grains (50 g/1000 kcal in each adiponectin concentration (r = .35, p = .04)
diet); 33 subjects completed the trial; (Fig. 14.4). Substantial reductions in body
five participants taking weight, fat loss, systolic blood pressure,
antihypertensive medication were total cholesterol, and LDL cholesterol were
instructed to maintain medication observed during both diet periods, with no
usage throughout the study significant differences between the two diets
(continued)
Tighe et al. (2010). Examine Parallel RCT: 233 healthy adults; After 6 and 12 weeks, the whole-grain
the effects of consumption of mean age 52 years, mean BMI 28; wheat and oats group significantly
whole-grains vs. refined foods mean BP 131/79 mm Hg; diets: three decreased systolic BP by 3.7 mm Hg more
on markers of cardiovascular daily servings of wheat and oats (one than the refined group. There was no
disease including blood serving of whole wheat bread and two significant diastolic BP difference between
pressure (UK) [64] servings of oat cereals); three the groups
servings whole wheat bread and
cereals or three servings of refined
cereals and white bread; 12 weeks
Maki et al. (2007). Assess Double-blind, Parallel RCT: 97 Obese subjects consuming foods rich in oat
the clinical effects of overweight/obese hypertensive β-glucan had significantly lowered systolic
consuming foods containing subjects; 56% women; mean age BP by 8.3 mm Hg and diastolic BP by
oat β-glucan on BP (US) [65] 49 years; mean systolic BP 3.9 mm Hg vs. the control group. No
130–179 mm Hg/diastolic BP significant differences in blood pressure
85–109 mm Hg; 7.7 g oat β-glucan responses were observed in normal BMI
foods or control foods with 0 g subjects
β-glucan; 12 weeks
Behall et al. (2006). Compare Crossover RCT: 25 mildly AHA step 1 diet with whole-grains lowered
the effects of whole wheat and hypercholesterolemic, overweight BP compared to a AHA step 1 diet with
brown rice vs. barley adults; seven men, nine premenopausal white rice; whole-grains lowered systolic
whole-grain (β-glucan) diet on and nine post-menopausal women; BP by 1.4–6.7 mm Hg and diastolic BP
BP (US) [66] mean age 47 years; mean BP 2.9–3.7 mm Hg
117/71 mm Hg; AHA Step 1 diet with
white rice control; 20% of energy
replaced with whole-grains: whole
wheat/brown rice, barley, or mixture of
both; 5 weeks
He et al. (2004). Investigate Double-blind, Parallel RCT: 110 Oat β-glucan intake significantly reduced
the effects of whole grain oats healthy adults; age 30–65 years; stage systolic BP by 2.0 mm Hg and
on BP (US) [67] 1 hypertension; foods containing insignificantly lowered diastolic BP by
whole-grain oats with 8 g β-glucan or 1.0 mm Hg
control foods with 0 g β-glucan; 6 and
12 weeks
Davy et al. (2002). Evaluate Parallel RCT: 36 overweight/obese Oat consumption insignificantly lowered
the clinical effect of whole- men; mean age 59 years; elevated BP; systolic and diastolic BP by 1 mm Hg
grain oat beta-glucan on BP addition of 14 g/day of fiber including compared to wheat. Also, no significant
(US) [68] oat cereal (5.5 g β-glucan) or wheat differences in 24-h, daytime and nighttime
cereals (no β-glucan); resting and BP observed between oat and wheat in this
ambulatory BP; 12 weeks study. Subjects in both groups significantly
increased body weight by 0.8 kg
Pins et al. (2002). Study the Parallel RCT: 88 adults on 73% of oat β-glucan participants vs. 42% in
effects of whole-grain antihypertensive medications; mean the wheat group either stopped or reduced
oat-based cereals vs. refined age 48 years; mean BP below their β blocker medications by half. Those
grain wheat-based cereals to 140/88 mm Hg; oatmeal or oat in the oats group who did not experience a β
determine their effects on BP squares (3 g β-glucan) vs. wheat blocker reduction had a significant 6 mm
and β blocker medications crisps or hot wheat cereal (0 Hg decrease in systolic BP vs. the wheat
(US) [24] β-glucan); 12 weeks group
Keenan et al. (2002). Parallel RCT: 18 hypertensives and The oat cereal group significantly lowered
Evaluate the anti-hypertensive hyperinsulinemic overweight/obese systolic BP by 7.5 mm Hg and diastolic BP
effects of β-glucan-rich whole adults, age 20–70 years; diets: oat by 5.5 mm Hg compared to the control
oat cereals when added to a cereal group (standardized to 5.5 g group. In the oat cereal group, a trend
standard American diet β-glucan/day) or a low-fiber cereal toward higher insulin sensitivity suggested
(US) [69] control; 6 weeks a possible mechanism for BP lowering
Saltzman et al. (2001). Parallel RCT: 43 overweight/obese The hypocaloric oat diet significantly
Investigate the effects of a adults; mean age 45 years; mean lowered systolic BP by 5 mm Hg and
hypo-caloric diet with and baseline BP of 118/71 mm Hg; insignificantly decreased diastolic BP by
without oats on BP (US) [70] 8 weeks); hypocaloric diets: 45 g 1 mm Hg vs. the control diet. There was no
oats/day or no added oat control; significant difference in weight loss
6-weeks between the two groups
Wheat Bran
Kestin et al. (1990). Double-blind, Crossover RCT: 24 The baseline BP was unaltered by the
Investigate the clinical effects healthy men; mean BP addition of wheat bran
of cereal brans on BP 125/79 mm Hg; elevated blood lipids;
(Australia) [71] 11.8 g fiber/day from wheat bran vs.
baseline diet; 4 weeks
Fehily et al. (1986). To Crossover RCT: 201 healthy adults; There was no significant difference in BP
evaluate the clinical effect of mean age approx. 40 years; 73% men; between the whole-grain wheat and refined
wheat bran rich food on BP mean baseline BP of 132/80 mm Hg; wheat diets over 4 weeks
(UK) [72] diet: 19 g cereal fiber/day (whole
meal bread, whole-grain breakfast
cereals with bran) vs. six cereal fiber
g/day (refined white bread and
breakfast cereals); 4 weeks
Diastolic BP (mm Hg) Systolic BP (mm Hg) Adiponectin (ug/ml)

0
50 g Whole-grains/1,000 kcal 50 g Refined grains/1,000 kcal
Change from Baseline over 8 Weeks
–1
–2
–3
–4
–5
–6
Fig. 14.4 Effect of whole vs. refined grains on blood pressure (BP) and plasma adiponectin in overweight and obese
men and women (diastolic BP p < .01; systolic BP p > .05; adiponectin p = .05) [63]
matched, except for the inclusion of either whole compared with a 1.4 μg/mL decline after refined-
grains or refined grains (50 g/1000 kcal); each grain intake. Decreases in diastolic BP corre-
diet for 8 weeks; 10-week washout) found that lated with the circulating adiponectin
the whole-grain diet significantly reduced dia- concentration. In general, whole-grain rich diets,
stolic BP by 5.8 mm Hg compared to a reduction especially those containing β-glucan, are more
of 1.6 mm Hg after the refined grain control diet effective at reducing BP in adults ≥45 years,
(Fig. 14.4) [63]. Whole-grain intake preserved hypertensive, or obese than in adults <45 years,
circulating total adiponectin concentrations normotensive or lean [24, 62, 63, 65–73]. A RCT
(88 adults taking β-blockers; mean age 48 years; vegetable campaigns in Europe, the US and
mean BP below 140/88 mm Hg; 12 weeks) Australia [73]. The USDA Dietary Guidelines
showed that the consumption of oatmeal or oat for Americans MyPlate educational concept,
squares (3 g β-glucan) vs. wheat crisps or hot devotes one-half the plate to the fruits and veg-
wheat cereal (0 g β-glucan) resulted in subjects etables, partially as a displacement of other
either stopping or reducing β blocker by half in foods of higher energy density from the diet,
73% of oat β-glucan participants vs. 42% in the and also as a good habit to establish for healthy
wheat group [24]. Those in the oat group who aging or healthy eating at any age [74]. However,
did not experience a β blocker reduction had a globally, fruit and vegetable consumption is at
significantly lower systolic BP by 6 mm Hg vs. only a small fraction of the recommended levels
the wheat group. One RCT (36 overweight and [75]. In the US, >85% of the population fall
obese men; mean age 59 years; elevated BP; short of meeting the daily fruit and vegetable
5.5 g β-glucan; 12 weeks) showed whole oats to intake recommendation [41].
have an insignificant lowering effect on BP,
which may have been confounded by a signifi- 14.2.4.2 Prospective Cohort Studies
cant 0.8 kg increase in body weight during the The inverse association of fruit and vegetables
study [68]. A USDA trial (25 mildly hypercho- intake and hypertension risk appears to be most
lesterolemic, overweight adults [seven men, nine strongly associated with fruit intake and weight
premenopausal and nine postmenopausal control/loss.
women; mean age 47 years; BP 117/71 mm Hg;
5 weeks) found that the consumption of an AHA Meta-analyses
step 1 diet plus barley (β-glucans), whole wheat Two meta-analyses show that fruit and vegetable
or brown rice (insoluble fibers), or a 50/50 mix- consumption is inversely associated with hyper-
ture resulted in similar significantly lowered sys- tension incidence [76, 77]. One meta-analysis
tolic and diastolic BPs compared to the white (25 studies; 334,468 participants; 41,713 hyper-
rice control [66]. tension cases) demonstrated that the mean risk
of hypertension was significantly reduced for
total fruit and vegetable intake by 19%, for fruit
14.2.4 Fruits and Vegetables intake by 27%, and vegetables by 3% (highest
vs. the lowest intake) [76]. For fruit intake, sub-
14.2.4.1 Background group analysis found a significant inverse rela-
Adequate intake of fruits and vegetables is an tionship between the level of fruit intake and the
important component of most global dietary risk of hypertension for studies carried out in
guidance recommendations for health and BP Asia by 30% but not in the EU or the US which
control because of their concentrations of: anti- showed a 6% lower risk. The other meta-analysis
oxidant vitamins and phytochemicals, espe- (seven cohort studies; 185,676 participants)
cially vitamins C and A, and carotenoids; showed that fruits or vegetables separately or
minerals (especially electrolytes potassium and total fruits and vegetables were inversely associ-
magnesium, and low sodium); and fiber [31, ated with hypertension risk (highest vs. lowest
34]. Fruit and vegetables include a diverse group intake) with fruits reducing risk by 13%, vegeta-
of plant foods that vary greatly in content of bles by 12%, and total fruits and vegetables by
energy, fiber, glycemic index value and nutri- 10% [77]. A dose response analysis estimated
ents. The World Health Organization (WHO) that each daily serving lowered hypertension
report recommends a minimum daily intake of risk for fruit by 1.9% and for total fruit and veg-
400 g of fruits and vegetables, based on evi- etables by 1.2%. Clearly, these meta-analyses
dence that higher levels are protective against support a role for increased fruit and vegetable
cardiovascular disease (CVD) [34]. This led to intake in reducing the risk of developing
the launch of various “eat 5 or more” fruit and hypertension.
rospective Cohort and Cross-Sectional

P risk after adjustment for lifestyle and dietary fac-
Studies tors but adding BMI to the models eliminated all
Prospective cohort studies consistently indicate significant associations [79]. This study also
that specific types of fruits and vegetables differ observed that green leafy and dark-yellow vege-
in their hypertension risk reduction capacity. An tables, and apples, oranges and raisins were
analysis of three large, long-term cohorts among the most effective in lowering risk of
(187,453 participants from Nurses’ Health Study: hypertension. The Japanese Ohasama study (745
62,175 women, Nurses’ Health Study II: 88,475 participants; mean age 56 years; mean follow-up
women, and Health Professionals Follow-up 4 years; 222 incidents of hypertension) observed
Study: 36,803 men; >20 years of follow-up) that high intake of fruit such as citrus, apples,
found a lower mean risk of hypertension for total grapes and watermelon was associated with a sig-
whole fruit by 8% and total vegetable intake by nificant 60% lower risk of hypertension, after
5% (≥4 servings/day vs. ≤4 servings/week) [78]. multivariate adjustments [80]. The Spanish
The consumption levels of broccoli, carrots, tofu Seguimiento University of Navarra (SUN) study
or soybeans, raisins, grapes and apples associated (8594 participants; mean age 41 years; median
with significantly lower hypertension risk for ≥4 duration 6 years) found a significant inverse asso-
servings/week vs. <1 serving/month (Figs. 14.5 ciation between fruit intake and BP, but not veg-
and 14.6). The Women’s Health Study (29,082 etable intake and hypertension risk [81]. In the
US health professionals; mean age 54 years; population-based International Study on Macro/
12.9 years of follow-up; 13,633 cases of hyper- Micronutrients and Blood Pressure (INTERMAP)
tension), found that women who had a higher cross-sectional, cross-cultural, population-based
intake of all fruits but not all vegetables were sig- study of middle-aged individuals (4680 partici-
nificantly associated with reduced hypertension pants from Japan, China, UK and US; mean age
Raisins or grapes (p<.001) Apples or pears (p<.001) Blueberries (p=.01)

Avocados (p<.001) Cantaloupe (p=.002)
1.08
1.06
1.04
Hazard Ratio for Hypertension
1.02
0.98
0.96
0.94
0.92
0.9
< 1 per month 1-3 per month 1-3 per week > = 4 per week
Fruit Servings
Fig. 14.5 Association between specific fruits and hypertension risk in US men and women from the Nurses’ Health
Studies and Health Professionals Follow-up Study (adapted from [78])
Brussel Sprouts (p=.03) Carrots (p=.003)

Tofu or soybeans (p=.001) Cauliflower (p<.001)
1.3
Hazard Ratio for Hypertension 1.25
1.2
1.15
1.1
1.05
1
0.95
0.9
0.85
0.8
< 1 per month 1-3 per month 1-3 per week > = 4 per week
Vegetable Servings
Fig. 14.6 Association between specific vegetables and hypertension risk in US men and women from the Nurses’
Health Studies and Health Professionals Follow-up Study (adapted from [78])
49 years), raw fruit intake was not inversely asso- /month) or occasional juice consumption (1 to <6
ciated with BP, but this is expected since the times /week) but these results are expected to
median fruit intake was low, 46.5 g/1000 kcals, vary depending on the type of fruit juice con-
which was below the threshold to expect a reduc- sumed (e.g., added sugar vs natural sugar or low
tion in BP [82]. For vegetables, the INTERMAP vs high polyphenol content) [85].
study (2195 participants; age 40–59 years)
observed a significant inverse relationship for 14.2.4.3 Randomized Controlled
both total raw and cooked vegetables and BP in Trials (RCTs)
multivariate-adjusted models [83]. The consump- Fruits and vegetables and BP RCTs are summa-
tion of raw vegetables (68 g per 1000 kcal) and rized in (Table 14.3) [17, 86–96].
cooked vegetables (92 g per 1000 kcal) were
associated with significant average systolic BP ruit and Vegetables
F
reductions by 0.9–1.3 mm Hg in the fully adjusted These studies generally show that increased fruit
model including BMI. In this analysis, the most and vegetable intake result in significant reduction
effective vegetables associated with BP lowering of BP in older, overweight, prehypertensive or
included tomatoes, carrots, peas, celery, and scal- hypertensive subjects. A Cochrane systematic
lions. Gazpacho consumption was associated review (two RCTs; 891 general healthy popula-
with a 15% lower prevalence of hypertension for tion; 6–12 months) showed that the consumption
each 250 g/week increase [84]. A 2015 crosssec- of ≥5 fruit and vegetable portions significantly
tional study (146 participants; mean age 59 years; reduced systolic BP by 3.0 mm Hg compared to
12 month dietary recall questionnaire for total the control lower intake [86]. A large RCT (459
fruit juice intake) indicated that daily fruit juice healthy adults; mean age 44 years; 50% women;
consumption (1 to >3 times/day) was associated mean BMI 28 kg/m2; mean BP 132/85 mm Hg;
with significantly higher increased central sys- 8-weeks duration) showed that 8–10 daily serv-
tolic BP, central pulse pressure, and central aug- ings of fruits and vegetables significantly reduced
mentation pressure compared to rarely (<3 times systolic BP by 2.8 mm Hg and borderline
Table 14.3 Summaries of fruits and vegetables RCTs on blood pressure (BP)
Cochrane Systematic Review and Meta-analysis
Hartley et al. (2013). 2 RCTs; 891 generally healthy subjects; Advice to eat fruits and vegetables
Determine the effectiveness 6 months to 1 year significantly reduced systolic BP by
of increased fruits and mean 3.0 mm Hg vs. low fruits and
vegetables intake for the vegetables control but the reduction
primary prevention of CVD in diastolic BP by 0.90 mm Hg was
including BP [86] not statistically significant
Liu et al. (2013). 8 RCTs; 197 participants; stage 1 Fruit juice significantly reduced
Quantitatively evaluate the hypertension; juices: pomegranate, concord mean diastolic BP values by 2.1 mm
effect of fruit juice on BP in grape juice, cranberry, orange and blueberry Hg but insignificantly lowered
adults (meta-analysis) [87] juice, muscadine grape juice systolic BP
Representative RCTs
Jovanovski et al. (2015). Crossover RCT: 27 healthy adults; mean age Spinach soup (rich in nitrates)
Investigate the effect of 25 years; mean BMI 23; 59% women; mean significantly reduced mean central
spinach intake on BP BP 116/69; studied high-nitrate (spinach systolic BP by 3.4 mm Hg and
(Canada) [88] (250 g); 845 mg nitrate/day) or low-nitrate diastolic BP by 2.6 mm Hg after
soup (asparagus (260 g); 0.6 mg nitrate/day); 7-days of supplementation compared
BP at 3 h post-prandial; 7-day duration; 7-day to the low-nitrate control. This
washout highlights the potential importance
of consuming specific types of
components in vegetables associated
with lower BP control
Tjelle et al. (2015). Assess Double-blinded, Parallel RCT: 134 healthy Compared with the placebo group,
the effects of polyphenol- individuals, mean age 62 years, with high the polyphenol rich juices
rich juices on BP (Norway) normal range BP 130/85–139/89 mm Hg; significantly reduced systolic BP in
[89] 500 mL/day of either a commercially available the hypertensive subjects by 2 mm
polyphenol-rich juice based on red grapes, Hg at 6 weeks and 2.8 mm Hg at
cherries, chokeberries and bilberries, a similar 12 weeks. In normotensive subjects,
juice enriched with polyphenol-rich extracts a significant difference between
from blackcurrants or a placebo juice; placebo and polyphenolic rich juices
12 weeks was not observed
Novotny et al. (2015). Double-blinded, Parallel RCT: 56 healthy Low-calorie cranberry juice
Determine the potential of subjects; mean age 50 years; BMI 28; about significantly reduced diastolic BP by
low-calorie cranberry juice 60% female; low-calorie cranberry juice vs. 2.4 mm Hg vs. placebo
on BP (US) [90] flavor/color/energy-matched placebo beverage;
Twice daily volunteers consumed 240 mL
beverage, containing 173 or 62 mg of phenolic
compounds and 6.5 or 7.5 g of total sugar per
240-mL serving; 8 weeks
Vinson et al. (2012). Evaluate Crossover RCT: 18 healthy adults; average Purple potatoes decreased diastolic
the clinical effect of purple age 54 years; 61% women; average BMI 29.4; BP by 4 mm Hg (4.3%) and systolic
potatoes rich in antioxidants 140 g of microwaved purple-pigmented BP by 5 mm Hg (3.5%) vs. control.
(e.g., phenolics, anthocyanins, potatoes twice daily vs. control biscuits; New cultivars of potatoes rich in
carotenoids) on BP in 4 weeks antioxidants may help to control BP
hypertensive subjects (US) [91]
Berry et al. (2010). Crossover RCT: 57 healthy adults; mean age No significant differences in resting
Investigate the effects of 45 years; mean BP 137/88 mm Hg untreated; or ambulatory 24 h BP and vascular
increasing potassium (K) additional 20 or 40 mmol K+/day from fruits function between any of the fruit and
intake above usual levels by and vegetables or 40 mmol potassium citrate vegetable diets. The present study
increased intakes of fruits capsules/day vs. low fruits and vegetables provides no evidence to support
and vegetables or control; 6 weeks; 5-week washout dietary advice to increase K intake
supplements on BP in above usual intakes in subjects with
subjects with elevated BP early stages of hypertension
(UK) [92]
(continued)
McCall et al. (2009). Parallel RCT: 108 adults; mean age 54 years; Among hypertensive participants,
Evaluate a potential mean BMI 29; mean BP 143/83 mm Hg; 1, 3, there was a significant dose-response
dose-dependent effect of or 6 daily fruit and vegetable portions added to relationship between fruit and
fruits and vegetables on BP the run-in diet. Compliance was monitored vegetable intake and vasodilation,
in hypertensive subjects with 4-day food records and by measuring which was correlated with serum
(UK) [93] concentrations of circulating carotenoids; concentrations of lutein and
4-week run-in on low fruit and vegetable diets; β- cryptoxanthin. Also, blood flow
8 weeks response increased by 6% after
consuming just one extra daily
portion of fruits and vegetables in
hypertensive subjects suggesting that
small increases in fruit and vegetable
intake has potential benefits
John et al. (2002). Study the Parallel RCT: 690 healthy adults; mean age Increased fruit and vegetable
effect of an increase in fruit 46 years; current smokers 55 + %; mean BMI portions significantly reduced
and vegetable intake on 26; mean BP of 130/79 mm Hg; diets: systolic BP by 4.0 mm Hg and
plasma antioxidant levels increased daily fruit and vegetable portions by diastolic BP by 1.5 mm Hg vs.
and BP (UK) [94] 1.4 servings vs. control diet increased by 0.1 control. At the population level, a
servings; 6 months reduction in diastolic BP by
2 mm Hg may lower the incidence
of hypertension by 17%. Plasma
lutein, α-carotene, β-carotene,
β-cryptoxanthin, and ascorbic acid
were significantly increased vs.
controls
Broekmans et al. (2001) Parallel RCT: 48 healthy adults; mean age Both fruit and vegetable diets
Determine the effect of fruits 50 years; 50% women; mean BMI 26; mean significantly reduced BP for systolic
and vegetables and juice BP of 126/80 mm Hg; two high fiber diets BPs by 5.8–7.7 mm Hg and diastolic
intake on BP (The with 500 g fruits and vegetables/day and BP 3.9–4.4 mm Hg
Netherlands) [95] 200 mL fruit juice/day vs. 100 g fruits and
vegetables/day and no juice; 4 weeks
Smith-Warner et al. (2000). Parallel RCT: 201 healthy adults; mean age There were no changes in BP
Evaluate the effect of 60 years; 71% male; BP 128/76 mm Hg; between groups, which is not
increased fruit and vegetable baseline diet about seven servings fruits and unexpected since these normotensive
intake above recommended vegetables; intervention increased fruits and subjects were all consuming
servings on normotensive vegetables by about five servings and control recommended levels of fruits and
individuals (US) [96] decreased fruits and vegetables by about 0.5 vegetables
serving; 1 year
Appel et al. (1999). Parallel RCT: 459 healthy adults; mean age The increased fruit and vegetable
Investigate the effects of 44 years; 50% women; 60% black; mean BP diet significantly reduced systolic
fruits and vegetables rich 132/85 mm Hg; diet: fruit and vegetable rich BP by 2.8 mm Hg and borderline
diets, the DASH diet and the diet (8–10 portions; 31 g fiber/day), DASH significantly reduced diastolic BP by
American diet on BP [17] diet (high fruit and vegetable rich diet (8–10 1.1 mm Hg vs. control diet
portions; 31 g fiber/day) vs. control American (Fig. 14.7). The DASH diet rich in
diet (3.6 portions of fruits and vegetables; 9 g fruits and vegetables, and low-fat
fiber/day); 3-week American diet run-in; dairy products lowered systolic BP
8 weeks by 5.5 mm Hg and diastolic BP by
3.0 mm Hg vs. control diet
significantly reduced diastolic BP by 1.1 mm Hg measurements did not see a significant change in
more than the control diet with 3.6 portions of fruit BP in UK diets supplemented with potassium rich
and vegetables/day in all subjects (non-hyperten- fruit and vegetables (20 and 40 mmol K+/day) or
sive and hypertensive combined) (Fig. 14.7) [17]. potassium citrate (40 mmol K+/day) added to the
However, when only hypertensive subjects were standard UK diets compared to the unsupple-
assessed, the BP reduction increased for systolic mented standard UK diet [92]. A parallel RCT
BP to 7.2 mm Hg (p < .001) and diastolic BP to (48 adults; mean age 50 years; mean BMI 26 kg/
2.8 mm Hg (p = .01) compared to the control diet. m2; mean baseline BP of 126/80 mm Hg; 4-weeks
A dose response study indicated that three and six duration) found that subjects adding 400 g fruits
servings of fruit and vegetables lowered systolic and vegetables and 200 ml fruit juice/day to a low
BP by 2.2 and 4.5 mm Hg, respectively, and each fruit and vegetable diet (100 g/day) showed no
fruit and vegetable serving significantly increased significant
differences in BP lowering effects,
forearm blood flow by 6% [93]. A longer-term which was most likely due to the short study dura-
RCT (690 healthy adults; mean age 46; current tion and normal BP levels of subjects [95]. A
smokers 55%; mean BMI 26 kg/m2; mean BP long-term RCT (201 healthy participants; mean
130/79 mm Hg; 6 months) found the intake of ≥5 age 60 years; normotensive; 1 year) observed no
fruit and vegetable portions/day significantly difference in BP with seven compared to five
reduced systolic BP by 4.0 mm Hg and diastolic daily servings of fruit and vegetables [96].
BP by 1.5 mm Hg more than the low fruit and veg-
etable controls [94]. RCTs have found that spin- Fruit Juices
ach, rich in nitrates, reduces BP by augmenting Fruit juice effects on BP are heterogeneous
nitric oxide status [88] and purple potatoes, rich in depending on the juice composition such as the
phenolic antioxidants, reduce BP by promoting type of fruit, concentration and type of polypheno-
vascular health [91]. lics, as well as the BP status of the subject (normo-
Several trials have suggested no effects from tensive vs. hypertensive) [85, 89, 90]. A
increased intake of fruit and vegetables on BP. A meta-analysis of fruit juices (19 RCTs; 618 sub-
crossover RCT (57 adults; mean age 45 years; jects; polyphenols contained in fruit juice ranged
BMI 28 kg/m2; mean BP 137/88 mm Hg; from 65 to 2660 mg/day (median: 927 mg/day);
86 weeks) using both resting and ambulatory BP juices: pomegranate, concord grape, cranberry,
0
Hypertensive Non-hypertensive All men All women
–1
–2
Change in mm Hg
–3
–4
High vs Low Fruit and Vegetable Intake
–5
–6
–7
–8
Fig. 14.7 Difference in blood pressure (BP) between a high fruit and vegetable diet and a lower fruit and vegetable
control diet after 8 weeks in men and women with mean age 44 years and BMI 28 kg/m2 (p < .001 for hypertensive
subjects) (adapted from [17])
orange, blueberry juice, and muscadine grape which contains 5–10 g of fiber and 7–8 g of pro-
juice; 2 weeks to 3 months with median: 6 weeks) tein. Most legumes contain <5% of energy as fat,
found a significantly mean lower diastolic BP by with the exception of chickpeas and soybeans
2.1 mm Hg but insignificant reductions in total which have 15 and 47% energy from fat, respec-
cholesterol, HDL and LDL-cholesterol, and sys- tively. Legumes, compared to cereal grains, are
tolic BP [87]. A RCT on high polyphenol juices rich in high-quality protein [99]. The protein
(134 healthy individuals, mean age 62 years, with content of legumes ranges from 17 to 20% (dry
high normal range BP 130/85–139/89 mm Hg; weight) in peas and beans which is similar to
12 weeks) found that 500 mL/day significantly meats 18–25%, and 38–40% in soybeans. This
reduced systolic BP in the hypertensive subjects contrasts with the protein content of cereals,
by 2 mm Hg at 6 weeks and 2.8 mm Hg at which is about 7–13%. A meta-analysis of 29
12 weeks compared to placebo but there was no observational studies and RCTs of dietary pro-
difference in BP in normotensive subjects [89]. In tein effects on BP and hypertension concluded
normotensive individuals (56 healthy men and that the substitution of either animal or plant
women; mean age 50 years; 8 weeks) 240 mL of based protein for carbohydrate reduced the
low-calorie cranberry juice twice daily signifi- pooled systolic BP by 2.1 mm Hg for a weighted
cantly lowered diastolic BP by 2.4 mm Hg vs. pla- mean protein intake of 41 g/day [100]. Legumes
cebo [90]. A 2017 review of 100% fruit and contain nutritionally important amounts of the B
vegetable juices concluded that although their vitamins and minerals, such as iron, calcium and
composition is different from that of whole fruits potassium. They also contain bioactive phyto-
and vegetables, they contain polyphenols, miner- chemicals such as phenolics, saponins and iso-
als such as potassium, vitamins such as vitamin C, flavones (especially in soy foods). Legumes are
and/or nitrates from fruits and vegetables that can often consumed as a lower energy dense, lower
lower BP [97]. The main mechanisms of action saturated fat, and high fiber meat or milk replacer.
included antioxidant effects, improvement of Legumes consumption has been in decline with
blood flow, inhibition of platelet aggregation, and the global shift to Western-style diets [101]. For
anti-inflammatory effects. This review identified example, between the 1960s and 1990s, legume
three 100% fruit and vegetable juices with BP intake decreased by 40% in India and by 24% in
lowering effects: (1) high flavonoid (naringin and Mexico. Legumes are infrequently consumed by
naritutin) sweetie fruit (a hybrid between grape- North Americans and northern Europeans, with
fruit and pummelo). (2) 100% (phenolic antioxi- <8% of Americans consuming them on any
dant rich) pomegranate juice (Punica granatum L.) given day.
and (3) nitrate rich beetroot juice. A 2017 meta-
analysis of 100% pomegranate juice (8 RCTs; 547 14.2.5.2 Non-Soy Legumes
hypertensive, CVD or diabetes subjects; 50 to
500ml/day; 2 weeks to 18 months) showed signifi- rospective Cohort Studies
P
cant reductions in systolic BP by 5 mm Hg and Although there are no specific prospective stud-
diastolic BP by 2 mm Hg after pomegranate juice ies on the effects of non-soy legumes on hyper-
consumption [98]. tension risk, several studies have reported that
non-soy legumes reduced risk of CVD compared
to control diets. Further, a meta-analysis of five
14.2.5 Legumes prospective studies reported that 4 weekly serv-
ings of 100 g non-soy legumes was associated
14.2.5.1 Background with a 14% lower ischemic heart disease risk,
Legumes, including pulses (e.g., pinto beans, which is associated with hypertension [102].
split peas, lentils, chickpeas) and soybeans, are
rich in fiber and protein with relatively low gly- andomized Controlled Trials (RCTs)
R
cemic response properties [35, 99]. A serving of A systematic review and meta-analysis 8 isocalo-
legumes is 1⁄2 cup or 90–100 g cooked legumes, ric dietary pulse (beans, peas, chickpeas, and
lentils) RCTs; 554 mid dle aged subjects; median andomized Controlled Trials (RCTs)
R
increase of 1 2/3 servings (160 g/day; range Intervention trials consistently show that the
81–275 g/day; median 10 weeks) found that consumption of soy products lowers BP. A meta-
dietary pulses significantly mean lowered sys- analysis (27 RCTs; soy protein varied from 18 to
tolic BP by 2.25 mm Hg and arterial BP by 66 g/day with median of 30 g/day and 100 mg
0.75 mm Hg in total subjects with or without isoflavones/day vs. control groups received
hypertension [103]. To achieve BP reductions casein or milk; 4–52 weeks with a median of
similar to those observed in this systematic 8 weeks) concluded that the consumption of soy
review and meta-analysis, an increase in con- protein products significantly lowered the pooled
sumption of at least two servings (1 cup)/day mean for systolic BP by 2.2 mm Hg and diastolic
above current average intakes (0.1–0.3 servings/ BP by 1.4 mm Hg compared to control diets
day) would need to be recommended. Dietary [108]. Furthermore, the mean BP decrease was
pulses may lower BP by increasing dietary levels markedly greater in the hypertensive group by
of fiber, plant protein and potassium, and aid in 8.6 mm Hg (p = .010) for systolic BP and 5.2 mm
weight control, all of which confer BP-lowering Hg (p = .014) for diastolic BP. Three double-
effects [104, 105]. blind RCTs suggest that partially replacing car-
bohydrate with soy or milk protein might help to
14.2.5.3 Soy Products lower elevated BP [109–111]: (1) In hyperten-
sive adults (40 adults; mean age 48 years; 60%
Observational Studies men; mean BP 153/100 mm Hg; 3 months) the
A limited number of observational studies sug- consumption of soy milk significantly decreased
gest that soy food consumption is inversely BP compared to the cow’s milk group (Fig. 14.8)
associated with BP. The large Shanghai [109]. (2) In a crossover trial (352 adults; mean
Women’s Health Study observed that the intake age 48 years; 59% men; mean BMI 29; 33%
of foods containing 25 g soy protein/day vs. blacks; mean BP 127/82 mm Hg; 8 weeks) 40 g/
<2.5 g/day was inversely associated with both day of soy protein and milk protein significantly
systolic and diastolic BP, especially among reduced systolic BP by 2.0 and 2.3 mm Hg,
women >60 years [106]. In a Japanese cross- respectively, compared to the complex carbohy-
sectional study of about 600 men, and premeno- drate control [110]. (3) In a parallel trial (302
pausal and postmenopausal women, there was adults; mean age 51 years; mean BMI 27; 53%
an inverse association with soy product intake women; mean BP 135/85 mm Hg; 3 months)
and BP in men [107]. 40 g of isolated soy protein significantly lowered
0
–2 Soy milk Cow's milk
–4
–6
Change in mm Hg
–8
–10
–12
–14
–16
–18
–20
Fig. 14.8 Effect of soy milk or cow’s milk-based diet (500 mL twice daily) on blood pressure (BP) in mid-age adults
with essential hypertension after 3 months (p < .0001) (adapted from [109])
systolic BP by 4.3 mm Hg and diastolic BP by 14.2.6.2 Prospective Cohort Studies

2.8 mm Hg compared to a complex carbohydrate A number of prospective studies have reported
control; hypertensive subjects had even greater that nuts are positively associated with CVD
significant lowering for systolic BP by 7.9 mm Hg health and reduced risk of ischemic heart disease,
and for diastolic BP by 5.3 mm Hg [111]. compared to nut-free controls [102, 119, 120]. In
Iranian adults, more frequent nut consumption
(≥4 servings/week vs. no nut intake) was signifi-
14.2.6 Nuts and Seeds cantly associated with a 34% lower incidence of
hypertension [121]. In US male physicians, the
14.2.6.1 Background intake of ≥7 servings of nuts/week vs. no nut
Nuts and seeds are rich sources of potentially consumption was associated with a significantly
hypertensive protective macronutrients, lower risk of hypertension [122]. In the
micronutrients and phytochemicals [32, 112,
Atherosclerosis Risk of Communities study
113]. They are rich in healthy dietary fat con- (9913 African-American and Caucasian adults
tent, with a total fat content ranging from 46% in aged 45–64 years and free of hypertension at
cashews and pistachios to 76% in macadamia baseline) diets rich in nuts and dairy and lower in
nuts. The fatty acid composition of nuts is of a meat were associated with a decreased risk of
healthy type since the saturated fat levels are developing hypertension over nine years [123].
low (range 4–16%) and with 84–96% of the total However, in a Mediterranean cohort, there was
fat content made up of unsaturated fat, mostly no association between nut consumption (2+
monounsaturated fatty acids (MUFAs), and servings/week vs. never/rare nut intake) and inci-
polyunsaturated fatty acids (PUFAs, predomi- dence of hypertension after multivariate adjust-
nantly linoleic acid). Walnuts and flaxseed are ments [124].
rich sources of alpha-linolenic acid, the plant
n-3 fatty acid. Nuts and seeds are good sources 14.2.6.3 Randomized Controlled
of protein and are often rich in fiber, ranging Trials (RCTs)
from 1 to 7 g/28 g serving. Despite their rela-
tively high Atwater metabolizable energy value Nuts
of 5–6 kcal/g, nuts and seeds often have a Overall nut intake generally tends to have a
10–25% lower net metabolizable energy than modest direct clinically meaningful effect on
the label value [114–116]. Among nut types, controlling BP but the association with increased
there are significant amounts of essential micro- nut intake to reduce the risk of weight gain may
nutrients including the B-vitamin folate, antiox- have longer-term implication on lower hyper-
idant vitamins (e.g., tocopherols), polyphenols tensive risk [23, 125]. A 2015 systematic review
and carotenoids such as the lutein in pistachios and dose response meta-analysis on tree nuts
[112]. Also, nuts are an important source of key (61 RCTs; walnuts, almonds, pistachios, maca-
shortfall minerals such as magnesium, potas- damia, pecans, cashews, hazel nuts and Brazil
sium and calcium and contain very low levels of nuts) found an insignificant lower pooled mean
sodium, which can contribute to improved BP BP per serving of tree nuts for systolic BP by
control. Overall, nut consumers have a signifi- 0.3 mm Hg and for diastolic BP by 0.39 mm Hg
cantly higher nutrient quality score than non-nut [125]. Another 2015 systematic review and
consumers, which can lead to overall better meta-analysis (21 RCTs; 1652 diabetic and non-
health and wellness [117]. However, in the US diabetic adults; walnuts, almonds, pistachios,
slightly over 60% of men and women do not cashews, hazelnuts, macadamia nuts, pecans,
consume any nuts on a given day and only14.4% peanuts, and soy nuts; range 30–108 g/day)
of men and 11.8% of women consume ≥1.5 found that nut consumption resulted in an insig-
ounces of nuts/daily, which is similar to global nificant mean lower systolic BP by 0.9 mm Hg,
nut intake [118]. but there was significant heterogeneity among
14.3 Tea and Coffee 407
studies [23]. In a subsequent revised analysis in diastolic BP by 5.8 mm Hg. To reduce the hetero-
only non-diabetic subjects, higher nut intake had geneity, the secondary meta-analysis was limited
modest significantly lowered systolic BP by to four high methodology quality trials which
1.3 mm Hg. A subgroup analysis of different nut showed a significant reduction in systolic BP by
types found that pistachio nuts were the most 3.2 mm Hg and a non-significant reduction in
effective, significantly reducing systolic BP by diastolic BP by 2.1 mm Hg [127]. A 2016 sys-
1.8 mm Hg. Pistachios and mixed nuts signifi- tematic review demonstrated that of the seven
cantly reduced diastolic BP by 0.8 and 1.2 mm Hg, RCTs evaluated five RCTs in hypertensive indi-
respectively. viduals had a significant reduction in systolic
and/or diastolic BP [128].
Seeds
Flax and sesame seeds are more effective than
nuts in lowering BP [21, 126–128]. 14.3 Tea and Coffee
Flaxseeds Coffee and tea are among the most popular bev-
RCTs suggest that flaxseeds are effective in low- erages in the world so the effects of how drinking
ering BP because they are a rich dietary source of them affects incident hypertension is an impor-
ω-3 fatty acids and α-linolenic acid, lignans, and tant public health question.
fiber, with combined potential for BP lowering
effects [21, 126]. A systematic review and meta-
analysis (12 RCTs; 1004 health participants) 14.3.1 Tea
showed that flaxseed consumption modestly but
significantly reduced mean systolic BP by Tea is the second most commonly consumed bev-
1.8 mm Hg and mean diastolic BP by 1.6 mm Hg erage globally, after water, and has been tradition-
[21]. A double-blinded RCT (110 adults; mean ally associated with CVD protective properties
age 67 years; mean BMI 28 kg/m2; mean BP such as antihypertensive effects in individuals
143/78 mm Hg; 75% hypertensive; 6 months) with elevated BP (hypertensive or prehyperten-
reported that the daily consumption of a variety of sive) [129]. Tea is rich in a class of polyphenolic
foods that contained 30 g of milled flaxseed sig- compounds known as flavonoids. Tea comes from
nificantly lowered systolic BP by 10 mm Hg and the plant Camellia sinensis and is a beverage pre-
diastolic BP by 7 mm Hg compared with a pla- pared by pouring hot or boiling water over its
cebo (0 g flaxseed) [127]. Hypertensive subjects cured leaves or leaf buds. Green and black teas
(systolic BP > 140 mm Hg at baseline) had a more have similar total flavonoid content but have dif-
significant lowering of systolic BP by 15 mm Hg ferent chemical structures. In green tea, flavo-
from flaxseed intake compared to the overall study noids are normally found in the form of catechins,
population mean. Body weight did not differ while in black tea, flavonoids are typically found
between the two groups, whereas plasma levels of as theaflavins. A 2015 meta-analysis of black and
ω-3 fatty acid and α-linolenic acid and enterolig- green tea combined (10 RCTs; 834 hypertensive
nans increased 2- to 50-fold in the flaxseed-fed or pre-hypertensive subjects; median duration 3
group compared to the control group [127]. months) found that the median dosage of 456 mg
of flavonoids/d (approximately 2 cups of black or
Sesame Seeds green tea per day) significantly reduced systolic
Sesame seed consumption can reduce BP due to BP by 2.4 mm Hg and diastolic BP by 1.8 mm Hg
its high polyunsaturated fatty acids, fiber, phytos- [130]. A 2014 meta-analysis of green tea (13
terol and lignans content [127, 128]. A 2017 sys- RCTs; 1,367 hypertensive or pre-hypertensive
tematic review and meta-analysis (8 RCTs; 843 subjects; 3 months) showed that median intake of
participants) found that sesame consumption sig- 583 mg of flavonoids/d (approximately 2 ½ cups)
nificantly reduced systolic BP by 7.8 mm Hg and significantly reduced both systolic and diastolic
BP by 2 mm Hg [131]. A 2014 meta-analysis of hypertension risk in a baseline healthy popula-

black tea (11 RCTs; 378 normotensive, hyper- tion. A 2011 meta-analysis (6 cohort studies;
tensive and pre-hypertensive subjects; > 1 week) 172,567 participants; mean follow-up 6.4 to 33
found that daily consumption of 4 to 5 cups of years) found that habitual coffee consumption of
black tea significantly reduced systolic BP by 2 >3 cups/day was not associated with a significant
mm Hg and diastolic BP by 1 mm Hg [132]. increased hypertension risk compared with con-
Another 2014 meta-analysis (25 RCTs; 1,476 sumption of <1 cup/day, but a slightly elevated
normotensive, hypertensive and pre-hyperten- risk appeared to be associated with light-to-mod-
sive subjects) found that >12 weeks consump- erate consumption (1–3 cups/day) [138]. The
tion of tea (green and black tea) resulted in a Women’s Health Initiative Observational Study
significant reduction of systolic BP by 2.6 mm (29,985 normotensive women; mean age 63
Hg and diastolic BP by 2.2 mm Hg which was years; 3-year follow-up) found that the consump-
not influenced by ethnicity, caffeine intake, tea tion of caffeinated coffee, decaffeinated coffee,
polyphenol doses, health status of participants and caffeine were not associated with the risk of
[133]. At the population level, reductions in BP postmenopausal incident hypertension (p-trend >
2 to 3mm Hg are associated with lower incidence 0.05 for all) [139]. In hypertensive individuals, a
of coronary artery disease by 8% and stroke by systematic review and meta-analysis (5 RCTs;
15% [134]. 200–300 mg caffeine/day) showed a mean
increase of 8.1 mm Hg in systolic BP and of 5.7
mm Hg in diastolic BP observed in the first hour
14.3.2 Coffee after caffeine intake and lasting approximately 3
hrs [140]. However, after 2 weeks of coffee con-
Coffee is a brewed drink prepared from roasted sumption, no increase in BP was observed after
coffee beans, which are the seeds of berries from coffee was compared with a caffeine-free diet or
the Coffea plant. Coffee beverages contain a mix- decaffeinated coffee. There is a need to investi-
ture of several pharmacologically-bioactive com- gate directly the influence of coffee or caffeine on
pounds, including caffeine, chlorogenic acid, the degree of BP control in hypertensive individ-
phenolic acids, and the diterpene alcohols, cafes- uals and its possible interaction with antihyper-
tol and kahweol, can also have long term effects tensive medications.
on health [135]. A link between coffee drinking
and BP was first reported nearly 75 yrs ago [136], Conclusions
but whether coffee intake is associated with BP By 2025, because of population growth and an
or hypertension risk has mixed results. A moder- aging population, it is projected that about 1.5
ate (400–600 mg/d) caffeine intake is not associ- billion individuals globally will have hyper-
ated with increased risks of total cardiovascular tension. Lifestyle factors such as meeting the
disease, or BP among regular caffeine consumers dietary recommendation for whole plant foods
and hypertension in baseline healthy populations is the most important action one can take to
[137]. Populations at risk for hypertension, or lower risk of hypertension. Whole (and mini-
already with hypertension, may be more sensitive mally processed) plant foods usually contain
to some effects of caffeine. Pre- or hypertensive some mixture of BP lowering bioactive nutri-
populations may experience acute increases in ents and phytochemicals such as dietary fiber,
BP following caffeine intake ranging from 100 to potassium, magnesium, polyphenols, unsatu-
<400 mg/d compared to normotensive people. rated fat, and plant protein and are lower in
Also, pre- or hypertensive populations may have sodium and sugar compared to highly pro-
a small increased relative risk of sustained hyper- cessed plant foods. Whole (and minimally
tension related to high caffeine beverage con- processed) plant foods usually contain a num-
sumption. Overall, studies generally suggest no ber of BP lowering bioactive nutrients and
association between caffeine consumption and phytochemicals such as fiber, potassium,
Appendix A 409
magnesium, carotenoids, polyphenols, unsat- have been shown to lower systolic BP; and 2
urated fat, and plant protein and are lower in daily servings of dietary pulses or 40 g soy
sodium and sugar compared to highly pro- protein are effective in lowering BP. Flaxseeds
cessed plant foods. Whole plant foods are and sesame seeds tend to be more effective
more effective at reducing BP in adults who than nuts in lowering BP. Tea and coffee have
are ≥45 years, hypertensive and obese than different effects on BP. Both black and green
adults <45 years, normotensive, or lean. tea (>2 cups/day) modestly lower BP in hyper-
Prospective studies show that the consump- tensive individuals. Coffee (>3 cups/day) does
tion of healthy diets with ≥3 daily servings of not increase hypertension risk in normoten-
whole grains, especially grains rich in sive people but hypertensive individuals may
β-glucan, and ≥5 daily servings of fruits and be more sensitive to acute increases in BP
vegetables, especially when including ≥4 after coffee consumption. The potential mech-
weekly servings of broccoli, carrots, tofu or anisms by which whole plant foods may
soybeans, raisins, grapes and apples, are asso- reduce blood pressure and hypertension risk
ciated with lower hypertension risk compared are; reducing the risk of weight gain, enhanc-
to Western diets. RCTs support the effective- ing insulin sensitivity, improving vascular
ness of whole grains in lowering BP espe- endothelial function, slowing the rate of arte-
cially at 50 g/1000 kcals or those rich in rial plaque build-up, maintaining electrolyte
β-glucan; fruits and vegetables rich in poly- balance, and stimulating a healthier microbi-
phenols and nitrates and their 100% juices ota ecosystem.
ppendix A: Estimated Range of Energy, Fiber, Nutrients

A
Whole-
Components grains Fresh fruit Dried fruit Vegetables Legumes Nuts/seeds
Nutrients/ Wheat, oat, Apples, Dates, dried Potatoes, spinach, Lentils, Almonds, Brazil
phytochemicals barley, rye, pears, figs, carrots, peppers, chickpeas, nuts, cashews,
brown rice, bananas, apricots, lettuce, green split peas, hazelnuts,
whole grain grapes, cranberries, beans, cabbage, black beans, macadamias,
bread, oranges, raisins and onions, cucumber, pinto beans, pecans, walnuts,
cereal, blueberries, prunes cauliflower, and soy peanuts,
pasta, rolls strawberries, mushrooms, and beans sunflower seeds,
and crackers and avocados broccoli and flaxseed
Energy (kcals) 110–350 30–170 240–310 10–115 85–170 520–700
Protein (g) 2.5–16 0.5–2.0 0.1–3.4 0.2–5.0 5.0–17 7.8–24
Available 23–77 1.0–25 64–82 0.2–25 10–27 12–33
carbohydrate (g)
Fiber (g) 3.5–18 2.0–7.0 5.7–10 1.2–9.5 5.0–11 3.0–27
Total fat (g) 0.9–6.5 0.0–15 0.4–1.4 0.2–1.5 0.2–9.0 46–76
SFAa (g) 0.2–1.0 0.0–2.1 0.0 0.0–0.1 0.1–1.3 4.0–12
MUFAa (g) 0.2–2.0 0.0–9.8 0.0–0.2 0.1–1.0 0.1–2.0 9.0–60
PUFAa (g) 0.3–2.5 0.0–1.8 0.0–0.7 0.0.0.4 0.1–5.0 1.5–47
Folate (μg) 4.0–44 <5.0–61 2–20 8.0–160 50–210 10–230
Tocopherols (mg) 0.1–3.0 0.1–1.0 0.1–4.5 0.0–1.7 0.0–1.0 1.0–35
Potassium (mg) 40–720 60–500 40–1160 100–680 200–520 360–1050
Calcium (mg) 7.0–50 3.0–25 10–160 5.0–200 20–100 20–265
Magnesium (mg) 40–160 3.0–30 5.0–70 3.0–80 40–90 120–400
Phytosterols (mg) 30–90 1.0–83 – 1.0–54 110–120 70–215
Whole-
Polyphenols (mg) 70–100 50–800 – 24–1250 120–6500 130–1820
Carotenoids (μg) – 25–6600 1.0–2160 10–20,000 50–600 1.0–1200
Nutrient Database for Standard Reference, Release 27. http://www.ars. usda.gov. /nutrientdata. Accessed 17 February
2015
Circulation. 2013;128:553–65
Sci Nutr. 2016;56(7):1097–1107
Gebhardt SE, Thomas RG. Nutritive value of foods. 2002; U.S. Department of Agriculture, Agricultural Research
An. 1999;12:169–96
tent during maturation and from different growing areas. J Agric Food Chem. 2009; 57(21):10408–13
8. Dietary Guidelines Advisory Committee (DGAC).

References Scientific report. Advisory report to the Secretary
of Health and Human Services and the Secretary
1. WHO. In: Mendis S, Puska P, Norrving B, editors. of Agriculture. Part D. Chapter 2: Dietary patterns,
Global atlas on cardiovascular disease prevention foods and nutrients and health outcomes. 2015:1–35.
and control. Geneva: World Health Organization; 9. Frisoli TM, Schmieder RE, Grodzicki T, Messerli
2011. FH. Beyond salt: lifestyle modifications and blood
2. World Health Organization. A global brief on hyper- pressure. Eur Heart J. 2011;32:3081–7.
tension. Silent killer, global public health crisis. 10. Rosvall M, Persson M, Ostling G, et al. Risk fac-
Geneva: WHO; 2013. http://www.who.int/cardio- tors for the progression of carotid intima-media
vascular_diseases/publications/global brief. hyper- thickness over a 16-year follow-up period: the
tension/en/. Accessed April 2017 Malmo Diet and Cancer Study. Atherosclerosis.
3. Mozaffarian D, Benjamin EJ, Go AS, on behalf of the 2015;239:615–21. https://doi.org/10.1016/j.
American Heart Association Statistics Committee atherosclerosis.2015.01.030.
and Stroke Statistics Subcommittee. Heart dis- 11. Dickinson HO, Mason JM, Nicolson DJ, et al.
ease and stroke statistics-2016 update: a report Lifestyle interventions to reduce raised blood pres-
from the American Heart Association. Circulation. sure: a systematic review of randomized controlled
2016;133:e38–e360. trials. J Hypertens. 2006;24(2):215–33.
4. Atlas SA. The renin-angiotensin aldosterone system: 12. Liu X, Zhang D, Liu Y, et al. Dose-response associa-
pathophysiological role and pharmacologic inhibition between physical activity and incident hyperten-
tion. J Manag Care Pharm. 2007;13(8):S9–S20. sion. A systematic review and meta-analysis of cohort
5. Livingstone KM, McNaughton SA. Dietary pat- studies. Hypertension. 2017;69(5):813–20. https://
terns by reduced rank regression are associated with doi.org/10.1161/HypertensionAHA.116.08994.
obesity and hypertension in Australian adults. Br 13. Briasoulis A, Agarwal V, Messerli FH. Alcohol
Nutr J. 2017;117:248–59. https://doi.org/10.1017/ consumption and the risk of hypertension in men
S0007114516004505. and women: a systematic review and meta-analysis.
6. Castro I, Waclawovsky G, Marcadenti A. Nutrition J Clin Hypertens (Greenwich). 2012;14:792–8.
and physical activity on hypertension: implication 14. Steffen LM, Kroenke CH, Yu X, et al. Associations
of current evidence and guidelines. Curr Hypertens of plant food, dairy product, and meat intakes with
Rev. 2015;11(2):91–9. 15-y incidence of elevated blood pressure in young
7. Savica V, Bellinghieri G, Kopple JD. Effect black and white adults: the Coronary Artery Risk
of nutrition on blood pressure. Ann Rev Nutr. Development in Young Adults (CARDIA) Study.
2010;30:365–402. Am J Clin Nutr. 2005;82:1169–77.
References 411
15. Gay HC, Rao SG, Vaccarino V, Ali MK. Effects of 29. Soedamah-Muthu SS, Verberne LDM, Ding EL,
different dietary interventions on blood pressure sys- et al. Dairy consumption and incidence of hyperten-
tematic review and meta-analysis of randomized con- sion a dose-response meta-analysis of prospective
trolled trials. Hypertension. 2016;67:733–9. https:// cohort studies. Hypertension. 2012;60:1131–7.
doi.org/10.1161/HypertensionAHA.115.06853. 30. Ballard KD, Bruno RS. Protective role of dairy and
16. Alonso A, de la Fuente C, Martin-Arnau AM, its constituents on vascular function independent
et al. Fruit and vegetable consumption is inversely of blood pressure-lowering activities. Nutr Rev.
associated with blood pressure in a Mediterranean 2014;3(1):36–50.
population with a high vegetable-fat intake: the 31. U.S. Department of Agriculture, Agriculture
Seguimiento Universidad de Navarra (SUN) Study. Research Service, Nutrient Data Laboratory. 2014.
Br J Nutr. 2004;92:311–9. USDA National Nutrient Database for Standard
17. Appel LJ, Moore TJ, Obarzanek E, et al. A clinical Reference, Release 27. http://www.ars. usda.gov./
trial of the effects of dietary patterns on blood pres- nutrientdata. Accessed 17 Feb 2015.
sure. N Engl J Med. 1997;336:1117–24. 32. Ros E, Hu FB. Consumption of plant seeds and car-
18. Feng J, He FJ, Li J, MacGregor GA. Effect of lon- diovascular health epidemiological and clinical trial
ger term modest salt reduction on blood pressure: evidence. Circulation. 2013;128:553–65.
Cochrane systematic review and meta-analysis of 33. USDA. What we eat in America, NHANES 2011-
randomised trials. BMJ. 2013;346:f1325. https://doi. 2012, individuals 2 years and over (excluding breast-
org/10.1136/bmj.f1325. fed children). Available: www.ars.usda.gov/nea/
19. Doménech M, Roman P, Lapetra J, et al. bhnrc/fsrg.
Mediterranean diet reduces 24-hour ambulatory 34. Slavin JL, Lloyd B. Health benefits of fruits and veg-
blood pressure, blood glucose, and lipids one- etables. Adv Nutr. 2012;3:506–16.
year randomized, clinical trial. Hypertension. 35. Rebello CJ, Greenway FL, Finley JW. A review of
2014;64:69–76. the nutritional value of legumes and their effects on
20. Jayalath VH, de Souza RJ, Sievenpiper JL, et al. obesity and its related co-morbidities. Obes Rev.
Effect of dietary pulses on blood pressure: a system- 2014;15:392–407.
atic review and meta-analysis of controlled feeding 36. Gebhardt SE, Thomas RG. Nutritive value of foods.
trials. Am J Hypertens. 2014;27(1):56–64. Washington, DC: U.S. Department of Agriculture,
21. Khalesi S, Irwin C, Schubert M. Flaxseed con- Agricultural Research Service, Home and Garden
sumption may reduce blood pressure: a systematic Bulletin 72; 2002.
review and meta-analysis of controlled trials. J Nutr. 37. Holden JM, Eldridge AL, Beecher GR, et al.
2015;145(4):758–65. https://doi.org/10.3945/jn.114. Carotenoid content of U.S. foods: an update of the
205302. database. J Food Comp An. 1999;12:169–96.
22. Kochar J, Gaziano M, Djoussé L. Breakfast cereals 38. Lu Q-Y, Zhang Y, Wang Y, et al. California Hass
and risk of hypertension in the physicians’ health avocado: profiling of carotenoids, tocopherol,
study. Clin Nutr. 2012;31(1):89–92. fatty acid, and fat content during maturation and
23. Mohammadifard N, Salehi-Abargouei A, Salas- from different growing areas. J Agric Food Chem.
Salvadó J, et al. The effect of tree nut, peanut, and 2009;57(21):10408–13.
soy nut consumption on blood pressure: a systematic 39. Wu X, Beecher GR, Holden JM, et al. Lipophilic
review and meta-analysis of randomized controlled and hydrophilic antioxidant capacities of common
clinical trials. Am J Clin Nutr. 2015;101:966–82. foods in the United States. J Agric Food Chem.
24. Pins JJ, Geleva D, Keenan JM. Do whole-grain oat 2004;52:4026–37.
cereals reduce the need for antihypertensive medi- 40. Bhupathiraju SN, Tucker KL. Coronary heart dis-
cations and improve blood pressure control? J Fam ease prevention: nutrients, foods, and dietary pat-
Pract. 2002;14:353–9. terns. Clin Chim Acta. 2011;412:1493–514.
25. Ahhmed AM, Muguruma M. A review of meat 41. Dietary Guidelines Advisory Committee. Scientific
protein hydrolysates and hypertension. Meat Sci. Report. Advisory Report to the Secretary of Health
2010;86(1):110–8. and Human Services and the Secretary of Agriculture.
26. Appleby PN, Davey GK, Key TJ. Hypertension and Part D. Chapter 1: Food and nutrient intakes, and
blood pressure among meat eaters, fish eaters, veg- health: current status and trends 2015;1–78. http://
etarians and vegans in EPIC–Oxford. Public Health health.gov/ dietary guidelines/2015/guidelines/.
Nutr. 2002;5(5):645–54. Accessed 26 Jan 2016.
27. Mattei J, Noel S, Tucker KL. A meat, processed 42. Sanchez-Muniz FJ. Dietary fibre and cardiovascular
meat, and French fries dietary pattern is associ- health. Nutr Hosp. 2012;27(1):31–45.
ated with high allostatic load in Puerto Rican older 43. Yang T, Santisteban MM, Rodriguez V, et al. Gut
adults. J Am Diet Assoc. 2011;111(10):1498–506. dysbiosis is linked to hypertension. Hypertension.
28. Lajous M, Bijon A, Fagherazzi G, et al. 2015;65:1331–40.
Processed and unprocessed red meat consump- 44. Pluznick JL. A novel SCFA receptor, the microbiota,
tion and hypertension in women. Am J Clin Nutr. and blood pressure regulation. Gut Microbes.
2014;100:948–52. 2014;5(2):202–7.
45. Leermakers ETM, Darweesh SKL, Baena CP, et al. randomized controlled trials of health individuals.
The effects of lutein on cardiometabolic health J Hypertension. 2015;33(5):897–911.
across the life course: a systematic review and meta- 62. Thies F, Masson LF, Boffetta P, Kris-Etherton
analysis. Am J Clin Nutr. 2016;103:481–94. P. Oats and CVD risk markers: a systematic litera-
46. Lillioja S, Neal AL, Tapsell L, et al. Whole grains, ture review. Br J Nutr. 2014;112:S19–30.
type 2 diabetes, coronary heart disease, and hyper- 63. Kirwan JP, Malin SK, Scelsi AR, et al. Whole-grain
tension: links to the aleurone preferred over indigest- diet reduces cardiovascular risk factors in over-
ible fiber. Biofactors. 2013;39(3):242–58. weight and obese adults: a randomized controlled
47. Karl JP, Saltzman E. The role of whole trial. J Nutr. 2016;146:2244–51.
grains in body weight regulation. Adv Nutr. 64. Tighe P, Duthie G, Vaughan N, et al. Effect of
2012;3:697–707. increased consumption of whole-grain foods on
48. Seal CJ. Whole grains and CVD risk. Proc Nutr Soc. blood pressure and other cardiovascular risk markers
2006;65:24–34. in healthy middle-aged persons: a randomized con-
49. Medina-Remón A, Tresserra-Rimbau A, Pons A, trolled trial. Am J Clin Nutr. 2010;92:733–40.
et al. Effects of total dietary polyphenols on plasma 65. Maki KC, Galant R, Samuel P, et al. Effects of con-
nitric oxide and blood pressure in a high cardio- suming foods containing oat beta-glucan on blood
vascular risk cohort. The PREDIMED randomized pressure, carbohydrate metabolism and biomark-
trial. Nutr Metab Cardiovasc Dis. 2015;25(1):60–7. ers of oxidative stress in men and women with
https://doi.org/10.1016/j.numecd. 2014.09.001. elevated blood pressure. Eur J Clin Nutr. 2007;61:
50. Clark JL, Zahradka P, Taylor CG. Efficacy of fla- 786–95.
vonoids in the management of high blood pressure. 66. Behall KM, Scholfield DJ, Hallfrisch J. Whole-
Nutr Rev. 2015;73(12):799–822. grain diets reduce blood pressure in mildly hyper-
51. Hobbs DA, George TW, Lovegrove JA. The effects cholesterolemic men and women. J Am Diet Assoc.
of dietary nitrate on blood pressure and endothelial 2006;106:1445–9.
function: a review of human intervention studies. 67. He J, Streiffer RH, Muntner P, et al. Effect of
Nutr Res Rev. 2013;26:210–22. dietary fiber intake on blood pressure: a randomized,
52. Micha R, Peñalvo JL, Cudhea F, et al. Association double-blind, placebo-controlled trial. J Hypertens.
between dietary factors and mortality from heart 2004;22:73–80.
disease, stroke, and type 2 diabetes in the United 68. Davy BM, Melby CL, Beske SD, et al. Oat con-
States. JAMA. 2017;317(9):912–24. https://doi. sumption does not affect resting casual and ambu-
org/10.1001/jama.2017.0947. latory 24-h arterial blood pressure in men with
53. http://wholegrainscouncil.org/whole-grains-101/ high-normal blood pressure to stage I hypertension.
what-counts-as-a-serving. Accessed 26 Dec 2015. J Nutr. 2002;132:394–8.
54. Seal CJ, Brownlee IA. Whole-grain foods and 69. Keenan JM, Joel J, Pins JJ, et al. Oat ingestion
chronic disease: evidence from epidemiological and reduces systolic and diastolic blood pressure in
intervention studies. Proc Nutr Soc. 2015;74:313–9. patients with mild or borderline hypertension: a pilot
55. Karl JP, Meydani M, Barnett JB, et al. Substituting trial. J Fam Pract. 2002;51:369–75.
whole grains for refined grains in a 6-wk random- 70. Saltzman E, Das SK, Lichtenstein AH, et al. An
ized trial favorably affects energy-balance metrics in oat-containing hypocaloric diet reduces systolic
healthy men and postmenopausal women. Am J Clin blood pressure and improves lipid profile beyond
Nutr. 2017;105:589–99. effects of weight loss in men and women. J Nutr.
56. Mellen PB, Walsh TF, Herrington DM. Whole grain 2001;131:1465–70.
intake and cardiovascular disease: a meta-analysis. 71. Kestin M, Moss R, Clifton PM, Nestel
Nutr Metab Cardiovasc Dis. 2008;18(4):283–90. PJ. Comparative effects of three cereal brans on
57. Tang G, Wang D, Long J, et al. Meta-analysis plasma lipids, blood pressure, and glucose metabo-
of the association between whole grain intake lism in mildly hypercholesterolemic men. Am J Clin
and coronary heart disease risk. Am J Cardiol. Nutr. 1990;52:661–6.
2015;115(5):625–9. 72. Fehily AM, Burr ML, Butland BK, Eastham RD. A
58. Ye EQ, Chacko SA, Chou EL, et al. Greater whole- randomised controlled trial to investigate the effect
grain intake is associated with lower risk of type 2 of a high fibre diet on blood pressure and plasma
diabetes, cardiovascular disease, and weight gain. fibrinogen. J Epid Commun Health. 1986;40:334–7.
J Nutr. 2012;142:1304–13. 73. Rooney C, McKinley MC, Appleton KM, et al. How
59. Flint AJ, FB H, Glynn RJ, et al. Whole grains and much is ‘5-a-day’? A qualitative investigation into
incident hypertension in men. Am J Clin Nutr. consumer understanding of fruit and vegetable intake
2009;90:493–8. guidelines. J Hum Nutr Diet. 2017;30(1):105–13.
60. Wang L, Gaziano JM, Liu S, et al. Whole- and https://doi.org/10.1111/jhn.12393.
refined-grain intakes and the risk of hypertension in 74. United States Department of Agriculture. 2013.
women. Am J Clin Nutr. 2007;86:472–9. Choose my plate. http://www choosemyplategov/.
61. Evans CEL, Greenwood DC, Threapleton DE, Accessed 17 Feb 2015.
et al. Effects of dietary fibre type on blood pres- 75. Micha R, Khatibzadeh S, Shi P, et al. Global, regional
sure: a systematic review and meta-analysis of and national consumption of major food groups in
References 413
1990 and 2010: a systematic analysis including 266 89. Tjelle TE, Holtung L, Bøhn SK, et al. Polyphenol-
country-specific nutrition surveys worldwide. BMJ rich juices reduce blood pressure measures in a
Open. 2015;5(9):e008705. https://doi.org/10.1136/ randomised controlled trial in high normal and hyper-
bmjopen-2015-008705. tensive volunteers. Br J Nutr. 2015;114:1054–63.
76. Li B, Li F, Wang L, Zhang D. Fruit and vegetables con- 90. Novotny JA, Baer DJ, Khoo C, et al. Cranberry
sumption and risk of hypertension: a meta-analysis. juice consumption lowers markers of cardiometa-
J Clin Hypertens (Greenwich). 2016;18(5):468–76. bolic risk, including blood pressure and circulating
https://doi.org/10.1111/jch.12777. C-reactive protein, triglyceride, and glucose concen-
77. Wu L, Sun D, He Y. Fruit and vegetable consump- trations in adults. J Nutr. 2015;145:1185–93.
tion and incident hypertension: dose response 91. Vinson JA, Demkosky CA, Navarre DA, Smyda
meta- analysis of prospective cohort studies. J MA. High-antioxidant potatoes: acute in vivo anti-
Hum Hypertens. 2016;30(10):573–80. https://doi. oxidant source and hypotensive agent in humans
org/10.1038/jhh.2016.44. after supplementation to hypertensive subjects. J
78. Borgi L, Muraki I, Satija A, et al. Fruit and vegeta- Agric Food Chem. 2012;60:6749–54.
ble consumption and the incidence of hypertension 92. Berry SE, Mulla UZ, Chowienczyk PJ, et al.
in three prospective cohort studies. Hypertension. Increased potassium intake from fruit and vegeta-
2016;67:288–93. bles or supplements does not lower blood pressure
79. Wang L, Manson JE, Gaziano JM, et al. Fruit and or improve vascular function in UK men and women
vegetable intake and the risk of hypertension in with early hypertension: a randomised controlled
middle-aged and older women. Am J Hypertens. trial. Br J Nutr. 2010;104:1839–47.
2012;25(2):180–9. 93. McCall DO, McGartland CP, McKinley MC,
80. Tsubota-Utsugi M, Ohkubo T, Kikuya M. High et al. Dietary intake of fruits and vegetables
fruit intake is associated with a lower risk of future improves microvascular function in hypertensive
hypertension determined by home blood pres-
subjects in a dose-dependent manner. Circulation.
sure measurement: the OHASAMA study. J Hum 2009;119:2153–60.
Hypertens. 2011;25(3):164–71. 94. John JH, Ziebland S, Yudkin P, et al. Effects of fruit
81. Nunez-Cordoba JM, Alonso A, Beunza JJ, et al. Role and vegetable consumption on plasma antioxidant
of vegetables and fruits in Mediterranean diets to pre- concentrations and blood pressure: a randomized
vent hypertension. Eur J Clin Nutr. 2009;63:605–12. controlled trial. Lancet. 2002;359:1969–74.
82. Griep LMO, Stamler J, Chan Q, et al. Association 95. Broekmans WMR, Klopping-Ketelaars WAA, Kluft
of raw fruit and fruit juice consumption with blood C, et al. Fruit and vegetables and cardiovascular FV
pressure: the INTERMAP study. Am J Clin Nutr. intake and risk of CVD risk profile: a diet controlled
2013;97:1083–91. intervention study. Eur J Clin Nutr. 2001;55:636–42.
83. Chan Q, Stamler J, Brown IJ, et al. Relation of raw 96. Smith-Warner SA, Elmer PJ, Tharp TM, et al.
and cooked vegetable consumption to blood pres- Increasing vegetable and fruit intake: random-
sure: the INTERMAP study. J Hum Hypertens. ized intervention and monitoring in an at-risk
2014;28:353–9. population. Cancer Epidemiol Biomarkers Prev.
84. Medina-Remon A, Vallverdu-Queralt A, Arranz 2000;9(3):307–17.
S, et al. Gazpacho consumption is associated with 97. Zheng J, Zhou Y, Li S, et al. Effects and mecha-
lower blood pressure and reduced hypertension in nisms of fruit and vegetable juices on cardiovascu-
a high cardiovascular risk cohort. Cross-sectional lar diseases. Int J Mol Sci. 2017;18:555. https://doi.
study of the PREDIMED trial. Nutr Metab Cardiov org/10.3390/ijms18030555.
Dis. 2013;23(10):944–52. 98. Sahebkar A, Ferri C, Giorgini P, et al. Effects of
85. Pase MP, Grima N, Cockerell R, Pipingas A. Habitual pomegranate juice on blood pressure: a systematic
intake of fruit juice predicts central blood pressure. review and meta-analysis of randomized controlled
Appetite. 2015;84:68–72. trials. Pharmacological Res. 2017;115:149–61.
86. Hartley L, Igbinedion E, Holmes J, et al. Increased 99. Bouchenak M, Lamri-Senhadji M. Nutritional
consumption of fruit and vegetables for the primary quality of legumes, and their role in cardiometa-
prevention of cardiovascular diseases. Cochrane bolic risk prevention: a review. J Med Food.
Database of Systematic Rev. 2013;6:CD009874. 2013;16(3):185–98.
https://doi.org/10.1002/14651858.CD009874.pub2. 100. Tielemans SMAJ, Altorf-van der Kuil W, Engberink
87. Liu K, Xing A, Chen K. Effect of fruit juice on cho- MF, et al. Intake of total protein, plant protein and
lesterol and blood pressure in adults: a meta-analysis animal protein in relation to blood pressure: a meta-
of 19 randomized controlled trials. PLoS One. analysis of observational and intervention studies.
2013;8(4). https://doi.org/10.1371/journal. pone. J Human Hypertension. 2013;27:564–71.
0061420. 101. Messina V. Nutritional and health benefits of dried
88. Jovanovski E, Bosco L, Khan K, et al. Effect of spin- beans. Am J Clin Nutr. 2014;100(suppl 1):437S–42S.
ach, a high dietary nitrate source, on arterial stiffness 102. Afshin A, Micha R, Khatibzadeh S, Mozaffarian
and related hemodynamic measures: a randomized, D. Consumption of nuts and legumes and risk of
controlled trial in healthy adults. Clin Nutr Res. incident ischemic heart disease, stroke, and diabetes:
2015;4:160–7. a systematic review and meta-analysis. Am J Clin
Nutr. 2014;100(1):278–88.
103. Jayalath VH, de Souza RJ, Sievenpiper JL, et al. 120. Ma L, Wang F, Guo W, et al. Nut consumption and
Effect of dietary pulses on blood pressure: a system- the risk of coronary artery disease: a dose-response
atic review and meta-analysis of controlled feeding meta-analysis of 13 prospective studies. Thromb
trials. Am J Hypertension. 2014;27(1):56–64. Res. 2014;134(4):790–4.
104. Lee YP, Puddey IB, Hodgson JM. Protein, fibre and 121. Yazdekhasti N, Mohammadifard N, Sarrafzadegan
blood pressure: potential benefit of legumes. Clin N, et al. The relationship between nut con-
Exp Pharm Physiol. 2008;35:473–6. sumption and blood pressure in an Iranian adult
105. Appel LJ, Sacks FM, Carey VJ, et al. Effects of population: Isfahan Healthy Heart. Program.
protein, monounsaturated fat, and carbohydrate 2013;23(10):929–36.
intake on blood pressure and serum lipids results 122. Djoussé L, Rudich T, Gaziano JM. Nut consumption
of the OmniHeart Randomized Trial. JAMA. and risk of hypertension in US male physicians. Clin
2005;294:2455–64. Nutr. 2009;28(1):10–4.
106. Yang G, Shu XO, Jin F, et al. Longitudinal study of 123. Weng L-C, Steffen LM, Szklo M, et al. Diet pattern
soy food intake and blood pressure among middle- with more dairy and nuts, but less meat is related
aged and elderly Chinese women. Am J Clin Nutr. to lower risk of developing hypertension in
2005;81:1012–7. middle-aged adults: the Atherosclerosis Risk in
107. Nagata C, Shimizu H, Takami R, et al. Association Communities (ARIC) study. Forum Nutr. 2013;5:
of blood pressure with intake of soy products and 1719–33.
other food groups in Japanese men and women. Prev 124. Martınez-Lapiscina EH, Pimenta AM, Beunza
Med. 2003;36:692–7. JJ, et al. Nut consumption and incidence of
108. Dong J-Y, Tong X, Z-W W, et al. Meta-analysis: hypertension: the SUN prospective cohort. Nutr

effect of soya protein on blood pressure: a meta- Metab Cardiov Dis. 2010;20(5):359–65.
analysis of randomised controlled trials. Br J Nutr. 125. Del Gobbo LC, Falk MC, Feldman R, et al. Effects
2011;106:317–26. of tree nuts on blood lipids, apolipoproteins, and
109. Rivas M, Garay RP, Escanero JF, et al. Soy milk blood pressure: systematic review, meta-analysis,
lowers blood pressure in men and women with and dose-response of 61 controlled intervention tri-
mild to moderate essential hypertension. J Nutr. als. Am J Clin Nutr. 2015;102(6):1347–56. https://
2002;132:1900–2. doi.org/10.3945/ajcn. 115.110965.
110. He J, Wofford MR, Reynolds K, et al. Effect of 126. Rodriguez-Leyva D, Weighell W, Edel AL, et al.
dietary protein supplementation on blood pres- Potent antihypertensive action of dietary flaxseed
sure: a randomized controlled trial. Circulation. in hypertensive patients. Hypertension. 2013;62:
2011;124(5):589–95. 1081–89.
111. He J, Gu D, Xigui W, et al. Effect of soybean protein 127. Khosravi-Boroujeni H, Nikbakht E, Natanelov E,
on blood pressure: a randomized, controlled trial. Khalesi S. Can sesame consumption improve blood
Ann Intern Med. 2005;143(1):1–9. pressure? A systematic review and meta-analysis of
112. Ros E. Nuts and CVD. Br J Nutr. 2015;113:S111–20. controlled trials. J Sci Food Agric. 2017; doi: https://
113. Prasad K. Flaxseed and cardiovascular disease. J doi.org/10.1002/jsfa.8361.
Cardiovasc Pharmacol. 2009;54(5):369–77. 128. Cardoso CA, de Oliveira GM, Gouveia LA, et al. The
114. Novotny JA, Gebauer SK, Baer DJ. Discrepancy effect of dietary intake of sesame (Sesamumindicum
between the Atwater factor predicted and empiri- L.) derivatives related to the lipid profile and blood
cally measured energy values of almonds in human pressure: a systematic review. Crit Rev Food Sci
diets. Am J Clin Nutr. 2012;96:296–301. Nutr. 2016; doi: https://doi.org/10.1080/10408398.
115. Mattes RD, Kris-Etherton PM, Foster 2015.1137858.
GD. Impact of peanuts and tree nuts on body 129. Yarmolinsky J, Gon G, Edwards P. Effect of tea
weight and healthy weight loss in adults. J Nutr. on blood pressure for secondary prevention of car-
2008;138(suppl):1741S–5S. diovascular disease: a systematic review and meta-
116. Traoret CJ, Lokko P, Cruz ACRF, et al. Peanut diges- analysis of randomized controlled trials. Nutr Rev.
tion and energy balance. Int J Obes. 2008;32:322–8. 2015;73(4):236–46.
117. Brown RC, Tey SL, Gray AR, et al. Nut consump- 130. Peng X, Zhou R, Wang B, et al. Effect of green
tion is associated with better nutrient intakes: results tea consumption on blood pressure: a meta-anal-
from the 2008/09 New Zealand Adult Nutrition ysis of 13 randomized controlled trials. Sci Rep.
Survey. Br J Nutr. 2016;115:105–12. 2014;4:6251.
118. Nielsen SJ, Kit BK, Ogden CL. Nut consumption 131. Greyling A, Ras RT, Zock PL et al. The effect of
among U.S. adults, 2009-2010. NCHS data brief, black tea on blood pressure: a systematic rReview
no 176. Hyattsville, MD: National Center for Health with meta-analysis of randomized controlled trials.
Statistics 2014. PLoS ONE. 2014;9(7):e103247.
119. Luu HN, Blot WJ, Xiang Y-B, et al. Prospective eval- 132. Liu G, Mi X-N, Zheng X-X, et al. Effects of tea
uation of the association of nut/peanut consumption intake on blood pressure: a meta-analysis of ran-
with total and cause-specific mortality. JAMA Intern domised controlled trials. Br J Nutr. 2014;112(7):
Med. 2015;175(5):755–66. 1043–54.
References 415
133. Cook NR. Implications of small reductions in dia- 137. Turnbull D, Rodricks JV, Mariano GF, Chowdhury
stolic blood pressure for primary prevention. Arch F. Caffeine and cardiovascular health. Regul Toxicol
Intern Med. 1995;155(7):701–9. Pharmacol. 2017;89:165–85.
134. Miranda A, Steluti J, Fisberg R, Marchioni D. 138. Zhang Z, Hu G, Caballero B, et al. Habitual coffee
Association between coffee consumption and its consumption and risk of hypertension: a systematic
polyphenols with cardiovascular risk factors: a pop- review and meta-analysis of prospective observa-
ulation-based study. Nutrients. 2017;9(3):276. tional studies. Am JClin Nutr. 2011;93(6):1212–19.
135. Loader TB, Taylor cG, Zahradka P, Jones PJH. 139. Rhee JJ, Qin F, Hedlin HK, et al. Coffee and caf-
Chlorogenic acid from coffee beans: evaluating the feine consumption and the risk of hyperten-
evidence for a blood pressure – regulating health sion in postmenopausal women. Am J Clin Nutr.
claim. Nutr Rev. 2017;75(2):114–33. 2016;103(1):210–17.
136. Horst K, Robinson WD, Jenkins WL, Bao DL. The 140. Mesas AE, Leon-Munoz LM, Rodriguez-Artalejo F,
effect of caffeine, coffee and decaffeinated coffee Lopez-Garcia E. The effect of coffee on blood pres-
upon blood pressure, pulse rate and certain motor sure and cardiovascular disease in hypertensive indi-
reactions of normal young men. J Pharmacol Exp. viduals: a systematic review and meta-analysis. Am
1934;52:307–21. J Clin Nutr. 2011;94(4):1113–26.
Dietary Patterns, Foods
and Beverages in Chronic Kidney 15
Disease
Keywords
Chronic kidney disease • Dietary fiber • Dietary patterns • Hypertension
• Inflammation • Microalbuminuria • Obesity • Glomercular filtration rate
• Whole foods
Key Points • Increased fiber intake triggers a number of

• The prevalence of chronic kidney disease physiologic processes in both the colon
(CKD) is high (estimated 200 million people microbiota and systemically that support the
worldwide) and steadily increasing, especially detoxification of the kidneys, via influences
in older populations, and it is associated with on the gut barrier, gastrointestinal immune
increased risk of renal cancer, cardiovascular and endocrine responses, nitrogen cycling,
disease, and bone disorders and fractures. and microbial metabolism which alter the
• The Western diet is associated with increased physiology and biochemistry of the kidneys
renal dysfunction, CKD risk and progression to help re-establish homeostasis.
to end stage renal disease (ESRD). • Healthy dietary patterns including fiber-rich
• A healthy diet for CKD patients should help to whole-grains, fruits and vegetables may
slow the rate of progression of kidney failure, improve renal function, and decrease meta-
reduce uremic toxicity, decrease proteinuria, bolic acidosis compared to poor quality diets
and lower the risk of secondary complications low in fruits and vegetables and high in pro-
including cardiovascular disease, bone dis- cessed foods and animal products. High
ease, and hypertension. adherence to healthy dietary patterns such as
• Lower dietary energy density and higher fiber the Dietary Approaches to Stop Hypertension
healthy dietary patterns can play a role in low- (DASH), especially a modified version of the
ering the risk of CKD. Protein sources vary in DASH diet for people with CKD, and the
their effect on CKD risk with red and pro- Mediterranean diet (MedDiet) may help to
cessed meat consumption significantly increas- reduce CKD risk, progression to later stages
ing risk whereas higher intake of nuts, legumes, and mortality.
and low-fat dairy products may lower risk.

418 15 Dietary Patterns, Foods and Beverages in Chronic Kidney Disease
15.1 Introduction filtration rate (eGFR) ≥90 mL/min per 1.73 m2)

with persistent proteinuria (≥3 months); (2) kid-
The kidney is a highly-vascularized organ, which ney damage with mild loss of kidney function
plays a major role regulating electrolyte concen- (eGFR 60–89 mL/min per 1.73 m2) and persistent
trations, blood pressure, energy metabolic hor- proteinuria (≥3 months); (3) mild-to-severe loss
mones, and excretion of waste metabolites. The of kidney function (eGFR 30–59 mL/min per
prevalence of chronic kidney disease (CKD) is 1.73 m2); (4) severe loss of kidney function (eGFR
high (estimated 200 million people worldwide) 15–29 mL/min per 1.73 m2); and (5) kidney fail-
and steadily increasing, especially in older popu- ure or ESRD (eGFR <15 mL/min per 1.73 m2)
lations, and it is associated with increased risk of requiring dialysis or transplant for survival [7].
renal cancer, cardiovascular disease, and bone Western dietary patterns, diabetes and obesity
disorders and fractures [1–9]. In the US, the esti- are important risk factors associated with the
mates of CKD in the general population is 14% increased risk of CKD [10, 11]. Western dietary
with approximately 500,000 individuals on dialy- patterns, rich in refined carbohydrates, salt, fat
sis and 200,000 living with kidney transplants [7]. and protein from red meat, and low in fiber-rich
The prevalence of CKD in individuals over age foods, are generally associated with increased
60 is estimated to be up to 25% [3–5]. African risk of CKD. A systematic review and meta-
Americans have a four-fold higher risk of CKD analysis (39 cohort studies; 630,677 participants;
compared to Caucasians. CKD patients experi- mean 6.8 years of follow-up) found that obesity
ence an increased rate of mortality by 59% com- in the general population increased risk of low
pared with healthy individuals as more people die eGFR by 28% and albuminuria by 51% [12]. A
from CKD each year than from breast or prostate high BMI predicts onset of albuminuria without
cancer. CKD usually develops over many years kidney failure (CKD stages 1–2) as well as CKD
and leads to end-stage kidney (or renal) disease stages 3 and higher. Abdominal obesity is consid-
(ESRD) [7]. There are 5 stages of CKD: (1) nor- ered a major risk factor for CKD development
mal kidney function (estimated glomerular and progression (Fig. 15.1) [13–16]. An analysis
Increased risk of insulin Abdominal obesity, visceral fat, Elevated blood

resistance, type ectopic fat, large waist size, pressure,
2 diabetes metabolic syndrome hypertension risk
Glomerular hyper-filtration/perfusion
Microalbuminuria and/or reduced

glomerular filtration rate
Altered metabolism and Chronic kidney End stage kidney

adipokines disease (CKD)risk disease (ESKD)
Fig. 15.1 Effect of abdominal obesity related conditions and chronic kidney disease (CDK) risk (adapted from
[13–16])
15.3 Protein and Dietary Fiber 419
of data from the US NHANES 1999–2010 (6918 limiting or avoiding sugar-sweetened beverages,
young adults; ages 20–40 years) found that has been shown to reduce CKD risk [19]. The
abdominal obesity in young adults, especially in Atherosclerosis Risk in Communities (ARIC)
Mexican-Americans, was independently associ- cohort study (14,832 participants; mean age
ated with 3.5-fold higher odds of the CKD risk 54 years; 55% female; 26% blacks; BMI range
factor albuminuria even with normal blood pres- 23–31; median 22-year follow-up; 2743 CKD
sures, normoglycemia and normal insulin levels cases) showed a significant inverse association
[17]. Initially, ectopic fat causes renal vasodila- between the adherence to Life’s Simple 7 goals
tion and glomerular hyper-filtration, which act as and incident CKD (Fig. 15.2) [19]. Individual
compensatory mechanisms to maintain sodium factors significantly associated with lower CKD
balance but the increased arterial pressure and risk were non-cigarette smoking, BMI < 25,
metabolic abnormalities, may ultimately lead to increased physical activity, and lower BP and
glomerular injury [18]. Specifically, increased fasting blood glucose but not healthy diet score
ectopic fat in the kidney can cause physical com- or total cholesterol. The Chronic Renal
pression of the kidney’s renal vein and artery that Insufficiency Cohort (CRIC) Study (3006 per-
pass through the renal sinus increasing renal sons with mild-to-moderate CKD; mean age
interstitial pressure, decreasing sodium excretion 58 years; 48% female; 47% non-Hispanic white,
and stimulating inflammation, oxidative stress, 45% diabetes; median follow-up of 4 years; 726
and lipotoxicity factors that may also contribute CKD progression events, 353 atherosclerotic
to renal dysfunction associated with hypertension events, and 437 deaths) found that greater adher-
and CKD. ence to all components of a healthy lifestyle were
Healthy diets are important for patients with associated with 68% reduced risk for adverse
CKD as proper nutrition helps to prevent infection outcomes, including progression of CKD, athero-
and prevent kidney disease from worsening. sclerotic events and all-cause mortality [20].
Nutrition requirements differ depending on the
level of kidney function and the presence of co-
morbid conditions, including hypertension, diabe- 15.3 Protein and Dietary Fiber
tes, and cardiovascular disease. The CKD patients’
diet should help to slow the rate of progression of 15.3.1 Protein
kidney failure, reduce uremic toxicity, decrease
proteinuria, and lower the risk of secondary com- Dietary proteins are digested to amino acids,
plications including cardiovascular disease, bone including acidic amino acids such as aspartic and
disease, and hypertension. The objective of this glutamic and basic amino acids with an amine
chapter is to comprehensively review the effects of functional such as lysine, arginine, and histidine
dietary patterns, foods, nutrients and beverages on [21]. Proteins from meat products (from a typical
CKD risk and management. Western diet) generate predominantly acidic
products including hydrogen chloride, sulfuric
acid, and phosphoric acids. These acids are non-
15.2 L
ifestyle and Chronic Kidney volatile and rely on the kidneys for their excre-
Disease (CKD) tion (primarily in the form of ammonium salts
and phosphoric salts). A healthy individual gen-
The American Heart Association (AHA) Life’s erates net acids, approximately 1 mEq/kg/day
Simple 7, which includes: (1) non-smoking or (mmol/kg/day), which is rapidly buffered by
quit >1 year ago; (2) BMI < 25; (3) blood pres- sodium bicarbonate to form sodium salts. During
sure (BP) <120/80 mm Hg; (4) ≥ 150 min/week this process, bicarbonate is consumed, which
of physical activity; (5) a healthy dietary pattern needs to be regenerated, a task accomplished by
(high in fruits and vegetables, fish, fiber-rich the kidneys. In patients with reduced kidney
whole grains, (6) low intake of sodium, and (7) function, nonvolatile acids can accumulate
0.8
0.7
Risk of CKD incidence 0.6
0.5
0.4
0.3
0.2
0.1
0
1 2 3 4 5 6 or 7
Number of Ideal Healthy Factors*
Fig. 15.2 Dose-response relationship between number of ideal Life’s Simple 7 health factors and chronic kidney dis-
ease (CKD) incidence (p-trend <0.001; multivariate adjusted) (adapted from [19]). * (1) non-smoking or quit >1 year
ago; (2) BMI < 25; (3) blood pressure (BP) <120/80 mm Hg; (4) ≥ 150 min/week of physical activity; (5) healthy
dietary pattern (high in fruits and vegetables, fish, fiber-rich whole grains; (6) low intake of sodium; and (7) low intake
or avoidance of sugar-sweetened beverages
causing metabolic acidosis and leading to pro- eGFR [23]. Intakes of total and plant protein, and
gressive kidney damage associated with defects dairy products are generally not associated with
in acid excretion, systemic inflammation, end- eGFR changes.
organ hormone resistance and uremic toxin
accumulation. Consequently, dietary protein
restriction is recommended for patients with 15.3.2 Dietary Fiber
moderate to severe renal insufficiency with plant
and low-fat dairy protein sources more effective Lower dietary energy density and higher fiber
than red and processed meat sources in reducing intake from fruits , vegetables, whole grains,
the risk and progression of CKD [22]. In the 2017 legumes and nuts are associated with better
US Atherosclerosis Risk in Communities study overall health and lower risk of CKD risk and
(11,952 adults; aged 44–66 years; estimated progression [24–26]. However, fruits and vege-
baseline glomerular filtration rate (eGFR) tables, important fiber sources but rich in potas-
≥ 60 mL/min/1.73 m2; 23 years of follow-up; sium, are often restricted in advanced CKD to
2.632 CKD cases) found that red and processed prevent or correct hyperkalemia [26]. Examples,
meat consumption was associated with signifi- of fiber-rich lower potassium fruit and vegeta-
cantly increased CKD risk by 23% whereas bles include apples, blueberries, cabbage, red
higher intake of nuts, legumes, and low-fat dairy bell peppers, cauliflower and onions. Adequate
products was associated with a significantly fiber intake is particularly important for pro-
lower CKD risk by 12 to 17% (Fig. 15.3) [22]. moting kidney health and protecting against
The Netherlands Doetinchem Cohort study (3798 advanced stage CKD by triggering a number of
participants, mean baseline age 45 years; 52% physiologic processes in both the colon micro-
women; examined 3 times 5 years apart) found biota and systemically that lead to the detoxi-
that the daily consumption of ≥2 servings milk fication of kidneys, via its influences on the
and milk products or low-fat dairy was associated gut barrier, gastrointestinal immune and endo-
with less annual decline in the eGFR in the sub- crine responses, nitrogen cycling and micro-
group of participants with mildly decreased bial metabolism which alter the physiology and
15.3 Protein and Dietary Fiber 421
30
20
10
CKD Risk (%)
0
Red/Processed Nuts Legumes Low Fat Dairy
Meats (p-trend = (p-trend <.001) (p-trend =.03) (p-trend <.001)
.01)
–10
–20
–30
Fig. 15.3 Association between protein source and chronic kidney disease (CKD) risk (adapted from [22])
biochemistry of the kidneys to help re-establish those without CKD but was inversely related
homeostasis [27]. Further, higher fiber intake to mortality in those with kidney disease.
increases microbial sequestration of nitrogen in Each 10 g/day increase in intake was associ-
the colon, resulting in increased fecal nitrogen ated with reduced overall mortality risk for
excretion and reduced concentrations of nitrog- total fiber by 17%, for insoluble fiber by 23%
enous metabolites in the blood, which reduces and soluble fiber by 33%. This study suggests
nitrogenous burden on the kidneys helping that increased fiber intake in CKD individuals
reduce risk of CKD development or progres- may lower systemic inflammation and mortality
sion. A systematic review and meta-analysis (14 risk. The Uppsala Longitudinal Study of Adult
RCTs;143 CKD patients; median age 52 years; Men (1110 community-dwelling elderly men
fiber supplemented/high fiber diets vs. non-sup- from Sweden; mean age 71 years; mean BMI
plemented/ low-fiber diets; median fiber 27 g/ 26; median 10 years of follow-up; 300 deaths,
day; median protein 60 g/day; median follow- 138 cardiovascular disease, 111cancer, 19 infec-
up 4.5 weeks) found that fiber supplementation tions, 33 other causes) showed that high fiber
significantly reduced serum urea by 1.8 mmol/L intake was associated with significantly bet-
and serum creatinine by 22.8 mmol/L [28]. ter kidney function (Fig. 15.4), lower odds of
Several observational studies demonstrated that having C-reactive protein (CRP) >3 mg/L and
increased fiber intake may be especially effec- reduced risk of mortality [30]. Total fiber was
tive at reducing systemic inflammation in CKD independently and directly associated with sig-
patients and mortality risk [29, 30]. The Third nificantly improved eGFR (adjusted difference,
National Health and Nutrition Examination 2.6 mL/min per 1.73m2) per 10 g/day higher
Survey (NHANES III) (14,533 adults; mean intake. Meta-analysis (14 RCTs) showed that
age 45 years; 48% males; prevalence of CKD increased fiber intake by 8 g/day from supple-
5.8%) found that for each 10 g/day increase in ments or fiber-rich foods compared with con-
total fiber intake, the odds of elevated serum trol significantly reduced CRP levels by about
C-reactive protein levels were decreased in indi- 0.5 mg/L and for those individuals with elevated
viduals without CKD by 11% and with CKD by CRP the reduction was 0.72 mg/dL (p = 0.06)
38% [29]. Also, total fiber intake was not sig- [31]. Lowering CRP has been associated with
nificantly associated with overall mortality in reduced incidence and complication of CKD as
3.5
Kidney Function (eGFR/1.73 m2) 2.5
1.5
0.5
0
> 14.5 -16.8 g >16.8 -19.2 g > 19.2 g
Total Fiber (g/day)
Fig. 15.4 Association between fiber intake/day in elderly men with chronic kidney disease (CKD) and kidney
glomerular filtration rate (eGFR) (p = 0.02) (adapted from [30])
elevated CRP has been associated with impaired 15.4 Whole Plant Foods
eGFR in hypertensive adults with a mean age of
60 years [32]. There is increasing clinical evi- 15.4.1 Whole-grains
dence that individuals at risk or patients with
CKD have distinctly dysbiotic colonic micro- Organizations such as the National Kidney
biota, which can activate a cascade of metabolic Foundation, the American Kidney Fund, the
abnormalities, including uremic toxin produc- National Institute of Diabetes and Digestive and
tion, inflammation, and immunosuppression, Kidney Diseases, and the US Department of
that may lead to CKD or ultimately promotes Health and Human Services have previously rec-
progressive kidney failure [33, 34]. Several ommended not including whole grains as part of
human studies show that increased fiber intake the renal diet because of the potential risk of
can lower circulating p-cresol and indoxyl sul- excessive phosphorus intake [37]. However, the
fate [34, 35]. A single-blind RCT (6 males and phosphorus content in whole-grains is cova-
7 females; mean age 65 years; eGFR <50 mL/ lently bound to organic molecules (primarily
min/1.73m2) showed that increasing fiber intake phytate) and requires the enzyme phytase to be
from 17 g to 27 g/day significantly reduced cir- released to become available for absorption.
culating p-cresol by 20% [35]. In a cross-sec- While some phytase is contained in some raw
tional analysis (40 CKD patients; mean age 69; whole grains (corn, oats, and millet have little to
60% male; 45% diabetic; mean estimated eGFR no phytase activity), the enzyme is decreased in
of 24 mL/min/1.73 m2), it was demonstrated milling, food preparation and over time. Also,
that total fiber intake was significantly associ- since the enzyme required for the release of
ated with lower free and total serum p-cresol phosphorus from phytate is not present in the
sulfate but not indoxyl sulfate [36]. This kid- human intestinal lumen, when ingesting cooked
ney-colonic axis may provide new nutritional food, the bioavailability of phosphorus from
CKD therapeutic opportunities involving the whole grains is low. The Australian Blue
microbiota and fiber, prebiotics, probiotics, and Mountains Eye Study (2600 participants; aged
symbiotics. Appendix A provides a list of 50 top ≥50 years; 19.4% had moderate CKD and 80.6%
fiber-rich plant foods. did not have CKD; 5 years of follow-up) found
15.5 Food Ingredients and Beverages 423
Cereal fiber (p-trend =.03) Energy-dense, nutrient poor diet (p-trend =.01)
Sugar (p-trend =.06)
4
incidence of CKD (eGFR < 60 ml/min 1.73m2) 3.5
2.5
1.5
0.5
0
1 2 3 4
Intake Quartile
Fig. 15.5 Dietary factors associated with 5-year incidence of chronic kidney disease (CKD) from a cross-sectional
analysis of 2600 Australian adults (≥50 years) in the Blue Mountains Eye Study (adapted from [38])
that fiber from cereal (predominantly from rolled equally effective in reducing urinary acidosis and
oats and wholemeal/wholegrain breads) signifi- in preserving glomerular filtration rate [41, 42].
cantly lowered adjusted incidence of CKD by An Australian prospective cohort study (145
50% and energy dense, nutrient poor foods sig- adults with stage 3 or 4 CKD; 3 years of follow-
nificantly increased risk by 220% (Fig. 15.5) up) found that increased fruits and vegetables
[38]. Consequently, it has been proposed that lowered risk of CKD progression by 39% and
modest consumption of whole-grains by CKD improved survival by 65% [43].
patients may provide benefits to help protect
against CKD and other chronic diseases. A par-
allel RCT (52 CKD patients; control diet vs. 15.5 Food Ingredients
control plus 50 g oats/day) found that the added and Beverages
oats beneficially lowered serum albumin and
serum potassium [39], This is consistent with 15.5.1 Minerals
the fiber -colon microbiota- kidney detoxifica-
tion axis described previously [28]. 15.5.1.1 Sodium (Salt)
High salt intake is associated with both increased
blood pressure (BP) and worsening of kidney
15.4.2 Fruits and Vegetables function. A 2015 Cochrane Systematic Review (8
RCTs; 258 subjects with CKD; 1 to 26 weeks)
A highly adhered to Western-type diet is deficient found that salt reduction in people with CKD
in fruits and vegetables and contains excessive reduced BP, 24-hour sodium excretion and pro-
red and processed meat products which can proteinuria, which are predictive of long-term
mote metabolic acidosis, which increases pro- improvement of kidney function, slower progres-
gressively with aging due to the physiological sion to ESRD and reduced mortality risk [44].
decline in kidney function [40]. Two RCTs in However, longer-term RCTs with CKD patients
stage 3 or 4 CKD patients suggest that increased on restricted salt diets are needed to confirm the
fruit and vegetable intake or oral bicarbonate are benefits suggested by the current short-term RCTs.
15.5.1.2 Phosphorous (Convenience 15.5.2.2 Alcohol

and Fast Foods) A systematic review and meta-analysis (20
In the Western diet, there is increased use of cohort studies; 292,431 subjects) found that
phosphorus-containing ingredients in processed CKD, proteinuria and ESRD were not associated
convenience food and fast foods. Acute studies with moderate alcohol intake but excessive alco-
in healthy young adults demonstrate that phos- hol intake or alcoholism is detrimental to health
phorus intakes in excess of nutrient needs may overall including the kidneys [49].
significantly disrupt the hormonal regulation of
phosphorus contributing to dysfunctional min- 15.5.2.3 Coffee
eral metabolism, vascular calcification, bone A meta-analysis (4 observational studies; 14,898
loss, and impaired kidney function [45, 46]. individuals) found that consuming coffee in the
Although physiological systems try to maintain overall population reduced CKD risk by 29%, and
serum phosphorus at relatively constant levels, coffee intake >1 cup/day was associated with
through the influence of multiple factors-such slightly higher eGFR in middle age and older
as aging, parathyroid hormone, fibroblast adults (> 46 years) [50]. The subgroup analysis
growth factor 23, and vitamin D, a phosphorus showed that coffee was more effective in reduc-
imbalance can affect kidney, bone health ing CKD risk in women than in men. Overall, this
(osteoporosis), and the digestive system. study concludes that coffee consumption should
Excessive dietary phosphate intake can elevate not be a concern for CKD risk in the general popu-
serum phosphate even in individuals with nor- lation. The various components of coffee that may
mal kidney function and higher serum phospho- preserve the glomerular endothelial cells from oxi-
rus has been associated with increased adverse dative stress include caffeine, hydroxychloro-
events and cardiovascular- related mortality quine, quinides, niacin and chlorogenic acid.
both in people with CKD and in those with no
evidence of disease.
15.6 Dietary Patterns
15.5.1.3 Magnesium
The Healthy Aging in Neighborhoods of Western diets have resulted in increased intake of
Diversity Across the Life Span Study (177 par- animal protein, refined carbohydrates, phosphate-
ticipants; mean age 47 years; 5-year follow-up) and sodium-based preservatives, and led to higher
found that low magnesium intake was signifi- risk of abdominal obesity, and systemic inflam-
cantly associated with a doubling of the odds for mation associated with CKD risk and progression
rapid eGFR decline, after multivariate adjust- to end stage renal disease (ESRD) [10, 11, 41].
ments [47]. Good sources of magnesium Healthy dietary patterns consisting of mostly
include: green leafy vegetables, such as spinach, whole plant foods and lower in processed foods
legumes, nuts, seeds, and whole grains. and meats have been shown to improve renal
function, reduce CKD risk and delay progression
and mortality risk in older adults at increased risk
15.5.2 Beverages of CKD. While healthy dietary patterns, such as
the DASH or Mediterranean diets (MedDiets),
15.5.2.1 Sugar Sweetened Beverages may help prevent CKD progression in early stage,
A systemic review and meta-analysis (5 specific renal diets with lower protein, sodium
sugar and 4 artificially sweetened cohorts) and phosphorus may be required in advance
found that high intake of sugar sweetened bev- stages of renal insufficiency. Table 15.1 provides
erages was associated with a 25% higher risk a summary of 2 meta- analyses, 9 prospective
CKD progression than artificially sweetened cohort studies and 2 intervention trials including
sodas [48]. one DASH diet and one MedDiet [51–63].
Table 15.1 Summary of dietary quality and dietary pattern studies on the risk and progression of chronic kidney dis-
ease (CKD) and mortality
Meta-analyses
Kelly et al. (2017) 7 cohort studies; 15,285 participants Six studies showed that healthy dietary
Assess the effect of healthy with CKD; 3983 events; healthy patterns were consistently associated with
dietary patterns on CKD risk dietary patterns generally consisted 27% lower mortality risk. One study found
of end stage renal disease of higher fruits and vegetables, fish, no significant association with a healthy diet
(ESRD) and mortality [51] legumes, cereals, whole grains and and ESRD risk
fiber and lower in red meats, salt
and added sugar
Lui et al. (2017) 1534 participants; mean age High energy intake and low DASH diet
Explore the association 48 years; 59% African-American; intake in participants with hypertension had
between a DASH diet and baseline estimated globular filtration a 68% increased risk of rapid decline of
CKD end points ([Healthy rate (eGFR) > 60 mL/min/1.73 m2; eGFR but not among participants without
Aging in Neighborhoods of 5 years of follow-up (multivariate hypertension
Diversity Across the Life Span adjusted)
study; US) [52]
Rebholz et al. (2016) 3720 participants with baseline A low DASH diet score was associated with
Assess the effect of the DASH eGFR >60 mL/min/1.73 m2; 16% higher risk of CKD (p-trend <0.001;
diet on risk of CKD 23 years of follow-up (multivariate multivariate adjusted)
(Atherosclerosis Risk in adjusted)
Communities (ARIC) Study;
US) [53]
Smyth et al. (2016) 544,635 participants; age All the healthy dietary patterns were
Investigate the effect of diet 51–70 years; diets: Alternate associated with significantly improved renal
quality on renal outcomes Healthy Eating Index (AHEI), function by 18–29%. Greater than 3.6 g
(NIH-AARP Diet and Health Healthy Eating Index (HEI), sodium/day was associated with a 17%
Study; US) [54] Mediterranean diet (MedDiet) score, increased risk of renal dysfunction and
Dietary Approach to Stop adequate potassium was associated with a
Hypertension (DASH) scores; 17% reduced risk of renal dysfunction
14.3 years of follow-up (multivariate
adjusted)
Banerjee et al. (2015) 1486 adults with CKD; median Higher levels of dietary acid load typical of
Examine the association 14.2 years of follow-up; 311 the Western diet were associated with a
between dietary acid load and (20.9%) participants developed doubling of the risk of ESRD; for the highest
progression to ESRD (US ESRD (multivariate adjusted) tertile and 81% for the middle tertile
National Health and Nutrition compared with the lowest tertile in the fully
Examination adjusted model
Survey III; US) [55]
Foster et al. (2015) 1802 participants; mean age Higher diet quality was associated with a
Determine the association of 59 years, 54.8% women; measures significant 37% reduced odds of renal
lifestyle characteristics with of diet quality, physical activity, dysfunction in fully adjusted models. Higher
estimated glomerular filtration alcohol intake, current smoking diet quality was associated with 31% lower
rate (eGFR) <60 mL/ status; 6.6 years of follow-up; 9.5% risk of rapid eGFR decline.
min/1.73 m2 and rapid eGFR of participants developed incident No associations were observed with physical
decline in older adults eGFR <60 (multivariate adjusted) activity, smoking status, or alcohol intake
(Framingham Offspring Study; with incident eGFR <60 or rapid eGFR
US) [56] decline
(continued)
Gutierrez et al. (2014) 3972 participants with CKD; mean Higher plant-based pattern scores were
Assess relationships between age 69 years; 5 empirically derived associated with lower risk of mortality by
dietary patterns and health dietary patterns identified via factor 23% whereas higher Southern pattern scores
outcomes in persons with analysis: convenience (Chinese and were associated with greater risk of mortality
CKD. (Reasons for Mexican foods, pizza, other mixed by 51% (highest vs. lowest quartiles)
Geographic and Racial dishes), plant-based (fruits, (Fig. 15.6). There were no associations of
Differences in Stroke vegetables); sweets/fats (sugary dietary patterns with incident ESRD in
(REGARDS) study; US) [57] foods); Southern (fried foods, organ multivariable-adjusted models
meats, sweetened beverages); and
alcohol/salads (alcohol, green-leafy
vegetables, salad dressing); 6 years
of follow-up; 816 deaths and 141
ESRD events (multivariate adjusted)
Chang et al. (2013) 2354 African-American and white Poor diet quality and obesity were
Investigate the effect of the participants; mean age 35 years; significantly associated with about a 100%
DASH diet on risk of coronary 47% male; DASH vs. Western diets; increased risk of microalbuminuria after
heart disease and CKD 15-year follow-up; 3.3% developed multivariate adjustments. Also, compared to
(Coronary Artery Risk incident Microalbuminuria individuals with no unhealthy lifestyle-
Development in Young Adults (multivariate adjusted) related factors (poor diet quality, current
[CARDIA] Study; US) [58] smoking and obesity), those with 1, 2 and 3
unhealthy lifestyle factors had increased risk
of microalbuminuria by 31%, 173%, and
534%, respectively
Huang et al. (2013) 1110 men; mean age 70 years; Adherence to the MedDiet was associated with
Test the hypothesis that MedDiet Score; follow-up of lower odds of CKD or mortality in elderly men.
adherence to the MedDiet may 9.9 years; 168 deaths (multivariate Compared with low adherence, medium and
better preserved kidney adjusted) high MedDiet adherences were significantly
function (Uppsala associated with lower risk of CKD by 23% and
Longitudinal Study of Adult 42%, respectively (Fig. 15.7). Also, moderate
Men cohort; Sweden) [59] and high adherence had lower mortality risk by
25% and 23%, respectively. Phosphate intake
and net endogenous acid production were
progressively lower across increasingly
adherent groups
Lin et al. (2011) 3121 women, mean age 67 years; The Western pattern score was directly
Evaluate the effect of healthier 54% hypertensive and 23% diabetic associated with microalbuminuria by 117%
eating patterns vs. the Western subjects; microalbuminuria or eGFR and rapid eGFR decline of ≥3 mL/
dietary pattern (Nurses’ Health decline; 11 years of follow-up min/1.73 m2 per year by 77% (high vs. low
Study; US) [60] (multivariate adjusted) quartile; multivariate adjusted). The DASH
score decreased risk for rapid eGFR decline
by 45%, but had no association with
microalbuminuria (high vs. low quartile;
multivariate adjusted; Fig. 15.8). The general
healthy dietary pattern was not associated
with microalbuminuria or eGFR decline
Intervention trials
Meta-analysis
Oyabu et al. (2016) 9 RCTs, 1687 participants; 46% The mean eGFR in the low carbohydrate diet
Determine the effect of a male; 4 studies in diabetic patients; group was improved compared to the higher
low-carbohydrate diet (LCD) 861 were fed LCD and 826 were fed carbohydrate control diet by 0.13 mL/
on renal function in the control diet; carbohydrate min/1.73m2
overweight and obese consumption was 4–45% of total
individuals without CKD [61] energy intake; 6–24 months
Representative trials
Tyson et al. (2016) 11 adults with an estimated The DASH diet had no significant effect on
Test the effects of the DASH glomerular filtration rate of potassium and serum bicarbonate was
diet on serum electrolytes and 30–59 mL/min/1.73 m2 and significantly reduced by 2.5 mg/dL at
blood pressure (BP) in adults medication-treated hypertension; 2 weeks, compared to baseline values.
with moderate CVD [62] reduced-sodium, run-in diet for Neither incident hyperkalemia nor new onset
1 week followed by a reduced metabolic acidosis was observed. Clinic BP
sodium, DASH diet for 2 weeks and mean 24-h ambulatory BP was
unchanged. DASH significantly reduced
mean nighttime BP by 5.3 mm Hg
Díaz-López et al. (2012) Prevención con Dieta The 3 dietary approaches were associated
Investigate the effects of Med Mediterránea (PREDIMED) with improved kidney function, with similar
Diets on kidney function sub-RCT: average increases in eGFR, but no changes
(Spain) [63] 785 participants; 55% women; mean in urinary albumin-creatinine ratio after full
age 67 years; diets: a MedDiet adjustment. Both the MedDiet and low-fat
supplemented with extra virgin olive diet are equally beneficial in elderly
oil or mixed nuts, or a control individuals at high cardiovascular risk
low-fat diet; 1-year
15.6.1 Prospective Cohort Studies [53]. Dietary patterns with greater than 3.6 g
sodium/day were associated with a 17% increased
15.6.1.1 Meta-analysis risk of renal dysfunction and adequate potassium
A meta-analysis (7 cohort studies; 15,285 sub- was associated with a 17% reduced risk of renal
jects with CKD) found that CKD patients with a dysfunction [54]. This study also showed that
healthy dietary pattern, generally higher in fruit higher Alternate Healthy Eating Index (AHEI),
and vegetables, fish, legumes, cereals, whole Healthy Eating Index (HEI), MedDiet score, and
grains, and fiber, and lower in red meat, and added DASH diet scores improved renal function by
salt and sugar, had a 27% lower risk of mortality 18–29% compared to lower adherence [54]. The
compared consuming Western diets [51]. US National Health and Nutrition Examination
Survey III (1486 CKD patients; 14.2 years of fol-
15.6.1.2 Prospective Studies low-up) found that a high dietary acid load typical
Nine prospective studies provide strong evidence of the Western dietary pattern doubled the risk of
that healthy dietary patterns reduce and Western end stage renal disease for the highest tertile [55].
diets increase CKD risk, progression and mortal- The Framingham Offspring Study (1802 partici-
ity [52–60]. All variations of higher quality or pants; mean age 59 years; 6.6 years of follow-up)
healthy dietary patterns based on minimizing showed that higher diet quality was associated
meat, salt, added sugar, and heavily processed with a significant 37% reduced odds of renal dys-
foods while emphasizing phytochemical-rich function in fully adjusted models [56]. Higher
whole plant foods have generally similar signifi- diet quality was associated with 31% lower risk of
cant effects on improving renal function [54, 56, rapid eGFR decline. A US study in patients with
57, 59, 60]. In contrast, Western dietary patterns CKD consuming higher intake of plant based
are associated with increased CKD risk and poor diets found reduced mortality by 23% whereas
renal function [52, 53, 55, 57, 58, 60]. Two stud- Western diets increased mortality risk by 51%
ies show that low Dietary Approaches to Stop [57]. The US Coronary Artery Risk Development
Hypertension (DASH) scores increase risk of in Young Adults (CARDIA) (2354 African
CKD by 16% [52] or increase rapid decline in American and Caucasian participants; mean base-
eGFR by 68% in individuals with hypertension line age of 35 years; 15 years of follow- up)
demonstrated that that poor diet quality and obe- high adherence to the Western diet had a 117%
sity was associated with a significant 100% increased risk of microalbuminuria whereas those
increased risk of microalbuminuria, a risk factor who followed healthy diets, especially the DASH
for CKD [58]. The Swedish, Uppsala Longitudinal diet, had no association with microalbuminuria
Study of Adult Men (1110 men; mean baseline after an 11-year follow-up (Fig. 15.6) [60]. Also,
age 70 years; 9.9 years of follow-up) found in women with high Western diet scores had a sig-
older men that the higher adherence to a MedDiet nificantly increased risk of rapid eGFR decline of
lowered the risk of CDK by 23% and risk of mor- ≥3 mL/min/1.73 m2 per year by 77% compared to
tality by 42% [59]. The Nurses’ Health Study a significant risk reduction by 45% in women
(3121 women; mean baseline age 67 years; with the highest DASH scores, or no eGFR
11 years of follow-up) showed that those with decline with a generally healthy diet.
60
50
40
CKD Mortality Risk (%)
30
20
10
0
Healthy Plant Based Diet US Southern Traditional Diet
–10
–20
–30
Fig. 15.6 Association between higher dietary pattern and chronic kidney disease (CKD) mortality risk in older US
adults (baseline age 69 years; 6 years of follow-up) (adapted from [57])
1.2
1
Odds Ratios for CKD
0.8
0.6
0.4
0.2
0
Low Medium High
Mediterranean Diet Adherence
Fig. 15.7 Association between adherence to the Mediterranean diet (MedDiet) and chronic kidney disease (CKD) risk
in older Swedish adults (mean baseline age 70 years; 9.9-years follow-up (p-trend = 0.04) (adapted from [59])
Western Diet General Healthy Diet DASH Diet

2.5
Microalbuminuria Risk 2
1.5
0.5
0
1 2 3 4
Dietary Pattern Adherence (Quintile)
Fig. 15.8 Association between dietary pattern adherence and microalbuminuria risk in older women (mean baseline
age 67 years; after 11 years; p-trend = 0.01) (adapted from [60])
15.6.2 Randomized Controlled Trials full adjustment after 1 year [63]. A 2017
Cochrane Systematic Review (17 RCTs; 1,639
A meta-analysis (9 RCTs; 1687 overweight or subjects with CKD; median 12 months) found
obese non-CKD subjects) showed that low- that healthier dietary patterns improved CKD
carbohydrate dietary patterns improved renal patient quality of life, and lowered BP and total
glomerular function vs. typical or higher carbo- cholesterol compared to usual diet controls [64].
hydrate dietary patterns [61]. A 2016 pilot
intervention trial (11 adults with moderate
CKD and medication-treated hypertension; 15.6.3 Nutritional Guidelines
reduced- sodium, run-in diet for 1 week fol-
lowed by a reduced sodium, DASH diet for A modified version of the DASH diet is available
2 weeks) found that the DASH diet had no sig- for people with CKD with a protein intake of
nificant effect on potassium and that serum 0.6–0.8 g/kg of body weight/day, as well as lower
bicarbonate was significantly reduced by phosphorus (0.8–1.0 g/day) and potassium
2.5 mg/dL at 2 weeks, compared with baseline (2–4 g/day) intake [10, 65]. Protein intake may
[62]. Also, no incident hyperkalemia, new onset be restricted to 0.6 g/kg body weight per day
metabolic acidosis or change in blood pressure when eGFR decreases to <60 mL/min/1.73 m2.
(BP) and mean 24-h ambulatory BP were High-protein diets should be avoided in persons
observed. A 2012 Prevención con Dieta with established later CKD stages who are not
Mediterránea (PREDIMED) RCT (785 partici- receiving dialysis. Adequate fiber intake should
pants with high CVD risk; 55% women; mean be encouraged for CKD patients. No specific lev-
age 67 years; diets: a MedDiet supplemented els of fiber intake are suggested, but the adequate
with extra virgin olive oil or mixed nuts, or a intake level of 14 g/1000 kcal should be a good
control low-fat diet; 1-year) found that target intake level, which is typical of most
MedDiets and low-fat diets were equally effec- healthy diets, including increased fruit and vege-
tive for improving kidney function, with similar table intake which may help to reduce risk of
average increases in eGFR, but no changes in metabolic acidosis and reduce urine albumin and
urinary albumin-creatinine ratio was noted at slow loss of renal function.
Conclusions sources vary in their effect on CKD risk with

The kidney is a highly-vascularized organ, red and processed meat consumption signifi-
which plays a major role in regulating electro- cantly increasing risk whereas higher intake of
lyte concentrations and blood pressure, lipid nuts, legumes, and low-fat dairy products sig-
metabolism, production and utilization of sys- nificantly lower risk. Increased fiber intake trig-
temic glucose, degradation of hormones, and gers a number of physiologic processes in both
excretion of waste metabolites, and is affected the colon microbiota and systemically that sup-
to a large degree by the nutritional quality of port the detoxification of the kidneys, via influ-
the diet. The prevalence of CKD is high (esti- ences on the gut barrier, gastrointestinal
mated 200 million people worldwide) and immune and endocrine responses, nitrogen
steadily increasing, especially in older popula- cycling, and microbial metabolism which alter
tions, and it is associated with increased risk of the physiology and biochemistry of the kidneys
renal cancer, cardiovascular disease, and bone to help re- establish homeostasis. Healthy
disorders and fractures. The Western diet is dietary patterns including fiber-rich whole-
associated with increased renal dysfunction, grains, fruits and vegetables appear to improve
CKD risk and progression to ESRD. The CKD renal function, and decrease metabolic acidosis
patients’ diet should help to slow the rate of compared to poor quality diets low in fruits and
progression of kidney failure, reduce uremic vegetables and high in processed foods and ani-
toxicity, decrease proteinuria, and lower the mal products. High adherence to healthy dietary
risk of secondary complications including car- patterns such as the DASH diet, especially a
diovascular disease, bone disease, and hyper- modified version of the DASH diet for people
tension. Healthy lower dietary energy density with CKD, and the Mediterranean diet
and higher fiber healthy dietary patterns lower (MedDiet) may help to reduce CKD risk, pro-
risk of CKD and its progression. Protein gression to later stages and mortality.
ppendix A: Fifty high fiber whole or minimally processed plant foods

A
ranked by amount of fiber per standard food portion size
Shredded wheat ready-to-eat cereal 1–1 1/4 cup (50-60 g) 5.0–9.0 155–220 3.2–3.7
Appendix A 431

Pumpkin seeds. whole, roasted 1 ounce (about 28 g) 5.3 126 4.5
Orange 1 medium (130 g) 3.1 69 0.5
Banana 1 medium (118 g) 3.1 105 0.9
Dates 1/4 cup (about 38 g) 2.9 104 2.8
Nutr Diet. 2015;115:1861–70
trends. 2015;97–98; Table D1.8
Slavin JL. Position of the American Dietetic Association: Health implications of dietary fiber. J Am Diet Assoc
2008;108:1716–31
accessed 22 Aug 2015
References metabolic syndrome risk in the general population:

analysis using multi-frequency bioimpedance. Kidney
Blood Press Res. 2015;40:223–30.
1. Jha V, Garcia-Garcia G, Iseki K, et al. Chronic kidney
17. Sarathy H, Henriquez G, Abramowitz MK, et al.

disease: global dimension and perspectives. Lancet.
Abdominal obesity, race and chronic kidney disease
2013;382:260–72.
in young adults: results from NHANES 1999-2010.
2. Lozano R, Naghavi M, Foreman K, et al. Global and
PLoS One. 2016;11(5). https://doi.org/10.1371/jour-
regional mortality from 235 causes of death for 20
nal.pone.0153588.
age groups in 1990 and 2010: a systematic analysis
for the Global Burden of Disease Study 2010. Lancet. 18. de Vries AP, Ruggenenti P, Ruan XZ, et al. Fatty
2013;380:2095–128. kidney: emerging role of ectopic lipid in obesity-
3. Ojo A. Addressing the global burden of chronic kid- related renal disease. Lancet Diabetes Endocrinol.
ney disease through clinical and translational research. 2014;2(5):417–26.
Trans Am Clin Climatol Assoc. 2014;125:229–46. 19. Rebholz CM, Anderson CAM, Grams ME, et al.

4. Sharma K. Obesity, oxidative stress, and fibro- Relationship of the American Heart Association’s
sis in chronic kidney disease. Kidney Int Suppl. impact goals (Life’s Simple 7) with risk of chronic
2014;4:113–7. kidney disease: results from the Atherosclerosis Risk
5. Hall ME, de Carmo JM, da Silva AA, et al. Obesity, in Communities (ARIC) Cohort Study. J Am Heart
hypertension, and chronic kidney disease. Int J Assoc. 2016;5:e003192. https://doi.org/10.1161/
Nephrol Renovasc Dis. 2014;7:75–88. JAHA.116.003192.
6. McCullough K, Sharma P, Ali T, et al. Measuring the 20. Ricardo AC, Anderson CA, Yang W, et al. Healthy
population burden of chronic kidney disease: a sys- lifestyle and risk of kidney disease progression, ath-
tematic literature review of the estimated prevalence erosclerotic events, and death in CKD: findings from
of impaired kidney function. Nephrol Dial Transplant. the chronic renal insufficiency cohort (CRIC) study.
2012;27(5):1812–21. Am J Kidney Dis. 2015;65(3):412–24.
7. National Institutes for Diabetes and Digestive and 21. Zha Y, Qian Q. Protein nutrition and malnutrition in
Kidney Diseases. Kidney disease. Kidney disease CKD and ESRD. Forum Nutr. 2017;9:208. https://doi.
statistics for the United States. https://www.niddk. org/10.3390/nu9030208.
nih.gov/health-information/health-statistics/Pages/ 22. Haring B, Selvin E, Liang M, et al. Dietary protein
kidney-disease-statistics-united-states.aspx. Accessed sources and risk for incident chronic kidney disease:
April 2017. results from the Atherosclerosis Risk in Communities
8. American Kidney Fund. End stage renal disease. (ARIC) Study. J Ren Nutr. 2017;27(4):233–42.
2015. http://www.kidneyfund.org/kidney-disease/ https://doi.org/10.1053/j.jrn.2016.11.004.
kidney-failure/end-stage-renal-disease/. Accessed 31 23. Herber-Gast G-CM, Biesbroek S, Verschuren WMM,
May 2015. et al. Association of dietary protein and dairy intakes
9. Said S, Hernandez GT. The link between chronic and change in renal function: results from the
kidney disease and cardiovascular disease. J population- based longitudinal Doetinchem cohort
Nephropathol. 2014;3(3):99–104. study. Am J Clin Nutr. 2016;104:1712–9.
10. Hariharan D, Vellanki K, Kramer H. The Western diet and 24. Rouhani MH, Najafabadi MM, Esmaillzadeh A, et al.
chronic kidney disease. Curr Hypertens Rep. 2015;17:16. Dietary energy density, renal function, and progression
https://doi.org/10.1007/s11906-014-0529-6. of chronic kidney disease. Adv Med. 2016;2675345.
11. Odermatt A. The Western-style diet: a major risk factor https://doi.org/10.1155/2016/2675345.
for impaired kidney function and chronic kidney dis- 25. Díaz-López A, Bulló M, Basora J, et al. Cross-

ease. Am J Physiol Renal Physiol. 2011;301:919–31. sectional associations between macronutrient intake
12. Garofalo C, Borrelli S, Minutolo R, et al. A system- and chronic kidney disease in a population at high car-
atic review and meta-analysis suggest obesity predicts diovascular risk. Clin Nutr. 2013;32:606e612. https://
onset of chronic kidney disease in the general popu- doi.org/10.1016/j.clnu.2012.10.013.
lation. Kidney Int. 2017;91(5):1224–35. https://doi. 26. Evenepoel P, Meijers BK. Dietary fiber and protein:
org/10.1016/k.kint.2016.12.013. nutritional therapy in chronic kidney disease and
13. Garland JS. Elevated body mass index as a risk fac- beyond. Kidney Int. 2012;81:227–9.
tor for chronic kidney disease: current perspectives. 27. Kieffer DA, Martin RJ, Adams SH. Impact of dietary
Diabetes Metab Syndr Obes. 2014;7:347–55. fibers on nutrient management and detoxification
14. Navarroa G, Ardiles L. Association between obe-
organs: gut, liver, and kidneys. Adv Nutr. 2016;7:1111–
sity and chronic renal disease. Rev Med Chil. 21. https://doi.org/10.3945/an.116.013219.
2015;143:77–84. 28. Chiavaroli L, Mirrahimi A, Sievenpiper JL, et al.

15. Foster MC, Hwang S-J, Porter SA, et al. Fatty kid- Dietary fiber effects in chronic kidney disease: a
ney, hypertension, and chronic kidney disease: systematic review and meta-analysis of controlled
The Framingham Heart Study. Hypertension. feeding trials. Eur J Clin Nutr. 2014. https://doi.
2011;58(5):784–90. org/10.1038/ejcn.2014.237.
16. Kang SH, Cho KH, Park JW, et al. Association of 29. Krishnamurthy VMR, Wei G, Baird BC, et al. High
visceral fat area with chronic kidney disease and dietary fiber intake is associated with decreased
References 433
inflammation and all-cause mortality in patients Rev. 2015;(2). https://doi.org/10.1002/14651858.

with chronic kidney disease. Kidney Int. 2012;81: CD010070.pub2.
300–6. 45. Uribarri J, Calvo MS. Dietary phosphorus excess: a
30. Xu H, Huang X, Riserus U, et al. Dietary fiber, kidney risk factor in chronic bone, kidney, and cardiovas-
function, inflammation, and mortality risk. Clin J Am cular disease? Adv Nutr. 2013;4:542–4. https://doi.
Soc Nephrol. 2014;9(12):2104–10. org/10.3945/an.113.004234.
31. Jiao J, Xu J, Zhang W, et al. Effect of dietary fiber 46.
Suki WN, Moore LW. Phosphorus regulation
on circulating C-reactive protein in over-weight in chronic kidney disease. Methodist Debakey
and obese adults: a meta-analysis of randomized Cardiovasc J. 2016;12(4 Suppl):6–9. 10.14797/
controlled trials. Int J Food Sci Nutr. 2015;66(1): mdcj-12-4s1-6.
114–9. 47. Rebholz CM, Tin A, Liu Y, et al. Dietary magne-
32. Shu H-S, Tai Y-Y, Chang K-T, et al. Plasma high- sium and kidney function decline: the healthy aging
sensitivity C-reactive protein level is associated with in neighborhoods of diversity across the life span
impaired estimated glomerular filtration rate in hyper- study. Am J Nephrol. 2016;44(5):381–7. https://doi.
tensives. Acta Cardiol Sin. 2015;31:91–7. org/10.1159/000450861.
33. Rossi M, Johnson DW, Campbell KL. The kidney- 48. Cheungpasitporn W, Thongprayoon C, O’Corragain
gut axis: implications for nutrition care. J Ren Nutr. OA, et al. Associations of sugar-sweetened and arti-
2015;25(5):399–403. ficially sweetened soda with chronic kidney disease:
34. Ramezani A, Raj DS. The gut microbiome, kid-
a systematic review and meta-analysis. Nephrology.
ney disease, and targeted interventions. J Am Soc 2014;19(12):791–7. https://doi.org/10.1111/nep.12343.
Nephrol. 2014;25:657–70. 49. Cheungpasitporn W, Thongprayoon C,
35. Salmean YA, Segal MS, Palil SP, Dahl WJ. Fiber sup- Kittanamongkolchai W, et al. High alcohol consump-
plementation lowers plasma p-cresol in chronic kid- tion and the risk of renal damage: a systematic review
ney disease patients. J Ren Nutr. 2015;25(3):316–20. and meta-analysis. Q J Med. 2015;108:539–48.
36. Rossi M, Johnson DW, Xu H, et al. Dietary protein- https://doi.org/10.1093/qjmed/hcu247.
fiber ratio with circulating levels of indoxyl sulfate 50. Wijampreecha K, Thongprayoon C, Thamcharoen N,
and p-cresyl sulfate in chronic kidney disease patients. et al. Association of coffee consumption and chronic
Nutr Metab Cardiovasc Dis. 2015;25(9):860–5. kidney disease: a meta-analysis. Int J Clin Pract.
37. Williams C, Ronco C, Kotanko P. Whole grains in the 2017;71:e12919. https://doi.org/10.1111/ijcp.12919.
renal diet - is it time to reevaluate their role? Blood 51. Kelly JT, Palmer SC, Wai SN, et al. Healthy dietary
Purif. 2013;36:210–4. patterns and risk of mortality and ESRD in CKD:
38. Gopinath B, Harris DC, Flood VM, et al. Carbohydrate a meta-analysis of cohort studies. Clin J Am Soc
nutrition is associated with the 5-year incidence of Nephrol. 2017;12(2):272–9. https://doi.org/10.2215/
chronic kidney disease. J Nutr. 2011;141:433–9. CJN.06190616.
39. Rouhani MH, Mortazavi Najafabadi M, et al. The 52. Liu Y, Kuczmarski MF, Miller ER 3rd, et al. Dietary
impact of oat (Avena sativa) consumption on biomark- habits and risk of kidney function decline in urban
ers of renal function in patients with chronic kidney population. J Ren Nutr. 2017;27(1):16–25. https://doi.
disease: A parallel randomized clinical trial. Clin Nutr. org/10.1053/j.jm.2016.08.007.
2016. https://doi.org/10.1016/j.clnu.2016.11.022. 53. Rebholz CM, Crews DC, Grams ME, et al. DASH
40. Adeva MM, Souto G. Diet-induced metabolic acido- (Dietary Approaches to Stop Hypertension) diet and
sis. Clin Nutr. 2011;30:416–21. risk of subsequent kidney disease. Am J Kidney
41. Goraya N, Simoni J, Jo C-H, Wesson DE. A com- Dis. 2016;68(6):853–61. https://doi.org/10.1053/j.
parison of treating metabolic acidosis in CKD stage ajkd.2016.05.019.
4 hypertensive kidney disease with fruits and vegeta- 54. Smyth A, Griffin M, Yusuf S, et al. Diet and major renal
bles or sodium bicarbonate. Clin J Am Soc Nephrol. outcomes: a prospective cohort study. The NIH-AARP
2013;8:371–81. Diet and Health Study. J Ren Nutr. 2016;26(5):288–
42. Goraya N, Simoni J, Jo C-H, Wesson DE. Treatment 98. https://doi.org/10.1053/ j.jrn.2016.01.016.
of metabolic acidosis in patients with stage 3 55. Banerjee T, Crews DC, Wesson DE, et al. High

chronic kidney disease with fruits and vegetables dietary acid load predicts ESRD among adults with
or oral bicarbonate reduces urine angiotensinogen CKD. J Am Soc Nephrol. 2015;26:1693–700.
and preserves glomerular filtration rate. Kidney Int. 56. Foster MC, Hwang SJ, Massaro JM, et al. Lifestyle fac-
2014;86(5):1031–8. tors and indices of kidney function in the Framingham
43. Wai SN, Kelly JT, Johnson DW, Campbell KL.
heart study. Am J Nephrol. 2015;41:267–74.
Dietary patterns and clinical outcomes in chronic kid- 57. Gutierrez OM, Muntner P, Rizk DV, et al. Dietary pat-
ney disease; The CKD.QLD Nutrition Study. J Ren terns and risk of death and progression to ESRD in
Nutr. 2016;27(3):175–82. https://doi.org/10.1053/j. individuals with CKD: a cohort study. Am J Kidney
jm.2016.10.005. Dis. 2014;64(2):204–13.
44. McMahon EJ, Campbell KL, Bauer JD, Mudge
5 8. Chang A, Van Horn L, Jacobs DR, et al.

DW. Altered dietary salt intake for people with chronic Lifestyle-r elated factors, obesity, and incident
kidney disease. Cochrane Database of Systematic microalbuminuria: the CARDIA (Coronary
Artery Risk Development in Young Adults) 62. Tyson CC, Lin P-H, Corsino L, et al. Short-term effects
study. Am J Kidney Dis. 2013;62(2):267–75. of the DASH diet in adults with moderate chronic
59. Huang X, Jimenez-Moleon JJ, Lindholm B, et al. kidney disease: a pilot feeding study. Clin Kidney J.
Mediterranean diet, kidney function, and mortal- 2016;9(4):592–8. https://doi.org/10.1093/ckj/sfw046.
ity in men with CKD. Clin J Am Soc Nephrol. 63. Díaz-López A, Bulló M, Martínez-González MA,

2013;8:1548–55. et al. Effects of Mediterranean diets on kidney func-
60. Lin J, Fung TT, Hu FB, Curhan GC. Association of tion: a report from the PREDIMED trial. Am J Kidney
dietary patterns with albuminuria and kidney function Dis. 2012;60(3):380–9.
decline in older white women: a subgroup analysis 64. Palmer SC, Maggo JK, Campbell KL, et al. Dietary
from the Nurses’ Health Study. Am J Kidney Dis. interventions for adults with chronic kidney disease.
2011;57(2):245–54. Cochrane Database of Systematic Rev. 2017; Issue 4.
61. Oyabu C, Hashimoto Y, Fukuda T. Impact of low- Art No.CD011998.
carbohydrate diet on renal function: a meta-analysis 65. Kidney Disease Outcomes Quality Initiative (K/
of over 1000 individuals from nine randomised DOQI). K/DOQI clinical practice guidelines on hyper-
controlled trials. Br J Nutr. 2016;116(4):632–8.
tension and antihypertensive agents in chronic kidney
https://doi.org/10.1017/S00071145516002178. disease. Am J Kidney Dis. 2004;43(5 Suppl 1):1–290.
PMID:15114537.
Dietary Patterns and Stroke Risk
16
Keywords
Ischemic stroke risk • Life’s Simple 7 • Healthy dietary patterns • Western
diet • DASH diet • Mediterranean diet • Healthy Nordic diet • Obesity
• Inactivity
Key Points v ascular function, and blood lipids and lipo-

protein profiles; and contributing to a healthier
• Approximately 75% of stroke risk is attribut- colonic microbiota ecosystem and lower sys-
able to smoking, poor diet quality, low physi- temic inflammation and oxidative stress.
cal activity, and excessive body weight,
which are associated with sub-optimal car-
diometabolic health. 16.1 Introduction
• Cohort studies show that healthy dietary pat-
terns are significantly protective against stroke Stroke is the brain equivalent of a heart attack and
risk by 11–23% whereas the Western dietary the leading cause of neurological functional impair-
pattern increases stroke risk by 5–6%. ment by a vascular cause and includes blockage
• A large Spanish PREDIMED trial found that (ischemic stroke) or rupture of a blood vessel
the Mediterranean diet (MedDiet) signifi- (hemorrhagic stroke). Stroke is a major cause of
cantly reduced stroke risk by 39% in older disability and death worldwide, and changes the
adults with high cardiovascular disease risk lives not only of the stroke victims but also of their
over 4.8 years. families as many stroke victims become dependent
• Prospective studies show that high adherence in their activities of daily living due to significant
to MedDiets, Dietary Approaches to Stop stroke related cognitive and physical disabilities
Hypertension (DASH) and the Healthy Nordic [1–5]. Stroke is the second most common cause of
Diet Index significantly reduces total or isch- death worldwide, accounting for 6.2 million deaths
emic stroke risk. or about 11% of total deaths [1–3]. Forecast projec-
• Potential healthy dietary pattern mechanisms tions estimate that there will be a 20% increase in
for lowering stroke risk are through aiding in stroke prevalence by 2030 compared to 2012 rates,
weight and glycemic control; improving because of the increase in aging populations.

436 16 Dietary Patterns and Stroke Risk
The objective of this chapter is to comprehensively stroke and mortality risk [6, 8, 9]. A National
assess the effects of dietary patterns on stroke Health and Nutrition Examination Survey pro-
events and mortality risk. spective study (420 US adults with previous
stroke experience; mean baseline age ≥65 years;
median 9 years of follow-up) found an inverse
16.2 Modifiable Behavior Risk dose-dependent relationship between the number
Factors of “Life’s Simple 7” metrics met and 10-year
mortality after a stroke (Fig. 16.1) [6]. The
Over 90% of stroke risk is attributable to modifi- Reasons for Geographic and Racial Differences
able risk factors, including 74% due to behav- in Stroke Study (22,914 participants; with data
ioral factors such as smoking, poor diet, and on “Life’s Simple 7” and no previous CVD; mean
low physical activity [5]. The American Heart age 65 years, 42% were black, 58% women; 432
Association developed the “Life’s Simple 7” plan incident strokes; mean 4.9 years of follow-up)
for ideal cardiovascular health and lower stroke found that a 1-point higher “Life’s Simple 7”
risk which includes the following criteria: (1) score was associated with an 8% lower risk of
non-smoking or quit >1 year ago; (2) BMI < stroke (Fig. 16.2) [8]. Those who met ≥4 health
25; (3) blood pressure (BP) <120/80 mmHg; metrics had significantly lower all-cause mortal-
(4) ≥ 150 min/week of physical activity; (5) ity by 49% than those who met 0 to 1. Specifically
healthy dietary pattern (high in fruits and vegeta- adhering to “Life’s Simple 7” goals might be the
bles, fish, fiber-rich whole grains, (6) low intake most effective approach for reducing stroke risk
of sodium, and (7) limiting or avoiding sugar- and extending survival after stroke. Also, the
sweetened beverages [6]. Less than 2% of the US Women’s Health Study (37,634 US women with-
population meets all the criteria for cardiovascu- out stroke at baseline; 17.2 years of follow-up,
lar health with adherence to a healthy dietary pat- 867 total stroke cases) found that higher healthy
tern the most difficult to achieve and maintain lifestyle index measures including non-smoking,
[7, 8]. regular physical activity, healthier BMI, limited
Adherence to “Life’s Simple 7” or similar alcohol consumption, and a healthy diet were sig-
healthy lifestyle criteria has been shown to reduce nificantly associated with reduced stroke risk [9].
60
55
10 Year Mortality Risk (%)
50
45
40
35
30
25
20
0 to 1 2 3 4 >=5
"Life's Simple 7" criteria*
Fig. 16.1 Association between the number of “Life’s Simple 7” criteria adopted and long-term mortality risk after stroke
(p-trend = 0.022; multivariate adjusted) (adapted from [6]). *Non-smoking, regular physical activity, healthy diet, maintain-
ing normal weight, and controlling cholesterol, blood pressure, and blood glucose levels
1.1
Hazard Ratio Total Stroke 0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0 1 2 3 4 5 6+7
"Life's Simple 7" Criteria*
Fig. 16.2 Association between the number of “Life’s Simple 7” criteria met and total stroke risk in 22,914 healthy
subjects (mean age 65 years; 58% women; 4.9 years of follow-up) (adapted from [8]). *Not smoking, regular physical
activity, healthy diet, maintaining normal weight, and controlling cholesterol, blood pressure, and blood glucose levels
Those women in the highest healthy lifestyle vitamins E and C, trace minerals, carotenoids,
adherence category had reduced total stroke risk flavonoids and phytosterols are characterizing
by up to 57% compared to women in the lowest components of healthy dietary patterns [22–28].
adherence group. Even women in the modest Potential mechanisms for healthy dietary pat-
healthy lifestyle index group had significant terns to lower stroke risk are: (1) reduce dietary
decreases in total and ischemic stroke risk. energy density and increase satiety and satia-
tion, which reduces the risk of weight gain or
obesity; (2) enhance insulin sensitivity, which
16.3 Dietary Patterns may improve vascular and endothelial function,
(3) promote healthier LDL-C, HDL-C and tri-
16.3.1 Background glyceride profiles for improved endothelial
health and a slower rate of arterial plaque build-
The type of dietary pattern adhered to has a up, (4) attenuate elevated systemic inflamma-
major effect on body weight, central obesity, tion and LDL-oxidation, and (5) stimulate
blood lipids, blood pressure, and glycemic and healthier colonic microbiota which is associated
insulinemic control, which are important factors with increased fiber fermentation to short chain
associated with stroke risk [2–23]. An overview fatty acids for colon and cardiometabolic health
of characteristics and composition of common [29–49].
dietary patterns including the Western diet,
dietary guideline based diets, Dietary
Approaches to Stop Hypertension (DASH), the 16.3.2 Healthy vs. Western Dietary
Mediterranean diet (MedDiet), lacto-ovo-vege- Patterns
tarian, and healthy vegan dietary patterns are
summarized in Appendix A. Whole (minimally Studies on the effects of healthy vs. Western diets
processed) plant foods rich in stroke protective on stroke risk are summarized in Table 16.1
nutrients and phytochemicals such as fiber, [50–55].
Table 16.1 Summary of healthy vs. western dietary patterns studies in stroke risk
Meta-analyses
Zhang et al. (20105). 21 cohort studies; 1,023,131 Healthy dietary patterns significantly reduced stroke
Investigate the association subjects; age range risk by 23% and a Western pattern insignificantly
between dietary patterns and 20–79 years (multivariate increased stroke risk by 5% (highest to the lowest
stroke risk [50] adjusted) intake)
Rodriguez et al. (2015). 19 cohorts; 610,691 subjects; Healthy dietary patterns reduced risk for CVD by
Assess the association between follow-up of 1–18 years 31%, for CHD by 17% and stroke by14%. Western
dietary patterns and CVD, (multivariate adjusted) dietary patterns increased risk for CVD by 14%,
CHD and stroke risk [51] CHD by 3% and stroke by 5%. All comparisons were
highest to the lowest intake
Li et al. (2015). Elucidate the 13 cohort studies; 338,787 Healthy dietary patterns significantly reduced risk of
potential associations subjects; mean follow-up of all-cause mortality by 24% and CVD by 19%, and
between dietary patterns and 5.7–18 years; 9465 cases of showed an inverse trend for stroke mortality by 11%
the risk of all-cause, CVD all-cause death, 5543 cases (highest vs. lowest intake). However, no significant
and stroke mortality [52] of CVD death and 1918 associations were observed between the Western
cases of stroke death dietary pattern and increased risk of mortality from
(multivariate adjusted) all-cause by 7%, CVD by 1% and stroke by 6%
Stricker et al. (2013). 40,011 men and women; The high adherence to healthy dietary patterns (high
Evaluate different dietary 13 years of follow-up; 588 intake of fish, high fiber foods, raw vegetables and
patterns and CHD and stroke stroke cases (multivariate moderate wine intake) reduced stroke risk by 21%
risk (Dutch cohorts of the adjusted) compared to the Western dietary pattern (high intake
European Prospective of French fries, fast food, low-fiber products, soft
Investigation into Cancer drinks with sugar, and alcoholic drinks)
(EPIC) Study) [53]
Judd et al. (2013). Assess the 30,239 American adults; Higher adherence to the healthy (plant-based pattern)
effect of US dietary patterns aged ≥45 years; median was significantly associated with lower stroke risk by
on stroke risk (REasons for 5.7 years of follow-up; 490 29%. This association was attenuated to 15–25%
Geographic and Racial stroke cases (multivariate after addition of income, education, total energy
Differences in Stroke adjusted) intake, smoking, and sedentary behavior. Participants
(REGARDS) cohort with a higher adherence to the Southern pattern had a
study) [54] significant 39% increased risk of stroke
Fung et al. (2004). 71,768 women; 38–63 years; The Western diet significantly increased risk for total
Investigate the effect of 29% hypertension; 23% stroke by 58% and for ischemic stroke by 56%. The
dietary patterns on stroke risk smokers; 14 years of healthy patterns reduced risk for total stroke by 22%
in women (The Nurses’ follow-up; 791 stroke cases and ischemic stroke by 26% (p = 0.13 for both)
Health Study) [55] (multivariate adjusted) (Fig. 16.3). Healthy dietary patterns were
characterized by higher intakes of fruits, vegetables,
legumes, fish, and whole-grains, whereas the Western
pattern by higher intakes of red and processed meats,
refined grains, and sweets and desserts. All
comparisons highest vs. lowest adherence
16.3.2.1 Prospective Cohort Studies 1,023,131 subjects; age range 20–79 years)

found that higher adherence to healthy dietary
ystematic Reviews and Meta-analyses
S patterns significantly reduced stroke risk by
Three meta-analyses support the protective 23% and higher adherence to a Western pat-
effects of consuming health dietary patterns tern insignificantly increased stroke risk by
and the adverse effects of a Western dietary 5% [50]. A second 2015 meta-analysis (19
pattern on stroke risk or mortality [50–52]. A cohorts; 610,691 subjects; follow-up of
2015 meta- analysis (21 cohort studies; 1–18 years) showed that higher adherence to
Healthy dietary pattern Western dietary pattern

1.8
Relative Risk for Total Stroke 1.6
1.4
1.2
0.8
0.6
1 2 3 4 5
Dietary Pattern Adherence (Quintiles)
Fig. 16.3 Association between level of dietary pattern adherence and total stroke risk in US women (age ranging from
38 to 63 years; 14 years of follow-up) (adapted from [55])
healthy dietary patterns reduced risk for CVD in Stroke (REGARDS) study (30,239
by 31%, for CHD by 17% and stroke by14% Americans; aged ≥45 years; median 5.7 years
whereas higher adherence to Western dietary of follow-up) observed that greater adherence to
patterns increased risk for CVD by 14%, CHD a healthy (plant-based pattern) significantly
by 3% and stroke by 5% [51]. A third 2015 lowered stroke risk by 29% whereas higher
meta-analysis (13 cohort studies; 338,787 adherence to the Southern pattern had a signifi-
subjects; mean follow-up of 6–18 years) indi- cantly increased risk of stroke by 39% [54]. A
cated that higher adherence to healthy dietary 2004 Nurses’ Health Study (71,768 women;
patterns lowered stroke mortality by 11% baseline age 38–63 years; 14 years of follow-
whereas higher adherence to a Western dietary up) observed that high adherence to the Western
pattern increased risk of mortality from stroke diet significantly increased risk for total stroke
by 6% [52]. by 58% and for ischemic stroke by 56% whereas
high adherence to healthy patterns reduced risk
Specific Studies for total stroke by 22% and ischemic stroke by
Three prospective studies provide insights on 26% (Fig. 16.3) [55].
dietary patterns and stroke risk [53–55]. A 2013
Dutch European Prospective Investigation into
Cancer [EPIC] Study (40,011 men and women; 16.3.3 Mediterranean Diet (MedDiet)
13 years of follow-up) observed that high adher-
ence to healthy dietary patterns (high intake of A primary prevention trial and cohort studies on
fish, high fiber foods, raw vegetables and mod- the MedDiet and stroke risk are summarized in
erate wine intake) significantly reduced stroke Table 16.2 [56–63]. Notable beneficial nutrients
risk by 21% compared to the Western diet pat- that are abundant in the MedDiet are monoun-
tern (high intake of French fries, fast food, low- saturated fatty acids, a balanced ratio of omega-6/
fiber products, soft drinks with sugar, and omega-3 essential fatty acids, high amounts of
alcoholic drinks) [53]. A 2013 Southern US fiber, and antioxidants such as vitamins E and C,
REasons for Geographic and Racial Differences and polyphenols.
Table 16.2 Summary of Mediterranean diet (MedDiet) studies in stroke risk

Primary prevention RCT
Estruch et al. (2013). 7447 adults with high CVD risk; The MedDiet significantly reduced the
Investigate the effect of the mean age 67 years; 57% women; incidence of primary major cardiovascular
MedDiet pattern on the two MedDiets supplemented with events (composite of myocardial infarction,
primary prevention of 30 g/day mixed nuts or 1 L/week stroke, and death from cardiovascular
cardiovascular events extra virgin olive oil vs. low fat causes) by 30%. Stroke risk was significantly
([Prevención con Diet diet; median follow-up of lowered by39%, which was the only specific
Mediterránea [PREDIMED]; 4.8 years outcome to reach statistical significance
Spain) [56] (Fig. 16.4)
Meta-analysis
Psaltopoulou et al. (2013). 9 cohorts; 162,092 subjects, 3176 High MedDiet adherence resulted in a 16%
Evaluate the association stroke cases (multivariate lower stroke risk and moderate adherence in
between the MedDiet and adjusted) 4% lower stroke risk
stroke risk [57]
Specific cohort studies
Tsivgoulis et al. (2015). 30,239 US adults; mean baseline High adherence to the MedDiet (5–9 on
Evaluate the effect of age of 65 years; 44% male; 56% MedDiet scale) was associated with lower
adherence to MedDiet on from the stroke-belt region; risk of incident ischemic stroke by 21%
stroke risk (REasons for MedDiet score 0–9; 6.5 years of (p = 0.016) after adjustment for
Geographic and Racial follow-up; 565 stroke cases demographics, vascular risk factors, blood
Differences in Stroke (multivariate adjusted) pressure levels, and antihypertensive
(REGARDS) study; US) [58] medications. When the MedDiet was
evaluated as a continuous variable, a 1-point
increase in MedDiet score was independently
associated with a 5% reduction in the risk of
incident ischemic stroke
Hoevenaar-Blom (2012). 40,011 participants; MedDiet There was an inverse association between
Evaluate the effect of score 0–9; mean 11.8 years of the MedDiet and stroke incidence by 30%
MedDiet score on incidence follow-up; 4881 CVD events (highest compared to lowest adherence) and
of total and specific CVD and (multivariate adjusted) a 2-point increase in MedDiet score reduced
stroke risk (Dutch EPIC stroke risk by 12%
Study) [59]
Misirli et al. (2012). 23,601 participants; age Compared to a low MedDiet score of 0–3
Investigate the effect of the 42% > 55 years; 76% ≥ 25 BMI; points, a high score of 6–9 significantly
traditional MedDiet on 59% women; MedDiet score 0–9; lowered ischemic stroke risk by 46%.
cerebrovascular disease risk in median 10.6 years of follow-up; MedDiets appear to be more effective in
a Mediterranean population 395 stroke incident cases and 196 lowering stroke risk in women than men
(Greek segment of the EPIC stroke deaths (multivariate (Fig. 16.5)
Study) [60] adjusted)
Agnoli et al. (2011). Assess 47,021 Italians adults; mean Higher adherence to the Italian MedDiet
the association between the baseline age 50 years; 68% Index reduced risk for total stroke by 53%
Italian MedDiet Index and women; 37% with hypertension; ischemic stroke by 63% and hemorrhagic
stroke risk (EPIC Italy) [61] 25% smokers; mean 7.9 years of stroke by 49%
follow-up; 178 stroke cases were
diagnosed (multivariate adjusted)
Gardner et al. (2011). 2568 adults; mean age 69 years; The MedDiet score was inversely associated
Examine the MedDiet effects 64% women; mean 9 years of with risk of the composite outcome of
on vascular events (Northern follow-up; 171 ischemic stroke ischemic stroke and myocardial infarction
Manhattan Study; US) [62] cases, 133 myocardial infarctions, (p-trend = 0.04) and with vascular related
314 vascular deaths (multivariate death (p-trend = 0.02). Moderate and high
adjusted) MedDiet scores were marginally associated
with decreased risk of MI. There was no
significant association between MedDiet
scores and risk of ischemic stroke
Fung et al. (2009). Evaluate 74,886 women; age range Women in the top Alternate MedDiet Score
the effects of the MedDiet on 38–63 years; 20 years of quintile were at significantly lower risk for
CHD and stroke incidence follow-up; 2391 incident cases of CHD by 29% and stroke by 13% vs. lowest
and mortality in women CHD, 1763 incident cases of score; CVD mortality by 39% vs. the lowest
(Nurses’ Health Study; US) stroke, and 1077 cardiovascular quintile
[63] disease deaths (multivariate
adjusted)
1
MedDiet vs. Low-Fat Diet
0.9
0.8
0.7
Hazard Ratio
0.6
0.5
0.4
0.3
0.2
0.1
0
Stroke Myocardial infarction CVD mortality
(p =.005) (p =.20) (p =.41)
Fig. 16.4 Association between high adherence to a Mediterranean diet (MedDiet) vs. low-fat diet and stroke, myocar-
dial infarction, and cardiovascular disease (CVD) mortality from the PREDIMED Primary Prevention Trial, multivari-
ate adjusted (adapted from [56])
was associated with reduced risk for stroke by

16.3.3.1 Randomized Controlled 16% whereas moderate adherence only reduced
Trial (RCT) risk by 4% [57].
A multicenter Spanish PREDIMED (Prevención
con Dieta Mediterránea) trial (7500 adults with
high CVD risk; mean age 67 years; 57% women; Specific Studies
diets: a MedDiet supplemented with extra-virgin Six prospective cohort studies consistently
olive oil, a MedDiet supplemented with mixed show that the MedDiet is protective against
nuts, or a control low fat diet; 4.8 years) found that stroke risk [58–63]. A 2015 US Southern
an energy-unrestricted MedDiets significantly regional REGARDS study (30,239 US adults;
reduced stroke risk by 39% (Fig. 16.4) [56]. mean baseline age of 65 years; 6.5 years of fol-
low-up) observed that high adherence to the
16.3.3.2 Prospective Cohort Studies MedDiet (5–9 on MedDiet scale) was associ-
ated with significantly lower ischemic stroke
Meta-analysis risk by 21% [58]. When the MedDiet was eval-
A meta-analysis (nine cohorts; 162,092 subjects) uated as a continuous variable, a 1-point
found that high adherence to the MedDiet pattern increase in MedDiet score was independently
associated with a 5% reduction in ischemic observed that women with high adherence to
stroke risk. A 2012 Dutch EPIC Study (40,011 the Alternate MedDiet Score had significantly
adults; mean 11.8 years of follow-up) found lower stroke risk by 13% and lower CVD mor-
that higher adherence to the MedDiet signifi- tality by 39% [63].
cantly reduced stroke incidence by 30% and a
2-point increase in MedDiet score reduced
stroke risk by 12% [59]. Another 2012 EPIC 16.3.4 DASH Diet
study in Greece (23,601 adults; median
10.6 years of follow-up) showed that higher The DASH diet was designed and clinically vali-
MedDiet adherence scores significantly low- dated to prevent and treat hypertension by empha-
ered ischemic stroke risk by 46% [60]. MedDiets sizing high intake of fruits, vegetables, grains,
appear to be more effective in lowering stroke low-fat dairy products, nuts, chicken, and fish
risk in women than men (Fig. 16.5). A 2011 and low intake of red meat, sweets, and refined
Italian EPIC study (47,021 Italians adults; mean carbohydrates [64–67]. Prospective cohort stud-
baseline age 50 years; mean 7.9 years of fol- ies on the DASH diet and stroke risk are summa-
low-up) observed that higher adherence to the rized in Table 16.3 [68–73]. A meta-analysis (six
Italian MedDiet Index significantly reduced cohort studies; 150,000 participants) found that
risk for total stroke by 53%, ischemic stroke by greater adherence to DASH-like diets was shown
63% and hemorrhagic stroke by 49% [61]. A to significantly protect against stroke, and heart
2011 US Northern Manhattan Study (2568 failure risk by 19% [68]. The pooled data from
adults; mean age 69 years; 64% women; mean the Cohort of Swedish Men and the Swedish
9 years of follow-up) observed that the higher Mammography Cohort studies (74,404 adults;
adherence to the MedDiet was inversely associ- mean age 60 years; 50% had BMI ≥ 25; hyper-
ated with risk of the composite outcome of tension 20%; mean 11.9 years of follow-up; 3896
ischemic stroke and myocardial infarction and ischemic strokes) observed that the modified
with vascular related death [62]. A 2009 US DASH score was significantly inversely associ-
Nurses’ Health Study (74,886 women; age ated with ischemic stroke risk with a reduction of
range 38–63 years; 20 years of follow- up) 14% (highest vs. lowest score quartile) (Fig. 16.6)
1
Incidence Mortality
Hazard Ratio for Total Stroke
0.8
0.6
0.4
0.2
0
Men Women Men Women
Fig. 16.5 Effect per 2-point increase in adherence to Mediterranean diet on total stroke incidence and mortality
Table 16.3 Summary of DASH diet cohort studies in stroke risk

Salehi-Abargouei (2013). 3 cohort studies for stroke; 150,191 The DASH-style diets significantly
Evaluate and quantify the adults; 8–24 years of follow-up decreased stroke risk by 19%
effects of the DASH diet on (multivariate adjusted)
stroke risk [68]
Specific prospective cohort studies
Larsson et al. (2016). Assess 74,404 adults; mean age 60 years; High adherence to the DASH diet was
the effect of adherence to the mean 11.9 years of follow-up; 3896 significantly associated with a reduced
DASH diet and stroke risk ischemic strokes, 560 intracerebral risk of ischemic stroke in generally
(Cohort of Swedish Men and hemorrhages, and 176 subarachnoid healthy adults (p-trend = 0.002), with a
the Swedish Mammography hemorrhages (multivariate adjusted) multivariable relative risk reduction by
Cohort studies) [69] 14% (highest vs. lowest quartile)
(Fig. 16.6)
Struijk et al. (2014). 33,671 Dutch adults; mean age Higher adherence to the DASH diet
Investigate the effects of 49 years; 74% women; 30% significantly lowered risk of CVD by
dietary patterns such as the smokers; 21% hypertension; 14%, CHD by 18% and stroke by13%
DASH diet and risk of CVD, average 12.2 years of follow-up; (highest vs. lowest tertile; or per standard
CHD or stroke (Dutch EPIC 2752 CVD cases including 1630 deviation)
study cohort) [70]. CHD and 527 stroke cases
Chan et al. (2013). Examine 2735 Chinese adults; The DASH diet score was borderline
the effect of the MedDiet and age ≥ 65 years; 51% women; mean significantly (p = 0.068) associated with
DASH diet scores on stroke follow-up of 5.7 years; 156 stroke lower stroke risk of 38% (≥4.5 vs. ≤4
risk in older Chinese adults cases (multivariate adjusted) score) in men but not in women
[71].
Fung et al. (2008). Evaluate 88,517 women; age 34–59 years; The DASH score was significantly
the association between the 24 years of follow-up; 2317 cases of inversely associated with CHD risk by
DASH diet score and CHD and CHD and 1682 cases of stroke, of 24% and stroke risk by 18% for extremes
stroke risk (The Nurses’ Health which 1242 were ischemic and 440 of adherence (Fig. 16.7). Cross-sectional
Study; US) [72] hemorrhagic (multivariate adjusted) analysis showed that the DASH score was
significantly associated with lower
plasma levels of C-reactive protein and
interleukin 6
Folsom et al. (2007). Evaluate 20,993 women; mean baseline age Risk of stroke death was insignificantly
the effect of DASH diet and risk 61 years; mean 16 years of reduced by 18%
of CVD mortality (Iowa follow-up; 236 stroke death cases
Women’s Health Study; US) [73] (multivariate adjusted)
[69]. In an older Chinese cohort (2735 Chinese significantly associated with lower plasma levels
adults; age baseline age ≥65 years; 51% women; of C-reactive protein (CRP) and interleukin 6
mean follow-up of 5.7 years) it was observed that (IL-6).
a higher DASH diet score (≥4.5 vs.≤4 score) was
associated with a 38% lower stroke risk in men,
but not in women [71]. The Nurses’ Health Study 16.3.5 Healthy Nordic Dietary
(88,517 women; age 34–59 years; 24 years of Pattern
follow-up) found that higher adherence to the
DASH-style diet was inversely associated with a The Healthy Nordic diet is based on healthy foods
lower risk of CHD and stroke among middle traditionally eaten in the Nordic countries and
aged women (Fig. 16.7) [72]. A sub-group analy- contains high intakes of fish, apples and pears,
sis of women suggests that the DASH score was cabbages, root vegetables, whole grains from oats,
Men (p-trend =.06) Women (p-trend =.005)

1.1
Relative Risk for Ischemic Stroke

1
0.9
0.8
0.7
1 2 3 4
DASH Diet Score (Quintile)
Fig. 16.6 Association between DASH diet adherence and ischemic stroke risk in men and women (mean baseline age
60 years; mean follow-up of 11.9 years) from the Cohort of Swedish Men and the Swedish Mammography Cohort
Total CHD (p-trend <.001) Total Stroke (p-trend =.002)

1.1
1
Relative Risk
0.9
0.8
0.7
Q-1 Q-2 Q-3 Q-4 Q-5
DASH Diet Score
Fig. 16.7 Association between DASH score on total coronary heart disease (CHD) and stroke risk in mid-life women
followed for 24 years from the Nurses’ Health Study (adapted from [72])
barley and rye, and berries [74]. The vegetable oil subjects) to have beneficial effects on cardiovascu-
used is rapeseed oil (Canola oil), which has been lar risk markers, such as blood pressure, choles-
shown in several RCTs of high-risk populations terol, and adiposity [75]. The 2017 Danish Diet,
(obese, hypercholesterolemic, or hyperlipidemic Cancer and Health cohort (55,338 men and
Total Stroke (p-trend =.004) Ischemic Stroke (p-trend =.017)

Large-Artery Atherosclerosis (p-trend =.023)
1.1
0.9
Hazards Ratio
0.8
0.7
0.6
0.5
0 to 1 2 to 3 4 to 6
Healthy Nordic Diet Index
Fig. 16.8 Association between the Healthy Nordic Food Index score and stroke risk (adapted from [74])
women; baseline age 56 years; 52% women; 11–23% whereas the Western dietary pattern
median 13.5 years of follow-up) found that a increases stroke risk by 5–6%. A large Spanish
higher healthy Nordic diet index score was associ- PREDIMED trial found that the MedDiet sig-
ated with a lower risk of stroke and large-artery nificantly reduced stroke risk by 39% in older
atherosclerosis (Fig. 16.8) [74]. adults with high cardiovascular disease risk over
4.8 years. Prospective studies show that high
Conclusions adherence to MedDiets, DASH diets and the
Stroke is the second most common cause of Healthy Nordic Diet Index significantly reduces
death worldwide and forecast projections esti- total or ischemic stroke risk. Med Diets and
mate that there will be a 20% increase in stroke DASH diets tend to be more effective at lower-
prevalence by 2030 compared to 2012 rates, ing stroke risk in women than men. Potential
because of the increase in aging populations. healthy dietary pattern mechanisms for lowering
Approximately 75% of stroke risk is attributable stroke risk are through aiding in weight and gly-
to smoking, poor diet quality, low physical activ- cemic control; improving vascular function, and
ity, and excessive body weight, which are asso- blood lipids and lipoprotein profiles; and con-
ciated with sub-optimal cardiometabolic health. tributing to a healthier colonic microbiota eco-
Cohort studies show that healthy dietary p atterns system and lower systemic inflammation and
are significantly protective against stroke risk by oxidative stress.

A
Healthy
Vegetarian
Western Healthy pattern
dietary USDA Base DASH Diet Mediterranean (Lact-ovo Vegan
Components pattern (US) pattern pattern pattern based) pattern
Emphasizes Refined Vegetables, Potassium rich Whole grains, Vegetables, Plant foods:
grains, low fruit, whole- vegetables, vegetables, fruit, whole- vegetables,
fiber foods, grain, and fruits, and low fruit, dairy grains, legumes, fruits, whole
red meats, low-fat milk fat milk products, olive nuts, seeds, grains, nuts,
sweets, and products oil, and milk products, seeds, and
solid fats moderate wine and soy foods soy foods
Includes Processed Enriched Whole-grain, Fish, nuts, Eggs, non-dairy Non-dairy
meats, sugar grains, lean poultry, fish, seeds, and milk milk
sweetened meat, fish, nuts and seeds pulses alternatives, and alternatives
beverages, nuts, seeds, and vegetable oils
and fast foods vegetable oils
vegetables, added sugars sweets, and refined grains, meats, or fish; products
and sugar-sweetened and sweets limited sweets
whole-grains beverages
(% total kcal)
Protein (% total 16 17 18 16 14 13
kcal)
Total fat (% 33 32 27 34 32 30
total kcal)
(% total kcal)
Unsat. fat (% 22 25 21 24 26 25
total kcal)
Fiber (g) 16 31 29+ 31 35+ 40+
Potassium (mg) 2800 3350 4400 3350 3300 3650
Vegetable oils 19 27 25 27 19–27 18–27
(g)
Sodium (mg) 3600 1790 1100 1690 1400 1225
Added sugar 79 (20 tsp) 32 (8 tsp) 12 (3 tsp) 32 (8 tsp) 32 (8 tsp) 32 (8 tsp)
(g)
References 447
Healthy
Vegetarian
Western Healthy pattern
dietary USDA Base DASH Diet Mediterranean (Lact-ovo Vegan
Components pattern (US) pattern pattern pattern based) pattern
Plant food groups
Fruit (cup) ≤1.0 2.0 2.5 2.5 2.0 2.0
Vegetables ≤1.5 2.5 2.1 2.5 2.5 2.5
(cup)
Whole-grains 0.5 3.0 4.0 3.0 3.0 3.0
(oz.)
Legumes (oz.) − 1.5 0.5 1.5 3.0 3.0+
Nuts/seeds 0.5 0.6 1.0 0.6 1.0 2.0
(oz.)
Soy products 0.0 0.5 − − 1.1 1.5
(oz.)
gov/accessed 8.22.2015
U.S. Department of Agriculture, Agriculture Research Service, Nutrient Data Laboratory. 2014. USDA National Nutrient
Database for Standard Reference, Release 27. https://www.ars.usda.gov/nutrientdata. Accessed 17 February 2015
terns. 2015; 1–9
7. Younus A, Aneni EC, Spatz ES, et al. A systematic

References review of the prevalence and outcomes of ideal car-
diovascular health in US and non-US populations.
1. Mendis S, Puska P, Norrving B (editors). Global atlas Mayo Clin Proc. 2016;91(5):649–70.
on cardiovascular disease prevention and control. The 8. Kulshreshtha A, Vaccarino V, Judd SE, et al. Life’s
underlying pathology of heart attacks and strokes. Simple 7 and risk of incident stroke the reasons for
Geneva: World Health Organization. 2011. ISBN geographic and racial differences in Stroke study.
978-92-4-156437-3. Stroke. 2013;44:1909–14.
2. Mozaffarian D, Benjamin EJ, Go AS, et al. Heart 9. Rist PM, Buring JE, Kase CS, Kurth T. Healthy life-
disease and stroke statistics-2015 update: a report style and functional outcomes from stroke in women.
from the American Heart Association. Circulation. Am J Med. 2016;129(7):715–724.e2. https://doi.
2015;131:29–322. org/10.1016/j.amjmed.2016.02.002.
3. Meschia JF, Bushnell C, Boden-Albala B, et al. 10. Sipahi I, Swaminathan A, Natesan V, et al. Effect of
Guidelines for the primary prevention of stroke. antihypertensive therapy on incident stroke in cohorts
a statement for healthcare professionals from the with prehypertensive blood pressure levels: a meta-
American Heart Association/American Stroke analysis of randomized controlled trials. Stroke.
Association. Stroke. 2014;45:3754–832. 2012;43:432–40.
4. Goldstein LB, Bushnell CD, Adams RJ, et al. 11. Lackland DT, Roccella EJ, Deutsch A, et al.

Guidelines for the primary prevention of stroke. Factors influencing the decline in stroke mortality:
a guideline for healthcare professionals from the a statement from the American Heart Association/
American Heart Association/American Stroke American Stroke Association. Circulation. 2014;45:
Association. Stroke. 2011;42:517–84. 315–53.
5. Feigin VL, Roth GA, Naghavi M, et al. Global burden 12. Chen X, Zhou L, Zhang Y, et al. Risk factors of stroke
of stroke and risk factors in 188 countries, during 1990- in Western and Asian countries: a systematic review
2013: a systematic analysis for the global burden of and meta-analysis of prospective cohort studies. BMC
disease study 2013. Lancet Neurol. 2016;15(9):913– Public Health. 2014;14(776.) http://www.biomedcen-
24. https://doi.org/10.1016/S1474-4422(16)30073-4. tral.com/1471-2458/14/776
6. Lin MP, Ovbiagele B, Markovic D, Towfighi A. 13. Strazzullo P, D’Elia L, Cairella G, et al. Excess body
“Life’s Simple 7” and long-term mortality after weight and incidence of stroke: meta-analysis of pro-
stroke. J Am Heart Assoc. 2015;4(11). https://doi. spective studies with 2 million participants. Stroke.
org/10.1161/JAHA.114.001470. 2010;41:418–26.
14. Neter JE, Stam BE, Kok FJ, et al. Influence of weight 29. Ascherio A, Rimm EB, Hernan MA, et al. Intake of
reduction on blood pressure: a meta-analysis of ran- potassium, magnesium, calcium, and fiber and risk of
domized controlled trials. Hypertension. 2003;42: stroke among US men. Circulation. 1998;98:1198–204.
878–84. 30. Crump C, Sundquist J, Winkleby MA, Sundquist

15. Lee CD, Folsom AR, Blair SN. Physical activity and K. Interactive effects of physical fitness and body mass
stroke risk: a meta-analysis. Stroke. 2003;34:2475–81. index on risk of stroke: a national cohort study. Int J
16. Wendel-Vos GC, Schuit AJ, Feskens EJ, et al. Physical Stroke. 2016;11(6):683–94. pii: 1747493016641961
activity and stroke: a meta-analysis of observational 31. Evans CEL, Greenwood DC, Threapleton DE, et al.
data. Int J Epidemiol. 2004;33:787–98. Effects of dietary fibre type on blood pressure: a
17. Hopper I, Billah B, Skiba M, Krum H. Prevention systematic review and meta-analysis of randomized
of diabetes and reduction in major cardiovascular controlled trials of healthy individuals. J Hypertens.
events in studies of subjects with prediabetes: meta- 2015;33(5):897–911.
analysis of randomised controlled clinical trials. Eur J 32. Davis JN, Hodges VA, Gillham MB. Normal-weight
Cardiovasc Prev Rehabil. 2011;18:813–23. adults consume more fiber and fruit than their age-
18. Zhang Y, Tuomilehto J, Jousilahti P, et al. Total and and height-matched overweight/obese counterparts.
high-density lipoprotein cholesterol and stroke risk. J Am Diet Assoc. 2006;106:833–40.
Stroke. 2012;43:1768–74. 33. Slavin JL. Dietary fiber and body weight. Nutrition.
19. Varbo A, Nordestgaard BG, Tybjaeg-Hansen A,
2005;21:411–8.
et al. Nonfasting triglycerides, cholesterol, and isch- 34. Fogelholm M, Anderssen S, Gunnarsdottir I, Lahti-

emic stroke in the general population. Ann Neurol. Koski M. Dietary macronutrients and food consumption
2011;69(4):628–34. as determinants of long-term weight change in adult
20. Lu Y, Hajifathalian K, Ezzati M, et al. Metabolic medi- populations: a systematic literature review. Food Nutr
ators of the effects of body-mass index, overweight, Res. 2012;56. https://doi.org/10.3402/fnr.v56i0.19103.
and obesity on coronary heart disease and stroke: a 35. Genser B, Silbernage G, De Backer G, et al. Plant ste-
pooled analysis of 97 prospective cohorts with 1.8 rols and cardiovascular disease: a systematic review
million participants. Lancet. 2014;383(9921):970–83. and meta-analysis. Eur Heart J. 2012;33:444–51.
21. Peters SA, Huxley RR, Woodward M. Smoking as a 36. Lau C, Faerch K, Glumer C, et al. Dietary glyce-
risk factor for stroke in women compared with men: mic index, glycemic load, fiber, simple sugars, and
a systematic review and meta-analysis of 81 cohorts, insulin resistance: the Inter99 study. Diabetes Care.
including 3,980,359 individuals and 42,401 strokes. 2005;28:1397–403.
Stroke. 2013;44(10):2821–8. 37. Tosh SM. Review of human studies investigating

22. Kontogianni MD, Panagiotakos DB. Dietary patterns the post-prandial blood-glucose lowering ability
and stroke: A systematic review and re-meta-analysis. of oat and barley food products. Eur J Clin Nutr.
Maturitas. 2014;79:41–7. 2013;67:310–7.
23. Sherzai A, Heim LT, Boothby C, Sherzai AD. Stroke, 38. Silva FM, Kramer CK, de Almeida JC, et al. Fiber intake
food groups, and dietary patterns: a systematic review. and glycemic control in patients with type 2 diabetes mel-
Nutr Rev. 2012;70(8):423–35. litus: a systematic review with meta-analysis of random-
24. Dietary Guidelines Advisory Committee. Scientific ized controlled trials. Nutr Rev. 2013;71(12):790–801.
Report. Advisory Report to the Secretary of Health 39. Sen A, Marsche G, Freudenberger P, et al. Association
and Human Services and the Secretary of Agriculture. between higher plasma lutein, zeaxanthin, and vita-
Part D. Chapter 1: Food and nutrient intakes, and min C concentrations and longer telomere length:
health: current status and trends 2015; 1–78. results of the Austrian Stroke Prevention Study. J Am
25. Dietary Guidelines Advisory Committee. Scientific Geriatr Soc. 2014;62:222–9.
Report. Advisory Report to the Secretary of Health 40. Winek K, Meisel A, Dirnagl U. Gut microbiota impact
and Human Services and the Secretary of Agriculture. on stroke outcome: fad or fact? J Cereb Blood Flow
Appendix E-37: Developing vegetarian and Metab. 2016;36(5):891–8. https://doi.org/10.1177/02
Mediterranean-style food patterns 2015; 1–9. 71678X16636890.
26. USDA. What We Eat in America, NHANES 2011– 41. Whitehead A, Beck EJ, Tosh S, Wolever TMS.

2012, individuals 2 years and over (excluding breast- Cholesterol-lowering effects of oat β-glucan: a meta-
fed children). Available: www.ars.usda.gov/nea/ analysis of randomized controlled trials. Am J Clin
bhnrc/fsrg. Nutr. 2014;100:1413–21.
27. U.S. Department of Agriculture, Agriculture Research 42. Leermaker ET, Darweesh SK, Baena CP, et al. The
Service, Nutrient Data Laboratory. 2014. USDA effects of lutein on cardiometabolic health across the
National Nutrient Database for Standard Reference, life course: a systematic review and meta-analysis.
Release 27. http://www.ars. usda.gov/nutrientdata. Am J Clin Nutr. 2016;103(2):481–94.
Accessed 17 Feb 2015. 43. Berni Canani R, Di Costanzo M, Leone L, et al.

28. U.S. Department of Agriculture and U.S. Department Potential beneficial effects of butyrate in intestinal
of Health and Human Services. Dietary Guidelines and extraintestinal diseases. World J Gastroenterol.
for Americans, 2010. 7th edition, Washington, DC: 2011;17(12):1519–28.
U.S. Government Printing Office. 2010; Table B2.4; 44. Zhou Y, Han W, Gong D, Fan Y. Hs-CRP in stroke: a
http://www.choosemyplate.gov/accessed 8.22.2015. meta-analysis. Clin Chim Acta. 2016;435:21–7.
References 449
45. McCall DO, McGartland CP, McKinley MC, et al. ease in a Mediterranean population. Am J Epidemiol.
Dietary intake of fruits and vegetables improves 2012;176(12):1185–92.
microvascular function in hypertensive subjects in 61. Agnoli C, Krogh V, Grioni S, et al. A priori-defined
a dose-dependent manner. Circulation. 2009;119: dietary patterns are associated with reduced risk of
2153–60. stroke in a large Italian cohort. J Nutr. 2011;141:
46. Noad RL, Rooney C, McCall D, et al. Beneficial effect 1552–8.
of a polyphenol-rich diet on cardiovascular risk: a ran- 62. Gardener H, et al. Mediterranean-style diet and risk of
domised control trial. Heart. 2016;102(17):1371–9. ischemic stroke, myocardial infarction, and vascular
https://doi.org/10.1136/heartjnl-2015-309218. death: the Northern Manhattan Study. Am J Clin Nutr.
47. Jiao J, JY X, Zhang W, et al. Effect of dietary fiber on 2011;94(6):1458–64.
circulating C-reactive protein in overweight and obese 63. Fung TT, Rexrode KM, Mantzoros CS, et al.

adults: a meta-analysis of randomized controlled tri- Mediterranean diet and incidence of and mortality
als. Int J Food Sci Nutr. 2015;66(1):114–9. from coronary heart disease and stroke in women.
48. North CJ, Venter CS, Jerling JC. The effects of dietary Circulation. 2009;119:1093–100.
fibre on C reactive protein, an inflammation marker 64. Appel LJ, Moore TJ, Obarzanek E, et al. A clinical
predicting cardiovascular disease. Eur J Clin Nutr. trial of the effects of dietary patterns on blood pres-
2009;63:921–33. sure. N Engl J Med. 1997;336:1117–24.
49. Maukonen J, Saarela M. Human gut microbiota: does 65. Appel LJ. ASH position paper: dietary approaches
diet matter? Proc Nutr Soc. 2015;74:23–36. to lower blood pressure. J Clin Hypertension.
50. Zhang X, Shu L, Si C, et al. Dietary patterns and 2009;11(7):358–68.
risk of stroke in adults: a systematic review and 66. Moore TJ, Conlin PR, Ard J, Svetkey LP. DASH

meta-analysis of prospective cohort studies. J Stroke (Dietary Approaches to Stop Hypertension) diet is
Cerebrovascular Dis. 2015;24(10):2173–82. effective treatment for stage 1 isolated systolic hyper-
51. Rodríguez-Monforte M, Flores-Mateo G, Sánchez
tension. Hypertension. 2001;38:155–8.
E. Dietary patterns and CVD: a systematic review 67. Sacks FM, Obarzanek E, Windhauser MM, et al.
and meta-analysis of observational studies. Br J Nutr. Rationale and design of the Dietary Approaches
2015;114:1341–59. to Stop Hypertension (DASH) trial. A multicenter
52. Li F, Hou L, Chen W, et al. Associations of dietary controlled-feeding study of dietary patterns to lower
patterns with the risk of all-cause, CVD and stroke blood pressure. Ann Epidemiol. 1995;5:108–18.
mortality: a meta-analysis of prospective cohort stud- 68.
Salehi-Abargouei A, Maghsoudi Z, Shirani F,
ies. Br J Nutr. 2015;13:16–24. Azadbakht L. Effects of Dietary Approaches to Stop
53. Stricker MD, Onland-Moret NC, Boer JM, et al.
Hypertension (DASH)-style diet on fatal or nonfa-
Dietary patterns derived from principal component tal cardiovascular diseases-incidence: a systematic
and k-means cluster analysis: long-term association review and meta-analysis on observational prospec-
with coronary heart disease and stroke. Nutr Metab tive studies. Nutrition. 2013;29:611–8.
Cardiovasc Dis. 2013;23(3):250–6. 69. Larsson SC, Wallin A, Wolk A. Dietary Approaches to
54. Judd SE, Gutiérrez OM, Newby PK, et al. Dietary Stop Hypertension diet and incidence of stroke results
patterns are associated with incident stroke and con- from 2 prospective cohorts. Stroke. 2016;47(4):986–90.
tribute to excess risk of stroke in black Americans. https://doi.org/10.1161/STROKEAHA. 116.012675.
Stroke. 2013;44:3305–11. 70. Struijk EA, May AM, Wezenbeek N, et al. Adherence
55. Fung TT, Stampfer MJ, Manson JE, et al. Prospective to dietary guidelines and cardiovascular disease risk in
study of major dietary patterns and stroke risk in the EPIC-NL cohort. Int J Cardiol. 2014;176(2):354–9.
women. Stroke. 2004;35:2014–9. 71. Chan R, Chan D, Woo J. The association of a priori
56. Estruch R, Ros E, Salas-Salvado J, for the PREDIMED and posterior dietary patterns with the risk of incident
Study Investigators, et al. Primary prevention of car- stroke in Chinese older people in Hong Kong. J Nutr
diovascular disease with a Mediterranean diet. N Engl Health Aging. 2013;17:866–74.
J Med. 2013;368:1279–90. 72. Fung TT, Chiuve SE, McCullough ML, et al.

57. Psaltopoulou T, Sergentanis TN, Panagiotakos DB, Adherence to a DASH-style diet and risk of coronary
et al. Mediterranean diet, stroke, cognitive impair- heart disease and stroke in women. Arch Intern Med.
ment, and depression: a meta-analysis. Ann Neurol. 2008;168(7):713–20.
2013;74:580–91. 73. Folsom AR, Parker ED, Harnack LJ. Degree of con-
58. Tsivgoulis G, Psaltopoulou T, Wadley VG, et al.
cordance with DASH diet guidelines and incidence of
Adherence to a Mediterranean diet and prediction of hypertension and fatal cardiovascular disease. Am J
incident stroke. Stroke. 2015;46:780–5. Hypertens. 2007;20:225–32.
59. Hoevenaar-Blom MP, Nooyens ACJ, Kromhout D, 74. Hansen CP, Overvad K, Kyrø C, et al. Adherence
et al. Mediterranean style diet and 12-year incidence to a healthy Nordic diet and risk of stroke a Danish
of cardiovascular diseases: the EPIC-NL cohort study. cohort study. Stroke. 2017;48:259–64. https://doi.
PLoS One. 2012;7(9). https://doi.org/10.1371/jour- org/10.1161/STROKEAHA.116.015019.
nal.pone.0045458. 75. Lin L, Allemekinders H, Dansby A, et al. Evidence of
60. Misirli G, Benetou V, Lagiou P, et al. Relation of the health benefits of canola oil. Nutr Rev. 2013;71(6):370–
traditional Mediterranean diet to cerebrovascular dis- 85. https://doi.org/10.1111/nure.12033.
Whole Plant Foods and Stroke Risk
17
Keywords
Stroke • Ischemic stroke • Hemorrhagic stroke • Whole plant foods
Whole-grains • Fruits • Vegetables • Green leafy vegetables • Apples
Legumes • Nuts • Peanuts • Tea • Coffee
Key Points vegetables, white fruits and vegetables (e.g.,

• Stroke and related cerebrovascular diseases apples, cauliflower, mushrooms) and vitamin
make-up the second most common cause of C rich fruits and vegetables (e.g., citrus, bell
death worldwide, accounting for 6.2 million peppers, broccoli).
deaths (11% of total deaths). Regular con- • Plant protein sources such as legumes and
sumption of specific foods and beverages may nuts have mixed effects on stroke risk. Dietary
have significant effects on stroke risk with pulses or total legumes are not associated with
potential stroke protective foods including stroke risk but nuts and soy foods are associ-
low fat dairy, whole grains, fruits, vegetables, ated with lower stroke risk.
nuts, tea and coffee whereas high intake of red • Whole plant foods containing a variety of
meat, sugar-sweetened beverages and alcohol nutrients and phytochemicals such as fiber,
may increase stroke risk. antioxidant vitamins, potassium, magne-
• The consumption of ≥3 daily servings of sium, carotenoids, flavonoids and phytoster-
whole grains is associated with lower total ols may provide potential stroke protection
stroke risk by 8–14% and ischemic stroke by by mechanisms associated with promoting
25% compared to never or rare intake. vascular health by attenuating elevated
• Increased consumption by 200 g (or 2 1/2 blood pressure, lowering LDL-cholesterol
servings)/day reduces total stroke for total levels and systemic inflammation associated
fruits and vegetables by 16%, fruits by 18% with atherosclerosis, and promoting better
and vegetables by 13%. Raw fruits and vegeta- insulin sensitivity, blood glucose control,
bles are more effective than processed forms; weight control, and microbiota health com-
examples of effective specific varieties for pared to less healthy or highly processed
reducing ischemic stroke risk are green leafy plant foods.

452 17 Whole Plant Foods and Stroke Risk
17.1 Introduction Table 17.1 [13–37]. The primary objective of this

chapter is to comprehensively review the effects of
Strokes are caused by a disruption of the blood whole and minimally processed plant foods
supply to the brain due to either blockage (isch- (whole plant foods) on stroke risk and prevention.
emic stroke) or rupture of a blood vessel (hemor-
rhagic stroke) [1]. Of all strokes, 87% are ischemic
[2]. Stroke and related cerebrovascular diseases 17.2 Whole Plant Foods
make-up the second most common cause of death
worldwide, accounting for 6.2 million deaths Whole plant foods contain a variety of nutrients
(11% of total deaths) [2]. Over 90% of stroke risk and phytochemicals that may be associated with
can be attributed to modifiable risk factors, includ- lower stroke risk, such as fiber, vitamins E and C,
ing 74% due to behavioral factors (smoking, poor trace minerals, carotenoids, flavonoids and phy-
diet, and low physical activity) and their associa- tosterols (Appendix A) [38–46]. Potential mecha-
tion with metabolic factors including high systolic nisms by which whole plant foods amplify support
blood pressure (BP), high BMI, high fasting for reduced stroke risk in contrast to the effects of
plasma glucose, high total cholesterol, and low highly processed refined foods by: (1) reducing
glomerular filtration rate [2–12]. Stroke is a major dietary energy density and increasing satiety and
cause of disability and death worldwide, and satiation, which reduces the risk of weight gain or
changes the lives not only of the stroke victims but obesity; (2) enhancing insulin sensitivity, which
also of their families as many become dependent may improve vascular and endothelial functions,
in their activities of daily living due to significant (3) promoting healthier LDL-C, HDL-C and tri-
stroke related cognitive and physical deficits [1– glyceride profiles for improved endothelial health
5]. Forecasts suggest a >20% increase in stroke and a slower rate of arterial plaque build-up, (4)
prevalence by 2030 compared to 2012 rates, attenuating elevated systemic inflammation and
because of the increase in aging populations [2– LDL-oxidation, and (5) stimulating healthier
4]. Stroke carries a high risk of death and survi- microbiota and increased fermentation to short
vors can experience loss of vision and/or speech, chain fatty acids (SCFAs) such as butyrate for
paralysis, and confusion. Intervention trials and colon and cardiometabolic health and body
observation studies suggest that lifestyle habits are weight control [47–69]. However, whole plant
the primary factors associated with stroke risk. foods are consumed at low levels in the typical
The type and amount of foods and beverages con- Western diet with an estimated >90% of US adults
sumed can have an important effect on stroke risk not meeting the recommended minimal levels of
and prevention. An overview of commonly con- whole-grains, fruits, vegetables, legumes, or nuts
sumed foods and beverages that may have poten- and seeds to maintain optimal health and weight
tial effects on stroke risk are summarized in control outcomes [47–49].
Table 17.1 Overview of foods and beverages with potential effects on stroke risk
Increase stroke risk Little or No effect on risk Decrease stroke risk
Total red meat >50 g/day and Moderate whole-fat dairy, butter Total dairy, low fat diary and cheese ≥2
fresh red meat >70 g/day or cream [20, 21] servings daily [22, 23]
[26–28] Refined grains (≤ 3 servings/ Extra virgin olive oil 1 L/week [19]
Sugar sweetened beverages day) [29, 30] Dark chocolate >50 g/week [18, 19].
≥2 servings/day [35–37] Lean, fatty or total fish [33, 34] Fruits and vegetables ≥200 g/day [31, 32]
Heavy alcohol intake [13, 14] Moderate alcohol intake [13, 14] Whole-grains ≥3 servings [29, 30]
Nuts ≥5 days of the week [24, 25]
Tea ≥3 cups/day; green or black [17]
Coffee 3–5 cups/day [15, 16]
Low alcohol intake ≤1 drink (10 g ethanol)/
day [13, 14]
17.2.1 Whole-Grains intake on stroke risk [76–80]. Two Iowa Women’s

Health Studies in post-menopausal women
17.2.1.1 Background (approx. 30,000 women; mean baseline age
Whole-grain products (brown rice, oatmeal, 61 years; 9–17 years of follow-up) observed that
whole oats, popcorn, whole rye, and graham and ≥2.7 servings/day of whole-grains had a trend for
whole wheat flour) consist of the starchy endo- reduced stroke mortality risk by 13–15% and
sperm, germ and bran found in similar propor- similar increased intake of refined grain intake
tions to that of the whole intact kernel with fiber, showed an increased trend for stroke mortality
vitamins, minerals and phytochemicals [70, 71]. risk by 30–33% [76, 80]. A 2005 Nurses’ Health
In contrast, refined grain products (white rice and Study (78,770 women;18 years of follow-up)
flour, and white bread, pastry, and low fiber showed that whole-grain cereal fiber was
breakfast cereals) are mainly comprised of the inversely associated with stroke risk outcomes
starchy endosperm with most of the fiber, vita- [77]. A 2000 Nurses’ Health Study (75,521 US
mins, minerals and phytonutrients removed dur- women; mean baseline age 50 years;12 years of
ing processing. The US dietary guidelines follow-up) observed that increased intake of
recommend ≥3 servings/whole-grains/day and whole-grains was inversely associated with isch-
≤3 servings of refined grains/day to promote emic stroke risk in women and increased refined
health and wellness associated with reduced risk grain intake positively associated with increased
of various chronic diseases [48, 49]. However, risk (Fig. 17.1) [29]. The Atherosclerosis Risk in
only about 1% of Americans follow the recom- Communities cohort (15,792 adults; mean base-
mendation for whole-grain intake as the average line age 53 years; 11 years of follow-up) showed
American’s intake is <1 ounce whole grains/day that ≥3 servings/day of whole-grains reduced
and 70% exceed the recommended intake for ischemic stroke risk by 25%, whereas 3–5 serv-
refined grains [48, 72]. ings of refined-grains did not significantly
increase ischemic stroke risk [78].
17.2.1.2 Prospective Cohort Studies
A summary of prospective cohort studies exam-
ining the effects of whole-grain and refined grains 17.2.2 Fruits and Vegetables
on stroke risk is provided in Table 17.2
[30, 73–80]. 17.2.2.1 Background
Adequate intake of fruits and vegetables is an
Systematic Reviews and Meta-analyses important component of most global dietary
Four meta-analyses provide insights on the guidance recommendations for health and BP
effects of increased intake of whole-grains and control because of their concentrations of: anti-
stroke risk [30, 73–75]. Three analyses conclude oxidant vitamins and phytochemicals, especially
that the consumption of ≥3 daily servings of vitamins C and A, and carotenoids; minerals
whole grains is associated with lower total stroke (especially electrolytes, high potassium and mag-
risk by 8–14% and ischemic stroke by 25% vs. nesium, and low sodium); and dietary fiber and
infrequent intake [30, 73, 74]. The effects of lower energy density; but their composition is
whole-grain intake on stroke risk was non-linear highly variable leading to diverse BP and hyper-
with no additional reduction in risk above 120– tension effects [38, 41]. The World Health
150 g/day (4–5 servings) [73]. The increased Organization (WHO) report recommended a
intake of refined grains was not significantly minimum daily intake of 400 g of fruit and veg-
associated with stroke risk [75]. etables in 1990s, based on emerging evidence
that higher levels are protective against cardio-
Specific Prospective Cohort Studies vascular disease (CVD) [81]. This led to the
Five prospective cohort studies provide specific launch of various “eat 5 or more” fruit and vege-
insights into the effect of increased whole-grain table campaigns in Europe, the US and Australia
Table 17.2 Summary of prospective cohort studies on whole- and refined-grain intake in stroke risk
Aune et al. (2016). 6 cohort studies; 245,012 For high vs. low intake of whole
Quantify the dose-response relation participants; 2337 stroke cases. grains, there was a 13% lower
between intake of whole grain and stroke risk; and a 12% lower risk
stroke risk [73]. per three servings (90 g/day).
There was evidence of non-
linearity between whole grain and
risk of stroke (p < 0.001) with no
further reduction in risk above
120–150 g/day
Chen et al. (2016). 7 cohort studies; 446,451 Total stroke risk was reduced for
Investigate the association of whole participants; 5892 stroke events). total grains by 5%, for whole grain
and refined grain and stroke risk [30]. by 8%, and for refined grains by
1% (highest vs. lowest intake).
Whole grains were inversely
associated with ischemic stroke
risk by 25%. Refined grain was not
associated with total stroke risk.
Fang et al. (2015). 6 cohort studies; 247,487 subjects; Whole-grains were significantly
Examine the evidence for the effects mean age range 38–74 years; 1635 associated with reduced risk of
of whole-grains on stroke risk [74]. stroke cases, mean follow-up of stroke by 14% (≥2.7 servings vs.
5.5–24 years; 5 studies from US and rare intake).
1 study from Finland.
Wu et al. (2015). 8 cohort studies; 410,821 Diets higher in refined-grains had
Evaluate the association between participants; 8284 stroke events. insignificantly increased stroke
refined-grains and stroke risk [75]. risk by 2% in men and women.
Jacobs et al. (2007). 27,312 healthy postmenopausal Compared with women who rarely
Assess the association between women; mean baseline age 61 years, or never ate whole-grain foods,
whole-grain intake and inflammatory mean BMI 27; 17 years of follow-up; women consuming ≥2.7 servings/
disease and stroke mortality risk (US 1071 deaths from inflammatory day whole-grains had a trend for
Iowa Women’s Health Study) [76]. causes (multivariate adjusted). lower stroke mortality risk by15%
and similar intakes of refined
grains showed a trend toward
increased stroke mortality risk by
30%.
Oh et al. (2005). 78,779 women; mean baseline age Carbohydrate intake and glycemic
Assess the association between high 46 years;18-years of follow-up; 1020 load were significantly associated
carbohydrate intake, a high glycemic stroke cases (multivariate adjusted). with elevated risk of hemorrhagic
index diet, and a high glycemic load stroke for extreme quintiles but not
diet and stroke risk (Nurses’ Health with ischemic stroke, especially
Study; US) [77]. among women with a BMI of ≥25.
Cereal fiber intake was inversely
associated with total stroke by
34% and hemorrhagic stroke risk
by 49% at highest vs. lowest
intake.
Steffen et al. (2003). 15,792 adults; mean baseline age Whole-grain intake was inversely
Evaluate the association between 53 years; mean BMI 27; 11 years of associated with ischemic stroke
whole grain intake on risk of total follow-up; 214 ischemic strokes incidence. Subjects consuming ≥3
mortality and the incidence of (multivariate adjusted). whole-grain foods servings / day
coronary artery disease (CAD) and had significantly reduced ischemic
ischemic stroke (Atherosclerosis Risk stroke risk by 25% (p-trend =0.15)
in Communities cohort; US) [78]. Compared to rare or never intake.
Refined grain was not associated
with ischemic stroke incidence.
Liu et al. (2003). 86,190 men; mean baseline age The consumption of ≥1 serving/
Evaluate the association between 56 years; mean follow-up of day of total, whole-grain or refined
whole- and refined-grain breakfast 5.5 years; 146 stroke deaths grain breakfast cereal vs. rarely or
cereal intakes and mortality in men (multivariate adjusted). never consuming breakfast cereal
(The Physicians’ Health Study; US) was insignificantly associated with
[80]. stroke mortality risk.
Liu et al. (2000). 75,521 US women; mean baseline Whole-grain intake was inversely
Investigate the relationship between age 50 years; 12 years of follow-up; associated with ischemic stroke
whole-grain intake and ischemic 352 ischemic stroke incident cases risk (median intake: 2.7 servings/
stroke risk (The Nurses’ Health (multivariate adjusted). day vs. 0.13 servings/day) with
Study) [29]. risk reductions by 31%
(p-trend = 0.08) whereas refined
grain insignificantly reduced
stroke risk by 3% (Fig. 17.1).
Jacobs et al. (1999). 34,333 healthy postmenopausal Women in the highest category of
Investigate the association between women; mean baseline age whole-grain cereal intake (≥3
whole-grains and mortality risk (The 61.5 years; 9 years of follow-up servings/day) had trend for lower
Iowa Women’s Health Study; US) (multivariate adjusted). stroke mortality risk by 13% and
[80]. similar refined grains intake
showed an increased trend for
stroke mortality risk by 33%.
Whole grain (p-trend =.08) Refined grain (p-trend =.58)

Total grain (p-trend =.16)
1.3
1.2
Relative Risk for Ischemic Stroke
1.1
0.9
0.8
0.7
0.6
0.5
0.4
1 2 3 4 5
Intake Quintile
Fig. 17.1 Association between whole, refined and total grain intake and ischemic stroke risk in women from the
Nurses’ Health Study after 12 years (adapted from [29])
[82]. The USDA MyPlate educational concept, ing at any age [83]. However, globally, fruit and
which devotes one-half the plate to fruits and vegetable consumption is only a small fraction of
vegetables, as a displacement of other foods of the recommended levels [84]. In the US, >85% of
higher energy density from the diet, is a good the population fall short of meeting the daily fruit
habit to establish for healthy aging or healthy eat- and vegetable intake recommendation [48].
17.2.2.2 Prospective Studies Systemic Reviews and Meta-analyses

Fruit and vegetable intake is consistently shown Four meta-analyses all show that fruits and veg-
to be inversely related to stroke risk with certain etables are inversely associated with stroke risk
specific fruits and vegetables being uniquely [31, 85–87]. A 2017 dose-response meta-analysis
more effective than others (Table 17.3) [31, 32, (43 cohort studies) found that 200 g/day reduced
78, 85–93]. total stroke risk for fruits and vegetables by 16%,
Table 17.3 Summary of fruit and vegetable intake studies in stroke risk
Aune et al. (2017). 43 cohort studies; >2 million Per 200 g fruits and vegetables/day the total stroke
Conduct a systematic participants. risk was reduced by 16%; fruits by 18% and
review and meta- vegetables by 13%. There was evidence of a
analysis to clarify the non-linear association for total stroke, risk was
effects of fruit and reduced for 800 g/day of fruits and vegetables
vegetable intake on combined by 33% and for 200–350 g/day by 20%
stroke risk [85]. whereas 500 g/day of vegetables reduced risk by
28%. For ischemic stroke, risk was reduced per
200 g/day for fruits by 12% and for vegetables by
14%. For hemorrhagic stroke, risk was reduced per
200 g/day for fruits by 34% and vegetables by 24%.
The effects of individual fruits and vegetables on
stroke risk are summarized in Fig. 17.2.
Hu et al. (2014). 20 cohort studies published up to Stroke risk was reduced for total fruits and
Investigate the effect of January 2014; 760,629 subjects; vegetables by 21%, for fruits by 23% and for
fruits and vegetables on mean duration of follow-up was vegetables by 14% (highest vs. lowest intake). The
stroke risk [31]. 3–37 years; 16,981 stroke cases; six linear dose–response relationship showed that the
studies were from the United States, risk of stroke decreased for every 200 g/day
eight from Europe, and six from increment of fruits by 32% and vegetables by 11%
Asia (Japan and China); average (Fig. 17.3). Citrus fruits, apples, pears, and leafy
serving was calculated as 77 g for vegetables appear to be uniquely effective for
vegetables and 80 g for fruits. stroke protection.
He et al. (2006). 9 cohort studies; 257,551 Compared with individuals who consumed ≤3
Evaluate the association participants; age ranged from servings of fruits and vegetables/day, stroke risk
between fruits and 25–103 years; average follow-up of was reduced for 3–5 servings/day by 11% and >5
vegetables and stroke 13 years; 4917 stroke cases. servings/day by 26%. Subgroup analyses showed
risk [86]. that fruits and vegetables had a significant
protective effect on both ischemic and hemorrhagic
stroke. Men and women had similar stroke risk
reduction for >5 servings/day.
Dauchet et al. (2005). 7 cohort studies; 90,513 men, The risk of stroke was significantly decreased by
Assess the effects of 141,536 women; mean follow-up of 11% for each additional serving/day of fruits, by
fruits and vegetables on 3 to 20 years; 2955 stroke cases. 5% for fruit and vegetables, and insignificantly by
stroke risk [87]. 3% for vegetables. The association between fruits
only or fruits and vegetables and stroke was linear,
suggesting a dose-response relationship.
Larsson et al. (2013). 74,961 adults; mean baseline age Total stroke risk for fruit and vegetable intake was
Examine the relationship 60 years; 46% women; 50% significantly reduced by13% (> 6 servings/day vs.
between intake of specific overweight; 25% current smokers; <2.3 servings/day). Sub-group analysis showed
fruit and vegetable 20% with hypertension; 10.2 years that the risk for normotensive individuals was
subgroups and stroke risk of follow-up; 4089 stroke cases significantly reduced by 19%. The reduced total
in a cohort of Swedish (multivariate adjusted). stroke risk per serving for apples and pears was
women and men 11% and for green leafy vegetables was 8%
(Swedish Mammography (Fig. 17.4).
Cohort and the Cohort of
Swedish Men) [32].
Oude Grieo et al. 20,069 adults; mean baseline age Higher intake of white fruits and vegetables was
(2011). 41 years, 45% men; mean BMI 25; inversely associated with stroke risk by 52%
Investigate the effect of 37% smokers; 10 years of (>171 g/day vs. ≤78 g/day). Each 25-g/d increase
fruit and vegetable color follow-up; 233 stroke cases in white fruits and vegetables consumption was
groups and stroke risk (multivariate adjusted). associated with a 9% lower risk of stroke. Apples
(Dutch cohorts of the and pears were the most commonly consumed of
European Prospective the white fruit and vegetables (55%) (Fig. 17.5).
Investigation into Cancer
[EPIC] Study) [88].
Oude Grieo et al. 20,069 adults; mean baseline age High intake of raw fruits and vegetables may
(2011). 41 years, 45% men; mean BMI 25; protect against stroke (median intake: 337 g/day
Evaluate effect of raw or 37% smokers; 10 years of vs. 56 g/day) whereas no significant inverse
processed fruits and follow-up; 233 stroke cases association was observed between processed fruit
vegetables and stroke (multivariate adjusted). and vegetable consumption and stroke risk (median
risk (Dutch cohorts of intake: 301/day vs. 86 g/day) (Fig. 17.6).
the EPIC study) [89].
Mizrahi et al. (2009). 3932 adults; mean baseline age The intake of fruits, especially citrus, and
Study the effect of 54 years; 53% men; mean BMI 27; cruciferous vegetables may protect against
whole plant foods on 30% current smokers; 24% with cerebrovascular diseases. An inverse association
cerebrovascular disease hypertension; 24 years of follow-up; was found between fruit consumption and the
incidence (Finnish 625 cases of cerebrovascular incidence of cerebrovascular diseases by 25%,
Mobile Clinic Health incidence or death (multivariate ischemic stroke by 27% and intracerebral
Examination Survey adjusted). hemorrhage by 53%, which was primarily due to
Follow-up Study) [90]. the consumption of citrus fruits. The consumption
of cruciferous vegetables, reduced risk of ischemic
stroke by 33% and intracerebral hemorrhage by
51%.
Sauvaget et al. (2003). 40,349 Japanese adults; mean Daily consumption of green-yellow vegetables and
Investigate effects of a baseline age 56 years; 38% men; fruits was associated with a lower risk of total
diet rich in fruits and 73% from Hiroshima city; mean stroke, ischemic and stroke mortality. A daily
vegetables on total radiation dose115mSv; 18 years of serving of green-yellow vegetables was associated
stroke mortality follow-up; 1926 stroke deaths with a significant 26% reduction in the risk of death
(Hiroshima/Nagasaki (multivariate adjusted). from total stroke in men and women compared with
Life Span Study) [91]. an intake of ≤1 serving/week. A daily fruit serving
was associated with a significant 35% reduction in
risk of total stroke death in men and a 25%
reduction in women vs. ≤1 serving/week.
Steffen et al. (2003). 15,792 adults; mean baseline age Compared with a mean intake of 1.5 servings of
Evaluate the effect of fruit 53 years; 11 years of follow-up; 270 fruits and vegetables/day, those with a mean intake
and vegetable intakes on fatal or nonfatal incident stroke, of 7.5 servings/day had a 22% lower risk of total
the risk of total mortality with 21 ischemic strokes mortality. There was no association between the
and the incidence of (multivariate adjusted). intake of fruit and vegetables and the risk of
coronary artery disease incident ischemic stroke.
(CAD) and ischemic
stroke (Atherosclerosis
Risk in Communities
cohort; US) [78].
Bazzano et al. (2002). 9608 adults; mean baseline age Consuming fruits and vegetables ≥3 times/day
Investigate the effect of 50 years; 40% men; mean BMI 25; compared with < once/day was associated with a
fruit and vegetable intake mean BP 134/84 mm Hg; 19 years significant 27% lower stroke incidence and a
and CVD and stroke risk of follow-up; 888 stroke cases significant 42% lower stroke mortality.
(National Health and (multivariate adjusted).
Nutrition Examination
Survey Epidemiologic
Follow-up Study; US)
[92].
(continued)
Joshipura et al. (1999). 75,596 women; mean baseline age Pooled data from men and women showed that
Examine the effect of 46 years; 14 years of follow-up; 366 each daily serving of fruit and vegetables was
fruit and vegetable stroke cases; and 38,683 men, mean associated with lower ischemic stroke risk by 6%.
intake on ischemic baseline age 54 years; 8 years of Risk reductions per serving of specific fruit and
stroke risk (Nurses’ follow-up; 204 stroke cases vegetable subtypes are shown in Fig. 17.7.
Health Study and Health (multivariate adjusted).
Professionals’ Follow-up
Study; US) [93].
Per 100g/day Intake High vs Low Intake
Vit C-rich F/V
Root vegetables
Potatoes
Allium vegetables
100% Citrus fruit juices
Citrus fruit
Berries
–40 –35 –30 –25 –20 –15 –10 –5 0 5

Ischemic Stroke Risk (%)
Fig. 17.2 Association between fruit and vegetable (F/V) subtypes and ischemic stroke risk from a 2017 meta-analysis
of 43 cohort studies (adapted from [85])
for fruits by 18% and vegetables by 13% [85]. reduced for fruit and vegetables combined by
There was evidence of a nonlinear association 8%, for fruits by 12% and for vegetables by 14%.
between total fruits and vegetables, fruit, and The effects of individual fruits and vegetable sub-
vegetables, and lower total stroke risk. Total risk types on stroke risk are summarized in Fig. 17.2.
was reduced for 800 g/day of fruits and vegeta- A 2014 meta-analysis (20 cohort studies; 760,629
bles by 33%, for 500 g/day of vegetables by 28% adults; mean follow-up of 3–37 years) found that
and for 200–350 g/day of fruits and vegetables by total stroke risk was reduced for higher total fruits
20%. Per 200 g/day, ischemic stroke risk was and vegetables consumption by 21%, for fruits by
Total fruit intake Total vegetable intake
1
Relative Risk for Total Stroke
0.8
0.6
0.4
0.2
0 50 100 150 200 250 300 400
Intake (g/day)
Fig. 17.3 The dose-response analysis between fruit and vegetable consumption and stroke risk from a meta-analysis of
20 prospective cohort studies (760,629 participants) (adapted from [31])
23% and for vegetables by 14% vs. the lowest lic, leeks and onions) as the most effective at reduc-
intake [31]. The linear dose-response relationship ing stroke risk by 52% (>171 g/day vs. ≤78 g/day)
showed that every daily 200 g (2 1/2 servings) of (Fig. 17.5) [88]. Each 25 g/day increased intake of
fruits reduced total stroke risk by 32% and vege- apples and pears was associated with a 7% lower
tables reduced risk by 11% (Fig. 17.3). Data sug- stroke risk. There was also an inverse association
gests that citrus fruits, apples, pears, and green with stroke for raw fruits and vegetables, with a
leafy vegetables might be uniquely effective for significantly reduced stroke risk by 31% compared
stroke protection. Two other meta-analyses report to an 11% increased risk trend for processed fruits
similar findings on the effects of fruits and vege- and vegetables (Fig. 17.6) [89]. The Hiroshima/
tables on stroke risk [86, 87]. Nagasaki Life Span Study (40,349 adults; mean
baseline age 56 years; 38% men; 18 years of fol-
Specific Prospective Cohort Studies low-up) found that daily fruit intake was associated
Eight cohort studies provide important insights with a reduction in total stroke mortality risk in
regarding increased intake of total and specific men by 35% and in women by 25% [91]. For veg-
fruits and vegetables on stroke risk [32, 78, 88–93]. etables, green-yellow vegetables were the most
A 2013 Swedish Mammography Cohort and the effective for stroke protection, reducing mortality
Cohort of Swedish Men studies (74,961 adults; risk by 26% (daily intake vs. ≤1/week). A 1999
mean baseline age 60 years; 46% women; 50% pooled analysis of Nurses’ Health Study and
overweight; 10.2 years of follow-up) identified Health Professionals’ Follow-up Study (75,596
apples, pears and green leafy vegetables most women; mean baseline age 46 years;14 years of
effective at lowering total stroke risk (Fig. 17.4) follow-up, 366 stroke cases; 38,683 men, mean
[32]. The Dutch cohorts of the European baseline age 54 years, mean BMI 26, 8 years of
Prospective Investigation into Cancer (EPIC) follow-up) found that in US men and women each
Study (20,069 adults; mean baseline age 41 years; serving of fruits and vegetables was associated
45% men; 10 years of follow-up) identified white with a 6% lower risk of ischemic stroke [93]. Risk
fruits (e.g., apples, pears, bananas) and vegetables reductions per serving of specific fruits and vegeta-
(e.g., cauliflower, cucumber, and mushrooms, gar- bles are shown in Fig. 17.7.
Apples/pears (p-trend =.02) Green leafy vegetables (p-trend =.03)

1.02
1
0.98
0.96
0.94
0.92
0.9
0.88
0.86
0.84
0.1 0.2 0.5 1
Median servings/day
Fig. 17.4 Association between apples and pears, and green leafy vegetables and total stroke risk among men and
women (adapted from [32])
1.2
1
Hazard Ratio for Total Stroke
0.8
0.6
0.4
0.2
57 98 142 216
Total White Fruit and Vegetable Intake* (g/day)
Fig. 17.5 Association between total white fruits and vegetables intake and total stroke risk (p-trend = 0.002) (adapted
from [88]). *55% apples, pears, apple sauce, and cloudy apple juice; 35% bananas, cauliflower, cucumber, and mush-
rooms; 10% garlic, leek and onion
17.2.3 Dietary Protein Sources Study (84,010 women aged 30–55 years at base-
line and 43,150 men aged 40–75 years at baseline;
Dietary protein sources have variable effects on during 26 and 22 years of follow-up) found that
stroke risk [94]. The pooled data from the Nurse’s the intake of red meat and potentially legumes
Health Study and Health Professionals Follow-up increased total stroke risk whereas intake of poul-
Raw (p-trend =.03) Processed (p-trend =.41) Total (p-trend =.49)

1.3
Hazard Ratio for Total Stroke incidence

1.2
1.1
0.9
0.8
0.7
0.6
100 150 200 250
Fruit & Vegetable Intake (g/day)
Fig. 17.6 Association between raw and processed fruits and vegetables and total stroke risk (multivariate adjusted)
(adapted from [89])
Potatoes
Legumes
Citrus fruit juices
Total citrus fruits
All vegetables
All fruits
–40 –30 –20 –10 0 10 20 30

Risk of Ischemic Stroke (%) per Daily Serving
Fig. 17.7 Association between specific fruits and vegetables per serving and ischemic stroke risk in women and men
(adapted from [93])
try, nuts, fish, and both whole-fat and low-fat dairy 17.2.3.1 Legumes
lowered risk (Fig. 17.8) [94]. These associations Total legumes or dietary pulses show variable
were independent of other major stroke risk fac- effects but they are generally not associated with
tors. The two-primary plant based protein sources, total stroke risk whereas soy products have been
legumes and nuts, will be reviewed in more detail. shown to lower total stroke risk [25, 95–98].
1.2
1
0.8
0.6
0.4
0.2
0
Total red Poultry Nuts Fish Low fat Whole-fat Legumes
meat dairy diary
Protein Source
Fig. 17.8 Association between protein source (per serving) and total stroke risk in men and women from the pooled
Nurses’ Health Study and Health Professionals Follow-up Study data (adapted from [94])
Total Legume and Dietary Pulses sis of nuts (11 cohort studies; 396,768 partici-
Several meta-analyses consistently show that pants) found that high vs. low nut intake lowered
higher total legume or dietary pulse intakes are stroke risk by 11% and each daily nut serving
not significantly associated with lower total lowered risk by 7% [99]. Sub-group analysis
stroke risk [25, 95–97]. A 2017 systematic review showed that high vs. low intake of tree nuts low-
and meta-analysis (6 legume studies; 254,628 ered risk by 7% compared to 17% for peanuts. A
participants) showed that 4 weekly 100 g servings 2014 meta-analysis (8 cohort studies; 468,887
of legumes (dietary pulses) was insignificantly subjects) showed that a diet containing greater
associated with lowered stroke risk by 2% [95]. amounts of nuts significantly lowered risk of
stroke by 10% [25]. Gender significantly modi-
Soy Products fied the effects of nut consumption on stroke
Two meta-analyses found that increased soy prod- risk, and high nut intake was associated with
ucts intake significantly lowers total stroke risk [97, reduced risk of stroke in women by 15% com-
98]. A 2017 meta-analysis (10 cohort and 7 case- pared to only 5% for men. Similar findings in
control studies) found that high intake of soy prod- another meta-analysis (6 articles including 9
ucts resulted in a significantly lower risk for total cohorts; 476,181 participants) found a 10%
stroke by 18% [98]. A 2016 meta-analysis (5 cohort reduction in stroke risk (highest vs. lowest
studies and 6 case-control studies) showed a sig- intake), which was significant for females only
nificant inverse association between soy intake and [24]. Another 2014 meta-analysis (3 cohort
total stroke risk but there was not an association studies and one RCT; 155,685 participants)
between soy isoflavones intake and stroke risk [97]. showed a trend that 4 weekly servings of nuts
lowered total stroke risk by 11% [96]. A system-
17.2.3.2 Tree Nuts and Peanuts atic review and meta-analysis (32 cohort stud-
ies; 841,211 participants) found that higher
Meta-analyses intake of monounsaturated fat (MUFA), oleic
Meta-analyses of tree nuts and peanuts consis- acid, and higher MUFA: SFA ratio reduced
tently show a stroke risk protective effect [24, stroke risk by 17%, which is characteristic of
25, 96, 99]. A 2016 dose-response meta-analy- most nuts [100].
17.3 Tea and Coffee on Stroke Risk 463
Prospective Cohort Studies age 58 years at baseline; 13.6 years of follow-up)

Several cohort studies provide insights on the showed that those drinking > 2 cups tea/day sig-
effects of nuts on stroke risk [101, 102]. Three nificantly reduced ischemic stroke by 21% com-
large cohorts (71,764 US participants in the pared to non-tea drinkers [104].
Southern Community Cohort Study in the south-
eastern United States, and the other 2 cohorts 17.3.1.2 Coffee
included 134,265 participants in the Shanghai Coffee is a complex mixture of biologically
Women’s Health Study and the Shanghai Men’s active substances with both potential beneficial
Health Study in Shanghai, China; 5.4–12.2 years and adverse effects on cardiovascular or cerebro-
of follow-up) found that peanuts significantly vascular health [15, 17]. The phenolic com-
lowered ischemic stroke and hemorrhagic stroke pounds in coffee, such as caffeic, ferulic, and
by 23% in Asians [101]. The US Physicians’ chlorogenic ac ids, have a strong antioxidant
Health Study (21,078 men; mean baseline age activity and may reduce the oxidation of low den-
55 years; 21 years of follow up) found that the sity lipoprotein cholesterol and systemic inflam-
consumption of ≥7 weekly nut servings showed matory markers, and increase insulin sensitivity
a trend toward reduced ischemic stroke risk by with inconsistent effects on BP depending on
7% [102]. whether the subjects are normotensive or hyper-
tensive and new or habitual coffee drinkers.
A 2011 meta-analysis (11 cohort studies; 479,689
17.3 T
ea and Coffee on participants; 2-24 years of follow-up) found that
Stroke Risk the consumption of 1 to 6 cups of coffee/d was
significantly inversely associated with risk of
Tea and coffee are the most frequently consumed total stroke, with the strongest association (17%
beverages worldwide and also important sources lower risk) being observed for 3 to 4 cups/day
of polyphenols, which have potential beneficial [15]. Heavy coffee consumption >7 cups/day was
effects on cardiovascular health [17]. The poly- not significantly associated with total stroke risk.
phenols in these beverages may reduce the risk of The associations were similar for ischemic stroke
stroke through multiple beneficial mechanisms, and hemorrhagic stroke, but only results for isch-
including antihypertensive, hypo-cholesterolemic, emic stroke was significant. The Finnish Alpha-
antioxidative and anti-inflammatory, vascular Tocopherol, Beta-Carotene Cancer Prevention
endothelial function and insulin sensitivity effects. Study (26,556 male smokers; mean age 58 years
at baseline; 13.6 years of follow-up) showed that
drinking > 8 cups coffee/d significantly reduced
17.3.1 Prospective Cohort Studies ischemic stroke by 23% compared to those drink-
ing < 2 cups/day [104].
17.3.1.1 Tea
There is a large body of scientific evidence that Conclusions
tea is associated with reduced hypertension risk Stroke and related cerebrovascular diseases
and other major risk factors associated with make-up the second most common cause of
stroke [17]. A 2012 meta-analysis (14 cohort death worldwide, accounting for 6.2 million
studies; 513,804 participants; median 11.5 years deaths (11% of total deaths). Over 90% of
of follow-up) showed that 3 cups of green and stroke risk is attributable to modifiable risk
black tea were associated with a reduction in total factors such as smoking, poor diet, and low
stroke risk for green tea by 17% (p < 0.01) and physical activity. Stroke is a major cause of
black tea by 9% (p = 0.17) [103]. Three cups of disability and death worldwide, which
overall tea consumption were associated with a changes the lives not only of the stroke vic-
significant 13% lower risk of total stroke and a tims but also of their families as many
24% lower ischemic stroke risk. The Finnish become dependent in their activities of daily
Alpha-Tocopherol, Beta-Carotene Cancer living due to significant stroke related cog-
Prevention Study (26,556 male smokers; mean nitive and other physical deficits. Forecasts
suggest a >20% increase in stroke preva- vegetables, white fruits and vegetables (e.g.,
lence by 2030 compared to 2012 rates, apples, cauliflower, mushrooms) and vitamin
because of the increase in aging populations. C rich fruits and vegetables (e.g., citrus, bell
Regular consumption of specific foods and peppers, broccoli). Plant protein sources such
beverages may have significant effects on as legumes and nuts have mixed effects on
stroke risk with potential stroke protective stroke risk. Dietary pulses or total legumes are
foods including low fat dairy, whole grains, not associated with stroke risk but nuts and
fruits, vegetables, nuts, tea and coffee whereas soy foods are associated with lower stroke
high intake of red meat, sugar-sweetened bev- risk. Whole plant foods containing a variety of
erages and alcohol may increase stroke risk. nutrients and phytochemicals such as fiber,
The consumption of ≥3 daily servings of antioxidant vitamins, potassium, magnesium,
whole grains is associated with lower total carotenoids, flavonoids and phytosterols may
stroke risk by 8–14% and ischemic stroke by provide potential stroke protection by mecha-
25% compared to never or rare intake. nisms associated with promoting vascular
Increased consumption by 200 g (or 2 1/2 health by attenuating elevated blood pressure,
servings)/day reduces total stroke for fruits lowering LDL-cholesterol levels and systemic
and vegetables by 16%, fruits by 18% and inflammation associated with atherosclerosis,
vegetables by 13%. Raw fruits and vegetables and promoting better insulin sensitivity, blood
are more effective than processed forms; glucose control, weight control, and microbi-
examples of effective specific varieties for ota health compared to less healthy or refined
reducing ischemic stroke risk are green leafy plant foods.
Appendix A 465
ppendix A: Estimated Range of Energy, Fiber, Nutrients

A
Components Whole-Grains Fresh Fruit Dried Fruit Vegetables Legumes Nuts/Seeds

Nutrients/ Wheat, oat, Apples, Dates, dried Potatoes, Lentils, Almonds, Brazil
Phytochemicals barley, rye, pears, figs, spinach, carrots, chickpeas, nuts, cashews,
brown rice, bananas, apricots, peppers, lettuce, split peas, hazelnuts,
whole grain grapes, cranberries, green beans, black beans, macadamias,
bread, cereal, oranges, raisins, and cabbage, pinto beans, pecans, walnuts,
pasta, rolls, blueberries, prunes onions, and soy beans peanuts,
and crackers strawberries, cucumber, sunflower seeds,
and avocados cauliflower, and flaxseed
mushrooms,
and broccoli
Energy (kcals) 110–350 30–170 240–310 10–115 85–170 520–700
Protein (g) 2.5–16 0.5–2.0 0.1–3.4 0.2–5.0 5.0–17 7.8–24
Available 23–77 1.0–25 64–82 0.2–25 10–27 12–33
carbohydrate (g)
Fiber (g) 3.5–18 2.0–7.0 5.7–10 1.2–9.5 5.0–11 3.0–27
Total fat (g) 0.9–6.5 0.0–15 0.4–1.4 0.2–1.5 0.2–9.0 46–76
SFA (g) 0.2–1.0 0.0–2.1 0.0 0.0–0.1 0.1–1.3 4.0–12
MUFA (g) 0.2–2.0 0.0–9.8 0.0–0.2 0.1–1.0 0.1–2.0 9.0–60
PUFA (g) 0.3–2.5 0.0–1.8 0.0–0.7 0.0.0.4 0.1–5.0 1.5–47
Folate (ug) 4.0–44 <5.0–61 2–20 8.0–160 50–210 10–230
Tocopherols (mg) 0.1–3.0 0.1–1.0 0.1–4.5 0.0–1.7 0.0–1.0 1.0–35
Potassium (mg) 40–720 60–500 40–1160 100–680 200–520 360–1050
Calcium (mg) 7.0–50 3.0–25 10–160 5.0–200 20–100 20–265
Magnesium (mg) 40–160 3.0–30 5.0–70 3.0–80 40–90 120–400
Phytosterols (mg) 30–90 1.0–83 — 1.0–54 110–120 70–215
Polyphenols (mg) 70–100 50–800 — 24–1250 120–6500 130–1820
Carotenoids (ug) — 25–6600 1.0–2160 10–20,000 50–600 1.0–1200
Circulation. 2013;128: 553–565
Sci Nutr. 2016; 56(7):1097–1107
Service, Home and Garden Bulletin 72.
An. 1999; 12:169–196
Lu Q-Y, Zhang Y, Wang Y, et al. California Hass avocado: profiling of carotenoids, tocopherol, fatty acid, and fat content
during maturation and from different growing areas. J Agric Food Chem. 2009; 57(21):10408–10413
States. J Agric Food Chem. 2004; 52: 4026–4037
References 15. Larsson SC, Orsini N. Coffee consumption and risk

of stroke: a dose-response meta-analysis of prospec-
tive studies. Am J Epidemiol. 2011;174:993–1001.
1. Mendis S, Puska P, Norrving B, editors. Global
16. Kim B, Nam Y, Kim J, et al. Coffee consumption
atlas on cardiovascular disease prevention and
and stroke risk: a meta-analysis of epidemiologic
control. The underlying pathology of heart attacks
studies. Korean J Fam Med. 2012;33:356–65.
and strokes. Geneva: World Health Organization;
17. Larsson SC. Coffee, tea, and cocoa and risk of
2011.
stroke. Stroke. 2014;45:309–14.
2. Mozaffarian D, Benjamin EJ, Go AS, et al. Heart
18. Larsson SC, Virtamo J, Wolk A. Chocolate con-
disease and stroke statistics-2015 update: a report
sumption and risk of stroke: a prospective cohort of
from the American Heart Association. Circulation.
men and women. Neurology. 2012;79(12):1223–9.
2015;131:29–322.
19. Martınez-Gonzalez MA, Dominguez LJ, Delgado-
3. Feigin VL, Roth GA, Naghavi M, et al. Global burden
Rodriguez M. Olive oil consumption and risk of
of stroke and risk factors in 188 countries, during 1990-
CHD and/or stroke: a meta-analysis of case–con-
2013: a systematic analysis for the Global Burden of
trol, cohort and intervention studies. Br J Nutr.
Disease Study. Lancet Neurol. 2016;15(9):913–24.
2014;112:248–59.
https://doi.org/10.1016/S1474-4422(16)30073-4.
20. Yakoob MY, Shi P, Hu FB, et al. Circulating bio-
4. Meschia JF, Bushnell C, Boden-Albala B, et al.
markers of dairy fat and risk of incident stroke in
Guidelines for the primary prevention of stroke.
U.S. men and women in 2 large prospective cohorts.
A statement for healthcare professionals from the
Am J Clin Nutr. 2014;100(6):1437–47.
American Heart Association/American Stroke
21. Pimpin L, Wu JHY, Haskelberg H, et al. Is but-
Association. Stroke. 2014;45(12):3754–832. https://
ter back? A systematic review and meta-analysis
doi.org/10.1161/STR.0000000000000046.
of butter consumption and risk of cardiovascular
5. Goldstein LB, Bushnell CD, Adams RJ, et al.
disease, diabetes, and total mortality. PLoS One.
Guidelines for the primary prevention of stroke.
2016;11(6):e0158118. https://doi.org/10.1371/ jour-
A guideline for healthcare professionals from the
nal. pone. 0158118.
American Heart Association/American Stroke
22. Hu D, Huang J, Wang Y, et al. Dairy foods and risk
Association. Stroke. 2011;42(2):517–84. https://doi.
of stroke: a meta-analysis of prospective cohort stud-
org/10.1161/STR.0b013e3181fcb238.
ies. Nutr Metab Cardiovasc Dis. 2014;24(5):460–9.
6. Niewadaa M, Michel P. Lifestyle modification
23. de Goede J, Soedamah-Muthu SS, Pan A, et al. Dairy
for stroke prevention: facts and fiction. Curr Opin
consumption and risk of stroke: a systematic review
Neurol. 2016;29:9–13.
and updated dose–response meta-analysis of pro-
7. Kontogianni MD, Panagiotakos DB. Dietary pat-
spective cohort studies. J Am Heart Assoc. 2016;5(5).
terns and stroke: a systematic review and re-meta-
https://doi.org/10.1161/JAHA.115.002787.
analysis. Maturitas. 2014;79:41–7.
24. Zhang Z, Xu G, Wei Y, et al. Nut consumption and
8. Neter JE, Stam BE, Kok FJ, et al. Influence of
risk of stroke. Eur J Epidemiol. 2015;30(3):189–96.
weight reduction on blood pressure: a meta-analysis
25. Shi ZQ, Tang JJ, Wu H, et al. Consumption of nuts
of randomized controlled trials. Hypertension.
and legumes and risk of stroke: a meta-analysis of
2003;42:878–84.
prospective cohort studies. Nutr Metab Cardiovasc
9. Strazzullo P, D’Elia L, Cairella G, et al. Excess
Dis. 2014;24(12):1262–71.
body weight and incidence of stroke: meta-analysis
26. Kaluza J, Wolk A, Larsson SC. Red meat consump-
of prospective studies with 2 million participants.
tion and risk of stroke a meta-analysis of prospective
Stroke. 2010;41:e418–26.
studies. Stroke. 2012;43:2556–60.
10. Lee CD, Folsom AR, Blair SN. Physical activity
27. Yang C, Pan L, Sun C, et al. Red meat consumption
and stroke risk: a meta-analysis. Stroke. 2003;34:
and the risk of stroke: a dose-response meta-analysis
2475–81.
of prospective cohort studies. J Stroke Cerebrovasc
11. Wendel-Vos GC, Schuit AJ, Feskens EJ, et al.
Dis. 2016;25(5):1177–86.
Physical activity and stroke: a meta-analysis of
28. Chen GC, Lv DB, Pang Z, Liu QF. Red and pro-
observational data. Int J Epidemiol. 2004;33:
cessed meat consumption and risk of stroke: a meta-
787–98.
analysis of prospective cohort studies. Eur J Clin
12. Masson P, Webster AX, Hong M, et al. Chronic
Nutr. 2013;91(1):91–5.
kidney disease and the risk of stroke: a systematic
29. Liu S, Manson JE, Stampfer MJ, et al. Whole grain
review and meta-analysis. Nephrol Dial Transplant.
consumption and risk of ischemic stroke in women a
2015;30(7):1162–9.
prospective study. JAMA. 2000;284:1534–40.
13. Reynolds K, Lewis B, Nolen JD, et al. Alcohol con-
30. Chen J, Huang Q, Shi W, et al. Meta-analysis of the
sumption and risk of stroke: a meta-analysis. JAMA.
association between whole and refined grain con-
2003;289:579–88.
sumption and stroke risk based on prospective cohort
14. Zhang C, Qin YY, Chen Q, et al. Alcohol intake and
studies. Asia Pac J Public Health. 2016;28(7):563–75.
risk of stroke: a dose-response meta-analysis of pro-
https://doi.org/10.1177/1010539516650722.
spective studies. Int J Cardiol. 2014;174(3):669–77.
References 467
31. Hu D, Huang J, Wang Y, et al. Fruits and vegetables foods in the United States. J Agric Food Chem.
consumption and risk of stroke a meta-analysis of 2004;52:4026–37.
prospective cohort studies. Stroke. 2014;45:1613–9. 47. Bhupathiraju SN, Tucker KL. Coronary heart dis-
32. Larsson SC, Virtamo J, Wolk A. Total and specific ease prevention: nutrients, foods, and dietary pat-
fruit and vegetable consumption and risk of stroke: a terns. Clin Chim Acta. 2011;412:1493–514.
prospective study. Atherosclerosis. 2013;227:147–52. 48. Dietary Guidelines Advisory Committee. Scientific
33. Chowdhury R, Stevens S, Gorman D, et al. Report. Advisory Report to the Secretary of
Association between fish consumption, long chain Health and Human Services and the Secretary of
omega 3 fatty acids, and risk of cerebrovascular Agriculture. Part D. Chapter 1: Food and nutrient
disease: systematic review and meta-analysis. BMJ. intakes, and health: current status and trends; 2015.
2012;345:e6698. https://doi.org/10.1136/bmj.e6698. p. 1–78.
34. Amiano P, Chamosa S, Etxezarreta N, et al. No asso- 49. http://health.gov/dietary guidelines/2015/guidelines/.
ciation between fish consumption and risk of stroke Accessed 26 Jan 2016.
in the Spanish cohort of the European Prospective 50. Ascherio A, Rimm EB, Hernan MA, et al. Intake
Investigation into Cancer and Nutrition (EPIC- of potassium, magnesium, calcium, and fiber
Spain): a 13.8-year follow-up study. Public Health and risk of stroke among US men. Circulation.
Nutr. 2015;19(4):674–81. 1998;98:1198–204.
35. Larsson SC, Akesson A, Wolk A. Sweetened bever- 51. Lackland DT, Roccella EJ, Deutsch A, et al. Factors
age consumption is associated with increased risk of influencing the decline in stroke mortality: a state-
stroke in women and men. J Nutr. 2014;144:856–60. ment from the American Heart Association/American
36. Bernstein AM, de Koning L, Flint AJ, et al. Soda Stroke Association. Circulation. 2014;45:315–53.
consumption and the risk of stroke in men and 52. Evans CEL, Greenwood DC, Threapleton DE, et al.
women. Am J Clin Nutr. 2012;95:1190–9. Effects of dietary fibre type on blood pressure: a
37. Jayalath VH, de Souza RJ, Ha V, et al. Sugar- systematic review and meta-analysis of randomized
sweetened beverage consumption and inci- controlled trials of healthy individuals. J Hypertens.
dent hypertension: a systematic review and 2015;33(5):897–911.
meta-analysis of prospective cohorts. Am J Clin 53. Davis JN, Hodges VA, Gillham MB. Normal-weight
Nutr. 2015;102:914–21. adults consume more fiber and fruit than their age-
38. U.S. Department of Agriculture, Agriculture and height-matched overweight/obese counterparts.
Research Service, Nutrient Data Laboratory. USDA J Am Diet Assoc. 2006;106:833–40.
National Nutrient Database for Standard Reference, 54. Slavin JL. Dietary fiber and body weight. Nutrition.
Release 27; 2014. http://www.ars.usda.gov/nutrient- 2005;21:411–8.
data. Accessed 17 Feb 2015. 55. Fogelholm M, Anderssen S, Gunnarsdottir I, Lahti-
39. Ros E, Hu FB. Consumption of plant seeds and car- Koski M. Dietary macronutrients and food consump-
diovascular health epidemiological and clinical trial tion as determinants of long-term weight change in
evidence. Circulation. 2013;128(5):553–65. https:// adult populations: a systematic literature review.
doi.org/10.1161/CIRCULATIONAHA.112.001119. Food Nutr Res. 2012;56. https://doi.org/10.3402/fnr.
40. USDA. What We Eat in America, NHANES 2011– v56i0.19103.
2012, individuals 2 years and over (excluding breast- 56. Genser B, Silbernage G, De Backer G, et al. Plant ste-
fed children). www.ars.usda.gov/nea/bhnrc/fsrg. rols and cardiovascular disease: a systematic review
41. Slavin JL, Lloyd B. Health benefits of fruits and veg- and meta-analysis. Eur Heart J. 2012;33:444–51.
etables. Adv Nutr. 2012;3:506–16. 57. Lau C, Faerch K, Glumer C, et al. Dietary glyce-
42. Rebello CJ, Greenway FL, Finley JW. A review of mic index, glycemic load, fiber, simple sugars, and
the nutritional value of legumes and their effects on insulin resistance: the Inter99 study. Diabetes Care.
obesity and its related co-morbidities. Obes Rev. 2005;28:1397–403.
2014;15:392–407. 58. Tosh SM. Review of human studies investigating
43. Gebhardt SE, Thomas RG. Nutritive value of foods. the post-prandial blood-glucose lowering ability
U.S. Department of Agriculture, Agricultural Research of oat and barley food products. Eur J Clin Nutr.
Service, Home and Garden Bulletin 72; 2002. 2013;67:310–7.
44. Holden JM, Eldridge AL, Beecher GR, et al. 59. Silva FM, Kramer CK, de Almeida JC, et al. Fiber
Carotenoid content of U.S. foods: an update of the intake and glycemic control in patients with type 2
database. J Food Comp Anal. 1999;12:169–96. diabetes mellitus: a systematic review with meta-
45. Lu Q-Y, Zhang Y, Wang Y, et al. California Hass analysis of randomized controlled trials. Nutr Rev.
avocado: profiling of carotenoids, tocopherol, 2013;71(12):790–801.
fatty acid, and fat content during maturation and 60. Sen A, Marsche G, Freudenberger P, et al. Association
from different growing areas. J Agric Food Chem. between higher plasma lutein, zeaxanthin, and vita-
2009;57(21):10408–13. min C concentrations and longer telomere length:
46. Wu X, Beecher GR, Holden JM, et al. Lipophilic results of the Austrian Stroke Prevention Study.
and hydrophilic antioxidant capacities of common J Am Geriatr Soc. 2014;62:222–9.
61. Varbo A, Nordestgaard BG, Tybjaeg-Hansen A, of non-cardiovascular, non-cancer death attributed to

et al. Nonfasting triglycerides, cholesterol, and isch- inflammatory diseases in the Iowa Women’s Health
emic stroke in the general population. Ann Neurol. Study. Am J Clin Nutr. 2007;85:1606–14.
2011;69(4):628–34. 77. Oh K, Hu FB, Cho E, et al. Carbohydrate intake,
62. Whitehead A, Beck EJ, Tosh S, Wolever TMS. glycemic index, glycemic load, and dietary fiber in
Cholesterol-lowering effects of oat β-glucan: a meta- relation to risk of stroke in women. Am J Epidemiol.
analysis of randomized controlled trials. Am J Clin 2005;161(2):161–9.
Nutr. 2014;100:1413–21. 78. Steffen LM, Jacobs DR Jr, Stevens J, et al. Associations
63. Leermaker ET, Darweesh SK, Baena CP, et al. The of whole-grain, refined-grain, and fruit and vegetable
effects of lutein on cardiometabolic health across the consumption with risks of all-cause mortality and
life course: a systematic review and meta-analysis. incident coronary artery disease and ischemic stroke:
Am J Clin Nutr. 2016;103(2):481–94. The Atherosclerosis Risk in Communities (ARIC)
64. Berni Canani R, Di Costanzo M, Leone L, et al. Study. Am J Clin Nutr. 2003;78:383–90.
Potential beneficial effects of butyrate in intestinal 79. Liu S, Sesso HD, Manson JE, et al. Is intake of
and extraintestinal diseases. World J Gastroenterol. breakfast cereals related to total and cause-specific
2011;17(12):1519–28. mortality in men? Am J Clin Nutr. 2003;77:594–9.
65. Zhou Y, Han W, Gong D, Fan Y. Hs-CRP in stroke: a 80. Jacobs DR, Meyer KA, Kushi LH, Folsom AR. Is
meta-analysis. Clin Chim Acta. 2016;435:21–7. whole grain intake associated with reduced total
66. McCall DO, McGartland CP, McKinley MC, et al. and cause-specific death rates in older women? The
Dietary intake of fruits and vegetables improves Iowa Women’s Health Study. Am J Public Health.
microvascular function in hypertensive sub- 1999;89:322–9.
jects in a dose-dependent manner. Circulation. 81. World Health Organization. Diet, nutrition, and
2009;119:2153–60. the prevention of chronic diseases. Geneva: World
67. Noad RL, Rooney C, McCall D, et al. Beneficial effect Health Organization; 1990. http://www.who.int/
of a polyphenol-rich diet on cardiovascular risk: a nutrition/publications/obesity/WHO_TRS_797/en/
randomised control trial. Heart. 2016;102(17):1371– index.html. Accessed 16 April 2015.
9. https://doi.org/10.1136/heartjnl-2015-309218. 82. Rooney C, McKinley MC, Appleton KM, et al. How
68. Jiao J, Xu JY, Zhang W, et al. Effect of dietary fiber much is ‘5-a-day’? A qualitative investigation into
on circulating C-reactive protein in overweight and consumer understanding of fruit and vegetable intake
obese adults: a meta-analysis of randomized con- guidelines. J Hum Nutr Diet. 2017;30(1):105–13.
trolled trials. Int J Food Sci Nutr. 2015;66(1):114–9. https://doi.org/10.1111/jhn.12393.
69. Maukonen J, Saarela M. Human gut microbiota: 83. United States Department of Agriculture. Choose
does diet matter? Proc Nutr Soc. 2015;74:23–36. my plate. 2013. http://www.choosemyplate.gov/.
70. http://wholegrainscouncil.org/whole-grains-101/ Accessed 17 Feb 2015.
what-counts-as-a-serving. Accessed 26 Dec 2015. 84. Micha R, Khatibzadeh S, Shi P, et al. Global, regional
71. Seal CJ, Brownlee IA. Whole-grain foods and and national consumption of major food groups
chronic disease: evidence from epidemiological and in 1990 and 2010: a systematic analysis including
intervention studies. Proc Nutr Soc. 2015;74:313–9. 266 country-specific nutrition surveys w orldwide.
72. McGill CR, Fulgoni VL III, Devareddy L. Ten-year BMJ Open. 2015;5. https://doi.org/10.1136/
trends in fiber and whole grain intakes and food bmjopen-2015-008705.
sources for the United States population: National 85. Aune D, Giovannucci E, Boffetta P, et al. Fruit
Health and Nutrition Examination Survey 2001– and vegetable intake and the risk of cardiovascular
2010. Forum Nutr. 2015;7:1119–30. disease, total cancer and all-cause mortality-a sys-
73. Aune D, Keum NN, Giovannucci E, et al. Whole grain tematic review and dose response meta-analysis of
consumption and risk of cardiovascular disease, can- prospective studies. Int J Epidemiol. 2017:1–28.
cer, and all cause and cause specific mortality: sys- https://doi.org/10.1093/ije/dyw319.
tematic review and dose-response meta-analysis of 86. He FJ, Nowson CA, MacGregor GA. Fruit and veg-
prospective studies. BMJ. 2016;353:i2716. https:// etable consumption and stroke: meta-analysis of
doi.org/10.1136/bmj.i2716. cohort studies. Lancet. 2006;367:320–6.
74. Fang L, Li W, Zhang W, et al. Association 87. Dauchet L, Amouyel P, Dallongeville J. Fruit and veg-
between whole grain intake and stroke risk: evi- etable consumption and risk of stroke: a meta-analysis
dence from a meta-analysis. Int J Clin Exp Med. of cohort studies. Neurology. 2005;65(8):1193–7.
2015;8(9):16978–83. 88. Oude Griep LM, Verschuren WM, Kromhout D,
75. Wu D, Guan Y, Lv S, et al. No evidence of increased et al. Colors of fruit and vegetables and 10-year inci-
risk of stroke with consumption of refined grains: a dence of stroke. Stroke. 2011;42:3190–5.
meta-analysis of prospective cohort studies. J Stroke 89. Oude Griep LM, Verschuren WM, Kromhout
Cerebrovasc Dis. 2015;24(12):2738–46. https://doi. D, et al. Raw and processed fruit and vegetable
org/10.1016/j.jstrokecerebrovasdis.2015.08.004. consumption and 10-year stroke incidence in a
76. Jacobs DR Jr, Andersen LF, Blomhoff R. Whole- population-based cohort study in the Netherlands.
grain consumption is associated with a reduced risk Eur J Clin Nutr. 2011;5:791–9.
References 469
90. Mizrahi A, Knekt P, Montonen J, et al. Plant foods of observational studies. Neuroepidemiology.
and the risk of cerebrovascular diseases: a poten- 2016;46(4):242–52.
tial protection of fruit consumption. Br J Nutr. 98. Yan Z, Zhang X, Li C, et al. Association between
2009;102:1075–83. consumption of soy and risk of cardiovascular dis-
91. Sauvaget C, Nagano J, Allen N, Kodama K. ease: a meta-analysis of observational studies. Eur
Vegetable and fruit intake and stroke mortality in J Prev Cardiol. 2017;24(7):735–47. https://doi.
the Hiroshima/Nagasaki Life Span Study. Stroke. org/10.1177/2047487316686441.
2003;34:2355–60. 99. Aune D, Keum NN, Giovannucci E, et al. Nut con-
92. Bazzano LA, He J, Ogden LG, et al. Fruit and veg- sumption and risk of cardiovascular disease, total
etable intake and risk of cardiovascular disease in cancer, all-cause and causespecific mortality: a
US adults: the first National Health and Nutrition systematic review and dose-response meta-analysis
Examination Survey Epidemiologic Follow-up of prospective studies. BMC Med. 2016;14:207.
Study. Am J Clin Nutr. 2002;76:93–9. https://doi.org/10.1186/s12916-016-0730-3.
93. Joshipura KJ, Ascherio A, Manson JE, et al. Fruit 100. Schwingshackl L, Hoffmann G. Monounsaturated
and vegetable intake in relation to risk of ischemic fatty acids, olive oil and health status: a system-
stroke. JAMA. 1999;282:1233–9. atic review and meta-analysis of cohort stud-
94. Bernstein AM, Pan A, Rexrode KM, et al. Dietary ies. Lipids Health Dis. 2014;13:154. https://doi.
protein sources and the risk of stroke in men and org/10.1186/1476-511X-13-154.
women. Stroke. 2012;43:637–44. https://doi. 101. Luu HN, Blot WJ, Xiang YB, et al. Prospective eval-
org/10.1161/STROKEAHA.111.633404. uation of the association of nut/peanut consumption
95. Marventano S, Pulido MI, Sánchez-González C, with total and cause-specific mortality. JAMA Intern
et al. Legume consumption and CVD risk: a sys- Med. 2015;175(5):755–66. https://doi.org/10.1001/
tematic review and meta-analysis. Public Health jamainternmed.2014.8347.
Nutr. 2017;20(2):245–54. https://doi.org/10.1017/ 102. Djoussé L, Gaziano JM, Kase C, Kurth T. Nut con-
S1368980016002299. sumption and risk of stroke in US male physicians.
96. Afshin A, Micha R, Khatibzadeh S, Mozaffarian Clin Nutr. 2010;29(5):605–9.
D. Consumption of nuts and legumes and risk of 103. Shen L, Song L-G, Hong MA, et al. Tea consump-
incident ischemic heart disease, stroke, and diabetes: tion and risk of stroke: a dose-response meta-
a systematic review and meta-analysis. Am J Clin analysis of prospective studies. Biomed Biotech.
Nutr. 2014;100(1):278–88. 2012;13(8):652–62.
97. Lou D, Li Y, Yao G, et al. Soy consumption with risk 104. Larsson SC, Mannisto S, Virtanen MJ, et al. Coffee
of coronary heart disease and stroke: a meta-analysis and tea consumption and risk of stroke subtypes in
male smokers. Stroke. 2008;39(6):1681–87
Dietary Patterns, Foods
and Beverages in Age-Related 18
Cognitive Performance
and Dementia
Keywords
Western diet • Diet quality • Healthy diet • Mediterranean diet • DASH
diet • MIND diet • Fruits • Vegetables • Soy • Nuts • Dairy • 100% fruit
juice • Coffee • Tea • Cocoa • Alcohol • Polyphenols • Flavanols •
Anthocyanins • Lutein • Macular pigment density
Key Points • The Mediterranean diet (MedDiet), DASH

diet, and MIND diet (a hybrid of both the
• All measures of cognitive performance decline MedDiet and DASH diet with an emphasis on
with age, with executive functioning (e.g., work- specific brain protective foods) are effective in
ing memory, reasoning, task flexibility, problem protecting cognitive performance with aging.
solving and planning) showing the largest rate of • A number of randomized controlled trials
decline with every successive decade of age. (RCTs) and prospective cohort studies support
There is a considerable degree of heterogeneity the benefits of high polyphenolic fruits and veg-
in cognitive performance across populations, etables, dairy (especially yogurt), 100% juices
which can be significantly affected by dietary (polyphenol rich), coffee, tea, flavanol-
rich
pattern and specific foods and beverages. cocoa beverages, and low-moderate wine con-
• In general, following dietary advice for lower- sumption on improving age-related cognitive
ing the risk of cardiovascular and metabolic performance and reducing risk of dementia, but
disorders, such as consuming high levels of excessive alcohol consumption can have nega-
healthy fats from fish or vegetable oils, non- tive effects on cognitive performance and lead
starchy vegetables, low glycemic index fruits to higher risk of dementia.
and a diet low in foods with added sugars • Lutein has been shown to preferentially accu-
should be encouraged for cognitive health. mulate in the human brain and its content in
There is significant evidence from human neural tissue as reflected in macular pigment
studies that low quality diets reduce and high- density has been positively correlated with
quality diets enhance global cognitive perfor- cognitive performance and reduced risk of
mance with aging. dementia.

472 18 Dietary Patterns, Foods and Beverages in Age-Related Cognitive Performance and Dementia
18.1 Introduction tion of amyloid-β peptides) or stroke [2–8]. With

aging populations, there are eight million new
All measures of cognitive performance decline dementia cases each year and the burden of ill-
with age, with the most rapid loss rate for infor- ness it creates approaches pandemic proportions.
mation processing speed [1]. Executive function- Dementia affects each person in a different way,
ing (e.g., working memory, reasoning, task depending upon the impact of the disease and the
flexibility, problem solving and planning) shows person’s personality. There are typically three
the largest rate of decline with every further stages: (1) The early stage of dementia is often
decade of age. There is a considerable degree of gradual and often overlooked; common symp-
heterogeneity in cognitive performance across toms include: forgetfulness, losing track of the
populations. In the Mini–Mental State time and becoming lost in familiar places; (2) As
Examination (MMSE), a 30-point questionnaire dementia progresses to the middle stage, the
that is used extensively in clinical and research signs and symptoms become clearer and more
settings to measure cognitive impairment and to restricting; common symptoms include: becom-
estimate the severity and progression of cognitive ing forgetful of recent events and people’s names,
impairment, males declined at a slightly slower becoming lost at home, and having increasing
rate than females, and every additional year of difficulty with communication such as repeating
education was associated with a slightly slower questions and personal care; and (3) Late stage of
rate of MMSE decline. Although apolipoprotein dementia is one of near total dependence and
E (Apo-E) is a major cholesterol carrier that sup- inactivity; common symptoms include: becom-
ports lipid transport and injury repair in the brain, ing unaware of the time and place, having diffi-
individuals that are APOE ε4 allele carriers tend culty recognizing relatives and friends, having an
to have a slightly more rapid decline in most cog- increasing need for assisted self-care, having dif-
nitive measures than those who are non-APOE ε4 ficulty walking and experiencing behavior
allele carriers due to increased risk of cerebral changes such as increased aggression.
amyloid angiopathy and age-related cognitive Alzheimer’s disease is the most common form of
impairment, especially in processing speed [1]. dementia and may contribute to 60–70% of late
The World Health Organization (WHO) reports stage cases. Lifestyle factors such as diet quality,
that 50 million people are affected by dementia levels of physical activity and smoking play
worldwide and the number of cases is expected to important roles in determining the risk of age-
triple by 2050 [2]. Dementia is a syndrome, typi- related cognitive impairment [2, 6–8]. Figure 18.1
cally chronic or progressive in development, in provides an overview of the biological process.
which there is deterioration in cognitive function Poor quality Western diets adversely influence
beyond what might be expected from normal cognitive health and are a major factor in cogni-
aging and which progressively interferes with a tive decline and dementia and the opposite is
person’s ability to function at work or in other associated with high quality healthy diets [7, 8].
everyday activities [2]. It affects memory, think- In general, following dietary advice including
ing, orientation, comprehension, calculation, the consumption of healthy fats from fish or veg-
learning capacity, language, and judgement. etable oils, non-starchy vegetables, low glycemic
Pathological mechanisms for age related demen- index fruits and a diet low in foods with added
tia are complex including; (1) reduced vascular sugars should be encouraged for lowering the
function related to chronic systemic inflamma- risk of cardiovascular and metabolic disorders to
tion, hyperinsulinemia or elevated systolic blood help support cognitive health and reduce the risk
pressure; (2) neuroinflammation, microglial dys- of dementia [6]. The objective of this chapter is to
function and brain cellular senescence are associ- comprehensively review the effects of dietary
ated with overweight and obesity related colonic pattern quality, foods, beverages, and individual
microbiota dysfunction; and (3) brain injuries dietary components on age-related cognitive
such as Alzheimer’s disease (cerebral accumula- function and dementia.
18.2 Diet Quality 473
Western Dietary Pattern:

Low in fiber-rich whole-foods and high in saturated fat, added sugar and sodium
Increased Disease Risk:

Obesity, diabetes, dyslipidemia, hypertension, and non-alcoholic fatty liver disease
Mechanisms:
Microbiota dysbiosis, altered brain blood barrier, insulin resistance, microvascular changes,
neuroinflammation, obesity induced oxidative stress, arterial stiffness, and microglial dysfunction
Brain Structure Changes:

White matter and hippocampus dysfunction
Cognitive Decline and Dementia
Fig. 18.1 Mechanisms associated with the Western diet and cognitive decline and dementia (adapted from [7, 8])
18.2 Diet Quality sured at 3–5 years) showed that diets high in red/
processed meat, gravy, and potatoes/potato dishes
Table 18.1 summarizes the findings from 10 pro- or butter may predispose very old adults to cogni-
spective cohort studies and two randomized con- tive impairment [11]. The Australian Personality
trolled trials (RCTs) on the protective effects of and Total Health Through Life project longitudi-
higher quality diets on cognitive function and nal investigation of older adults (255 subjects;
dementia [9–20]. A 2017 UK Whitehall II Study aged 63 years at baseline; 4 years of follow-up)
(5083 participants; mean baseline age 56 years; showed that those p articipants with a high adher-
29% women; 10 years of follow-up) found that a ence to unhealthy diets or moderate adherence to
major inflammatory diet consisted of higher red healthy diets had significantly smaller left hippo-
and processed meats, peas, fried foods and lower campal volumes compared to those with high
whole grains, which was associated with higher adherence to healthy diets (Fig. 18.3) [12]. The
systemic inflammation and accelerated cognitive Swedish National Study on Aging and Care (2223
decline [9]. The US Reasons for Geographic and healthy participants; mean baseline age 71 years;
Racial Differences in Stroke (REGARDS) study 39% men; 6-years follow-up) demonstrated that
in the Southeast (18,080 subjects; baseline higher adherence to the Western diet was associ-
age ≥ 49 years; cross-sectional analysis) found ated with faster cognitive decline whereas higher
that diets rich in plant-based foods were associ- adherence to higher quality healthy diets was
ated with higher cognitive scores and reduced risk inversely associated with cognitive decline [13].
of cognitive impairment whereas diets high in The Taiwan Longitudinal Study of Aging (1926
fried food and processed meat were associated Chinese men and 1744 Chinese women; baseline
with lower cognitive scores and significantly age > 65 years; 8-years follow-up) showed that
increased risk of cognitive impairment (Fig. 18.2) Western diets or high meat and low fish, beans/
[10]. The UK Newcastle 85+ Study (302 men and legumes, vegetables and fruits diets were posi-
489 women; 85 years +; global cognition mea- tively associated with up to an eightfold increase
Table 18.1 Summaries of studies on diet quality and age-related cognitive performance
Prospective Studies
Ozawa et al. (2017). 5083 participants; 29% women; A major inflammatory dietary pattern
Investigate the association mean baseline age 56 years; characterized by higher intake of red meat,
between diet quality and 10 years of follow-up (multivariate processed meat, peas, and fried food, and
inflammation and cognitive adjusted) lower intake of whole grains was associated
decline (The Whitehall II with high circulating IL-6 levels and
Study; UK) [9] accelerated cognitive decline (significant
reductions in reasoning and global cognition
after adjustment for demographics and
health related factors)
Pearson et al. (2016). 18,080 black and white subjects; Diets rich in plant-based foods were
Evaluate associations baseline age ≥ 49 years associated with higher cognitive scores and
between empirically derived (multivariate adjusted) reduced risk of cognitive impairment
dietary patterns and whereas diets rich in fried food and
cognitive function in the processed meat typical of a Southern diet
Southeast USA known as was associated with lower scores and
the stroke belt. (Reasons for significantly increased risk of cognitive
Geographic and Racial impairment (Fig. 18.2)
Differences in Stroke
(REGARDS) cohort; US)
[10]
Granic et al. (2016). 302 men and 489 women; 85 years Diets high in red/processed meat, gravy, and
Investigate the association +; global cognition [measured by potatoes/potato dishes or butter may
between diet quality and the Standardized Mini-Mental State predisposed older adults to cognitive
global and attention-specific Examination (SMMSE)] over 5 y impairment
cognition (Newcastle 85+ and attention assessed by the
Study; UK) [11] cognitive drug research attention
battery over 3 years (multivariate
adjusted)
Jacka et al. (2015). 255 healthy adults; mean baseline Increased adherence to a healthy quality diet
Examine the association age 63 years; follow-up 4 years was associated with a larger left
between diet quality and (multivariate adjusted) hippocampal volume, while higher
hippocampal volume in adherence to an unhealthy Western diet was
humans (Personality and (independently) associated with a smaller
Total Health Through Life left hippocampal volume (Fig. 18.3)
project; Australian) [12]
Shakersaina et al. (2015). 2223 healthy participants; mean Higher adherence to the Western diet vs.
Evaluate the effect of diet baseline age 71 years; 39% men; lower adherence was associated with
quality on cognitive changes 6-years follow-up; mini-mental cognitive decline whereas higher adherence
with aging (Swedish state examination to a healthy diet was inversely associated
National study on Aging and with cognitive decline
Care) [13]
Tsai (2015). 1926 Chinese men and 1744 A Western diet was positively associated
Examine the effect of diet Chinese women; baseline with an eightfold increased risk of cognitive
quality on cognitive decline age > 65 years; 8-years of follow-up decline over 8 years (adjusted) whereas
in older Taiwanese (Taiwan (multivariate adjusted) traditional and healthy diets were not. Diets
Longitudinal Study of rich in meats and infrequent consumption of
Aging [14] fish, beans/legumes, vegetables and fruits
may adversely affect cognitive function in
older Taiwanese
18.2 Diet Quality 475
Gardener (2015). 527 subjects; mean baseline age The primary findings from this study is that
Investigate the associations 69 years; 40% male; 3 years of higher baseline adherence to a ‘healthy’ diet
of diet quality and change in follow-up; 3 diets: Australian is associated with less decline in the
cognitive performance MedDiet, prudent, and Western executive function composite score in
(Australian Imaging, diets (multivariate adjusted) apolipoprotein E (APO-E) ε4 allele carriers.
Biomarkers and Lifestyle By contrast, higher baseline adherence to the
study of Ageing) [15] ‘unhealthy’ Western diet score is associated
with increased cognitive decline in the
visuospatial functioning composite in
APO-E ε4 allele non-carriers
Zhu et al. (2015). 2435 participants; 18–30 years at Those participants with higher quality diet at
Evaluate the associations baseline; 5 and 25 years of age 18–30 maintained or improved cognitive
between diet quality and follow-up (multivariate adjusted) performance 5 and 25 years later; with
cognitive function in middle especially better word memory recall
age years (The Coronary
Artery Risk Development in
Young Adults (CARDIA)
Study; US) [16]
Ozawa et al. (2013). 1006 healthy community-dwelling High quality diets, high intakes of soybeans
Investigate the associations Japanese subjects; baseline age and soybean products, vegetables, seaweed,
of diet quality and risk of 60–79 years; median follow-up milk and dairy products and a low intake of
dementia in older Japanese 15 years;144 Alzheimer disease rice was associated with 35% lower risk of
(Hisayama Study) [17] cases, 88 vascular dementia cases Alzheimer’s disease and a 55% lower risk of
(multivariate adjusted) vascular dementia
Wengreen et al. (2009). 3634 resident men and women; Participants in the highest quartile of RFS
Examine associations mean baseline age 75 years; scored 1.8 points higher on the baseline
between diet quality and recommended food score (RFS) and cognitive function test than those in the
cognitive performance non-RFS; cognition assessed by the lowest quartile of RFS (p < .001). This effect
among elderly men and modified mini-mental state was strengthened over 11 years of follow-up.
women (Cache County examination (3MS) at baseline and Those with the highest RFS declined by 3.4
Study on Memory and 3 subsequent interviews over points over 11 years compared with the
Aging in Utah) [18] 11 years (multivariate adjusted) 5.2-point decline experienced by those with
the lowest RFS (p = .0013). Consuming a
healthy diet with a variety of recommended
foods may help to attenuate age-related
cognitive decline among the elderly
RCTs
Attuquayefio et al. (2017). Parallel RCT: The experimental group showed a marked
Investigate the impacts of 102 participants completed the trial; increase in blood glucose and triglyceride
high saturated fat and added mean age 20 years; mean BMI 20; response to their breakfast compared to
sugar from the Western diet breakfasts high in saturated fat and controls. Larger changes in blood glucose
on hippocampal related added sugar (experimental group) were associated with greater reductions in
functioning (Australia) [19] vs. similar breakfast food types but hippocampal-dependent learning and
significantly lower in saturated fat memory, similar to findings from animal
and added sugar (control group); studies (Fig. 18.4)
4 days
(continued)
Smyth et al. (2015). 2 Large parallel RCTs: Those in the highest healthy eating quintile
Determine the association of 27,860 men and women; 26% had significantly lower risk of cognitive
dietary factors and risk of women; 56 months of follow-up, decline by 24% compared with the lowest
cognitive decline in a 4699 cases of cognitive decline; quintile. The healthy diet was primarily
population at high risk of Alternative Healthy Eating Index associated with an improvement in attention
cardiovascular disease and calculation ability. Lower risk of
(CVD) (2 international cognitive decline was consistent regardless
parallel trials of the of baseline cognitive level. The interaction
ONTARGET (Ongoing with diet quality and activity is summarized
Telmisartan Alone and in in Fig. 18.5
Combination with Ramipril
Global Endpoint Trial) and
TRANSCEND (Telmisartan
Randomised Assessment
Study in ACE Intolerant
Subjects with CVD) [20]
Plant-based/Healthy DP (p-trend =.23) Southern/Unhealthy DP (p-trend =.05)
1.2
1.15
Odds of Cognitive Impairment
1.1
1.05
1
0.95
0.9
0.85
0.8
0.75
0.7
1 2 3 4 5
Dietary Pattern (DP) Adherence Quintiles
Fig. 18.2 Risk of cognitive impairment and type of dietary pattern (DP) from the Reasons for Geographic and Racial
Differences in Stroke (REGARDS) cohort (adapted from [10])
in the rate of cognitive decline over 8 years com- Coronary Artery Risk Development in Young
pared to healthy or traditional diets [14]. The Adults (CARDIA) Study (2435 participants;
Australian Imaging, Biomarkers and Lifestyle 18–30 years at baseline; 5 and 25 years of follow-
Study of Ageing (527 subjects; mean baseline age up) showed that those participants with a higher
69 years; 40% male; 3-years follow-up) found quality diet at age 18–30 maintained or improved
that higher adherence to a healthy diet is impor- cognitive performance 5 and 25 years later and
tant to reduce risk for cognitive decline, whereas were especially better in word memory recall
the Western diet was associated with faster [16]. The Japanese Hisayama Study (1006 healthy
decline, especially related to executive function community-dwelling Japanese subjects; baseline
and visuospatial functioning [15]. The US age 60–79 years; median follow-up 15 years)
Baseline 4 Year Follow-up

3200
Left Hippocampus Volume (mm3)

3000
2800
2600
2400
2200
2000
Western Diet Healthy Diet Healthy Diet
(Moderate Adherence) (High Adherence)
Fig. 18.3 Association between dietary pattern and left hippocampus volume in adults aged 60–64 years at baseline and
at 4 years of follow-up (p = .008 high healthy diet adherence vs. other diets) (adapted from [12])
demonstrated that a healthy diet with high intake decline) found that those subjects in the healthiest
of soybeans and soybean products, vegetables, dietary quintile of the modified Alternative
seaweed, milk and dairy products and a low intake Healthy Eating Index had significantly lower risk
of rice, was associated with 35% lower risk of of cognitive decline by 24% compared with those
Alzheimer’s disease and a 55% lower risk of vas- in the lowest quintile. The healthy diet was pri-
cular dementia [17]. The Cache County Study on marily associated with an improvement in atten-
Memory and Aging in Utah (3634 resident men tion and calculation [20]. The interaction of diet
and women; mean baseline age 75 years; recom- quality and activity is summarized in Fig. 18.5.
mended food score (RFS) and non-RFS; 11 years Studies on higher quality diets generally show
of follow-up) found that participants in the high- they help to attenuate age-related cognitive
est quartile of RFS had significantly slower cogni- decline and protect against dementia (among the
tive declined by 34% compared with those with elderly).
the lowest RFS (p = .0013) [18]. A 2017 RCT
(102 participants; mean age 20 years; breakfasts
high in saturated fat and added sugar (experimen- 18.3 Healthy Dietary Patterns
tal group) vs. a similar breakfast food type but
significantly lower in saturated fat and added Numerous prospective cohort studies and several
sugar (control group); 4 days) demonstrated that RCTs support the beneficial effects of high
the experimental group showed a marked increase adherence to the Mediterranean diet, (MedDiet)
in blood glucose and triglyceride responses to Dietary Approaches to Stopping Hypertension
their breakfast, which was associated with greater (DASH) diet, MIND diet (a hybrid of MedDiet
reductions in hippocampal-dependent learning and DASH) and/or the Nordic diet in protecting
and memory compared to the controls (Fig. 18.4) against age-related cognitive decline and demen-
[19]. The pooled data from two large international tia which are summarized in Table 18.2 [15, 21–
trials with populations at high risk of cardiovascu- 42]. The composition of the Western diet vs
lar disease (27,860 men and women; 26% women; common healthy dietary patterns is summarized
56 months of follow-up, 4699 cases of cognitive in Appendix A.
Healthier Control Diet Western Diet

100
90
Hopkins-Verbal Learning Score (%) 80
70
60
50
40
30
20
10
0
Day 1 Day 4
Fig. 18.4 Effect of four days on a high saturated fat and added sugar Western diet on hippocampal dependent learning
in young adults (p < .001) (adapted from [19])
Overall Sedentary Moderate Activity High Activity

1.1
1.05
Hazard Ratio of Cognitive Decline
1
0.95
0.9
0.85
0.8
0.75
0.7
0.65
0.6
1 2 3 4 5
Healthy Diet (Quintile)
Fig. 18.5 Interaction between adherence to a healthy diet and level of physical activity in 27,860 older adults (mean
baseline age 66 years) over 56 months (adapted from [20])
18.3.1 Mediterranean Diet (MedDiet) a cognitively healthy older population (24 studies
including 16 cohort studies, 3 cross-sectional
18.3.1.1 Systematic Reviews studies, and 5 RCTs) found that higher adherence
and Meta-Analyses to the MedDiet was associated with better cogni-
Seven systematic reviews and meta-analyses con- tive outcomes in 75% of the studies. Highly con-
sistently show that MedDiets are associated with sumed foods in the MedDiet, such as fish, nuts
lower risk for age-related cognitive decline and and vegetables, had protective effects against cog-
the development of Alzheimer disease [21–27]. A nitive decline across the studies [21]. Also,
2017 analysis of the effect of (1) the MedDiet and increased dietary and circulating blood omega-3
(2) omega-3 fatty acids on cognitive outcomes in fatty acids were shown to improve cognition in
Table 18.2 Summary of specific dietary patterns studies in age-related cognitive performance
Mediterranean (MedDiet)
Masana et al. (2017). 24 studies including 16 cohort (1) Adherence to the MedDiet was associated with
Evaluate the effect of studies (8.5 months to 11 years), 3 better cognitive outcomes in 75% of the studies,
(1) the MedDiet and cross-sectional studies, 5 RCTs while 25% found no association. Also, there was
(2) omega-3 fatty (12 weeks to 6.5 years); 45–16,058 some evidence that certain food groups that are
acids on cognitive participants highly consumed in the MedDiet, such as fish, nuts
outcomes in a and vegetables, had a protective effect on cognitive
cognitively healthy decline across included studies. (2) Increased dietary
aged population [21] and circulating blood omega-3 fatty acids were
shown to improve cognition in 62.5% of the studies
and 37.5% of the studies found no association
Wu and Sun (2017). 9 cohort studies; 34,168 Compared with the lowest category, the pooled
Evaluate the participants; baseline analysis showed that the highest MedDiet scores
association and age ≥ 45 years (45–98 years); were inversely associated with developing cognitive
dose-response of the 2–12 years of follow-up disorders by 21%. Subgroup analysis showed that
MedDiet and cognitive the MedDiet was inversely associated with mild
function [22] cognitive impairment by 17% and Alzheimer’s
disease by 40% (Fig. 18.6)
Hardman et al. 18 longitudinal and prospective Of the 18 studies, 13 showed that higher adherence
(2016). studies to a MedDiet was related to slowing the rate of
Evaluate the effect of cognitive decline, minimizing the conversion to AD,
the MedDiet on or improving the cognitive function. Five of the 18
cognitive processes studies did not demonstrate that MedDiet adherence
and Alzheimer’s had a protective effect against cognitive decline or
disease (AD) over show improved cognition. The specific cognitive
time [23] domains that benefit from higher MedDiet score
were memory (delayed recognition, long-term, and
working memory), executive function, and visual
constructs
Petersson et al. 32 studies (5 RCTs and 27 A majority of studies found that the MedDiet was
(2016). observational studies) associated with improved cognitive function, a
Provide an update on decreased risk of cognitive impairment, dementia, or
the effects of the AD. Five studies found no association between the
MedDiet on cognitive MedDiet and cognitive function; 3studies found no
function, cognitive correlation between the MedDiet and AD, and
impairment, AD, and another 3 studies found no association between the
all-type dementia [24] MedDiet and cognitive impairment. Although there
was large heterogeneity and differences in study
quality, the authors concluded that higher adherence
to the MedDiet is associated with better cognitive
performance
Singh et al. (2014). 6 cohorts; 8019 subjects, mean Subjects in the highest MedDiet tertile had
Determine whether baseline age mid-70s; cognitively significantly lower adjusted risk of cognitive
there is an association normal; follow-up of 4–8 years impairment by 33% as compared to the lowest
between the tertile. Among cognitively normal individuals,
Mediterranean diet higher adherence to the MedDiet was associated
(MedDiet) and risk of with significantly reduced risk of mild cognitive
cognitive impairment impairment by 27% and Alzheimer’s disease by 36%
[25]
(continued)
Lourida et al. (2013). 13 studies [7 cohort, 1 Higher adherence to the MedDiet was associated
Evaluate the longitudinal, 1 nested case-control, with better cognitive function, lower rates of
association between 3 cross-sectional studies and 1 cognitive decline, and reduced Alzheimer disease
MedDiet adherence single blind RCT]; 25–3790 risk in 10 out of 13 studies, whereas results for mild
and cognitive function subjects; mean follow-up ranged cognitive impairment were inconsistent
or dementia [26] from 10 days to 8 years
Sofi et al. (2010). 18 cohorts; 2,190,627 subjects; A 2-point increase in adherence to the MedDiet was
Assess the effect of follow-up of 4–20 years. associated with a significant reduction in
adherence to the neurodegenerative diseases by 13%
MedDiet on
neurodegenerative
diseases [27]
Prospective Studies
Gardener et al. 527 subjects; mean baseline age Of the diets, higher adherence to the MedDiet was
(2015). 69 years; 40% male; 3-years of significantly associated after 36 months with better
Investigate the follow-up (multivariate adjusted) performance in the executive function in
association of three apolipoprotein E (APOE) ε4 allele carriers compared
well-recognised to higher Western diet adherence, which was
dietary patterns with significantly associated with greater cognitive
cognitive change decline
(Australian Imaging,
Biomarkers and
Lifestyle study of
Ageing) [15]
Galbete et al. (2015). 823 participants; mean baseline Higher cognitive decline was observed among
Evaluate the age 62 years; 10-point (0–9) participants with low or moderate baseline
association between MedDiet Score was used to adherence to the MedDiet than among those with
adherence to the categorize adherence to MedDiet; better adherence. A higher adherence to the MedDiet
MedDiet and telephone interview of cognitive might be associated with better cognitive function
cognitive function status-modified range 0–54 points; but the differences were of small magnitude
(SUN Project; Spain) 2-year change (multivariate
[28] adjusted)
Tangney et al. (2014). 826 participants; mean baseline A 1-unit increase in DASH score was associated
Estimate the effects of age 81.5 years; 26% men; with a slower rate of global cognitive decline by
DASH and MedDiet 4.1 years of follow-up; 0.007 standardized units (p = .03). Similarly, a
on age-related (multivariate adjusted) 1-unit-higher MedDiet score was associated with a
cognitive function slower rate of global cognitive decline by 0.002
(Memory and Aging standardized units (p = .01)
Project; US) [29]
Samieri et al. (2013). 16,058 women; mean baseline age Each higher quintile of MedDiet score was linearly
Assess the long-term 74 years; 6 years of follow-up positively associated with an increase in mean
effect of the MedDiet (multivariate adjusted) cognitive global cognition and verbal memory, a loss
on cognitive function in these cognitive functions are strong early
and decline (Nurses’ predictors of Alzheimer disease at older age
Health Study; US) (Fig. 18.7). These associations were similar to those
[30] observed in women 1–1.5 years of younger age
Wengreen et al. 3831 men and women; Mean Higher DASH and MedDiet scores were associated
(2013). baseline age 74 years and BMI 26; with higher average cognitive function. People in
Examine associations cognitive function was assessed by quintile 5 of DASH averaged 0.97 point higher than
between DASH and using the Modified Mini-Mental those in quintile 1 (p = .001). The corresponding
MedDiet dietary State Exam; 4 times over 11 years difference for MedDiet quintiles was 0.94 (p = .001).
patterns and age- (multivariate adjusted) Higher intakes of whole grains, nuts, and legumes
related cognitive were also associated with higher average cognitive
change in a function scores by approx. 20% each
prospective,
population-based study
(Cache County Study
on Memory, Health
and Aging; US) [31]
Samieri et al. (2013). 6174 women; baseline age The aMedDiet score was not associated with
Examine the effect of 72 years; 5 years of follow-up. trajectories of repeated cognitive scores (p-trend
adherence to the Alternate MedDiet (aMedDiet) across quintiles = .26 and .40 for global cognition
alternate MedDiet on adherence is based on intakes of: and verbal memory, respectively). Overall global
cognitive function and vegetables, fruits, legumes, whole cognition and verbal memory at older ages were
decline (Women’s grains, nuts, fish and lower intake assessed by averaging the three cognitive measures
Health Study; US) of red and processed meats, (p-trend =.63 and .44, respectively). Among
[32] moderate alcohol, and the healthy alternate MedDiet components, higher
ratio of monounsaturated-to- monounsaturated-to-saturated fats ratio (or higher
saturated fats (multivariate olive oil) was associated with more favorable
adjusted) cognitive trajectories (p-trend = .03 and 0.05 for
global cognition and verbal memory, respectively).
Greater whole-grain intake was associated with
better average global cognition (p-trend = .02)
Valls-Pedret et al. Parallel RCT: The MedDiet supplemented with olive oil or mixed
(2015). 447 cognitively healthy volunteers nuts was associated with significantly improved
Investigate the effect with high cardiovascular disease cognitive function. Compared with the low-fat
of adherence to the (CVD) risk; mean baseline age control diets, MedDiet plus tree nuts significantly
MedDiet on cognitive 67 years; 51% women; randomly improved memory and MedDiet plus extra virgin
function in older assigned to a MedDiet plus olive oil improved frontal and global cognition
adults (PREDIMED extra-virgin olive oil (1 L/week) or
trial; Spain) [33] mixed tree nuts (30 g/day) vs. a
control diet (advice to reduce
dietary fat); median of 4.1 years
Martinez-Lapiscina Parallel RCT: The MedDiet supplemented with extra virgin olive
et al. (2013). 522 participants at high CVD risk; oil or mixed nuts significantly enhanced cognitive
Assess effects of mean baseline age 75 years; 45% function as measured by the Mini-Mental State
MedDiets on men; randomly assigned to a Exam and clock draw test vs. a low-fat diet
cognitive function MedDiet plus extra virgin olive oil
(PREDIMED- or mixed tree nuts vs. advice to
NAVARRA; Spain) reduce dietary fat; 6.5 years of
[34] follow-up (multivariate adjusted)
Martinez-Lapiscina Parallel RCT: Better post-trial cognitive performance versus
et al. (2013). 285 high CVD risk participants; control in all cognitive domains and significantly
Assess the effect of 45% men; mean baseline age better performance across fluency and memory tasks
MedDiets on 74 years; 3 diets: MedDiets plus were observed for participants allocated to the
cognition in a extra-virgin olive oil; MedDiets MedDiet extra virgin olive oil vs. control; mild
controlled intervention plus mixed tree nuts; a low-fat cognitive impairment was significantly reduced by
(PREDIMED- control diet; 6.5 years (multivariate 66% compared with control group (Fig. 18.8).
NAVARRA; Spain) adjusted) Participants assigned to MedDiet plus tree nuts
[35] group did not differ from controls
(continued)
Valls-Pedret et al. Cross-sectional assessment of MedDiet adherence as measured by increasing
(2012). Parallel RCT: consumption of polyphenol rich foods and beverages
Assess whether 447 cognitively healthy volunteers was associated with better cognitive performance.
consumption of with high CVD risk; mean Higher intakes of extra virgin olive oil, coffee,
antioxidant-rich foods baseline age 67 years; 52% walnuts, and wine known to be rich in polyphenols
in the MedDiet relates women; MedDiet (multivariate were associated with better memory function and
to cognitive function adjusted) global cognition. Increased urinary polyphenol
in the elderly excretion directly reflected cognitive/memory
(PREDIMED trial; improvements as summarized in Figs. 18.9 and
Spain) [36] 18.10
Dietary Approaches to Stop Hypertension (DASH) Diet
Berendsen et al. 16,144 women; mean baseline age Greater DASH diet score adherence was associated
(2017). 74 years; 6 years of follow-up with better average cognitive function, irrespective
Examine the (multivariate adjusted) of apolipoprotein E 4 allele status. High DASH diet
association between adherence was not associated with a decline in
long-term adherence cognitive function over 6 years and the data suggests
to the Dietary an improved cognitive function of 1 year younger in
Approaches to Stop age
Hypertension (DASH)
diet and cognitive
function and decline
in older American
women (The Nurses’
Health Study; US)
[37]
Haring et al. (2016). 6425 postmenopausal women; The DASH diet was insignificantly associated with
Determine the effect baseline age 65–79 years; DASH cognitive decline in older women especially those
of dietary patterns on diet score and other healthy women with hypertension
cognitive function in dietary patterns; median follow-up
older women of 9.1 years; 499 cases of mild
(Women’s Health cognitive impairment and 390
Initiative Memory probable dementia cases
Study; Germany) [38] (multivariate adjusted)
Mediterranean-DASH Diet Intervention for Neurodegenerative Delay [MIND] Diet
Cross-over Study
McEvoy et al. 5907 community-dwelling older In this large nationally representative US population
(2017). adults; mean age 68 years of older adults, greater adherence to the MedDiet
evaluate the (multivariate adjusted) and MIND diet was associated with improved
association between The MIND diets emphasizes brain cognitive performance and lower risk of cognitive
the Mediterranean healthy foods (green leafy impairment by 35% (p-trend <.001)
diet (MedDiet) and vegetables, other vegetables, nuts,
the Mediterranean- berries, beans, whole grains,
DASH diet seafood, poultry, olive oil and
Intervention for wine) and restricts unhealthy foods
Neurodegeneration (red meats, stick margarine,
Delay (MIND diet) pastries and sweets, and fried/fast
and cognition in a food)
nationally
representative
population of older
adults (Health and
Retirement Study;
US) [39]
Morris et al. (2015). 960 participants; average baseline The MIND score was positively associated with
Assess the effects of 81 years of age; 75% women; slower decline in global cognitive score (p < .0001)
combining key effects 4.7 years of follow-up (Fig. 18.11). The difference in decline rates for
of the MedDiet and (multivariate adjusted) being in the top tertile of MIND diet scores vs. the
DASH diet on lowest was equivalent to being 7.5 years younger in
slowing cognitive cognitive age
decline with the
MIND diet; Memory
and Aging Project
[MAP]; US) [40]
Morris et al. (2015). 923 participants; average baseline Participants in the top and middle tertile of MIND
Assess combining key age 81 years; 75% women; diet scores had significant 53 and 35% reductions in
effects of the 4.5 years of follow-up; 144 cases the rate of developing Alzheimers’ disease compared
MedDiet and DASH of Alzheimers’ disease with participants in the lowest tertile (Fig. 18.12).
diet on Alzheimers’ (multivariate adjusted) These data suggest that even modest adherence to
disease incidence the MIND diet score may help substantially in the
with the MIND diet; prevention of Alzheimers’ disease. By contrast, only
MAP; US) [41] the highest adherence to the DASH and MedDiet
diets were associated with Alzheimers’ disease
prevention
Nordic Diet – Prospective Cohort Study
Mannikko et al. 1140 women and men; mean A higher Nordic diet score was positively associated
(2015). baseline age 66 years and BMI 27; with verbal fluency (p = .039) and word list learning
Estimate the 4 years of follow-up (multivariate (p = .022) but better global cognitive performance
cross-sectional and adjusted) and lower Alzheimer’s disease related cognitive
longitudinal The Nordic diet is characterized indicators were only significant, after excluding
associations of the by a wide selection of berries, root individuals with impaired cognition at baseline
Nordic diet with vegetables, whole-grain products
cognitive function (primarily rye, oat and barley
(Sweden and Finland) being eaten in bread and porridge)
[42] with higher fiber contents;
rapeseed (canola) oil with oleic
acid and essential fatty acids
linoleic acid and α-linolenic acid
at 2- and 20-fold higher amounts,
respectively, compared with olive
oil used in the MedDiet
62.5% of the studies. A 2017 dose-response anal- that 13 of these studies showed that higher adher-
ysis of the MedDiet on cognitive function (9 ence to a MedDiet either slowed the rate of cogni-
cohort studies; 34,168 participants; baseline tive decline, minimized the conversion to
age ≥ 45 years; 2–12 years of follow-up) showed Alzheimer’s disease or improved the cognitive
that the highest MedDiet scores were inversely function [23]. The specific cognitive domains
associated with the developing of cognitive disor- benefiting from higher MedDiet score were mem-
ders by 21% compared to low scores; subgroup ory (delayed recognition, long-term, and working
analysis showed that the MedDiet was inversely memory), executive function, and visual con-
associated with mild cognitive impairment by structs. Another 2016 analysis (32 studies (5
17% and Alzheimer’s disease by 40% (Fig. 18.6) RCTs and 27 observational studies) showed that
[22]. A 2016 analysis of the effects of the MedDiet 21 of the 32 studies found that the MedDiet was
on cognitive processes and Alzheimer’s disease associated with improved cognitive function, a
(18 longitudinal and prospective studies) found decreased risk of cognitive impairment or
Mediterranean Diet (MedDiet)

0
Total Cognitive Mild Cognitive Alzheimer’s Disease
–5 Disorders (p = .0003) Impairment (p = .001) (p = .0001)
–10
Risk Reduction (%)
–15
–20
–25
–30
–35
–40
–45
Fig. 18.6 Association between high adherence to the MedDiet and risk of age-related cognitive decline and dementia
in adults (≥ 45 years; highest vs. lowest adherence) from a metaanalysis of 9 cohort studies (adapted from [22])
decreased risk of dementia or Alzheimer’s disease 18.3.1.2 Prospective Cohort Studies

[24]. Although there was large heterogeneity and Six prospective studies summarize the range of
differences in study quality, the authors concluded effects of the MedDiet on age related cognitive per-
that higher adherence to the MedDiet is associ- formance [15, 28–32]. The Australian Imaging,
ated with better cognitive performance. A 2014 Biomarkers and Lifestyle Study of Ageing (527
analysis (6 cohorts; 8019 subjects, mean baseline subjects; mean baseline age 69 years; 40% male;
age mid-70s; cognitively normal; follow-up 3-years follow-up; 3 diets: MedDiet, prudent and
4–8 years) found that the highest MedDiet tertile Western diets) showed that higher baseline adher-
had significantly lower adjusted risk of cognitive ence to the MedDiet was associated with less
impairment by 33% as compared to the lowest decline in the executive function composite score
tertile [25]. Among cognitively normal individu- in apolipoprotein E (APO-E) ε4 allele carriers. By
als, higher adherence to the MedDiet was associ- contrast, higher baseline adherence to the Western
ated with significantly reduced risk of mild diet score was associated with increased cognitive
cognitive impairment by 27% and Alzheimer’s decline in the visuospatial functioning composite
disease by 36%. A 2013 analysis (12 studies: 7 in APO-E ε4 allele non-carriers [15]. Overall, this
cohort, 1 longitudinal, 1 nested case-control, 3 study shows that adherence to the MedDiet is
cross-sectional studies and 1 single blind RCT; important to reduce risk for cognitive decline, espe-
25–3790 subjects; mean follow-up ranged from cially in executive function and visuospatial func-
10 days to 8 years) showed that higher adherence tioning. A 2015 Spanish study (823 participants;
to the MedDiet was associated with better cogni- mean baseline age 62 years; 2 year changes; tele-
tive function, lower rates of cognitive decline, and phone interviews of cognitive status-modified cog-
reduced Alzheimer disease risk in 9 out of 12 nition assessment) found a faster rate of cognitive
studies, whereas results for mild cognitive impair- decline with lower to moderate adherence to the
ment were inconsistent [26]. A 2010 analysis (18 MedDiet than for those with high adherence [28].
cohorts; 2,190,627 participants; follow-up for The 2014 US Memory and Aging Project (826 par-
4–20 years) demonstrated a 2-point increase in ticipants; mean baseline age 82.5 years; 74%
adherence to the MedDiet was associated with a women; 4.2 years of follow-up) found that each
significant reduction in neurodegenerative dis- 1-unit-higher MedDiet score was associated with a
eases by 13% [27]. significantly slower rate of global cognitive decline
by 0.002 standardized units [29]. A 2013 Nurses’ improved global cognitive performance and verbal
Health Study (16,058 women; mean baseline age memory scores and greater whole-grain intake was
74 years; 6 years of follow-up) showed that each associated with significantly improved global cog-
higher quintile of MedDiet score was linearly asso- nitive performance [32].
ciated with better mean global cognition and verbal
memory (multivariable- adjusted), declines in 18.3.1.3 Randomized Controlled
which are strong early predictors of Alzheimer dis- Trials
ease at older age (Fig. 18.7) [30]. The 2013 US Three Spanish RCTs demonstrated consistent pro-
Cache County Study on Memory, Health and tective effects of the MedDiet on age related cog-
Aging (3831 men and women; mean baseline age nitive performance in older subjects with high
74 years and BMI 26; 11 years of follow-up; cogni- CVD risk [33–36]. A 2015 PREDIMED trial (447
tive function was assessed by using the modified cognitively healthy volunteers with high CVD
mini-mental state examination) found that higher risk; mean baseline age 67 years; 51% women;
MedDiet scores were associated with higher aver- randomly assigned to a MedDiet supplemented
age cognitive function by 0.94 units (p = .001) [31]. with extra virgin olive oil or (one L/week) or
Also, higher intakes of whole grains, nuts, and mixed nuts (30 g/day) vs. a control diet (advice to
legumes were also associated with higher average reduce dietary fat); median of 4.1 years) found that
cognitive function scores by approximately 20%. both MedDiets were associated with significantly
A 2013 Women’s Health Study (6174 women; improved cognitive function compared to the
baseline age 72 years; 5 years of follow-up) showed lower-fat control [33]. Also, a 2013 PREDIMED
that higher adherence to the alternative MedDiet, a trial from Navarra, Spain (522 participants at high
simplifıed guidance based on the intake of vegeta- vascular risk; mean baseline age 75 years; 45%
bles, fruits, legumes, whole-grains, nuts, fish, red men; 6.5 years of follow-up) showed that MedDiets
meat, moderate alcohol and higher ratio of mono- supplemented with extra virgin olive oil or mixed
unsaturated (MUFA)-to-saturated fats was not sig- nuts significantly enhanced cognitive function as
nificantly associated with better global cognition or measured by the Mini-Mental State Exam and
verbal memory scores [32]. However, specifically clock draw test vs. a low-fat diet [34]. However,
higher MUFA to saturated fat ratio significantly another 2013 PREDIMED-Navarra analysis (285
Verbal Memory Score (p-trend <.001) Global Cognitive Score (p-trend =.002)
0.07
0.06
Better Cognitive Performance
0.05
0.04
0.03
0.02
0.01
0
1 2 3 4 5
MedDiet Adherence (Quintile)
Fig. 18.7 Association between cognitive performance and Mediterranean diet (MedDiet) adherence in women from
the Nurses’ Health Study (mean baseline age 74 years; 6 years of follow-up) (adapted from [30])
high vascular risk participants; 45% men; mean 67 years; 52% women) found that MedDiet adher-
baseline age 74 years: 6.5 years) found that only ence with increasing consumption of polyphenol
the MedDiet plus extra virgin olive oil group sig- rich foods and beverages was associated with bet-
nificantly reduced mild cognitive impairment by ter cognitive performance [36]. Higher intakes of
66% compared to the lower fat control diet extra virgin olive oil, coffee, walnuts, and wine
(Fig. 18.8) [35]. Results from the participants were associated with better memory function and
assigned to MedDiet plus nuts group did not differ global cognition. The relationship between
from the control group. A cross-sectional analysis increasing urinary polyphenols excretion and cog-
of a PREDIMED trial (447 cognitively healthy nitive performance are summarized in Figs. 18.9
volunteers with high CVD risk; mean baseline age and 18.10.
1.2
1
Odds Ratio for MCI
0.8
0.6
0.4
0.2
0
Low-Fat Diet MedDiet plus tree nuts MedDiet + extra virgin
(p =.226) olive oil (p =.044)
Fig. 18.8 Risk of mild cognitive impairment (MCI) after 6.5 years of two MedDiets (supplemented with 30 g mixed
tree nuts/day or 1 liter extra virgin olive oil/week) vs. guidance for a low-fat diet (PREDIMED-Navarra RCT) (adapted
from [35])
41.5
41
Immediate Memory Recall Score
40.5
40
39.5
39
38.5
38
37.5
37
36.5
1 2 3 4 5
Urinary Polyphenol Excretion Concentration Quintile
Fig. 18.9 Association between Mediterranean diet (MedDiet) adherence as measured by urinary polyphenol excretion
and immediate memory recall score (p-trend = .018) (adapted from [36])
7.5
Delayed Memory Recall Score

6.5
5.5
4.5
4
1 2 3 4 5
Urinary Polyphenol Excretion Concentration Quintile
Fig. 18.10 Association between Mediterranean diet (MedDiet) adherence as measured by urinary polyphenol excre-
tion and delayed memory recall score (p-trend = .003) (adapted from [36])
18.3.2 Dietary Approaches to Stop scores by approximately 20%. However, a 2016

Hypertension (DASH) Diets German study (6425 postmenopausal women;
baseline age 65–79; 9.1 years of follow-up) found
Several prospective cohort studies generally show that higher adherence to the DASH diet was not
that higher adherence to the DASH diet improves significantly associated with less cognitive decline
cognitive function in older adults [29, 31, 37, 38]. especially in women with hypertension [38].
A 2017 The Nurses’ Health Study (16,144 women;
mean baseline age 74 years; 6 years of follow-up)
found that high adherence to the DASH diet was 18.3.3 Mediterranean-DASH Diet
not associated with a decline in cognitive function Intervention for
over 6 years and suggested the equivalent to being Neurodegenerative Delay
1 year younger in age [37]. Also, greater adherence (MIND)
to the DASH diet was associated with better cogni-
tive function irrespective of apolipoprotein E ε4 The MIND diet score combines the cognitive
allele status. The 2014 US Memory and Aging promoting dietary components from both the
Project (826 participants; mean baseline age MedDiet and DASH diet, with emphasizes on
81.5 years; 26% men; 4.1 years of follow-up) brain healthy foods including green leafy vegeta-
found that a 1-unit increase in DASH score was bles, other vegetables, nuts, berries, beans, whole
associated with a slower rate of global cognitive grains, seafood, poultry, extra virgin olive oil and
decline by 0.007 standardized units (p = .03) [29]. red wine or high polyphenolic 100% juices (in
The US Cache County Study on Memory, Health moderation) and the restriction of unhealthy foods
and Aging (3831 men and women; mean baseline (red meats, stick margarine, pastries and sweets,
age 74 years and BMI 26; 11 years of follow-up; and fried/fast foods) [20]. Several studies support
cognitive function by modified mini-mental state the MIND diet as being one of the most effective
examination) showed that those participants with dietary patterns for promoting better age related
the highest DASH diet averaged 0.97 points higher cognitive performance and reducing the risk of
in cognitive function than those in the lowest dementia such a Alzheimer’s disease [39–41].
DASH diet quintile (p = .001) [31]. Higher intakes The Health and Retirement Study, a large US pop-
of whole grains, nuts, and legumes were also asso- ulation-based cross-sectional study (5907 partici-
ciated with higher average cognitive function pants; mean age 68 years) showed that high
adherence scores to both the MIND diet and the age; 75% women; 4.7 years of follow-up) showed
MedDiet were independently associated with sig- that participants in the top tertile of MIND diet
nificantly better cognitive function in a dose scores vs. the lowest tertile had slower rates of
response manner (p < .001, all) [39]. The partici- cognitive decline equivalent to being 7.5 years
pants in this study with high dietary adherence younger in cognitive age (Fig. 18.11) [40]. For
were 35% less likely to have poor cognitive per- Alzheimer’s disease, a MIND study (923 partici-
formance. Also, two 2015 US Memory and Aging pants; average baseline age 81 years; 75% women;
Project prospective studies support the effective- 4.5 years of follow-up) found that participants in
ness of the MIND diet on slowing cognitive the top and middle tertile of MIND diet scores had
decline and delaying Alzheimers disease [39, 40]. significant 53 and 35% reductions in the rate of
For slowing cognitive decline, a MIND diet study developing Alzheimers disease compared with
(960 participants; average baseline 81 years of participants in the lowest tertile (Fig. 18.12) [41].
Baseline 4.7 years

0.4
0.3
0.2
Global Cognitive Score
0.1
0
MIND Score High MIND Score Middle MIND Score Low
–0.1
–0.2
–0.3
–0.4
Fig. 18.11 Change in global cognitive score as a function of MIND diet score over 4.7 years with an average baseline
age of 81 years (p = .01) (adapted from [40])
MIND Diet Score (p-trend =.002) DASH Diet Score (p-trend =.07)
MedDiet Score (p-trend =.006)
1.1
Hazard Ratio for Alzheimer's Disease
0.9
0.8
0.7
0.6
0.5
0.4
1 2 3
Dietary Score (Tertile)
Fig. 18.12 Risk of Alzheimer’s disease as a function of dietary pattern over 4.5 years w ith an average baseline age of
81 years (adapted from [41])
18.4 Foods and Beverages 489
18.3.4 Nordic Diets thereafter, whereas the associations were linear

for mushrooms. Individual plant foods, carrots,
The Nordic diet is characterized by a wide selec- cruciferous vegetables, citrus fruits and high-
tion of berries, root vegetables, and whole-grain fiber bread were most effective at improving cog-
products (primarily rye, oat and barley) being nitive performance whereas high intake of white
eaten in bread and porridge (i.e. products that bread was negatively associated with cognitive
have high fiber contents). The vegetable oil used function. Tables 18.3 and 18.4 provide summa-
is rapeseed (canola), which is high in unsaturated ries of the effects of whole (minimally processed)
fatty acids including oleic acid and essential fatty plant foods, and soy and dairy products on cogni-
acids linoleic acid and α-linolenic acid which are tive performance. The composition of whole
2- and 20 times higher, respectively, compared plant foods is summarized in Appendix B.
with olive oil used in the MedDiet [42]. A longi-
tudinal assessment of the Nordic diet (1140 18.4.1.1 Fruits and Vegetables
women and men from Sweden and Finland; mean Two systematic reviews, 5 prospective cohort stud-
baseline age 66 years; 4 years of follow-up) ies, and 2 RCTs summarize the range of positive
showed that a higher Nordic diet score was posi- effects of increased fruit and vegetable intake on
tively associated with verbal fluency (p = .039) age-related cognitive performance (Table 18.3)
and word list learning (p = .022) but only better [45–53]. Fruit and vegetable sources with higher
global cognitive performance and lower levels of nutrient and phytochemical antioxidants
Alzheimer’s disease related cognitive indicators tend to be more effective in attenuating brain oxi-
were significant, after excluding individuals with dative or inflammatory stress or elevating serum
impaired cognition at baseline [42]. levels of brain-derived neurotrophic factor.
Systematic Reviews
18.4 Foods and Beverages Two systematic reviews support the protective
effect of fruit and vegetable intake on improving
18.4.1 Whole Plant Foods and age related cognitive performance [45, 46]. A
Dairy Products 2014 systematic review found statistically signifi-
cant benefits of fruits, vegetables, or 100% juice
Aging brain protective foods include higher consumption for improved cognitive function in
intake of vegetables, fruit, whole grains, fish, low older adults reported in 17 of 19 observational
fat dairy products and nuts, and lower intake of studies and 3 of 6 intervention trials [45]. However,
sweets, fried potatoes, processed meat, high-fat there was a high degree of variability in cognitive
dairies and butter [43]. A Western Norway cross- effects depending on the type of fruit, vegetable or
sectional study (2031 older subjects; aged juice consumed. A 2012 systematic review
70–74 years; 55% women; extensive cognitive (6 cohort studies; ≥ 6 months of follow- up)
testing; completed a comprehensive validated showed in 5 of 6 studies that higher consumption
and self-reported food frequency questionnaire) of vegetables, but not fruit, was associated with a
found that participants with higher intakes of decreased risk of cognitive decline or dementia
fruits, vegetables, whole grain products and [46]. In these studies, the vegetables most associ-
mushrooms performed significantly better in ated with slower cognitive decline included cruci-
cognitive tests than those with very low or no ferous vegetables, legumes, and green leafy
intake [44]. The associations were strongest vegetables, particularly cabbage, zucchini, squash,
between cognition and the combined intake of broccoli, and lettuce, at a daily intake of 3 servings
fruits and vegetables, with a marked dose- (200 g). The effect of vegetables may be associ-
dependent relationship up to about 500 g/day. ated with the fact that people frequently consume
The dose-related increase of intakes of whole vegetables with added healthy oils, which aid in
grain products and potatoes reached a plateau at absorption of fat soluble antioxidants such as
about 100–150 g/day, levelling off or decreasing vitamins A and E, and carotenoids.
Table 18.3 Summaries of fruit, vegetable, whole-grain and nut studies in age-related cognitive performance
Fruits and Vegetables
Systematic Reviews
Lamport et al. 19 observational studies and 6 17 observational studies and 3 intervention studies
al. (2014). intervention studies reported significant benefits of fruit, vegetable, or juice
Summarize the consumption for cognitive performance. The data suggest
association that chronic intake of fruits, vegetables, and juices are
between beneficial for cognition in healthy older adults. However,
polyphenol intake there was a high degree of variability in cognitive effects
from fruit, depending on the type of fruit, vegetable or juice
vegetable, and consumed
juice consumption
and cognition [45]
Loef and Walach 6 cohort studies on fruits and Five of the 6 studies that analyzed fruit and vegetable
(2012). vegetables with a follow-up of consumption separately found that higher consumption
Summarize the 6 months or longer of vegetables, but not fruit, was associated with a
effects of fruit decreased risk of cognitive decline or dementia. In these
and vegetable studies, the vegetables most associated with slower
intake on age cognitive decline included cruciferous vegetables,
related cognitive legumes, and green leafy vegetables, particularly
function [46] cabbage, zucchini, squash, broccoli, and lettuce, at a
daily intake of 3 servings (200 g) a day. The authors
suggest that these beneficial effects might be due to
higher intake of flavonoids or antioxidants in both fruits
and vegetables, or increased vitamin E in vegetables,
compared to fruits, which have more vitamin
C. Furthermore, people frequently consume vegetables
with added fats (e.g., oils) which may aid in absorption
of nutrients/phytochemicals
Devore et al. 16,010 participants, mean baseline Greater intakes of blueberries and strawberries were
(2012). age 74 years; follow-up associated with slower rates of cognitive decline. For
Evaluate the assessments were conducted blueberries, high intake improved mean global score
effect of berries twice, at two-year intervals for (averaging six cognitive tests) by 0.04 units (p-trend
and flavonoids on 4 years of follow-up (multivariate = .014 (Fig. 18.13). For strawberries, high intake
cognitive decline adjusted) improved mean global cognitive score by 0.03 units
(The Nurses’ (p-trend = .022). Berries delayed cognitive aging by
Health Study; 1.5–2.5 years of age. Greater intakes of anthocyanidins
US) [47] and total flavonoids were associated with slower rates of
cognitive decline (p-trends = .015 and .053)
Chen et al. 6911 illiterate participants; Lower intakes of vegetables and legumes were associated
(2012). baseline age ≥ 65; 3 years of with cognitive decline among illiterate elderly Chinese.
Investigate the follow-up (multivariate adjusted) Intake of vegetables and legumes were inversely
association associated with cognitive decline with significantly lower
between dietary risk of cognitive decline for high intake of vegetables by
habits and 34% and legumes by 22%
declines in
cognitive function
among Chinese
illiterate elderly
(Chinese
Longitudinal
Health Longevity
Study) [48]
Nooyens et al. 2613 men and women; baseline Higher vegetable intake was associated with smaller
(2011). age 43–70 years (mean 55 years); decline in information processing speed (p < .01) and
Evaluate the examined for cognitive function global cognitive function (p = .02) over 5 years. High
effect of habitual twice over a 5-year follow-up intakes of some subgroups of vegetables (i.e. cabbage
fruit and (multivariate adjusted) and root vegetables such as carrots, red beets,
vegetable intake mushrooms) were associated with a smaller decline in
during mid-age cognitive function. Total intakes of fruits, legumes and
on cognitive juices were not associated with change in cognitive
function function
(Doetinchem
Cohort Study;
The Netherlands)
[49]
Peneau et al. 2533 subjects; baseline age Higher intakes of fruit and vitamin C–rich fruits and
(2011). 45–60 years; mean age at vegetables were associated with better verbal memory. In
Examine the evaluation 66 years; 13 years of contrast, higher intakes of vegetables, and β-carotene-
association follow-up (multivariate adjusted) rich fruits and vegetables were associated with poorer
between fruit and executive function. Further research is required to better
vegetable intake understand the complex associations between different
and cognitive groups of fruits and vegetables on specific elements of
performance in a age related cognitive function
sample of adults
(Supplementation
with Antioxidant
Vitamins and
Minerals 2;
France) [50]
Kang et al. 13,388 women; mean baseline age Specific vegetable intake was significantly associated
(2005). 67 years; 6 years of follow-up with a lower rate of cognitive decline. For the highest
Examine the (multivariate adjusted) quintile of cruciferous vegetables, the decline was slower
effect of fruit and by 0.04 units (p-trend = .01) compared to the lowest
vegetable intake intake. Women consuming the highest intake of green
on cognitive leafy vegetables had a slower decline by 0.05 units
function and (p-trend <.001). The cognitive improvement was equal to
decline in older 1–2 years of younger cognitive age
women (The
Nurses’ Health
Study; US) [51]
RCTs
Lee et al. (2016). Double-blinded, Parallel RCT: Polyphenol rich grapes increased metabolic activity in
Assess effects of 10 subjects with mild decline in the right superior parietal cortex and left inferior anterior
polyphenol rich cognition; mean age 72 years; temporal cortex which was correlated with improvements
grapes on 50% women; high polyphenol in attention/working memory, as measured with WAIS-III
regional cerebral grape formulation vs. placebo Digital Span (r = −0.69, p = .04). The placebo low
metabolism in grape formulation free of polyphenol grapes lowered metabolic activity in brain
older adults (US) polyphenols; 6 months; cognitive area of cognitive function especially the right posterior
[52] performance was measured cingulate cortex (p = .01), and left superior posterolateral
through neuropsychological temporal cortex metabolic activity (p = 0.04). The
assessments. Changes in brain placebo group also showed significant declines in the left
metabolism occurring with each prefrontal, cingulate, and left superior posterolateral
therapy regimen were assessed by temporal cortex (p < .01). Although there was no
brain PET scans with the significant difference in behavior cognitive performance,
radiotracer this study suggests a protective effect of polyphenol rich
grapes against early brain pathologic metabolic decline
(continued)
Neshatdousta Parallel RCT: High-flavonoid intake from fruits and vegetables induced
et al. (2016). 154 men and women; aged significant improvements in cognitive performance and
Investigate effect 26–70 years; intervention diet increases in serum brain-derived neurotrophic factor
of fruit and averaged 3 portions of fruit and (BDNF) levels (p = <.001) compared to low flavonoid
vegetable vegetables per day to deliver fruits and vegetables habitual diets (Figs. 18.14 and
flavonoid levels high-or low-levels of flavonoids; 18.15)
on serum intake was increased by 2 portions Flavonoid rich fruits and vegetables include: apples,
brain-derived every 6 weeks; 2, 4 and 6 portions pears, berries, oranges, peppers, broccoli, onions,
neurotrophic were 49, 121 and 198 mg cabbage
factor (BDNF) flavonoids/day, respectively
and cognitive vs < 6 mg flavonoids/day (habitual
performance diet); 18-weeks
(UK) [53]
Whole-grains
Samieri et al. 6174 women; baseline age Greater whole-grain intake was associated with better
(2013). 72 years; 5 years of follow-up average global cognition (p-trend = .02)
Examine the
effect of
adherence to the
alternate MedDiet
on cognitive
function and
decline (Women’s
Health Study;
US) [35]
Wengreen et al. 3831 men and women; Mean Higher intakes of whole grains were associated with
(2013). baseline age 74 years and BMI 26; higher average cognitive function scores by approx. 20%
Evaluate cognitive function was assessed
associations by using the Modified Mini-
between DASH Mental State Examination 4 times
and MedDiet over 11 years
dietary patterns
and age-related
cognitive changes
in a prospective,
population-based
study (Cache
County Study on
Memory, Health
and Aging; US)
[37]
Nuts
O’Brien et al. 16,010 women; mean baseline age Women consuming ≥5 servings of nuts/week had higher
(2014). 74 years; 6 years of follow-up cognitive scores than non-consumers with a mean
Examine the (multivariate adjusted) improvement by 0.08 standard units (p-trend = .003).
effects of This mean difference of 0.08 is equivalent to a mean
long-term nuts 2-year improved cognitive age
intake on
cognition in older
women (The
Nurses’ Health
Study; US) [54]
RCT
Cardoso et al. Parallel RCT: Selenium levels were increased with Brazil nuts whereas
(2016). 31 older adults with mild the control group had no change. Among the parameters
Evaluate whether cognitive impairment (MCI); related to the antioxidant system, only erythrocyte GPx
Brazil nuts, rich mean age 78 years; 70% women; activity change was significantly different between the
in selenium, randomly assigned to ingestion of groups (p = .006). Improvements in verbal fluency
preserve Brazil nuts (290 μg selenium/day) (p = .007) and constructional praxis (the ability to build,
cognitive function or to the low selenium control; assemble, or draw objects) (p = .031) were significantly
in older adults via blood selenium concentrations, greater in the Brazil nuts group when compared with the
enhancement of erythrocyte glutathione peroxidase control group. These findings suggest that the intake of
selenoprotein (GPx) activity, oxygen radical Brazil nuts helps to correct selenium deficiency and
antioxidant absorbance capacity; 6 months provides preliminary evidence that Brazil nut
systems (Brazil) consumption can have positive effects on some cognitive
[55] functions of older adults with MCI
Valls-Pedret Parallel RCT: Multivariate analysis found that the MedDiet plus nuts
et al. (2015). 447 cognitively healthy adults at improved performance above the baseline in memory test
Examine the high CVD risk; mean baseline age (p = .04) whereas the subjects on MedDiet plus extra
effect of 67 years; 3 dietary interventions: virgin olive oil performed better in tests of frontal
MedDiets control (low-fat) diet, MedDiet (p < .003) and global cognition (p < .005). All cognitive
supplemented supplemented with virgin olive tests on the lower fat control diets were significantly
with nuts or olive oil, or MedDiet supplemented decreased (p < .05)
oil vs. lower fat with nuts; median follow-up
diets on 4.1 years
age-related
cognitive decline
(PREDIMED;
Spain) [33]
Sanchez-Villegas Parallel RCT: Participants assigned to MedDiet + nuts showed a
et al. (2011). 243 cognitively healthy adults at significantly lower risk by 78% of low plasma BDNF
Assess the role of high CVD risk; mean baseline age values (<13 μg/mL) as compared to the low-fat control
a MedDiet on 67 years; 3 dietary interventions: group. Among participants with depression at baseline,
plasma brain- control (low-fat) diet, MedDiet significantly higher BDNF levels were found for those
derived plus extra virgin olive oil, or assigned to the MedDiet plus nuts
neurotrophic MedDiet plus nuts. Plasma BDNF
factor (BDNF) levels were measured after 3 years
levels of intervention
(PREDIMED;
Spain) [56]
Prospective Cohort Studies decline. A 2005 Nurses’ Health Study (13,388

Four representative prospective studies illustrate women; mean baseline age 67 years; 6 years of
the effects of increased fruit and vegetable intake follow-up) showed that specific vegetables were
on age-related cognitive function [47–51]. A 2012 significantly associated with a lower rate of cog-
Nurses’ Health Study (16,010 participants, mean nitive decline [51]. For cruciferous and green
baseline age 74 years; 4 years of follow-up) found leafy vegetables, the highest quintile of intake sig-
that higher intake of berries was associated with nificantly reduced cognitive decline by 0.04 and
slower rates of cognitive decline [47]. For blue- 0.05 units, respectively, compared to the lowest
berries, high intake significantly improved mean intake. The Chinese Longitudinal Health
global score (averaging six cognitive tests) Longevity Study (6911 illiterate participants;
(Fig. 18.13) and slightly less similar findings were baseline age ≥ 65; 3 years of follow-up) found
observed for strawberries. Greater intakes of that the intake of vegetables and legumes was
anthocyanidins and total flavonoids were associ- inversely associated with cognitive decline with
ated with significantly slower rates of cognitive significant lower risk of cognitive decline for high
Table 18.4 Summary of soy food/supplements and dairy foods studies in age related cognitive performance
Soy Products
Meta-Analysis
Cheng et al. (2015). 10 placebo-controlled RCTs; 1024 Overall global cognitive function was
Quantify the effects of soy participants; 6 weeks to 30 months significantly improved by 0.08 units (p = .014).
isoflavone intake on Visual memory was significantly improved by
improving cognitive 0.10 units (p = .016). In subgroup analyses,
function in post- statistically significant improvements were
menopausal women [57] observed in non-US countries and with a mean
age younger than 60 years
RCT
St John et al. (2014). Placebo-controlled, double-blind Mean increased urine excretion of isoflavonoids
Determine effects of RCT: from baseline was not significantly associated with
change in urine excretion 350 healthy postmenopausal change in a composite score of global cognition.
of isoflavonoids on women; mean age 61 years; 25 g Secondary analyses indicated that change in urine
cognitive change of isoflavone-rich soy protein excretion of isoflavonoids was inversely associated
(Women’s Isoflavone Soy (91 mg of aglycone weight with general intelligence, but not with factor
Health Clinical Trial; US) isoflavones) or milk protein- scores representing verbal or visual episodic
[58] matched placebo, provided daily; memory. Postmenopausal women considering
mean 2.5 years long-term soy protein supplementation should
consider consuming less than 25 g soy protein or
91 mg of isoflavones per day
Henderson et al. (2012). Placebo-controlled, double-blind Women consuming soy protein or milk protein
Determine the cognitive RCT: placebo showed improved global cognitive
effects of long-term dietary 350 healthy postmenopausal composite scores. Soy protein improved global
soy isoflavones in a daily women; mean age 61 years; 25 g cognition from baseline by 58% whereas milk
dose comparable to that of of isoflavone-rich soy protein protein improved score by 69% but there was no
traditional Asian diets (91 mg of aglycone weight significant difference between the two groups.
(Women’s Isoflavone Soy isoflavones) or milk protein- Secondary analyses indicated that soy protein
Health Clinical Trial; US) matched placebo, provided daily; improved visual memory by 67% but there were
[59] mean 2.5 years no significant between-group differences on
other cognitive factors or individual test scores
Fournier et al. (2007). Double-blind, placebo-controlled Soy milk or soy isoflavone supplements over a
Investigate the effect of RCT: 16-week period did not improve or appreciably
soy isoflavones (soy milk 79 postmenopausal women; mean affect cognitive functioning in healthy,
and supplement) on age 56 years; diets: cow’s milk postmenopausal women compared to cow’s milk.
cognitive functioning in (control); soy milk, (72 mg Soy milk and supplements did not improve
healthy, postmenopausal isoflavones/day); isoflavone selective attention (Stroop task), visual long-term
women (US) [60] supplement (70 mg isoflavones/ memory (pattern recognition), short-term
day); placebo supplement; visuospatial memory (Benton Visual Retention
16 weeks Test), or visuo-spatial working memory (color
match task). Also, the soy milk group showed a
decline in verbal working memory (Digit
Ordering Task) compared to cow’s milk groups
Dairy Products
Systematic Review
Crichton et al. (2010). 3 cross-sectional and 5 prospective Poorer cognitive performance and an increased
Systematic review to studies; 449–4809 subjects; risk for vascular dementia were found to be
assess evidence for an subjects >60 years in prospective associated with a lower consumption of milk or
association between dairy studies dairy products. However, the consumption of
intake and cognitive whole-fat dairy products may be associated
functioning [61] with cognitive decline in the elderly. No
significant associations were found between
milk and yogurt, or cheese consumption and
cognitive decline
Prospective Cohort Studies and RCTs
Kesse-Guyot et al. (2016). 3076 participants from the general Higher yogurt consumption was significantly
Examine the associations population; mean age 66 years; associated with better verbal memory
of total and specific dairy cognitive assessment; 13 years of performance
product intake with follow-up; dairy product
cognitive performance in consumption was estimated using
aging adults 10 × 24 h recalls
(Supplémentation en
Vitamines et Minéraux
Antioxydants (SU.
VI. MAX) RCT and the
SU.VI.MAX 2
observational follow-up
study; France) [62]
Crichton et al. (2012). Crossover RCT: Spatial working memory performance was
Determine the effect of 38 participants with habitually low improved by 8% (p = .046) following the high
reduced fat dairy on dairy intakes (<2 servings/day); dairy diet compared with the low dairy diet.
cognitive performance 60% women; mean age 52 years; Spatial working memory requires simultaneous
(Australia) [63] mean BMI 31.5; randomized to 4 storing and manipulating visual-spatial
vs. one servings/day of reduced fat information mediated by the prefrontal cortex
dairy foods; 6 months
Global Score (p-trend =.014)

Verbal Memory Score (p-trend =.022)
Telephone Interview of Cognitive Status (p-trend =.022)
0.18
0.16
Mean Cognitive Improvement
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
< 1 serving/mo 1-3 servings/mo at least 1 serving/week
Blueberry Intake Frequency
Fig. 18.13 Association between blueberry intake frequency and cognitive function in older women (baseline age
74 years) over 4 years from the Nurses’ Health Study (adapted from [47])
intake of vegetables by 34% and legumes by 22% in information processing speed and global cogni-
[48]. The Dutch Doetinchem Cohort Study (2613 tive function [49]. A French cohort study (2533
adults; mean age 55 years; 5 years of follow-up) subjects; mean age at evaluation 66 years; 13 years
reported that higher vegetable intake, especially of follow-up) found that higher intake of fruit and
from root vegetables, such as carrots, red beets vitamin C rich fruits and vegetables was associ-
and mushrooms, significantly slowed the decline ated with better verbal memory [50].
Randomized Controlled Trials (RCTs) (Table 18.3) [35, 37]. The Women’s Health Study
Two RCTs show the effects of flavonoid-rich fruits (6174 women; baseline age 72 years; 5 years of
and vegetables on age related cognitive performance follow-up) showed that greater whole-grain
with aging [52, 53]. A 2017 RCT (10 subjects with intake was associated with better average global
mild decline in cognition; mean age 72 years; 50% cognition (p-trend = 0.02) [35]. The Cache
women; polyphenol-rich grapes vs. placebo low County Study on Memory, Health and Aging
polyphenol grapes; 6 months; brain PET scans with (3831 men and women; mean baseline age
the radiotracer) found that the polyphenol rich 74 years and BMI 26; cognitive function assessed
grapes increased metabolic activity in the right supe- over 11 years) found that higher intakes of whole
rior parietal cortex and left inferior anterior temporal grains were associated with higher average cog-
cortex which was correlated with improvements in nitive function scores by approx. 20% [37].
attention/working memory whereas subjects con-
suming the low polyphenol grape placebo had sig- 18.4.1.3 Nuts
nificant metabolic decline in brain areas of cognitive One prospective cohort study and three RCTs
interest [52]. This study suggests a protective effect show the effects of increased nut intake on
of grapes against early pathologic brain metabolic improving age related cognitive performance
decline. A 2016 RCT (154 adults; aged 20–70 years; (Table 18.3) [33, 54–56]. A 2014 Nurses’ Health
18 weeks) found that high flavonoid fruits and veg- Study (15,467 women; mean age 74 years;
etables such as berries, oranges, apples, pears, pep- 15–21 years of follow-up) found that higher total
pers, or broccoli significantly improved cognitive nut intake was associated with better average cog-
performance and serum brain-derived neurotrophic nitive status for all cognitive attributes measured
factor compared to a usual low fruit and vegetable [54]. Women consuming ≥5 servings of nuts/
diet or low flavonoid fruits and vegetables week had significantly higher cognitive scores
(Figs. 18.14 and 18.15) [53]. equivalent to a 2-year improved cognitive age vs.
non-consumers. For Brazil nuts, a 2016 RCT (31
18.4.1.2 Whole-Grains older adults with mild cognitive impairment
There are only a limited number of studies on the (MCI); mean age 78 years; 70% women; randomly
effects of whole grains on cognitive performance assigned to ingestion of Brazil nuts (290 μg sele-
Control Low Flavonoid F/V High Flavonoid F/V

0.25
0.2
0.15
Cognitive Function (z-score)
0.1
0.05
0
Baseline 6 weeks 12 weeks 18 weeks
–0.05
–0.1
–0.15
–0.2
Fig. 18.14 Effect of fruits and vegetables (F/V) flavonoid level on cognitive function after 18 weeks from a RCT with
154 adults (p < 0.001 for high flavonoid F/V at 12 and 18 weeks) (adapted from [53])
Control Low Flavonoids F/V High Flavonoid F/V

450
400
350
300
BDNF (pg/ml)
250
200
150
100
50
0
Baseline 6 12 18
Weeks
Fig. 18.15 Effect of fruits and vegetables (F/V) flavonoid level on serum brain-derived neurotrophic factor (BDNF)
concentration over 18 weeks from a randomized controlled trial with 154 adults (p < .001 for high flavonoid F/V after
18 weeks) (adapted from [53])
nium/day) or to the low selenium control; of life and cognitive performance in post-
6 months) found that Brazil nuts significantly menopausal women [57–60]. A 2015 meta-
improved verbal fluency and constructional praxis analysis (10 placebo-controlled RCTs; 1024
(the ability to build, assemble, or draw objects) participants; 6 weeks to 30 months) found that
compared with the control group [55]. These find- soy protein significantly improved global cogni-
ings support a possible role for Brazil nuts in cor- tive performance by 0.08 units and visual mem-
recting selenium deficiency and preserving ory by 0.10 units. In subgroup analyses, the
cognitive function in older adults via its enhance- statistically significant improvements were
ment of selenoprotein antioxidant systems. Two observed for non-US countries and mean age
Spanish PREDIMED RCTs (243 and 447 cogni- younger than 60 years [57]. The 2014 Women’s
tive healthy adults at high risk of CVD; mean Isoflavone Soy Health Clinical Trial (350 healthy
baseline age 67 years; 30 g mixed nuts/day; postmenopausal women; mean baseline age
3–4.1 years) found that the MedDiet plus nuts sig- 61 years; 25 g of soy protein or milk protein-
nificantly improved memory and increased levels matched placebo, provided daily; 2.5 years)
of plasma brain derived neurotrophic factor, found that women with increased urine excretion
known to improve brain function compared to of isoflavonoids from baseline showed an insig-
lower fat control diets [33, 56]. nificant trend for better composite score of global
cognition [58]. However, a secondary analysis
18.4.1.4 Soy vs. Milk Protein suggests that higher urinary excretion of isoflavo-
Table 18.4 summarizes a meta-analysis, system- noids was associated with lower general intelli-
atic review and key studies on the effects of soy gence, which needs to be confirmed. A 2012
and milk protein or foods on cognitive perfor- Women’s Isoflavone Soy Health Clinical Trial
mance [57–63]. found that both soy protein and the milk protein
placebo were associated with similarly improved
Isolated Soy Protein global cognitive composite scores from baseline
Isoflavone-rich isolated soy protein (soy protein) levels [59]. Specifically, soy protein improved
intake is often considered for improving quality global cognition from baseline by 58% whereas
milk protein improved scores by 69% but there Minéraux Antioxydants Study (3076 participants
was no significant difference between the two from the general population; mean age 66 years;
groups. Secondary analyses indicated that soy cognitive assessment; 13 years of follow-up)
protein improved visual memory by 67% but found no association between total or specific
there were no significant between-group differ- dairy product consumption and working memory
ences on other cognitive factors or individual test [62]. However, higher yogurt consumption was
scores. A 2007 RCT (79 postmenopausal women; significantly associated with better verbal mem-
mean age 56 years; diets: soy milk (soy milk, ory performance. Higher cheese consumption
72 mg isoflavones/day), cow’s milk, and isofla- was also significantly associated with better ver-
vone supplement (isoflavone supplement, 70 mg bal memory performance but this association did
isoflavones/day); 16 weeks) showed that soy not remain significant after adjustment. A 2012
milk or soy isoflavone supplements over a Australian study (38 participants with habitually
16-week period did not improve or appreciably low dairy intakes (<2 servings/day); 11 men and
affect cognitive functioning in healthy, post- 27 women; mean age 52 years; mean BMI 31.5;
menopausal women compared to cow’s milk randomised to a high (4 servings/day) or low (1
[60]. Also, the soy milk group showed a decline servings/day) intake of reduced fat dairy;
in verbal working memory compared to the cow’s 6 months each diet) showed that spatial working
milk group. memory performance was significantly improved
by 8% following the high dairy diet compared
Dairy Products with the low dairy diet [63]. Spatial working
Dairy products are shown to have variable effects memory requires simultaneously storing and
on age related cognitive performance [61–63]. A manipulating visual-spatial information mediated
2010 systematic review (3 cross-sectional and 5 by the prefrontal cortex.
prospective studies; 449–4809 subjects; subjects
>60 years in prospective studies) found that
poorer cognitive performance and an increased 18.4.2 Beverages
risk for vascular dementia were found to be asso-
ciated with a lower consumption of milk or dairy Table 18.5 summarizes the studies associated
products [61]. However, the consumption of with polyphenol-rich beverages and alcohol on
whole-fat dairy products may be associated with age-related cognitive performance and dementia
cognitive decline in the elderly. No significant [64–83]. Poor cardiovascular health such as
associations were found between milk and yogurt, endothelial dysfunction or hypertension plays an
or cheese consumption and cognitive decline. A important role in the development and progression
2016 French Supplémentation en Vitamines et of age related cognitive decline [5–9]. Over the
Table 18.5 Summaries of common polyphenol rich beverages studies in age-related cognitive performance
100% Fruit and Vegetable Juices
Dai et al. (2006). 1836 Japanese Americans in King After adjustment for potential confounders,
Assess the effects of County, Washington, who were participants who drank juices ≥3 times per week
consumption of fruit and dementia-free; mean baseline age had a significantly 76% lower risk for AD
vegetable juices, as rich 72 years; 54% women; 10 years of compared with those who drank juice < once per
sources of polyphenols, follow-up; >80% of subjects drank week. Also, participants who drank juices 1–2
in Alzheimer’s disease tea at least once per week; 65% times per week had a 16% lower risk of AD. This
(AD) protection (Kame drank 100% fruit or vegetable inverse association tended to be more pronounced
Project cohort US) [64] juices at least once per week; and among those with an apolipoprotein Eε-4 allele
20% of subjects possessed one or and those who were not physically active.
more ApoE ε-4 alleles Conversely, no association was observed for
dietary intake of vitamins E, C, or β- carotene or
tea consumption
Kent et al. (2017). Single-Blind, Parallel RCT: Cherry juice improved cognitive function in older
Assess whether daily 49 subjects with mild-to-moderate adults with Alzheimer’s type dementia. Trends
consumption of dementia; mean age 80 years; 24 were evident, showing improvements in cognitive
anthocyanin-rich cherry women, 25 men; consumption of performance across all tasks with regular cherry
juice can change 200 mL/day of either cherry juice juice consumption. Significant improvements in
cognitive function in or control juice with negligible verbal fluency, short-term memory and long-term
older adults with anthocyanin content; 12-weeks memory were found in the cherry juice group.
dementia (Australia) [65] Also, cherry juice significantly reduced systolic
BP by 4.7 mm Hg compared to the control juice
Alharbi et al. (2016). Double-blind, Crossover RCT: Flavonoid rich orange juice was shown to acutely
Investigate if the 24 healthy men; mean age enhance objective and subjective cognition over
consumption of flavonoid 51 years; consumed 240-mL the course of 6 h in healthy middle-aged adults.
rich orange juice is flavonoid rich orange juice The effects on global cognitive performance are
associated with acute (272 mg flavonoids) and a placebo summarized in Fig. 18.16
cognitive benefits in control juice; 2 days separated by
healthy middle-aged men a 2-week washout; cognitive
(UK) [66] function and subjective mood were
assessed at baseline (prior to drink
consumption) and 2 and 6 h post
consumption
Lamport et al. (2016). Crossover RCT: CGJ was associated with better immediate spatial
Examine the effects of the 25 healthy mothers of preteen memory and 2 aspects of driving performance;
daily consumption of children who were employed for increased steering accuracy in combination with a
100% Concord grape ≥30 h/week; mean age 43 years; faster response time to changes in lead vehicle
juice (CGJ) for 12 weeks consumed 12 ounces (355 mL) of behavior during car following. The observed
on cognitive function, either CGJ (containing 777 mg effects would account for a reduction in stopping
driving performance, and total polyphenols) or an energy-, distance of approx 11 m at the speeds driven
blood pressure in healthy, taste-, and appearance-matched (40–60 mph; 64–97 km/h), which is an important
middle-aged working placebo; daily for 12 weeks; safety benefit
mothers (UK) [67] 4-week washout. Verbal and
spatial memory, executive
function, attention, blood pressure,
and mood were assessed at
baseline and at 6 and 12 weeks;
driving performance assessment in
a driving simulator
(continued)
Kean et al. (2015). Double-blind, Crossover RCT: In healthy older adults, the daily consumption of
Investigate the effects of 37 healthy older adults; mean age flavanoid-rich 100% orange juice significantly
flavonoid rich orange 67 years; high-flavanoid (305 mg) improved global cognitive and executive function
juice on cognitive 100% orange juice and an compared to a low-flavanoid control orange juice
function (US) [68] equicaloric low-flavanoid (37 mg) after 8 weeks
orange-flavored drink (500 mL)
were consumed daily; 8 weeks;
4 weeks washout
Bookheimer et al. Double-blind, Parallel RCT: Only the pomegranate group showed a significant
(2013). 32 subjects (28 completers); improvement in the Buschke selective reminding
Assess the effect of 100% randomly assigned to drink 8 test of verbal memory and a significant increase in
pomegranate juice on ounces of either 100% plasma trolox-equivalent antioxidant capacity and
cognitive performance pomegranate juice or a flavor- urolithin α-glucuronide. Also, compared to the
(US) [69] matched placebo drink; 4 weeks; placebo group, the pomegranate group had
memory testing, fMRI scans increased f-MRI activity during verbal and visual
during cognitive tasks, and blood memory tasks. These results suggest a role for
draws for peripheral biomarkers pomegranate juice in augmenting memory function
before and after the intervention via task-related increases in functional brain activity
Kelly et al. (2013). Double-blind, crossover RCT: Short-term dietary nitrate intake from beet juice
Evaluate the effects of 12 healthy adults; age range increased plasma nitrate by fourfold and
short-term consumption 60–70 years; supplemented their significantly reduced resting blood pressure in
of nitrate rich beet juice diet with either nitrate-rich normotensive older adults. These results suggest
on blood pressure, concentrated beetroot juice (2 × that beet juice may have potential in reducing the
oxygen uptake kinetics, 70 mL/day, 9.6 mmol/day nitrate) risk of hypertension and cardiovascular disease
and muscle and cognitive or a nitrate-depleted beetroot juice in older adults. The VO2 kinetics was
function in older adults placebo (2 ×70 mL/day, accelerated during treadmill walking, although
(US) [70] 0.01 mmol/day nitrates); 3 days this did not translate into significant enhanced
physical performance and cognitive function
Krikorian et al. (2010). Double-blind, Parallel RCT: Concord grape juice significantly improved
Assess the effect of 12 older adults with memory verbal learning (p = 0.04) with non-significant
supplementation with decline but not dementia: mean enhancement of verbal and spatial recall. There
100% Concord grape age 78 years; 8 men and 4 women; was no appreciable effect of the intervention on
juice on memory 444–621 mL of 100% Concord depression symptoms or weight or waist
performance in older grape juice/day; 12 week circumference. A small increase in fasting
adults with early insulin was observed in those that consumed
age-related memory grape juice. These preliminary findings suggest
decline (US) [71] that supplementation with Concord grape juice
may help to enhance cognitive function for older
adults with early memory decline
Coffee and Tea
Meta-Analysis and Systematic Review
Wu et al. (2017). 9 cohort studies; 34,282 Coffee consumption exhibited a J-shaped
Assess the effects of participants; baseline age > 60 association for incident cognitive disorders
coffee consumption on years; 1.3 to 28 years of follow-up (Alzheimers disease, dementia, cognitive decline
dose-response and impairment). Compared to <1 cup coffee/d,
development of cognitive daily consumption of 1–2 cups was associated
disorders [74] with a significant 18% lower risk of cognitive
disorders, whereas >3 cups of coffee were not
significantly different than <1 cup
Liu et al. (2017). 17 studies including 6 cohort, Tea consumption was inversely and linearly
Evaluate the dose- 3 case-control and 8 cross-over related to risk of cognitive disorders (dementia,
response relationship of studies; 48,435 participants; cognitive decline and impairment). High tea
tea consumption and baseline age 50 to 93 years intake was significantly associated with a
cognitive disorders [75] significant 27% reduction in cognitive disorder
risk with green tea lowering risk by 36%
compared to 25% for black tea. The dose response
showed a linearly reduced risk of cognitive
disorder for tea at 100 ml/d by 6%, at 300 ml/d by
19%, and at 500 ml/d by 29%
Gelber et al. (2011). 3494 men; mean baseline age Men with high coffee and caffeine intake in
Examine associations of 52 years; 25 years of follow-up; midlife were not associated with cognitive
coffee and caffeine intake 226 cases of dementia and 347 impairment, dementia, or neuropathologic lesions.
in midlife with risk of cases of cognitive impairment Also, men in the highest quartile of caffeine intake
dementia, its (multivariate adjusted) (≥411.0 mg/day) vs. the lowest quartile
neuropathologic (≤137.0 mg/day) had a 55% lower risk of any
correlates, and cognitive type of brain lesion (p-trend = .04
impairment (Honolulu-
Asia Aging Study; US)
[76]
Feng et al. (2010). 716 Chinese adults; mean age Total tea consumption was independently
Examine the relationship 65 years; 56% women associated with better performances on global
between tea consumption (multivariate adjusted) cognition, memory, executive function, and
and cognitive function in information processing speed. Both black/oolong
older adults (The and green tea consumption were associated with
Singapore Longitudinal better cognitive performance
Aging Studies) [77]
Ng et al. (2008). 1438 cognitively intact Total tea intake was significantly associated with a
Examine the relation participants in longitudinal lower prevalence of cognitive impairment,
between tea intake and analysis; 1–2 years of follow-up independent of other risk factors. Compared with
cognitive impairment and (multivariate adjusted) rare or no tea intake, the risk for cognitive
decline (The Singapore impairment was reduced for low intake by 44%,
Longitudinal Aging moderate intake by 55%, and high intake by 63%
Studies) [78] (p- trend <.001). For cognitive decline, the
corresponding risk reductions were 26%, 22% and
43% (p-trend = .042). These effects were most
evident for black (fermented) and oolong
(semi-fermented) teas, the predominant types
consumed by this cohort
Flavanol-rich Cocoa Beverages
RCTs
Lamport et al. (2015). Double-blind, Crossover RCT: Significant increases in regional perfusion across
Explore the effect of a 18 subjects; mean age 61 years; 10 the brain were observed following consumption of
single acute dose of men; low (23 mg) or high the high flavanol drink relative to the low flavanol
cocoa flavanols on (494 mg) 330 mL equicaloric drink, within 2 hours, particularly in the anterior
regional cerebral blood flavanol drinks matched for cingulate cortex and the central opercular cortex
flow (UK) [79] caffeine, theobromine, taste and of the parietal lobe. This increased cerebral blood
appearance; pre- and 2 h flow provides a possible acute mechanism by
post-consumption; arterial spin which cocoa flavanols benefit cognitive
labelling functional magnetic performance
resonance imaging
(continued)
Mastroiacovo et al. Double-blind, Parallel RCT: High cocoa flavanol significantly improved the
(2015). 90 older subjects; mean age overall composite cognitive function, z score.
Evaluate the effect of 70 years; daily a drink containing Higher cocoa flavanol consumption significantly
cocoa flavanol 993 mg [high cocoa flavanol improved blood pressure and reduced insulin
consumption on cognitive (HF)], 520 mg [intermediate cocoa resistance and lipid peroxidation which can be
performance in flavanol (IF)], or 48 mg [low linked to improved cognitive function. Reduction
cognitively intact elderly cocoa flavanol (LF)]; 8 weeks in insulin resistance explained 17% of
subjects (the Cocoa, improvements in composite cognitive
Cognition, and Aging performance z score
Study; Italy) [80]
Sorond et al. (2013). Double-blind, Parallel RCT: Regular high flavanol cocoa beverage intake
Investigate the 60 older subjects; aged 73 years; 2 promoted better neurovascular coupling and
relationship between cups of cocoa daily (flavanol-rich greater white matter structural integrity in
neurovascular coupling cocoa 609 mg, and flavanol-poor individuals with baseline impairments with
measured by Magnetic cocoa 13 mg flavanols/serving; improved neurovascular coupling by 5.6% vs.
resonance imaging (MRI) 30 days) −2.4% (p = .001)
and cognitive function in
elderly individuals with
vascular risk factors and
cocoa consumption
(US) [81]
last decade human intervention trials consistently tended to be more pronounced among those with
show the potential for a variety of beverages rich an apolipoprotein Eε-4 allele and those who were
in polyphenols to have protective effects on age- not physically active.
related cognitive performance and against the
development of neurodegenerative diseases via RCTs
improving cardiovascular health regulatory Seven blinded RCTs show the effects of 100%
mechanisms [64–79]. Alcohol has variable juices on age-related cognitive performance [65–
effects on cognitive functions depending on the 71]. A 2017 parallel RCT (49 subjects with mild-
level of intake and type of alcoholic drink to-
moderate dementia; mean age 80 years; 24
consumed [80–83]. women, 25 men; consumption of 200 mL/day of
either cherry juice or control juice; 12 weeks)
18.4.2.1 100% Juices found that cherry juice improved cognitive func-
tion in older adults with Alzheimer’s type demen-
Prospective Cohort Study tia [65]. Significant improvements in verbal
The US Kame Project cohort (1836 Japanese fluency, short-term memory and long-term mem-
Americans in King County, Washington, who ory were found in the cherry juice group, accom-
were dementia-free; mean baseline age 72 years; panied by a significant reduction in systolic BP by
54% women; 10 years of follow-up) showed that 4.7 mm Hg compared to the control juice. A 2016
participants who drank 100% fruit or vegetable UK crossover RCT (24 healthy men; mean age
juices ≥3 times per week had a significantly 76% 51 years; consumed a 240-mL flavonoid rich
lower risk for Alzheimer’s disease compared orange juice (272 mg flavonoids) and a placebo
with those who drank 100% juice < once per control juice; 2 days; separated by a 2-week wash-
week [64]. Also, participants who drank 100% out) showed that orange juice improved global
juices 1–2 times per week had a 16% lower risk cognition within 6 h after consumption (Fig. 18.16)
of Alzheimer’s disease. This inverse association [66]. A 2016 UK cross-over RCT (25 healthy
mothers of preteen children who were employed improvement in the Buschke selective reminding
for ≥30 h/week; mean age 43 years; 12 ounces test of verbal memory and a significant increase in
(355 mL) of either 100% Concord grape juice plasma trolox-equivalent antioxidant capacity and
(containing 777 mg total polyphenols) or an urolithin A-glucuronide [69]. Also, the pomegran-
energy-, taste-, and appearance-matched placebo; ate group had increased fMRI activity during ver-
daily for 12 weeks) found that Concord grape juice bal and visual memory tasks compared to the
was associated with better immediate spatial placebo group, suggesting a role for pomegranate
memory and driving performance including steer- juice in augmenting memory function via task-
ing accuracy in combination with a faster response related increases in functional brain activity. A
time to changes in lead following vehicle behavior 2013 crossover RCT (12 healthy adults; age range
[67]. Also, concord grape juice consumers had 60–70 years; supplemented their diet with either
improved car braking reaction time with a reduc- nitrate-rich concentrated beetroot juice (2 ×
tion in stopping distance of 11 m at the speeds 70 mL/day, 9.6 mmol/day nitrate) or a nitrate-
driven (40–60 mph; 64–97 km/h), which is an depleted beetroot juice placebo (2 × 70 mL/day,
important safety benefit. A 2015 crossover RCT 0.01 mmol/day nitrates); 3 days) showed that
(37 healthy adults; mean age 67 years; high-flavo- dietary nitrate intake from beet juice increased
noid (305 mg) 100% orange juice and an equica- plasma nitrate by 4-fold and significantly reduced
loric low-flavonoid (37 mg) orange-flavored drink resting blood pressure in normotensive older
(500 mL) were consumed daily; 8 weeks) showed adults but this did not translate into significant
that daily orange juice consumption significantly enhanced improvements in brain metabolism and
improved global cognitive and executive function cognitive function [70]. A 2010 parallel RCT (12
compared to the low-flavonoid control orange fla- older adults with memory decline but not demen-
vored drink after 8 weeks [68]. A 2013 parallel tia; mean age 78 years; 8 men and 4 women; 444–
RCT (32 subjects; randomly assigned to drink 8 621 mL 100% Concord grape juice/day; 12 weeks)
ounces of either 100% pomegranate juice or a fla- found that Concord grape juice significantly
vor-matched placebo drink; 4 weeks; memory improved verbal learning (p = .04) with trends for
testing, fMRI scans during cognitive tasks) found enhanced verbal and spatial recall suggesting that
that the pomegranate group showed a significant supplementation with Concord grape juice may
Placebo Flavonoid-rich Orange Juice

0.15
0.1
Global Cognitive Performance
0.05
Post Juice Consumption

0
2 hours post 6 hours post
–0.05
–0.1
–0.15
Fig. 18.16 Acute effect of the consumption of high flavonoid orange juice on global cognitive z score in middle-aged
men (p = .09) (adapted from [66])
help to enhance cognitive function for older adults Meta-analysis and Systematic Review
with early memory decline [71]. Two 2017 dose response meta-analyses of obser-
vational studies consistently show that coffee and
18.4.2.2 Coffee and Tea tea improved age related cognitive performance
Caffeinated beverages, such as coffee and tea are and most common cognitive disorders [74, 75].
among the most popular beverages in the world. For coffee, a dose-response meta-analysis (9
In addition to caffeine they contain polyphenols, cohort studies; 34,282 participants; 13 to 28 years
chlorogenic acid, theaflavins and other com- of follow-up) found a J-shaped association
pound that can have direct effects on the brain or between coffee consumption and the incidence of
influence cardio- and cerebrovascular health to cognitive disorders (e.g., Alzheimers disease,
improve cognitive function. Thus, the effects of dementia, cognitive decline or impairment [74].
coffee and tea on cognitive performance and Compared to < 1 daily cup of coffee, the consump-
dementia risk is of public health importance. The tion of 1-2 daily cups was associated with a signifi-
effects of caffeine have been the most widely cant 18% lower risk of cognitive disorders,
evaluated for safety and cognitive function [72]. whereas >4 cups of coffee were not significantly
A 2017 comprehensive systematic review of the different than <1 cup of coffee. For tea, a dose-
potential adverse effects of caffeine consumption response meta-analysis (17 studies including 6
found that the evidence generally supports that cohort, 3 case-control, and 8 cross-sectional stud-
consumption of up to 400 mg caffeine/day (esti- ies; 48,435 participants) found that tea consump-
mated 4–5 cups of coffee or 5–6 cups of black tion was inversely and linearly associated with the
tea) in healthy adults is not associated with overt, risk of cognitive disorders [75]. The highest tea
adverse cardiovascular effects, behavioral effects, intake was significantly associated with a 27%
reproductive and developmental effects, acute lower cognitive disorder risk with green tea lower-
effects, or bone status [72]. This review also sup- ing risk by 36% compared to 25% for black tea.
ports consumption of up to 300 mg caffeine/day The dose response analysis for increasing tea
in healthy pregnant women as an intake that was intake on reducing cognitive disorder risk at 100
generally not found to be associated with adverse ml/day by 6%, at 300 ml/day by 19% and at 500
reproductive and developmental effects. For opti- ml/day by 29% [75].
mal cognitive function, caffeine exerts its effects
by blocking adenosine receptors [73]. The intake Prospective Studies
of low (∼40 mg) to moderate (∼300 mg) caffeine Three observational studies are representative of
doses showed general improvements in alertness, coffee and tea effects on age related cognitive
attention, and reaction time, but less consistent performance [76–78]. The 2011 Honolulu-Asia
effects were observed in memory and higher- Aging Study (3494 men; mean baseline age
order executive function, such as judgment and 52 years; 25 years of follow-up) found that that
decision making. Caffeine’s effect on physical men with high coffee and caffeine intake in
performance metrics such as time-to-exhaustion, midlife had a lower risk of cognitive impairment,
time-trial, muscle strength and endurance, and dementia, or neuropathologic lesions. In the
high-intensity sprints typical of team sports were highest quartile of caffeine intake (≥411.0 mg/
evident following doses ≥200 mg. Also, physi- day) vs. the lowest quartile (≤137.0 mg/day) men
cally and mentally demanding occupations, had a 55% lower risk of any type of brain lesion
including military, first responders, transport by 55% (p-trend = 0.04; multivariate adjusted)
workers, and shift workers, can require optimal [76]. A 2010 cross-sectional analysis of The
physical and cognitive function, and repeated Singapore Longitudinal Aging Studies (716
administration of caffeine may be an effective Chinese adults; mean age 65 years; 56% women)
option to maintain the required physical and showed that total tea consumption was indepen-
cognitive performance. dently associated with better performances on the
Min i-Mental State Examination measures of high cocoa flavanol beverage significantly
global cognition, memory, executive function, improved the overall composite cognitive func-
and information processing speed after adjusting tion z score as well as significantly improved
for confounding factors [77]. Black/oolong and blood pressure and reduced insulin resistance and
green tea consumption promoted similar cogni- lipid peroxidation [80]. A 2013 RCT (60 older
tive performance benefits. A 2008 examination of US subjects; aged 73 years; 2 cups of cocoa daily
The Singapore Longitudinal Aging Studies (1438 (flavanol-rich cocoa 609 mg vs. flavanol-poor
cognitively normal participants in longitudinal cocoa 13 mg flavanols/serving; 30 days) demon-
analysis; mean baseline age 65 years; 1–2 years strated that regular high flavanol cocoa beverage
of follow-up) showed that total tea intake was intake promoted better neurovascular coupling
significantly associated with a lower prevalence and greater white matter structural integrity in
of cognitive impairment, independent of other individuals with baseline impairments [81].
risk factors. For age-related dementia, low intake
significantly reduced risk by 44%, moderate 18.4.2.4 Alcohol
intake by 55%, and high intake by 63% compared Despite chronic excessive alcohol consumption
to no or rare tea intake [78]. For normal cognitive causing progressively increasing neurodegenera-
decline, the corresponding risk reductions for tive disease risk, a number of studies suggest that
low, moderate and high tea consumption were low-moderate alcohol consumption, within limits
26%, 22% and 43% (p-trend = 0.042). These and/or of certain types, may be associated with a
effects were most evident for black (fermented) decreased cognitive decline and risk of dementia
and oolong (semi-fermented) teas. [82]. There are many mechanisms proposed to
explain this low to moderate alcohol effect
18.4.2.3 Flavanol-Rich Cocoa including: (1) the antioxidant properties of the
Beverages flavonoids in red wine may help prevent the oxi-
dative damage implicated in dementia; (2) by
RCTs increasing levels of HDL cholesterol and fibrino-
Three double-blind RCTs are representative of lytic factors leading to lower platelet aggregation
the benefits of flavanol-rich cocoa beverages on and possibly lower risk of ischemic stroke; and
age-related cognitive performance [79–81]. A (3) by decreasing cardiovascular risk via the
2015 UK crossover RCT (18 subjects; mean age mechanisms described above and possibly
61 years; 10 men; pre- and post-consumption of enhancing insulin sensitivity or reducing sys-
low cocoa flavanol (23 mg) or high cocoa flava- temic inflammatory response [82]. Table 18.6
nol (494 mg) 330 mL equicaloric flavanol drinks summarizes prospective cohort studies on the
matched for caffeine, theobromine, taste and effects of low to moderate alcohol intake by older
appearance) found that the high cocoa flavanol adults in cognitive performance and dementia
drink improved regional cerebral perfusion, espe- risk [82–88].
cially in the anterior cingulate cortex and the cen-
tral opercular cortex of the parietal lobe identified Systematic Review and Meta-Analysis
by arterial spin labelling functional magnetic One systematic review and meta-analysis (20 pro-
resonance imaging within 2 h [79]. This increased spective cohort studies, and 3 retrospective
cerebral blood flow provides a possible acute matched case-control nested in a cohort; adults
mechanism by which cocoa flavanols benefit >65 years; 1–25 years of follow-up) indicates that
cognitive performance. Another 2015 RCT (90 intake of small to moderate amounts of alcohol
older Italian subjects; mean age 70 years; daily beverage in older adults intake may be protective
drink containing 993 mg [high] cocoa flavanol, against dementia by 37% and Alzheimer’s disease
520 mg [intermediate] cocoa flavanol, or 48 mg risk by 43% [82]. The evidence is strongest for
[low] cocoa flavanol; 8 weeks) showed that the red wine consumption but not conclusive.
Table 18.6 Summaries of alcohol beverage studies on age-related cognitive performance

Peters et al. (2008). 20 prospective cohorts and 3 This meta-analysis suggests that small to
Evaluate the evidence retrospective matched case-control moderate amounts of alcohol intake may be
for any relationship nested in cohort studies; participants protective against dementia by 37% and
between cognitive ≥65 years; 1–25 years of follow-up Alzheimer’s disease by 43%. The evidence is
decline or dementia in strongest for wine consumption but not
the elderly and alcohol conclusive
consumption [82]
Prospective Studies
Heymann et al. (2016). 360 patients with early Alzheimer’s Heavy drinkers (8 or more alcoholic drinks/
Examine the disease in New York, Boston, week) had a faster cognitive decline,
relationship between Baltimore and Paris; followed-up deteriorating by 1.85 more points on their
alcohol, both the biannually for up to 19 years global cognitive score annually compared to
amount and type, and (multivariate adjusted) abstainers (p = .001), or 2.44 more points
cognitive decline in a compared to mild-moderate drinkers (1–7
cohort of Alzheimer’s alcoholic drinks/week) (p = .008). There was
disease patients (US no significant difference when comparing
and France) [83] mild-moderate drinkers to abstainers.
Increasing standard drinks of hard liquor, but
not beer or wine, was also associated with a
faster rate of cognitive decline (β = −0.117
p = .001)
Downer et al. (2015). 664 subjects; mean baseline age of Light alcohol consumption during late life is
Examine the 42–75 years; follow-up for 33 years associated with higher episodic memory, but
relationship between (multivariate adjusted) not executive function or global cognition, after
midlife and late-life controlling for the effects of age, gender,
alcohol intake, educational attainment, smoking, and APOE e4
cognitive functioning, status. Also, older adults who consumed
and regional brain moderate amounts of alcohol had significantly
volumes among older larger hippocampal volume compared to older
adults without dementia adults who did not consume alcohol
or a history of abusing
alcohol (Framingham
Heart Study Offspring
Cohort; US) [84]
Ritchie et al. (2014). 1932 subjects completed the same There was a significant gene × alcohol
Evaluate the effect of intelligence test at age 11 and consumption interaction on lifetime cognitive
alcohol consumption on 70 years; alcohol consumption in later change (p = .007). Subjects with higher genetic
cognitive function and life and genotype for a set of four ability to process alcohol showed relative
genetic differences in single-nucleotide polymorphisms in 3 improvements in cognitive ability with more
the ability to metabolize alcohol dehydrogenase genes were consumption, whereas those with low
alcohol (Lothian Birth determined processing capacity showed a negative
Cohort; UK) [85] relationship between cognitive change and
higher alcohol consumption. The effect of
alcohol consumption on cognitive change
appears to depend on genetic differences in the
ability to metabolize alcohol
Arntzen et al. (2010). 5033 stroke-free men and women; Light to moderate wine intake was
Examine the effect of 7 years of follow-up; verbal memory, independently associated with better
different alcoholic digit-symbol coding test, and tapping performance on all cognitive tests in both men
beverages on cognitive test (multivariate adjusted) and women. There was no consistent
performance in a association between beer and spirits intake and
population based study cognitive performance results. Alcohol
(Tromso Study; abstention was associated with a modest lower
Norway) [86] cognitive function in women
Solfrizzi et al. (2007). 121 MCI patients; aged 65–84 years; Patients with MCI who consumed up to 1
Estimate the impact of 3.5 years of follow-up (multivariate drink/day had a reduction in the rate of
alcohol consumption on adjusted) progression to dementia in comparison with
the incidence of mild patients with MCI who never consumed
cognitive impairment alcohol. Overall, vs. nondrinkers, patients with
(MCI) and its MCI who consumed up to 1 drink/day, derived
progression to dementia mostly from wine, had a decrease in the rate of
(Italian Longitudinal progression to dementia by 85%
Study on Aging) [87]
Stampfer et al. (2005). 12,480 women; mean baseline age Moderate drinkers (about one drink/day) had
Examine the effects of 74 years; 2 years of follow-up better mean cognitive scores than nondrinkers,
moderate alcoholic (multivariate adjusted) a reduced risk of general cognitive impairment
consumption in women by 33% and global cognitive score combining
(Nurses’ Health Study; the results of all tests by 19%. There were no
US) [88] significant associations between 2 daily drinks
and the risk of cognitive impairment or decline
vs. non-drinkers. There were no significant risk
differences for wine vs. beer
Prospective Studies differences in the ability to metabolize alcohol [85].

Six prospective studies are representative of the Subjects with higher genetic ability to process alco-
effects of alcohol intake on age related cognitive hol had relative improvements in cognitive ability
performance [83–88]. A 2016 study (360 patients with alcohol consumption, whereas those with low
with early Alzheimers’ disease in New York, processing capacity had a relatively negative effect
Boston, Baltimore and Paris; followed-up bi- on cognition with alcohol consumption. A 2010
annually for up to 19 years) found that heavy drink- Norwegian study (5033 adults from the general pop-
ers (8 or more alcoholic drinks/week) had a faster ulation; 7 years of follow-up) showed that moderate
cognitive decline in their annual global cognitive wine consumption was independently associated
score compared to abstainers (p = 0.001) or mild- with better performance on all cognitive measures in
moderate drinkers (1–7 alcoholic drinks/week) both men and women [86]. The 2007 Italian
(p = 0.008) [83]. There was no significant difference Longitudinal Study on Aging (121 patients with
when comparing mild-moderate drinkers to abstain- mild cognitive impairment (MCI); aged 65–84 years;
ers. Also, higher consumption of hard liquor, but not 3.5 years of follow-up) found that patients with MCI
beer or wine, was associated with a faster rate of who consumed up to 1 drink/day had a reduction in
cognitive decline. A 2015 Framingham Heart Study the rate of progression to dementia in comparison
Offspring Cohort (664 subjects; follow-up for with patients with MCI who never consumed alco-
33 years from mean age of 42–75 years) showed hol [87]. Overall, compared to non-drinkers, patients
that light alcohol consumption during late life is with MCI who consumed 1 drink/day, derived
associated with higher multivariate adjusted epi- mostly from wine, had a decrease in the rate of pro-
sodic memory, but not executive function or global gression to dementia by 85%. The Nurses’ Health
cognition [84]. Also, older adults who consumed Study (12,480 women; mean baseline age 74 years;
moderate amounts of alcohol had significantly 2 years of follow-up) showed that moderate drinkers
larger hippocampal volume compared to older (about one drink/day) had better mean cognitive
adults who did not consume alcohol. The 2014 UK scores than nondrinkers; moderate drinkers, as com-
Lothian Birth Cohort (1932 subjects who completed pared with nondrinkers had a reduced risk of general
the same intelligence test at age 11 and 70 years; cognitive impairment by 33% and global cognitive
alcohol consumption in later life and genotype score combining the results of all tests by 19% (after
determined) found that the effect of alcohol con- multivariate adjustment) [88]. There were no sig-
sumption on cognitive change depends on genetic nificant risk differences for wine vs. beer.
18.4.3 Lutein tively correlated with cognitive performance and

reduced risk of dementia. Table 18.7 summarizes
Lutein and zeaxanthin are carotenoids that are RCTs and observational studies on the effects of
highly concentrated in human macular pigment in lutein intake by older adults in cognitive perfor-
the retina, attenuating blue light exposure, provid- mance and dementia risk [90–99].
ing protection from photo-oxidation and enhanc-
ing visual performance [89]. Recently, interest in 18.4.3.1 Randomized Controlled
lutein has expanded beyond the retina to its possi- Trials (RCTs)
ble contributions to brain development and func- Two RCTs support the benefits of increased lutein
tion. Only humans and other primates accumulate on cognitive performance in humans [90, 91]. A
lutein within the brain. Recently, lutein has been US RCT (39 adults; mean age 63 years; 70%
shown to preferentially accumulate in the human women; daily intake of 1 medium avocado vs. 1
brain and its content in neural tissue has been posi- medium potato or 1 cup of chickpeas (control);
Table 18.7 Summary of studies on lutein and age-related cognitive performance and dementia
Randomized Controlled Trials (RCT)
Johnson et al. (2015). Parallel RCT: At baseline, there were no significant differences
Test the effects of daily 39 adults mean age 63 years; 70% between subjects in the avocado and control
intake of 1 medium women; daily intake of 1 medium groups in any of the study measures. At 6 months,
avocado on macular avocado vs. alternating medium serum lutein levels in the avocado group
pigment density and potato or cup of chickpeas significantly increased from baseline by 20%
cognitive function (US) (control); on serum lutein levels, (p < .0005) whereas the potato/chickpea control
[90] macular pigment density and group increased by 7% (p < .03). After 6 months,
cognition in healthy older adults; there was a significant increase in macular
serum carotenoids were measured pigment density in the avocado group but no
by HPLC; computerized cognitive change in the control group. In the avocado group
assessment battery was used for the change in macular pigment density was
cognition measures; 6 months significantly related to an improved spatial
working memory (p < .009) and problem
approaching efficiency (p < .036). No significant
changes in cognitive function were observed in the
control group. These data suggest that an
intervention with avocados to increase neural
lutein is an effective dietary strategy for cognitive
health
Bovier et al. (2014). Double-blind Parallel RCT: Significant correlations were found between
Evaluate the effect of 92 young adults; average age retinal lutein and zeaxanthin (macular pigment
increased lutein and 22 years; 36 males and 56 [MP] density) and critical flicker fusion (CFF)
zeaxanthin on visual females; 3 interventions: thresholds (p < .01) and visual motor performance
processing speed and supplements: 20 mg zeaxanthin/ (overall p < .01). In general, increasing MP
efficiency in younger day; 26 mg zeaxanthin, 8 mg density through supplementation (average increase
adults (US) [91] lutein, 190 mg mixed n-3 fatty of about 0.09 log units) resulted in significantly
acids/day; placebo; 4 months improved visual processing speed
Zamroziewicz et al. 76 cognitively intact adults; mean Serum lutein levels were significantly positively
(2016). age 69 years; 67% women; mean associated with gray matter volume of the right
Evaluate the effect of serum lutein 454 pmol/mL (range para-hippocampal cortex and with crystallized
serum lutein levels on 120–1328 pmol/mol) intelligence (the ability to use learned knowledge
cognitive performance and experience)
and intelligence in
healthy older adults
(cross-sectional study;
US) [92]
Feart et al. (2016). 1092 nondemented older After adjustment for sociodemographic data, diet
Analyze the relation participants; average baseline age quality, and clinical variables, including baseline
between plasma 74 years; average 8.8 years of cognitive performances, only higher lutein
carotenoids and the risk follow-up; 132 Alzheimers’ concentration, considered as a function of plasma
of dementia and disease cases lipids, was consistently significantly associated
Alzheimer’s disease with a decreased risk of all-cause dementia by
(Three-City-Bordeaux 20% (p = .024) and Alzheimers’ disease by 24%
cohort; French) [93] (p = .021)
Lindbergh et al. (2016). 43 community-dwelling older This study indicates that higher lutein and
Investigate underlying adults; mean age 72 years; 58% zeaxanthin levels, measured both acutely (serum)
neural mechanisms for female; 100% Caucasian); asked and acquired (retinal), enhance neural efficiency
lutein and zeaxanthin to learn and recall pairs of during verbal learning and memory in older adults.
using functional magnetic unrelated words in an f MRI- Lutein and zeaxanthin were significantly related to
resonance imaging adapted paradigm. Lutein and more and efficient blood-oxygen-level-dependent
(f-MRI) during cognitive zeaxanthin levels were measured signals in central and parietal operculum cortex,
performance (cross- in retina by macular pigment cerebellum and other brain regions
sectional study; US) [94] optical density and serum using
validated procedures
Vishwanathan et al. Donated human brain tissue and Macular pigment carotenoids (lutein, meso-
(2016). matched retina for 13 individuals zeaxanthin, and zeaxanthin combined) in the
Evaluate the relationship with a mean age at death of retina were significantly related to the combined
between retinal and brain 75 years with normal cognitive concentrations of lutein and zeaxanthin in the
lutein and zeaxanthin in function or Alzheimer’s disease occipital cortex. When analyzed separately, only
humans (US) [95] were obtained from the National retinal lutein (plus meso-zeaxanthin), not
Disease Research Interchange zeaxanthin, was significantly related to lutein in
the occipital cortex. No correlations were observed
with lutein and zeaxanthin in the hippocampus.
Total macular pigment density measured via
non-invasive, psychophysical techniques can be
used as a biomarker to assess brain lutein and
zeaxanthin status in clinical studies
Kelly et al. (2015). 105 subjects free of retinal Significant correlations were evident between
Investigate the disease with low macular pigment macular pigment levels and various measures of
relationship between and mean age 47 years; and 121 cognitive function in both groups (r = −0.27–0.26,
macular pigment, serum subjects with age-related macular p ≤ .05, for all). Both serum lutein and zeaxanthin
concentrations of lutein degeneration (AMD) and mean levels correlated significantly (r = 0.19, p ≤ .05
and zeaxanthin, and age 65 years and r = 0.20; p ≤ .05, respectively) with semantic
cognitive function in fluency cognitive scores. For AMD subjects,
subjects free of retinal serum lutein levels also correlated significantly
disease with low macular with verbal recognition memory learning slope
pigment levels (cross- (r = 0.20, p = .031)
sectional study; Ireland)
[96]
Renzi et al. (2014). 24 subjects with mild cognitive In healthy older adults, macular pigment density
Test whether retinal impairment were compared with was significantly related to visual-spatial and
measures of macular 24 matched controls; subjects constructional abilities (p = .04). For subjects with
pigment density can be matched with respect to age, mild cognitive impairment, macular pigment
used as a surrogate for BMI, ethnicity, sex, and smoking density was broadly and significantly related to
brain lutein and status; degree of cognitive cognition including the composite score on the
zeaxanthin levels related impairment and cognitive ability mini-mental state exam, visual-spatial and
to cognitive function when was determined via structured constructional abilities, language ability, attention,
comparing healthy older clinical interview; macular and the total scale on the repeatable battery
adults with mildly pigment density was measured assessment of neuropsychological status
cognitively impaired older psychophysically
adults (cross-sectional
study; US) [97]
(continued)
Vishwanathan et al. 108 adults, mean age 78 years Macular pigment density levels were significantly
(2014). sampled from the age-related associated with better global cognition, verbal
Determine whether maculopathy ancillary study; learning and fluency, recall, processing speed and
macular pigment density, serum carotenoids were measured perceptual speed, whereas serum lutein and
lutein and zeaxanthin in using HPLC; macular pigment zeaxanthin were significantly related to only
the macula are related to density was assessed using verbal fluency
cognitive function in heterochromatic flicker
older adults (cross- photometry; 8 cognitive tests
sectional study Healthy designed to evaluate several
Aging and Body cognitive domains including
Composition Study; US) memory and processing speed
[98] were administered
Johnson et al. (2013). Sera from 78 octogenarians and Serum lutein, zeaxanthin, and 𝛽-carotene
Assess the relationship 220 centenarians; brain tissues concentrations were most consistently related to
between serum and brain from 47 centenarian decedents; better cognition (p < .05) in the whole population
carotenoids on cognitive samples were analyzed for and in the centenarians. Only serum lutein was
performance (cross carotenoids, 𝛼-tocopherol, and significantly related to better cognition in the
sectional study; Oldest retinol; global cognition, octogenarians. In the brain, lutein was consistently
Old from the Georgia dementia, depression and associated with a range of cognitive measures.
Centenarian Study; US) cognitive domains (memory, There were fewer significant relationships for
[99] processing speed, attention, and 𝛼-tocopherol. These findings suggest that the
executive functioning) status of lutein in the old may especially reflect
their cognitive function
6 month-duration) found that serum lutein levels 18.4.3.2 Observational Studies

in the avocado group were significantly increased Eight observation studies consistently support
from baseline by 20% compared to an insignifi- the beneficial effects of increased lutein on cog-
cant increase in the control group were by 7% nitive performance and reduced Alzheimer’s
[90]. After 6 months, there was a significant disease risk [92–99]. A 2016 US cross-sectional
increase in macular pigment density (a biomarker study (76 cognitively intact adults; mean age
of brain lutein levels) in the avocado group but no 69 years; 67% women) found that serum lutein
change in the control group. In the avocado group levels were significantly positively associated
the change in macular pigment density was sig- with gray matter volume of the right parahip-
nificantly related to an improved spatial working pocampal cortex and with crystallized intelli-
memory and problem approaching efficiency, gence (the ability to use learned knowledge and
whereas no significant changes in cognitive func- experience) [92]. A 2016 French prospective
tion were observed in the control group. These study (1092 nondemented older participants;
data suggest that an intervention with avocados to average baseline age 74 years; average 8.8 years
increase neural lutein is an effective dietary strat- of follow-up) showed that of all the diet quality
egy for cognitive health. Also, a US double- variables only higher lutein concentration as a
blinded RCT (92 young adults; average age function of plasma lipids was consistently sig-
22 years; 36 males and 56 females; 3 interven- nificantly associated with a decreased risk of
tions: supplement containing 20 mg zeaxanthin/ all-
cause dementia by 20% and Alzheimers’
day; supplement containing 26 mg zeaxanthin, disease by 24% [93]. Another US cross-sec-
8 mg lutein, 190 mg mixed n-3 fatty acids/day; tional study (43 community-dwelling older
and a placebo; 4 months) showed significant cor- adults; mean age 72 years; 58% female; 100%
relations between retinal lutein and zeaxanthin Caucasian; asked to learn and recall pairs of
macular pigment density and improved visual unrelated words in an fMRI-adapted paradigm)
processing speed [91]. found that higher lutein and zeaxanthin levels,
18.5 Lifestyle Factors 511
measured both acutely (serum) and acquired whereas serum lutein and zeaxanthin were
(retinal), enhanced neural efficiency during significantly related to only verbal fluency

verbal learning and memory in older adults [98]. The US Oldest Old from the Georgia
with improved brain functional signaling effi- Centenarian Study (78 octogenarians and 220
ciency [94]. In a 2016 evaluation of brain and centenarians; brain tissues from 47 deceased
retina tissue from 13 individuals with a mean centenarians) showed that serum lutein, zea-
age at death of 75 years with normal cognitive xanthin, and 𝛽-carotene concentrations were
function or Alzheimer’s disease donated from most consistently related to better cognition
the US National Disease Research Interchange (p < 0.05) in the whole population and in the
showed that lutein was significantly concen- centenarians whereas only serum lutein was
trated in both the brain’s occipital cortex and significantly related to better cognition in the
retina macular pigment to confirm use of macu- octogenarians [99]. These studies indicate that
lar pigment density as a biomarker of brain higher serum levels of certain carotenoids may
lutein status in clinical studies [95]. A 2015 reflect better cognitive function in older age.
Irish cross-sectional study (105 subjects free of
retinal disease with low macular pigment; mean
age 47 years; 121 subjects with age-related 18.5 Lifestyle Factors
macular degeneration [AMD] and mean age
65 years) found that significant correlations CVD risk factors such as hypertension, diabetes,
were evident between macular pigment levels and hyperlipidemia increase an individuals’ risk
and various measures of cognitive function in for stroke, neurocognitive impairment and
both groups (r = −0.27–0.26, p ≤ 0.05, for all) dementia [100]. Also, the effects of these CVD
and both serum lutein and zeaxanthin levels risk factors on brain dysfunction are additive.
correlated significantly (r = 0.19, p ≤ .05 and Lifestyle behaviors, including physical activity
r = 0.20; p ≤ .05, respectively) with semantic and healthy dietary habits are recommended to
fluency cognitive scores [96]. For AMD sub- improve CVD risk factors and may aid in the
jects, serum lutein levels also correlated signifi- prevention of neurocognitive decline. A US
cantly with verbal recognition memory cross-sectional analysis of baseline data from
learning slope (r = 0.20, p = .031). A 2014 Exercise and Nutritional Interventions for
cross-sectional study (24 subjects with mild Neurocognitive Health Enhancement
cognitive impairment were compared with 24 (ENLIGHTEN) trial (160 adults with cognitive
matched controls) showed that macular pig- impairment, without dementia; 67% women;
ment density was broadly and significantly mean age 65 years; completed neurocognitive
related to cognition including the composite assessments of executive function, processing
score on the mini-mental state examination, speed, and memory; physical activity using
language ability, attention, and a repeatable accelerometry, aerobic capacity determined by
assessment of neuropsychological status [97]. exercise testing, and dietary habits for adherence
The 2014 US Healthy Aging and Body to the Mediterranean and DASH diets) found
Composition Study (108 adults, mean age that greater aerobic capacity and daily physical
78 years sampled from the agerelated macu- activity were associated with better executive
lopathy ancillary study; serum carotenoids; functioning, processing speed and verbal mem-
macular pigment density; eight cognitive tests ory, DASH diet adherence was associated with
designed to evaluate several cognitive domains better verbal memory whereas higher hs-CRP
including memory and processing speed) was associated with poorer executive function-
found that macular pigment density was asso- ing, processing speed and verbal memory [100].
ciated with significantly better global cogni- These findings support the adoption of healthy
tion, verbal learning and fluency, recall, lifestyle habits to reduce the risk of neurocogni-
processing speed and perceptual speed, tive decline in older adults.
Conclusions MedDiet, DASH diet, and MIND diet (a

All measures of cognitive performance hybrid of both the MedDiet and DASH diet
decline with age, with executive functioning with an emphasis on specific brain protective
(e.g., working memory, reasoning, task flexi- foods) are effective in protecting cognitive
bility, problem solving and planning) show- performance with aging. A number of RCTs
ing the largest rate of decline with every and prospective cohort studies support the
successive decade of age. There is a consider- benefits of high polyphenolic fruits and veg-
able degree of heterogeneity in cognitive per- etables, dairy (especially yogurt), 100%
formance across populations, which can be juices (polyphenol rich), coffee, tea, flavanol-
significantly affected by dietary pattern and rich cocoa beverages, and low-moderate wine
specific foods and beverages. In general, fol- consumption on improving age-related cogni-
lowing dietary advice for lowering the risk of tive performance and reducing risk of demen-
cardiovascular and metabolic disorders, such tia, but excessive alcohol consumption can
as consuming high levels of healthy fats from have negative effects on cognitive perfor-
fish or vegetable oils, non-starchy vegetables, mance and lead to higher risk of dementia.
low glycemic index fruits and a diet low in Lutein has been shown to preferentially accu-
foods with added sugars should be encour- mulate in the human brain and its content in
aged for cognitive health. There is significant neural tissue as reflected in macular pigment
evidence from human studies that low quality density has been positively correlated with
diets reduce and high-quality diets enhance cognitive performance and reduced risk of
global cognitive performance with aging. The dementia.
Appendix A 513

A
Healthy
vegetarian
Healthy pattern
Western dietary USDA base DASH diet Mediterranean (Lact-ovo
Emphasizes Refined grains, Vegetables, Potassium rich Whole grains, Vegetables, Plant foods:
low fiber foods, fruits, whole- vegetables, fruits, vegetables, fruit, whole- vegetables,
red meats, sweets, grains, and and low fat milk fruits, dairy grains, legumes, fruits, whole
and solid fats low-fat milk products products, olive nuts, seeds, milk grains, nuts,
oil, and products, and seeds, and soy
moderate wine soy foods foods
Includes Processed meats, Enriched grains, Whole-grains, Fish, nuts, Eggs, non-dairy Non-dairy
sugar sweetened lean meat, fish, poultry, fish, seeds, and milk milk
beverages, and nuts, seeds, and nuts, and seeds pulses alternatives, and alternatives
fast foods vegetable oils vegetable oils
Limits Fruits and Solid fats and Red meats, Red meats, No red and No animal
vegetables, and added sugars sweets and refined grains, white meats, or products
whole-grains sugar-sweetened and sweets fish; limited
beverages sweets
(% Total kcal)
Protein (% Total 16 17 18 16 14 13
kcal)
Total fat (% Total 33 32 27 34 32 30
kcal)
(% Total kcal)
Unsat. fat 22 25 21 24 26 25
(% Total kcal)
Fiber (g) 16 31 29+ 31 35+ 40+
Potassium (mg) 2800 3350 4400 3350 3300 3650
Vegetable oils (g) 19 27 25 27 19–27 18–27
Sodium (mg) 3600 1790 1100 1690 1400 1225
Plant Food Groups
Fruit (cup) ≤1.0 2.0 2.5 2.5 2.0 2.0
Vegetables (cup) ≤1.5 2.5 2.1 2.5 2.5 2.5
Whole-grains 0.5 3.0 4.0 3.0 3.0 3.0
(oz.)
Legumes (oz.) − 1.5 0.5 1.5 3.0 3.0+
Nuts/Seeds (oz.) 0.5 0.6 1.0 0.6 1.0 2.0
Soy products 0.0 0.5 − − 1.1 1.5
(oz.)
gov/ accessed 8.22.2015
2015
patterns. 2015;1–9
trends. 2015;1–78
Bhupathiraju SN, Tucker KL. Coronary heart disease prevention: nutrients, foods, and dietary patterns. Clinica Chimica
Acta. 2011;412: 1493–1514
ppendix B: Estimated Range of Energy, Fiber, Nutrients and Phytochemicals

A
Composition of Whole Plant Foods/100 g Edible Portion

Nutrients/ Wheat, oat, Apples, pears, Dates, dried Potatoes, Lentils, Almonds, Brazil
Phytochemicals barley, rye, bananas, figs, spinach, carrots, chickpeas, nuts, cashews,
brown rice, grapes, apricots, peppers, lettuce, split peas, hazelnuts,
whole grain oranges, cranberries, green beans, black macadamias,
bread, cereal, blueberries, raisins and cabbage, onions, beans, pecans, walnuts,
pasta, rolls and strawberries, prunes cucumber, pinto peanuts,
crackers and avocados cauliflower, beans, and sunflower seeds,
mushrooms, and soy beans and flaxseed
broccoli
Energy (kcal) 110–350 30–170 240–310 10–115 85–170 520–700
Protein (g) 2.5–16 0.5–2.0 0.1–3.4 0.2–5.0 5.0–17 7.8–24
Available 23–77 1.0–25 64–82 0.2–25 10–27 12–33
Carbohydrate (g)
Fiber (g) 3.5–18 2.0–7.0 5.7–10 1.2–9.5 5.0–11 3.0–27
Total fat (g) 0.9–6.5 0.0–15 0.4–1.4 0.2–1.5 0.2–9.0 46–76
SFAa (g) 0.2–1.0 0.0–2.1 0.0 0.0–0.1 0.1–1.3 4.0–12
MUFAa (g) 0.2–2.0 0.0–9.8 0.0–0.2 0.1–1.0 0.1–2.0 9.0–60
PUFAa (g) 0.3–2.5 0.0–1.8 0.0–0.7 0.0.0.4 0.1–5.0 1.5–47
Folate (μg) 4.0–44 <5.0–61 2–20 8.0–160 50–210 10–230
Tocopherols (mg) 0.1–3.0 0.1–1.0 0.1–4.5 0.0–1.7 0.0–1.0 1.0–35
Potassium (mg) 40–720 60–500 40–1160 100–680 200–520 360–1050
Calcium (mg) 7.0–50 3.0–25 10–160 5.0–200 20–100 20–265
Magnesium (mg) 40–160 3.0–30 5.0–70 3.0–80 40–90 120–400
Phytosterols (mg) 30–90 1.0–83 – 1.0–54 110–120 70–215
Polyphenols (mg) 70–100 50–800 – 24–1250 120–6500 130–1820
Carotenoids (μg) – 25–6600 1.0–2160 10–20,000 50–600 1.0–1200
a
Database for Standard Reference, Release 27. http://www.ars.usda.gov/nutrientdata. Accessed 17 February 2015
Circulation. 2013;128: 553–565
Slavin JL, Lloyd B. Health benefits of fruits and vegetables. Adv Nutr. 2012;3:506–516
An. 1999;12:169–196
Acta. 2011;412:1493–1514
trends. 2015;1–78
http://health.gov/dietaryguidelines/2015/guidelines/ Accessed 1.26.2016
References 515
References Dement. 2016;12(2):100–9. https://doi.org/10.1016/

j.jalz.2015.08.002.
14. Tsai HJ. Dietary patterns and cognition decline in
1. Lipnicki DM, Crawford JD, Dutta R, et al. Age-
Taiwanese aged 65 years and older. Int J Geriatr
related cognitive decline and associations with
Psychiatry. 2015;30(5):523–30.
sex, education and apolipoprotein E genotype
15. Gardener SL, Rainey-Smith SR, Barnes MB,
across ethnocultural groups and geographic
et al. Dietary patterns and cognitive decline in
regions: a collaborative cohort study. PLoS Med.
an Australian study of ageing. Mol Psychiatry.
2017;14(3):e1002261. https://doi.org/10.1371/jour-
2015;20(7):860–6.
nal.pmed.1002261.
16. Zhu N, Jacobs DR, Meyer KA, et al. Cognitive func-
2. World Health Organization (WHO). Dementia:
tion in a middle-aged cohort is related to higher
a public health priority (2012). World Health
quality dietary patterns 5 and 25 years earlier: the
Organization Dementia: Fact Sheet No. 362 (2016).
CARDIA study. J Nutr Health Aging. 2015;19(1):33–
http://www.who.int/mediacentr0e/factsheets/fs362/
8. https://doi.org/10.1007/s12603-014-0491-7.
en/. Accessed Apr 2017.
17. Ozawa M, Ninomiya T, Ohara T, et al. Dietary pat-
3. Koellhoffer EC, McCullough LD, Ritzel RM. Old
terns and risk of dementia in an elderly Japanese
maids: aging and its impact on microglia function.
population: The Hisayama Study. Am J Clin Nutr.
Int J Mol Sci. 2017;18:769. https://doi.org/10.3390/
2013;97:1076–82.
ijms18040769.
18. Wengreen HJ, Neilson C, Munger R, Corcoran
4. Lopez-Valdes HE, Martinez-Coria H. The role of
C. Diet quality is associated with better cognitive
neuroinflammation in age-related dementias. Rev
test performance among aging men and women. J
Inves Clin. 2016;68:40–8.
Nutr. 2009;139:1944–9. https://doi.org/10.3945/
5. Santo CY, Snyder PJ, Wu W-C, et al. Pathophysiological
jn.109.106427.
relationship between Alzheimer’s disease, cerebrovas-
19. Attuquayefio T, Stevenson RJ, Oaten MJ, Francis
cular disease, and cardiovascular risk: a review and syn-
HM. A four-day Western-style dietary intervention
thesis. Alzheimers Dementia (Amst). 2017;7:69–87.
causes reductions in hippocampal-dependent learn-
6. Solfrizzi V, Panza F, Frisardi V, et al. Diet and
ing and memory and interoceptive sensitivity. PLoS
Alzheimer’s disease risk factors or prevention: the cur-
One. 2017;12(2):e0172645. https://doi.org/10.1371/
rent evidence. Expert Rev Neurother. 2011;11(5):677–
journal.pone.0172645.
708. https://doi.org/10.1586/ern.11.56.
20. Smyth A, Dehghan M, O’Donnell M, et al.
7. Noble EE, Hau TM, Kanoski SE. Gut to brain dysbi-
Healthy eating and reduced risk of cognitive
osis: mechanisms linking Western diet consumption,
decline a cohort from 40 countries. Neurology.
the microbiome, and cognitive impairment. Front
2015;84:2258–65.
Behav Neuorsci. 2017;11:9. https://doi.org/10.3389/
21. Masana ME, Koyanagi A, Haro JM, Tyrovolas S.
fnbah.2017.00009.
n-3 fatty acids, Mediterranean diet and cognitive
8. Wang M, Norman JE, Srinivasan VJ, Rutledge
function in normal aging: a systematic review. Exp
JC. Metabolic, inflammatory, and microvascular
Gerontol. 2017;91:39–50. https://doi.org/10.1016/
determinants of white matter disease and cognitive
Jexger.2017o2.008.
decline. Am J Neurodegener Dis. 2016;5(5):171–7.
22. Wu L, Sun D. Adherence to Mediterranean diet and
9. Ozawa M, Shipley M, Kivimaki M, et al. Dietary
risk of developing cognitive disorders: an updated
pattern, inflammation and cognitive decline: The
systematic review and meta-analysis of prospective
Whitehall II prospective cohort study. Clin Nutr.
cohort studies. Sci Rep. 2017;7:41317. https://doi.
2017;36:506–12.
org/10.1038/srep41317.
10. Pearson KE, Wadley VG, McClure LA, et al. Dietary
23. Hardman RJ, Kennedy G, Macpherson H, et al.
patterns are associated with cognitive function in the
Adherence to a Mediterranean-style diet and effects
REasons for Geographic and Racial Differences in
on cognition in adults: a qualitative evaluation and
Stroke (REGARDS) cohort. J Nutr Sci. 2016;5:e38.
systematic review of longitudinal and prospective
https://doi.org/10.1017/jns.2016.27.
trials. Front Nutr. 2016;3:22. https://doi.org/10.3389/
11. Granic A, Davies K, Adams A, et al. Dietary pat-
fnut.2016.00022.
terns high in red meat, potato, gravy, and butter are
24. Petersson SD, Philippou E. Mediterranean diet, cog-
associated with poor cognitive functioning but not
nition function, and dementia: a systematic review of
with rate of cognitive decline in very old adults. J
the evidence. Adv Nutr. 2016;7:889–904. https://doi.
Nutr. 2016;146:265–74. https://doi.org/10.3945/
org/10.3945/an.16.012138.
jn.115.216952.
25. Singh B, Parsaik AK, Mielke MM, et al. Association
12. Jacka FN, Cherbuin N, Anstey KJ, et al. Western
of Mediterranean diet with mild cognitive impair-
diet is associated with a smaller hippocampus: a
ment and Alzheimer’s disease: a systematic review
longitudinal investigation. BMC Med. 2015;13:215.
and meta-analysis. J Alzheimers Dis. 2014;39(2):
https://doi.org/10.1186/s12916-015-0461-x.
271–82.
13. Shakersaina B, Santonia G, Larssonb SC, et al.
26. Lourida I, Soni M, Thompson-Coon J, et al.
Prudent diet may attenuate the adverse effects of
Mediterranean diet, cognitive function, and
Western diet on cognitive decline. Alzheimers
dementia. A systematic review. Epidemiology.
2013;24(4):479–89. https://doi.org/10.1097/EDE. ter cognitive function: The Health and Retirement

0b013e3182944410. Study. J Am Geriatr Soc. 2017. https://doi.
27. Sofi F, Abbate R, Gensini GF, Cesari F. Accruing evi- org/10.1111/jgs.14922.
dence on benefits of adherence to the Mediterranean 40. Morris MC, Tangney CC, Wang Y, et al. MIND
diet on health: an updated systematic review and diet slows cognitive decline with aging. Alzheimers
meta-analysis. Am J Clin Nutr. 2010;92:1189–96. Dement. 2015;11(9):1015–22. https://doi.org/10.1016/j.
28. Galbete C, Toledo E, Toledo JB, et al. Mediterranean jalz.2015.04.011.
diet and cognitive function: the SUN project. J Nutr 41. Morris MC, Tangney CC, Wang Y, et al. MIND diet
Health Aging. 2015;19(3):305–12. associated with reduced incidence of Alzheimer’s
29. Tangney CC, Li H, Wang Y, et al. Relation of disease. Alzheimers Dement. 2015;11(9):1007–14.
DASH- and Mediterranean-like dietary patterns https://doi.org/10.1016/j.jalz.2014.11.001.
to cognitive decline in older persons. Neurology. 42. Mannikko R, Komulainen P, Schwab U, et al. The
2014;83:1410–6. Nordic diet and cognition—The DR’s EXTRA
30. Samieri C, Okereke OL, Devore EE, et al. Long- Study. Br J Nutr. 2015;114:231–9. https://doi.
term adherence to the Mediterranean diet is associ- org/10.1017/S0007114515001890.
ated with overall cognitive status, but not cognitive 43. Berti V, Murray J, Davies M, et al. Nutrient pattern
decline, in women. J Nutr. 2013;143:493–9. https:// and brain biomarkers of Alzheimer’s disease in cog-
doi.org/10.3945/jn.112.169896. nitively normal individuals. J Nutr Health Aging.
31. Wengreen H, Munger RG, Cutler A, et al. 2015;19(40):413–23. https://doi.org/10.1007/
Prospective study of Dietary Approaches to Stop s12603-014-0534-0.
Hypertension - and Mediterranean-style dietary pat- 44. Nurk E, Refsum H, Drevon CA, et al. Cognitive per-
terns and age-related cognitive change: the Cache formance among the elderly in relation to the intake
County Study on memory, health and aging. Am J of plant foods. The Hordaland Health Study. Br J
Clin Nutr. 2013;98:1263–71. Nutr. 2010;104:1190–201. https://doi.org/10.1017/
32. Samieri C, Grodstein F, Rosner BA, et al. Mediterranean S0007114510001807.
diet and cognitive function in older age: results 45. Lamport DJ, Saunders C, Butler LT, Spencer
from the Women’s Health Study. Epidemiology. JPR. Fruits, vegetables, 100% juices, and cognitive
2013;24(4):490–9. https://doi.org/10.1097/EDE. function. Nutr Rev. 2014;2(12):774–89.
0b013e18294a065. 46. Loef M, Walach H. Fruit, vegetable and prevention of
33. Valls-Pedret C, Sala-Vila A, Serra-Mir M, et al. cognitive decline or dementia: a systematic review of
Mediterranean diet and age-related cognitive cohort studies. J Nutr Health Aging. 2012;16(7):625–30.
decline. a randomized clinical trial. JAMA Intern 47. Devore EE, Kang JH, Breteler MMB. Dietary intake
Med. 2015;175(7):1094–103. of berries and flavonoids in relation to cognitive
34. Martinez-Lapiscina EH, Clavero P, Toledo R, decline. Ann Neurol. 2012;72(1):135–43. https://
et al. Mediterranean diet improves cognition: the doi.org/10.1002/ana.23594.
PREDIMED-NAVARRA randomized trial. J Neurol 48. Chen X, Huang Y, Cheng HG. Lower intake of vege-
Neurosurg Psychiatry. 2013;84(12):1318–25. tables and legumes associated with cognitive decline
35. Martinez-Lapiscina EH, Clavero P, Toledo E, et al. among illiterate elderly Chinese: a 3-year cohort
Virgin olive oil supplementation and long-term cog- study. J Nutr Health Aging. 2012;16(6):549–52.
nition: The PREDIMED-Navarra randomized trial. J 49. Nooyens ACJ, Bueno-de-Mesquita HB, van Boxtel
Nutr Health Aging. 2013;17(6):544–52. https://doi. MPJ. Fruit and vegetable intake and cognitive decline
org/10.1007/s12603-013-0027-6. in middle-aged men and women: the Doetinchem
36. Valls-Pedret C, Lamuela-Raventos RM, Alexander Cohort Study. Br J Nutr. 2011;106:752–61.
Medina-Remon A, et al. Polyphenol-rich foods in 50. Peneau S, Galan P, Jeandel C, et al. Fruit and
the Mediterranean diet are associated with better vegetable intake and cognitive function in the
cognitive function in elderly subjects at high cardio- SU.VI.MAX 2 prospective study. Am J Clin Nutr.
vascular risk. J Alzheimers Dis. 2012;29:773–82. 2011;94:1295–303.
https://doi.org/10.3233/JAD-2012-111799. 51. Kang JH, Ascherio A, Grodstein F. Fruit and veg-
37. Berendsen AM, Kang JH, van de Rest O, et al. The etable consumption and cognitive decline in aging
dietary approaches to stop hypertension diet, cogni- women. Ann Neurol. 2005;57(5):713–20.
tive function, and cognitive decline in American older 52. Lee J, Torosyan N, Silverman DH. Examining the
women. J Am Med Dir Assoc. 2017;18(5):427–32. impact of grape consumption on brain metabolism
https://doi.org/10.1016/j.jamada.2016.11.026. and cognitive function in patients with mild decline
38. Haring B, Wu C, Mossavar-Rahmani Y, et al. No in cognition: A double-blinded placebo controlled
association between dietary patterns and risk for pilot study. Exp Gerontol. 2017;87:121–8. https://
cognitive decline in older women with 9-year fol- doi.org/10.1016/j.exger.2016. 10.004.
low-up: data from the Women’s Health Initiative 53. Neshatdousta S, Saunders C, Castle SM, et al. High-
Memory Study. J Acad Nutr Diet. 2016;116:921–30. flavonoid intake induces cognitive improvements
39. McEvoy CT, Guyer H, Langa KM, Yaffe linked to changes in serum brain-derived neuro-
K. Neuroprotective diets are associated with bet- trophic factor: two randomised, controlled trials.
References 517
Nutr Healthy Aging. 2016;4:81–93. https://doi. males. Eur J Nutr. 2016;55:2021–9. https://doi.
org/10.3233/NHA-1615. org/10.1007/s00394-015-1016-9.
54. O'Brien J, Okereke O, Devore E, et al. Long-term 67. Lamport DJ, Lawton CL, Merat N, et al. Concord
intake of nuts in relation to cognitive function in older grape juice, cognitive function, and driving per-
women. J Nutr Health Aging. 2014;18(5):496–502. formance: a 12-wk, placebo-controlled, random-
55. Cardoso BR, Apolinário D, da Silva BV, et al. ized crossover trial in mothers of preteen children.
Effects of Brazil nut consumption on selenium sta- Am J Clin Nutr. 2016;103:775–83. https://doi.
tus and cognitive performance in older adults with org/10.3945/ajcn.115.114553.
mild cognitive impairment: a randomized controlled 68. Kean RJ, Lamport DJ, Dodd GF, et al. Chronic con-
pilot trial. Eur J Nutr. 2016;55(1):107–16. https:// sumption of flavanone-rich orange juice is associ-
doi.org/10.1007/s00394-014-0829-2. ated with cognitive benefits: an 8-wk, randomized,
56. Sánchez-Villegas A, Galbete C, Martinez-González double-blind, placebo-controlled trial in healthy
MA. The effect of the Mediterranean diet on plasma older adults. Am J Clin Nutr. 2015;101:506–14.
brain-derived neurotrophic factor (BDNF) levels: 69. Bookheimer SY, Renner BA, Ekstrom A, et al.
The PREDIMED-NAVARRA Randomized Trial. Pomegranate juice augments memory and fMRI activ-
Nutr Neurosci. 2011;14(5):195–201. ity in middle-aged and older adults with mild mem-
57. Cheng P-F, Chen J-J, Zhou X-Y, et al. Do soy isofla- ory complaints. Evid Based Complement Alternat
vones improve cognitive function in postmenopausal Med. 2013;946298. https://doi.org/10.1155/2013/
women? A meta-analysis. Menopause. 2015;22(2):198– 946298.
206. https://doi.org/10.1097/gme.00000000000290. 70. Kelly J, Fulford J, Vanhatalo A, et al. Effects of
58. St. John JA, Henderson VW, Hodis HN, et al. short-term dietary nitrate supplementation on blood
Associations of urine excretion of isoflavonoids with pressure, O2 uptake kinetics, and muscle and cog-
cognition in postmenopausal women in the Women’s nitive function in older adults. Am J Physiol Regul
Isoflavone Soy Health Clinical Trial. J Am Geriatr Integr Comp Physiol. 2013;304:R73–83. https://doi.
Soc. 2014;62(4):629–35. https://doi.org/10.1111/ org/10.1152/ajpregu.00406.2012.
jgs.12752. 71. Krikorian R, Nash TA, Shidler MD, et al. Concord
59. Henderson VW, St. John JA, Hodis HN, et al. Long- grape juice supplementation improves memory func-
term soy isoflavone supplementation and cognition tion in older adults with mild cognitive impairment.
in women: a randomized, controlled trial. Neurology. Br J Nutr. 2010;103:730–4. https://doi.org/10.1017/
2012;78:1841–8. S0007114509992364.
60. Fournier LR, Ryan-Borchers TA, Robison LM, et al. 72. Wikoff D, Welsh BT, Henderson R, et al. Systematic
The effects of soy milk and isoflavone supplements on review of the potential adverse effects of caffeine con-
cognitive performance in healthy postmenopausal sumption in healthy adults, pregnant women, adoles-
women. J Nutr Health Aging. 2007;11(2):155–64. cents, and children. Food Chem Toxicol. 2017:1–64.
61. Crichton GE, Bryan J, Murphy KJ, Buckley https://doi.org/10.1016/j.fct.2017.04.002.
J. Review of dairy consumption and cognitive per- 73. McLellan TM, Caldwell JA, Lieberman HR. A
formance in adults: findings and methodological review of caffeine’s effects on cognitive, physical
issues. Dement Geriatr Cogn Disord. 2010;30:352– and occupational performance. Neurosci Biobehav
61. https://doi.org/10.1159/000320987. Rev. 2016;71:294–312. https://doi.org/10.1016/j.
62. Kesse-Guyot E, Assmann KE, Andreeva VA, et al. neubiorev.2016.09.01.
Consumption of dairy products and cognitive func- 74. Wu L, Sun D, He Y. Coffee intake and the incident
tioning: findings from the SU.VI.MAX 2 study. J risk of cognitive disorders: a dose-response meta-
Nutr Health Aging. 2016;20(2):128–37. https://doi. analysis of nine prospective cohort studies. Clin
org/10.1007/s12603-015-0593-x. Nutr. 2017;36:730–36.
63. Crichton GE, Murphy KJ, Howe PRC, et al. Dairy 75. Liu X, Du X, Han G, Gao W. Association between
consumption and working memory performance in tea consumption and risk of cognitive disorders: a
overweight and obese adults. Appetite. 2012;59:34– dose - response meta-analysis of observational stud-
40. https://doi.org/10.1016/j.appet.2012.03.019. ies. Oncotarget. 2017;8(26):43306–21.
64. Dai Q, Borenstein AR, Wu Y, et al. Fruit and veg- 76. Gelber RP, Petrovitch H, Masaki KH, et al.
etable juices and Alzheimer’s disease: The Kame Coffee intake in midlife and risk of dementia
Project. Am J Med. 2006;119(9):751–9. and its neuropathologic correlates. J Alzheimers
65. Kent K, Charlton K, Roodenrys S, et al. Dis. 2011;23(4):607–15. https://doi.org/10.3233/
Consumption of anthocyanin-rich cherry juice JAD-2010-101428.
for 12 weeks improves memory and cognition in 77. Feng L, Gwee X, Kua E-H, Ng T-P. Cognitive func-
older adults with mild-to-moderate dementia. Eur tion and tea consumption in community dwelling
J Nutr. 2017;56(1):333–41. https://doi.org/10.1007/ older Chinese in Singapore. J Nutr Health Aging.
s00394-015-1083-y. 2010;14(6):433–8.
66. Alharbi MH, Lamport DJ, Dodd GF, et al. Flavonoid- 78. Ng T-P, Feng L, Niti M, et al. Tea consumption and
rich orange juice is associated with acute improve- cognitive impairment and decline in older Chinese
ments in cognitive function in healthy middle-aged adults. Am J Clin Nutr. 2008;88:224–31.
79. Lamport DJ, Pal D, Moutsiana C, et al. The effect 91. Bovier ER, Renzi LM, Hammond BR. A double-
of flavanol-rich cocoa on cerebral perfusion in blind, placebo-controlled study on the effects of
healthy older adults during conscious resting lutein and zeaxanthin on neural processing speed
state: a placebo controlled, crossover, acute trial. and efficiency. PLoS One. 2014;9:e108178. https://
Psychopharmacology. 2015;232:3227–34. https:// doi.org/10.1371/journal.pone.0108178.
doi.org/10.1007/s00213-015-3972-4. 92. Zamroziewicz MK, Paul EJ, Zwilling CE, et al.
80. Mastroiacovo D, Kwik-Uribe C, Grassi D, et al. Para-hippocampal cortex mediates the rela-
Cocoa flavanol consumption improves cognitive tionship between lutein and crystallized intel-
function, blood pressure control, and metabolic pro- ligence in healthy, older adults. Front Aging
file in elderly subjects: the Cocoa, Cognition, and Neurosci. 2016;8:297. https://doi.org/10.3389/
Aging (CoCoA) Study—a randomized controlled fnagi.2016.00297.
trial. Am J Clin Nutr. 2015;101:538–48. 93. Feart C, Letenneur L, Helmer C, et al. Plasma carot-
81. Sorond F, Hurwitz S, Salar DH, et al. Neurovascular enoids are inversely associated with dementia risk
coupling, cerebral white matter integrity, and in an elderly French cohort. J Gerontol A Biol Sci
response to cocoa in older people. Neurology. Med Sci. 2016;71(5):683–8. https://doi.org/10.1093/
2013;81:904–9. gerona/glv135.
82. Peters R, Peters J, Warner J, et al. Alcohol, demen- 94. Lindbergh CA, Mewborn CM, Hammond BR, et al.
tia and cognitive decline in the elderly: a systematic Relationship of lutein and zeaxanthin levels to neu-
review. Age Ageing. 2008;37:505–12. https://doi. rocognitive functioning: an fMRI study of older
org/10.1093/ageing/afn095. adults. JINS. 2016;22:1–12. https://doi.org/10.1017/
83. Heymann D, Stern Y, Cosentino S, et al. The S1355617716000850.
association between alcohol use and the progres- 95. Vishwanathan R, Schalch W, Johnson EJ. Macular
sion of Alzheimer’s disease. Curr Alzheimer Res. pigment carotenoids in the retina and occipi-
2016;13(12):1356–62. tal cortex are related in humans. Nutr Neurosci.
84. Downer B, Jiang Y, Zanjani F, Fardo D. Effects of 2016;19(3):95–101. https://doi.org/10.1179/147683
alcohol consumption on cognition and regional 0514Y.0000000141.
brain volumes among older adults. Am J Alzheimers 96. Kelly D, Coen RF, Akuffo KO, et al. Cognitive func-
Dis Other Demen. 2015;30(4):364–74. https://doi. tion and its relationship with macular pigment optical
org/10.1177/1533317514549411. density and serum concentrations of its constituent
85. Ritchie SJ, Bates TC, Corley J, et al. Alcohol consump- carotenoids. J Alzheimers Dis. 2015;48:261–77.
tion and lifetime change in cognitive ability: a gene × https://doi.org/10.3233/JAD-150199.
environment interaction study. Age. 2014;36:1493– 97. Renzi LM, Dengler MJ, Puente A, et al.
502. https://doi.org/10.1007/s11357-014-9638-z. Relationships between macular pigment optical
86. Arntzen SA, Schirmer H, Wilsgaard T, Mathiesen density and cognitive function in unimpaired and
EB. Moderate wine consumption is associated mildly cognitively impaired older adults. Neurobiol
with better cognitive test results; a 7-year follow- Aging. 2014;35:1695–9. https://doi.org/10.1016/j.
up of 5033 subjects in the Tromso Study. Acta
neurobiolaging.2013.12.024.
Neurol Scand Suppl. 2010;190:23–9. https://doi. 98. Vishwanathan R, Iannaccone A, Scott TM, et al.
org/10.1111/j.1600-0404.2010.01371.x. Macular pigment optical density is related to cognitive
87. Solfrizzi V, D’Introno A, Colacicco AM, et al. function in older people. Age Ageing. 2014;43:271–
Alcohol consumption, mild cognitive impair- 5. https://doi.org/10.1093/ageing/aft210.
ment, and progression to dementia. Neurology. 99. Johnson EJ, Vishwanathan R, Johnson MA, et al.
2007;68:1790–9. Relationship between serum and brain carotenoids,
88. Stampfer MJ, Kang JH, Chen J, et al. Effects of α-tocopherol, and retinol concentrations and cogni-
moderate alcohol consumption on cognitive function tive performance in the oldest old from the Georgia
in women. N Engl J Med. 2005;352:245–53. Centenarian Study. J Aging Res. 2013;2013:951786.
89. Erdman JW, Smith JW, Kuchan MJ, et al. Lutein and https://doi.org/10.1155/2013/951786.
brain function. Foods. 2015;4:547–64. https://doi. 100. Blumenthal JA, Smith PJ, Mabe S, et al. Lifestyle
org/10.3390/foods4040547. and neurocognition in older adults with cardio-
90. Johnson E, Vishwanathan R, Mohn E, et al. vascular risk factors and cognitive impairment.
Avocado consumption increases neural lutein and Psychosom Med. 2017;79(6):719–27. https://doi.
improves cognitive function. FASEB J. 2015; 29(1) org/10.1097/PSY.0000000000000474.
Supplement 32.8.
Part V
Cancer Prevention and Survival
Dietary Patterns, Whole Plant
Foods, Nutrients 19
and Phytochemicals in Colorectal
Cancer Prevention
and Management
Keywords
Colorectal cancer • Colorectal adenomas • Dietary fiber • Antioxidants
vitamins • Calcium • Magnesium • Selenium • Folate • Isoflavonoids •
Carotenoids • Phenolics • Dietary patterns • Whole plant foods • Fruits •
Vegetables • Legumes • Soy products • Whole-grains • Peanuts • Tree nuts
• Seeds
Key Points Hypertension (DASH), Healthy Eating

Indices, pesco-vegetarian and low inflamma-
• Globally colorectal cancer (CRC) rates have
tory index diets can significantly reduce risk
doubled since the 1970s and incidence is
of colorectal adenoma and CRC.
strongly associated with the Western lifestyle
• Survivors of CRC with high intake of Western
and aging populations.
dietary patterns had significantly higher odds
• As much as 90% of CRC cases may be attrib-
of CRC mortality and recurrence compared to
utable to dietary factors. A number of nutri-
ents and phytochemicals are considered to be those consuming healthy diets.
potentially protective against CRC to various • Dietary patterns rich in fruits, vegetables
degrees including fiber, isoflavones, flavo- (including green leafy vegetables, cruciferous
noids, antioxidant vitamins, carotenoids, and allium vegetables), legumes (including
folate, calcium, magnesium, and selenium. soy), whole-grains (≥3 servings/day) and
• Higher adherence to a Western dietary pattern, peanuts may have protective effects against
which can stimulate a proinflammatory sys- colorectal adenomas and CRC risk.
temic response, can significantly increase risk • Diets rich in dietary fiber have been related to
of colorectal adenomas and CRC, especially a lower CRC risk due in large part to benefi-
in diets high in red or processed meats. In con- cial effects of butyrate, derived from fiber
trast, higher adherence to healthy dietary pat- fermentation by colonic microflora, an inhib-
terns including the Mediterranean diet itor of colonocyte tumor cell initiation and
(MedDiet), Dietary Approaches to Stop progression.

522 19 Dietary Patterns, Whole Plant Foods, Nutrients and Phytochemicals
19.1 Introduction excessive alcohol intake [6–8]. The 2017 World

Cancer Research Fund International Continuous
Colorectal cancer (CRC) is among the most Update Project systematic review and meta-
common cancers worldwide [1]. CRC rates have analysis (45 meta-analyses of prospective stud-
doubled since the 1970s and rose by more than ies; 15 different foods or food groups; 111 unique
200,000 new cases per year from 1990 to 2012 cohort studies) found that foods associated with
[1–4]. In 2015, there were 1.7 million new CRC an increased CRC risk were red and processed
cases globally and 832,000 deaths resulting from meat and alcohol and foods associated with a
CRC. Most cases of CRC are detected in Western decreased CRC risk were whole grains, vegeta-
countries, but this is changing due to the adop- bles, dairy and fish [9]. Foods not associated
tion of Western dietary habits in developing with CRC risk were fruits, coffee, tea, poultry,
countries over the past few decades [1–4] cheese and legumes. Consistent with these
Incidence is strongly associated with the Western findings, the 2017 National Institutes of
lifestyle and aging populations with 2.4 million Health - American Association of Retired
new CRC cases projected to be diagnosed in Persons (NIH)-AARP) Diet and Health Study
2035 [2–4]. By gender, CRC is the second most (398,458 middle-aged and older adults; 10 years
common cancer in women (9.2%) and the third of follow-up) found that among normal-weight
in men (10%). Despite an increasing number of and overweight men, CRC risk was 25%-30%
tumors now diagnosed, the CRC mortality rate lower with high adherence to healthy dietary pat-
has decreased because of more appropriate and terns [10]. High adherence to Western dietary
available information, earlier diagnosis, and patterns characterized as energy dense, high red/
improvements in treatment. A recent meta- processed meat, salty foods and sweets, and
analysis of 43 studies showed CRC risk was refined grains or higher alcohol consumption are
associated with a variety of lifestyle factors associated with an increased CRC risk, whereas
including: diet, smoking, alcohol, body fatness, high adherence to a healthy dietary pattern char-
physical activity, medication and/or hormone acterized as low to moderate energy dense, high
replacement therapy [5]. As much as 90% of in dietary fiber, fruits and vegetables, legumes,
CRC cases appear to be attributable to dietary whole-grains, and nuts are associated with
factors including unhealthy diet, obesity and decreased CRC risk (Figs. 19.1 and 19.2) [11, 12].
30
25
20
Risk of Colorectal Adenoma (%)
15
10
0
Healthy (Plant-based) Dietary Unhealthy (Western) Dietary
–5 Pattern Pattern
–10
–15
–20
–25
Fig. 19.1 Association between dietary pattern quality and colorectal adenoma risk from a meta-analysis of seven-
cohort and five-case-control studies (p < .004) (adapted from [11])
19.2 Nutrients and Phytochemicals 523
50
40
30
Colorectal Cancer Risk (%)
20
10
0
Healthy Western-style Alcohol-consumption
–10
–20
Dietary Patterns
–30
Fig. 19.2 Association between dietary pattern and colorectal cancer risk from a meta-analysis of 40 observational
studies (p <.003) (adapted from [12])
An overview of the foods and nutrient composi- array of cancer protective cellular regulatory
tion of the Western and common healthy dietary vprocesses and promote a healthier colonic micro-
patterns are provided in Appendix A. The objec- biome ecosystem. Table 19.1 summarizes nutrient
tive of this chapter is to provide a comprehensive and phytochemical systematic reviews and meta-
review of the effect of dietary patterns and whole and pooled analyses in CRC and colorectal ade-
(minimally processed) foods, nutrients and phy- noma risk [17–38].
tochemicals on CRC risk.
19.2.1 Dietary Fiber

19.2 Nutrients and Phytochemicals
The fiber protective effects in CRC and colorectal
A number of nutrients and phytochemicals are adenoma risk have been extensively evaluated in
potentially associated with lower CRC risk to systematic reviews and meta-and pooled analyses
various degrees including by slowing or prevent- of RCTs and observational studies [17–20].
ing adenoma-carcinoma progression and protect- A 2017 meta-analysis (5 RCTs; 4798 subjects;
ing normal epithelium cells from acquiring wheat bran, psyllium, higher fiber diets;
genetic and epigenetic mutations in specific onco- 2–8 years) found no statistically significant low-
genes or tumor suppressor genes [13–16]. The ering of adenomatous polyp recurrence or inci-
change from adenoma polyp to adenocarcinomas dence with increased fiber intake; however, the
is usually a slow process over an average period authors concluded that these results should be
of 10 years. Healthy plant-based dietary patterns interpreted with caution, because of the: (1) high
rich in fruits, vegetables, legumes, whole-grains, subject loss rate to follow-up; (2) issues with
and nuts are important sources of bioactive com- dietary compliance; and (3) questions about the
pounds, which protect against the development of reliability of adenomatous polyps as a surrogate
CRC and include fiber, isoflavones, flavonoids, biomarker of CRC risk RCTs that are <10 years
antioxidant vitamins, carotenoids, folate, calcium, duration [17]. A 2006 pooled analysis of the
magnesium, and selenium. These plant phyto- Wheat Bran Fiber Trial (WBFT) and Polyp
chemicals and nutrients may influence a complex Prevention Trial (PPT) (3209 participants, mean
Table 19.1 Summaries of systematic reviews and meta-and pooled analyses on nutrients and phytochemicals in
colorectal cancer (CRC) and colorectal adenoma risk
Dietary fiber (fiber)
Yao et al. (2017) . Assess the 5 RCTs; 4798 participants, age This pooled analysis of RCTs showed no
effect of fiber on the recurrence range 56–66 years; wheat bran, statistically significant difference between the
of colorectal adenoma in psyllium, or a comprehensive fiber-rich intervention and lower fiber control
people with a known history of dietary intervention with high fiber groups. However, these results may be
adenomatous polyps and on the whole food sources alone or in unreliable and should be interpreted
incidence of CRC compared to combination vs. low-fiber diets, cautiously, not only because of the high rate of
placebo (Cochrane Database placebo, or a regular diet; subjects loss to follow-up and issues with
Systematic Review) [17] 2–8 years of follow-up dietary compliance
Ben et al. (2014). Conduct a 20 studies (16 case-control studies High fiber intake, especially fruit and cereal
meta-analysis primarily of and 4 cohort studies); approx. fiber, was associated with reduced colorectal
case-control studies to analyze 150,000 subjects and 10,948 adenoma risk. The pooled analysis found a
the association between fiber subjects with colorectal adenoma significantly inverse colorectal adenoma risk
intake and risk of colorectal effect per 10 g/day of total fiber by 9%, fruit
adenoma [18] fiber by 21%, and cereal fiber by 30%. The
effects were similar for both early and
advanced colorectal adenoma
Aune et al. (2011). Assess the 25 prospective cohort studies; 19 A 10 g/day increase in total and cereal fiber
association between and fiber were included in the dose-response intake was associated with a 10% lower CRC
intake and CRC risk from analyses; 14,514 CRC cases among risk with a significant reduction in risk of
prospective cohort studies 1,985,552 subjects; ranges of fiber colon cancer but not rectal cancer. Fruit,
(World Cancer Research Fund/ intake: total fiber (6.3–21.4 g/day), vegetable, and legume fiber intake were
American Institute for Cancer fruit fiber (1.8–15.5 g/day), suggestive for CRC risk reduction but not
Research Continuous Update vegetable fiber (1.9–16.8 g/day), confirmed.
Report) [19] cereal fiber (3.0–16.9 g/day), and
legume fiber (1.3–3.8 g/day)
Jacobs et al. (2006). Determine 3209 participants combined from 2 This pooled analysis shows that increased
the pooled effects of increased RCTs; mean baseline age 64 years; fiber intake was more effective in lowering
fiber intake on colorectal approx. 65% men; analyzed with colorectal adenoma recurrence in men than in
adenoma recurrence in men and logistic regression models to women, which may help to explain some of
women from 2 large clinical examine the effect of a dietary the discrepant results reported from previous
intervention trials (The Wheat intervention on colorectal trials. For the total pooled population, the
Bran Fiber Trial (WBF) and the adenoma recurrence adjusted adenoma recurrence risk was
Polyp Prevention Trial (PPT); 1. The WBF trial subjects with insignificantly reduced by 9%. For men, the
US) [20] recent colorectal adenomas intervention was associated with statistically
removal were randomly assigned significantly reduced risk of recurrence by
to receive either 13.5 or 2.0 g fiber/ 19%; for women, no significant association
day as a breakfast cereal (wheat was observed. There was a statistically
bran) supplement for 3-years. significant interaction between fiber intake
level and sex (p = .03)
2. The PPT subjects with recent
colonic polyp removal were
randomized into an intervention
diet of decreased fat and increased
fiber, fruit, and vegetables or
control diet for 4-years
Soy Isoflavones
Jiang et al. (2017). Meta- 5 cohort studies and 5 case-control Case-control studies showed lower CRC risk
analysis was conducted to studies for isoflavones by 23%, but for cohort studies
quantify the association the pooled CRC risk was reduced by 6%.
between isoflavones and CRC Dose–response analysis yielded an 8% reduced
risk [21] risk of colorectal neoplasms for every 20 mg/
day increase in isoflavones intake in Asians
Yu et al. (2016). Perform a 17 studies (13 case-control and 4 This analysis found that soy isoflavone
meta-analysis to assess the prospective cohort studies); consumption reduced CRC risk by 22%
association between soy 273,765 subjects (p = .024). Based on subgroup analyses, a
isoflavone consumption and significant protective effect was observed with
CRC risk in humans [22] in participants with higher intake of soy foods/
products and in Asian populations by 21%
Tse and Eslick (2016). 22 case-control and 18 cohort Although this analysis showed that soy food
Determine the potential studies; 633,476 participants and intake was only associated with a small
relationship between dietary 13,639 CRC cases reduction in CRC risk by 8% (p = .3),
soy intake and CRC risk with subgroup analysis for isoflavone intake
an evaluation of the effects of suggests a strong inverse association with a
isoflavone as an active soy significant reduced CRC risk by 24%
constituent [23]
Flavonoids (Phenolics)
Grosso et al (2016). Summarize 20 studies (9 case-control and 11 Cohort studies show that flavonoids
observational studies on the cohort studies) insignificantly lower CRC risk, while several
association between dietary significant associations were found among
flavonoid intake and cancer risk case-control studies, including flavonols,
using meta-analysis [24] quercetin, subclasses of flavanols
proanthocyanidins and catechins, and
anthocyanins
Jin et al. (2012). Assess the 8 studies (5 cohort studies, 2 Increased intake of flavan-3-ols, epicatechins
effect of dietary flavonoids on case-control studies and one significantly reduced the risk of both CRC
the incidence of colorectal RCT); 390,769 participants and/or colorectal adenomas. There was
adenoma and CRC (Cochrane medium quality evidence to support that
Database Systematic Review) increased intake of procyanidin lowers the
[25] incidence of CRC but no significant evidence
for anthocyanin intake
Antioxidant vitamins
Xu et al. (2013). Assess the 13 studies (12 case–control studies In highest vs. lowest analysis, dietary intake
association between dietary and one nested case–control of vitamin C reduced the risk of colorectal
intake of vitamins A, C, and E study); 3832 individuals with adenoma by 22% independent of BMI,
and the risk of colorectal colorectal adenoma smoking status, and dietary energy intake
adenoma, a potential precursor (p = .005); with a 9% risk reduction per
of CRC by meta-analysis [26] 100 mg vitamin C. Also, dietary intake of
β-carotene was inversely associated with the
risk of colorectal adenoma by 53% (p = .009).
However, dietary intake of vitamins A and E
were insignificantly associated with the risk of
colorectal adenoma by 13% each
Pais and Dumitrascu ( 2013). 20 RCTs: 12 on CRC incidences; Overall antioxidant supplement intake had an
Estimate the effectiveness of 250,676 participants; and eight insignificant effect on CRC risk or colorectal
antioxidants on CRC incidence analyzed colorectal adenoma adenoma recurrence with a pooled reduction
and adenomatous polyp recurrence; 17,914 participants by 6% (p = .32). Also, beta carotene was
recurrence [27] insignificantly associated with CRC risk
Park et al. (2010). Evaluate 13 cohort studies; 209,263 men Colon cancer risk was significantly reduced
the associations between and 466,878 women; 7–20 years by 12% (>4000 vs. ≤1000 μg/day) for vitamin
intakes of vitamins A, C, and E of follow-up; 5454 colon cancer A, 19% (>600 vs. ≤100 mg/day) for vitamin
and risk of colon cancer (the cases C, and 22% (>200 vs. ≤6 mg/day) for vitamin
Pooling Project of Prospective E. Adjustment for total folate intake
Studies of Diet and Cancer) attenuated these associations, but the vitamins
[28] C and E inverse associations retained their
significant lowering effects
(continued)
Carotenoids
Panic et al. (2017). Systematically
22 studies (16 case-control studies Cohort studies showed no significant
review the epidemiological and 6 cohort studies); 486,393 association between intake of specific
evidence for the association participants carotenoids from natural sources, as well as
between carotenoid intake from combined carotenoids, and the risk of CRC
natural sources and CRC risk overall and by anatomic subsite with no
[29] evidence of heterogeneity among studies.
Case-control studies found insignificant
inverse associations with colon cancer
incidence for β-carotene by 22%, lycopene by
5%, and lutein/ zeaxanthin by 11%
Mannisto et al. (2006). 11 cohort studies (in North Lutein/zeaxanthin (highest vs. lowest quintile
Analyze the associations America and Europe); 702,647 of intake) showed a lower pooled multivariate
between intakes of specific participants; 6–20 years of CRC risk by 8% (p-trend = .08). There were
carotenoids (α-carotene, follow-up; 7885 CRC cases; no other borderline significant trends for other
beta-carotene, β-cryptoxanthin, carotenoid intakes were estimated carotenoids
lutein and zeaxanthin, and from food frequency
lycopene) and CRC risk [30] questionnaires (FFQ) at baseline in
each study
Folate
Figueiredo et al. (2011). 3 large RCTs; 2632 men and After 42 months of folic acid use, there was
Assess the effect folic acid women with a history of no clear decrease or increase in the occurrence
intake on the prevention of adenomas; either 0.5 or 1.0 mg/ of new adenomas in patients with a history of
colonic adenomas by pooled day of folic acid vs. placebo; mean adenoma. In participants with high alcohol
analysis of RCTs [31] follow-up of 6–42 months (mean intake, there was a trend for decreasing
of 31 months) colorectal adenoma with higher folic acid
intake. During the early follow-up reported
here, more deaths occurred in the placebo
group than in the folic acid group (1.7% vs.
0.5%, p = .002)
Kim et al. (2010). Evaluate the 13 cohort studies; 725,134 In dose-response analysis, every 100 μg/day
overall and dose effects of participants (229,466 men and increase in total folate intake reduced CRC
folate intake on CRC risk [32] 495,668 women); 7–20 years of risk by 2%. Comparing the highest vs. lowest
follow-up; 5720 colon cancer quintile of intake showed reduced pooled risk
cases by 8%. Comparing daily total folate intake of
≥560 mcg vs. <240 mcg, CRC risk was
reduced by 13% (p-trend = .009)
Calcium
Bonovas et al. (2016). 4 RCTs; daily calcium ranged Higher calcium intake significantly lowered
Determine the efficacy of from 1200 to 2000 mg/day; risk of colonic adenomas by 13% (p < .05)
calcium supplementation in 36–60 months and insignificantly lowered advanced
reducing the recurrence of adenomas by 8%
colorectal adenomas [33]
Keum et al. (2015). Evaluate 8 prospective studies; 11,005 For total calcium intake, each 300 mg/day
dose-response effects of subjects; total calcium intake reduced adenoma risk by 5%; evidence of
calcium intake on colorectal ranging from 333 to 2229 mg/day nonlinearity was indicated with risk
adenoma risk [34] reductions for 1000 mg/day by 8% and for
1450 mg/day by 13% (p-nonlinearity < .01).
Calcium intake effects were stronger for
high-risk adenomas (≥1 cm in diameter) with
1000 mg calcium/ day reducing risk by 23%
Magnesium
Chen et al. (2012). Assess the 8 prospective studies; 338,979 For the highest vs. lowest magnesium intake,
association between participants; 7.9–28 years of the pooled CRC risk was lowered by 11%. For
magnesium intake and CRC follow-up; 8000 CRC cases colon cancer, the pooled risk was reduced by
risk [35] 19%. For rectal cancer, the pooled risk was
lowered by 6%. In the dose-response analyses,
50 mg magnesium/day lowered colon cancer
risk by 7%
Wark et al. (2012). Evaluate 6 prospective cohort studies on Each 100-mg/day increase in magnesium
the dose-response effect of CRC; three case-control studies of intake was associated with 13% lower
dietary magnesium on CRC colorectal adenomas colorectal adenomas risk and 12% lower CRC
risk [36] risk. Magnesium intake was inversely
associated with risk of colorectal adenomas
and CRC
Selenium
Cai et al. (2016). Investigate 10 studies (2 RCTs; 4 cohort This analysis showed that higher selenium
the association between studies; 4 case-control studies) exposure was associated with an
selenium intake and cancer risk insignificantly lower CRC risk by 11% with
by meta-analysis [37] moderate heterogeneity
Pais and Dumitrascu (2013). 5 RCTs (3 analyzed CRC Selenium supplementation was associated
Estimate the effectiveness of incidences and 2 analyzed with an insignificant lowering of both CRC
selenium on CRC incidence colorectal adenoma recurrence) incidence by 12% (p = .59) and colorectal
and adenomatous polyp adenoma recurrence by 30% (p = .16)
recurrence [27]
Takata et al. (2012). Evaluate WHI Observational Study: 804 Within the WHI, selenium concentrations
the effect of selenium intake on CRC cases and 805 matched were relatively high (mean 135.6 mg/L) and
CRC risk in women in a nested controls; meta-analysis: 12 were not associated with CRC risk
case-control study (Women’s observational studies and 2 RCTs (p-trend = .10). Meta-analysis (highest vs.
Health Initiative (WHI) lowest quantile) showed that increased
Observational Study) and selenium intake insignificantly reduced CRC
meta-analysis of men and risk in women by 3% and in men CRC risk
women [38] was significantly reduced by 32% (p = .01)
baseline age 64 years, 64% men; WBFT: of 10 g fiber significantly reduced colorectal ade-
increased wheat bran 13.5 vs. 2.0 g/day from nomas risk for total fiber by 9%, fruit fiber by
cereal, 3 years; and PPT: diet lower in fat, higher 21%, and cereal fiber by 30% [18]. A 2011 dose
in fiber and fruits and vegetables, 4 years;) response meta-analysis affiliated with the World
showed that men in the study had significantly Cancer Research Fund (WCRF) and American
reduced colon adenoma recurrence risk by 19% Institute for Cancer Research (AICR) Continuous
whereas no significant association was observed Update Project (19 cohort studies used for dose
for women (multivariate adjusted) [20]. Using a response analysis; 1,985,552 participants) found
likelihood-ratio test, a statistically significant a modest, statistically significant fiber dose
interaction between fiber intake and sex was response for a 10% lower CRC risk per 10 g of
found (p = .03). Two systematic reviews and total and cereal fiber consumed daily [19].
meta-analyses of cohort and case-control studies There are a number of potential fiber related
demonstrate the efficacy of increased fiber intake CRC prevention mechanisms. Fiber and fiber
on CRC risk [18, 19]. A 2014 dose meta-analysis metabolites can provide direct protection
on the effects of fiber intake and colorectal ade- throughout the colorectal tract by [39–50] includ-
nomas (16 case-control and 4 cohort studies; ing: (1) reducing colorectal exposure to carcino-
150,000 subjects) found that an increased intake gens by increasing stool weight, decreasing
transit time and diluting colonic carcinogenic the stage of CRC, dietary isoflavones appear to
concentrations, (2) increasing colonic butyrate increase estrogen receptor-β expression and
levels (via fiber fermentation) which is the pre- exhibit anti-proliferative effects.
ferred energy source for colon epithelial cells for
optimal barrier protection and anti-inflammatory
effects, (3) lowering colonic pH to protect against 19.2.3 Flavonoids (Phenolics)
colonic pathogenic bacterial infections and
reduce bacterial enzymes, including Whole plant foods, including soy, berries, vege-
7α-dehydroxylase, which is involved in the for- tables, cereals and nuts, as well as tea and coffee
mation of secondary bile acids, and other bacte- are rich in flavonoids or other phenolic com-
rial enzymes known to convert relatively harmless pounds [13]. Because most of these plant pheno-
compounds to reactive toxic metabolites and (4) lics are poorly absorbed, they may accumulate in
promoting colonic anti-cancer activity by increas- the colon up to the millimolar range and become
ing the Warburg effect, arresting the growth of subjected to conversion by microorganisms into
tumors by histone deacetylase (HDAC) inhibitor metabolites with potential anti-cancer biological
and anti-inflammatory properties mediated by activity. Phenolic compounds in the colon have
suppressing the activation of nuclear factor-kB (a been found to alter the microbial population by
transcription factor controlling the expression of suppressing the growth of Clostridium and
genes encoding proinflammatory responses and Bacteroides species to promote a healthier micro-
inflammatory mediators like tumor necrosis biota that may promote reduced CRC risk but
factor-α (TNF-α) and nitric oxide) [37–43]. Also, two meta-analyses of prospective cohorts do not
increased fiber intake is associated with lower confirm a significant protective role for higher
dietary energy density and metabolizable energy flavonoid intake against CRC risk [24, 25].
for better weight maintenance, reduced risk of Cohort studies show that higher flavonoid intake
central obesity, and attenuation of colonic micro- is not significantly associated with lower CRC
biota inflammatory activity [45–48]. and adenomas risk while case-control studies
suggest a possible protective CRC role for flavo-
nols, quercetin, subclasses of flavanols proantho-
19.2.2 Soy Isoflavones cyanidins and epicatechins, and procyanidins.
Three meta-analyses consistently support CRC

risk lowering effects for soy isoflavones by a 19.2.4 Antioxidant Vitamins
pooled mean of 23% [21–23]. The association of
isoflavones with reduced CRC risk was statisti- Antioxidant vitamins A, C, and E, have potent
cally significant only in case-control studies but antioxidative and anti-inflammatory properties
not in cohort studies, which are less subject to [6] with the potential to reduce CRC risk by act-
recall and selection bias. Dose response analysis ing as reactive oxygen species scavengers and
showed an 8% lower risk of colorectal neoplasms protecting cells from oxidative stress and inflam-
per 20 mg/day increase of isoflavones in Asian mation, which can initiate and promote carcino-
populations. The mechanism by which soy isofla- genesis by inducing gene mutations, DNA
vones protect against the development of CRC damage, genome instability, and cell prolifera-
remains unclear [21, 22]. Isoflavones have simi- tion. A 2010 meta-analysis of antioxidant vita-
lar chemical structure to estrogen, which allows mins from foods and supplements (10 cohort
their binding to estrogen receptors with an espe- studies; 209,263 men and 466,878 women;
cially high binding affinity to estrogen receptor β, 7–20-years of follow-up) found that the highest
the predominant estrogen receptor in normal intake of antioxidant vitamins C and E lowered
colon mucosa. Though estrogen receptor β pro- CRC risk by 12–22% compared to the lowest
gressively decreases with advancing grade and intake, after adjusting for total folate intake [28].
A 2013 meta-analysis (13 case-control studies) genomic hypomethylation and defects in DNA
showed that higher intake of vitamin C signifi- synthesis, which may contribute to colonic carci-
cantly reduced the risk of colorectal adenomas nogenesis. Dietary intake and circulating levels of
by 22%, with a 9% risk reduction per 100 mg folate have been inversely associated with CRC
vitamin C [26]. Also, higher dietary intake of and adenoma risk in observational studies with the
β-carotene was inversely associated with the risk association being more effective among heavy
of colorectal adenomas by 53% (p = .009). In alcohol consumers because of alcohol’s impair-
contrast, a 2013 meta-analysis of RCTs (12 ment of folate mediated DNA methylation. A 2010
RCTs on CRC risk; 50,676 subjects; and 8 RCTs meta-analysis of folate intake effects on CRC risk
on colorectal adenoma recurrence; 7914 sub- (13 cohort studies; 725,134 participants; 7–20 years
jects) found increased overall antioxidant sup- of follow-up) found a significant reduction in CRC
plement intake had an insignificant 6% lowering risk by 2% per every 100 μg/day increase in total
effect on CRC incidence or colorectal adenoma folate intake [31]. Comparing daily total folate
recurrence risk (p = .32) [27]. intake of ≥560 mcg vs. <240 mcg was associated
with a significantly lower CRC risk by 13%
(p-trend = .009). However, three large RCTs (2632
19.2.5 Carotenoids men and women with a history of adenomas; either
0.5 or 1.0 mg/day of folic acid or placebo;
Carotenoids represent a diverse group of natural ≥42 months of follow-up endoscopy) found, after
pigments present in non-starchy vegetables and 42 months of folic acid use there was no clear
fruit. The carotenoids from natural sources can be decrease or increase in the occurrence of new ade-
classified into two groups: hydrocarbons, such as nomas in patients with a history of adenoma [32].
α-carotene, β-carotene, and lycopene; and xan-
thophylls, such as β-cryptoxanthin, lutein, and
zeaxanthin [29]. Several potential mechanisms 19.2.7 Calcium
support protective effects of carotenoids in CRC
development by functioning as a provitamin-A Calcium’s CRC protective mechanisms include
and influencing cellular differentiation and prolif- its: (1) involvement in the formation of insoluble
eration, or by neutralizing free radicals to prevent soaps with potential carcinogenic free fatty acids
colorectal colonic cell and tissue damage. A 2017 and bile acids in the colonic lumen and (2) contri-
systematic review and meta-analysis (16 were bution to the integrity of the intestinal barrier
case-control studies and 6 cohort studies) found function and homeostasis between the microbiota
no association between the intake of individual and their effect on immune response, thought to
and total carotenoids and the risk of CRC overall be mediated by extracellular calcium-sensing
and by anatomic subsite [29]. One cohort study receptor signaling [6]. Meta-analysis of both
found for lutein and zeaxanthin (highest vs. low- cohort studies and RCTs show that calcium sig-
est quintile of intake) a trend toward reduced CRC nificantly lowers the risk of new or recurrent
risk by 8% (p-trend = .08) [30]. colorectal adenomas [33, 34]. A pooled analysis
of RCTs (four RCTs; calcium 1200–2000 mg/
day, 36–60-month duration) found that increased
19.2.6 Folate calcium significantly reduced colonic adenomas
by 13% (p < .05). A dose response meta-analysis
Since folate (vitamin B-9) is essential for DNA (Eight cohort studies; 11,000 subjects) showed
methylation, synthesis, stability, and repair, it has that each 300 mg calcium/day lowered adenoma
been extensively investigated for its protective role risk by 5% and there were additional protective
in CRC [6]. The primary sources of folate are from effects against high-risk adenomas (≥1 cm in
the diet or supplements and by production of folate diameter) with 1000 mg calcium/day further
by colonic bacteria. Folate deficiency results in reducing risk to 23% [34].
19.2.8 Magnesium however, in men, there was a significant inverse

association with a 32% lower risk [38].
Magnesium is an essential mineral, which is most
notably present in foods rich in dietary fiber,
non-starchy vegetables, fruits, and nuts [35]. 19.3 Dietary Patterns
Magnesium is required for many physiologic
processes that affect CRC risk, including DNA 19.3.1 Colorectal Cancer Risk (CRC)
synthesis and repair, glucose metabolism and
insulin sensitivity, the regulation of cell prolifera- Table 19.2 summarizes seven systematic reviews,
tion and apoptosis and defense against oxidative meta- or pooled analyses and seven representa-
stress and inflammatory responses, which may tive prospective cohort studies on the effects of
influence colorectal carcinogenesis. Several dietary pattern quality and CRC risk [11, 12,
meta-analyses of cohort studies indicate that 51–62]. Meta-analyses consistently show that
increased magnesium intake was inversely asso- dietary pattern quality has a significant effect on
ciated with risk of colorectal adenomas and CRC colorectal adenomas and CRC risk (Figs. 19.1
[35, 36]. Each 100 mg/day increase in magne- and 19.2) [11, 12, 51–55]. A higher adherence
sium intake was associated with 13% lower score to a Western dietary pattern or proinflam-
colorectal adenomas risk and 12% lower CRC matory diet significantly increased risk of adeno-
risk [36]. mas by 13% a nd CRC risk by 29–65%, especially
with high intake of red or processed meats.
Higher adherence to healthy dietary patterns
19.2.9 Selenium including the Mediterranean diet (MedDiet),
Dietary Approaches to Stop Hypertension
Selenium is not an antioxidant by itself but is (DASH), Healthy Eating Indices (HEI), vegetar-
required for the anti-oxidative activity of seleno- ian, and low inflammatory index diets can signifi-
enzymes [6]. There is evidence for a U-shaped cantly reduce colorectal adenoma risk by 19%
relationship between selenium status and protec- and CRC risk by 8–65%. Among vegetarian
tion from cancer, with an optimal circulating diets, the pesco-vegetarian diet was most effec-
level of selenium within the range of 130– tive at lowering CRC risk by 33% compared to a
150 mcg/L. Several potential anticarcinogenic non-vegetarian diet [51]. Specific prospective
mechanisms of selenium include: (1) contribut- cohort studies show the effects of healthy dietary
ing to the antioxidant function of glutathione per- patterns on colorectal adenoma and CRC risk
oxidases and thioredoxin reductases: (2) [56–62]. In postmenopausal women, the HEI-
association with the regulation of protein folding 2010 and DASH diet significantly lowered CRC
via the function of the endoplasmic reticulum to risk by 22–28% [56]. The Adventist Health Study
influence the process of necrosis and apoptosis of (77,659 subjects; mean 7.3 years of follow-up)
malignant cells; (3) effects on DNA stability. found a wide variation of CRC lowering effects
Conversely, the adverse effects of excess sele- for the various types of vegetarian diets com-
nium intake are possibly increasing risk of diabe- pared to non-vegetarians (Fig. 19.3) [58]. The
tes, glaucoma, and dermatologic alterations [6, pesco-vegetarian diet reduced CRC risk by 43%
37, 38]. Three meta-analysis including both vs. 22% for the composite of all vegetarian diets.
observational studies and RCTs showed that Vegetarians and 1 day per week meat eaters had
increased selenium insignificantly reduces CRC 27% lower CRC risk compared to 6–7 day/week
risk in the total population [27, 37, 38]. One meat eaters [57]. The pooled data from the
meta-analyses (highest vs. lowest quantile) found Nurses’ Health and Health Professional
that selenium intake was insignificantly associ- Follow-up Studies (87,256 women and 45,490
ated with reduced CRC risk in women by 3%; men; up to 26-years of follow-up) found that the
Table 19.2 Summaries of dietary pattern studies in colorectal adenomas and colorectal cancer (CRC) risk
Systematic reviews and meta or pooled analyses
Godos et al. (2016). 7 cohort studies: 94,217 Higher adherence to healthy dietary patterns was
Evaluate the association subjects; 3–12 years of significantly associated with lower risk of colorectal
between diet quality follow-up; 7384 colorectal adenomas by 19% and unhealthy dietary patterns had an
(plant based dietary adenoma cases; and 5 increased risk by 13% (Fig. 19.1) with no evidence of
patterns vs. meat-based or case-control studies: 3682 heterogeneity between type of dietary patterns
other Western dietary controls and 1578 cases of
patterns) and colorectal colorectal adenomas
adenoma risk [11]
Feng et al. (2016). 40 observational studies (22 For the highest vs. lowest adherence to dietary patterns,
Review the association cohort studies, 27 the ‘healthy’ pattern was associated with a lower risk
between dietary patterns case-control studies, and 1 for CRC by 25% (p < .00001), Western-style pattern
and CRC risk [12] cross-sectional study) increased risk of CRC by 40% (p < .00001) and alcohol
consumption pattern raised CRC risk by 44% (p = .003)
(Fig. 19.2)
Godos et al. (2016). 4 cohort studies; approx. There was a significantly lower risk of CRC associated
Assess the association 642,000 subjects; with a semi-vegetarian diet by 14%
between vegetarian diets 7–20 years of follow-up (p-heterogeneity = .82) and a pesco-vegetarian diet by
and colorectal cancer risk 33% (p-heterogeneity = .46) compared to a non-
[51] vegetarian diet
Steck et al. (2015). 5 case-control studies and Comparing highest to lowest score groups, higher
Examine the effects of seven cohort studies MedDiet scores were associated with an 8–54% lower
index-based dietary CRC risk, higher HEIs were associated with a 20–56%
patterns on CRC risk lower CRC risk. Elevated DII proinflammatory diet
(Mediterranean Diet scores were associated with a 12–65% higher CRC risk
(MedDiet) score, Healthy compared with anti-inflammatory rich diets. Low CRC
Eating Index (HEI), and risk diets consisted of higher intake of plant-based
Dietary Inflammatory foods and lower intake of red or processed meats
Index (DII)) [52]
Azeem et al. (2015). 16 observational studies in 1. There was a positive (adverse) association between
Assess correlations Asian populations intake of red meats, processed meats, preserved foods,
between various diet saturated and animal fats, cholesterol, high sugar foods,
types, food or nutrients spicy foods, tubers or refined carbohydrates and CRC
and colorectal cancer risk risk. 2. There was inverse (protective) association for
among Asian populations the intake of calcium/dairy foods, vitamin D, general
[53] vegetable/fruit/fiber intake, cruciferous vegetables, soy
bean/soy products, selenium, vitamins C, E and B12,
lycopene, alpha-carotene, beta-carotene, and folate and
CRC risk
Yusof et al. (2012). Six cohort studies;1.2 Protective CRC dietary patterns include: healthy and
Identify associations million subjects; 5–10-year prudent patterns; healthy eating index, alternate healthy
between specific dietary follow-up eating index, and Mediterranean score and protective
patterns and risk of CRC foods included: fruits and vegetables, whole-grains, fat
[54] reduced diets, dairy, fish, poultry. Elevated CRC risks
are shown from Western and traditional meat-eating
patterns and foods including: pork processed meat,
potatoes, and refined grains
Magalhaes et al. (2012). 8 prospective cohort studies 1. The healthy pattern, rich in fruit and vegetables,
Evaluate the relationship and 8t case-control studies reduced risk of colon cancer by 20% but had no effect
between dietary patterns on rectal cancer
and CRC risk [55] 2. The Western pattern, high in red and processed
meats, increased risk of colon cancer by 29% and rectal
cancer by 19%
(continued)
Colorectal cancer (CRC) risk
Vargas et al. (2016). 78,273 postmenopausal Higher adherence to HEI-2010 and DASH dietary
Examine associations women; mean baseline age recommendations were inversely associated with risk of
between dietary pattern 63 years and BMI 27; mean CRC in this cohort of postmenopausal women by
scores and CRC risk 12.4-years follow-up; 938 22–28% lower risk of CRC (highest vs. lowest quintile;
(Women’s Health diagnosed with CRC and p < .01). No associations were observed between any
Initiative Observational 238 died from CRC diet quality score and CRC specific mortality
Study; US) [56] (multivariate adjusted)
Gilsing et al. (2016). 10,210 individuals; 1040 Vegetarians and 1 day/week meat eaters showed a
Evaluate the effect of self-defined vegetarians; modest, borderline significantly decreased CRC risk by
vegetarian and low meat 20.3-year of follow-up; 437 approx 27% compared to daily meat eaters. Most of
diets on CRC risk CRC cases (multivariate CRC risk lowering was explained by intake of higher
(Netherlands Cohort adjusted) intake of fiber and soy products
Study -Meat Investigation
Cohort) [57]
Orlich et al. (2015). 77,659 subjects; mean All vegetarians combined vs. non-vegetarians showed
Evaluate the association 7.3 years of follow-up; 380 reduced risk by 22% for all colorectal cancers, colon
between vegetarian cases of colon cancer and cancer by 19% and rectal cancer by 29%; CRC risk was
dietary patterns and CRC 110 cases of rectal cancer reduced in vegans by 16%, lacto-ovo vegetarians by
incidence (The Adventist (multivariate adjusted) 18%, pesco-vegetarians by 43% and in semi-vegetarians
Health Study 2; US) [58] by 8% compared with nonvegetarians (Fig. 19.3). Effect
estimates were similar for men and women and for
black and nonblack individuals
Nimptsch et al. (2014). 17,221 women completed a A higher “prudent” pattern during high school,
Examine associations retrospective high school characterized by high consumption of vegetables, fruit
between adolescent food frequency and fish was associated with a significantly lower risk of
dietary patterns (derived questionnaire in 1998 when rectal adenomas by 55% (p-trend = .005), but not colon
using factor analysis) and they were 34–51 years old, adenomas. A higher Western dietary pattern during high
risk of colorectal and subsequently school, characterized by high consumption of desserts
adenoma in middle underwent an endoscopy by and sweets, snack foods and red and processed meat,
adulthood (Nurses’ 2007; 1299 women were was significantly associated with increased risk of rectal
Health Study II; US) [59] diagnosed with ≥1 adenomas by 78% (p-trend = .005) and advanced
colorectal adenoma adenomas by 58% (p-trend = .08), but not associated
(multivariate adjusted) with colon or non-advanced adenomas
Fung et al. (2010). 87,256 women and 45,490 Comparing higher to lower quintiles of the DASH
Evaluate associations men (baseline age score, the pooled risk was reduced for CRC by 20%
between the Alternate 30–55 years for women and (p- trend = .0001) and for colon cancer by 19%
MedDiet (aMed) and the baseline age 40–75 years (p-trend = .002); higher adherence to the aMed score
DASH-style diet scores for men) without a history was associated with borderline significant lower CRC
with risk of CRC in of cancer; up to 26-year risk by 11% (p = .06) (Fig. 19.4)
middle-aged men and follow-up; aMed and
women (Nurses’ Health DASH scores were
Study and Health calculated up to seven times
Professionals Follow-Up during follow-up; 1432
Study; US) [60] cases of incident colorectal
cancer among women and
1032 cases in men
Reedy et al. (2008). 492,306 subjects; mean For men, there was a significant decreased CRC risk by
Compare healthy dietary baseline age 62 years and 25–30% that was comparable across all indexes when
indices: Healthy Eating BMI 27; 5 years of comparing the highest vs. lowest quintile scores. For
Indices, MedDiet Score, follow-up, 3110 CRC cases women, a decreased risk was associated with higher
and Recommended Food (multivariate adjusted) adherence to Healthy Eating Indices
Score on CRC risk
(National Institutes of
Health-AARP Diet and
Health Study; US) [61]
Kim et al. (2005). 20,300 men and 21,812 The Western dietary pattern increased colon cancer risk
Investigate the women; three major dietary by 200% in women. There was no significant
associations between patterns: traditional, association between healthy patterns and increased
dietary patterns and the Western and healthy pattern colon cancer risk in men or women. There was a
risk of CRC (Japan scores; 10-years of positive association between a traditional Japanese diet
Public Health Center- follow-up, 370 CRC cases and colon cancer in women but not in men
Cohort I) [62] (multivariate adjusted)
0
Vegan Lacto-ovo Pesco Semi
−5 (p =.32) (p =.08) (p =.002) (p =.69)
% Colorectal Cancer Risk Reduction
−10
−15
−20
−25
−30
−35
−40
−45
−50
Fig. 19.3 Colorectal cancer risk for various vegetarian diets compared to a non-vegetarian dietary pattern (adapted
from [58])
DASH diet was more effective than the MedDiet 19.3.2 CRC Recurrence or Survival
in reducing CRC risk (Fig. 19.4) [60]. For post-
menopausal women, the DASH and Healthy Table 19.3 summaries four prospective cohort stud-
Eating Indices diets were the dietary patterns ies on the effects of dietary patterns on CRC recur-
most effective in lowering CRC risk by 22–28% rence and survival [63–66]. CRC survivors with
(p < .01) [56, 61]. In women, the higher intake of high intake of Western dietary patterns had signifi-
a Western diet in high school was associated with cantly increased odds of CRC mortality and recur-
a 78% increased risk of rectal adenomas rence (Fig. 19.5) and a healthy or p rudent diet was
(p < .005) [59]. In Japanese women, a high associated with lower but insignificant odds for
Western diet intake was associated with a 3.5- CRC mortality or recurrence [66]. In postmeno-
fold increased risk of distal colon cancer [62]. pausal women, higher adherence to Healthy Eating
DASH Diet Score (p-trend =.001)

Alt. Mediterranean Diet Score (p-trend =.06)
1.05
0.95
Multivariate Relative Risk
0.9
0.85
0.8
0.75
0.7
0.65
0.6
1 2 3 4 5
Dietary Pattern Adherence Quintiles
Fig. 19.4 Dietary pattern adherence and colorectal cancer risk in women and men from pooled data from the Nurses’
Health Study and Health Professionals Follow-Up Study (adapted from [60])
Indices was associated with a 29% lower risk of no difference in the rate of colorectal adenoma
mortality (p-trend = .01), which was primarily recurrence [71], subsequent secondary analyses
related to lower intake of alcohol and sugar sweet- suggested potential beneficial effects on adenoma
ened beverages [64] and MedDiet scores were recurrence with the low fat, high fiber fruits and
associated with 14% reduced risk [63]. Also, there vegetables diet [68–70]. These secondary analy-
was a three-fold increase in women’s risk of CRC ses of the Polyp Prevention Trial show that: (1)
mortality with higher intake of processed meat super dietary compliers consuming about 12 g
compared to only 22% increased risk for men [65]. fiber and about three fruit and vegetable serv-
ings/1000 kcals had a significantly lower risk of
adenoma recurrence by 32% and multiple and
19.3.3 CRC Risk Biomarkers advanced adenoma recurrence by 50% compared
to lower compliant controls (p < .05) after 4 years
Table 19.4 shows summaries of six RCTs on the [68]; (2) an 8-year follow-up found an insignifi-
effects of dietary patterns on CRC risk biomark- cant 2% lower colonic adenoma recurrence risk
ers [67–72]. A crossover RCT comparing the for those consuming higher fruit, vegetable, and
high-fat, low fiber Western diet vs. the traditional fiber dietary patterns vs. the control group [69];
high-fiber, low-fat rural African diet over 2 weeks and (3) higher fiber fruit, vegetable, low-fat diets
found that the traditional diet improved colonic with increased bean intake was inversely associ-
health by increasing butyrogenic bacteria to ated with the risk of advanced a denoma recur-
increase fecal butyrate concentration and sup- rence by 65% (Fig. 19.6) [70]. A Canadian trial
pressed secondary bile acid formation, which are (201 subjects; mean baseline age 58 years; 55%
biomarkers of lower CRC risk [67]. Although the men; 2 years) found that low-fat and high fiber
Polyp Prevention Trial (2079 subjects mean base- plant food diets reduced adenoma recurrence by
line age 61 years; 64% men; low fat, high fiber 50% along with reduced fecal bile acid concen-
and fruit and vegetables diet vs. usual diet control trations in women but not in men [72]. A 2017
with minimal guidance for healthy eating) found Cochrane systematic review (5 RCTs; 4,798
Table 19.3 Summary of dietary pattern prospective cohort studies in CRC recurrence and survival
Jacobs et al. (2016). >215,000 African-American, A higher alternate Mediterranean Diet score was
Investigate the Native Hawaiian, Japanese- associated with lower CRC-specific mortality in
association of four American, Latino, and white women by 14% but not in men. Healthy Eating
pre- diagnostic a priori adults living in Hawaii and Indices and DASH index were not significantly
diet quality indexes with California; mean age diagnosis associated with CRC-specific mortality
CRC-specific and 71 years; mean 6-year
all-cause mortality follow-up; 4204 CRC cases
(Multi-ethnic Cohort; (multivariate adjusted)
US) [63]
Fung et al. (2014). 1201 women diagnosed with Only a higher AHEI-2010 score was significantly
Prospectively examine stage I–III CRC, median associated with lower overall mortality by 29%
the association between baseline age 66.5 years, BMI (p-trend = .01) and borderline significantly lower risk of
diet quality scores, 25; median follow-up 11.2 years CRC mortality by 28% (p-trend = .07). A sub-
dietary patterns and and median survival 8.0 years; component analysis showed that low-moderate alcohol
CRC survival (NHS; 162 died from CRC intake and lower intake of sugar sweetened beverages
US) [64] (multivariate adjusted) were the primary factors associated with lower CRC risk
Zhu et al. (2013). 529 newly diagnosed CRC CRC disease-free survival was significantly
Examine the patients from Newfoundland; decreased among patients with a high processed meat
association between mean baseline age 60 years; dietary pattern by 82%. No significant associations
dietary patterns and 50% women; median 6.4 years were observed with the prudent vegetable or the
CRC disease-free of follow-up; 30 cases of CRC high-sugar patterns and CRC disease-free survival
survival (Canada) [65] cancer recurrence or metastasis
Meyerhardt et al. 1009 patients with stage III Patients with high adherence to the Western dietary
(2007). Determine the colon cancer; median baseline pattern had significantly increased colon cancer
association of dietary age 60 years; median 5.3 years mortality risk by 225% (p-trend <.001) and colon
patterns with cancer of follow-up; 324 patients had cancer recurrence risk by 185% (p-trend <.001)
recurrences and colon cancer recurrence, 223 patients (Fig. 19.5). In contrast, the prudent dietary patterns
cancer recurrence or died with cancer recurrence were insignificantly associated with cancer
mortality (US) [66] (multivariate adjusted) recurrence or mortality
Colon Cancer Survival (Mortality) Colon Cancer Recurrence

3.5
2.5
Hazard Ratio
1.5
0.5
0
1 2 3 4 5
Western Dietary Pattern Adherence (Quintile)
Fig. 19.5 Associations between colon cancer mortality and recurrence with adherence to the Western dietary pattern
(p-trend < .001; both) (adapted from [66])
Table 19.4 Summaries of dietary pattern randomized controlled trials (RCTs) in colorectal cancer (CRC) risk
biomarkers
O’Keefe et al. (2015). Crossover RCT: 20 healthy African African Americans switching to a rural
Investigate the acute effects Americans and 20 rural Africans; age traditional diet had increased saccharolytic
of drastic changes in dietary range 40–65 years; BMI 18–35; fermentation bacteria and fecal butyrate
pattern quality in CRC risk 2-weeks; African Americans fed a concentrations, and suppressed secondary
(US) [67] high-fiber, low-fat African-style diet bile acid synthesis associated with lower
and rural Africans fed a high-fat, CRC risk. The opposite effect was observed
low-fiber Western-style diet, which in the rural Africans as they switched to an
reversed their usual dietary patterns American style Western diet with an
increased colon cancer risk profile
Sansbury et al. (2009). Parallel RCT Secondary Analysis: The intervention diet, higher fiber super-
Examine the effect of strict 821 subjects completed 4-year compliers had 32% lower fully adjusted risk
adherence to a low-fat, follow-up; super compliers (25.6%), of adenoma recurrence and 50% lower risk
high-fiber, high-fruit and inconsistent compliers (44.6%) and of multiple and advanced adenoma
-vegetable intervention on poor compliers (29.8%); mean recurrence compared with lower dietary
adenoma recurrence risk baseline age 61 years; 63% men; BMI fiber controls (p < .05)
(Polyp Prevention Trial; US) 27.5; super compliers averaged
[68] baseline intake of 11.7 g/1000 kcal for
fiber, and 2.8 servings/1000 kcal of
fruit and vegetables
Lanza et al. (2007) . Follow Parallel RCT Continuation: 1192 This sub-group trial continuation from 4 to
a sub-cohort of the original subjects (63% of the original cohort); 8 years showed an insignificant 2% lower
cohort for an 801 confirmed colonoscopy reports; adenoma recurrence risk for the higher fiber,
additional 4 years to further mean baseline age 60 years; 66% men; fruit and vegetable and lower fat eating
assess the effect of fruit and high fiber, high-fruit and -vegetable, pattern vs. the control group
vegetables and low-fat diets and low-fat intervention vs. control;
on recurrence of one or more 8-years
adenomas (The Polyp
Prevention Trial- Continued
Follow-up Study; US) [69]
Lanza et al. (2006). Assess Multicenter Parallel RCT: 1905 Higher intake of beans (dietary pulses;
the association between subjects; mean baseline age 61 years; median intake 42 vs. 12 g/day) was
specific fruits, vegetables, 64% men; mean baseline BMI 27.6; inversely associated with risk for advanced
and dried pulses on colorectal low-fat, high-fiber, high-fruit, and adenoma recurrence by 65%
adenoma recurrence (Polyp vegetable vs. control American diet; (p-trend < .001; multivariate adjusted;
Prevention Trial; US) [70] 4 years Fig. 19.6). In addition, vegetables, green
beans and peas, and green salad were
associated with lower risk for advanced
adenoma recurrence
Schatzkin et al. (2000). Multi-center Parallel RCT: 2079 Adopting a diet low in fat and high in fiber,
Investigate the effect of subjects; inclusion criteria > = 1 large fruits, and vegetables did not significantly
healthy fiber-rich dietary bowel adenoma removed within affect the risk of colorectal adenomas
patterns on recurrent 6 months, polyp free colon post recurrence as 40% of subjects in both
colorectal adenomas colonoscopy; mean baseline age 61 years; groups had at least one recurrent adenoma;
development (Polyp 64% males; mean baseline BMI 27.6; the mean number of adenomas was 1.9 in
Prevention Trial; USA) [71] diet goals: 20% energy from fat and 5–8 both groups and the rate of recurrence of
servings of fruits and vegetables daily vs. large adenomas was similar in both groups
usual diet given a standard brochure on
healthy eating; 4–year duration
McKeown-Eyssen et al. Parallel RCT:201 subjects; mean No significant difference in recurrence of
(1994). Assess the effect of a baseline age 58 years; 55% men; diet adenoma polyp incidence rates was found
low fat and high fiber dietary guidance lower fat to 20% of energy between the two dietary groups. A subgroup
pattern on colorectal polyp’s and increase fiber to 50 g/day vs. analysis conducted among 142 subjects with
recurrence (Canada) [72] Western diet; actual diet estimates: high adherence to the low-fat and higher
25% vs. 33% of energy from fat and fiber diet showed that the women in this
35 g vs. 16 g fiber/day; 2-year group had a 50% reduced risk of adenoma
follow-up polyp recurrence, associated with reduced
fecal bile acids concentration
Fruit & vegetables (p =.09) Fruits (p =.23)

Vegetables (p =.09) Dry beans (p = .001)
Green beans & peas (p =.01) Green salad (p =.05)
1.5
Odds Ratio for Advanced Adenome Recurrence
1.3
1.1
0.9
0.7
0.5
0.3
1 2 3 4
Intake Quartile
Fig. 19.6 Association between total fruit and vegetables and specific effects of dry edible beans, green beans and peas,
and green salad and advanced adenoma recurrence risk from the Polyp Prevention Trial (adapted from [70])
subjects with a history of adenomatous polyps (Fig. 19.7] [74]. Table 19.5 summarizes the pro-
removed prior to the trial; mean baseline age 56 spective cohort studies on the effects of s pecific
to 66 years; 2-8 years) found fiber-rich dietary whole plant foods and risk of colorectal adeno-
patterns supplemented with wheat bran cereals or mas and CRC. An overview of the nutrient and
psyllium insignificantly reduced colorectal ade- pytochemical composition of whole plant foods
nomas recurrence [73]. However, the Cochrane are pro vided in Appendix B.
report authors indicated that these studies should
be interpreted with caution because of the high
study subject loss to follow-up and adenomatous 19.4.1 Fruits and Vegetables
polyps are a surrogate outcome for CRC that
appears to require long-term RCTs to assure There were relatively heterogenous effects of
more confident conclusions. increased fruits and vegetables intake on
colorectal adenomas and CRC risk from seven
systematic reviews and meta-analyses of cohort
19.4 Whole Plant Foods and case-control studies [76–82] and eight
specific cohort studies in various populations
Healthy diets rich in whole foods such as whole [83–91]
grains, limited intake of meat products, and
increasing fruit, vegetable, and legume intake 19.4.1.1 Systematic Reviews
may have CRC protective effects [74, 75]. A and Meta-Analyses
2011 Adventist Health Study (2818 subjects;
average 26 years of follow-up) found that high olorectal Adenoma Risk
C
frequency of cooked green vegetables, dried Two meta-analyses evaluated the effects of
fruit, legumes, and brown rice was associated fruits and vegetables on colorectal adenomas
with a decreased risk of colorectal polyps risk [76, 79]. A 2015 meta-analysis (5 cohort
Cooked Green Vegetables Legumes Dried Fruit Brown Rice

1.1
Mutrivariate Odds Ratio 1
0.9
0.8
0.7
0.6
0.5
1 2 3 4
Frequency of Intake Category
Fig. 19.7 Odds of colon/rectal polyps by specific whole plant foods from The Adventist Health Study (p < .05)
(adapted from [74])
Table 19.5 Summaries of whole plant foods studies in colorectal cancer (CRC) and colorectal adenoma risk
Fruits and vegetables
Ben et al. (2015). Evaluate the 5 cohort studies; 126,999 Colorectal adenoma risk was reduced with higher
association between fruit and subjects; 2–26-years of intake of vegetables by 9%, combined vegetables
vegetables and colorectal follow-up and 17 case-control and fruits by 18%, and fruits by 21% (Fig. 19.8).
adenoma risk [76] studies; 11,696 colorectal Also, a linear dose-response analysis showed
adenoma cases reduced risk for each 100-g/day of fruit by 6% and
for vegetables by 2%
Kashino et al. (2015). Assess 6 cohort studies; 488,596 There was insufficient evidence to support an
the effects of vegetables on subjects; 3602 cases or deaths; association between intake of vegetables and CRC
CRC risk among the Japanese and 11 case-control studies risk among the Japanese population as there was
population [77] 0% mean reduced risk in cohort studies and a 25%
mean reduced risk in case-control studies
Tse and Eslick (2014). 11 prospective cohort studies Higher total cruciferous vegetable intake
Examine the observational and 18 case-control studies; significantly reduced colon cancer risk by 16%;
association between 5994 CRC cases and 814 with broccoli lowering risk by 20% (Fig. 19.9).
cruciferous vegetable intake colonic adenoma cases There were insignificant effects of higher
and risk of developing CRC cruciferous vegetables intake on colorectal cancer,
[78] colonic adenoma and rectal cancer
Turati et al. (2014). Evaluate 6 cohort studies and ten CRC risk was reduced with high intakes of garlic
the effect of allium vegetables case-control studies; 13,333 by 15%, onions by 15% and total allium vegetables
on CRC and adenomatous CRC cases by 22%, primarily in case-control studies and with
polyps [79] significant heterogeneity. For colorectal adenoma,
total allium vegetables reduced risk by 12% with
no heterogeneity
Wu et al. (2013). Assess the 11 cohort studies and 24 Cruciferous vegetables intake had a significantly
relationship between case-control studies; 1,295,063lower CRC risk by 18% (high vs. low intake).
cruciferous vegetables and subjects; 24,275 CRC cases Specific analysis reduced CRC risk for cabbage by
CRC risk [80] 24% and broccoli by 18%. The results from the
cohort studies showed borderline statistical
significance
Aune et al. (2011). Summarize 19 cohort studies; >1.5 million There is a weak and nonlinear inverse association
the CRC risk evidence from participants and 11,800– between intake of fruits and vegetables up to
cohort studies in categorical, 16,000 cases; 5–15 years of 500–600 g/day and CRC risk, with the greatest
linear, and nonlinear, duration reduction in risk when increasing intake from very
dose-response meta-analyses low levels (Fig. 19.10).
[81]
Koushik et al. (2007). 14 cohort studies; 756,212 Total combined fruit and vegetable intakes of
Examine the associations subjects; 6–20 years of ≥800 g/day vs. <200 g/day reduced distal colon
between fruit and vegetable follow-up; 5838 colon cancer cancer risk by 26% (p-trend = .02) and total colon
intakes and risk of colon cases cancer risk by 9% (p-trend = .19). Similar trends
cancer [82] were observed for total fruits and total vegetables
Kunzmann et al. (2016). 57,774 individuals; mean age An inverse dose-response effect was shown
Evaluate the association 63 years; median 12.1-year between incident and recurrent colorectal adenoma
between fruit and vegetable follow-up; 1.004 colorectal risk and increased total fruit and vegetable intake
intake and the risk of incident adenoma and 738 recurrent by 19% (p-trend = .07) with a significant reduced
and recurrent colorectal adenoma cases; the median risk of multiple adenomas by 39% (p-trend = .04).
adenoma and CRC (Prostate, dietary intake of fruit and Also, higher fruit and vegetable intakes including
Lung, Colorectal and Ovarian vegetables was 2.4 fruit juice were associated with a borderline
Cancer (PLCOC) Screening cups/1000 kcal; 373 g/day reduced CRC risk by 18% (p-trend = .05), with risk
Trial; US) [83] (multivariate adjusted) reduction increasing to 26% among those with
processed meat intake above the median. Higher
intake of citrus fruits, melons and berries was
associated with a reduced CRC risk by 15%
(p-trend = .09)
Leenders et al. (2015). 40,880 participants; mean A lower risk of colon cancer was observed with
Examine the effects of fruits baseline age 51 years; 13-year higher intake of fruits and vegetables combined by
and vegetables on colon and follow-up; 3370 diagnosed 13% (p-trend = .02), but no consistent association
rectal cancer risk with colon or rectal cancer cases was observed for separate intake of fruits and
extended follow-up and variety (multivariate adjusted) vegetables. A high variety of fruits and vegetables
of intake assessments (EPIC; consumed was not associated with colon or rectal
EU) [84] cancer risk
Aoyama et al. (2014). 45,516 subjects; 58% women; This Japanese population-based cohort study found
Evaluate the effect of low average baseline age 57 years; no significant protective associations between the
intake of vegetables and fruits approx. 20 years of follow-up; intake frequency of vegetables or fruits and CRC
on CRC risk (Japan 806 CRC cases (multivariate incidence. These results are consistent with several
Collaborative Cohort Study for adjusted) previous cohort studies in Japanese subjects
Evaluation of Cancer Risk)
[85]
Vogtmann et al. (2013). 61,274 men; baseline age High fruit intake was associated with reduced CRC
Evaluate the association of 40–70 years; median follow-up risk by 33% (p-trend =.03), whereas vegetable
fruit and vegetable intake with of 6.3 years; 398 CRC cases intake was not significantly associated with risk. In
the risk of colorectal cancer (multivariate adjusted) subgroup analyses only fruits and legumes were
Chinese men (Shanghai Men’s generally inversely associated with the risk of
Health Study; China) [86] colon and rectal cancer among middle age and
older Chinese men
(continued)
Van Duijnhoven et al. (2009). 452,755 subjects; 70% Higher fruits and vegetables intake was found to
Examine the relationship women; 8.8 years of significantly reduce CRC risk by 14%
between self-reported usual follow-up; 2819 CRC cases (p-trend = .04) and colon cancer by 24%
consumption of fruit and (multivariate adjusted) (p-trend < .01). Also, after exclusion of the first
vegetables and CRC incidence 2 years of follow-up, a 100-g increase in
(EPIC; EU) [87] consumption fruits and vegetables reduced risk for
CRC by 5% (p =.04) and colon cancer by 6%
(p = .02)
Nomura et al. (2008). 85,903 men and 105,108 In men, CRC risk was reduced for high intake of
Investigate the association of women; average follow-up of fruits and vegetables combined significantly by
vegetable and fruit intakes 7.3 years; CRC cases 1138 26%, for fruit by 20% (p-trend =.09), and for
with CRC risk (Multiethnic men and 972 women vegetables by 15% 0.85 (p-trend = .05). The
Cohort Study; US) [88] (multivariate adjusted) inverse associations were stronger for colon than
for rectal cancer. In women, none of the
associations with vegetables, fruit, or vegetables
and fruit combined were significant
Millen et al. (2007). Evaluate 32,470 subjects; 47% women; Total fruit intake (5.7 vs. 1.2 serving/day)
effect of fruit, vegetables, or mean baseline age 63 years; significantly lowered risk of distal adenoma by
their subgroups on colorectal 9 years of follow-up; 3057 25% (Fig. 19.11). Although total vegetable intake
adenoma incidence (PPCOC cases with at least one distal was insignificantly associated with colorectal
Screening Trial; US) [89] large bowel adenoma adenoma risk, high intakes of deep-yellow
(multivariate adjusted) vegetables, dark-green vegetables, and onions and
garlic were related to lower colorectal adenoma
risk (Fig. 19.12)
Park et al. (2007). Assess the 488,043 men and women; Higher vegetable intake reduced CRC risk by 18%
relationship between fruit and baseline aged 50–71 years; for men (p-trend = .03) but there was an
vegetable intakes and CRC mean 4.3 years of follow-up; insignificant effect for women. There was an
risk (NIH–AARP Diet and 2972 CRC cases (2048 in men increased CRC risk for very low intake of total
Health Study; US) [90] and 924 in women) fruits and vegetables for men by 26%
(multivariate adjusted) (p-trend = .006). Among subgroups of vegetables,
higher intake of green leafy vegetables was
associated with a lower CRC risk for men by 14%
(p-trend = .04). Intake of fruits was not related to
CRC risk in men or women
Michels et al. (2006). 34,467 women who had Women consuming ≥5 fruit servings/day had 40%
Examine the relationship undergone colonoscopy or lower risk for developing colorectal adenomas
between fruit and vegetable sigmoidoscopy; 19 years of compared with women who consumed only ≤1
consumption and the follow-up; 1720 adenomas of fruit servings/day (p-trend = .001). The respective
prevalence and incidence of the distal colon and rectum colorectal adenoma risk was reduced by 18%
colorectal adenoma (Nurses’ cases (multivariate adjusted) (p-trend = .1) for vegetable consumption
Legumes
Total legumes
Zhu et al. (2015). Investigate 14 cohort studies; 1,903,459 Higher legume consumption was associated with a
the association between participants; 12,261 CRC decreased risk of CRC by 9% (p = .01); legume
dietary legume consumption cases fiber lowered risk by 15% (p = .05). Subgroup
and risk of CRC [92] analyses indicate that higher legume consumption
was inversely associated with CRC risk in Asians
by 18% (p <.01) and soybean intake was associated
with a decreased risk of CRC by 15% (p = .04)
Wang et al. (2013). Assess the Three cohort studies and 11 Higher intake of legumes was associated with a
association between legume case control studies; 101,856 statistically significant 17% decreased risk of
intake and colorectal adenoma participants; 8380 colorectal colorectal adenoma. There was no difference
risk [93] adenoma cases between men and women
Soy products
Tse and Eslick (2016). 22 case control and 18 cohort Increase in soy foods showed a modest reduced
Determine the association studies; 633,476 subjects; CRC risk by 8% but a subgroup analysis found that
between dietary soy and CRC 13,639 CRC cases the high intake of isoflavones lowered CRC risk by
risk [23] 24% vs. low intake
Yan et al. (2010). Determine 4 cohort studies and 7 Soy foods were associated with a 21% reduction in
the relationship between soy case-control studies CRC risk in women (p = .026), but not in men
intake and CRC risk [94]
Yang et al. (2009). Investigate 68,412 women, mean baseline Each 5-g/day intake of soy foods as assessed by
the effect of soy foods on CRC age 52 years; mean 6.4 years dry weight (equivalent to 1 oz. (28.35 g) tofu/day)
risk (Shanghai Women’s of follow-up, 321 CRC cases was associated with an 8% reduced CRC risk.
Health Study; China [95]) (multivariate adjusted) Women in the highest tertile of intake had a
reduced CRC risk by 33% compared with those in
the lowest tertile (p-trend = .008). This inverse
association was primarily confined to post-
menopausal women. Similar results were also
found for intakes of soy protein and isoflavones
Akhter et al. (2008). Examine 83,063 Japanese men and The highest vs. lowest intake of isoflavone, miso
the association between soy women, mean baseline age soup, and soy food reduced CRC risk in men by
food intake and CRC risk 57 years; 5 year-follow-up; approx. 11% (p <.05). In women, there was no
(Japan Public Health 886 cases of CRC (291 association observed with CRC risk for any of
Center-Based Prospective proximal colon, 286 distal- these soy products or components
Study) [96] colon and 277 rectum)
Michels et al. (2006). 34,467 women who had Women who consumed ≥4 servings of legumes/
Examine the effect of legume undergone colonoscopy or week had a lower colorectal adenoma incidence
intake on the prevalence and sigmoidoscopy; 19-year than women who reported intake of ≤1 serving/
incidence of colorectal follow-up; 1720 adenomas of week by 33% (p-trend = .005)
adenoma (Nurses’ Health the distal colon and rectum
Study; US) [91] cases (multivariate adjusted)
Lin et al. (2005). Examined 39,876 healthy women; aged Higher intake of legume fiber was associated with
the association between ≥45 years at baseline; average a lower CRC risk by 40% (highest vs. lowest
dietary intakes of fruit, follow-up of 10 years; 223 quintile; p-trend = .02). One legume serving was
vegetables, and fiber and CRC CRC cases (multivariate associated with reduced CRC risk by 17%
risk in women (Women’s adjusted) (p-trend = .19)
Whole-grains
Aune et al. (2011). Investigate 6 prospective cohort studies; The consumption of 3 servings (90 g)/day of
the association between intake 774,806 participants; whole-grains reduced risk for CRC by 17%, colon
of whole grains and CRC risk 4.5–26-years of follow-up; cancer by 14% and rectal cancer by 20%
using dose-response meta- 7941 CRC cases; total whole
analysis [19] grains included whole grain
rye breads, other whole grain
breads, oatmeal, whole grain
cereals, high fiber cereals,
brown rice, and porridge;
61–128 g whole-grains/day
(continued)
Haas et al. (2009). Evaluate 11 cohort studies; 1,719,590 Increased intake of whole-grains reduced CRC risk
the effectiveness of whole subjects; baseline age in women by 8% and men by 7%; overall lower
grain intake on CRC risk [98] 25–76 years; 6–16-years of risk was reduced for colon cancer by 7% and rectal
follow-up; 7745 CRC cases cancer by 11%
Bakken et al. (2016). 78,254 women; median Higher whole-grain bread intake by Norwegian
Investigate the association baseline age 55 years; median women was insignificantly associated with reduced
between whole-grain bread 9 years of follow-up; 795 CRC CRC risk by 11–14%. A cancer subsite analysis
consumption and CRC cases (multivariate adjusted) showed higher whole-grain bread intake was
incidence among Norwegian weakly associated with a lower risk of proximal
women (the Norwegian colon cancer (p-trend = .09)
Women and Cancer Study) [99]
Abe et al. (2014). Examine 73,501 men and women; High intake of japonica rice reduced rectal cancer
associations between japonica average follow-up of 11 years; risk in men by 39% (p-trend = .085). No clear
round rice vs. bread, noodles 1276 CRC cases (multivariate patterns of association were found for bread,
and cereal intake on CRC risk adjusted) noodles and cereal intake
among Japanese adults (the
Japan Public Health Center-
based prospective Study) [100]
Kyra et al. (2013). Investigate 108,000 Danish, Swedish, and Per 50 g whole-grain products intake, CRC risk
the association between Norwegian subjects; mean was significantly reduced by 6%. Intake of
whole-grain intake and CRC baseline age 52 years; median whole-grain wheat was associated with a lower
risk (Scandinavian HELGA 11 years of follow-up; 1123 CRC incidence (highest vs. lowest quartile of
cohort) [101] CRC cases (multivariate intake) by 34% but the effect was non-linear
adjusted) (p-trend = .18)
Egeberg et al. (2010). 26,630 men and 29,189 Per daily 50 g increment whole grain product
Evaluate the association women; mean baseline age intake reduced risk of colon cancer by 15% and
between intake of total and 53 years; median 10.6 years of rectal cancer by 10% in men. Each 25 g/day of
individual whole-grain follow-up; 461 colon cancer whole-grain bread significantly lowered risk of
products in relation to risk of cases and 283 rectal cancer colon cancer by 11%. For women, no consistent
colon and rectal cancer cases (multivariate adjusted) associations between total or individual whole-
(Danish Diet, Cancer and grains product consumption and colon or rectal
Health prospective cohort cancer risk were observed
study) [102]
Schatzkin et al. (2007) 291,988 men and 197,623 Whole-grain intake was associated with lower CRC
Investigate the relation women; mean baseline age risk by 21% for the total cohort (p-trend <.001). In
between fiber and whole-grain 63 years; 5-year of follow-up; a sub-group analysis, CRC risk was reduced for
food intakes and CRC 2974 CRC cases (multivariate men by 21% and for women by 13%. The
(National Institutes of adjusted) association with whole grain was stronger for rectal
Health-AARP Diet and Health than for colon cancer
Study) [103]
Larsson et al. (2005). 61,433 women; mean In women, high intake of whole grains was
Examine prospectively the follow-up of 14.8 years; 805 associated with a lower risk of colon cancer, but
association between whole CRC cases (multivariate not of rectal cancer. Colon cancer risk (≥4.5 vs.
grain consumption and CRC adjusted) <1.5 whole-grain servings/day) was reduced by
risk in women (Swedish 33% (p-trend = .06). After excluding cases
Mammography Cohort) [104] occurring within the first 2 years of follow-up, the
risk was further reduced to 35% (p-trend = .04).
Women in the top quintile of cereal fiber intake
(>13.6 g/day) had a 27% reduced colon cancer risk
(p-trend = .03) compared with those in the lowest
quintile (<7.3 g/day)
Nuts (including peanuts)
Yang et al. (2016). Examine 75,680 women; nut intake CRC risk was reduced for the intake of nuts ≥2
the association of long-term assessed at baseline and times/week vs. rarely consuming nuts by 14%
nut consumption with CRC updated every 2–4 years for (p-trend = .04), which was attenuated after further
risk (Nurses’ Health Study; 30 years; 1503 CRC cases adjusting for BMI and diabetes to 13%
US) [106] (multivariate adjusted) (p-trend = .06). No association was observed for
peanut butter
Yeh et al. (2006). Examine the 12,026 men and 11,917 CRC risk was reduced for higher peanut intake for
relationship of peanut intake women; baseline age men by 27% and for women by 58%
and CRC risk (Taiwan 30–65 years; 10 years of
community-based cancer follow-up; 107 CRC cases
screening cohort) [107] (multivariate adjusted)
Jenab et al. (2004). Determine 478,040 subjects (141,988 There was no association between higher intake of
the effects of nut and seed men, 336,052 women); nuts and seeds and risk of CRC, colon, and rectal
intake on CRC risk (EPIC; 4.8-year follow-up; 1329 CRC cancers in men and women combined. However, a
EU) [108] cases (multivariate adjusted) subgroup analysis indicated that higher nut and
seed intake in women reduced colon cancer risk by
31% (p-trend = .04)
and 17 case-control studies) found a significant nificantly reduced distal colon cancer by 26%
inverse association for increased fruit intake on with similar trends for fruits and vegetables [82].
colorectal adenomas risk by 21% but only 9% Several meta-analyses have observed that
for vegetables (higher vs. lower intake; increased intake of cruciferous vegetables such
Fig. 19.8); each 100 g/day of total fruit lowered as broccoli and cabbage reduce CRC risk, espe-
colorectal adenomas risk by 6% compared to cially for colon cancer (Fig. 19.9) [78, 80]. A
only 2% for total vegetables [76]. In a 2014 2014 meta-analysis (six cohort and ten case-con-
meta-analysis (six cohort and ten case-controlled studies) found for highest vs. lowest
trolled studies) total allium vegetables (e.g., intake reduced CRC risk for garlic by 15%,
garlic and onion) was shown to reduce the inci- onions by 15% and total allium vegetables by
dence of colorectal adenomas by 12% with no 22% (primarily in response to case-control stud-
heterogeneity [79]. ies and with significant heterogeneity) [79].
However, a Japanese meta-analysis (6 cohort and
olorectal Cancer Risk
C 11 case-control studies) showed that there is
Five meta-analyses evaluated the effects of fruits insufficient evidence supporting a protective
and vegetables on CRC risk (Figs. 19.9 and effect of increased vegetable intake on CRC risk
19.10) [77, 78, 80–82]. A 2011 meta-analysis in a Japanese population [77].
(19 cohort studies; 1.7 million subjects;
5–15 years of follow-up) showed a weak and 19.4.1.2 S
pecific Prospective Cohort
nonlinear inverse association between intake of Studies
fruits and vegetables in CRC risk with a benefit
threshold at between 500 and 600 g/day, with the US Populations
greatest reduction in risk when increasing intake Five cohort studies evaluated the effect of fruits
from very low levels (Fig. 19.10) [81]. A 2007 and vegetables on colorectal adenomas and CRC
meta-analysis (14 cohort studies; 746,212 sub- risk [83, 88–91]. Two prospective analyses from
jects) found that very high combined fruit and the Prostate, Lung, Colorectal and Ovarian
vegetable intake (800 g/day) vs. 200 g/day sig- Cancer Screening Trial: (1) a 2016 prospective
0
Fruits Vegetables Fruits & Vegetables
Combined
−5
Colorectal Adenoma Risk (%)
−10
−15
−20
−25
Fig. 19.8 Association between vegetables and fruits intake and colorectal adenoma risk (high vs. low intake) from a
meta-analysis (adapted from [76])
Total Cruciferous Vegetables (p <.05) Broccoli (p <.05) Cabbage (p >.05)

0
−5
Risk of Colon Cancer (%)
−10
−15
−20
−25
Fig. 19.9 Associations between cruciferous vegetables intake and colon cancer risk based on a meta-analysis (adapted
from [78])
analysis (57,774 subjects; 12 years of follow-up) green vegetables, and onions and garlic were
showed an inverse dose-response trend between related to lower colorectal adenomas risk
total fruit and vegetable intake and colorectal ade- (Fig. 19.12) [89]. Also, higher fruit and vegetable
nomas risk (p-trend = .07) and the risk of multiple intakes including 100% fruit juice were associ-
adenomas was significantly reduced by 39% [83]; ated with a borderline reduced CRC risk by 18%
and (2) a 2007 analysis (32,470 subjects; 9 years (p-trend = .05), which was stronger among those
of follow-up) found that total fruit intake (5.7 vs. with intakes of processed meats above the median
1.2 serving/day) significantly lowered risk of dis- by 26% vs. 10% for those with processed meat
tal colon adenoma by 25% (Fig. 19.11) [89]. intakes below median. A 2006 Nurses’ Health
Although total vegetable intake was insignifi- Study reported that increased total fruit intake
cantly associated with colorectal adenomas risk, (≥5 vs. ≤1 serving(s)/day) significantly reduced
high intakes of deep-yellow vegetables, dark- colorectal adenomas risk by 40% whereas total
Fruits Vegetables
1.05
Relative Risk for CRC 1
0.95
0.9
0.85
0.8
0 100 200 300 400 500
g/day
Fig. 19.10 Relationship between fruit and vegetable intake and colorectal cancer (CRC) risk from a meta-analysis of
cohort studies (adapted from [81])
Total Fruit (p < .001) Fruit w/o Juice (p <.001)

Fruit Juice (p .04) Total Vegetables (p =.24)
Total Vegetables w/o Potatoes (p =.08)
1.1
Odds Ratio for Colorectal Adenoma
1.05
1
0.95
0.9
0.85
0.8
0.75
0.7
0.65
1 2 3 4 5
Intake (Quintile)
Fig. 19.11 Association between fruits and vegetables and colorectal adenomas from the Prostate, Lung, Colorectal,
and Ovarian Cancer Screening Study (adapted from [89])
vegetable intake insignificantly reduced risk by EU Populations

18% [91]. Two other cohort studies suggest that Two cohort studies from the EPIC study popula-
increased fruit and vegetable intake only signifi- tions evaluated the effect of fruit and vegetables
cantly reduced CRC risk in men (especially green on CRC risk [84, 87]. A 2009 analysis (452,755
leafy vegetables) but not in women, which may be subjects; 8.8 years of follow-up) found that higher
related to the amount consumed [88, 90]. total fruits and vegetables intake significantly
Deep-Yellow Vegetables (p = .004) Dark-green Vegetables (p =.07)

Dry Beans (p =.1) Onions & Garlic (p =.01)
1.05
Odds Ratio for Colorectal Adenoma
0.95
0.9
0.85
0.8
1 2 3 4 5
Intake (Quintile)
Fig. 19.12 Association between specific vegetables and colorectal adenomas risk from the Prostate, Lung, Colorectal,
and Ovarian Cancer Screening Study (adapted from [89])
reduced risk of CRC by 14% and colon cancer by colon and rectal cancer risk among middle age
24%. Also, after exclusion of the first 2 years of and older Chinese men [86].
follow-up, a 100 g increase in total fruit and veg-
etable consumption significantly reduced risk for
CRC by 5% and colon cancer by 6%. A 2015 19.4.2 L
egumes (including Soy
EPIC analysis (40,880 participants; 13 years of Products)
follow-up) showed that increased total fruits and
vegetables significantly reduced colon cancer risk Table 19.5 summarizes the effects of increased
by 13%, but no consistent associations were legume intake on colorectal adenomas and CRC
observed for separate intake of fruits and risk. Four meta-analyses of observational studies
vegetables. consistently show that legumes reduce colorectal
adenomas and CRC risk [23, 92–94]. A 2015
Asian Populations meta-analysis (14 cohort studies; 1,903,459 par-
Two cohort studies from Asian countries ticipants) found that higher legume consumption
assessed the effects of fruits and vegetables on was associated with a significantly lower CRC
CRC risk [85, 86]. A 2014 Japanese study risk by 9% and legume fiber significantly low-
(45,516 subject; 20 years of follow-up) found ered CRC risk by 15% [92]. Also, in Asian popu-
that increased intake of fruits and vegetables was lations higher legume consumption was
not significantly associated with CRC risk in associated with significantly reduced CRC risk
men or women, which is consistent with other by 18% and soybean intake was associated with a
previous Japanese cohort studies [85]. In Chinese significantly decreased risk of CRC by 15%. A
men, the Shanghai Men’s Health Study (61,274 2013 meta-analysis (3 cohort and 11 case-control
men; 6.3 years of follow-up) found that increased studies; 101,856 subjects) showed that higher
fruit intake was significantly inversely associ- legume intake significantly lowered colorectal
ated with CRC risk, whereas vegetable intake adenomas risk by 17% for both men and women
was not associated with CRC risk except for [93]. A 2010 meta-analysis found that increased
legumes, which were inversely associated with soy intake significantly reduced CRC risk only in
women by 21% [94], but a 2016 meta-analysis risk of colon cancer by 7% and rectal cancer by
(18 cohort and 22 case-control studies; 633,476 11% [98]. Four cohort studies evaluated the CRC
subjects) found a modest reduced CRC risk by protective effect of increased whole-grains con-
8% in men and women with a higher intake of sumption in Scandinavian populations [99, 101,
isoflavones significantly lowering CRC risk by 102, 104]. Two cohort studies evaluated the
24% [23]. Two cohort studies in women consis- effect of whole-grain intake on both men and
tently show that increased intake of legumes or women: (1) the 2013 Scandinavian HELGA
legume fiber reduces both the risk of colorectal cohort (108,000 Danish, Swedish, and Norwegian
adenomas and CRC [91, 97]. The Nurses’ Health men and women; median 11 years of follow-up)
Study (34,467 women; 19 years of follow-up) found that per 50 g whole grain products CRC
showed that women consuming higher legume risk was significantly reduced by 6% [101]; and
intake (≥4 vs. ≤1 weekly servings) significantly (2) the 2010 Danish Diet, Cancer and Health
reduced risk of colorectal adenomas by 33% Study (26,630 men and 29,189 women; median
[91]. Similar findings were observed in the 10.6 years of follow-up) showed that daily 50 g
Women’s Health Study with higher intake of intake of whole grain products was associated
legume fiber and significantly lower CRC risk with lower risk of colon cancer by 15% and rec-
[97]. Two cohort studies evaluated the effect of tal cancer by 10% in men. Higher whole-grain
increased soy product intake on CRC risk in bread was statistically significantly associated
Chinese women and in a Japanese population with a lower risk of colon cancer with each intake
[95, 96]. The Shanghai Women’s Health Study of 25 g/day by 11% [102]. However, for women,
(64,412 women; 6.2 years of follow-up) found in no consistent associations between total or indi-
women that each 1 oz of tofu was associated with vidual whole-grains product consumption and
a significant 8% lower CRC risk and the highest colon or rectal cancer risks were observed. Two
tertile of intake was associated with a 33% lower studies in women showed somewhat conflicting
risk [95]. The Japanese Public Health Prospective findings: (1) a 2016 Norwegian Women and
Study (83,063 subjects; 5 years of follow-up) Cancer Study (78,254 women; 6–16 years of
showed that higher intake of isoflavones, miso follow-up) showed an insignificant CRC risk
soup, and soy food significantly reduced CRC reduction by 11–14% for increased whole- grain
risk in men by 11%, but no association was bread intake [99]; and (2) the Swedish
observed for CRC risk in women [96]. Mammography Cohort (61,433 women; mean
follow-up of 14.8 years) found that colon cancer
risk was reduced by 33% (p-trend = .06) (≥4.5
19.4.3 Whole-Grain Products vs. <1.5 whole-grain servings/day) [104]. After
excluding cases occurring within the first 2 years
Table 19.5 summarizes the effects of increased of follow-up, the risk was further reduced to 35%
whole-grain intake on CRC risk for observational (p-trend = .04). In a 2014 nested case-control
studies. Two meta-analyses show that higher study within the Scandinavian HELGA cohort
intake of whole-grains is associated with lower assessment (450 cases vs. 450 controls), no sig-
CRC risk [19, 98]. A 2011 meta-analysis nificant associations were observed across
(6 cohort studies; 774,806 participants; 4.5–36 increasing quartiles of whole-grain intake with
years of follow-up) found that three servings any subtype of colorectal cancer whereas blood
daily (90 g/day) of whole-grains reduced risk for alkylresorcinol concentrations (a biomarker of
CRC by 17%, colon cancer by 14% and rectal short and medium term whole-grain intake) were
cancer by 20% [19]. A 2009 meta-analysis associated with significantly lower distal colon
(11 cohort studies; 1,719,590 participants; 6–16- cancer risk by 66% (highest vs. lowest quartiles)
years of follow-up) showed that higher intake of [105]. Two cohort studies assessed the effect of
whole-grains reduced CRC risk in women by 8% increased whole-grain intake on CRC risk
and men by 7% with an overall population lower [100, 103]. The 2014 Japan Public Health
Centered based Study (73,501 men and women; colon cancer in women by 31% with increased
average follow-up of 11 years) found no clear nut and seed intake [108].
associations between higher intakes of whole-
grains Japonica round and short grain rice, bread, Conclusions
noodle and cereal intake on CRC risk and its sub- Globally CRC rates have doubled since the
sites in men or women [100]. However, an 1970s and incidence is strongly associated with
inverse trend was observed between Japonica the Western lifestyle and aging populations. As
rice intake and rectal cancer in men with reduced much as 90% of CRC cases may be attributable
risk by 39% (p-trend = .085). In the US, the 2007 to dietary factors. A number of nutrients and
National Institutes of Health-AARP Diet and phytochemicals are considered to be poten-
Health Study (291,988 men and 197,623; 5 years tially protective against CRC to various degrees
of follow-up) demonstrated that whole-grain including fiber, isoflavones, flavonoids, antiox-
intake was associated with significantly lower idant vitamins, carotenoids, folate, calcium,
CRC risk by 21% for the total cohort; a subgroup magnesium, and selenium. Higher adherence to
analysis indicated that women had a 13% CRC a Western dietary pattern, which can stimulate a
lower risk [103]. proinflammatory systemic response, can sig-
nificantly increase risk of colorectal adenomas
and CRC, especially in diets high in red or pro-
19.4.4 Nuts (Including Peanuts) cessed meats. In contrast, higher adherence to
healthy dietary patterns including the MedDiet,
Table 19.5 summarizes the effects of three pro- DASH, Healthy Eating Indices, pesco-vegetar-
spective cohort studies on the relationship ian and low inflammatory index diets can sig-
between increased tree nut and peanut intake and nificantly reduce risk of colorectal adenoma
CRC risk [106–108]. The US 2016 Nurses’ and CRC. Survivors of CRC with high intake
Health Study (75,680 women; 30 years of follow- of Western dietary patterns had significantly
up) found that the CRC risk was significantly higher odds of CRC mortality and recurrence
reduced for nuts ≥2 times/week vs. rarely con- compared to those consuming healthy diets.
suming nuts by 14%, which was attenuated after Dietary patterns rich in fruits, vegetables
further adjusting for BMI and diabetes to 13% (including green leafy vegetables, cruciferous
(p-trend = .06) [106]. The Taiwan Community- and allium vegetables), legumes (including
based Cancer Screening cohort (12,026 men and soy), whole-grains (≥3 servings/day) and pea-
11,917 women) showed that increased peanut nuts may have protective effects against
intake significantly reduced CRC risk in men by colorectal adenomas and CRC risk. Diets rich
27% and in women by 58% [107]. An EPIC study in dietary fiber have been related to a lower
(478,040 subjects; 4.8 years of follow-up) dem- CRC risk due in large part to beneficial effects
onstrated no association between nut and seed of butyrate, derived from fiber fermentation by
intake and CRC risk for men and women com- colonic microflora, an inhibitor of colonocyte
bined, but there was a significantly lower risk of tumor cell initiation and progression.
Appendix A 549
ppendix A: Comparison of Western and Healthy Dietary Patterns

A
per 2000 kcals (Approximated Values)
Healthy
Vegetarian
Western Healthy Pattern
Dietary USDA Base DASH Diet Mediterranean (Lact-ovo Vegan
Components Pattern (US) Pattern Pattern Pattern based) Pattern
Emphasizes Refined Vegetables, Potassium Whole grains, Vegetables, Plant foods:
grains, low fruits, rich vegetables, fruit, whole- vegetables,
fiber foods, whole-grains, vegetables, fruits, dairy grains, legumes, fruits, whole
red meats, and low-fat fruits a products, olive nuts, seeds, milk grains, nuts,
sweets and milk low-fat milk oil, and products and seeds and
solid fats products moderate wine soy foods soy foods
Includes Processed Enriched Whole- Fish, nuts, Eggs, non-dairy Non-dairy
meats, sugar grains, lean grains, seeds, and milk alternatives milk
sweetened meat, fish, poultry, fish, pulses and vegetable alternatives
beverages, nuts, seeds and nuts and oils
and fast foods vegetable oils seeds
vegetables, added sugars sweets and refined grains, meats, or fish; products
whole-grains sugar- and sweets limited sweets
sweetened
beverages
Carbohydrates (% 51 51 55 50 54 57
Total kcal)
Protein (% Total 16 17 18 16 14 13
kcal)
Total fat (% Total 33 32 27 34 32 30
kcal)
Total kcal)
Unsat. fat (% Total 22 25 21 24 26 25
kcal)
Fiber (g) 16 31 29+ 31 35+ 40+
Potassium (mg) 2800 3350 4400 3350 3300 3650
Vegetable oils (g) 19 27 25 27 19–27 18–27
Sodium (mg) 3600 1790 1100 1690 1400 1225
Plant food groups
Fruit (cup) ≤1.0 2.0 2.5 2.5 2.0 2.0
Vegetables (cup) ≤1.5 2.5 2.1 2.5 2.5 2.5
Whole-grains (oz.) 0.5 3.0 4.0 3.0 3.0 3.0
Legumes (oz.) − 1.5 0.5 1.5 3.0 3.0+
Nuts/Seeds (oz.) 0.5 0.6 1.0 0.6 1.0 2.0
Soy products (oz.) 0.0 0.5 − − 1.1 1.5
gov/ accessed 8.22.2015.
Nutrient Database for Standard Reference, Release 27. http://www.ars. usda.gov./nutrientdata. accessed 17 February
2015.
patterns.2015;1–9.
trends. 2015;1–78.
Acta. 2011;412: 1493–1514.

A
Nutrients/ Wheat, oat, Apples, pears, Dates, dried Potatoes, spinach, Lentils, Almonds, Brazil
phytochemicals barley, rye, bananas, figs, carrots, peppers, chickpeas, nuts, cashews,
brown rice, grapes, apricots, lettuce, green split peas, hazelnuts,
whole grain oranges, cranberries, beans, cabbage, black beans, macadamias,
bread, cereal, blueberries, raisins and onions, cucumber, pinto beans, pecans, walnuts,
pasta, rolls strawberries, prunes cauliflower, and soy beans peanuts,
and crackers and avocados mushrooms, and sunflower seeds,
broccoli and flaxseed
Energy (kcals) 110–350 30–170 240–310 10–115 85–170 520–700
Protein (g) 2.5–16 0.5–2.0 0.1–3.4 0.2–5.0 5.0–17 7.8-24
Available 23–77 1.0–25 64–82 0.2–25 10–27 12–33
Carbohydrate (g)
Fiber (g) 3.5–18 2.0–7.0 5.7–10 1.2–9.5 5.0–11 3.0–27
Total fat (g) 0.9–6.5 0.0–15 0.4–1.4 0.2–1.5 0.2–9.0 46–76
SFAa (g) 0.2–1.0 0.0–2.1 0.0 0.0–0.1 0.1–1.3 4.0–12
MUFAa (g) 0.2–2.0 0.0–9.8 0.0–0.2 0.1–1.0 0.1–2.0 9.0–60
PUFAa (g) 0.3-2.5 0.0–1.8 0.0–0.7 0.0.0.4 0.1–5.0 1.5–47
Folate (μg) 4.0–44 <5.0–61 2–20 8.0–160 50–210 10–230
Tocopherols 0.1–3.0 0.1–1.0 0.1–4.5 0.0–1.7 0.0–1.0 1.0–35
(mg)
Potassium (mg) 40–720 60–500 40–1160 100–680 200–520 360–1050
Calcium (mg) 7.0–50 3.0–25 10–160 5.0–200 20–100 20–265
Magnesium (mg) 40–160 3.0–30 5.0–70 3.0–80 40–90 120–400
Phytosterols 30–90 1.0–83 – 1.0–54 110–120 70–215
(mg)
Polyphenols 70–100 50–800 – 24–1250 120–6500 130–1820
(mg)
Carotenoids (μg) – 25–6600 1.0–2160 10–20,000 50–600 1.0–1200
a
Nutrient Database for Standard Reference, Release 27. http://www.ars. usda.gov./nutrientdata. Accessed 17 February
2015
Circulation. 2013;128: 553-565.
USDA. What we eat in America, NHANES 2011-2012, individuals 2 years and over (excluding breast-fed children).
Available: www.ars.usda.gov/nea/bhnrc/fsrg.
Slavin JL, Lloyd B. Health benefits of fruits and vegetables. Adv Nutr. 2012; 3:506-516.
References 551
related co-morbidities. Obes Rev. 2014;15: 392-407.
Service, Home and Garden Bulletin 72.
Holden JM, Eldridge AL, Beecher GR, et al. Carotenoid content of U.S. foods: An update of the database. J Food Comp
An. 1999; 12:169-196.
during maturation and from different growing areas. J Agric Food Chem. 2009; 57(21):10,408-10,413.
States. J Agric Food Chem. 2004; 52: 4026-4037.
Acta. 2011;412: 1493-1514.
trends. 2015;1-78.
http://health.gov/dietary guidelines/2015/guidelines/accessed 1.26.2016.
10. Stone RAT, Waring ME, Cutrona SL, et al. The asso-
References ciation of dietary with colorectal cancer among nor-
mal weight, overweight and obese men and women:
1. Global Burden of Disease Cancer Collaboration. a prospective longitudinal study in the USA. BMJ
Global, regional, and national cancer incidence, mor- Open. 2017;7;e015619.
tality, years of life lost, years lived with disability, and 11. Godos J, Bella F, Torrisi A, et al. Dietary patterns and
disability-adjusted life-years for 32 cancer groups, risk of colorectal adenoma: a systematic review and
1990 to a systematic analysis for the global burden meta-analysis of observational studies. Hum Nutr Diet.
of disease study. JAMA Oncol. 2015;3(4):524–48. 2016;29:757–67. https://doi.org/10.1111/jhn.12395.
https://doi.org/10.1001/ jamaoncol.2016.5688. 12. Feng Y-L, Shua L, Zhenga P-F, et al. Dietary patterns
2. Mármol I, Sánchez-de-Diego C, Dieste AP, et al. and colorectal cancer risk: a meta-analysis. J Cancer
Colorectal carcinoma: a general overview and future Prev. 2016;26(3):201–11. https://doi.org/10.1097/
perspectives in colorectal cancer. Int J Mol Sci. CEJ.0000000000000245.
2017;18:197. https://doi.org/10.3390/ ijms1800197. 13. MacDonald RS, Wagner K. Influence of dietary
3. American Cancer Society (ACS). Colorectal can- phytochemicals and microbiota on colon cancer
cer facts & figures 2014-2016. Atlanta: American risk. J Agric Food Chem. 2012;60:6728–35. doi.
Cancer Society; 2014. org/10.1021/jf204230r
4. World Cancer Research Fund International (WCRF). 14. De Rosa M, Rega D, Costabile V, et al. The biologi-
Colorectal cancer statistics. www.wcrf.org/int/ cal complexity of colorectal cancer: insights into
cancer-facts-figures/data specific-cancer/colorectal biomarkers for early detection and personalized
cancer-statistics. Accessed 6 May 2015. care. Ther Adv Gastroenterol. 2016;9(6):861–86.
5. Bailie L, Loughrey MB, Coleman HG. Lifestyle https://doi.org/10.1177/1756283X16659790.
risk factors for serrated colorectal polyps: a system- 15. Arvelo F, Sojo F, Cotte C. Biology of colorectal can-
atic review and meta-analysis. Gastroenterology. cer. Ecancermedicalscience. 2015;9:520.
2017;152(1):92–104. https://doi.org/10.1053/j. 16. Brenner H, Kloor M, Pox C. Colorectal cancer.
gastro.2016.09.003. Lancet. 2014;383:1490–502.
6. Song M, Garrett WS, Chen AT. Nutrients, foods, 17. Yao Y, Suo T, Andersson R, et al. Dietary fibre
and colorectal cancer prevention. Gastroenterology. for the prevention of recurrent colorectal adeno-
2015;148:1244–60. mas and carcinomas. Cochrane Database Syst Rev.
7. Baena R, Salinas P. Diet and colorectal cancer. 2017;1:CD003430. https://doi.org/10.1002/14651858.
Maturitas. 2015;80:258–64. CD003430.pub2.
8. Makarem N, Lin Y, Bandera EV, et al. Concordance 18. Ben Q, Sun Y, Chai R, et al. Dietary fiber intake
with World Cancer Research Fund/American reduces risk for colorectal adenoma: a meta-analysis.
Institute for Cancer Research (WCRF/AICR) guide- Gastroenterology. 2014;146:689–99.
lines for cancer prevention and obesity-related cancer 19. Aune D, Chan DSM, Lau R, et al. Dietary fibre,
risk in the Framingham Offspring cohort (1991– whole grains, and risk of colorectal cancer: system-
2008). Cancer Causes Control. 2015;26(2):277–86. atic review and dose-response meta-analysis of pro-
https://doi.org/10.1007/s10552-014-0509. spective studies. BMJ. 2011;343:d6617.
9. Vieira AR, Abar L, Chan DSM, et al. Foods and bev- 20. Jacobs ET, Lanza E, Alberts DS, et al. Fiber, sex,
erages and colorectal cancer risk: a systematic review and colorectal adenoma: results of a pooled analysis.
and meta-analysis of cohort studies, an update of the Am J Clin Nutr. 2006;83:343–9.
evidence of the WCRF-AICR Continuous Update 21. Jiang R, Botma A, Rudolph A, et al. Phyto-
Project. Annals Oncology. 2017;28:1788–1802. oestrogens and colorectal cancer risk: a systematic
review and dose–response meta-analysis of observa- meta-analysis of prospective observational studies.

tional studies. Br J Nutr. 2017;116:2115–28. Int J Cancer. 2015;136:1680–7.
22. Yu Y, Jing X, Li H, et al. Soy isoflavone consump- 35. Chen G-C, Pang Z, Liu Q-F. Magnesium intake and
tion and colorectal cancer risk: a systematic review risk of colorectal cancer: a meta-analysis of pro-
and meta-analysis. Sci Rep. 2016;6:25939. https:// spective studies. Eur J Clin Nutr. 2012;66:1182–6.
doi.org/10.1038/srep25939. https://doi.org/10.1038/ejcn.2012.135.
23. Tse G, Eslick GD. Soy and isoflavone consump- 36. Wark PA, Lau R, Norat T, Kampman E. Magnesium
tion and risk of gastrointestinal cancer: a systematic intake and colorectal tumor risk: a case-con-
review and meta-analysis. Eur J Nutr. 2016;55(1):63– trol study and meta-analysis. Am J Clin Nutr.
73. https://doi.org/10.1007/s00394-014-0824-7. 2012;96:622–31.
24. Grosso G, Godos J, Lamuela-Raventos R, et al. A 37. Cai X, Wang C, Yu W, et al. Selenium exposure and
comprehensive meta-analysis on dietary flavonoid cancer risk: an updated meta-analysis and meta-
and lignan intake and cancer risk: level of evidence regression. Sci Reports. 2016;6:19213. https://doi.
and limitations. Mol Nutr Food Res. 2017;61(4). org/10.1038/srep.19213.
https://doi.org/10.1002/mnfr.201600930. 38. Takata Y, Kristal AR, King IB, et al. Serum sele-
25. Jin H, Leng Q, Li C. Dietary flavonoid for prevent- nium, genetic variation in selenoenzymes, and risk
ing colorectal neoplasms. Cochrane Database Syst of colorectal cancer: primary analysis from the
Rev. 2012;(8):CD009350. doi: 10.1002/14651858. Women’s Health Initiative Observational Study and
CD009350. pub2. meta-analysis. Cancer Epidemiol Biomark Prev.
26. Xu X, Yu E, Liu L, et al. Dietary intake of vitamins 2011;20(9):1822–30.
A, C, and E and the risk of colorectal adenoma: a 39. Vipperla K, O’Keefe SJ. Diet, microbiota, and dys-
meta-analysis of observational studies. Eur J Cancer biosis: a ‘recipe’ for colorectal cancer. Food Funct.
Prev. 2013;22:529–39. 2016;7:1731–40.
27. Pais R, Dumitraşcu DL. Do antioxidants prevent 40. O’Keefe SJD, Li JV, Lahti L, et al. Fat, fibre and
colorectal cancer? A meta-analysis. Rom J Intern cancer risk in African Americans and rural Africans.
Med. 2013;51(3-4):152–63. Nat Commun. 2015;6:6342. https://doi.org/10.1038/
28. Park Y, Spiegelman D, Hunter DJ, et al. Intakes of ncomms7342.
vitamins A, C, and E and use of multiple vitamin 41. Borges-Canha M. Role of colonic microbiota in
supplements and risk of colon cancer: a pooled colorectal carcinogenesis: a systematic review. Rev
analysis of prospective cohort studies. Cancer Esp Enferm Dig. 2015;107(11):659–71.
Causes Control. 2010;21(11):1745–57. https://doi. 42. Encarnação JC, Abrantes AM, Pires AS, Botelho
org/10.1007/s10552-010-9549-y. MF. Revisit dietary fiber on colorectal cancer:
29. Panic N, Nedovic D, Pastorino R, et al. Carotenoid butyrate and its role on prevention and treatment.
intake from natural sources and colorectal cancer: a Cancer Metastasis Rev. 2015;34:465–78. https://doi.
systematic review and meta-analysis of epidemiolog- org/10.1007/s10555-015-9578-9.
ical studies. Eur J Cancer Prev. 2017;26(1):27–37. 43. Kelly CJ, Zheng L, Campbell EL, et al. Crosstalk
https://doi.org/10.1097/CEJ.0000000000000251. between microbiota-derived short-chain fatty acids
30. Mannisto S, Yaun S-S, Hunter DJ, et al. Dietary and intestinal epithelial HIF augments tissue barrier
carotenoids and risk of colorectal cancer in a pooled function. Cell Host Microbe. 2015;17:662–71.
analysis of 11 cohort studies. Am J Epidemiol. 44. Sebastián C, Mostoslavsky R. Untangling the fiber
2007;165:246–55. yarn: butyrate feeds Warburg to suppress colorectal
31. Kim D-H, Smith-Warner SA, Spiegelman D, et al. cancer. Cancer Discov. 2014;4(12):1368–70.
Pooled analyses of 13 prospective cohort stud- 45. Keum N, Lee DH, Kim R, et al. Visceral adiposity and
ies on folate intake and colon cancer. Cancer colorectal adenomas: dose-response meta-analysis of
Causes Control. 2010;21(11):1919–30. https://doi. observational studies. Ann Oncol. 2015;26:1101–9.
org/10.1007/s10552-010-9620-8. https://doi.org/10.1093/annonc/mdu563.
32. Figueiredo JC, Mott LA, Giovannucci E, et al. Folic 46. Riondino S, Roselli M, Palmirotta R, et al. Obesity
acid and prevention of colorectal adenomas: a com- and colorectal cancer: role of adipokines in tumor
bined analysis of randomized clinical trials. Int J initiation and progression. World J Gastroenterol.
Cancer. 2011;129:192–203. 2014;20(18):5177–90. https://doi.org/10.3748/wjg.
33. Bonovas S, Fiorino G, Lytras T, et al. Calcium v20.i18.5177.
supplementation for the prevention of colorectal 47. Tucker LA, Thomas KS. Increasing total fiber intake
adenomas: a systematic review and meta-analysis of reduces risk of weight and fat gains in women. J
randomized controlled trials. World J Gastroenterol. Nutr. 2009;139:576–81.
2016;22(18):4594–603. https://doi.org/10.3748/wjg. 48. Baer DJ, Rumpler WV, Miles CW, Fahey GC Jr.
v22.i18.4594. Dietary fiber decreases the metabolizable energy
34. Keum N, Lee DH, Greenwood DC, et al. Calcium content and nutrient digestibility of mixed diets fed
intake and colorectal adenoma risk: dose-response to humans. J Nutr. 1997;127:579–86.
References 553
49. Verspreet J, Damen B, Broekaert WF, et al. A critical 63. Jacobs S, Harmon BE, Ollberding NJ, et al. Among 4
look at prebiotics within the dietary fiber concept. diet quality indexes, only the alternate Mediterranean
Annu Rev Food Sci Technol. 2016;7:167–90. diet score is associated with better colorectal
50. Zeng H, Lazarova DL, Bordonaro M. Mechanisms cancer survival and only in African American women
linking dietary fiber, gut microbiota and colon in the Multiethnic Cohort. J Nutr. 2016;146:
cancer prevention. World J Gastrointest Oncol. 1746–55.
2014;6(2):41–51. 64. Fung TT, Kashambwa R, Sato K, et al. Post diag-
51. Godos J, Bella F, Sciacca S, Galvano F, et al. nosis diet quality and colorectal cancer survival in
Vegetarianism and breast, colorectal and prostate women. PLoS One. 2014;9(12):e115377. https://doi.
cancer risk: an overview and meta-analysis of cohort org/10.1371/journal.pone.0115377.
studies. J Hum Nutr Diet. 2017;30(3):349–59. 65. Zhu Y, Wu H, Wang PP, et al. Dietary patterns and
https://doi.org/10.1111/jhn.12426. colorectal cancer recurrence and survival: a cohort
52. Steck SE, Guinter M, Zheng J, Thomson CA. Index- study. BMJ Open. 2013;3:e002270. https://doi.
based dietary patterns and colorectal cancer risk: a org/10.1136/bmjopen-2012-.
systematic review. Adv Nutr. 2015;6:763–673. 66. Meyerhardt JA, Niedzwiecki D, Hollis D, et al.
53. Azeem S, Gillani SW, Siddiqui A, et al. Diet Association of dietary patterns with cancer recur-
and colorectal cancer risk in Asia—a sys- rence and survival in patients with stage III colon
tematic review. Asian Pac J Cancer Prev. cancer. JAMA. 2007;298(7):754–64.
2015;16(13):5389–96. 67. O'Keefe SJD, Li JV, Lahti L, et al. Fat, fiber and
54. Yusof AS, Isa ZM, Shah SA. Dietary patterns and cancer risk in African Americans and rural Africans.
risk of colorectal cancer: a systematic review of Nat Commun. 2015;6:6342. https://doi.org/10.1038/
cohort studies (2000-2011). Asian Pacific J Cancer ncomms7342.
Prev. 2012;13(9):4713–7. https://doi.org/10.7314/ 68. Sansbury LB, Wanke K, Albert PS, et al. The effect
ACJCP.2012.13.9.4713. of strict adherence to a high-fiber, high-fruit and
55. Magalhaesa B, Peleteiro B, Lunet N. Dietary pat- -vegetable, and low-fat eating pattern on adenoma
terns and colorectal cancer: systematic review and recurrence. Am J Epid. 2009;170(5):576–84.
meta-analysis. Eur J Cancer Prev. 2012;21(1):15–23. 69. Lanza E, Yu B, Murphy G, et al. The Polyp Prevention
https://doi.org/10.1097/CEJ.0b013e 3283472241. Trial–Continued Follow-up Study: no effect of a
56. Vargas AJ, Neuhouser ML, George SM, et al. low-fat, high-fiber, high-fruit, and -vegetable diet on
Diet quality and colorectal cancer risk in the adenoma recurrence eight years after randomization.
Women’s Health Initiative Observational Study. Epidemiol Biomarkers Prev. 2007;16(9):1745–52.
Am J Epidemiol. 2016;184(1):23–32. https://doi. 70. Lanza E, Hartman TJ, Albert PS, et al. High dry
org/10.1093/aje/kwv304. bean intake and reduced risk of advanced colorectal
57. Gilsing AMJ, Schouten LJ, Goldbohm RA, et al. adenoma recurrence among participants in the polyp
Vegetarianism, low meat consumption and the risk of prevention trial. J Nutr. 2006;136:1896–903.
colorectal cancer in a population based cohort study. 71. Schatzkin A, Lanza E, Corle D, et al. Lack of effect
Sci Rep. 2015;5:13484. https://doi.org/10.1038/ of a low-fat, high-fiber diet on the recurrence of
srep13484. colorectal adenomas. Polyp prevention trial study
58. Orlich MJ, Singh PN, Sabate J, et al. Vegetarian group. N Engl J Med. 2000;342:1149–55.
dietary patterns and the risk of colorectal cancers. 72. McKeown-Eyssen GE, Bright-See E, Bruce
JAMA Intern Med. 2015;175(5):767–76. https://doi. WR, et al. A randomized trial of a low-fat high
org/10.1001/jamainternmed.2015.59. fibre diet in the recurrence of colorectal polyps.
59. Nimptsch K, Malik VS, Fung TT, et al. Dietary Toronto Polyp Prevention Group. J Clin Epidemiol.
patterns during high school and risk of colorectal 1994;47:525–36.
adenoma in a cohort of middle-aged women. Int J 73. Yao Y, Suo T, Andersson R, et al. Dietary fibre for
Cancer. 2014;134:2458–67. the prevention of recurrent adenomas and carcino-
60. Fung TT, FB H, Wu K, et al. The Mediterranean mas. Cochrane Database Syst Rev 2017; CD003430.
and Dietary Approaches to Stop Hypertension doi:https://doi.org/10.1002/14651858.CD03430. pub2.
(DASH) diets and colorectal cancer. Am J Clin 74. Tantamango YM, Knutsen SF, Beeson WL, et al.
Nutr. 2010;92(6):1429–35. https://doi.org/10.3945/ Foods and food groups associated with the incidence
ajcn.2010.29242. of colorectal polyps: the Adventist Health Study.
61. Reedy J, Mitrou PN, Krebs-Smith SM, et al. Index- Nutr Cancer. 2011;63(4):565–72. https://doi.org/10
based dietary patterns and risk of colorectal can- .1080/01635581.2011.55988.
cer the NIH-AARP diet and health study. Am J 75. Levi F, Pasche C, La Vecchia C, et al. Food
Epidemiol. 2008;168:38–48. groups and colorectal cancer risk. Br J Cancer.
62. Kim MK, Sasaki S, Otani T, et al. Dietary pat- 1999;79(7/8):1283–7.
terns and subsequent colorectal cancer risk by 76. Ben Q, Zhong J, Liu J, et al. Association between con-
subsite: a prospective cohort study. Int J Cancer. sumption of fruits and vegetables and risk of colorec-
2005;115:790–8. tal adenoma a PRISMA-compliant meta-analysis of
observational studies. Medicine. 2015;94(42):e1599. noma in a cancer screening trial. Am J Clin Nutr.
https://doi.org/10.1097/MD.000000000000.1599. 2007;86:1754–64.
77. Kashino I, Mizoue T, Tanaka K, et al. Vegetable 90. Park Y, Subar AF, Kipnis V, et al. Fruit and veg-
consumption and colorectal cancer risk: an evalua- etable intakes and risk of colorectal cancer in the
tion based on a systematic review and meta-analysis NIH–AARP diet and health study. Am J Epidemiol.
among the Japanese population. Jpn J Clin Oncol. 2007;166:170–80.
2015;45(10):973–9. https://doi.org/10.1093/jjco/ 91. Michels KB, Giovannucci E, Chan AT, et al. Fruit
hyv111. and vegetable consumption and colorectal adeno-
78. Tse G, Eslick GD. Cruciferous vegetables and risk mas in the Nurses’ Health Study. Cancer Res.
of colorectal neoplasms: a systematic review and 2006;66(7):3942–53.
meta-analysis. Nutr Cancer. 2014;66(1):128–39. 92. Zhu B, Sun Y, Qi L, et al. Dietary legume consump-
https://doi.org/10.1080/01635581. 2014.852686. tion reduces risk of colorectal cancer: evidence
79. Turati F, Guercio V, Pelucchi C, et al. Colorectal from a meta-analysis of cohort studies. Sci Rep.
cancer and adenomatous polyps in relation to allium 2015;5:8797. https://doi.org/10.1038/srep08797.
vegetables intake: a meta-analysis of observational 93. Wang Y, Wang Z, Fu L, et al. Legume consump-
studies. Mol Nutr Food Res. 2014;58:1907–14. tion and colorectal adenoma risk: a meta-analysis of
https://doi.org/10.1002/mnfr.201400169. observational studies. PLoS One. 2013;8(6):e67335.
80. QJ W, Yang Y, Vogtmann E, et al. Cruciferous veg- https://doi.org/10.1371/journal.pone. 0067335.
etables intake and the risk of colorectal cancer: a 94. Yan L, Spitznagel EL, Bosland MC. Soy con-
meta-analysis of observational studies. Ann Oncol. sumption and colorectal cancer risk in humans: a
2013;24:1079–87. https://doi.org/10.1093/annonc/ meta-analysis. Cancer Epidemiol Biomark Prev.
mds601. 2010;19(1):148–58.
81. Aune D, Lau R, Chan DSM, et al. Nonlinear reduc- 95. Yang G, Shu X-O, Li H, et al. Prospective cohort
tion in risk for colorectal cancer by fruit and veg- study of soy food intake and colorectal cancer risk in
etable intake based on meta-analysis of prospective women. Am J Clin Nutr. 2009;89:577–83.
studies. Gastroenterology. 2011;141:106–18. 96. Akhter M, Inoue M, Kurahashi N, et al. Dietary
82. Koushik A, Hunter DJ, Spiegelman D, et al. Fruits, soy and isoflavone intake and risk of colorectal
vegetables, and colon cancer risk in a pooled cancer in the Japan Public Health Center–Based
analysis of 14 cohort studies. J Natl Cancer Inst. Prospective Study. Cancer Epidemiol Biomark Prev.
2007;99:1471–83. 2008;17(8):2128–35.
83. Kunzmann AT, Coleman HG, Huang W-Y, et al. 97. Lin J, Zhang SM, Cook NR, et al. Dietary intakes
Fruit and vegetable intakes and risk of colorec- of fruit, vegetables, and fiber, and risk of colorec-
tal cancer and incident and recurrent adenomas tal cancer in a prospective cohort of women
in the PLCO cancer screening trial. Int J Cancer. (United States). Cancer Causes Control.
2016;138:1851–61. 2005;16:225–33.
84. Leenders M, Siersema PD, Overvad K, et al. 98. Haas P, Machado MJ, Anton AA, et al. Effectiveness
Subtypes of fruit and vegetables, variety in con- of whole grain consumption in the prevention of
sumption and risk of colon and rectal cancer in the colorectal cancer: meta-analysis of cohort studies.
European Prospective Investigation into Cancer and Int J Food Sci Nutr. 2009;60(Suppl. 6):1–13.
Nutrition. Int J Cancer. 2015;137:2705–14. 99. Bakken T, Braaten T, Olsen A, et al. Consumption
85. Aoyama N, Kawado M, Yamada H, et al. Low intake of whole-grain bread and risk of colorectal cancer
of vegetables and fruits and risk of colorectal cancer: among Norwegian women (the NOWAC Study).
the Japan Collaborative Cohort Study. J Epidemiol. Forum Nutr. 2016;8:40. https://doi.org/10.3390/nu
2014;24(5):353–60. 8010040.
86. Vogtmann E, Xiang Y-B, Li H-L, et al. Fruit and 100. Abe SK, Inoue M, Sawada N, et al. Rice, bread,
vegetable intake and the risk of colorectal cancer: noodle and cereal intake and colorectal cancer in
results from the Shanghai Men’s Health Study. Japanese men and women: the Japan Public Health
Cancer Causes Control. 2013;24(11):1935–45. Center-based prospective Study (JPHC Study).
87. van Duijnhoven FJB, Bueno-De-Mesquita HB, BJC. 2014;110:1316–21. https://doi.org/10.1038/
Ferrari P, et al. Fruit, vegetables, and colorectal bjc.2013.799.
cancer risk: the European Prospective Investigation 101. Kyrø C, Skeie G, Loft S, et al. Intake of whole
into Cancer and Nutrition. Am J Clin Nutr. grains from different cereal and food sources and
2009;89:1441–51. incidence of colorectal cancer in the Scandinavian
88. Nomura AMY, Wilkens LR, Murphy SP, et al. HELGA cohort. Cancer Causes Control.
Association of vegetable, fruit, and grain intakes 2013;24:1363–74.
with colorectal cancer: the Multiethnic Cohort 102. Egeberg R, Olsen A, Loft S, et al. Intake of who-
Study. Am J Clin Nutr. 2008;88:730–7. legrain products and risk of colorectal cancers
89. Millen AE, Subar AF, Graubard BI, et al. Fruit and in the diet, cancer and health cohort study. BJC.
vegetable intake and prevalence of colorectal ade- 2010;103:730–4.
References 555
103. Schatzkin A, Mouw T, Park Y, et al. Dietary fiber 106. Yang M, FB H, Giovannucci EL, et al. Nut consump-
and whole-grain consumption in relation to colorec- tion and risk of colorectal cancer in women. Eur J
tal cancer in the NIH-AARP Diet and Health Study. Clin Nutr. 2016;70:333–7.
Am J Clin Nutr. 2007;85:1353–60. 107. Yeh CC, You SL, Chen CJ, Sung FC. Peanut con-
104. Larsson SC, Giovannucci E, Bergkvist L, Wolk sumption and reduced risk of colorectal cancer in
A. Whole grain consumption and risk of colorectal women: a prospective study in Taiwan. World J
cancer: a population-based cohort of 60,000 women. Gastroenterol. 2006;12(2):222–7.
BJC. 2005;92:1803–7. 108. Jenab M, Ferrari P, Slimani N, et al. Association of
105. Knudsen MD, Kyrø C, Olsen A, et al. Self-reported nut and seed intake with colorectal cancer risk in
whole-grain intake and plasma alkylresorcinol the European Prospective Investigation into Cancer
concentrations in combination in relation to the and Nutrition. Cancer Epidemiol Biomark Prev.
incidence of colorectal cancer. Am J Epidemiol. 2004;13(10):1595–603.
2014;179(10):1188–96.
Dietary Patterns, Whole Plant
Foods, Nutrients 20
and Phytochemicals in Breast
Cancer Prevention
and Management
Keywords
Fruits • Vegetables • Soy foods • Whole-grains • Seeds • Dietary fiber •
Vitamins • Protein source • Carotenoids • Flavonoids • Isoflavonoids •
Lignan • Dietary patterns • Premenopausal • Postmenopausal • Body mass
index • Mortality • Estrogen
Key Points processes, which can influence BC initiation

• Dietary choices including: (1) level of adherence and progression..
to healthy vs Western dietary patterns; (2) high • Lifestyle indicators, which are associated with
vs low dietary energy density intake; (3) type increased BC risk, recurrence or mortality,
and level of dietary fat, fiber and protein con- especially for postmenopausal women, may
sumed; (4) adequate vs inadequate intake of cal- include having an overweight or obese BMI,
cium, folate and α-tocopherol; (5) type and weight gain by over 15 lbs. over 4 years, and
levels of non-starchy vegetables and fruits con- physical inactivity. Patients with BC are most
taining dietary carotenoids and flavonoids often either overweight or obese at diagnosis
intake; (6) level of phytoestrogen containing and obesity increases mortality risk in both
legumes and seeds consumed; and (7) higher pre- and postmenopausal women with BC.
vs lower intake of alcohol or coffee are exam- • Meta-analyses reported that healthy dietary
ples of dietary factors that may influence breast patterns reduced overall BC risk, whereas a
cancer (BC) risk, recurrence or mortality. high consumption of alcohol and a Western
• Biological factors and mechanisms associated diet increased BC risk. Healthy dietary pat-
with diet and BC risk and survival include: terns, especially the Mediterranean diet,
body weight and central adiposity, tumor DASH diet and the vegan diet are effective in
advancement, systemic and tissue lipid/fatty reducing BC risk and improving odds for sur-
acid peroxidation and inflammation, epigenetic vival. Key adverse dietary components for BC
and transcriptional regulation, hormone levels risk and survival include high intake of red
(e.g., estrogen, insulin, leptin, adiponectin and and processed meats, high energy dense and
growth factor cascades), insulin resistance, and high glycemic foods and beverages and >1
various endometabolic and colonic microbiota alcoholic beverage/day.

558 20 Dietary Patterns, Whole Plant Foods, Nutrients and Phytochemicals in Breast Cancer Prevention
• Highly colored non-starchy vegetables rich in first full-term pregnancy, the use of exogenous
carotenoids and flavonoids have been associ- hormones (oral contraceptives and combined
ated with reduced BC risk, especially in estro- postmenopausal hormone replacement therapy),
gen receptor negative BC. After BC diagnosis, alcohol consumption, excess weight, insulin resis-
soy foods (>10 mg isoflavones/day or > ½ cup tance, diet, and physical activity [2–6].
of soy milk or 2 ounces of tofu/day) may help Survival in women diagnosed with early-
to reduce risk of BC recurrence or mortality in stage, invasive breast cancer has improved dra-
both Asian and Western women. matically in the past 25 years [2–7]. This is
largely due to the use of evolving pharmacologi-
cal therapies targeting a reduction in estrogen
20.1 Introduction action and exposure. However, it is estimated that
up to 90% of overall cancer risk may be attribut-
In women, breast cancer (BC) is the most com- able to environmental and lifestyle factors [6].
mon cancer worldwide, with an estimated 2.4 mil- Women diagnosed with BC often change their
lion cases in 2015 [1]. Between 2005 and 2015, eating behavior towards healthier food choices in
BC remained the fifth leading global cause of can- an attempt to improve their overall health, well-
cer. Overall incident cases have increased by 43% being and survival but they are often unaware of
because of population growth (contributing an specific and effective dietary guidance [7]. A
additional 13%) and aging (contributing 15%). number of prospective cohort and randomized
The odds of developing BC between birth and controlled trials (RCTs) support BC protective
79 years are 1 in 14 for women globally but these benefits of healthy dietary patterns containing
odds increase to 1 in 9 for women in the highest fiber and phytochemical rich whole or minimally
income countries such as in North America, processed plant foods (whole plant foods),
Western Europe, and Australia. The worldwide healthy vegetable oils and omega 3 (n-3) fatty
rise in BC incidence, despite continuous improve- acids sources. Also, important is the adherence to
ments in BC prognosis, is primarily due to longer a healthy lifestyle and maintaining and achieving
life expectancy, increased aging populations, and a healthy weight which can significantly reduce
the adoption of Western diets and lifestyles [2, 3]. BC risk, recurrence and improve survival after
Hormones such as estrogens, progesterone, insu- BC diagnosis compared to Western diets and life-
lin, and growth factors, which peak with puberty, styles [7–10]. Dietary factors including energy
pregnancy, and lactation, may influence the life- density, type of fat, levels of dietary fiber (fiber),
time risk of BC because they modulate the struc- phytoestrogens, carotenoids, flavonoids, type of
ture, growth, and epigenetics of tumor cells. Risk protein source (red meat vs soy foods), beverages
doubles each decade until menopause, when the such as alcohol, and others components may play
risk slows down or remains stable, but breast can- an important role in both promoting and inhibit-
cer is more common after menopause. In many ing BC development. Biological factors and
countries, the 5-year survival rate for women mechanisms associated with diet and BC risk and
diagnosed with Stage I/II BC (only spread to tis- survival include: body weight and central adipos-
sues or nodes under the arm) is 80–90% but if the ity, tumor immunity, systemic and tissue lipid/
cancer stage is more advanced (spread to distant fatty acid peroxidation and inflammation, epi-
lymph nodes or organs) the survival rate falls to genetic and transcriptional regulation, hormone
about 25%. Breast cancer is a heterogeneous dis- levels (e.g., estrogen, insulin, leptin, adiponectin
ease with various subtypes [5]. Common BC and growth factor cascades), insulin resistance,
molecular subtype biological markers, include the and various endometabolic and colonic microbi-
presence or absence of estrogen and progesterone ota processes, which can influence BC initiation
receptors, or human epidermal growth factor and progression [2–12]. The objective of this
receptor 2 (HER2). More typically, BC risk and chapter is to comprehensively assess the effects
survival is associated with lifestyle, reproductive, of dietary patterns, whole plant foods, nutrients
and other environmental factors, including aging, and phytochemicals on BC risk, recurrence and
early age at menarche, lactation, late menopause, survival.
20.2 Specific Lifestyle, Dietary and Lifecyle Factors 559
20.2 S
pecific Lifestyle, Dietary reference, women meeting <3 recommendations
and Lifecyle Factors showed a 300+% increased BC risk, especially in
postmenopausal women. For premenopausal
20.2.1 Overview women, excessive intake of energy dense foods
and drinks that promote weight gain increased
Lifestyle habits play an important role in BC risk, BC risk by 200+% (p-interaction = 0.014). For
recurrence and mortality [11–16]. World Cancer postmenopausal women, low intake of healthy
Research Fund (WCRF)/American Institute for fiber-rich plant foods such as fruit, vegetables,
Cancer Research (AICR) [11, 12] and American whole-grains, and legumes increased BC risk by
Cancer Society (ACS) [13] have developed about 250%. The Women’s Health Initiative
guidelines, recommending a healthy weight, a (65,838 postmenopausal women; mean baseline
diet rich in fiber containing plant foods, and age 63 years at baseline; mean 12.6 years of fol-
physical activity as important for lowering over- low-up) found that women with the highest ACS
all cancer risk, including BC prevention and guideline scores had significantly lower risk for
improved post-diagnosis survival (Table 20.1). any cancer by 17%, for BC by 22%, and for
For the WCRF/AICR cancer prevention guide- colorectal cancer by 52% (Fig. 20.1) and similar
lines, the EpiGEICAM case-control study (973 risk reductions for cancer mortality (Fig. 20.2)
cases of BC and 973 controls from 17 Spanish [15]. A multinational European Prospective
regions; age range 22–71 years) found a linear Investigation into Cancer and Nutrition (EPIC)
association between the degree of diet and life- cohort investigated the effect of a healthy life-
style noncompliance and BC risk [14]. This study style index score [HLIS] (5 factors including
found that compared to women who met 6 or diet, physical activity, smoking avoidance, alco-
more healthy lifestyle recommendations as hol consumption and anthropometry; score range
Table 20.1 Adult guidelines for nutrition and physical activity in cancer prevention
World Cancer Research Fund (WCRF)/American
Institute for Cancer Research (AICR) [11, 12] American Cancer Society (ACS) Guidelines [13]
Maintain a healthy lean body weight without being Achieve and maintain a healthy lean body weight
underweight throughout life
Be physically active for at least 30 min every day Be as lean as possible throughout life without being
underweight
Limit consumption of energy dense foods Avoid excess weight gain at all ages. For those who are
(particularly processed foods high in added sugar, currently overweight or obese, losing even a small amount of
or low in fiber, or high in fat) weight has health benefits and is a good place to start
Eat mostly plant foods including a variety of Engage in regular physical activity and limit consumption of
vegetables, fruits, whole grains, and legumes high-calorie foods and beverages as key strategies for
maintaining a healthy weight
Limit animal foods such as red and processed meats Adopt a physically active lifestyle. Adults should engage in
at least 150 min of moderate-intensity or 75 min of
vigorous-intensity activity weekly spread over the week
Limit alcoholic beverages (2 for men and 1 for Limit sedentary behavior such as sitting, lying down,
women a day) watching television, or other forms of screen-based
entertainment. Doing some physical activity above usual
activities, no matter what one’s level of activity, can have
many health benefits
Limit consumption of salty foods and foods Choose foods and beverages in amounts that help achieve
processed with salt and maintain a healthy weight
Meet nutritional needs through diet Limit consumption of processed meat and red meat
Breastfeed exclusively for up to 6 months Eat at least 2.5 cups of vegetables and fruits each day
Cancer survivors should follow the Choose whole grains instead of refined grain products
recommendations for cancer If you drink alcoholic beverages, limit consumption. Drink no
prevention more than 1 drink per day for women or 2 per day for men
Any cancer (except skin cancer) Breast cancer Colorectal cancer

1.1
Fully Adjusted Hazard Ratio 1
0.9
0.8
0.7
0.6
0.5
0.4
0 to 2 3 4 5 6 7 to 8
American Cancer Society - Cancer Prevention Score
Fig. 20.1 For postmenopausal cancer risk, the effect of American Cancer Society cancer prevention score based on
adherence to nutrition and physical activity guidelines from the US Women’s Health Initiative (all cancers p < 0.001)
(adapted from [15)
Any cancer (except skin; p <.001) Breast cancer (p =.049)

Colorectal cancer (p <.001)
1.1
1
Fully Adjusted Hazard Ratio
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0 to 3 4 to 5 6 to 8
American Cancer Society - Cancer Prevention Score
Fig. 20.2 For postmenopausal cancer mortality, the effect of American Cancer Society (ACS) cancer prevention score
based on adherence to nutrition and physical activity guidelines from the US Women’s Health Initiative (adapted from
[15])
of 0–4 for each component with higher values that each 1 point increase in HLIS lowered BC
indicating healthier behaviors) on postmeno- risk by 3% [16]. The effect of increased HLIS
pausal BC risk [16]. This EPIC study (242,918 score on BC risk in all women is summarized in
postmenopausal women; mean baseline age Fig. 20.3. Also, the higher specific lifestyle
53 years; median 10.9 years of follow-up) found scores are associated with reduced risk of
1.3
1.2
Hazard Ratio for Post - menopausal BC

1.1
0.9
0.8
0.7
0.6
0.5
< = 5 points 6 to 10 points 11 to 15 points > = 16 points
Healthy Lifestyle Index Score*
Fig. 20.3 Association between Healthy Lifestyle Index Score (HLIS) and post-menopausal breast cancer (BC) risk
from the EPIC cohort study including 242,918 women; median 10.9 years of follow-up (p-trend <0.001; multivariate
adjusted) (adapted from [16]). * HLIS was constructed from five factors (diet, physical activity, smoking, alcohol con-
sumption and anthropometry) by assigning scores of 0–4 to categories of each component, for which higher values
indicate healthier behaviors
Diet (p < .005) BMI (p <.001)

Physical activity (p <.001) Alcohol (p <.001)
Smoking (p =.171)
1.05
Hazard Ratios for Post - menopausal BC
0.95
0.9
0.85
0.8
0 1 2 3 4
Healthy Lifestyle Index Score
Fig. 20.4 Association between Healthy Lifestyle Index Score (HLIS) and post-menopausal breast cancer (BC) risk
from EPIC cohort study including 242,918 women; median 10.9 years of follow-up (adapted from [16])
developing BC among postmenopausal women and postmenopausal women based on the 2017
(Fig. 20.4). Table 20.2 provides a summary of the CUP Breast Cancer Systematic Literature Review
overall findings on the effects of diet, nutrition, and the CUP Expert Panel discussions in June
physical activity and BC risk in premenopausal 2016 [2].
Table 20.2 Continuous Update Project (CUP) expert panel findings on diet, nutrition and physical activity and breast
cancer (BC) risk (adapted from [2])
Evidence Premenopausal BC Postmenopausal BC
Convincing Adult attained height: developmental factors Alcoholic beverages: one drink or 10 g alcohol
leading to greater linear growth by 5 cm increases increases BC risk by 9%
BC risk by 6% Body fatness: BMI increase by 5 kg/m2
elevates BC risk by 12% and mortality risk by
20%, 10 cm increase in waist circumference
elevates BC risk by 6%, and 0.1-unit higher
waist-to-hip ratio increases BC risk by 10%
Adult weight gain: 5 kg body weight gain
increases BC risk by 6%
Adult attained height: developmental factors
leading to greater linear growth by 5 cm
elevates BC risk by 8%
Probable Vigorous physical activity: 30 minutes of daily Total physical activity: higher physical
vigorous physical activity reduces BC risk by 9% activity reduces BC risk by 13%
Body fatness in young adulthood (18–30 years): Vigorous physical activity: 30 minutes/day of
5 kg/m2 increase in BMI decreases BC risk by 18% vigorous physical activity reduces BC risk by
Body fatness (before menopause): BMI increase 6%
by 5 kg/m2 reduces BC risk by 7%, 10 cm increased Body fatness in young adulthood (18–
in waist circumference by 10 cm or 0.1unit 30 years): BMI increase by 5 kg/m2 decreases
increased waist-to-hip ratio is not associated with BC risk by 18%
BC risk Lactation: 5 months duration reduces BC risk
Alcoholic beverages: one drink or 10 g alcohol/day by 2%
increases BC risk by 5%.
Lactation: 5 months duration reduces BC risk by
2%
Suggestive Non-starchy vegetables: 200 g/day decreases the Non-starchy vegetables: 200 g/day decreases
risk of estrogen receptor negative BC risk by the risk of estrogen receptor negative BC by
18–21% 18–21%
Foods containing carotenoids: inverse association Foods containing carotenoids: inverse
with beta-carotene, total carotenoids and lutein in association with beta-carotene, total
BC risk carotenoids and lutein in BC risk
Diets high in calcium: 300 mg/day reduces BC Diets high in calcium: 300 mg/day reduces BC
risk by 13% risk by 4%
Dairy Products: 200 g/day reduces BC risk by 5%
Total physical activity: higher physical activity
reduces BC risk by 7%
With improvements in medical diagnosis and prognosis. Dietary and exercise patterns associ-
treatment, there are increasing numbers of women ated with lower BC risk include: diets with
with BC risk factors and long-term BC survivors 45–65% energy from fiber rich carbohydrates
who are looking at dietary pattern modification to including whole grains, fruits, vegetables and
prevent occurrence or recurrence and mortality legumes and low in refined grains and added
[17]. Data from RCTs studying diet, exercise, or sugar; 10–35% energy from healthy dietary fats
combined diet and exercise interventions show
low in saturated fats; and 10–35% energy from
that the most consistent findings are that reduc- protein which is very low or devoid of processed
tions in adiposity, and maintaining or gain of skel- meats, plus avoiding long periods of physical inac-
etal muscle had the most beneficial protective tivity and including 150 minutes/week of moder-
effects on BC outcomes, including survival, risk of ate intensity aerobic activity or 75 minutes/week
recurrence, or biomarkers associated with of vigorous activity and resistance exercise at least
2 days/week [17]. However, the relationship all-cause mortality rates by 23% compared with
between dietary pattern and BC risk is complex maintaining body weight [20]. In contrast, a 2017
with differences between pre- and postmenopausal Kaiser Permanente study (12,590 stage I-III
women, including estrogen levels, BMI, dietary breast cancer patients; mean age 59 years)
energy density, diet during adolescence, level of observed that compared to weight maintenance,
wine or other types of alcoholic beverage, and ≥10% weight losses were associated with worse
fiber intake are all examples of factors which can survival by 163% increase in all-cause mortality
influence the effects of dietary pattern on BC risk, [21]. Also, increased waist circumference and
recurrence and survival. Unhealthy dietary pat- waist to hip ratio have been associated with
terns and obesity, especially among postmeno- increased BC risk in postmenopausal women [2].
pausal women, are associated with changes in
biomarkers, such as insulin resistance, lipopro- 20.2.2.2 W eight Change Across
teins, estradiol, and micr obiota dysbiosis that are the Lifespan
risk factors for BC and cardiovascular diseases A review of Nurses’ Health Studies findings on
[18]. A comparison of Western and Healthy BC incidence and survival show a complex rela-
dietary patterns characterizing food components tionship between weight change and BC risk
and nutrients are summarized in Appendix A. across the lifespan [10, 14]. Levels of body fat-
ness in childhood and high BMI at age 18 years
are inversely associated with adult plasma insulin-
20.2.2 Noteworthy Factors like growth factor 1 (IGF-1) levels, a hormone
similar in molecular structure to insulin which
This section highlights some specific factors plays an important role in childhood growth and
including elevated BMI in postmenopausal continues to have anabolic effects in adults.
women or after BC diagnosis, ≥15 kg increase in Although higher BMI at age 18 years was
body weight over 4 years, dietary energy density, inversely associated with both pre- and postmeno-
adolesent and early adulthood diet, and wine and pausal BC risk, weight gain after age 18 years was
other alcoholic beverages, coffee consumption, positively associated with risk after menopause, in
and physical activity are related factors that may those who never used hormone therapy (HT). In a
influence BC risk and survival. subsequent analysis with 26 years of follow-up, it
was observed that among women who never used
20.2.2.1 Body Mass Index (BMI) HT, those who had lost more than 10 kilograms
BMI levels outside the normal range in post- after menopause and maintained their weight loss
menopausal women and after BC diagnosis are had a lower risk of BC than women with stable
associated with increased BC risk, recurrence and weight since menopause. Short-term gain over
mortality [2, 10, 19–21]. A comprehensive sys- 4 years by ≥15 lbs. was associated with higher BC
tematic review and meta-analysis (82 observa- risk in premenopausal than postmenopausal
tional studies; 213,075 BC survivors; 41,477 women compared to no weight change.
deaths with 23,182 from BC) demonstrated that
higher BMI levels in BC survivors, especially in 20.2.2.3 Dietary Energy Density
the obesity range, have unfavorable effects on The intake of high dietary energy dense diets is
overall BC risk and survival in both pre- and post- associated with increased BC risk or high dense
menopausal BC [19]. For each 5 kg/m2 increment breast volume [22, 23]. The US Cancer Prevention
of BMI before, <12 months after, and ≥12 months Study II Nutrition Cohort (56,795 postmenopausal
after diagnosis, there is an increased BC mortality women; 11.7 years of follow-up) showed that high
risk observed by 18%, 14%, and 29%, respec- vs low dietary energy density (≥1.7 vs <1.2 kcal/g)
tively. Another systematic review and meta-anal- significantly increased BC risk by 20%
ysis (12 observational studies; 23,832 women) (p-trend = 0.03) independent of hormone receptor
found that weight gain of >10% after BC diagno- status, BMI, age or physical activity level [22].
sis is associated with significantly higher Also, in premenopausal women each 1 kcal/g
increase in dietary energy density is associated protective effect threshold for up-to 1 glass of
with a 26% increase in dense breast volume %, a wine/day above which there was a 0.6% increase
potential risk factor of BC, after multivariate in BC risk for each additional g of alcohol/day
adjustment (p = 0.01) [23]. [26]. A 2008 meta-analysis (dose-response analy-
sis including 4 cohorts and 12 case-control stud-
20.2.2.4 A dolescence and Early ies) showed that an increase in alcohol consumption
Adulthood of 10 g ethanol/day was associated with statisti-
The Nurses’ Health Study II found that diet cally significant increased BC risk for hormone
quality during adolescence and early adulthood receptor positive women by 7–15% [27]. A 2014
can influence BC risk in later adulthood [10, 24, meta-analysis (11 case-control studies) found that
25]. A 2017 Nurses’ Health Study (45,204 hormone receptor negative women diagnosed with
women; food frequency questionnaire in 1998 BC with low to moderate alcohol consumption
about their high school diet (HS-FFQ) and a had a lower risk of BC-specific mortality [28]. The
FFQ in 1991 when they were ages 27–44 years; 2016 Danish Diet, Cancer and Health Study
22 years of follow-up; 1477 BC cases) found (21,523 postmenopausal women; 11 years of fol-
that an adolescent and early adulthood inflam- low-up) found that increased alcohol consumption
matory dietary pattern characterized by sugar- over a 5-year period resulted in increased BC risk
sweetened and diet soft drinks, refined grains, and lower coronary heart disease risk compared to
red and processed meat, and margarine, and low stable moderate alcohol intake [29].
intake of green leafy vegetables, cruciferous
vegetables, and coffee, was associated with an 20.2.2.6 Coffee
increased incidence of premenopausal BC [24]. Coffee may impact risk of pre- and post-meno-
Women with a high inflammatory pattern score pausal BC differently [30–32]. A dose response
in adolescence and early adulthood had signifi- meta-analysis (37 published observational stud-
cantly increased risk for premenopausal BC by ies; 966,263 participants; 59,018 BC cases)
35% and by 41% compared with women with found a linear dose-response relationship for BC
healthy low inflammatory diets. This increased risk with coffee and caffeine, with the risk of BC
risk did not extend to postmenopausal BC and decreased by 2% (p = 0.05) for every 2 cups/day
was not significantly different by hormone increment in coffee intake [30]. In subgroup anal-
receptor subtype. A 2016 Nurses’ Health Study yses, higher coffee and caffeine intake by post-
(45,204 women who completed a 124-item food menopausal women significantly reduced BC
frequency questionnaire about their high-school risk by 6% and there was a significant 31% lower
diet; 22 years of follow-up; 863 cases of pre- BC risk for BRCA mutation carriers with higher
menopausal BC and 614 cases of postmeno- coffee consumption. A 2015 EPIC study of pre-
pausal BC) showed those consuming the highest and postmenopausal women (335,060 women;
quintile of the prudent dietary pattern, character- 11 years of follow-up; 1064 premenopausal BC
ized by high intake of vegetables, fruits, legumes, cases and 9134 postmenopausal cases) found that
fish and poultry, had a 16% lower risk of pre- higher coffee intake was associated with a sig-
menopausal BC (p-trend = 0.07) compared with nificant 10% lower BC risk [31]. Caffeinated and
the lowest quintile [25]. decaffeinated coffee were not associated with
premenopausal BC risk. A 2008 Nurses’ Health
20.2.2.5 W ine and Overall Alcohol Study (85,987 women; 22 years of follow-up;
Consumption 5272 BC cases) observed no significant associa-
Low intake of wine or other alcoholic beverages tion between caffeinated and decaffeinated cof-
may be protective against BC risk or mortality fee consumption and BC risk [32]. However, a
[26–29]. For wine, a 2016 meta-analysis (8 case- subgroup analysis showed that postmenopausal
control and 18 cohort studies; 21,149 cases) found women with the highest intake of caffeine-
a 36% increase in BC risk for highest vs lowest containing beverages had a significant 12% lower
intake with a non-linear dose response showing a BC risk.
20.2.2.7 Physical Activity least-active women. Also, Nurses Health studies

Physical inactivity increases postmenopausal and indicate that physical activity is important for
possibly premenopausal BC risk [10, 33]. A 2017 survival after breast cancer diagnosis [10]. For
meta-analysis (101 observational studies) found improved BC survival, benefits are observed for
that higher vs. lower levels of moderate-vigorous physical activity equivalent to walking 3–5 h
activity resulted in an approximately 20% lower weekly at an average pace or for those who fol-
risk of BC for both pre- and postmenopausal low the US recommendations of at least 30 min
women [33]. Physical activity is postulated to daily of moderate physical activity for at least
decrease BC risk by lowering ovarian hormone 5 days weekly, independent of activity level
levels. Nurses’ Health Study analyses suggest before diagnosis.
women who reported participating in ≥7 hours of
moderate or vigorous physical activity weekly
had an 18% lower BC risk [10]. In follow-up 20.3 Dietary Patterns
analyses, the cumulative and recent physical
activity of postmenopausal women were inversely Nine systematic reviews and meta-analyses from
associated with BC risk and among younger observational studies and intervention trials pro-
women lifetime physical activity was inversely vide insights on associations between diet quality
associated with risk of premenopausal BC with a and dietary patterns, and BC risk, recurrence and
33% risk reduction, comparing the most- with the mortality (Table 20.3) [34–42].
Table 20.3 Summaries of systematic reviews and meta-analyses on dietary quality and dietary pattern studies and
breast cancer (BC) risk, recurrence and mortality
Systematic reviews and meta- and pooled analyses
van den Brandt and Schulpen 6 cohort studies plus a large MedDiets were associated with lower BC
(2017). cohort (62,573 women; baseline risk for postmenopausal women by 6%
Investigate the relationship between age 55–69 years; mean 20 years and hormone receptor negative BC risk
adherence to the Mediterranean diet of follow-up; alternate MedDiet by 23–27% (highest vs. lowest
(MedDiet) and risk of score excluding alcohol) adherence), which is important because of
postmenopausal BC (and estrogen/ the relatively poor prognosis of these BC
progesterone receptor subtypes) subtypes. The large cohort study found
(The Netherlands Cohort Study and that high adherence to the MedDiet
a meta-analysis) [34] (excluding alcohol) lowered risk of
estrogen receptor negative BC by 40%
(p-trend = 0.032)
Bloomfield et al. (2016). For total cancer mortality 28 cohort This meta-analysis showed that high
Evaluate the association between the studies (2,262,786 participants; adherence to MedDiets lowered pooled
MedDiet with unrestricted fat intake follow-up periods of 4–40 years); for total cancer mortality by 14% and BC
and BC risk and total cancer BC incidence (13 observational incidence risk by 4%
mortality risk [35] studies including 3 cohort studies; PREDIMED trial found a significant
7152 subjects); and for Prevencion lower BC risk by 57%, CVD events by
con Dieta Mediterranea 29%, and diabetes by 30% in the two
(PREDIMED) RCT (4282 women; higher fat MedDiets combined vs a lower
baseline age 60–80 years; MedDiets fat control group
supplemented with extra virgin olive
oil or nuts vs guidance for low fat
diets; 4.8 years)
Pourmasoumi et al. (2016). 4 cohort studies; 9819 women; This analysis showed no significant
Investigate the relationship between mean baseline age 56 years; mean associations between adherence to these
the Healthy Eating Index (HEI)/ 7.7 years of follow-up healthy eating indices and risk of BC
Alternative Healthy Eating Index mortality or survival among
(AHEI) and BC mortality or survival postmenopausal women
rates [36]
(continued)
Wu et al. (2015). 21 case-control studies and 1 This analysis showed significant lowering
Summarize published Chinese data cohort study; 23,201 patients: of BC risk with higher intake of
on the relationship between BC and 10,566 in the experimental group vegetables by 23% and fruit and soy
healthy dietary quality factors [37] and 12,635 in the control group foods by 32% each
Xing et al. (2014). 2 RCTs and one large multi-center Post-diagnostic healthy “lower-fat
Review associations between prospective cohort study with dietary pattern” (with emphasis on
post-diagnostic healthy low-fat 9966 BC patients; 5–7.3 years of increased fruit, vegetables and fiber
dietary patterns and BC recurrence follow-up intake) reduced BC recurrence risk by
and all-cause mortality [38] 23% (p = 0.009) and all-cause mortality
by 17% (p = 0.05)
Liu et al. (2014). 31 case-control studies and 2 Consumption of both soy and fruit
Evaluate associations between cohort studies; 9299 cases and significantly reduced BC risk by 35%
dietary quality factors and BC risk 11,413 controls (p < 0.001) and 34% (p < 0.001),
in Chinese women [39] respectively. The consumption of
vegetables lowered BC risk by 28% with
significant heterogeneity
Farsinejad-Marj et al. (2015). 5 cohort and 3 case-control All pooled results show that MedDiets
Examine the association between studies were inversely associated with BC risk in
MedDiet adherence and BC risk pre- and postmenopausal women;
[40] however, prospective cohort study findings
were inconsistent; and case-control study
findings generally showed an inverse
association between the MedDiet and
pre- and postmenopausal BC risk
Albuquerque et al. (2013). 11 cohort studies and 15 Dietary patterns characterized by
Evaluate the association between case-control studies); 584,437 vegetables, fruit, fish, and soy products,
dietary pattern quality and BC risk women; 28,962 BC cases as well as dietary patterns designated as
[41] traditional MedDiet reduced BC risk
whereas alcoholic patterns were
associated with increased BC risk. Also,
the Western diet was positively associated
with BC risk but a majority of the studies
had insignificant trends
Brennan et al. (2010). 39 case-control and cohort The pooled higher vs lower adherence to:
Clarify the relationship between studies including prudent/healthy, (1) prudent/healthy dietary patterns
dietary patterns and breast cancer Western, and drinker dietary lowered BC risk by 11% (p = 0.02); (2)
risk [42] patterns drinker dietary pattern cohorts increased
BC risk by 21% (p = 0.01); and (3)
Western/unhealthy dietary patterns
increased BC risk by 9% (p = 0.12)
20.3.1 Systematic Reviews and of follow-up) found that high adherence to a

Meta-analyses MedDiet lowered BC risk in postmenopausal
women [34]. The Netherlands cohort showed a
20.3.1.1 Breast Cancer Risk statistically significant inverse association
between MedDiet adherence and risk of estrogen
Mediterranean diet (MedDiet) receptor negative BC, with a 40% lower risk for
Recently the MedDiet pattern has been shown to high vs. low MedDiet adherence whereas there
be effective in reducing BC risk especially in was a modest inverse trend for estrogen receptor
postmenopausal women [34, 35, 40, 41]. A 2017 positive or total BC risk (when excluding or lim-
meta-analysis (7 cohort studies plus The iting alcohol beverage consumption). Also, the
Netherlands Cohort Study including 62,573 meta-analysis found a 6% lower total postmeno-
women; baseline age 55–69 years; mean 20 years pausal BC risk for high vs. low MedDiet
adherence and a 23–27% lower risk of hormone or mortality [36, 38]. A 2014 meta-analysis (2
receptor negative BC, which may have important RCTs and one large multi-center prospective
prevention implications because of the poorer cohort study with 9966 BC patients; 5–7.3 years
prognosis of receptor negative BC subtypes. A of follow-up) found that high adherence to a
2016 meta-analysis (28 cohort studies on total post-BC diagnostic healthy lower-fat dietary pat-
cancer risk; >2 million women; 4–40 years of tern (with emphasis on increased fruit, vegetables
follow-up; 13 observational and 3 cohort studies and fiber intake) significantly reduced risk of BC
for BC incidence; and 7152 subjects) found that recurrence by 23% and all-cause mortality by
higher adherence to the MedDiet reduced total 17% [38]. However, a meta-analysis (4 cohort
cancer mortality by 14% and lowered BC studies; 9819 women; mean baseline age
incidence by 4%. The Prevencion con Dieta
56 years; 7.7 years of follow-up) showed that
Mediterranea (PREDIMED) trial (7447 partici- postmenopausal women with high adherence to
pants/4282 postmenopausal women; MedDiet the Healthy Eating Indices did not significantly
supplemented with extra-virgin olive oil or tree lower BC mortality or increase survival. However,
nuts vs a low-fat control diet; 4.8 years of follow- the studies are limited and heterogenous and in
up) showed that MedDiets reduced common
need of more and larger studies to confirm [36].
chronic disease with aging including lowering
BC risk by 57%, cardiovascular disease by 29%
and type 2 diabetes by 30% compared to guid- 20.3.2 Prospective Cohort Studies
ance of low-fat dietary intake [35]. Two other
meta-analyses of observational studies also sup- Thirteen cohort studies provide additional insight
port a modest inverse relationship between the on the association between diet quality and spe-
MedDiet and overall BC risk [41, 42]. cific dietary patterns, and BC risk, recurrence,
and mortality (Table 20.4) [43–55].
Healthy vs Western Diets
High adherence to healthy dietary patterns rich in 20.3.2.1 Breast Cancer Risk
fruits, vegetables, fish and soy products protect
against BC risk and adherence to Western dietary Western vs Healthy Dietary Patterns
patterns tend to increase BC risk [37, 39, 41, 42]. Three prospective cohort studies show that
A 2010 meta-analysis (39 case-control and cohort Western diets increase BC risk and a range of
studies) found that healthy dietary patterns sig- healthy dietary patterns decrease BC risk [45, 52,
nificantly reduced overall BC risk by 11%, 54]. The 2013 California Teachers Study (91,779
whereas a high consumption of alcohol signifi- women; mean baseline age 50 years; 14 years of
cantly increased BC risk by 21% and a Western follow-up) found that a plant-based pattern was
diet increased BC risk by 9% (p = 0.12) [42]. associated with a significant reduced BC risk by
Similar findings were observed in a 2013 meta- 15% (highest vs lowest intake) and an even
analysis (11 cohort studies and 15 case-control greater significant 34% reduced BC risk for hor-
studies; 584,437 women) [41]. Two meta- mone receptor negative tumors [45]. In contrast,
analyses primarily of case-control studies in a high wine dietary pattern was associated with
Chinese women found that diets rich in fruits and significant increased risk of hormone receptor
vegetables (emphasizing higher fiber intake) and positive BC risk by 29%. The 2010 UK Women’s
soy products significantly lowered BC risk by Cohort Study (35,372 women; mean baseline age
approx. 30% [37, 39]. 52 years; mean 9 years of follow-up) showed in
postmenopausal women significantly lower BC
20.3.1.2 B reast Cancer Recurrence risk dietary with patterns rich in fish by 40% and
and All-Cause Mortality vegetarian diets by 15% compared to high red
Several meta-analyses of RCTs and cohort stud- meat dietary patterns (Fig. 20.5) [52]. A 2009
ies support the protective effects of healthy French EPIC cohort (65,374 women; mean base-
dietary patterns in lowering risk of BC recurrence line age 53 years; 9.7 years of follow-up; 2381
Table 20.4 Summaries of prospective cohort studies on dietary quality and dietary patterns and breast cancer (BC)
risk, recurrence and mortality
Penniecook-Sawyers et al. (2016). 50,404 women; mean baseline As compared with non-vegetarians, all
Evaluate the association between age 63 years; 7.8 years of vegetarians combined had a 3%
vegetarian diets and BC risk follow-up; 26,193 vegetarians significant lower BC risk. However,
(Adventist Health Study −2; US) (multivariate adjusted) vegans had a 22% lower BC risk (but still
[43] non-significant) compared with
non-vegetarians (p = 0.09)
Hirko et al. (2016). 100,643 women; mean baseline A higher DASH diet pattern score or
Examine associations between age 54 years; 22 years of adherence was significantly associated
dietary quality indices and breast follow-up; 2372 BC cases that with 56% reduced risk of HER2-type BC
cancer risk by molecular subtype could be classified by molecular (p-trend = 0.02), which was primarily
among women (Nurses’ Health Study subtype; dietary quality scores for associated with increased fruit intake.
(NHS) cohort; US) [44] the alternative Healthy Eating There were no significant association
Index (AHEI), alternate between the other healthy diets and BC
Mediterranean diet (aMED), and risk
Dietary Approaches to Stop
Hypertension (DASH) dietary
patterns (multivariate adjusted)
Link et al. (2013). 91,779 women; mean baseline High adherence to healthy plant-based
Evaluate dietary patterns and their age 50 years; 14 years of dietary patterns was associated with
relation to BC risk in a large cohort follow-up; 4140 BC cases lower BC risk overall by 15% (p-trend =
of women (California Teachers Study (multivariate adjusted) 0.003) and hormone receptor negative
cohort; US) [45] tumors by 34% (p-trend = 0.03).
Additionally, a dietary pattern with high
wine intake was associated with an
increased risk of hormone receptor
positive tumors by 29%
Couto et al. (2013). 44,840 women; baseline age A 2-point increase in MedDiet score was
Investigate whether adherence to a range 30–49 years; 16 years of not significantly associated with BC risk
MedDiet pattern influences BC risk follow-up on average; 1278 BC for either premenopausal or
(The Swedish Women’s Lifestyle and cases (multivariate adjusted) postmenopausal women
Health cohort) [46]
Buckland et al. (2013). 335,062 women; 11 years of High adherence to an alcohol restricted
Investigate the association between follow-up; 9009 postmenopausal MedDiet was associated with a lower BC
adherence to the alcohol restricted and 1216 premenopausal first risk for all women by 6% (p = 0.048) and
MedDiet and risk of BC (European primary invasive BC were for postmenopausal women by 7%
Prospective Investigation into Cancer identified (multivariate adjusted) (p = 0.037). For postmenopausal women
and Nutrition cohort study) [47] with hormone receptor negative tumors
BC risk was reduced by 20%. In
premenopausal women, the MedDiet
score was not associated with BC risk
Vrieling et al. (2013). 2522 postmenopausal BC High adherence to an unhealthy dietary
Examine the effect of pre-diagnostic patients; mean 5.5 years of pattern was associated with an increased
dietary patterns with mortality and follow-up; 316 deaths occurred, risk of non-BC mortality (highest vs
BC recurrence in postmenopausal BC 235 due to BC and 81 due to lowest) by 269% (p-trend <0.001) but
survivors (the MARIE study; non-BC causes (multivariate there were no significant associations
German) [48] adjusted) with BC-specific mortality and BC
Two major dietary patterns were recurrence. Adherence to the healthy
identified: healthy (high intakes of dietary pattern was inversely associated
vegetables, fruits, vegetable oil, with non-BC mortality by 26% (p-trend
and soups) and unhealthy (high = 0.02) and 29% reduced BC recurrence
intakes of red meat, processed (p-trend = 0.02) in stage I-IIIa patients
meat, and deep-frying fat)
Izano et al. (2013). 4103 women with invasive stage Highest vs lowest adherence reduced
Assess the effect of diet quality after I–III BC; median length of non-BC mortality for the DASH diet by
BC diagnosis on BC survival follow-up was 112 months and 28% (p-trend = 0.03) and AHEI-2010
(Nurses’ Health Study; US) [49] maximum length of follow-up diets by 43% (p-trend <0.0001). Diet
was 277 months; DASH diet scores were not significantly associated
score, and the Alternative Healthy with BC mortality
Eating Index (AHEI)-2010; 981
women died, 453 BC deaths,
38 BC recurrences and 528
non-BC related deaths
Kim et al. (2011). 2729 postmenopausal women In postmenopausal women with low
Assess the associations between diet invasive stage I-III BC; up to physical activity, the highest aMedDiet
quality and postmenopausal BC 26 years of follow-up; 4 dietary score was associated with lower non-BC
survival (Nurses’ Health Study; US) quality scores: Alternative mortality risk by 61% (p-trend = 0.0004).
[50] Healthy Eating Index, Diet No other associations were observed
Quality Index-Revised, between diet quality indices and either
Recommended Food Score, and total or non-BC mortality
the alternative MedDiet score
Fung et al. (2011). 86,621 women; baseline age A higher DASH diet score was
Examine the associations of the 30–55 years; diet quality scores associated with a 20% reduced risk of
DASH score and low-carbohydrate calculated from up to 7 food estrogen receptor negative BC (p-trend =
diets on the risk of postmenopausal frequency questionnaires; 0.02), which was largely explained by
BC (Nurses’ Health Study; US) [51] 26 years of follow-up; 5522 BC higher fruits and vegetables intake. Also,
cases (multivariate adjusted) a higher vegetable-based, low-
carbohydrate-diet score was associated
with 19% lower risk of estrogen receptor
negative BC (p- trend = 0.03)
Cade et al. (2010). 35,372 women; aged between 35 In postmenopausal women, compared to
Assess the relationship of 4 common and 69 years (mean baseline age high red meat diets, the higher fish diets
dietary patterns to the risk of BC (UK 52 years); dietary patterns based significantly reduced BC risk by 40%
Women’s Cohort Study) [52] on: fish; red meat intake; and and the vegetarian diet insignificantly
vegetarian intake; mean 9 years reduced risk by 15% (Fig. 20.5)
of follow-up (multivariate
adjusted)
Trichopoulou et al. (2010). 14,807 women; average of For all women, every 2 points increase in
Evaluate the effect of the traditional 9.8 years of follow-up; 240 BC traditional MedDiet score directionally
MedDiet in a Mediterranean country cases (multivariate adjusted) lowered BC risk by 12%. In a subgroup
on BC risk (EPIC; Greece) [53] analysis, postmenopausal women had a
significantly lower BC risk by 22% for
every 2 points increase in score but there
was no effect on BC risk in
premenopausal women (Fig. 20.6)
Cottet et al. (2009). 65,374 women; mean baseline The MedDiet pattern was associated with
Assess the association between age 53 years; 9.7 years of lower BC risk by 15% (p-trend = 0.003)
postmenopausal dietary patterns and follow-up; 2381 postmenopausal and a Western dietary pattern was
BC risk (EPIC -France) [54] BC cases (multivariate adjusted) associated with an increased BC risk by
20% (p-trend <0.007). Adherence to a
diet comprising mostly of fruits,
vegetables, fish, and olive or sunflower
oil, along with avoidance of Western-type
foods, may contribute to a substantial
reduction in postmenopausal BC risk.
(continued)
Kwan et al. (2009). 1901 women diagnosed with High adherence to the prudent diet was
Determine the association of dietary early-stage BC; mean baseline associated with a significantly lower risk
patterns on BC recurrence and age 59 years; 2 dietary patterns of non-BC mortality by 65% whereas
mortality of early stage BC survivors were identified: prudent (high high adherence to the Western diet
(Life After Cancer Epidemiology intakes of fruits, vegetables, significantly increased risk of non-BC
Study; US) [55] whole grains, and poultry) and mortality by 115% (Fig. 20.7). Neither
Western (high intakes of red and dietary pattern was associated with risk
processed meats and refined of BC recurrence or BC mortality. These
grains); 6 years of follow-up; findings were independent of physical
268 BC recurrences and 226 activity, overweight, or smoking
all-cause deaths (multivariate
adjusted)
postmenopausal BC cases) found that Western an alcohol restricted MedDiet significantly low-
dietary patterns significantly increased BC risk ered BC risk in all women by 6% and in post-
by 20% and a healthy MedDiet pattern signifi- menopausal women by 7% compared to low
cantly lowered BC risk by 15% [54]. adherence [47]. In postmenopausal hormone
receptor negative women, the adherence to an
Vegetarian Dietary Patterns alcohol restricted MedDiet significantly reduced
The only prospective study on vegetarian dietary BC risk by 20%. A 2010 Greek EPIC cohort
patterns and BC in women is from the Adventist (14,807 women; average of 9.8 years of follow-
Health Study-2, which has rates of BC that are up) showed a significantly lowered BC risk

>20% lower than usual because of very low among postmenopausal women by 22% for
rates of tobacco and alcohol consumption, a every 2 points increase in MedDiet score adher-
wide diversity of dietary habits and overall good ence compared to no effect on BC risk in pre-
health [43]. Vegetarian diets are classified into menopausal women per 2 point MedDiet score
four patterns (vegan, lacto-ovo-vegetarian, increase (Fig. 20.6) [53]. A 2009 French EPIC
pesco-vegetarian, and semi-vegetarian). The US cohort (65,374 women; mean baseline age
Adventist Health Study (50,404 women; mean 53 years; 9.7 years of follow-up; 2381 post-
baseline age 63 years; 7.8 years of follow-up; menopausal BC cases) found that for postmeno-
26,193 vegetarians) found that compared with pausal women the MedDiet significantly lowered
non-vegetarians, all vegetarians combined had BC risk by 15% and Western dietary patterns
an insignificantly 3% lower BC risk [43]. significantly increased BC risk by 20% [54].
However, vegans had a 22% lower BC risk com- Adherence to a diet comprising mostly of fruits,
pared with non-vegetarians (p = 0.09). vegetables, fish, and olive and sunflower oil,
along with minimizing or avoiding Western-type
Mediterranean Diet (MedDiet) foods, such as processed meats, French fries,
Of the four European, prospective cohort studies appetizers, cakes and pies may contribute to a
assessing the effect of adherence to MedDiets on substantial reduction in postmenopausal BC risk.
BC risk, three show that higher adherence sig- The 2013 Swedish Women’s Lifestyle and
nificantly reduced BC risk and one study found Health study (44,840 women; baseline age range
an insignificant effect on BC risk [46, 47, 53, 30–49 years; 16 years of follow-up on average)
54]. A 2013 European Prospective Investigation found, in this cohort of relatively young women,
into Cancer and Nutrition (EPIC) study (335,062 that adherence to the MedDiet pattern was not
women; 11 years of follow-up; 9009 postmeno- statistically significantly associated with reduced
pausal and 1216 premenopausal first primary risk of BC overall, especially when alcohol was
invasive BC cases) found that high adherence to excluded [46].
Red Meat Eater Poultry Eater Fish Eater Vegetarian

1.2
1
Breast Cancer Relative Risk
0.8
0.6
0.4
0.2
0
Postmenopausal Premenopausal
Fig. 20.5 Association between dietary pattern type and postmenopausal vs premenopausal breast cancer (BC) risk
from the UK Women’s Cohort Study (post-menopausal fish eater risk P < 0.05 after multivariate adjustment) (adapted
from [52])
0
All Women (p =.12) Premenopausal Postmenopausal
Women (p =.91) Women (p =.03)
% Breast Cancer Risk
-5
–10
–15
–20
–25
Fig. 20.6 Adherence to the Mediterranean diet (MedDiet) on breast cancer risk from the Greek European Prospective
Investigation into Cancer and Nutrition (EPIC) Study (adapted from [53])
Dietary Approaches to Stop Hypertension adherence to the DASH dietary pattern, which
(DASH) Diet was primarily associated with higher fruit
Two prospective studies observed that high intake [44]. Also, a 2011 Nurses’ Health Study
adherence to a DASH dietary pattern can reduce (86,621 women; baseline age 30–55 years;
the risk of BC [44, 51]. A 2016 Nurses’ Health 26 years of follow-up) observed that a higher
Study (100,643 women; 22 years of follow-up) DASH score significantly lowered estrogen
observed a significant 56% reduced risk of receptor negative multivariate BC risk by 20%
HER2-type BC among women with a high [51].
20.3.2.2 B reast Cancer Recurrence associated with lower BC mortality [49]. A 2011
and Morality Nurses’ Health Study (2729 postmenopausal
Four prospective studies provide insights on the women with stage I-III BC) observed that a
effects of dietary patterns on BC recurrence, and higher alternate MedDiet score was associated
BC and non-BC mortality [48–50, 55]. The 2013 with a 61% lower risk of non-BC mortality in
German MARIE study (2522 postmenopausal women with low physical activity [50]. Finally,
BC patients; mean 5.5 years of follow-up) found the 2009 Life After Cancer Epidemiology Study
that increasing consumption of an unhealthy (1901 women with early stage BC; mean baseline
(Western) dietary pattern was associated with a age 59 years; 6 years of follow-up) observed that
significant increased risk of non-BC mortality a healthy diet significantly reduced all-cause
(highest vs lowest) by 269% whereas no signifi- mortality by 43% and non-BC mortality by 65%,
cant associations with BC-specific mortality and whereas a Western diet significantly increased
BC recurrence were observed [48]. In contrast, all-cause mortality by 53% and non-BC mortality
high adherence to the healthy dietary pattern was by 115% (Fig. 20.7) [55]. Neither of these dietary
associated with a reduced non-BC mortality by patterns were significantly associated with BC
26% (highest vs lowest) and 29% reduced risk of recurrence or BC mortality risk.
BC recurrence in stage I-IIIa patients. A 2013
Nurses’ Health Study (4103 women with inva-
sive stage I–III breast cancer; median length of 20.3.3 Randomized Controlled Trials
follow-up was 112 months and maximum length (RCTs)
of follow-up was 277 months) found a significant
lower non-BC risk for a higher DASH score by Eleven RCTs on dietary patterns and BC risk,
28% and an Alternate Healthy Eating Index by recurrence and survival are summarized in
43% but neither of these dietary patterns were Table 20.5 [56–66].
Prudent Diet All-Cause (p-trend = .02) Prudent Diet Non-BC (p-trend =.003)
Western Diet All -Cause (p-trend =.05) Western Diet Non-BC (p-trend =.02)
2.5
2
Relative Risk for Mortality
1.5
0.5
0
1 2 3 4
Dietary Pattern Quartile Score
Fig. 20.7 Association between dietary pattern score and all-cause mortality and non-breast cancer (BC) mortality in
1901 postmenopausal women with early stage breast cancer (adapted from [55])
Table 20.5 Summaries of RCTs on dietary patterns and breast cancer (BC) risk recurrence and mortality
Rock et al. (2016). Parallel RCT: All groups had significant mean weight loss
Examine the effects of 245 non-diabetic, overweight/obese after 1 year; the lower fat diet by 9.2%, lower
diet composition on women; mean age 50 years; diets: carbohydrate by 6.5%, and walnut-rich diet
weight loss and BC risk lower fat (20% energy), higher by 8.2% (p < 0.0001). Subgroup analysis
biomarkers [56] carbohydrate (65% energy); a lower showed that insulin sensitive women lost
carbohydrate (45% energy), higher fat more weight on the lower fat vs. lower
(35% energy); or a walnut-rich (18% carbohydrate group (7.5 kg vs. 4.3 kg;
energy) higher fat (35% energy), lower p = 0.06), and in the walnut-rich vs. lower
carbohydrate (45% energy); 1 year carbohydrate group (8.1 kg vs. 4.3 kg;
p = 0.04). Sex hormone binding globulin,
associated with a reduced BC risk, increased
within each group except for the lower
carbohydrate group (p < 0.01). Also, hs-CRP
and IL-6 decreased at follow-up in all groups
(p < 0.01)
Neuhouser et al. (2015). Multi-center Parallel RCT: Postmenopausal women with overweight
Investigate effect of 67,142 postmenopausal women aged and obese BMIs had an increased BC risk
overweight and obese 50–79 years were enrolled from compared to normal weight BMI women.
BMI on BC risk in 1993–1998 with a median of 13 years Obesity grade 2 and 3 was associated with
postmenopausal women of follow-up; 40 U.S. clinical centers; higher risk for advanced disease including
(Women’s Health 3388 BC cases larger tumor size by 112%, positive lymph
Initiative (WHI); US) [57] nodes by 89%, and mortality after BC by
111%. Women with baseline BMI <25 who
gained >5% of body weight over the
follow-up period had an increased BC risk
by 36%
Toledo et al. (2015). Parallel RCT: The multivariable-adjusted BC risk reduction
Evaluate the effect of 2 4282 postmenopausal women; baseline for the MedDiet with extra-virgin olive oil was
interventions with MedDiet age 68 years; mean BMI 30; < 3% used 68% (p = 0.02) and 41% for the MedDiet with
vs the advice to follow a hormone therapy; median follow-up of nuts (p = 0.24) vs low-fat control. Also, for
low-fat diet (control) on BC 4.8 years; 35 malignant BC cases each additional 5% of energy from extra-virgin
incidence (PREDIMED olive oil BC risk was reduced by 28%
trial; Spain) [58]
Martin et al. (2011). Parallel RCT: The lower dietary fat group (20% energy from
Assess the effect of 4690 women with elevated fat) had a 19% increased BC risk in women
low-fat and high- mammographic density; mean baseline with elevated mammographic breast density
carbohydrate diets on BC age 47 years; most were Caucasian and compared to a control group (30% energy
incidence in women at premenopausal at baseline; from fat)
increased risk (Canadian intervention group received intensive
Diet and Breast Cancer dietary counseling to reduce fat and
Prevention Study) [59] increase carbohydrate to 65% of
calories; average study duration
10 years
(continued)
Pierce et al. (2009). Multi-center Parallel RCT: The greatest effect on lowering additional BC
Secondary analysis of the 896 peri and postmenopausal events occurred among women who were
baseline quartiles of fiber, early-stage BC survivors without hot already eating significant amounts of
fiber to fat ratio and flashes at baseline (1/3 of total trial vegetables, fruits, and fiber at baseline rather
vegetable-fruit intake population); 1 and 4 years; intervention than the degree of intervention dietary change
effects in the hot flash increased servings of vegetables and that was achieved. Hot flash negative BC
negative BC survivor fruit daily, higher fiber, and lower survivors with the highest of baseline quartiles
subgroup (Women’s energy from fat vs 5-a-day diet of fiber, fiber to fat ratio and vegetable-fruit
Healthy Eating and intake had significantly fewer BC recurrence
Living (WHEL) Study; events for 1 and 4 years. For example, after
US) [60] 1 year, the highest baseline fiber consumers
(mean 32 g fiber/day) had a 36% lower risk of
BC recurrence compared to those with lower
fiber intake (mean 12 g/day), which suggests
that higher fiber intake may help to attenuate
systemic circulating estrogen levels in this
peri- and postmenopausal population
(p = 0.02). After 4-years, there were fewer BC
events observed across quartiles of vegetable-
fruit and fiber consumption compared to the
control group (p = 0.01)
Gold et al. (2008). Multi-center Parallel RCT: Peri- and postmenopausal women without hot
Secondary evaluation of 2967 BC survivors; age 18–70 years; flashes at baseline assigned to the intervention
the effect of a low-fat diet 900 women (30%) in the hot flash- had 31% fewer BC events than those assigned
high in vegetables, fruit, negative group; 7.3-years duration; to the 5-a-day control group (p = 0.02). These
and fiber on the prognosis intervention increased intake of effects may be related to fiber’s attenuating
of BC survivors with hot vegetable and fruit servings per day (10 effects on circulating estrogen concentrations
flashes or without hot vs 6.5 servings/day), fiber (25.5 vs
flashes at baseline 19.4 g/day), and reduced percent
(WHEL Study; US) [61] energy from fat (26.9% vs 31.3%) vs
5-a-day diet
Pierce et al. (2007). Multi-center Parallel RCT: Women in the intervention group had an
Assess effects of increased 3088 women; mean baseline age insignificant lower BC event risk by 4% and
vegetable, fruit, and fiber 53 years; mean 7.3-year follow-up; the lower mortality risk 9% compared to the
intake above the 5-a-day intervention group achieved and control group
recommendations along maintained the following dietary
with reduced fat intake on changes vs. the 5-a-day fruit and
BC recurrence and vegetable group through 4 years:
all-cause mortality among higher servings of vegetables by 65%,
women with previously fruit by 25%, and fiber by 30%, and
treated early stage BC lower energy intake from fat by 13%
(WHEL; US) [62]
Prentice et al. (2006). Multi-center Parallel RCT: Among postmenopausal women, a low-fat,
Examine the effects of a 48,835 postmenopausal women, and higher fruit, vegetable, and whole-grains
healthy low-fat dietary baseline age 50–79 years, without prior dietary pattern insignificantly lowered BC risk
pattern on BC incidence breast cancer, including 19% minority by 9% compared with women in the control
(WHI; US) [63] race/ethnicity; 40 US clinical sites; group. Secondary analyses provided greater
intervention diet goals to reduce intake evidence of BC risk reduction among women
of total fat to 20% of energy and having a high-fat diet at baseline
increase intake of vegetables and fruit
to at least 5 servings daily and grains to
at least 6 servings daily; comparison
group participants were not asked to
make dietary changes; average duration
of 8.1 years
Chlebowski et al. (2006). Phase III Multicenter RCT: This dietary intervention rich in fruit and
Evaluate the effect of a 2437 postmenopausal women with vegetables and lower in dietary fat
healthier dietary resected, early-stage BC receiving significantly prolonged BC relapse free
intervention significantly conventional cancer management; survival compared with the control group by
lower in dietary fat and median 60-month follow-up; dietary 24% (p = 0.034). The dietary intervention
energy, and higher in total change: fat (33.3 vs 51.3 g/day), daily improved relapse-free survival in women with
fiber from fruits and energy (1460 vs 1531 kcal/day) and estrogen receptor negative BC by reducing
vegetables on prolonged total fiber (19.5 vs 17.3 g/day); weight mortality risk by 42%
relapse-free survival in loss 2.3 kg
women with resected BC
(The Women’s
Intervention Nutrition
Study [WINS]; US) [64]
Rock et al. (2004). Multi-center Parallel RCT: The higher total fiber intervention group had
Assess the effects of 291 women with BC history; mean age significantly lower circulatory estradiol
post-BC diagnosis dietary 55 years; mean BMI 27; dietary goals concentration after 1 year by 16 pmol/L vs the
modifications including for the intervention group were control group (p = 0.05). Change in total fiber
increased fruit, increased total fiber, vegetable, and (but not fat) intake was significantly and
vegetables, and fiber on fruit intakes and reduced fat intake; independently related to estradiol levels, which
estrogen levels (WHEL 1 year may play a role in lowering the risk of BC
study; US) [65] The intervention group had lower recurrence
intake of energy from fat (21% vs
28%), higher total fiber intake from
fruits and vegetables (29 g/d vs 22 g/
day) (p < 0.001), and mean weight loss
of 1 kg
Rock et al. (2001). Multi-center Parallel RCT: After 1 year, the general guidance group lost
Evaluate the effect of a 1010 women with BC history; mean slightly more weight by 0.42 kg compared to
low-fat diet and advice age 54 years; diet intervention was the low-fat intervention group. For the total
for increased fruits and performed by telephone counseling and group, body weight was stable in 74% of the
vegetables on weight promoting a low-fat diet that also was subjects with 11% losing weight and 15%
change in women after high in fiber, vegetables, and fruit gaining weight (which was similar for both
BC diagnosis (WHEL intake vs general dietary guidelines to groups). The low-fat diet intervention was not
Study; US) [66] reduce disease risk (control); weight at associated with significant weight loss in
baseline and 12 months mid-aged women at risk for BC recurrence
20.3.3.1 B
reast Cancer Risk, lymph nodes by 89%, and mortality after BC by
Recurrence and Survival 111%. Women with baseline BMI <25.0 who
gained >5% of body weight over the follow-up
Obesity period had an increased BC risk by 36%, but
Approximately two-thirds of US and Western among women already overweight or obese there
countries’ women are overweight or obese, placing was no association of weight change (gain or loss)
them at increased risk for postmenopausal BC. A with BC during follow-up. There was no modifica-
2015 analysis of the US Women’s Health Initiative tion effect of the BMI-BC relationship by post-
(WHI) Multicenter Trials (67,142 postmenopausal menopausal hormone therapy and the direction of
women aged 50–79 years; median of 13 years of association across BMI categories was similar for
follow-up; 40 U.S. clinical centers; 3388 BC cases) never, past and current hormone therapy use.
showed that overweight and obese postmenopausal
women had an increased BC risk vs normal weight Weight Loss
women [57]. Obesity grade 2 + 3 was also associ- Two RCTs demonstrate that both low-fat and
ated with significantly higher risk for advanced dis- higher fat dietary patterns can promote weight
ease including larger tumor size by 112%, positive loss, a critical factor in reducing chronic
inflammation and BC risk and progression [56, to a target of 15% of calories and increase carbo-
66]. In both pre- and postmenpausal women hydrate to 65% of calories; average of 10 years of
(245 non- diabetic, overweight/obese women; duration) demonstrated that the lower fat diet
mean age 50 years; 3 diets: lower fat diet; a increased BC risk by 19% in women with ele-
lower carbohydrate diet; or a walnut-rich diet; vated mammographic breast density [62]. The
1 year) found that all groups had significant 2007 WHEL Trial (3088 women; mean baseline
mean percent weight loss of 9.2% in the lower age 53 years; guidance for higher vegetables and
fat diet, 6.5% in the lower carbohydrate diet, and fruit, and fiber intake, and lower energy intake
8.2% in the walnut-rich diet at 12 months regard- from fat vs. the 5-a-day fruit and vegetable con-
less of dietary fat level [56]. C-reactive protein trol group without guidance for low fat; mean
and interleukin-6 decreased at follow-up in all 7.3-years of follow-up) found that women in the
groups (p < 0.01). A 2001 Women’s Healthy intervention group had an insignificant 4% lower
Eating and Living (WHEL) Trial (1010 women BC event risk (p = 0.63) and a 9% lower mortal-
after BC diagnosis; diet intervention was per- ity risk (p = 0.43) compared to the 5-a-day con-
formed by telephone counseling and promoted a trol group [62]. The US WHI multi-center trial
low-fat high in fiber, vegetables, and fruit diet vs (48,835 postmenopausal women; baseline age
general health dietary guidance; 12-months) 50–79 years; average 8.1 years of follow-up)
found that the general dietary guidance (without showed that women consuming a low-fat, healthy
a focus on low-fat) reduced baseline body weight fruit, vegetable, and whole-grains rich dietary
by 0.42 kg more than the healthy low-fat diet pattern insignificantly lowered BC risk by 9%
intervention [66]. compared with those in the control group [63].
Secondary analyses of this trial indicated that
editerranean Diet (MedDiet)
M women with high-fat diets at baseline had lower
A 2015 Spanish PREDIMED Trial (4282 post- BC risk. The 2006 Women’s Intervention
menopausal women; baseline age 68 years; Nutrition Study (WINS) (2437 women with BC;
mean BMI 30; < 3% used hormone therapy; 2 60 months) found that lower energy intake (com-
interventions on MedDiets supplemented with bination of low-dietary fat and higher fiber) and
either extra virgin olive oil or mixed nuts com- an average weight loss of 2.3 kg significantly
pared to advice to follow a low-fat diet (con- improved overall BC survival by 24% and estro-
trol); median follow-up of 4.8 years) showed gen receptor negative BC risk by 42% compared
that the MedDiet with extra virgin olive oil sig- to the control diet [64].
nificantly lowered BC risk by 68% and the
MedDiet with nuts lowered BC risk by 41% vs enopause Hot Flash Status
M
the low-fat control [58]. Also, each additional Three RCTs indicate that increased fiber intake
5% of energy from extra virgin olive oil reduced helps to protect against BC recurrence, especially
BC risk by 28%. in women without menopause associated hot
flashes [60, 61, 65]. A 2004 WHEL sub-trial (291
ealthy, Low Fat Diets
H women with a history of BC; mean age 55 years;
Four RCTs show that lower fat diets inconsis- reduced fat, increased fruits and vegetables plus
tently lower BC risk in postmenopausal women fiber; 1 year) found weight loss of 1 kg vs control
and may increase risk, especially in women with diet, and significant reduced estrogen concentra-
elevated mammographic breast density [59, 62– tions associated with higher fiber intake [65]. A
64]. The 2011 Canadian Diet and Breast Cancer 2009 secondary analysis of the WHEL Study
Prevention Study (4690 women with elevated (896 early stage BC survivors with no hot flashes;
mammographic density; mean baseline age 1 and 4 years) found that hot flash negative BC
47 years; most were Caucasian and premeno- survivors with the highest baseline intake quar-
pausal at baseline; intervention group received tiles of fiber, fiber to fat ratio, and vegetable and
intensive dietary counseling to reduce fat intake fruit intake had significantly fewer BC recurrence
events after 1 and 4 years [60]. For example, after The relationship between circulating estrogen
1 year, the highest baseline fiber consumer (mean levels and BC risk is complex involving BMI,
32 g fiber/day) had a 36% lower risk of BC recur- menopausal status, diet during adolescence and
rence compared to those with lower fiber intake estrogen-only hormone replacement therapy. The
(mean 12 g/day), which suggests that higher fiber US Women’s Health Trial found that weight loss,
intake may help to attenuate systemic circulating especially of abdominal adipose tissue, and
estrogen levels in peri - and postmenopausal healthier dietary patterns with higher fiber to
women. Similar findings were observed in energy intake ratios can lower systemic estrogen
another WHEL Study secondary evaluation and estrogen metabolites levels in postmeno-
(2967 BC survivors; 900 hot flash-negative pausal women to lower BC risk [18]. The Finnish
women or 30% of population; 7.3-years duration) Diabetes Prevention Trial (522 obese adults;
which found that peri- and postmenopausal mean baseline age 55 years and BMI 31; 2/3
women without hot flashes at baseline assigned women; 3 years) found that high fiber, lower fat
to the intervention had significantly fewer BC and lower energy density diets are significant
events by 31% than those assigned to the control predictors of weight loss (Fig. 20.9) [67]. Fiber
group [61]. intake is a major short-fall ‘nutrient’, especially
in highly developed countries such as the US
with high energy dense diets where <5% of the
20.4 Nutrients and populations consume adequate fiber with the
Phytochemicals mean fiber intake at only about half of the recom-
mended level [68, 69]. The systematic reviews
The nutritional and phytochemical compositions and meta-analyses and prospective cohort studies
of whole plant foods are summarized in on the effects of increased fiber intake on BC
Appendix B. The effects on BC risk with specific risk, recurrence and survival are summarized in
bioactive nutritional and phytochemical compo- Table 20.6 [70–83].
nents of whole plant foods is reviewed in this
section. Systematic Reviews and Meta- and Pooled
Analyses
Several meta-analyses of prospective studies
20.4.1 Dietary Nutrients and RCTs generally demonstrate that increased
fiber intake has a role in reducing BC risk [70–
The effects of dietary fiber, fat and vitamins on 73]. A 2016 meta-analysis (20 cohort and 4
BC risk, recurrence, and survival are summarized case-control studies; 3,662,421 participants;
in Table 20.6. 51,939 BC cases; 1–20-year follow-up) showed
that in postmenopausal women the highest
20.4.1.1 Dietary Fiber total fiber intake significantly reduced BC risk
There are a number of potential biological mech- by 12% compared to the lowest intake [70].
anisms supporting a role for fiber in the preven- Three meta-analyses of prospective studies
tion, recurrence and survival of BC, especially in show a clear inverse association between
postmenopausal women by attenuating systemic increased fiber intake and BC risk with 10 g of
estrogen levels, body weight and abdominal fat; total fiber significantly reducing BC risk by
and lowering insulin resistance, and circulating 4–7% [70, 72, 73]. For fiber-sub-types, the
C-reactive protein (CRP) (Fig. 20.8) [2, 3, 17]. In dose-response analyses per 10 g show a lower
postmenopausal women, adequate fiber intake BC risk for soluble fiber by 9%, for insoluble
has been hypothesized to protect against BC inci- fiber by 5%, and for fruit and cereal fiber by
dence by decreasing the rate of estrogen (pro- 4–5% [72]. The BC risk reducing effects were
duced from fat cells) recirculation by binding only significant in studies with higher total
estrogen and increasing its excretion in the feces. fiber intake increased by ≥13–15 g/day vs
Table 20.6 Summaries of dietary fiber, fat, protein, folate and antioxidant vitamin studies on breast cancer (BC) risk,
recurrence and mortality
Dietary fiber
Chen et al. (2016). 20 cohort and 4 case-control Highest vs lowest total fiber intake significantly
Update previous meta- studies up to March 2016; reduced BC risk by 12%, particularly in
analysis on the effectiveness 3,662,421 participants; 51,939 postmenopausal women (p = 0.027). Dose-response
of total fiber intake on BC cases; US, Canada, Europe, analysis found that every 10-g total fiber intake was
risk [70] China, and Malaysia; associated with a 4% reduction in BC risk (p < 0.002)
1–20-year follow-up
Xing et al. (2014). 2 RCTs: 5525 women wit h This meta-analysis of the pooled data from 2 large
Evaluate the pooled BC (1) Women’s Intervention RCTs and one large prospective study found that
association between diets Nutrition Study (WINS); 2437 post-diagnostic diets lower in fat, and higher in fruits
lower in fat, and higher in women; median 60 months of and vegetables and fiber reduced the risk of BC
fruits, vegetables, and fiber duration and (2) Women’s recurrence by 23% (p = 0.009) and BC related
and breast cancer (BC) Healthy Eating and Living mortality by 17% (p = 0.05)
recurrence and survival by (WHEL) study; 3088 women;
meta-analysis [71] mean 7.3 years; plus,
Collaborative Women’s
Longevity Study, a large
multi-center prospective
cohort study with 4441 BC
patients for 5.5-year follow-up
Aune et al. (2012). 16 cohort studies, 500,000– This analysis found an overall 7% lower BC risk for
Assess the pooled 1,000,000 participants; higher total fiber intake but significant risk reduction
association between total 15,000–26,000 BC cases up to was only observed among studies with an increased
fiber and specific fiber August 2011 intake of fiber by 15.8 g or mean total fiber intake of
source intake and BC risk 29.4 g/day. Dose-response analyses showed per 10 g
by systematic review and lowered BC risk for total fiber by 5%, soluble fiber
dose-response meta- by 9%, insoluble fiber by 5%, fruit fiber by 5%, and
analysis [72] cereal fiber by 4% but only 1% for vegetable fiber
Dong et al. (2011). 10 cohort studies through For all women, increasing total fiber intake by15 g/
Examine the association January 2011; 712,195 day significant lowered BC risk by 11%. Dose-
between total fiber intake participants; 16,848 BC cases; response analysis found a significant 7% reduction in
and risk of BC by North America, Europe, and BC risk for every 10 g/day increment of fiber intake
conducting a meta-analysis China; studies with both (p-trend = 0.004) based on 6 of the studies with no
of prospective cohort premenopausal women, evidence of heterogeneity
studies [73] postmenopausal women, both;
follow-up period ranged from
4.3–18 years, with a median of
8 years
Prospective Studies
Farvid et al. (2016). 90,534 women; mean baseline Higher fiber intake during adolescence and early
Evaluate fiber intake during age 36.4 years; 20 year- adulthood may be especially important in reducing
adolescence and early follow-up; 2833women were BC risk (Fig. 20.10). Higher total fiber intake in
adulthood in relation to BC diagnosed with BC; 44,263 of adolescence and early adult life reduced BC risk by
risk (The Nurses’ Health these women had data on 25% (p-trend = 0.004). Higher total fiber intake
Study II; US) [74] adolescent fiber intake; 1118 during adolescence was associated with a 16% lower
women were diagnosed with BC risk (p-trend = 0.04). Among all women, higher
BC (multivariate adjusted) total fiber intake in early adulthood significantly
lowered BC risk by 19% (p-trend = 0.002); higher
intakes of soluble fiber lowered risk by14% (p-trend
= 0.02) and insoluble fiber reduced risk by 20%
(p-trend <0.001)
Chhim et al. (2015). 3771 women; completed at The combination of high alcohol consumption and
Investigate whether fiber least 6 valid 24-h dietary low fiber intake acted synergistically to increase
intake modulates the records during the first 2 years hormone dependent BC risk and the effects on BC
association between alcohol of follow-up; median 12 years risk when stratified by median intake is summarized
intake and hormone of follow-up; 158 BC cases in Fig. 20.11
dependent BC risk confirmed (multivariate
(Supplémentation en adjusted)
Vitamines et Minéraux
Antioxydants study; France)
[75]
Ferrari et al. (2013). 334,849 women; mean Diets rich in total fiber and, particularly, vegetable
Investigate associations baseline age approx. 50 years fiber modestly and significantly reduced BC risk,
between total fiber and its (range 35–70 years); tumor independently of menopausal status. Total fiber was
main food sources subtypes: hormone receptor inversely associated with BC risk by 7% for each
(vegetables, fruit, cereals, positive and negative tumors; 10-g total fiber intake. In premenopausal women,
and legumes) and BC risk median follow-up 11.5 years; there was a significant 34% lower risk (p-trend =
(The European Prospective 11,576 invasive BC cases 0.02) for those consuming both high-fiber (>26 g/
Investigation into Cancer (multivariate adjusted) day) and low-fat (< 63 g/day) diets compared with
and Nutrition [EPIC]) [76] high-fat (>89 g/day) and low-fiber (< 18 g/day) diets
but this interaction between fiber and fat intake was
not significant in post-menopausal women. For
vegetable fiber (excluding potatoes, legumes, soy and
tomato products), BC risk was reduced by 15% for
each 5 g/day intake (p-trend = 0.01), independent of
menopausal status or hormone receptor status
Park et al. (2009). 185,598 postmenpausal Fiber was associated with preventing BC through
Examine the relation of women; mean age: 62 years; non-estrogen pathways among post-menopausal
fiber intake to BC risk by average of 7-year follow-up; women. Higher total fiber intake was associated with
hormone receptor status and 5461 BC cases were a 13% lower BC risk (p-trend = 0.02). The
histologic type among identified, of which 3341 association was stronger for hormone negative
post-menopausal women cases had hormone dependent tumors with lower risk by 44% (p-trend = 0.008) than
(National Institutes of BC (multivariate adjusted) for hormone positive tumors with a lower risk by 5%
Health-AARP Diet and (p-trend = 0.47). Fruit was the most effective fiber
Health Study; US) [77] source in reducing BC risk. Soluble fiber intake was
inversely associated with BC by 17% (p-trend =
0.02). Total fiber effect on lowering the risk of BC
was independent of the level of dietary fat intake
(p = 0.08)
Cade et al. (2007). 35,792 women (17,781 In older premenopausal women, higher total fiber
Evaluate associations post-menopausal women and intake significantly lowered BC risk by 52% (p-trend
between total fiber and fiber 15,951 pre-menopausal = 0.01; ≥ 30 g vs < 20 g/day). Also, higher cereal
source on BC risk in a women at baseline; mean fiber reduced BC risk by 41% (p-trend = 0.05; ≥
cohort including large baseline age 52 years (mean 13 g/day vs < 4 g/day) and fruit fiber had a borderline
numbers of vegetarians 45 years premenopausal and inverse BC risk lowering effect by 19% (p-trend =
(The UK Women’s Cohort 59 years menopausal); 18% 0.09; ≥ 6 g/day vs < 2 g/day). No significant BC
Study) [78] vegetarian; 10-year follow-up; lowering effects were seen for post-menopausal
cases of invasive BC (350 women
post-menopausal and 257
pre-menopausal) (multivariate
adjusted)
(continued)
Mattisson et al. (2004). 11,726 postmenopausal This study supports the hypothesis that dietary
Investigate the associations women; 342 BC cases; patterns high in fiber and low/moderate in fat are
between intakes of plant 11-year follow-up associated with lower risk of post-menopausal
foods, fiber and dietary fat (multivariate adjusted) BC. High fiber intakes were associated with a lower
and BC risk (The Malmo¨ risk of postmenopausal BC by 42% (25.9 vs 12.5 g/
Diet and Cancer cohort; day; p-trend = 0.056; multivariate adjusted). A
Sweden) [79] significant interaction (p = 0.049) was found between
fiber- and fat-tertiles with higher fiber and lower-fat
dietary patterns having the optimal impact for
reducing post-menopausal BC risk
Holmes et al. (2004). 88,678 participants; mean age This study found no overall significant association
Evaluate the association of 56.5 years at assessment; 68% between the midlife intake of total fiber (25 g/day
total fiber, fiber fractions, post-menopausal and 38% on vs 12 g total fiber/day (mean intake 18.1 g/day) on
carbohydrate, glycemic hormone replacement; 18-year BC risk. However, these findings do not exclude the
index, and glycemic load follow-up; 4092 BC cases possibility that diets including a very high intake of
with the risk of BC (Nurses’ (multivariate adjusted) fiber (>30 g/day) may reduce BC risk. In
Health Study; US) [80] postmenopausal women, there was a positive
association between glycemic index and BC risk by
15% (p-trend = 0.02) with a stronger association
among women whose BMI was <25 with doubling
of risk to 28% (p-trend = 0.003). For specific fiber
types, higher fruit fiber had a moderate trend
toward reduced BC risk by 8% (p = 0.08)
Terry et al. (2002). 89,835 women ages Total fiber (26 g/day vs 15 g/day) and specific fiber
Examine the association 40–59 years; 16.2-year fractions or types appear to be weakly associated with
between total dietary fiber follow-up; 2536 BC cases reduced BC risk. Higher total fiber intake insignificantly
and dietary fiber fractions, were diagnosed; self- lowered BC risk by 8% (p = 0.16). There were similar
and BC risk (Canadian completed questionnaire insignificant risk reductions for intakes of specific fiber
National Breast Screening regarding diet and physical fractions, including soluble and insoluble fiber, fiber
Study) [81] activity (multivariate adjusted) from cereals, fruit, and vegetables
Willett et al. (1992). 89,494 women; 34–59 years Total fiber intake (≥ 22 g/day vs ≤ 11 g/day; mean
Evaluate the hypothesis that of age; 8-year followed up; intake 17 g/day) did not protect against BC risk in
dietary fat increases and 1439 incident cases of BC, pre- or postmenopausal women (p = 0.62). However,
fiber decreases BC risk including 774 among this study cannot exclude the possibility that higher
(Nurses’ Health Study; US) postmenopausal women total fiber intake or that some specific fraction may
[82] (multivariate adjusted) lower BC risk. Also, this study did not find an
adverse effect of high levels of dietary fat on BC risk
Dietary fat and fat subtypes
Brennan et al. (2017). 15 prospective cohort studies There was no difference in BC specific mortality or
Clarify the association all-cause mortality for women in the highest vs the
between dietary total fat and lowest category of total fat intake. However, BC
saturated fat on BC specific mortality was 51% higher for women in the
mortality [83] highest vs the lowest intake of saturated fat
Cao et al. (2016). For total dietary fat: 24 cohort Total dietary fat (highest vs. lowest category) was
Evaluate the effect of total studies; 1,387,366 subjects associated with a 10% increase in BC risk, which was
dietary fat and serum fatty and 38,262 BC cases; for insignificant after adjusting for traditional BC risk
acids on BC risk [84] individual fatty acids: 7 cohort factors. No significant association was observed
study articles; 3511 subjects; between animal fat, vegetable fat, saturated fatty
1334 BC cases acids, monounsaturated fatty acids, polyunsaturated
fatty acids, n-3 PUFA, n-6 PUFA, eicosapentaenoic
acid, docosahexaenoic acid, alpha-linolenic acid, oleic
acid, linoleic acid and arachidonic acid and BC risk
Xia et al. (2015). 52 studies (24 cohort studies High intake of saturated fat increased BC risk by
Determine quantitative and 28 case-control studies); > 18% in case-control studies and 4% in cohort studies
relationship between dietary 50,000 women diagnosed with vs lower intake. A subgroup analysis of case-control
saturated fat intake and BC BC studies found significant positive associations
risk [85] between higher saturated fat intake and BC risk in
Asians by 17%, Caucasians by 19%, and
postmenopausal women by 33%
Xin et al. (2015). 5 prospective cohort studies Higher consumption of total vegetable oils reduced
Investigate the association and 11 retrospective case- BC risk by 12%, and dose-response analyses showed
of high vegetable oils control studies; > 150,000 that each 10-g vegetable oil/day insignificantly
consumption and breast women; 11,161 BC events lowered BC risk by 2%. However, higher olive oil
cancer risk, and evaluate the intake significantly reduced BC risk by 24%
dose–response relationship
[86]
Yang et al. (2014). 6 prospective nested case- Women with higher dietary ratio of n-3/n-6 PUFAs
Ascertain the relationship control and 5 cohort studies; had a significantly lower risk of BC by 10%. One
between n-3/n-6 ratio and 274,135 women; 8331 BC USA study with higher ratio of n-3/n-6 in serum
BC risk [87] events; across different phospholipids showed significantly lowered BC risk
countries by 27% (p-trend = 0.004)
Farvid et al. (2014). 88,804 women; baseline age Higher total fat intake was not associated with BC
Examine the association 26–45 years; 20 years of risk overall except for a positive association observed
between fat intake and BC follow-up; 2830 BC cases between animal fat intake and BC risk by 18%
incidence (Nurses’ Health (multivariate adjusted) (p-trend = 0.01). This positive association with
Study II; US) [88] animal fat intake was seen among premenopausal
women, but not among postmenopausal women.
Also, any associations between BC risk and saturated
fat, monounsaturated fat and animal fat, were
attenuated and non-significant after adjustment for
red meat intake. Other types of fat were not
significantly associated with BC risk
Boeke et al. (2014). 88,804 women; baseline age Higher total fat intake was associated with a slightly
Evaluate intakes of total fat, 26–45 years; 20 years of lower BC mortality risk by 15% (p-trend = 0.05).
specific types of fat, and follow-up; 9979 invasive Specific types of fat were generally not associated
cholesterol prior to breast cancer cases developed, with BC mortality risk
diagnosis in relation to of which 1529 went on to
lethal BC risk (Nurses’ become lethal (multivariate
Health Study II; US) [89] adjusted)
Sieri et al. (2008). 319,826 women; baseline age In all women, high saturated fat intake increased BC
Investigate the association 20–70 years; mean of risk by 13% vs lowest intake (p-trend = 0.038). Also,
between fat consumption 8.8 years of follow-up; 7119 in postmenopausal women, high saturated fat intake
and BC risk (European women developed BC increased BC risk by 21% for nonusers of hormone
Prospective Investigation (multivariate adjusted) therapy (p-trend = 0.044)
into Cancer and Nutrition
[EPIC]) [90]
Chajes et al. (2008). 19,934 women; mean baseline High intake of industrial type trans-fat or serum trans
Assess the effect of trans vs age 57 years; 7 years of fatty acids increased postmenopausal BC risk by 75%
oleic monounsaturated fat follow-up; 363 BC cases (p-trend = 0.018). cis-Monounsaturated fatty acids
intake and BC risk (EPIC (multivariate adjusted) were unrelated to BC risk
-France) [91]
(continued)
Holmes et al. (1999). 88,795 women; baseline age Compared to women consuming 30.1–35% energy
Determine whether intake 30–55 years; 14 years of from fat, those consuming ≤20% of energy from fat
of fat and fatty acids are follow-up; 2956 BC cases had a 15% higher BC risk (p-trend = 0.03). A
associated with BC risk (multivariate adjusted) dose-response assessment found that a 5% increase in
(Nurses’ Health Study; US) total fat energy or specific fatty acid were not
[92] significantly associated with BC risk except for
trans-fat (from industrial processing) with each 1%
energy intake increasing BC risk by 9% in post-
menopausal women. Also, each 5% total fat
replacement of carbohydrate reduced BC risk by 4%
Protein source
Wu et al. (2016). 46 prospective cohort studies Higher total red meat, fresh red meat, and processed
Conduct a meta-analysis to meat intake may be risk factors for breast cancer,
assess the association whereas higher soy food and skim milk intake may
between protein source and reduce BC risk (Fig. 20.12). In dose-response
BC risk [93] analysis BC risk decreased by 9% for soy food, by
4% for skim milk, and 10% for yogurt, and increased
by 9% for processed meat and by 7% for total red
meat
Farvid et al. (2014). 88,803 women; 20 years of Higher intake of total red meat was associated with
Investigate the association follow-up; 2830 BC cases an increased overall BC risk by 22% (p-trend = 0.01).
between dietary protein (multivariate adjusted) However, higher intakes of poultry, fish, eggs,
sources in early adulthood legumes, and nuts were not related to BC risk overall.
and BC risk (Nurses’ Health When the association was evaluated by menopausal
Study II; US) [94] status, higher intake of poultry was associated with a
lower BC risk in postmenopausal women by 27%
(p-trend = 0.02), but not in premenopausal women
with a 7% lower risk (p-trend = 0.60). In estimating
the effects of exchanging different protein sources,
substituting one serving/day of legumes for a daily
serving of red meat lowered BC risk by 15% among
all women and 19% lower risk among premenopausal
women. Also, substituting one serving/day of poultry
for one serving/day of red meat was associated with a
17% lower BC risk and a 24% lower postmenopausal
BC risk. Also, substituting one serving/day of
combined legumes, nuts, poultry, and fish for one
serving/day of red meat was associated with a 14%
lower risk of BC risk overall
Dietary Folate intake
Zhang et al. (2014). 14 prospective cohort studies; Dose-response analysis showed a nonlinear
Assess the dose response 677,858 women relationship as a potential J-shaped correlation
association between folate between folate intake and BC risk (p = 0.007). The
and BC risk [95] analysis revealed that a daily folate intake of
200–320 mg was associated with a lower breast
cancer risk, but BC risk increased significantly with a
daily folate intake >400 mg
Dietary antioxidant vitamins
Hu et al. (2015). 40 studies including 28 There was a significant increased BC risk for women
Meta-analysis and regression case-control studies and 12 with the lowest plasma α-tocopherol 5.74–
to assess the association nested case–control studies 9.16 μmol/L and a significantly lower BC risk for
between plasma retinol, higher plasma vitamin C levels only in case-control
vitamin A, C and α-tocopherol studies. However, there was no significant association
levels, and BC risk [96] between plasma retinol and BC risk
Hu et al. (2011). 51 studies (3 RCTs, 9 cohort Higher total vitamin A and retinol intake could
Meta-analysis and studies; 2 nested case-control significantly lower BC risk by 17%. Although the
regression evaluation of the studies, and 36 case-control dietary vitamin E, and total vitamin E intake reduced
association between intake studies) BC risk significantly when data from all studies were
of retinol, vitamins A, C pooled, the results became insignificant when data
and E dietary intake, and from cohort studies were analyzed. No significant
BC risk and estimate their dose-response relationship was observed for the
dose-response effects [97] higher intake of these vitamins and reduced BC risk
Vitamin D Intake and Blood 25-hydroxyvitamin D [25(OH)D]
Kim and Je et al. (2014). 30 prospective studies Overall there was an insignificant, weak inverse
Assess the association association between vitamin D intake or blood
between vitamin D intake 25(OH)D levels and BC risk. However, among BC
and blood 25(OH)D on BC patients high blood 25(OH)D levels were
risk or mortality [98] significantly associated with a 42% lower risk of BC
mortality. There was no significant heterogeneity
among the studies
Calcium
Meta-analyses
Hidayat et al. (2016). 11 cohort studies; 872,895 Higher calcium intake reduced overall BC risk by 8%
Investigate the association participants; 26,606 BC cases with moderate heterogeneity (p = 0.026). In the
of serum calcium and BC subgroup analysis, premenopausal BC risk was
risk [99] reduced by 25% compared to 6% for postmenopausal
BC. Dose-response analysis revealed that each
300 mg/day increase in calcium intake was associated
with reduced BC risk for premenopausal women by
8% and postmenopausal BC by 2%
Healthy (Fiber-Rich/Low Energy Dense) Diet and Physical Activity
Body Fat Mass Skeletal Muscle
Lower Levels of Insulin, Insulin-Like Growth Factor-1, Leptin, Systemic hs-CRP,

Breast Tissue Inflammation, Aromatase, and Estrogen
Tumor Progression and Risk of Recurrence
Fig. 20.8 Potential mechanisms by which high adherence to a healthy fiber-rich dietary pattern and appropriate levels
physical activity routines on lowering breast cancer risk and recurrence (adapted from [17])
control diets or total fiber intake ≥25 g/day Women’s Longevity Study, 4441 BC patients,
[72, 73]. A pooled analysis (5525 women in 5.5-year follow-up) found that post-diagnostic
the Women’s Intervention Nutrition Study, diets lower in fat, and higher in fruits and veg-
median 60 months of duration; the Women’s etables and fiber significantly reduced the risk
Healthy Eating and Living study, mean of BC recurrence by 23% and BC-related mor-
7.3 years duration; and the Collaborative tality by 17% [71].
0
Low-fat/high fiber Low-fat/low fiber High fat/high fiber High fat/low fiber
(lowest energy (highest energy
-0.5 density) density)
-1
Weight Loss (kg)
-1.5
-2
-2.5
-3
-3.5
Fig. 20.9 Effect of dietary fiber and fat intake on energy density and weight loss in overweight and obese adults over
3 years (adjusted for age, sex, baseline weight, baseline fiber and fat intake, and baseline and follow-up physical activity)
(adapted from [67])
Prospective Cohort Studies Antioxydants study (3771 women; completed at

Nine prospective studies summarize the effect of least 6 valid 24-h dietary records during the first
increased dietary fiber intake on BC risk [74–82]. 2 years of follow-up; median 12-year follow-up)
Although the pre-2005 cohort studies generally showed that increased fiber intake is protective
observed a weak or insignificant association against the increased BC risk associated with
between fiber and BC risk, they did not preclude high alcohol intake (Fig. 20.11) [75]. In stratified
the possibility that fiber or certain fiber compo- analyses, the combination of low fiber intake and
nents, depending on a women’s age, menopausal high alcohol intake was directly associated with
or hormone receptor status and/or BMI level, hormone-dependent BC risk increase by 1.5-fold
may affect BC risk [79–82]. Cohort studies since but not among women with higher fiber intake. A
2005 generally report an inverse association 2013 EPIC (334,849 women; mean baseline age
between adequate fiber intake and BC risk [74– 50 years (range 35–70 years); median follow-up
78]. These studies have also uncovered important 11.5 years) found that diets rich in total fiber and
interactions between adequate fiber intake during particularly vegetable fiber modestly and signifi-
adolescence and pre- vs postmenopausal status, cantly reduced BC risk, independently of meno-
estrogen receptor affinity, alcohol intake or gly- pausal status [76]. Total fiber was inversely
cemic load and type of fiber-rich food consumed. associated with BC risk by 7% for each 10-g total
A 2016 Nurses’ Health Study II prospective anal- fiber intake. Vegetable fiber significantly lowered
ysis (90,534 women; mean baseline age 36 years; BC risk by 15% for each 5 g/day intake, exclud-
20 year-follow-up; 2833 women were diagnosed ing potatoes, legumes, soy and tomato products,
with BC; 44,263 of these women had data on independent of menopausal status. A 2009
adolescent fiber intake) found that higher fiber National Institutes of Health-AARP Diet and
intake during adolescence and early adulthood Health Study (185,598 postmenpausal women;
may be especially important in reducing BC risk mean baseline age 62 years; average of 7-year
(Fig. 20.10) [74]. Higher total fiber intake in ado- follow-up) showed that fiber can play a role in
lescence and early adult life significantly reduced preventing BC through non-estrogen pathways
BC risk by 25%. A 2015 analysis of the French among post-menopausal women [77]. Higher
Supplémentation en Vitamines et Minéraux total fiber intake was inversely associated with a
Adolescent (p-trend =.04) Early Adult (p-trend =.008)

1.05
Relative Risk for Premenopausal BC Risk

1
0.95
0.9
0.85
0.8
0.75
0.7
12.4 to 15.1 15.3 to 18.0 17.7 to 20.3 20.2 to 22.8 24.7 to 27.4
Median Fiber (g/day)
Fig. 20.10 Associations between adolescent and early adulthood fiber intake and breast cancer (BC) risk premeno-
pausal adult women from Nurses’ Health Study II (multivariate adjusted) (adapted from [74])
Dietary Fiber Intake < Median Dietary Fiber Intake > Median
3
Hazard Ratio for Hormone Dependent BC Risk
2.5
1.5
0.5
0
1 2 3
Tertiles of Alcohol Intake
Fig. 20.11 Associations between tertiles of alcohol intake and hormone dependent breast cancer (BC) risk stratified by
median dietary fiber intake (p-interaction = 0.01) (adapted from [75])
13% lower BC risk. The association was stronger most effective fiber source in reducing BC risk;
for hormone negative tumors with lower risk by (2) soluble fiber was inversely associated with
44% (p-trend = 0.008) than for hormone positive BC by 17%; and (3) total fiber effects on lower-
tumors with a lower risk by 5% (p-trend = 0.47). ing the risk of BC was independent of the level of
Subgroup analysis found that: (1) fruit was the dietary fat intake.
20.4.1.2 Dietary Fat [86]. A 2014 analysis (6 prospective nested case-

Dietary fat intake was long hypothesized to be control and 5 cohort studies; 274,135 women)
associated with higher rates of BC in affluent showed that women with a higher dietary ratio of
countries, based primarily on strong international n-3/n-6 PUFAs have a significantly lower risk of
correlations of BC incidence and fat intake, espe- BC by 10% and a subset analysis of USA women
cially in animal studies [10]. The systematic found that a higher ratio of n-3/n-6 in serum phos-
reviews and meta-analyses and prospective pholipids significantly lowered BC risk 27% [87].
cohort studies on total dietary fat and fat sub-
types on BC risk, recurrence and mortality are Prospective Cohort Studies
summarized Table 20.6 [83–92]. Five prospective cohort studies provide represen-
tative insights on the effects of total fat and fatty
Systematic Reviews and Meta-Analyses acid sub-types intake on BC risk [88–92]. A 2014
Five systematic reviews and/or meta-analyses pri- Nurses’ Health Study II (88,804 women; baseline
marily of cohort studies provide important age 26–45 years; 20 years of follow-up) found
insights on the effects of total dietary fat and spe- that higher total fat intake was not associated
cific fat sub-types on BC risk, recurrence and with BC risk overall except for a significant posi-
mortality [83–87]. These analyses indicated that tive association observed between animal fat
total dietary fat intake is less important than the intake and BC risk overall by 18% [88]. This
type of fat consumed, with lower intake of satu- positive association with animal fat intake was
rated fat and higher n-3 to n-6 ratio or higher seen among premenopausal women, but not
marine n-3 PUFA being the most effective dietary among postmenopausal women. Also, any asso-
fat considerations for protecting against BC risk ciations between saturated fat and animal fat,
or mortality. A 2017 analysis (15 prospective were attenuated and non-significant after adjust-
studies) concluded that total fat intake was not ment for red meat intake. A 2008 analysis of the
significantly associated with all-cause or BC spe- EPIC cohort (319,826 women; baseline age
cific mortality whereas higher saturated fat intake 20–70 years; mean of 8.8 years of follow-up)
was associated with a 51% higher risk of BC spe- found that in women high saturated fat intake was
cific mortality, especially in postmenopausal associated with a significant increase in BC risk
women [83]. A 2016 analysis (24 prospective by 13%, whereas no significant association with
studies; 1,387,366 subjects and 38,262 BC cases) total, monounsaturated, or polyunsaturated fat
found that neither higher total fat intake nor any was observed [90]. In this study, higher saturated
specific fatty acid types were statistically signifi- fat intake by postmenopausal women increased
cantly associated with BC risk, after multivariate BC risk by 21% for nonusers of hormone therapy.
adjustments [84]. A 2015 analysis (24 cohort Also, a 2008 French EPIC analysis (19,934
studies and 28 case-control studies) concluded women; mean baseline age 57 years; 7 years of
that high saturated fat intake increased BC risk by follow-up) found that increased intake of the
18% in case-control studies and 4% in cohort trans-
monounsaturated fatty acids (palmitoleic
studies [85]. In the case-control studies, there acid and elaidic acid found in industrially pro-
were significant positive associations between cessed foods) significantly increased BC risk in
higher saturated fat intake and BC risk in post- postmenopausal women by 75% [91]. A 1999
menopausal women by 33%, whereas there was Nurses’ Health Study (88,795 women; baseline
no association between higher dietary saturated age 30–55 years; 14 years) showed that a 5%
fat intake and BC risk among premenopausal increase in total fat energy or specific fatty acid
women in any studies. A 2015 analysis (5 pro- was not significantly associated with BC risk
spective and 11 case-control studies) found that except for trans-fat (from industrial processing)
higher vegetable oil intake reduced BC risk by which with each 1% energy intake increased BC
12% and higher olive oil appears to have a BC risk by 9% in post-menopausal women [92]. A
protective effect with a reduced BC risk by 24% 5% increase in total fat as a substitute for
carbohydrate reduced BC risk by 4%. There is lower BC risk among all women and a 19%
little support for lower total fat intake or specific lower risk among premenopausal women. Also,
fatty acids being associated with BC risk, expect substituting one serving/day of combined
possibly for industrial trans or saturated fat. legumes, nuts, poultry, and fish for one serving/
day of red meat was associated with a 14%
20.4.1.3 Dietary Protein Source lower risk of BC risk overall.
Protein is important in the maintenance of
human tissues and the regulation of various 20.4.1.4 Dietary Folate Intake
physiological functions, which may influence Previous observational studies regarding the
BC risk [93, 94]. A 2016 dose response meta- existence of an association between folate intake
analysis (46 prospective cohort studies) found and BC risk have been inconsistent [95]. A 2014
each serving of various protein sources can sig- dose-response meta-analysis (14 prospective
nificantly effect BC risk (Fig. 20.12) [93]. A cohort studies; 677,858 women) showed a non-
2014 Nurses’ Health Study II analysis (88,803 linear relationship between folate intake and the
women; 20 years of follow-up) found that higher risk of breast cancer, and discovered a potential
intake of total red meat was associated with a J-shaped correlation between folate intake and
significant increased overall BC risk by 22% BC risk [96]. This analysis also revealed that a
[94]. However, higher intakes of poultry, fish, daily folate intake of 200–320 mg was associated
eggs, legumes, and nuts were not related to BC with a lower breast cancer risk, but BC risk
risk overall. When the association was evaluated increased significantly with a daily folate intake
by menopausal status, higher intake of poultry >400 mg [95].
was associated with a significantly lower BC
risk in postmenopausal women by 27%, but not 20.4.1.5 Antioxidant Vitamins
in premenopausal women with a 7% lower risk Retinol, vitamins A, C and E are hypothesized
(p-trend = 0.60). In estimating the effects of to reduce BC risk due to their roles in the regu-
exchanging different protein sources, substitut- lation of cell growth, differentiation and apopto-
ing one serving/day of legumes for one serving/ sis (retinol, vitamin A) and enhancing immune
day of red meat was associated with a 15% function, anti-inflammation and antioxidant
Yogurt
Skim Milk
Per Protein Source Serving
Whole Milk
Nuts
Soy Food
Egg
Poultry
Processed Meat
Total Red Meat
–15 –10 –5 0 5 10 15
BC Risk (%)
Fig. 20.12 Association between protein source (per serving) and breast cancer (BC) risk from a meta-analysis of 46
prospective studies (adapted from [93])
activities (vitamin C and E) but experimental showed that each 300 mg/day increase in calcium
and epidemiologic studies have yielded conflict- intake was associated with a lower BC risk in
ing results on the relationship between dietary premenopausal women by 8% and postmeno-
antioxidant vitamins and BC risk [96, 97]. A pausal women by 2%.
2015 meta-analysis of plasma retinol, vitamins
A, C and α-tocopherol levels, and BC risk (40
studies including 28 case-control studies and 12 20.4.2 Phytochemicals
nested case-control studies) showed no signifi-
cant association between plasma retinol and BC The effects of phytochemicals found in fruits,
risk but a significant increased BC risk for vegetables, dietary pulses, soy, nuts and seeds
women with the lowest plasma α-tocopherol including carotenoids, flavonoids, and phytoes-
[96]. Also, there was a significant negative asso- trogens (isoflavonoids and lignans) on BC risk,
ciation between plasma vitamin C and BC risk recurrence, and survival are summarized in
only in case-control studies but not for other Table 20.7.
vitamins. A 2011 meta-analysis intake of reti-
nol, vitamin A, C and E levels and BC risk (3 20.4.2.1 Carotenoids
RCTs, 9 cohort studies; 2 nested case-control Carotenoids are naturally occurring richly col-
studies, and 36 case-control studies) showed ored pigments providing the yellow, orange, and
that higher total vitamin A and retinol intake red colors of many fruits and vegetables, which
could significantly lower BC risk by 17% [97]. can act as a biomarker of their intake. α-Carotene,
Although dietary vitamin E reduced BC risk β-carotene, β-cryptoxanthin, lutein, zeaxanthin,
significantly when data from all studies were and lycopene are the most prevalent, comprising
pooled, the results became insignificant when 90% of circulating carotenoids. They are believed
data from cohort studies were analyzed. to be anticarcinogenic, with possible biologic
activities including antioxidation, enhanced gap-
20.4.1.6 V itamin D and Blood junction intercellular communication, immu-
25-Hydroxyvitamin D noenhancement, inhibition of tumorigenesis and
[25(OH)D] malignant transformation, and metabolism to
Experimental studies suggest that vitamin D retinoids, which in turn contributes to cellular
has potential anti-carcinogenic properties differentiation [100].
against BC risk [98]. A meta-analysis (30 pro-
spective cohort studies) found that high vita- Systematic Reviews and Meta-and Pooled
min D status is weakly associated with lower Analyses
BC risk but strongly associated with improved Three 2012 analyses provide important insights
BC survival with a risk reduction of 42% for on the protective effects of higher dietary intake
BC mortality and 39% for all-cause mortality and plasma levels of carotenoids on BC risk
[98]. [100–102]. A pooled analysis of 8 prospective
cohort studies demonstrated that women with
20.4.1.7 Calcium higher circulating levels had significantly reduced
Calcium has been suggested as a potential BC BC risk for α-carotene by 13%, β-carotene by
protective factor from experimental and observa- 17%, lutein and zeaxanthin by 16%, lycopene by
tional studies [99]. A 2016 meta-analysis (11 22%, and total carotenoids by 19% [100]. A sys-
cohort studies; 872,895 participants) found that tematic review and meta-analysis of 25 prospec-
higher calcium intake reduced overall BC risk by tive studies found that plasma concentrations of
8% with moderate heterogeneity (p = 0.026) carotenoids are more strongly associated with
[99]. In the subgroup analysis, premenopausal lower BC risk than are carotenoids assessed by
BC risk was reduced by 25% compared to 6% for dietary questionnaires [101]. Only dietary intake
postmenopausal BC. Dose-response analysis of β-carotene was significantly associated with a
reduced BC risk by 5% per 5000 mg/day. In con- dietary intake (24 case-control studies, one nested
trast, plasma concentration measures showed sig- case-control study and 6 cohort studies) reported
nificantly lower BC risk for total carotenoids, that higher intake of dietary α-carotene signifi-
β-carotene, α-carotene, and lutein by 18–32%. A cantly reduced BC risk by 9% and dietary
dose-response meta-analysis of carotenoid β-carotene intake reduced the risk by 6% when
Table 20.7 Summary of phytochemicals studies on breast cancer (BC) risk, recurrence, and mortality
Carotenoids
Systematic review and meta- and pooled analyses
Eliassen et al. (2012). 8 cohort studies; 3055 Women with higher circulating levels of α-carotene,
Examine associations between cases and 3956 controls β-carotene, lutein and zeaxanthin, lycopene, and total
circulating carotenoids and BC carotenoids may be at reduced BC risk. Statistically
risk [100] significant inverse associations (highest vs lowest
quintile) with BC risk were found with α-carotene by
13%, β-carotene by 17%, lutein and zeaxanthin by
16%, lycopene by 22%, and total carotenoids by 19%
Aune et al. (2012). 25 prospective cohort Of the 6 carotenoids assessed by dietary intake
Systematic review and studies questionaires, only β-carotene was significantly
meta-analysis of prospective associated with a reduced BC risk by 5% per
studies of dietary intake and 5000 mg/day.
blood concentrations of In contrast, blood concentration measures showed
carotenoids and BC risk [101] lower BC risk for total carotenoids by 22% per
100 μg/dL, β-carotene by 26% per 50 μg/dL,
α-carotene by 18% per 10 μg/dL, and for lutein by
32% per 25 μg dL. Blood concentrations of carotenoids
are more strongly associated with lower BC risk than
are carotenoids assessed by dietary questionnaires
Hu et al. (2012). 24 case-control studies, Comparing the highest with the lowest intake: dietary
Comprehensively summarize 1 nested case-control α-carotene intake significantly reduced BC risk by
the associations between study and 6 cohort 9.0% (p = 0.01) and dietary β-carotene intake reduced
carotenoids and BC risk and studies the risk by 6.0% (p = 0.05) when data from cohort
quantitatively estimate their studies were pooled. Total α-carotene intake had the
dose–response relationships strongest association with reducing BC risk. There
[102] were no significant associations between dietary intake
of β-cryptoxanthin, lutein/zeaxanthin, or lycopene and
BC risk
Prospective cohort or nested case-control studies
Eliassen et al. (2015). 32,826 women donated Higher concentrations of α and β-carotenes lycopene,
Examine the timing of blood samples; > and total carotenoids were associated with 18–28%
carotenoid exposure and 20 years of follow-up; statistical significant lower risks of BC (p-trend
association with BC risk, 2767 BC cases were <0.001). Plasma carotenoid concentrations were
recurrence and mortality diagnosed and matched strongly inversely associated with BC recurrence and
(Nurses’ Health Study; US) with control subjects mortality with higher β-carotene intake reducing risk
[103] (multivariate adjusted) by 68% (p-trend <0.001)
Boeke et al. (2014). 6593 adolescent girls; β-Carotene intake was inversely associated with BBD;
Examine adolescent carotenoid baseline age 12 years; comparing the highest to lowest quartile, the
intake in relation to benign 10–12 years of multivariate-adjusted odds for BBD were reduced by
breast disease (BBD) in young follow-up; intakes of 42% (p-trend = 0.03). α-carotene and lutein/zeaxanthin
women (Growing Up Today α-carotene, β-carotene, were also inversely associated with BBD, but the
Study cohort; US) [104] β-cryptoxanthin, lutein/ associations were not statistically significant
zeaxanthin, and
lycopene from food-
frequency questionnaires
(continued)
RCT
Rock et al. (2005). 1551 women previously Women in the highest quartile of plasma total
Examine the relationship treated for BC were carotenoid concentration had significantly reduced risk
between plasma carotenoid randomly assigned to the for new BC events by 43%, after controlling for
concentration, as a biomarker of control or a higher fruit covariates influencing BC prognosis
vegetable and fruit intake, and and vegetable diet
risk for a new BC event in a intervention trial arm;
cohort of women with a history mean dietary carotenoid
of early-stage breast cancer intake in this cohort was
(Women’s Healthy Eating and 14 mg/day; 7-years; 205
Living Study; US) [105] women had recurrent BC
Flavonoids
Hui et al. (2013). 12 studies (6 cohort and The risk of BC was significantly decreased in women
Examine the association 6 case-control); 9513 with high intake of flavonols by 12% and flavones by
between flavonoids, each cases and 181,906 17% compared with a low intake of flavonols and
flavonoid subclass (except controls flavones. No significant association was found for
isoflavones) and BC risk by Common flavonoids flavan-3-ols, flavanones, anthocyanins or total
conducting a meta-analysis foods and beverages, flavonoids, which lowered BC risk by 2–7%. When
[106] include onions, broccoli, studies were stratified by menopausal status significant
tea, aromatic herbs, lower BC was observed only in post-menopausal
celery, chamomile tea, women
cocoa, red wine, grapes,
apples, green tea,
oranges, berries and
black currants
Phytoestrogens (Isoflavonoids and Lignans)
Systematic review and meta-analyses
Isoflavonoids
Chen et al. (2014). 10 cohort studies and 21 In Asian countries, higher soy isoflavone intake
Explore the soy isoflavone- case-control studies significantly lowered BC risk by 41% in both pre- and
breast cancer association post-menopausal women. However, in Western
stratified by menopausal status countries, soy isoflavone intake had a marginally
[107] significant protective BC risk lowering effect only for
postmenopausal women by 8%
Xie et al. (2013). 22 studies (7 cohort and Higher isoflavone intake significantly reduced the BC
Examine the association 15 case-control designs); risk in Asian women by 32% compared to an
between isoflavones intake and menopausal status in 14 insignificant reduction of 2% in Western women.
BC risk by meta-analysis [108] studies; 9 studies from Further analysis found that the intake of isoflavones
Asian populations and 5 reduced BC risk by 54% in postmenopausal Asian
studies from Western women and 37% in premenopausal women. The
populations observed insignificant effect of soy intake in studies of
Western women appears to be primarily due to the
relatively lower intake of soy isoflavones in general
Dong and Qin (2011). 14 cohort studies of BC Higher soy isoflavones consumption was associated
Examine the association of soy incidence (369,934 with 11% lower overall BC risk. A significant
isoflavones consumption and participants and protective effect of soy isoflavones was only observed
risk of BC incidence or 5828 BC cases) and 4 in studies conducted with Asian women with a lower
recurrence, by conducting a cohort studies of BC BC risk of 24%. In Western women, the BC risk was
meta-analysis of prospective recurrence (9656 BC only reduced by 3%. Also, soy isoflavones intake were
studies [109] patients and 1226 associated with lower overall BC recurrence risk by
recurrent cases) 16%
Lignans
Seibold et al. (2014). 6 prospective studies; Higher lignan intake by BC patients was significantly
Meta-analysis on the effect of 5–10 years of follow-up associated with lower mortality risk for all-cause by
lignans on post-menopausal BC 43% and BC by 46%. Also, high enterolactone
risk and mortality [110] concentrations per 10 nmol/L were significantly
Major dietary sources of associated with lower mortality for all-cause and BC by
lignans are flax, sunflower, 6% for both
pumpkin seeds and fiber-rich
cereal.
Buck et al. (2010). 11 prospective studies Lignan exposure was associated with an insignificant
Conduct a meta-analysis on the and 10 case-control 8% lower overall BC risk, whereas in postmenopausal
association of lignan intake and studies women, high lignan intake was associated with a
BC risk [111] significant 14% reduced BC risk
data from cohort studies were pooled but dietary 20.4.2.2 Flavonoids
intake of β-cryptoxanthin, lutein/zeaxanthin, or Flavonoids are a large family of polyphenolic
lycopene did not show reduced BC risk [102]. plant compounds, which may have biological
effects related to their ability to modulate a num-
Prospective Cohort or Nested Case- ber of cell-signaling cascades with antiinflamma-
Control Studies tory, anti-thrombogenic, antidiabetic, anticancer,
Two studies are representative of the prospective and neuroprotective activities [106]. Common
studies on the effects of carotenoids on BC risk flavonoid rich foods, include onions, broccoli,
recurrence, mortality [103, 104]. A 2015 Nurses’ aromatic herbs, celery, cocoa, red grapes, apples,
Health Study (32,826 women; > 20 years of fol- oranges and other citrus fruits, berries, and black
low-up) found that higher plasma concentrations currants. A 2013 meta-analysis (6 cohort and 6
of a α and β-carotenes, lycopene, and total case-control) showed that BC risk was signifi-
carotenoids were associated with 18–28% statis- cantly decreased in women with high intake of
tically significantly lower BC risk [103]. Plasma flavonols by 12% and flavones by 17% but no
carotenoid concentrations were strongly inversely significant association was found for flavan-3-
associated with BC recurrence and mortality with ols, flavanones, anthocyanins or total flavonoids
higher β-carotene intake reducing risk by 68%. intake which showed a 2–7% lower BC risk
The 2014 Growing Up Today Study (6593 ado- [106]. Higher flavonoid intake only significantly
lescent girls; baseline age 12 years; 10–12 years reduced BC risk in post-menopausal women.
of follow-up) showed that higher dietary
β-carotene int ake reduced the risk of benign 20.4.2.3 Phytoestrogens
breast disease by 42%; α-carotene and lutein/zea-
xanthin showed insignificant reduced BC risk Soy Isoflavonoids
trends [104]. Isoflavones are the major flavonoids found in
legumes, particularly soybeans [107, 108]. Soy
Randomized Controlled Trial (RCT) isoflavones are known to have weak estrogenic
A 2005 analysis of the Women’s Healthy Eating activity due to their structural similarity with
and Living Study (1551 women previously 17-β-estradiol. Estrogens are signaling mole-
treated for BC who were randomly assigned to cules that exert their effects by binding to estro-
higher fruit and vegetable diets vs control; gen receptors within cells. Soy isoflavones can
7 years) found that women in the highest quartile preferentially bind to estrogen receptor-β mim-
of plasma total carotenoid concentration had sig- icking the effects of estrogen in some tissues and
nificantly reduced risk for a new BC event by blocking the effects of estrogen in others. Three
43% after controlling for covariates influencing systematic reviews and meta-analyses provide
BC prognosis [105]. an overview on soy isoflavonoids and BC risk
[107–109]. A 2014 analysis (35 cohort and case- were significantly associated with lower all-cause
control studies) found that higher soy isoflavone mortality by 6%, and BC-specific for all-cause
intake by both pre- and post-menopausal Asian mortality by 6% [110]. A 2010 analysis (11 pro-
women significantly lowered BC risk by 41% spective studies and 10 case-control studies)
whereas in Western women increased soy isofla- showed that higher lignan intake was associated
vone intake marginally lowered risk by 8% with a significant 14% lower overall BC risk in
[107]. A 2013 analysis (7 cohort and 15 case- postmenopausal women [111].
control studies) found that higher isoflavone
intake significantly reduced the BC risk in post-
menopausal Asian women by 54% and in pre- 20.5 W
hole (Minimally Processed)
menopausal women by 37% compared to an Plant Foods
insignificant reduction of 2% in Western women
[108]. A 2011 analysis (14 cohort studies of BC 20.5.1 Fruits and Vegetables
incidence; 369,934 participants; and 4 cohort
studies of BC recurrence; 9656 BC patients) Fruits and vegetables contain numerous nutrients
concluded that in all women higher isoflavones and phytochemicals which may reduce BC risk,
consumption lowered BC risk by 11% and in including fiber which can bind estrogens during
Asian women higher isoflavones lowered risk by enterohepatic circulation, and a range of antioxi-
24% compared to 3% in Western women [109]. dants such as carotenoids, flavonoids, and vita-
Also, higher soy isoflavones intake lowered BC mins, which can prevent oxidative damage [112].
recurrence risk by 16% in all women studies. Table 20.8 summarizes the effect of fruits and
The general lack of significant effect of isofla- vegetable studies on BC risk, recurrence and sur-
vones on BC risk in the studies of Western vival [112–121].
women may be due to the relatively lower fre-
quency or level of isoflavone intake, as there is 20.5.1.1 Systematic Reviews
little evidence that Asian women respond differ- and Meta- and Pooled
ently to soy isoflavones than non-Asian women. Analyses
Five systematic reviews and meta- and pooled
20.4.2.4 Lignans analyses provide important insights into the
Plant lignans are found in seeds (sunflower, pump- effect of fruits and vegetables on BC risk, recur-
kin, sesame, flaxseeds), whole-grain cereals and rence and mortality [112–116]. A 2017 analysis
(fiber-rich) vegetables [110]. They are metabo- (9 cohort studies and 1 RCT; a total of
lized by the colonic microbiota to the biologically 31,210 BC cases; median follow-up of
active forms enterolactone and enterodiol. 6.6 years) showed a borderline inverse associa-
Because of structural similarities, lignans can bind tion between pre-diagnostic intake of fruit, and
to estrogen receptors and thus exert both estro- improved odds of overall BC survival by 17%.
genic and anti-estrogenic effects. Enterolactone, a Whereas intake of vegetables was not signifi-
lignan metabolite, has been positively correlated cantly associated with survival with improved
with sex hormone binding globulin, which may survival odds of 4% [112]. No significant asso-
lead to lower concentrations of circulating sex ciation was found between intake of vegetables
hormones by binding free estradiol. Two system- and fruits and breast cancer-specific mortality
atic reviews and meta-analyses provide an over- after BC-diagnoses. Also, intake of cruciferous
view on lignans and BC risk and mortality [110, vegetables was not found to be protective
111]. A 2014 analysis (6 prospective studies; against BC-specific mortality. Another 2017
5–10 years of follow-up) found that higher lignin analysis (12 cohort studies; 41,185 participants;
intake significantly lowered risk of mortality for follow-up ranged from 3 to 18 years) found no
all-cause by 43% and BC risk by 46%. Also, high significant associations between fruit and veg-
blood enterolactone concentrations per 10 nmol/L etable intake (fruits and vegetables combined,
20.5 Whole (Minimally Processed) Plant Foods 593
Table 20.8 Summaries of fruit and vegetables studies on breast cancer (BC) risk, recurrence and survival
He et al. (2017). 9 cohort studies and 1 RCT; There was a borderline significant inverse
Conduct a meta-analysis 31,210 BC cases; median association for pre-diagnostic intake of
investigating the association follow-up of 6.6 years fruits and 17% improved odds of BC
between consumption of survival, whereas intake of vegetables was
vegetables and fruits and BC not significantly associated with BC survival
survival [112] with improved odds of 4% [108]. No
significant association was found between
intake of vegetables and fruits and
BC-specific mortality. Also, intake of
cruciferous vegetables was not protective
against BC-specific mortality. Finally,
increased post-diagnostic intake of
vegetables and fruits was not significantly
associated with better odds of BC survival
Peng et al. (2017). 12 cohort studies; 41,185 Comparing the highest with the lowest intake,
Evaluate the overall effect of participants; length of the risk for all-cause mortality was reduced
fruit and vegetable intake on the follow-up ranged from 3 to for intake of total vegetables by 4%, for
prognosis of breast cancer [113] 18 years cruciferous vegetables by 1%, and fruits by
12%. BC-specific mortality risk was
insignificant for fruit and vegetable intake;
total vegetable intake reduced risk of BC
recurrence by 11% and cruciferous
vegetables reduced BC recurrence by 2%. No
significant associations were found between
fruit and vegetable intake (fruits and
vegetables combined, total vegetable intake,
cruciferous vegetable intake and fruit intake)
and BC prognosis (all-cause mortality,
BC-specific mortality and BC recurrence)
Jung et al. (2013). 20 prospective studies; Total fruit and vegetable intake was
Assess the association between follow-up of 11–20 years; statistically significantly inversely
fruit and vegetable intake and 34,526 BC cases identified associated with risk of estrogen receptor
risk of estrogen receptor negative among a total of 993,466 negative BC but not with the risk of BC
BC risk (Pooling Project of women; receptor status overall or estrogen receptor positive BC
Prospective Studies of Diet and information was available for tumors. The inverse association for estrogen
Cancer; US) [114] 24,673 BC patients receptor negative tumors was observed
primarily for vegetable consumption. The
pooled lower BC risk comparing the highest
vs lowest quintile of total vegetable
consumption was 18% for estrogen receptor
negative BC (p-trend <0.001). Higher total
fruit consumption was non-statistically
significantly associated with lower risk of
estrogen receptor negative BC by 6%
Aune et al. (2012). 15 prospective studies; 5 High intake of fruits and fruits and vegetables
Clarify the association between European, 7 North America and combined, but not vegetables, is associated
fruit and vegetable intake and 3 Asia studies with modest but significant reductions in
breast cancer risk [115] breast cancer risk for the highest versus the
lowest intake by 11% for fruit and vegetables
combined, 8% for fruits and 1% for
vegetables (including starchy vegetables). In
dose-response analyses, per 200 g/d reduced
risk by 4% for fruit and vegetables combined,
5% for fruits and 0% for vegetables
(continued)
Smith-Warner et al. (2001). 8 prospective studies; 351,825 For comparisons of the highest vs lowest
Examine the association between women; 7377 BC cases quartiles of intake (pooled multivariate),
breast cancer and total and insignificant associations with lower BC
specific fruit and vegetable group risks were observed for total fruits 7%
intakes [116] (p-trend = 0.08), total vegetables by 4%
(p-trend = 0.54), and total fruits and
vegetables by 7% (p-trend = 0.12)
Farvid et al. (2016). 90,476 premenopausal women; Total fruit intake during adolescence was
Evaluate the association between aged 27–44 years who associated with a significantly lower BC
fruit and vegetable intake during completed a questionnaire on risk by 25% (median intake 2.9 servings/
adolescence and early adulthood diet in 1991 and 44,223 of day vs 0.5 serving/day). Higher early
and BC risk (Nurses’ Health those women who completed a adulthood intake of fruits and vegetables
Study II; US) [117] questionnaire about their diet rich in α carotene (yellow-orange
during adolescence in 1998; vegetables such as carrots, sweet potatoes,
22 years of follow-up; 3235 BC pumpkin, winter squash and dark-green
cases (multivariate adjusted) vegetables such as broccoli, green beans,
green peas, spinach, and avocado) was
associated with lower risk of
premenopausal BC by 18% (median intake
0.5 serving/day vs median intake 0.03
serving/day). For individual fruits and
vegetables, greater consumption
significantly reduced BC risk (per 2
servings/week) for apples by 7%, bananas
by 9%, and grapes by19% during
adolescence and oranges by 7% and kale
by 30% during early adulthood
The association with adolescent fruit intake
was stronger for both estrogen and
progesterone receptor negative cancers than
for the positive counterparts
Emaus et al. (2016). 335,054 women; mean age Higher vegetable intake was associated with
Investigate the association 51 years; median 11.5 years of a 13% lower overall BC risk vs low
between vegetable and fruit follow-up; 10,197 BC cases vegetable intake. The inverse association
intake and steroid hormone (multivariate adjusted) was most apparent for hormone receptor
receptor–defined BC risk negative BC with lower risk by 26%
(European Prospective (highest vs lowest intake; multivariate
Investigation into Cancer and p-trend = 0.03). Fruit intake was not
Nutrition [EPIC] study) [118] significantly associated with total and
hormone receptor–defined BC risk
Fung et al. (2013). 75,929 women; baseline aged For every 2 servings/week consumption
Examine the associations of 38–63 years; followed for up to of total berries BC risk was reduced by
specific fruits and vegetables on 24 years; 792 incident cases of 18% (p = 0.01) and for ≥1 serving week
risk of estrogen receptor negative estrogen receptor negative of blueberries by 31% (p = 0.02)
BC risk in postmenopausal post-menopausal BC compared with non-consumers. Also,
women (Nurses’ Health Study; (multivariate adjusted) consuming ≥2 servings of peaches/
US) [119] nectarines/week lowered BC risk by 41%
(p = 0.02). Risk of estrogen receptor
negative BC was not associated with
intakes of other specific fruits or
vegetables (Fig. 20.13)
Bao et al. (2012). 3443 BC cases and 3474 Total vegetable intake was inversely related
Evaluate associations of fruits, healthy controls; 6 years of to BC risk with a 20% reduction for the
vegetables and animal foods with follow-up (multivariate highest quintile (p- trend = 0.02;
BC risk (Shanghai Breast Cancer adjusted) multivariate adjusted). Significantly reduced
Study; China) [120] BC risk between 16 and 24% was shown for
high intake of allium vegetables, fresh
legumes, citrus fruits and Rosaceae fruits,
but inconsistent associations were observed
for total fruit intake. Elevated risk was
observed for all types of meat and fish
intake, while intakes of eggs and milk were
associated with a decreased risk of breast
cancer
Masala et al. (2012) 31,510 women; baseline age There was an inverse association between all
Evaluate the relationship range 34–64 years; 45% “vegetables” and BC with a highest vs
between vegetables and fruit post-menopausal; median lowest intake reducing risk by 35% (p-trend
consumption, overall and by 11 years of follow-up; 1072 BC = 0.03). For vegetable sub-types, there was
specific types, and BC risk cases (multivariate adjusted) an inverse association between leafy green
(Italian section of the EPIC vegetables, fruiting vegetables (e.g., peppers
study; Mediterranean population) or eggplant) and raw tomatoes and BC risk
[121] (Fig. 20.14). However, there was no
protective association observed for fruit
overall or its subtypes
total vegetable intake, cruciferous vegetable associated with modest but significant reduc-
intake and fruit intake) and BC prognosis (all- tions in BC risk for the highest versus the low-
cause mortality, BC-specific mortality and BC est intake by 11% for fruit and vegetables
recurrence) [113]. A 2013 Pooling Project of combined, 8% for fruit and 1% for vegetables
Prospective Studies of Diet and Cancer analysis (including starchy vegetables) [115]. In dose-
(20 prospective studies; 993,466 women; fol- response analyses, per 200 g/day showed
low-up of 11–20 years) showed that increased reduced BC risk for fruit and vegetables com-
total fruit and vegetable intake was statistically bined by 4%, fruit by 5%, and vegetables by
significantly inversely associated with risk of 0%. An earlier 2001 analysis (8 prospective
estrogen receptor negative BC but not with the studies; 351,825 women; 7377 BC cases)
risk of BC overall or estrogen receptor positive showed highest vs lowest quartiles of fruit and
BC tumors [114]. The inverse association for vegetable intake (pooled multivariate) had
estrogen receptor negative tumors was observed insignificant associations with lower BC risks
primarily for vegetable consumption. Higher (total fruits 7%, total vegetables by 4%, and
total vegetable intake was associated with a sig- total fruits and vegetables by 7%) [116].
nificant 18% lower risk for estrogen receptor
negative BC but higher total fruit consumption 20.5.1.2 Prospective Cohort Studies
was associated with an insignificant 6% lower Five prospective cohort studies provide more
risk of estrogen receptor negative BC. A 2012 details of the effects of specific fruits and vegeta-
dose-response analysis (15 prospective studies) bles and their dose-response effects on BC risk
found that higher intake of fruits, and fruit and [117–121]. A 2016 Nurses’ Health Study II anal-
vegetables combined, but not vegetables, was ysis (90,476 premenopausal women; aged
27–44 years; 44,223 of those women completed a a pparent for hormone receptor negative BC with
questionnaire about their diet during adolescence a significant 26% lower BC risk (highest vs low-
in 1998; 22 years of follow-up) found that total est intake; multivariate adjusted) but the effect of
fruit intake during adolescence was associated fruit intake on BC risk was not significantly asso-
with a significantly lower BC risk by 25% ciated with total and hormone receptor BC risk.
(median intake 2.9 servings/day vs 0.5 serving/ A 2013 Nurses’ Health Study analysis (75,929
day) [117]. Higher early adulthood intake of women; baseline aged 38–63 years; followed for
fruits and vegetables rich in α carotene (yellow- up to 24 years) found that estrogen receptor nega-
orange vegetables such as carrots, sweet pota- tive BC risk was significantly reduced per 2 serv-
toes, pumpkin, winter squash and dark-green ings/week of total berries, peaches and nectarines
vegetables such as broccoli, green beans, green and ≥1 serving week of blueberries (Fig. 20.13)
peas, spinach, and avocado) was associated with [119]. A 2012 Shanghai Breast Cancer Study
lower risk of premenopausal BC by 18% (median analysis (3443 BC cases and 3474 healthy con-
intake 0.5 serving/day vs median intake 0.03 trols; 6 years of follow-up) showed that total veg-
serving/day). For individual fruits and vegeta- etable intake was inversely related to BC risk
bles, greater consumption significantly reduced with a 20% reduction for the highest quintile of
BC risk (per 2 servings/week) for apples by 7%, intake [120]. Significant reduced BC risk of
bananas by 9%, and grapes by19% during ado- between 16 and 24% was shown for high intake
lescence and oranges by 7% and kale by 30% of allium vegetables, fresh legumes, citrus fruits
during early adulthood. A 2016 EPIC analysis and Rosaceae fruits (apples, pears, berries) but
(335,054 women; mean age 51 years; median inconsistent associations were observed for total
11.5 years of follow-up) showed that higher veg- fruit intake. A 2012 Italian EPIC analysis (31,510
etable intake had a 13% lower (mainly hormone women; baseline age range 34–64 years; 45%
receptor–negative) BC risk [118]. An inverse post-menopausal; median 11 years of follow-up)
association for vegetable intake was most found an inverse association between all
Total Berries (p-trend =.01) Blueberries (p-trend =.02)

Peaches/Nectarines (p-trend =.02)
Serving Frequency
0
> 1+/week 1/week to < 2/week 2+/week
–5
–10
BC Risk Reduction (%)
–15
–20
–25
–30
–35
–40
–45
Fig. 20.13 Association between specific fruits and breast cancer (BC) risk in estrogen negative post-menopausal
women from the Nurses’ Health Study (adapted from [119])
Leafy Vegetables Raw (p-trend =.005) Leafy Vegetables Cooked (p-trend =.03)
Fruiting Vegetable (p-trend =.01) Raw Tomatoes (p-trend =.03)
1.05
0.95
Hazard Ratio for BC
0.9
0.85
0.8
0.75
0.7
1 2 3 4 5
"Vegetable" Intake Quintiles*
Fig. 20.14 Association between specific vegetables and breast cancer (BC) risk (adapted from [121]). * leafy vegeta-
bles raw intake range 6–32 g/day; leafy vegetables cooked intake range 5–25 g/day; fruiting (peppers/
eggplant) vegetable intake range 15–57 g/day; rae tomatoes 14–76 g/day
vegetables and BC risk with a significant 35% associated with significantly lower premeno-
reduction for highest vs lowest intake [121]. For pausal BC risk by 18%, but not postmenopausal
sub-types of vegetables, there was an inverse BC risk [122]. This association was no longer
association between BC risk and leafy green veg- significant after further adjustment for fiber
etables, fruiting vegetables (e.g., peppers or egg- intake. The average of adolescent and early
plant) and raw tomatoes (Fig. 20.14), but there adulthood whole grain food intake was sugges-
was no protective association observed for fruit tively associated with lower premenopausal BC
overall or its subtypes. risk by 26% (p-trend = 0.09). Adult consumption
of brown rice was associated with lower BC risk
for all women by 6% and for premenopausal
20.5.2 Whole Grains women by 9% (per each 2 servings/week). Adult
white bread intake was associated with increased
Whole-grains, are rich sources of fiber, micronu- overall BC risk by 2% (per each 2 servings/
trients and phytochemicals that may influence week). Whole-grain pasta intake was inversely
breast cancer risk. Table 20.9 summarizes the associated with overall BC risk. These results
findings of two prospective cohort whole-grain suggest that high whole-grain food intake may be
studies on BC risk [122, 123]. A 2016 Nurses’ associated with lower BC risk before menopause.
Health Study II analysis (90,516 premenopausal A 2009 Danish Diet, Cancer and Health Cohort
women; aged 27–44 years; 3235 BC cases and study analysis (25,278 postmenopausal women;
44,263 women reported their diet during high mean follow-up of 9.6 years) showed in post-
school, 1347 BC cases; 22 years of follow-up) menopausal women that higher intake of whole
found that adult intake of whole-grain foods was grain products was not significantly associated
Table 20.9 Whole-grain and soy food studies on breast cancer (BC) risk, recurrence and mortality
Whole-grains
Farvid et al. (2016). 90,516 premenopausal women; aged Higher adult intake of whole grain
Evaluate individual grain-containing 27–44 years; 3235 BC cases and foods was associated with lower
foods and whole and refined grain 44,263 women reported their diet premenopausal BC risk by 18%
intake during adolescence, early during high school, 1347 BC cases; (p-trend = 0.03), but not
adulthood, and premenopausal years 22 years of follow-up (multivariate postmenopausal BC. This
in relation to BC risk (Nurses’ adjusted) association was no longer
Health Study II; US) [122] significant after further adjustment
for fiber intake. The average of
adolescent and early adulthood
whole grain food intake was
suggestively associated with lower
premenopausal BC risk by 26%
(p-trend = 0.09). Total refined grain
food intake was not associated with
BC risk. Adult brown rice intake
was associated with lower BC risk
for all women by 6% and for
premenopausal women by 9% (per
each 2 servings/week). Adult white
bread intake was associated with
increased overall BC risk (per each
2 servings/week by 2%). Whole
grain pasta intake was inversely
associated with overall BC risk.
These results suggest that high
whole grain food intake may be
associated with lower BC risk
before menopause
Egeberg et al. (2009). 25,278 postmenopausal women; In postmenopausal women, higher
Investigate the association between mean follow-up of 9.6 years; intake of whole grain products
intake of whole grain products and 978 BC cases (multivariate adjusted) was not significantly associated
BC risk by hormone receptor with a lower BC risk. Intake of
status (Danish Diet, Cancer and rye bread, oatmeal and whole
Health Cohort study; Denmark) grain bread was not associated
[123] with BC risk
Soy Foods
Wu et al. (2015). 15 case-control studies; 11,283 Higher intake of soy foods
Assess the effect of diet on BC risk Chinese women: 4602 in the significantly lowered BC risk by
in Chinese women [37] experimental group and 6681 in the 32% (p = 0.02)
control group
Nagata et al. (2014). 5 cohort studies and 6 case-control The association between soy foods
Review epidemiological studies of studies and BC risk in Japanese women is
soy intake and breast cancer among inconsistent. Only 2 of the 5 cohort
Japanese women [124] studies associated soy foods with a
significantly lower risk of BC in
postmenopausal women. In the
case-control studies only 2 of 6
studies reported a significantly
reduced BC risk
Chi et al. (2013). 5 cohort studies; 11,206 patients Higher soy food intake after BC
Evaluate the associations between diagnosis was associated with better
soy food intake after cancer survival, reduced mortality by 15%
diagnosis with breast cancer and recurrence by 21%. Subgroup
survival [125] analysis of estrogen receptor status
showed that higher soy food intake
was associated with reduced
mortality in both estrogen receptor
negative and positive patients by
25% and 28%, respectively, in both
premenopausal and postmenopausal
women. Also, higher soy food intake
was associated with reduced BC
recurrence in estrogen receptor
negative and positive by 36% and
33%, respectively, and in
postmenopausal patients by 33%
Zhong and Zhang (2012). 23 case-control and 1 cohort studies The pooled analysis of all studies
Explore the association of soy food showed that the highest intake of soy
intake and BC risk by meta-analysis foods reduced BC risk by 14% vs
[126] the lowest intake. The protective
effect was only observed in
case-control studies. Also, the
protective effect of soy foods was
only observed in Asian women with
a reduced BC risk of 24% compared
to no association with Western
women. In Asian women, the cohort
studies found lower BC risk by 19%
in post-menopausal women
Trock et al. (2006). 12 case-control and 6 cohort or Among all Asian and Western
Meta-analysis of soy intake and BC nested case-control studies women, higher soy intake was
risk [127] modestly associated with reduced BC
risk by 14%. When analyzed by soy
protein intake (grams/day), there was
statistically significantly association
with lower BC risk by 6% only
among premenopausal women
Nechuta et al. (2012). 9514 BC survivors; mean baseline Despite large differences in soy
Evaluate the association between age 54 years; mean follow-up of isoflavone intake by country,
postdiagnosis soy food consumption 7.4 years; 1171 total deaths (881 isoflavone consumption was inversely
and breast cancer outcomes among from BC) and 1348 recurrences; soy associated with BC recurrence,
US and Chinese women (After isoflavone intake (mg/day) was among both US and Chinese women,
Breast Cancer Pooling Project) measured with validated food- regardless of whether data were
[128] frequency questionnaires analyzed separately by country or
(multivariate adjusted) combined. The consumption of
≥10 mg isoflavones/day from soy
foods was associated with reduced
risk of all-cause mortality by 13%,
breast cancer-specific mortality by
17% and a statistically significantly
reduced risk of recurrence by 25%
(continued)
Caan et al. (2011). 3088 early stage BC survivors; Overall mortality risk was inversely
Examine the Women’s Healthy followed for a median of 7.3 years; associated with isoflavone intake
Eating and Living (WHEL) study soy food isoflavone intakes were (p-trend = 0.02). Women at the
population for the effect of soy measured post-diagnosis by using a highest levels of isoflavone intake
intake on BC prognosis [129] food frequency questionnaire (>16.3 mg isoflavones, equivalent to
(multivariate adjusted) ≥ one-half cup soymilk or 2-oz. tofu
each day) had a 54% lower total
mortality risk compared with the
lowest quintile of soy intake but
there was an insignificant 22%
lowering of BC recurrence (p-trend
= 0.47)
Shu et al. (2009). 5042 female BC survivors; mean age Higher soy food intake, as measured
Evaluate the association of soy food 53 years; median follow-up of either by soy protein or soy
intake after diagnosis of BC with 3.9 years; 444 deaths and 534 isoflavone intake, lowered total
total mortality and BC recurrence relapses (multivariate adjusted) mortality and BC recurrence by 33%
(Shanghai Breast Cancer Survival vs lower intake
Study; China) [130]
with a lower BC risk. Intake of rye bread, oat- c ase-control studies significantly associated with
meal and whole grain bread was not associated lower risk of BC in postmenopausal women
with BC risk [123]. [124]. A 2013 analysis (5 cohort studies; 11,206
patients) found that higher soy food intake after
BC diagnosis was associated with better survival;
20.5.3 Soy Foods reduced BC mortality by 15% and reduced BC
recurrence by 21% [125]. Subgroup analysis of
Soy food might have potential BC inhibitory estrogen receptor status showed that soy food
effects as it is a rich source of isoflavones and a intake was associated with reduced mortality in
healthy protein source as well as having a range both estrogen receptor negative and positive
of other nutrients and phytochemicals associated patients in both pre- and postmenopausal women.
with reduced BC risk [37]. Table 20.9 summa- A 2012 analysis (23 case-control studies and 1
rizes systematic reviews and meta-analyses, and cohort study) found that soy foods intake was
representative prospective cohort studies on the inversely associated with BC risk with a 14%
effects of soy foods on BC risk, recurrence and reduction (highest vs lowest intake) [126]. Also,
mortality [37, 124–130]. the protective effect of soy foods was only
observed in Asian women with a reduced BC risk
20.5.3.1 Systematic Review of 24% compared to no association with Western
and Meta-Analyses women. In Asian women, the one cohort study
Five systematic reviews and meta-analyses pro- found that higher soy foods intake reduced BC
vide important insights on the associations risk by 19% in postmenopausal women. A 2006
between soy foods and BC risk and mortality. A meta-analysis (12 case-control and 6 cohort or
2015 analysis of studies of Chinese women (15 nested case - control studies) showed that high
case-control studies; 11,283 women) showed soy intake was modestly associated with reduced
higher intake of soy foods significantly lowered BC risk by 14% among all women [127]. When
BC risk by 32% [37]. A 2014 analysis of studies exposure was analyzed by soy protein intake in
of Japanese women (5 cohort studies and 6 case- grams/day, there was a statistically significant
control studies) found inconsistent results with lower BC risk by 6% among premenopausal
only 2 of the 5 cohort studies and 2 of 6 women.
20.5.3.2 Prospective Cohort Studies flavonoids intake; (6) level of phytoestrogen

Three prospective studies provide representative containing legumes and seeds consumed; and
insights into the effects of soy foods on BC risk (7) higher vs lower intake of alcohol or cof-
and recurrence [128–130]. A 2012 analysis of fee are examples of dietary factors that may
After Breast Cancer Pooling Project of Chinese influence breast cancer (BC) risk, recurrence
and US women (9514 BC survivors; mean base- or mortality. Biological factors and mecha-
line age 54 years; mean follow-up of 7.4 years) nisms associated with diet and BC risk and
showed that despite large differences in soy iso- survival include: body weight and central
flavone intake by country, isoflavone consump- adiposity, tumor advancement, systemic and
tion was inversely associated with BC recurrence tissue li pid/fatty acid peroxidation and
among both US and Chinese women, regardless inflammation, epigenetic and transcriptional
of whether data were analyzed separately by regulation, hormone levels (e.g., estrogen,
country or combined [128]. The consumption of insulin, leptin, adiponectin and growth factor
≥10 mg isoflavones/day from soy foods was cascades), insulin resistance, and various
associated with a nonsignificant reduced risk of endometabolic and colonic microbiota pro-
all-cause mortality by 13%, breast cancer-specific cesses, which can influence BC initiation and
mortality by 17% and a statistically significant progression. Lifestyle indicators, which are
reduced risk of recurrence by 25%. A 2011 pro- associated with increased BC risk, recur-
spective examination of the Women’s Healthy rence or mortality, especially for postmeno-
Eating and Living (WHEL) study population pausal women, may include having an
(3088 early stage BC survivors; followed for a overweight or obese BMI, weight gain by
median of 7.3 years) showed that overall mortal- over 15 lbs over 4 years, and physical inac-
ity risk was inversely associated with isoflavone tivity. Patients with BC are most often either
intake [129]. Women at the highest levels of iso- overweight or obese at diagnosis and obesity
flavone intake (>16.3 mg isoflavones, equivalent increases mortality risk in both pre- and post-
to ≥ one-half cup soymilk or 2-ounces of tofu menopausal women with BC. Meta-analyses
each day) had a significant 54% reduction in BC reported that healthy dietary patterns reduced
mortality risk compared with those in the lowest overall BC risk, whereas a high consumption
quintile of soy intake. A 2009 analysis of the of alcohol and a Western diet increased BC
Shanghai Breast Cancer Survival Study (5042 BC risk. Healthy dietary patterns, especially the
survivors; mean age 53 years; median follow-up Mediterranean diet, DASH diet and the vegan
of 3.9 years) showed that soy food intake, as diet are effective in reducing BC risk and
measured either by soy protein or soy isoflavone improving odds for survival. Highly colored
intake, was inversely associated with BC mortal- non-starchy vegetables rich in carotenoids
ity and recurrence with a 33% reduction (highest and flavonoids have been significantly asso-
vs lowest intake) [130]. ciated with reduced BC risk especially in
estrogen receptor negative BC. Key adverse
Conclusions dietary components for BC risk and survival
Dietary choices including: (1) level of adher- include high intake of red and processed
ence to healthy vs Western dietary patterns; meats, high energy dense and high glycemic
(2) high vs low dietary energy density intake; foods and beverages and >1 alcoholic bever-
(3) type and level of dietary fat, fiber and pro- age/day. After BC diagnosis, soy foods (>10
tein consumed; (4) adequate vs inadequate mg isoflavones/day or > ½ cup of soy milk or
intake of calcium, folate and α-tocopherol; 2 ounces of tofu/day) may help to reduce BC
(5) type and levels of non-starchy vegetables recurrence or mortality in both Asian and
and fruits containing dietary carotenoids and Western women.
602 20 Dietary Patterns, Whole Plant Foods, Nutrients and Phytochemicals in Breast Cancer Prevention…
ppendix A: Comparison of Western and Healthy Dietary Patterns per 2000

A
kcal (Approximated Values)
Healthy
Vegetarian
Western Healthy Pattern
Dietary Pattern USDA Base DASH Diet Mediterranean (Lact-ovo
Components (US) Pattern Pattern Pattern based) Vegan Pattern
Emphasizes Refined Vegetables, Potassium Whole grains, Vegetables, Plant foods:
grains, low fruits, rich vegetables, fruit, vegetables,
fiber foods, whole-grains, vegetables, fruits, dairy whole-grains, fruits, whole
red meats, and low-fat fruits, and products, legumes, nuts, grains, nuts,
sweets, and milk low fat milk olive oil, and seeds, milk seeds, and
solid fats products moderate products, and soy foods
wine soy foods
Includes Processed Enriched Whole- Fish, nuts, Eggs, Non-dairy
meats, sugar grains, lean grains, seeds, and non-dairy milk
sweetened meat, fish, poultry, fish, pulses milk alternatives
beverages, nuts, seeds, nuts, and alternatives,
and fast and vegetable seeds and vegetable
foods oils oils
Limits Fruits and Solid fats and Red meats, Red meats, No red or No animal
vegetables, added sugars sweets and refined grains, white meats, products
and whole- sugar- and sweets or fish;
grains sweetened limited sweets
beverages
(% Total kcal)
Protein (% Total 16 17 18 16 14 13
kcal)
Total fat (% 33 32 27 34 32 30
Total kcal)
Total kcal)
Unsat. fat (% 22 25 21 24 26 25
Total kcal)
Fiber (g) 16 31 29+ 31 35+ 40+
Potassium (mg) 2800 3350 4400 3350 3300 3650
Vegetable oils 19 27 25 27 19–27 18–27
(g)
Sodium (mg) 3600 1790 1100 1690 1400 1225
Plant food groups
Fruit (cup) ≤1.0 2.0 2.5 2.5 2.0 2.0
Vegetables ≤1.5 2.5 2.1 2.5 2.5 2.5
(cup)
Whole-grains 0.5 3.0 4.0 3.0 3.0 3.0
(oz.)
Legumes (oz.) − 1.5 0.5 1.5 3.0 3.0+
Nuts/Seeds (oz.) 0.5 0.6 1.0 0.6 1.0 2.0
Soy products 0.0 0.5 − − 1.1 1.5
(oz.)
gov/ accessed 8.22.2015
Appendix B 603
2015
patterns.2015;1–9
trends. 2015;1–78
Acta. 2011;412: 1493–1514

A
Whole-
Nutrients/ Wheat, Apples, Dates, dried Potatoes, Lentils, Almonds,
Phytochemicals oat, barley, pears, figs, spinach, chickpeas, Brazil nuts,
rye, brown bananas, apricots, carrots, split peas, cashews,
rice, whole grapes, cranberries, peppers, black beans, hazelnuts,
grain oranges, raisins and lettuce, green pinto beans, macadamias,
bread, blueberries, prunes beans, cabbage, and soy beans pecans,
cereal, strawberries, onions, walnuts,
pasta, rolls and cucumber, peanuts,
and avocados cauliflower, sunflower
crackers mushrooms, seeds, and
and broccoli flaxseed
Energy (kcals) 110–350 30–170 240–310 10–115 85–170 520–700
Protein (g) 2.5–16 0.5–2.0 0.1–3.4 0.2–5.0 5.0–17 7.8–24
Available 23–77 1.0–25 64–82 0.2–25 10–27 12–33
Carbohydrate (g)
Fiber (g) 3.5–18 2.0–7.0 5.7–10 1.2–9.5 5.0–11 3.0–27
Total fat (g) 0.9–6.5 0.0–15 0.4–1.4 0.2–1.5 0.2–9.0 46–76
SFAa (g) 0.2–1.0 0.0–2.1 0.0 0.0–0.1 0.1–1.3 4.0–12
MUFAa (g) 0.2–2.0 0.0–9.8 0.0–0.2 0.1–1.0 0.1–2.0 9.0–60
PUFAa (g) 0.3–2.5 0.0–1.8 0.0–0.7 0.0.0.4 0.1–5.0 1.5–47
Folate (ug) 4.0–44 <5.0–61 2–20 8.0–160 50–210 10–230
Tocopherols 0.1–3.0 0.1–1.0 0.1–4.5 0.0–1.7 0.0–1.0 1.0–35
(mg)
Potassium (mg) 40–720 60–500 40–1160 100–680 200–520 360–1050
Calcium (mg) 7.0–50 3.0–25 10–160 5.0–200 20–100 20–265
Magnesium (mg) 40–160 3.0–30 5.0–70 3.0–80 40–90 120–400
Phytosterols 30–90 1.0–83 – 1.0–54 110–120 70–215
(mg)
Polyphenols 70–100 50–800 – 24–1250 120–6500 130–1820
(mg)
Carotenoids (ug) – 25–6600 1.0–2160 10–20,000 50–600 1.0–1200
Database for Standard Reference, Release 27. http://www.ars. usda.gov./nutrientdata. Accessed 17 February 2015
Circulation. 2013;128: 553–565
Slavin JL, Lloyd B. Health benefits of fruits and vegetables. Adv Nutr. 2012; 3:506–516
Holden JM, Eldridge AL, Beecher GR, et al. Carotenoid content of U.S. foods: An update of the database. J Food Comp
An. 1999; 12:169–196
during maturation and from different growing areas. J Agric Food Chem. 2009; 57(21):10408–10413
States. J Agric Food Chem. 2004; 52: 4026–4037
Acta. 2011;412: 1493–1514
trends. 2015;1–78
http://health.gov/dietaryguidelines/2015/guidelines/ accessed 1.26.2016
a
hypertension score, and health outcomes: a system-

References atic review and meta-analysis of cohort studies. J
Acad Nutr Diet. 2015;115(5):780–800.e5. https://
1. Global Burden of Disease Cancer Collaboration. doi.org/10.1016/j.jand.2014.12.009.
Global, regional, and national cancer incidence, mor- 10. Rice MS, Eliassen AH, Hankinson SE, et al. Breast
tality, years of life lost, years lived with disability, and cancer research in the Nurses’ Health Studies:
disability-adjusted life-years for 32 cancer groups, 1990 exposures across the life course. Am J Public
to 2015. A systematic analysis for the Global Burden Health. 2016;106:1592–8. https://doi.org/10.2105/
of Disease Study. JAMA Oncol. 2017;3(4):524–48. AJPH.2016.303325.
https://doi.org/10.1001/jamaoncol.2016.5688. 11. World Cancer Research Fund/American Institute for
2. World Cancer Research Fund/American Institute Cancer Research. Food, nutrition, physical activity,
for Cancer Research. Continuous Update Project and the prevention of cancer: a global perspective;
Report. Diet, nutrition, physical activity, and breast 2007. p. 289–95.
cancer. 2017. wcrf.org/breast-cancer. 12. American Institute for Cancer Research.
3. World Cancer Research Fund International/ Recommendations for cancer prevention. www.aicr.
American Institute for Cancer Research. Continuous org/reduce-your-risk/recommendations-for-cancer-
Update Project Report: diet, nutrition, physical prevention/recommendations. Accessed 17 May
activity, and breast cancer survivors. 2014. 2015.
4. Rossi RE, Pericleous M, Mandair D, Whyand 13. Kushi LH, Doyle C, McCullough M, et al. American
TME. The role of dietary factors in prevention and pro- Cancer Society guidelines on nutrition and physical
gression of breast cancer. Anticancer Res. 2014;34: activity for cancer prevention: reducing the risk of
6861–76. cancer with healthy food choices and physical activ-
5. American Cancer Society. Breast cancer facts & fig- ity. CA Cancer J Clin. 2012;62(1):30–67.
ures 2013–2014. Atlanta: American Cancer Society, 14. Castelló A, Martín M, Ruiz A, et al. Lower breast
Inc; 2013. cancer risk among women following the World
6. Anand P, Kunnumakkara AB, Sundaram C, et al. Cancer Research Fund and American Institute
Cancer is a preventable disease that requires major for Cancer Research lifestyle recommendations:
lifestyle changes. Pharm Res. 2008;25:2097–116. EpiGEICAM Case-Control Study. PLoS One.
https://doi.org/10.1007/s11095-008-9661-9. 2015;10(5):e0126096. https://doi.org/10.1371/jour-
7. Thomson CA, Thompson PA. Dietary patterns, nal.pone.0126096.
risk and prognosis of breast cancer. Future Oncol. 15. Thomson CA, McCullough MJ, Wertheim BC, et al.
2009;5(8):1257–69. https://doi.org/10.2217/fon.09. Nutrition and physical activity cancer prevention
86. guidelines, cancer risk, and mortality in the Women’s
8. Harmon BE, Boushey CJ, Shvetsov YB, et al. Health Initiative. Cancer Prev Res. 2014;7(1):42–53.
Associations of key diet-quality indexes with mor- 16. McKenzie F, Ferrari P, Freisling H, et al. Healthy
tality in the Multiethnic Cohort: the Dietary Patterns lifestyle and risk of breast cancer among post-
Methods Project. Am J Clin Nutr. 2015;101(3):587– menopausal women in the European Prospective
97. https://doi.org/10.3945/ajcn.114. 090688. Investigation into Cancer and Nutrition cohort study.
9. Schwingshackl L, Hoffmann G. Diet quality as Int J Cancer. 2015;136:2640–8.
assessed by the healthy eating index, the alternate 17. Dieli-Conwright CM, Lee K, Kiwata JL. Reducing
healthy eating index, the dietary approaches to stop the risk of breast cancer recurrence: an evaluation
References 605
of the effects and mechanisms of diet and exercise. BMJ. 2016;353:i2314. https://doi.org/10.1113/bmj.
Curr Breast Cancer Rep. 2016;8:139–50. https://doi. i2314.
org/10.1007/s12609-016-218-3. 30. Jiang W, Wu Y, Jiang X. Coffee and caffeine intake
18. Bhargava A. Fiber intakes and anthropometric mea- and breast cancer risk: an updated dose- response
sures are predictors of circulating hormone, triglyc- meta-analysis of 37 published studies. Gynecol
eride, and cholesterol concentrations in the Women’s Oncol. 2013;129(3):620–9. https://doi.org/10.1016/j.
Health Trial. J Nutr. 2006;136:2249–54. ygyno.2013.03.014.
19. Chan DSM, Vieira AR, Aune D, et al. Body mass 31. Bhoo-Pathy N, Peeters PHM, Uiterwaal CSPM,
index and survival in women with breast cancer- et al. Coffee and tea consumption and risk of pre-
systematic literature review and meta-analysis of 82 and postmenopausal breast cancer in the European
follow-up studies. Ann Oncol. 2014;25:1901–14. Prospective Investigation into Cancer and Nutrition
https://doi.org/10.1093/annonc/mdu042. (EPIC) cohort study. Breast Cancer Res. 2015;17:15.
20. Playdon MC, Bracken MB, Sanft TB, et al. Weight https://doi.org/10.1186/s13058-015-0521-3.
gain after breast cancer diagnosis and all-cause mor- 32. Ganmaa D, Willett WC, Li TY, et al. Coffee, tea, caf-
tality: systematic review and meta-analysis. J Natl feine and risk of breast cancer: A 22-year follow-up.
Cancer Inst. 2015;107(120):djv275. https://doi. In J Cancer. 2008;122:2071–6.
org/10.1093/jnci/djv275. 33. Neilson HK, Farris MS, Stone CR, et al. Moderate-
21. Cespedes Feliciano EM, Kroenke CH, Bradshaw vigorous recreational physical activity and breast
PT, et al. Postdiagnosis weight change and survival cancer risk, stratified by menopause status: a sys-
following a diagnosis of early-stage breast cancer. tematic review and meta-analysis. Menopause.
Cancer Epidemiol Biomark Prev. 2017;26(1):44–50. 2017;24(3):322–44. https://doi.org/10.1097/
https://doi.org/10.1158/1055-9965.EPI-16-0150. GME.00000000000745.
22. Hartman TJ, Gapstur SM, Gaudet MM, et al. 34. van den Brandt PA, Schulpen M. Mediterranean diet
Dietary energy density and postmenopausal breast adherence and risk of postmenopausal breast cancer:
cancer incidence in the Cancer Prevention Study II results of a cohort study and meta-analysis. Int J
Nutrition Cohort. J Nutr. 2016;146:2045–50. Cancer. 2017;140:2220–31.
23. Jones JA, Hartman TJ, Klifa CJ, et al. Dietary 35. Bloomfield HE, Koeller E, Greer N, et al. Effects
energy density is positively associated with breast on health outcomes of a Mediterranean diet with no
density among young women. J Acad Nutr Diet. restriction on fat intake. A systematic review and
2015;115(3):353–9. https://doi.org/10.1016/j. meta-analysis. Ann Intern Med. 2016;165(7):491–
jand.2014.08.015. 500. https://doi.org/10.7326/M16-0361.
24. Harris HR, Willett WC, Vaidya RL, Michels KB. An 36. Pourmasoumi M, Karimbeiki R, Vosoughi N,
adolescent and early adulthood dietary pattern asso- et al. Healthy eating index/alternative healthy eat-
ciated with inflammation and the incidence of breast ing index and breast cancer mortality and sur-
cancer. Cancer Res. 2017;77(5):1179–87. https:// vival: a systematic review and meta-analysis. Asia
doi.org/10.1158/0008-5472.CAN-16-2273. Pac J Oncol Nurs. 2016;3:297–305. https://doi.
25. Harris HR, Willett WC, Vaidya RL, Michels org/10.4103/2347-5625.189819.
KB. Adolescent dietary patterns and premeno- 37. Wu Y-C, Zheng D, Sun JJ, et al. Meta-analysis of
pausal breast cancer incidence. Carcinogenesis. studies on breast cancer risk and diet in Chinese
2016;37(4):376–84. https://doi.org/10.1093/carcin/ women. Int J Clin Exp Med. 2015;8(1):73–85.
bgw023. 38. Xing M-Y, Xu S-Z, Shen P. Effect of low-fat diet
26. Chen J-Y, Zhu H-C, Guo Q, et al. Dose-dependent on breast cancer survival: a meta-analysis. Asia
association between wine drinking and breast can- Pac J Cancer Prev. 2014;15:1141–4. https://doi.
cer risk- meta-analysis findings. Asian Pac J Cancer org/10.7314/APJCP.2014.15.3.1141.
Prev. 17(30):1221–33. https://doi.org/10.7314/ 39. Liu X-O, Huang Y-B, Gao Y, et al. Association
APJCP.2016.17.3.1221. between dietary factors and breast cancer risk among
27. Suzuki R, Orsini N, Mignone L, et al. Alcohol intake Chinese females: systematic review and meta-
and risk of breast cancer defined by estrogen and pro- analysis. Asian Pac J Cancer Prev. 2015;15(3):1291–
gesterone receptor status—a meta-analysis of epide- 8. https://doi.org/10.7314/APJCP.2014.15.3.1291.
miological studies. Int J Cancer. 2008;122:1832–41. 40. Farsinejad-Marj M, Talebi S, Ghiyasvand R,
https://doi.org/10.1002/ijc.23184. Miraghajani M. Adherence to Mediterranean
28. Ali AMG, Schmidt MK, Bolla MK, et al. Alcohol diet and risk of breast cancer in premenopausal
consumption and survival after a breast cancer diag- and postmenopausal women. Arch Iran Med.
nosis: a literature -based meta-analysis and collab- 2015;18(11):786–92.
orative analysis of data for 29,239 cases. Cancer 41. Albuquerque RCR, Baltar VT, Marchioni
Epidemiol Biomark Prev. 2014;23(6):934–45. DML. Breast cancer and dietary patterns: a system-
https://doi.org/10.1158/1055-9965.EPI-13-0901. atic review. Nutr Rev. 2013;72(1):1–17.
29. Dam MK, Hvidtfeldt UA, Tjonneland A, et al. 42. Brennan SF, Cantwell MM, Cardwell CR, et al.
Five-year change in alcohol intake and risk of Dietary patterns and breast cancer risk: a system-
breast cancer and coronary heart disease among atic review and meta-analysis. Am J Clin Nutr.
postmenopausal women: prospective cohort study. 2010;91:1294–302.
43. Penniecook-Sawyers JA, Jaceldo-Siegl K, Fan J, 56. Rock CL, Flatt SW, Pakiz B, et al. Effects of diet
et al. Vegetarian dietary patterns and the risk of composition on weight loss, metabolic factors and
breast cancer in a low-risk population. Br J Nutr. biomarkers in a 1-year weight loss intervention in
2016;115(10):1790–7. https://doi.org/10.1017/ obese women examined by baseline insulin resis-
S0007114516000751. tance status. Metabolism. 2016;65(11):1605–13.
44. Hirko KA, Willett WC, Hankinson SE, et al. Healthy https://doi.org/10.1016/j.metabol. 2016.07.008.
dietary patterns and risk of breast cancer by molecu- 57. Neuhouser ML, Aragaki AK, Prentice RL, et al.
lar subtype. Breast Cancer Res. 2016;155(3):579– Overweight, Obesity and Postmenopausal Invasive
88. https://doi.org/10.1007/s10549-016-3706-2. Breast Cancer Risk. JAMA Oncol. 2015;1(5):611–
45. Link LB, Canchola AJ, Bernstein L, et al. Dietary 21. https://doi.org/10.1001/jamaoncol. 2015.1546.
patterns and breast cancer risk in the California 58. Toledo E, Salas-Salvadó J, Donat-Vargas C, et al.
Teachers Study cohort. Am J Clin Nutr. 2013;98: Mediterranean diet and invasive breast cancer risk
1524–32. among women at high cardiovascular risk in the
46. Couto E, Sandin S, Lof M, et al. Mediterranean PREDIMED trial. A randomized clinical trial.
dietary pattern and risk of breast cancer. PLoS One. JAMA Intern Med. 2015;175(11):1752–60. https://
2013;8(2):e55374. https://doi.org/10.1371/journal. doi.org/10.1001/jamainternmed.2015.4838.
pone.0055374. 59. Martin LJ, Li Q, Melnichouk O, et al. A randomized
47. Buckland G, Travier N, Cottet V, et al. Adherence trial of dietary intervention for breast cancer preven-
to the Mediterranean diet and risk of breast can- tion. Cancer Res. 2011;71(1):123–33. https://doi.
cer in the European Prospective Investigation into org/10.1158/0008-5472.CAN-10-1436.
Cancer and Nutrition cohort study. Int J Cancer. 60. Pierce JP, Natarajan L, Caan BJ, et al. Dietary change
2013;132:2918–27. and reduced breast cancer events among women
48. Vrieling A, Buck K, Seibold P, et al. Dietary patterns without hot flashes after treatment of early-stage
and survival in German postmenopausal breast can- breast cancer: subgroup analysis of the Women’s
cer survivors. Br J Cancer. 2013;108:188–92. https:// Healthy Eating and Living Study. Am J Clin Nutr.
doi.org/10.1038/bjc.2012.521. 2009;89(suppl):1565S–71S.
49. Izano MA, Fung TT, Chiuve SS, et al. Are diet 61. Gold EB, Pierce JP, Natarajan L, et al. Dietary pat-
quality after breast cancer diagnosis associated tern influences breast cancer prognosis in women
with improved breast cancer survival? Nutr Cancer. without hot flashes: The Women’s Healthy Eating
2013;65(6):820–6. https://doi.org/10.1080/0163558 and Living Trial. J Clin Oncol. 2008;27:352–9.
1.2013.804939. 62. Pierce JP, Natarajan L, Caan BJ, et al. Influence
50. Kim HER, Willett WC, Fung T, et al. Diet quality of a diet very high in vegetables, fruit, and fiber
indices and postmenopausal breast cancer survival. and low in fat on prognosis following treatment
Nutr Cancer. 2011;63(3):381–8. https://doi.org/10.1 for breast cancer: the Women’s Healthy Eating
080/01635581.2011.535963. and Living (WHEL) randomized trial. JAMA.
51. Fung TT, Hu FB, Hankinson SE, et al. Low carbohy- 2007;298:289–98.
drate diets, dietary approaches to stop hypertension 63. Prentice RL, Caan B, Chlebowski RT, et al. Low-
diets, and the risk of postmenopausal breast cancer. fat dietary pattern and risk of invasive breast can-
Am J Epidemiol. 2011;174(6):652–60. https://doi. cer The Women’s Health Initiative Randomized
org/10.1093/aje/kwr148. Controlled Dietary Modification Trial. JAMA.
52. Cade JE, Taylor EF, Burley VJ, Greenwood 2006;295(6):629–42.
DC. Common dietary patterns and risk of breast 64. Chlebowski R, Blackburn GL, Thomson CA, et al.
cancer: analysis from the United Kingdom Women’s Dietary fat reduction and breast cancer outcome:
Cohort Study. Nutr Cancer. 2010;62(3):300–6. interim efficacy results for the Women’s Intervention
https://doi.org/10.1080/01635580903441246. Nutrition Study (WINS). J Natl Cancer Inst.
53. Trichopoulou A, Bamia C, Lagiou P, Trichopoulos 2006;98:1767–76.
D. Conformity to traditional Mediterranean diet 65. Rock CL, Flatt SW, Thomson CA, et al. Effects
and breast cancer risk in the Greek EPIC (European of a high-fiber, low-fat diet intervention on serum
Prospective Investigation into Cancer and Nutrition) concentrations of reproductive steroid hormones in
cohort. Am J Clin Nutr. 2010;92:620–5. women with a history of breast cancer. J Clin Oncol.
54. Cottet V, Touvier M, Fournier A, et al. Postmeno 2004;22(12):2379–87.
pausal breast cancer risk and dietary patterns in the 66. Rock CL, Thomson C, Caan BJ, et al. Reduction in
E3N-EPIC Prospective Cohort Study. Am J Epide fat intake is not associated with weight loss in most
miol. 2009;170:1257–67. https://doi.org/10.1093/ women after breast cancer diagnosis: evidence from a
aje/kwp257. randomized controlled trial. Cancer. 2001;91:25–34.
55. Kwan ML, Welzien E, Kushi LH, et al. Dietary pat- 67. Lindstrom J, Peltonen M, Eriksson JG, et al. High-
terns and breast cancer recurrence and survival among fibre, low-fat diet predicts long-term weight loss and
women with early-stage breast cancer. J Clin Oncol. decreased type 2 diabetes risk: the Finnish Diabetes
2009;27:919–26. https://doi.org/10.1200/JCO.2008. Prevention Study. Diabetologia. 2006;49:912–20.
19.4035. https://doi.org/10.1007/s00125-006-0198-3.
References 607
68. Dahl WJ, Stewart ML. Position of the Academy review and meta-analysis. Crit Rev Food Sci Nutr.
of Nutrition and Dietetics: health implications of 2017;57(10):1999–2008.https://doi.org/10.1080/10408398.
dietary fiber. J Acad Nutr Diet. 2015;115:1861–70. 2012.724481.
69. Hoy MK, Goldman JD. Fiber intake in the US popu- 84. Cao Y, Hou L, Wang W. Dietary total fat and fatty
lation. What we eat in America, NHANES 2009– acids intake, serum fatty acids and risk of breast
2010. Food Surveys Research Group. 2014; Dietary cancer: a meta-analysis of prospective cohort stud-
Data Brief No 12:1–6. ies. Int J Cancer. 2016;138:1894–904. https://doi.
70. Chen S, Chen Y, Ma S, et al. Dietary fibre intake and org/10.1002/ijc.29938.
risk of breast cancer: a systematic review and meta- 85. Xia H, Ma S, Wang S, Sun G. Meta-analysis
analysis of epidemiological studies. Oncotarget. of saturated fatty acid intake and breast cancer
2016;7(49):80980–9. risk. Medicine. 2015;94(52):e2391. https://doi.
71. Xing M-Y, Xu S-Z, Shen P. Effect of low-fat diet org/10.1097/MD000000000002391.
on breast cancer survival: a meta-analysis. Asian 86. Xin Y, Li X-Y, Wan L-X, Huang T. Vegetable oil
Pac J Cancer Prev. 2014;15(3):1141–4. https://doi. intake and breast cancer risk: a meta-analysis. Asian
org/10.7314/APJCP.2014.15.3.1141. Pac J Cancer Prev. 2015;16(12):5125–1535. https://
72. Aune D, Chan DSM, Greenwood DC, et al. Dietary doi.org/10.7314/APJCP.2015.16.12.5125.
fiber and breast cancer risk: a systematic review and 87. Yang B, Ren X-L, Fu Y-Q, et al. Ratio of n-3/n-6
meta-analysis of prospective studies. Ann Oncol. PUFA and risk of breast cancer: a meta-analysis of
2012;23:1394–402. 274,135 adult females from 11 independent prospec-
73. Dong J-Y, He K, Wang P, Qin L-Q. Dietary fiber tive studies. BMC Cancer. 2014;14:105.
intake and risk of breast cancer: a meta-analysis 88. Farvid MS, Cho E, Chen WY, et al. Premenopausal
of prospective cohort studies. Am J Clin Nutr. dietary fat in relation to pre- and postmenopausal breast
2011;94:900–5. cancer. Breast Cancer Res Treat. 2014;145:255–5.
74. Farvid MS, Eliassen H, Cho E, et al. Dietary fiber intake https://doi.org/10.1007/s10549-014-2895-9.
in young adults and breast cancer risk. Pediatrics. 89. Boeke CE, Eliassen AH, Chen WY, et al. Dietary
2016;137(3):e20151226. https://doi.org/10.1542/ fat intake in relation to lethal breast cancer in two
peds.2015-1226. large prospective cohort studies. Breast Cancer Res
75. Chhim A-S, Fassier P, Latino-Martel P, et al. Treat. 2014;146(2):383–92. https://doi.org/10.1007/
Prospective association between alcohol intake and s10549-014-3005-8.
hormone dependent cancer risk: modulation by dietary 90. Sieri S, Krogh V, Ferrari P, et al. Dietary fat and
fiber intake. Am J Clin Nutr. 2015;102(2):182–9. breast cancer risk in the European Prospective
10.39445/ajcn.114.098418. Investigation into Cancer and Nutrition. Am J Clin
76. Ferrari P, Rinaldi S, Jenab M, et al. Dietary fiber Nutr. 2008;88:1304–12.
intake and risk of hormonal receptor–defined breast 91. Chajes V, Thiebaut ACM, Rotival M, et al.
cancer in the European Prospective Investigation Association between serum trans-monounsaturated
into Cancer and Nutrition Study. Am J Clin Nutr. fatty acids and breast cancer risk in the E3N-EPIC
2013;97:344–53. Study. Am J Epidemiol. 2008;167:1312–20. https://
77. Park Y, Brinton LA, Subar AF, et al. Dietary fiber doi.org/10.1093/aje/kwn069.
intake and risk of breast cancer in postmenopausal 92. Holmes MD, Hunter DJ, Colditz GA, et al.
women: the National Institutes of Health-AARP Diet Association of dietary intake of fat and fatty acids
and Health Study. Am J Clin Nutr. 2009;90:664–71. with risk of breast cancer. JAMA. 1999;281:914–20.
78. Cade JE, Burley VJ, Greenwood DC. Dietary fibre 93. Farvid MS, Cho E, Chen WY, et al. Dietary pro-
and risk of breast cancer in the UK Women’s Cohort tein sources in early adulthood and breast can-
Study. Int J Epidemiol. 2007;36:431–8. cer incidence: prospective cohort study. BMJ.
79. Mattisson I, Wirfalt E, Johansson U, et al. Intakes of 2014;348:g3437. https://doi.org/10.1136/bmj.
plant foods, fibre and fat and risk of breast cancer— g3437.
a prospective study in the Malmo¨ Diet and Cancer 94. Wu J, Zeng R, Huang J, et al. Dietary protein sources
cohort. Br J Cancer. 2004;90:122–7. and incidence of breast cancer: a dose-response
80. Holmes MD, Liu S, Hankinson SE, et al. Dietary meta-analysis of prospective studies. Forum Nutr.
carbohydrates, fiber, and breast cancer risk. Am J 2016;8:730. https://doi.org/10.3390/nu8110730.
Epidemiol. 2004;159:732–9. 95. Zhang Y-F, Shi W-W, Gao H-F, et al. Folate intake
81. Terry P, Jain M, Miller AB, et al. No association and the risk of breast cancer: a dose-response
among total dietary fiber, fiber fractions, and risk meta-analysis of prospective studies. PLoS One.
of breast cancer. Cancer Epidemiol Biomark Prev. 2014;9(6):e100044. https://doi.org/10.1371/journal.
2002;11:1507–8. pone.0100044.
82. Willett WC, Hunter DJ, Stampfer MJ, et al. Dietary 96. Hu F, Wu Z, Li G, et al. The plasma level of retinol,
fat and fiber in relation to risk of breast cancer an vitamins A, C and α-tocopherol could reduce breast
8-year follow-up. JAMA. 1992;268:2037–44. cancer risk? A meta-analysis and meta-regression. J
83. Brennan SF, Woodside JV, Lunny PM, et al. Dietary Cancer Res Clin Oncol. 2015;141:601–14. https://
fat and breast cancer mortality: a systematic doi.org/10.1007/s00432-014-1852-7.
97. Hu F, Jiang C, Baina WY, et al. Retinol, vita- 110. Seibold P, Vrieling A, Johnson TS, et al. Enterolactone
mins A, C and E and breast cancer risk: a meta- concentrations and prognosis after postmenopausal
analysis and meta-regression. Cancer Causes breast cancer: assessment of effect modification
Control. 2011;22:1383–96. https://doi.org/10.1007/ and meta-analysis. Int J Cancer. 2014;135:923–33.
s10552-011-9811-y. https://doi.org/10.1002/ijc.28729.
98. Kim Y, Je Y. Vitamin D intake, blood 25(OH)D 111. Buck K, Zaineddin AK, Vrieling A, et al. Meta-
levels, and breast cancer risk or mortality: a meta- analyses of lignans and enterolignans in relation to
analysis. Br J Cancer. 2014;110:2772–84. https:// breast cancer risk. Am J Clin Nutr. 2010;92:141–53.
doi.org/10.1038/bjc.2014.175. https://doi.org/10.3945/ajcn.2009.28573.
99. Hidayat K, Chen G-C, Zhang R, et al. Calcium 112. He J, Gu Y, Zhang S. Consumption of vegetables and
intake and breast cancer risk: meta-analysis of pro- fruits and breast cancer survival: a systematic review
spective cohort studies. Br J Nutr. 2016;116:158–66. and meta-analysis. Sci Rep. 2017;7:599. https://doi.
https://doi.org/10.1017/S00071145160001768. org/10.1038/s41598-017-00635-5.
100. Eliassen AH, Hendrickson SJ, Brinton LA, et al. 113. Peng C, Luo W-P, Zhang C-X. Fruit and vegetable
Circulating carotenoids and risk of breast can- intake and breast cancer prognosis: a meta-analysis of
cer: pooled analysis of eight prospective studies. J prospective cohort studies. Br J Nutr. 2017;117:737–
Natl Cancer Inst. 2012;104:1905–16. https://doi. 49. https://doi.org/10.1017/S0007114517000423.
org/10.1093/jnci/djs461. 114. Jung S, Spiegelman D, Baglietto L, et al. Fruit and
101. Aune D, Chan DSM, Vieira AR, et al. Dietary compared vegetable intake and risk of breast cancer by hormone
with blood concentrations of carotenoids and breast receptor status. J Natl Cancer Inst. 2013;105:219–
cancer risk: a systematic review and meta-analysis of 36. https://doi.org/10.1093/jnci/djs635.
prospective studies. Am J Clin Nutr. 2012;96:356–73. 115. Aune D, Chan DSM, Vieira AR, et al. Fruits, vegeta-
https://doi.org/10.3945/ajcn.112.034165. bles and breast cancer risk: a systematic review and
102. Hu F, Yi BW, Zhang W, et al. Carotenoids and breast meta-analysis of prospective studies. Breast Cancer
cancer risk: a meta-analysis and meta regression. Res Treat. 2012;134(2):479–93.
Breast Cancer Res Treat. 2012;131:239–53. https:// 116. Smith-Warner SA, Spiegelman D, Yaun SS, et al.
doi.org/10.1007/s10549-011-1723-8. Intake of fruits and vegetables and risk of breast
103. Eliassen AH, Liao X, Rosner B, et al. Plasma carot- cancer: a pooled analysis of cohort studies. JAMA.
enoids and risk of breast cancer over 20 y of follow- 2001;285:769–76.
up. Am J Clin Nutr. 2015;101:1197–205. https://doi. 117. Farvid MS, Chen WY, Michels KB, et al. Fruit and
org/10.3956/ajcn.114.105080. vegetable consumption in adolescence and early
104. Boeke CE, Tamini RM, Berkey CS, et al. adulthood and risk of breast cancer: population
Adolescent carotenoid intake and benign breast based cohort study. BMJ. 2016;353:i2343. https://
disease. Pediatrics. 2014;133:e1292–8. https://doi. doi.org/10.1136/bmj.i2343.
org/10.1542/peds.2013-3844. 118. Emaus MJ, Peeters PHM, Bakker MF, et al.
105. Rock CL, Flatt SW, Natarajan L, et al. Plasma carot- Vegetable and fruit consumption and the risk of hor-
enoids and recurrence-free survival in women with a mone receptor–defined breast cancer in the EPIC
history of breast cancer. J Clin Oncol. 2005;23:6631– cohort. Am J Clin Nutr. 2016;103:168–77.
8. https://doi.org/10.1200/JCO.2005.19.505. 119. Fung TT, Chiuve SE, Willett WC, et al. Intake of
106. Hui C, Qi X, Qianyong Z, et al. Flavonoids, fla- specific fruits and vegetables in relation to risk of
vonoid subclasses and breast cancer risk: a meta- estrogen receptor-negative breast cancer among
analysis of epidemiologic studies. PLoS One. postmenopausal women. Breast Cancer Res Treat.
2013;8(1):e54318. https://doi.org/10.1371/journal. 2013;138(3):925–30. https://doi.org/10.1007/
pone.0054318. s10549-013-2484-3.
107. Chen M, Rao Y, Zheng Y, et al. Association between 120. Bao P-P, Shu X-O, Zheng Y, et al. Fruit, veg-
soy isoflavone intake and breast cancer risk for pre- etable, and animal food intake and breast cancer
and post-menopausal women: a meta-analysis of epi- risk by hormone receptor status. Nutr Cancer.
demiological studies. PLoS One. 2014;9(2):e89288. 2012;64(6):806–19.
https://doi.org/10.1371/journal.pone.0089288. 121. Masala G, Assedi M, Bendinelli B, et al. Fruit
108. Xie Q, Chen M-L, Qin Y, et al. Isoflavone consump- and vegetables consumption and breast cancer
tion and risk of breast cancer: a dose-response meta- risk: the EPIC Italy study. Breast Cancer Res
analysis of observational studies. Asia Pac J Clin Treat. 2012;132:1127–36. https://doi.org/10.1007/
Nutr. 2013;22(1):118–27. https://doi.org/10.6133/ s10549-011-1939-7.
apjcn.2013.22.1.16. 122. Farvid MS, Cho E, Eliassen AH, et al. Lifetime
109. Dong J-Y, Qin L-Q. Soy isoflavones consump- grain consumption and breast cancer risk. Breast
tion and risk of breast cancer incidence or recur- Cancer Res Treat. 2016;159(2):335–45. https://doi.
rence: a meta-analysis of prospective studies. Breast org/10.1007/s10549-016-3910-0.
Cancer Res Treat. 2011;125:315–23. https://doi. 123. Egeberg R, Olsen A, Loft S, et al. Intake of whole
org/10.1007/s10549-010-1270-8. grain products and risk of breast cancer by hormone
References 609
receptor status and histology among postmenopausal 127. Trock BJ, Hilakivi-Clarke L, Clarke R. Meta-
women. Int J Cancer. 2009;124:745–50. https://doi. analysis of soy intake and breast cancer risk. J Natl
org/10.1002/ijc.23992. Cancer Inst. 2006;98(7):459–71.
124. Nagata C, Mizoue T, Tanaka K, et al. Soy intake 128. Nechuta SJ, Caan BJ, Chen WY, et al. Soy food
and breast cancer risk: an evaluation based on intake after diagnosis of breast cancer and survival:
a systematic review of epidemiologic evidence an in-depth analysis of combined evidence from
among the Japanese population. Jpn J Clin Oncol. cohort studies of US and Chinese women. Am J
2014;44(3):282–95. https://doi.org/10.1093/jjco/ Clin Nutr. 2012;96:123–32. https://doi.org/10.3945/
hyt203. ajcn.112.035972.
125. Chi F, Wu R, Zeng Y-C, et al. Post-diagnosis soy 129. Caan BJ, Natarajan L, Parker B, et al. Soy food
food intake and breast cancer survival: a meta- consumption and breast cancer prognosis. Cancer
analysis of cohort studies. Asian Pac J Cancer Prev. Epidemiol Biomark Prev. 2011;20(5):854–8. https://
2013;14(4):2407–12. doi.org/10.1158/1055-9965.EPI-10-1041.
126. Zhong X, Zhang C. Soy food intake and breast 130. Shu XO, Zheng Y, Cai H, et al. Soy food intake
cancer risk: a meta-analysis. Wei Sheng Yan Jiu. and breast cancer survival. JAMA. 2009;302(22):
2012;41(4):670–6. 2437–43. https://doi.org/10.1001/jama.2009.1783.
Index
A B
Adenoma recurrence, 531 Beverages, 492, 496
Age related dementia, 466, 472, 478–481, 483, 490–492, alcohol, 420
496–499, 501, 504, 505 coffee, 420
beverages, 492 flavanol-rich cocoa beverages, 499
alcohol, 499, 501 100% juices, 496
caffeinated, 498 RCTs, 496–498
flavanol-rich cocoa beverages, 499 sugar and sweetened soda, 420
100% juices, 496, 497 Biological age, 4
diet quality, 467, 471 Blood 25-hydroxyvitamin D [25(OH)D], 577
foods sources, 483 Blood pressure (BP), 370, 390, 398–401, 448
fruits and vegetables, 483, 490 DASH diet vs. Western-type diet, 378
nuts, 490 fruit juice, 401, 402
soy vs. milk protein, 491, 492 fruits and vegetables
whole grains, 490 intake, 398, 401
healthy dietary patterns, 471 RCTs, 399–400
DASH diets, 481 healthy dietary pattern mechanisms, 380
MedDiet, 472, 478–480 healthy lifestyle factors, 371
MIND diet, 481 hypocaloric DASH-type vs. low fat
Nordic diet, 483 weight-loss diet, 377
lifestyle factors, 505 soy milk/cow’s milk-based diet, 403
lutein, 501 whole vs. refined grains, 395
observation studies, 504, 505 whole-grain RCTs, 393–395
RCTs, 501, 504 Body mass index (BMI), 557
pathological mechanisms, 466 all-cause mortality risk, 9
Age-related cognitive performance, 468–470, 473–477, body composition, 10
484–489, 493–496, 500–504 disability at retirement, 11
Age-related macular degeneration (AMD), 504 Body weight, 370
Aging related diseases, 4, 6 Brain-derived neurotrophic factor (BDNF), 491
Alcohol consumption Breast cancer (BC), 552, 556
prospective studies, 499, 501 lifestyle factors, 552
systematic review and meta-analysis, 499 molecular subtype biological markers, 552
Aldosterone, 370 2017 World Cancer Research Fund International/
All-cause mortality, 567 American Institute of Cancer Research
Alternate Healthy Eating Index (AHEI), 40, 258, 264 Continuous Update (CUP) Project
Alzheimer’s disease, 477, 482 Report, 556
American Heart Association’s (AHA) healthy dietary Breast cancer risk, 566, 569–571
criteria, 314 healthy vs. Western diets, 561
Ancestral fiber-rich whole foods diets, 213 MedDiet pattern, 559
Anthocyanins, 493 RCTs, 566
Antioxidant vitamins, 522 body weight, 566
Apolipoprotein E (Apo-E), 466 healthy, low fat diets, 570
Australian Blue Mountains Eye Study, 34, 35, 41 MedDiet, 570

https://doi.org/10.1007/978-3-319-59180-3
612 Index
Breast cancer risk, (cont.) dysbiosis, 118

menopause hot flash status, 570, 571 fiber intake, 118, 119
weight loss, 569 fiber-rich dietary patterns (see Fiber-rich dietary
recurrence/mortality, 560–564 patterns)
British Whitehall II study, 35, 42 human biological functions, 118
lifestyle factors, 118, 119
proteobacteria, 118, 119
C Colorectal adenoma, 516, 518–521, 540
Caffeinated beverages, 498 dietary pattern studies, 525–527
Calcium meta-analyses, 577 systematic reviews and meta-analyses, 531
Calcium, CRC, 523 vegetables and fruits intake, 538
Canadian cost-of-illness analysis, 130 whole plant foods, 532–537
Cancer prevention, adult guidelines, 553 Colorectal cancer (CRC), 516–524, 528, 531, 537–542
Cardiometabolic deaths, 336 dietary pattern studies, 524–527
Cardiovascular disease (CVD), 351, 370, 390, 396, 449 nutrients and phytochemicals, 516
Carotenoids, 523, 582 cohort studies, 521
cohort studies, 585 mechanisms, 521–524
RCTs, 585 systematic reviews and meta pooled analyses, 517
systematic reviews and meta-analyses, 582, 583 RCTs, 517, 530
Celiac disease, 168 recurrence/survival, 527, 529
Central obesity, 8, 10 specific prospective cohort studies, 524, 527, 541
Childhood constipation, 146 Asian populations, 540, 541
Chinese Longitudinal Healthy Longevity Survey, 35 EU populations, 539
Chronic constipation, 146 legumes, 540
Chronic kidney disease (CKD), 414, 418–425 nuts (peanuts), 542
abdominal obesity, 414 risk biomarkers, 528, 531
dietary factors, 419 Scandinavian populations, 541, 542
dietary fiber, 416, 417 US populations, 537, 538, 541
dietary patterns, 420 systematic reviews and meta-analyses, 524, 537, 540,
cohort studies, 423, 424 541
nutritional guidelines, 425 colorectal adenoma risk, 531
RCTs, 425 vegetarian diets vs. non-vegetarian dietary pattern,
dose-response relationship, 416 527
fiber intake/day, 418 whole plant foods, 531–537
food ingredients fruits and vegetables, 531
beverages, 420 Constipation, 148–161
minerals, 419 Academy of Nutrition and Dietetics
healthy diets, 415 recommendations, 146
lifestyle, 415 American Academy of Pediatrics recommendations,
MedDiet, 424 146
protein source, 415–417 American College of Gastroenterology
systemic inflammation, 420 recommendations, 146
Western dietary patterns, 414 American Medical Association recommendations,
whole plant foods 146
fruits and vegetables, 419 causes, 146
whole grains, 418 childhood constipation, 146
Chronic Renal Insufficiency Cohort (CRIC), 415 chronic constipation, 146
Chronological age, 4 definition, 147
Clostridium difficile infections, 120–127 European Food Safety Authority recommendations,
Coffee, 558 146
Cohort studies, BC risk fecal bulking index and total fiber content, 149
DASH diet, 566 fiber laxation mechanisms, 147, 148
MedDiet, 564, 565 fiber-rich dietary patterns
recurrence/mortality, 566 breads, 157, 158
vegetarian dietary patterns, 564 breakfast cereals, 157
Western vs. healthy dietary patterns, 561, 564 cereal brans, 152, 153
Colonic microbiota, 118–120 cereal sources, 149–156
bacteroides-related bacteria, 118 chicory inulin, 155, 160, 161
butyrate-producing bacteria, 118 dose response RCT, 156
Index 613
fecal bulking index and total fiber content, 148 blood pressure, 374–375
food ingredient intervention trials, 150–155 CHD (see Coronary heart disease (CHD))
kiwi fruit, 153, 154, 159 dietary patterns, 372
oat bran, 158 full-fat/low-fat dairy foods, 376
Pajala porridge, 157 NAFLD, 305, 306
plant cell walls, 148 obesity, 201, 202, 204, 210
plant foods, 151 observational studies, 373
polydextrose, 154, 159, 160 RCTs, 373
prebiotics, 148, 156–158 meta-analyses, 373, 376
prunes, 153, 156, 158 type 2 diabetes prevention and management, 271, 272
psyllium, 155, 156, 160 Dietary energy density, 416, 558
soluble corn fiber, 154, 160 Dietary fat, 572–577
synbiotics, 156 Dietary fiber, 416, 417, 517, 572–578
systematic reviews, 149 Dietary folate intake, 576
wheat bran, 148 Dietary nutrients, 571, 577–580
whole fruits and vegetables vs. fruit and vegetable antioxidant vitamins, 581
juices, 153, 158 blood 25-hydroxyvitamin D [25(OH)D], 582
Coronary artery disease (CAD), 356, 360 calcium, 582
Coronary artery risk development in young adults dietary fat
(CARDIA), 470 cohort studies, 580
Coronary heart disease (CHD), 318–327, 337, 350, systematic reviews and meta-analyses, 579, 580
370, 440 dietary fiber, 571
AHA healthy dietary criteria, 314 cohort studies, 577, 578
ancient vs. modern grain effect, 344 meta-analyses, 571, 577
DASH diet folate intake, 581
prospective cohort studies, 323–325 protein source, 581
RCTs, 324–327 vitamin D, 582
death rates, 314 Dietary patterns (DP), 28, 371, 420, 423, 424, 470, 471,
dietary carbohydrates vs. dietary fat, 328–330 556–559
Elderly Dietary Index, 327, 328 chronic disease risk, 36–40
foods vs. beverages, 337 cohort studies
fruits and vegetables studies, 346–349 meta-analysis, 423
healthy behaviors, 336 prospective studies, 423, 424
Healthy Eating Index, 314, 315 dietary quality score, 28
healthy lifestyles, 314 dietary rules, 29
healthy vs. Western dietary patterns, 315–318, 330, 331 healthy dietary patterns, 28, 34–36
ideal healthy diet, 314 lifestyle factors, 557
MedDiet adolescence and early adulthood, 558
dietary patterns, 318 beverages, 558, 559
prospective studies, 318–322 BMI, 557
RCTs, 320–324 dietary energy density, 558
non-soy legumes studies, 353–354 physical activity, 559
nuts and seeds studies, 356–359 weight change across lifespan, 557
soy product studies, 354–355 meta-analyses, 30
suboptimal diet quality, 314 mortality risk, 30–35
vegetarian dietary patterns, 327 nutrient-dense diets, 28
whole-grain studies, 339–342 nutritional guidelines, 425
C-reactive protein (CRP), 338, 351, 571 pooled prospective data analyses, 30
RCTs, 425
telomere length, 49–51
D Western dietary pattern (see Western dietary
Dairy products, 492 patterns)
Dementia, 466, 467 Dietary protein sources, 415, 416, 456, 458, 459,
Diabesity, 14 572–577
Diet plans, 18 legumes, 457, 458
Dietary antioxidant vitamins, 576 nuts and peanuts
Dietary approaches to stop hypertension (DASH), 36, cohort studies, 459
201, 202, 204, 323–327, 373, 376, 382, 433, meta-analyses, 458
438–440, 481, 524, 566 Dietary pulses/total legumes, 459
614 Index
Diverticular disease, 178–186 Canadian cost-of-illness analysis, 130

alcoholic beverages, 177, 178 Clostridium difficile colonic infections, 125, 127
coffee, 177 colorectal cancer, 127, 128
complications, 166, 167 fecal short-chain fatty acids (SCFAs), 120, 125
diarrhea predominate IBS, 167 fruits and vegetables, 120
dietary patterns and foods, 176, 177 healthy aging, 130
fiber mechanisms hs-C-reactive protein (hs-CRP) levels, 130, 131
body weight regulation, 185, 186 inflammatory bowel disease, 127
colonic health, 184, 185 meta-analysis, 130
fiber-rich dietary patterns metabolic syndrome, 129, 130
hospital admission/death risk, 180, 181 microbiota composition and frailty, 132–135
intervention trials, 180, 183, 184 mortality risk, 130, 131
non-vegetarian vs. vegetarian diets, 180, 182 obesity
observational studies, 178–180 fiber intake and adult obesity risk, 213, 215
quality score and risk, 180, 181 high fiber foods ranking, 219, 226, 227
total fiber intake, 180, 182 observational studies, 214–217
FODMAPs highly fermentable fiber, 178 pre-agricultural vs. present day Western dietary
FODMAPs intake, 167 pattern, 213, 214
meta-analysis, 167 RCTs, 215, 218–220
NIDDK recommendations, 178 observational studies, 120–123
nuts and seeds, 176–178 pig-human colonic model, 120, 126
symptoms, 167 premature death, 130
RCTs, 120, 123, 124
type 2 diabetes, 129
E vegetarian vs. omnivore dietary patterns, 120, 126
Elderly Dietary Index, 327, 328 weight control, 127–129
Elevated blood pressure, 391 Western and healthy dietary patterns, 120, 136
End stage renal disease (ESRD), 414, 420 whole-grains, 120, 124, 125
Energy dense diets, 196 Finnish nutritional guidelines, 15
Estimated baseline glomerular filtration rate (eGFR), 416 Finnish prospective study, 11
European Prospective Investigation into Cancer and Flavanol-rich cocoa beverages, 499
Nutrition (EPIC), 455, 553 Flavonoids, 499, 522, 584, 585
Exercise and Nutritional Interventions for Flaxseeds, 405
Neurocognitive Health Enhancement Folate (vitamin B-9), 523, 572–577
(ENLIGHTEN), 505 Foods sources, 483, 487, 490
fruits and vegetables, 483
cohort studies, 487
F RCTs, 490
Fasting lipid profiles, 351 systematic reviews, 483
Fermentable oligosaccharides, disaccharides, French SU.VI.MAX study, 18
monosaccharides and polyols (FODMAPs) Fruits and vegetables, 74–89, 344, 345, 396, 419,
diets, 169, 170, 173–177 587–589
diverticular disease, 167, 178 chronic disease risk
IBS, 168 breast cancer, 79, 84
colonic related health concerns, 173 colorectal cancer, 79, 83
low FODMAPs diet vs. high diets & traditional CVD, 74–81
IBS dietary guidance, 174, 176, 177 diabetes, 78, 82, 83
observational studies, 173 hypertension and blood pressure, 76, 77, 80–82
potential food sources, 170, 174 ischemic stroke, 77, 78, 82
RCTs, 173–176 cohort studies, 345, 349
Fiber biological mechanisms, 224 meta-analyses, 345
colonic effects, 224 flavonoids, 70
eating and digestion rates, 223 healthy aging
energy density, 222 age related cognitive performance, 84–88
fiber and healthy dietary pattern mechanisms, 222, general aging and frailty, 86–89
223 mortality risk, 70–74
postprandial satiety signaling, 223, 224 RCTs, 349
satiety and energy metabolism, 224, 225 USDA MyPlate educational concept, 70
Fiber-rich dietary patterns, 213–220, 226, 227, 577 Fruit juices, 401
Index 615
H healthy dietary pattern mechanisms, 380

Healthy aging, 130, 396 legumes, 402
body weight, 16, 17 non-soy legumes, 402
dietary patterns, 17–19 RCTs, 392, 396
Finnish nutritional guidelines, 15 refined-grain intake, 393
healthy lifestyles, 15 soy foods, 403
physical activity, 19 whole plant foods, 390
Healthy dietary patterns, 197–214, 225, 226, 471 Hypocaloric diets, 196
age-related cognitive performance, 44–49
all-cause mortality risk, 36
Alzheimer’s disease, 44–49 I
DASH and Nordic diets, 39 Insulin-like growth factor 1 (IGF-1), 557
diet quality score, 41, 43 International Study on Macro/Micronutrients and Blood
fiber-rich dietary patterns, 120, 136 Pressure (INTERMAP), 397–398
frailty risk, 42, 44 Irritable bowel syndrome (IBS), 166, 168–170, 173, 186,
healthy aging and frailty, 41, 42 187
ideal aging, 42, 44 celiac disease, 168
MedDiet, 472 dietary fiber
non-US cohorts study, 34 high fiber whole/minimally processed plant foods,
obesity 186, 187
common dietary patterns, 197, 225, 226 psyllium, 169, 170, 173
DASH dietary pattern, 201, 202, 204 RCTs, 170
MedDiet, 200, 201, 203, 204 sub-type, 169, 170
overall diet quality, 198–204 wheat bran, 169, 170, 173
vegetarian dietary pattern, 202, 205 diverticular disease (see Diverticular disease)
RCTs fiber intake, 168
DASH diet, 208, 210 FODMAP diets (see Fermentable oligosaccharides,
MedDiet, 205–211 disaccharides, monosaccharides and polyols
Nordic diet, 208, 210–213 (FODMAPs) diets)
vegetarian diets, 209, 212, 214 foods and drinks, 167
US cohorts studies, 36 pathogenesis, 166
Western dietary pattern, 38 pathophysiology, 168, 169
diet modifications, 28, 29 postprandial worsening symptoms, 168
Healthy Eating Index, 314, 315 subtypes, 166
Healthy lifestyle factors, 19–22 symptoms, 166
Healthy Lifestyle Index Score (HLIS), 555 Ischemic heart disease (IHD), 344
Healthy Nordic diet, 439 Ischemic stroke, 431, 448
Healthy Nordic food index, 441 fruits and vegetables, subtypes, 454
Healthy vegan dietary patterns, 433 whole, refined and total grain intake, 451
Healthy vs. Western diets, 433, 561 Isoflavone-rich isolated soy protein (soy protein), 491,
Helsinki Businessmen Study, 17 492
Hemorrhagic stroke, 431, 448 Isoflavonoids, 584
Histone deacetylase (HDAC), 522 Isolated fiber ingredients, 219–222
Honolulu Heart Program/Honolulu Asia Aging Study, 16
Hormone therapy (HT), 557–558
100% juices, 496 L
Hypertension, 370, 372, 373, 376, 377, 381, 382, 390, Lacto-ovo-vegetarian, 433
396, 397 Legumes, 349, 350, 402, 457, 540
cohort studies, 392 age-related cognitive function, 93, 96
dietary patterns, 371 breast cancer risk, 93, 95
DASH, 372, 373, 376 cardiometabolic and type 2 diabetes risk
guidelines, 381 management, 91, 92, 94, 95
MedDiet, 377, 381 colorectal cancer risk, 92, 95
Nordic diet, 381 CVD and stroke risk, 89–94
potential mechanisms, 382 mortality risk, 89, 90
vegetarian diets, 381, 382 non-soy legumes, 350, 351
fruits and vegetables, 396 soy products, 351, 356, 458
cohort studies, 396 Life’s Simple 7, 432, 433
meta-analyses, 396, 397 Life-sustaining processes, 4
616 Index
Lignans, 585, 586 N

Low-density lipoprotein- cholesterol (LDL-C), 336, 338 National Institute of Diabetes and Digestive and Kidney
Lutein, 501 Diseases (NIDDK) recommendations, 178
observation studies, 504 National Institutes of Health (NIH)-AARP Diet and
RCTs, 501, 504 Health Study, 60
Nonalcoholic fatty liver disease (NAFLD), 296–304,
307, 308
M body weight and central adiposity, 294, 295
Macular pigment density, 504, 505 carotenoids, 298, 299
Magnesium, CRC, 523 clinical risk factors, 292
Mediterranean diet (MedDiet), 18, 36, 47, 48, 205–207, coffee, 298
209–211, 266–271, 318–323, 377, 378, 380, DASH, 305, 306
382, 424, 426, 433, 435–438, 479–481, 524, de novo lipogenesis, 293
559, 564 definition, 292
all-cause mortality, 20 dietary patterns
Alzheimer’s disease, 47 dietary pattern quality, 299–304
blood pressure, 379–380 energy restricted/weight loss diets, 299, 300
CHD (see Coronary heart disease (CHD)) low-carbohydrate diets, 299, 300
cognitive disorders, 47 Western and healthy dietary patterns, 307, 308
cohort studies, 478, 479 dietary tips, 292
meta-analysis, 30, 47 fiber-rich whole plant foods
NAFLD, 304, 305 body weight and central obesity control, 296, 297
obesity, 200, 201, 203, 204 systemic inflammation and insulin resistance
observational studies, 377 attenuation, 297, 298
protective effects, 44–48 flavonoids, 299
RCTs, 378, 479, 480 fructose intake, 293
high fiber-rich MedDiet, 207, 209, 211 habitual diet, 292
hypocaloric MedDiet, 207, 209, 211 Iranian case control study, 292
meta-analyses, 205, 206, 378, 380 mean liver fat, 294
non-energy restricted MedDiet, 207, 209, 210 MedDiet, 304, 305
systematic reviews, 205, 206 metabolic dysfunctions, 292
unrestricted MedDiet, 205, 206, 210 MUFAs, 296
score, 35 NASH, 292
systematic reviews and meta-analyses, 472, 478 omega-3 fatty acids, 295, 296
type 2 diabetes prevention and management pathogenesis, 292, 293
meta-analyses, 266–269 prevalence, 292
representative studies, 266–271 prevention and management, 292
Mediterranean-DASH diet intervention for risk factors, 293
neurodegenerative delay (MIND), 481, 482 soy based diet, 299
Merck Manual primary strategies, 8 SREBF1, 293
Meta-analyses, BC risk, 559, 561 vitamin E, 298
healthy vs. Western diets, 561 Nonalcoholic steatohepatitis (NASH), 292
MedDiet pattern, 559 Non-APOE ε4 allele, 466
recurrence/mortality, 561 Non-soy legumes
Metabolic dysfunction, 8 cohort studies, 350, 402
Metabolic syndrome, 14 RCTs, 351, 402
Microalbuminuria, 425 Nordic diet, 208, 210, 212, 213, 381, 382, 483
Microbiota ecosystem, 406 Nordic food index, 274, 275
Mild cognitive impairment (MCI), 490 Nurses’ Health Study, 10, 11, 17, 34
Minerals Nutrition science, 29
magnesium, 420 Nuts, 96–106
phosphorus-containing ingredients, 420 almonds, 96
sodium, 419 cardiometabolic disease and type 2 diabetes
Mini–Mental State Examination (MMSE), 466 biomarkers
Monounsaturated fatty acids (MUFAs), 296, 376, 404, age-related cognitive function, 102, 105
459 intervention trials, 101, 102, 104, 105
MyPlate method, 259 systematic review and meta-analyses, 100, 101, 104
MyPlate visual educational tool, 234 telomeres, 102, 105, 106
Index 617
chronic disease risk, 97–100, 103, 104 healthy, low fat diets, 570
mortality risk MedDiet, 479, 480, 570
prospective cohort studies, 96–103 menopause hot flash status, 570, 571
randomized controlled trial, 103 meta-analyses of, 561
and peanuts, 458 nuts intake, 404
and seeds, 359, 360, 404 sesame seeds, 405
weight loss, 569
REasons for Geographic and Racial Differences in
O Stroke (REGARDS), 435, 470
Obesity, 8, 197–213, 234 Renin-angiotensin aldosterone system (RAAS), 370
definition, 196
dietary approaches, weight loss, 197
healthy dietary patterns (see Healthy dietary S
patterns) Sarcopenia, 15
hypocaloric diet plan, 197 Scandinavian HELGA cohort studies, 62
long-term weight loss maintenance, 234 Sedentary lifestyle, 13
metabolic regulatory processes, 196, 234 Selenium, CRC, 524
thermogenesis, 196 Sesame seeds, 405
weight and body composition regulation (see Whole Short chain fatty acids (SCFAs), 390
plant foods) Soy foods, 458
weight maintenance determinants, 196 cohort studies, 352, 594, 595
weight regain metabolic processes, 196 observational studies, 403
Omega-3 fatty acids, 295 RCTs, 352, 403
Osteoarthritis Initiative study, 11 systematic reviews and meta-analyses, 594
Soy isoflavonoids, 522, 586
Soy vs. milk protein, 491
P Sterol regulatory element-binding transcription factor 1
Phenolics. See Flavonoids (SREBF1), 293
Physical and physiological changes of aging, 4, 5 Stroke, 431, 433–435, 437–439, 448, 449, 452, 455,
Physically active lifestyle, 19 456, 458, 459
Physicians’ Health Study, 34 dietary patterns, 433
Phytochemicals, 582–585 cohort studies, 433–435, 437, 438
carotenoids, 582 DASH diet, 438, 439
flavonoids, 585 healthy Nordic diet, 439
Phytoestrogens, 584 RCTs, 437
lignans, 586 specific studies, 435
soy isoflavonoids, 585, 586 dose-response analysis, 455
Plant based diet, 18 foods and beverages, 448
Polyphenols, 480 fruits and vegetables, 452–454
Polyunsaturated fatty acids (PUFAs), 404 healthy vs. Western diets, 433, 434
Post-menopausal breast cancer, 555 MedDiet, 435–437
cancer mortality, 554 modifiable risk factors, 432, 433
vs. pre-menopausal breast cancer, 565 risk in US women, 435
Post-menopausal women, stroke mortality, 449 whole-and refined-grain intake, 450–451
Prediabetes, 14, 15 whole plant foods, 448
Prevencion con Dieta Mediterranea (PREDIMED), 36, dietary protein sources, 456, 458, 459
380, 559 fruits and vegetables, 449, 452, 455
Proteobacteria, 118, 119, 129 whole-grains, 449
Public health policies, 4
T
R Thermogenesis, 196
Randomized controlled trials (RCTs), 341, 356, 392, Total legume/dietary pulse, 458
396, 437, 467, 552, 556, 567–569 Triglycerides (TG), 342
body weight, 566 2017 World Cancer Research Fund International/
carotenoids, 585 American Institute of Cancer
CRC risk, 530 Research Continuous Update (CUP)
flaxseeds, 405 Project Report, 556
618 Index
Type 2 diabetes, 14–15 W

diet quality Warburg effect, 522
AHEI, 259–264 Western dietary patterns, 12, 13, 466, 467, 472
healthy and unhealthy dietary patterns, 264, 265 diet modifications, 28, 29
plant-based dietary indices, 264 fiber-rich dietary patterns, 120, 136
prospective cohort studies, 261–263, 265 vs. healthy dietary patterns, 38, 52, 53, 383, 384, 441,
RCT, 263, 266 543, 564, 596
risk decreasing vs. increasing foods, 259, 263 vs. high-fiber, 528
systematic reviews and pooled/meta-analyses, Nurses’ Health Study, 34
259–265 Physicians’ Health Study, 34
dietary patterns, 258 risk factors, 28
fiber-rich foods, 283 Whole grains, 418, 490
general guidance, 258 Whole plant foods, 59, 60, 70, 89, 96, 234, 241–247,
MedDiet (see Mediterranean diet (MedDiet)) 249–251, 276–282, 336, 448, 531, 544, 571,
moderate vs. high carbohydrate diet, 271, 272 578–582, 585, 586
mortality rates, 258 biological mechanisms, 58
MyPlate method, 259 dietary nutrients, 571
Nordic food index, 274, 275 antioxidant vitamins, 581
Nurses’ Health Studies, 259 calcium, 582
PREDIMED intervention, 282 dietary fat, 579, 580
prevalence rate, 258 dietary fiber, 571, 578
prevention and management, 259–271, 276–282 dietary protein source, 581
adult-onset diabetes, 258 folate intake, 581
AHEI, 258 vitamin D, 582
anti-inflammatory and anti-oxidant dietary energy, fiber, nutrients and phytochemicals
patterns, 282 composition, 252, 253
DASH diet, 271, 272 fruits and vegetables, 419, 449, 452, 455, 531, 586,
risk assessment model, 258 590
vegetarian diets, 271–274 advantages, 241
Western and healthy dietary patterns, 259, 284 Dietary Guidelines for Americans
whole body inflammatory homeostasis, 282 recommendations, 241 (see Fruits and
whole plant foods vegetables)
fruits and vegetables, 276, 278–280 prospective cohort studies, 242, 243, 245–247
legumes, 276, 281 RCTs, 243, 244, 247
pulse consumption, 276 systematic reviews and meta-analyses, 241–245
tree nuts and flaxseeds, 281, 282 general recommendations, 58
US dietary guidelines, 276 legumes, 248, 249 (see Legumes)
whole-grains, 276–278 nutrient and phytochemical compositions, 58, 106,
107
nuts (see Nuts)
U phytochemicals, 582
Unhealthy aging phenotypes carotenoids, 582, 585
metabolic syndrome, 14 flavonoids, 585
prediabetes, 14, 15 phytoestrogens, 585, 586
sarcopenia, 15 protein foods, 247, 248
type 2 diabetes, 14–15 soy food, 594, 595
Unhealthy/premature aging, 8 specific whole/processed plant food choices, 234, 235
US dietary guidelines, 58 sub-optimal dietary intake, 59, 60
US National Health and Nutrition Examination Surveys total and specific nuts
(NHANES), 336 almonds, 249–251
US NIH-AARP Diet and Health Study, 62 Atwater energy tables, 249
US Nurses’ Health and Health Professionals Follow-Up meta-analysis, 249
studies, 60 prospective cohort studies, 249, 250
walnuts, 251
type 2 diabetes prevention and management
V (see Type 2 diabetes)
Vegetarian dietary patterns, 202, 564 type 2 diabetes risk, 58, 59
Vegetarian diets, 381, 382 Whole-grains, 63–70, 276–278, 337, 338, 343, 449,
hypertension risk vs. omnivore diet, 382 591, 592
Vitamin D intake, 577 American intake, 60
Index 619
brown vs. white rice, 237, 240 oat ready-to-eat, 240, 241
CHD, 338 observational studies, 235–237
chronic disease risk periodontal disease, 69, 70
colorectal cancer, 66, 68 RCTs, 237–241, 342, 343
CVD, 63, 64, 66 vs. refined grain products, 235
hypertension/BP, 63–67 Scandinavian HELGA cohort studies, 62
ischemic stroke, 65, 67 type 2 diabetes prevention and management
microbiota, 68, 69 RCTs, 277, 278
type 2 diabetes, 65, 67, 68 representative cohort studies, 277
visceral fat, 69, 70 systematic reviews and meta-analyses, 276, 277
cohort studies, 338, 342 US dietary guidelines, 235
meta-analyses, 338 US NIH-AARP Diet and Health Study, 62
components, 391 US Nurses’ Health and the Health Professionals
CRC risk, 541 Follow-Up studies, 60
dietary patterns, 59 Wine/alcoholic beverages, 558
disease specific mortality risk, 60–63 Women’s Healthy Eating and Living (WHEL), 570
energy-restricted dietary intervention, 237, 240
examples, 59
meta-analyses, 60 Z
myocardial infarction (MI), 343 Zeaxanthin, 501
National Institutes of Health (NIH)-AARP Diet and
Health Study, 60

2018 Book DietaryPatternsAndWholePlantFo

Uploaded by

Copyright:

Available Formats

2018 Book DietaryPatternsAndWholePlantFo

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2018 Book DietaryPatternsAndWholePlantFo

Uploaded by

Copyright:

Available Formats

Nutrition and Health

Series Editors: Adrianne Bendich · Connie W. Bales

More information about this series at http://www.springer.com/series/7659

Dietary Patterns and

Nutrition and Health

Library of Congress Control Number: 2017959638

© Springer International Publishing AG 2018

Printed on acid-free paper

This Humana Press imprint is published by Springer Nature

Wimberley, TX Mark L. Dreher

Part I: Overview of Aging and Disease

related relaxation lifestyles. The chapter concludes with an in-depth examina-

Part II: Gastrointestinal Tract

Part III: Weight Management and Related Diseases

and eighth decades of life. We learn, in Chap. 7, that obesity is a complex

 art IV: Cardiovascular and Cerebrovascular Diseases

of a blood vessel that results in a hemorrhagic stroke. With regard to clinical

Part V: Cancer Prevention and Survival

Chapters 19 and 20 provide objective, up-to-date reviews of the associations

cancer reviewed in Chap. 19 include higher than recommended intakes of

Of importance to physicians, nutritionists, dieticians, researchers, nurses, and

Morristown, NJ Adrianne Bendich, Ph.D., FACN, FASN

Adrianne Bendich, Ph.D., F.A.S.N.,

1. Dietary Fiber in Health and Disease, edited as well as written by Mark

10. Mediterranean Diet: Impact on Health and Disease, edited by Donato

31. B  ioactive Dietary Factors and Plant Extracts in Dermatology, edited

Earlier books included Vitamin D, Second Edition, edited by Dr.

Mark L. Dreher, Ph.D., is president and

I am profoundly appreciative to the hundreds of investigators who have stud-

Part I Overview of Aging and Disease

1 Major Lifestyles and Phenotypes in Aging and Disease������������ 3

Part II Gastrointestinal Tract

4 Fiber-Rich Dietary Patterns and Colonic Microbiota

Part III Weight Management and Related Diseases

7 Dietary Patterns and Fiber in Body Weight and Composition

Part IV Cardiovascular and Cerebrovascular Diseases,

11 Dietary Patterns and Coronary Heart Disease���������������������������� 315

13 Dietary Patterns and Hypertension���������������������������������������������� 371

Part V Cancer Prevention and Survival

19 Dietary Patterns, Whole Plant Foods, Nutrients

Key Points • The concept of healthy aging includes

© Springer International Publishing AG 2018 3

1.1 Introduction During the twentieth century, life expectancy

Number of Healthy Lifestyles Adopted

Hazard Ratio of All-Cause Mortality

All-cause CVD Cancer

Odds of Healthy Survival at 70 Years or Older

All-cause disability Musculoskeletal diability

Healthy (p-trend =.33) Western (p-trend <.001)

Relative Risk of Diabetes in Women 1.4

Healthy diet (p =.68) Western diet (p =.02)

Mobility disability All-cause mortality

1.3  ajor Unhealthy Aging

Odds of Developing Frailty

Odds for Healthy Aging 1.4

a­ ctivity, 75 minutes of vigorous activity, or some 1.5  ultiple Healthy Lifestyles

Vigorous exercise (sports) Moderate/vigorous (leisure time)

MedDiet score Physical activity

Harzard Ratios of All-cause Mortality

All-cause Cancer Cardiovascular

Wimberley, TX Mark L. Dreher

Part I: Overview of Aging and Disease

Part II: Gastrointestinal Tract

Part III: Weight Management and Related Diseases

art IV: Cardiovascular and Cerebrovascular Diseases

Part V: Cancer Prevention and Survival

Morristown, NJ Adrianne Bendich, Ph.D., FACN, FASN

10. Mediterranean Diet: Impact on Health and Disease, edited by Donato

31. B ioactive Dietary Factors and Plant Extracts in Dermatology, edited

1 Major Lifestyles and Phenotypes in Aging and Disease�� 3

4 Fiber-Rich Dietary Patterns and Colonic Microbiota

7 Dietary Patterns and Fiber in Body Weight and Composition

Part IV Cardiovascular and Cerebrovascular Diseases,

11 Dietary Patterns and Coronary Heart Disease�� 315

13 Dietary Patterns and Hypertension�� 371

Part V Cancer Prevention and Survival

19 Dietary Patterns, Whole Plant Foods, Nutrients

1.1 Introduction During the twentieth century, life expectancy

1.3 ajor Unhealthy Aging

a ctivity, 75 minutes of vigorous activity, or some 1.5 ultiple Healthy Lifestyles

2.1 Introduction meats and sugar-sweetened beverages, generally

the US Southern dietary pattern (characterized 2.2.2.2 Healthy Dietary Patterns

[56–58]. A 2017 meta-analysis (9 cohort studies; significantly reduced multivariate age-related

ppendix A: Comparison of Western and Healthy Dietary Patterns

3.2 hole Plant Foods in Aging

pecific Prospective Cohort Studies

Ischemic Stroke showed a significant linear inverse relationship

significantly related to mortality risk (Fig. 3.9) ardiovascular Diseases (CVD)

(Fig. 3.14) [107]. A dose-response meta-analysis Cancer

andomized Controlled Trial (RCT)

Conclusions endothelial health, hepatic function, adipo-

ppendix A: Estimated Range of Energy, Fiber, Nutrient and Phytochemical

4.1 Introduction colonic anti-inflammatory functions. When fiber