Popular Weight Loss Diets:

What’s The Evidence?

• Low Carbohydrate Diets
CJ Segal-Isaacson, EdD, RD

• High Protein Diets

Manny Noakes, PhD

• Low Fat Diets

Keith Ayoob, EdD, RD
Application for CME credit has
been filed with the American
Academy of Family Physicians.
Determination of credit is
pending The AAFP is
accredited by the Accreditation
Council for Continuing Medical
Education (ACCME) to sponsor
Continuing Medical Education.
 Focus of Each Talk:
• Define Diet

• Present The Evidence for It’s Weight

Loss Effects

• Present The Evidence for Cardiovascular

and Insulin Sensitivity Effects

• Conclusions
Low Carbohydrate
Weight Loss Diets

CJ Segal-Isaacson, EdD RD
Assistant Professor
Division of Nutrition and Health Behavior
Department of Epidemiology and Population Health
Albert Einstein College of Medicine, Yeshiva University
 Cathy...
On Low Carbohydrate Diets
by Cathy Guisewite
 Different Strokes For Different Folks

• One size does not fit all with weight loss diets --- we need a
variety of approaches to combat the obesity epidemic.
• It is possible that lower carbohydrate diets may be more
effective for people with Metabolic Syndrome.1-2

1. Samaha FF, Iqbal N, Seshadri P, et al. N Engl J Med. May 22 2003; 348
(21): 2074-2081.
2. Eckel RH, Drazin B. J. Investigative Med. 51:Suppl. 2.2003; S383.
 Popular Low Carbohydrate Diets
•The Dr. Atkins Diet (Has four levels):
–Induction (<20 g of carbs)
–Ongoing Weight Loss (Typically 25-45 grams carbs)
–Premaintenance (Typically 30-60 grams carbs)
–Maintenance (Typically 40-100 grams carbs)

•Carbohydrate Addicts’ Diet

•Protein Power

•Cyclic Ketogenic

•The Zone Diet (30-40-30)

•Neanderthin
 2001 International
Low Carbohydrate Weight Loss Survey*

• Internet-based questionnaire.
• Data collected June-July 2001.
• 6,088 low carbohydrate dieters responded.
*Segal-Isaacson CJ, Segal-Isaacson AE, Wylie-Rosett, J. The Journal
of The American Dietetic Association. 2002. 102: S45.
 Who Responded To The Survey?
Questionnaire Was Web-based, With Online Consent Form
•SAMPLE SIZE: 6,088 “Low Carbers” Responded, June-July 2001
•GENDER: 75.4% Women 24.6% Men
•RACE: 90.5% White 3.5% Hispanic 2.7% Black 3.3% Other
•MAIN GEOGRAPHIC LOCATIONS: 87.5% USA 0.5% Canada
•HIGHEST EDUCATIONAL DEGREE ATTAINED:
0.5% Elementary School
29.4% High School
21.3% Associates Degree
31.5% Bachelors Degree
13.0% Masters Degree
4.4% Doctoral Degree
 The Three Most Common Low Carb Diets
2001 Low Carbohydrate Weight Loss Diet Survey)*
N= 5177

100.0%

80.0%

60.0%

40.0%

20.0%

0.0%

Dr. Atkins Protein Power Carb. Addicts

*Segal-Isaacson CJ, Segal-Isaacson AE, Wylie-Rosett, J. The Journal of The American

Dietetic Association. 2002. 102: S45.
 Typical Macronutrient Profile of
Low Carbohydrate Diets
Nutrient Percentage of Calories
Carbohydrate: 5-25%
Protein: 25-35%
Fat: 55-65%
Alcohol: <5% (metabolized mostly as fat)

• Macronutrient percentages may distort what actually

occurs in low carbohydrate dieting.
• Due to increased satiety, less food may be eaten.
Carbohydrate intake is reduced while fat and protein intake
may only moderately increase. Calories are often
substantially reduced.
 Typical Low Carbohydrate Dinner
Food Portion Size Calories Carbs Fat (g) Protein
(g) (g)
Roast Chicken 1/4 Chicken 306 0 19 31

Green Beans with 3/4 Cup 102 9 9 3

Sunflower Seeds
and Olive Oil

Green Salad with 2 Cups with 74 4 4 2

Italian Dressing 1 Tablespoons

Diet Gelatin with 1 Cup with 70 2 6 3

Whipped Cream 1 Tablespoon

TOTALS : 550 15 38 39
 Foods On Low Carbohydrate Diets
• Unrestricted Foods: Poultry, fish, meat, eggs, protein
powders, low-starch vegetables, artificial sweeteners, diet
gelatins, oils, butter.
• Mildly Restricted Foods: Cheeses, tofu, nuts, berries,
yogurt, milk.
• Moderately Restricted Foods: Other fruits, cream,
legumes, carbohydrate-reduced soy and grain products.
• Excluded Foods: Grains, bread, rice, potatoes, pasta,
cereals, candies, pastries, cookies, pies.
 Micronutrient Intake
• Generally adequate for most vitamins and
minerals.

• If carbohydrates are severely restricted (<20

grams):
– Diet may be low in Vitamin C, Beta Carotene and
Fiber.
– Potassium and sodium levels should be monitored and
may need to be supplemented during the first month of
the diet, as they are additionally excreted in urine.
Let’s Switch Gears…..

And Talk About Ketogenic,

Very Low Carbohydrate Diets
 The Role of Ketones
In Human Physiology
• Diets containing less than 10-15% carbohydrates
usually cause modest amounts of urinary ketosis.
• Ketones are produced as fuel from metabolized fats
(acetyl-CoA) when carbohydrate-dervived pyruvate is
low.

• Enzymes are present within all cells to convert

ketones into fuel except in erythrocytes, cornea, lens
and retina.
Adapted from a slide by Eric Westman, MD.
 The Role of Ketones (continued)
Serum ketone levels are relatively low during low
carbohydrate diets:
–Fed State 0.1 mmol/L*

–Overnight Fast 0.3 mmol/L

–Low Carb Ketogenic Diet 1-3 mmol/L**

–>20 Days Fasting 10 mmol/L

–Diabetic Ketoacidosis >25 mmol/L

*Meckling et al. Can J Physiol Pharmacol 2002;80:1095-1105.
** Sharman MJ et al. J Nutr 2002;132:1879-1885.
Adapted from a slide by Eric Westman, MD.
 Short-Term Weight Loss Studies
• Early studies comparing low carb to low fat diets often
used caloric levels of <1000 cal/day. These very low
calorie studies did not show a weight loss difference
between diets.*

• Most comparison studies today are using diets with

moderate calorie deficits of 500-700 kcal/day.

• The carbohydrate intake in current studies is usually about

that of the Atkins “Induction” Diet (most strict level) and is
generally 30 g/day.
*Yang MU, Van Itallie TB. J Clin Invest. Sep 1976;58(3):722-730.
 Foster GD, Wyatt HR, Hill JO, et al. N Engl J
Med. May 22, 2003;348(21): 2082-2090.

• First published RCT. Six month trial with follow-up to one

year. Average baseline BMI of 34 kg/m2.

• Atkins Diet (all 4 levels) with ad lib kcal but controlled

carbohydrate intake.

• Low cal diet 1200-1500 kcal for women and 1500-1800 kcal for
men. Energy intake was 60% carb, 25% fat and 15% protein.

• N = 63; 43 women and 20 men. Low carb arm=33 and Low Cal
arm=30. Average age is 44 years.

• No type 2 diabetes, lipid-lowering meds or other serious illness.

 Weight Loss Results of Foster et al.
• Better participant retention in low carb arm at all measurement
points -- 3,6 & 12 months, p<0.05.

• Did not report on actual calorie intakes of subjects.

• Weight loss significantly better in low carb group at 3 and 6

months but not at 12 months:
3 Month Wt Loss:
Low Carb -6.8 ±5.0% Low Cal -2.7 ± 3.7% (p<0.001).

6 Month Wt Loss:
Low Carb -7.0 ± 6.5% Low Cal -3.2 ± 5.6% (p<0.02).

12 Month Wt Loss:
Low Carb -4.4 ± 6.7% Low Cal -2.5 ± 6.3% (p<0.26).
Weight Loss Results of Foster et al.

From p. 2085 of Foster et al., 2003 NEJM.

Lipid Results of Foster et al.

From p. 2088 of Foster et al., 2003 NEJM.

 Blood Pressure and
Insulin Sensitivity
Results of Foster et al.

Equivalent improvements for both groups at 3,6

and 12 months for:

•Systolic and diastolic blood pressure.

•Area under the glucose curve (OGT).

•Area under the insulin curve (OGT).

•Insulin sensitivity.
 Summary Of Results From Short-Term Studies

• Low carb diets consistently produced more weight loss in

RCTs that did not control the calorie levels between diets.
1. Brehm BJ, Seeley RJ, Daniels SR, D'Alessio DA. J Clin Endocrinol
Metab. Apr 2003;88(4):1617-1623.

2. Foster GD, Wyatt HR, Hill JO, et al. N Engl J Med. May 22,
2003;348(21): 2082-2090.

3. Samaha FF, Iqbal N, Seshadri P, et al. N Engl J Med. May 22

2003;348(21):2074-2081.

4. Westman EC, Yancy WS, Guyton JS. AHA. Nutrition Abstracts.

Circulation (Supplement II). 2002;106(19).
 Summary Of Results From Short-Term Studies cont.

• In a recent study where the calories were kept the same, the
weight loss was fairly comparable between the Atkins Diet
and the DASH Diet although there was a trend for greater
weight loss by 1.3 kg in the Atkins group.

• Due to the small sample size in each group (n=13) in the

Atkins group and (n=12) in the DASH group, and the short
time the results are not definitive.
1. Stadler D, Burden V, McMurry M, Gerhard G, Connor W, Karanja N.
Presentation at Experimental Biology annual conference, April, 2003.

• Current RCT trials at the Albert Einstein College of

Medicine and Harvard University are also trying to answer
the question of diet composition per se affects weight loss
rates.
 Summary Of Results From Short-Term Studies cont.

• Low carbohydrate diets may produce more satiety and

reduced appetite …. leading to decreased caloric intake.

• Increased satiety may explain generally lower drop-out

rates in low carbohydrate groups. If there is increased
satiety, is it due to protein?
1. Brehm BJ, Seeley RJ, Daniels SR, D'Alessio DA. J Clin Endocrinol
Metab. Apr 2003;88(4):1617-1623.

2. Foster GD, Wyatt HR, Hill JO, et al. N Engl J Med. May 22,
2003;348(21): 2082-2090.

3. Samaha FF, Iqbal N, Seshadri P, et al. N Engl J Med. May 22

2003;348(21):2074-2081.

4. Westman EC, Yancy WS, Guyton JS. AHA. Nutrition Abstracts.

Circulation (Supplement II). 2002;106(19).
 Summary Of Results From Short-Term Studies cont.

• A low carbohydrate weight loss diet produced a greater

reduction than a low fat weight loss diet in the
inflammatory markers C-reactive protein and serum
amyloid A. Both markers are associated with cardiovascular
risk.

• Unclear whether the better reduction in inflammatory

markers with the low carb diet was related to greater weight
loss in the low carb group or was independent of this effect.
1. O’Brien KD, Brehm BJ, Seeley RJ, Werner M, Daneils, D,D’Alessio
DA. AHA. Nutrition Abstracts. Circulation .(Supplement II).
2002;106(19).
 Summary Of Results From Short-Term Studies cont.

• Preliminary evidence from 1 month study showed greater

urinary acid and calcium excretion among overweight
volunteers (BMI = 29.4 kg/m2) on a very low carbohydrate
diet compared to their normal diet.

• Study found no metabolic acidosis. It did find increased

urinary acid excretion that was due to the increase in sulfur-
containing amino acids from protein foods and acidic
ketone bodies. This may increase the risk of renal stones.
1. Reddy ST, Wang CY, Sakhaee K, Brinkley L, Pak CY. Am J Kidney Dis.
Aug 2002;40(2):265-274.
 Summary Of Results From Short-Term Studies cont.

• Increased calcium excretion without compensatory increase

in intestinal calcium absorption may lead to bone loss.

• However, it’s unclear whether if a low carb diet was used

for a longer time period, whether there might be
physiological adaptation to it. Longer term studies are
needed.

• Do populations such as the Eskimos who have survived on

very low carb diets show such adaptations?
1. Reddy ST, Wang CY, Sakhaee K, Brinkley L, Pak CY. Am J Kidney Dis.
Aug 2002;40(2):265-274.
 Summary Of Results From Short-Term Studies cont.

• Low Carbohydrate Weight Loss Diets may work better for

people with Metabolic Syndrome -- better weight loss, and
greater improvements in triglyceride levels. Better weight
loss tracks with insulin sensitivity improvements.
• Insulin sensitive participants lost more weight on the
LF/HC diet compared to the HC/LC diet (13.56 ± 1.60 kg
compared to 6.12 ± 1.13 kg, p<0.01). Also there was a trend
for improvements in insulin sensitvity in already IS
participants on the LF/HC diet.
 Summary Of Results From Short-Term Studies cont.

• Insulin resistant participants lost more weight on the HF/LC

diet compared to the LF/HC diet (11.46 ± 1.37 kg compared
to 6.52 ± 0.98 kg, p<0.05). Also there was a trend for
improvements in insulin sensitvity in insulin resistant
participants on the HF/LC diet.
1. Samaha FF, Iqbal N, Seshadri P, et al. N Engl J Med. May 22
2003;348(21):2074-2081.

2. Eckel RH, Drazin B. J. Investigative Med. 51:Suppl 2.2003; S383.

 Summary Of Results From 2001 Low
Carbohydrate Weight Loss Survey, N=6,088
Segal-Isaacson CJ, Segal-Isaacson AE, Wylie-Rosett, J. The Journal of
The American Dietetic Association. 2002. 102: S45.

• More than half of the respondents (55.9%) were currently dieting to

lose weight and their BMI was higher than those who were not
trying (31.2  7.8 compared to 29.8  7.6; p < 0.001).

• Those who were not currently dieting to lose weight reported they
had lost an average of 36.2  25.6 lbs but had wanted to lose 56.3 
41.3 lbs.

• The amount of weight these respondents reported they kept off for
at least one year was at least 11-20 lbs for 53.2% of the respondents
and at least 21-30 lbs for 37.4% of the respondents.
 Summary Of Results From 2001 Low
Carbohydrate Weight Loss Survey
cont.
• Most respondents modified the diet they chose (60.2%). The most
common modifications were adding more vegetables (29.0%), fruit
(12.4%) or occasionally going off the diet (32.7%).

• The three most common diet likes were not feeling hungry (87.5%)
or deprived (85.6%) and losing weight easily (74.1%).

• The three most common diet dislikes were bad breath (30.9%), light
headedness (15.0%) and nausea (8.3%).
 Current Low Carbohydrate Research
At The Albert Einstein College of Medicine

• The CCARB Study: First long-term study on low carbohydrate dieters.

Internet-based -- will follow each participant for at least three years.
More than 50% of participants have objectively documented their
weight and height through their healthcare provider. Will also assess
lipid values in a subset. Website address is:
http://epi.aecom.yu.edu/ccarbs
• The Metabolic Impact Study: Pilot controlled feeding study that uses
isocaloric-isonitrogenous diets to determine whether a very low
carbohydrate diet burns body fat more quickly than a moderate low fat
diet. Uses stable isotopes to trace whether ingested fats are oxidized or
stored; indirect calorimetry to measure energy expenditure and and
MRI to measure body fat distribution.
 Current Low Carbohydrate Research
At The Albert Einstein College of Medicine cont.

• The Diabetes Dietary Study: Study on 74 adults with type 2

diabetes who also take insulin. Study will compare a one year
trial of an Atkins-style low carbohydrate diet to the low fat diet
used in the Diabetes Prevention Project.
• Call 718 430-2161 for more information on any of the three
studies or email at [email protected] .
Annotated Bibliography
Brehm BJ, Seeley RJ, Daniels SR, D'Alessio DA.
J Clin Endocrinol Metab. Apr 2003;88(4):1617-1623.
• Six month RCT with 53 obese women comparing a low carb (<20
g/day for first two weeks then 40-60 g/day) to a low calorie-30% fat
diet with calories at an estimate of resting energy expenditure.
• Weight loss was greater in the low carb group: 8.5 ± 1.0 kg compared
to 3.9 ±1.0 kg.
• Blood pressure was normal in both groups and remained so during the
trial.
• Lipids were normal in both groups but improved with weight loss in
both groups.
• Fasting insulin and glucose decreased similarily in both groups.
Eckel RH, Drazin B. J. Investigative Med.51:Suppl
2.2003; S383.
• Controlled 16 week feeding trial with 24 obese adults who were randomized
to either a low fat/high carb or a high fat/low carb diet. Participants were
also stratified by insulin levels: <10 U/ml = insulin sensitive; >15
U/ml = insulin resistant.

• Diets were matched for energy with a 400 kcal deficit. The composition of
the low fat/high carb diet (LF/LC) was 60% CHO, 20% fat, 20% protein.
The composition of the high fat/low carb diet (HF/LC) was 40% CHO, 40%
fat, 20% protein.

• All participants lost at least 5.82 kg.

• Insulin sensitive participants lost more weight on the LF/HC diet compared
to the HC/LC diet (13.56 ± 1.60 kg compared to 6.12 ± 1.13 kg, p<0.01).
Also there was a trend for improvements in insulin sensitvity in already IS
participants on the LF/HC diet.
Eckel RH, Drazin B. J. Investigative Med.51:Suppl
2.2003; S383 cont.
• Insulin resistant participants lost more weight on the HF/LC diet compared
to the LF/HC diet (11.46 ± 1.37 kg compared to 6.52 ± 0.98 kg, p<0.05).
Also there was a trend for improvements in insulin sensitvity in insulin
resistant participants on the HF/LC diet.

• There was no relationship between baseline fasting insulin levels and weight
loss.

• The improvement in insulin sensitivity predicted the amount of weight loss,

r=0.71.

• No significant changes in LDL or HDL. Triglycerides decreased in

everyone but insulin resistant participants on the LF/HC diet.

• A high fat, low carbohydrate diet may be a more effective weight loss diet
for insulin resistant obese people.
O’Brien KD, Brehm BJ, Seeley RJ, Werner M, Daneils,
D,D’Alessio DA. AHA. Nutrition Abstracts. Circulation
(Supplement II). 2002;106(19).

• RCT of 43 obese adults comparing effects of low fat diet versus low
carbohydrate diet on the inflammatory markers C-reactive protein and
serum amyloid A.

• Six month trial with adlib diet.

• Low fat diet: n=22, Energy distribution = CHO 55%; Fat 30%;
Protein 15%

• Low carb diet: n=21, Energy distribution = CHO <10%; no other

restriction.
O’Brien KD, Brehm BJ, Seeley RJ, Werner M, Daneils,
D,D’Alessio DA. AHA. Nutrition Abstracts. Circulation
(Supplement II). 2002;106(19). Continued.

• More weight loss in low carb group 6 months: -7.3 ± 4.7kg versus -2.8
± 4.4 kg in low fat group.

• No reduction in inflammatory markers in low fat group but significant

reduction in low carb group at 6 months: -33% reduction in CRP and
-21% in serum amyloid A at 3 months.

• Unclear from abstract to what extent reduction in inflammatory

markers was related to weight loss.
Reddy ST, Wang CY, Sakhaee K, Brinkley L, Pak CY.
Am J Kidney Dis. Aug 2002;40(2):265-274.

• Purpose of study was to evaluate the effect low carb, high protein diets
on acid-base balance, calcium metabolism and stone formation.

• 10 volunteers with an average BMI of 29.4 kg/m2.

• Eight week trial: 2 weeks normal diet, 2 weeks at <20 g/day of carbs
and 4 weeks at <35 g/day.

• No metabolic acidosis.

• Increased urinary acid excretion due to sulfer-containing amino acids

from protein foods as well as acidic ketone bodies.
Reddy ST, Wang CY, Sakhaee K, Brinkley L, Pak CY.
Am J Kidney Dis. Aug 2002;40(2):265-274. Continued.

• Also Increased urinary calcium excretion without compensatory

increases in intestinal calcium absorption.

• No changes in serum potassium, chloride, CO2, calcium or phosporus

but slightly lower sodium levels.

• No change in urinary uric acid, oxalate, sodium, potassium or total

volume of urine.

• Authors conclude the net increase in acid (50 mEq/d) through the
combined effects of a high protein-low carbohydrate diet may increase
risk for renal stones and bone loss.
Samaha FF, Iqbal N, Seshadri P, et al. N Engl J Med. May
22 2003;348(21):2074-2081.

• 132 severely obese adults with an average BMI of 43 kg/m2.

• 39% participants had diabetes and 43% metabolic syndrome.

• Six month RCT at a Phil. V.A. hospital.

• Low Carb: <30 g/day of carbs with no calorie restriction.

• Low Fat / LowCal: 30% dietary fat with 500 calorie deficit.

• Dietary compliance evaluated with 24 recalls.

Samaha FF, Iqbal N, Seshadri P, et al. N Engl J Med. May
22 2003;348(21):2074-2081. Continued.

• More weight lost on low carb diet: 5.8 ± 8.6 kg compared to 1.9 ± 4.2 kg;
p=0.002.

• Triglycerides decreased more in the low carb group.

• No other lipid differences.

• Fasting glucose decreased more in low carb group among its diabetic
subjects with concommittant reductions in DM meds.

• Insulin sensititivity improved more in low carb group (this was only
measured in non-diabetic participants of both groups).
 Stadler D, Burden V, McMurry M, Gerhard G,
Connor W, Karanja N. Presentation at Experimental
Biology annual conference, April, 2003.
• Trial matched the caloric levels of 13 participants on the Atkins diet
with another 12 participants on the DASH diet.

• The average ad lib calorie reduction on the Atkins diet was 68% of
usual and the calorie intake of participants on the DASH diet was
matched at 67% of usual.

• The participants were given meals prepared in a metabolic kitchen

for 42 days.

• Although weight loss was not significantly different between the two
diet groups, there was a trend for greater weight loss in the Atkins
group by 1.3 kg.
Westman EC, Yancy WS, Edman JS, Tomlin KF, Perkins
CE. Am J Med. Jul 2002;113(1):30-36.
• One-armed trial.

• 41 of 51 adults (80%) completed six month study.

• Atkins Diet (adlib) with initially <25 g carbs + Atkins MVI suppl.
Carbs increased to 50 g if 40% of wt loss achieved.

• Average weight loss 10.3 ± 5.9% p<0.001.

• Average fat loss of 2.9 ± 3.2% p<0.01 (skinfolds).

• Significant improvements in: systolic and diastolic BP, total

cholesterol, LDL, triglycerides and HDL.

• Increases in urinary calcium and uric acid.

Westman EC, Yancy WS, Guyton JS. AHA. Nutrition Abstracts.
Circulation (Supplement II). 2002;106(19).
•Six month RCT comparing Atkins Diet (<20 g/day Carbs) to Low
Fat/Low Cal Diet (no composition given in abstract).
•MVI, Borage, Flaxseed and Fish Oils supplements given to low carb
group.
•Mean BMI of 34.5 kg/m2 at baseline.
•Low Carb Group, n= 36; Low Fat Group n = 27.
•Greater weight loss in low carb group: 13.8% to 8.8%.
•Reduction in VLDL greater in low carb group.
•Both groups had a 73% reduction in small LDL and an overall increase
in LDL particle size.
•The low carb group had a larger increase in large LDL particles.
Popular Weight Loss Diets:
What’s The Evidence?
Moderately High Protein Lower Carbohydrate
(30% protein; <30% fat; 40% carbohydrate)

Dr Manny Noakes
Senior Research Dietitian
CSIRO Clinical Research Unit
Australia

Health Sciences and Nutrition

Rationale for Moderately High
Protein Intakes in Weight Loss?

Emerging scientific evidence for

effects on satiety

Dietary pattern that is nutritionally

adequate
 Emerging Scientific Evidence
 Lean body mass spared - glucose tolerant women
(Piatti et al,1994)

 Increased satiety - glucose tolerant women

(Latner & Schwartz, 1999)

 Thermic effects - obese, hyperinsulinaemic men

(Hwalla Baba et al,1999)

 High protein/Low GI -25% decrease in ad libitum intake – obese men

(Dumesnil et al 2001)

 Thermic effect of HP meal 28% greater – no change in REE - Type 2

diabetes (Luscombe et al 2002)

 Endocrine and metabolic improvements - women with PCOS

(Moran et al 2003)
 Zone vs Atkins vs Conventional

• Mean weight loss was 5.1 kg for those who completed the
12-week program.
• No significant differences in total weight, fat, or lean body
mass loss by diet group.

• 91 commenced study and 49 completed the 12 weeks.

Attrition was substantial for all plans
– Atkins 43%,
– Zone 60%,
– Conventional 36%
Landers P J Okla State Med Assoc. 2002
 Protein vs Carbohydrate in ad libitum
Fat Reduced Diet
Skov et al 1999

 High-carbohydrate
(HC, protein 12% energy) n=25

 High-protein
(HP, protein 25% energy) n=25

RESULTS AFTER 6 MONTHS

 5.1 kg in the HC group
 8.9 kg in the HP group (P < 0.001)
 More subjects lost > 10 kg in the HP group (35% vs 9%).
 HP diet decreased fasting plasma triglycerides significantly.
 Increased Protein/Carb ratio
in overweight women

CHO/protein ratio
3.5 (CHO Group)
68g protein
1.4 (Protein group)
125g protein

Layman et al J Nutr. 2003

 Increased Protein/Carb ratio
in overweight women

CHO/protein ratio
3.5 (CHO Group)
68g protein
1.4 (Protein group)
125g protein

Layman et al J Nutr. 2003

 0 4 8 12

HP Weight Loss Maintenance

LP Weight Loss Maintenance

Interventions: Fat <30%, sat fat <10% 6500kJ

High Protein (HP) - 28% E Protein, 42% E CHO

Low Protein (LP) - 16% E Protein, 55% E CHO

Parker B, Noakes M, Luscombe N, Clifton P.
Diabetes Care. 2002
Foods HP diet LP diet
Weetbix 1 1
Milk - 1% fat 250ml 250ml
Skim milk powder 30g -
Fruit Yoghurt - 0.1% fat 200g (7oz) -
Wholemeal bread 2 sl 3 sl
Fruit 2 small 3 small
Meat/fish/chicken 200g (7oz) 100g (3.5oz)
Low kJ Vegetables <2.5 cups <2.5 cups
Cheese - 3% fat 60g (2oz) -
Rice - 1 cup cooked
Canola lite margarine 2 tsp 3 tsp
Sunola oil 3 tsp 3 tsp
Shortbread biscuits 3 4
 Dietary Protein Effects in Type 2
Diabetes
Total fat mass significantly decreased in
women on HP diet (12.4%)

Abdominal fat mass significantly

decreased in women on HP diet (12%)

Significant decrease in LDL cholesterol

levels on HP diet (5.7%)
Parker B, Noakes M, Luscombe N, Clifton P.
Diabetes Care. 2002
 High Protein Diets and Renal
Function in D2
Urinary albumin excretion
was not affected by dietary
protein during weight loss:

n=19 with
microalbuminuria
HP diet: 24.2 to 19.8 mg/l
(n=12)
LP diet: 4.3 to 3.5 mg/l
(n=7) Parker B, Noakes M, Luscombe N, Clifton P.
Diabetes Care. 2002
 High Protein Diets In
Subjects With High Insulin
Diet effect p < 0.05
2
9.6%
Triglycerides (mMol/L)

1.8

1.6
High carb
1.4 High
Protein
1.2 22.8%

1
0 4 8 12 16

Week Farnsworth et al AJCN 2003

 Insulin Response to Test Meals
Diet effect p < 0.05
140
120
100
Insulin (mU/l)

80 High Protein
60 High Carb
40
20
0
0 30 60 120 180

Time (min) Farnsworth et al AJCN 2003

 Glucose Response to Test Meals

Diet effect p = 0.020

9
8
Glucose (mMol/l)

7 High Protein

6 High carb

5
4
0 30 60 120 180

Time (min) Farnsworth et al AJCN 2003

 Aim:
To compare a high protein
weight loss diet with a
high carbohydrate diet on

 weight loss and body

composition
 nutrient status
 bone turnover markers Diets:
 measures of heart disease risk
1. High
in women over 12 carbohydrate
weeks… ..and
subsequently follow up
for 1 year 2. High protein,
high red meat
High Meat Protein 5600 KJ (1330 kcal)
• Cereal
• Low fat milk (250ml) 8oz
• Wholemeal bread (2 slices)
• Fruit (2)
• Beef/lamb 200g (7oz)-
34% protein • dinner
20% fat • Chicken/fish/meat 100g (3.5oz)
lunch
46% carbohydrate
• Vegetables 2.5 cups
• Diet Yoghurt 200g (7oz)
• Canola oil 3 tsp
• Wine 2 glasses/week (optional)

109g protein 31g fat 161g carbohydrate

High Carbohydrate 5600 KJ (1330 kcal)
• Cereal
• SKIM milk (250ml) 8oz
• Wholemeal bread (3 slices)
• Fruit (3)
17% protein • Chicken/pork/fish 80g - 3oz
20% fat • Vegetables 2.5 cups
64% carbohydrate • Canola oil 3 tsp
• Pasta/rice 120g/4oz cooked
• Low fat biscuits 3
• Wine 2 glasses/week (optional)

57g protein 31g fat 229 g carbohydrate

 Baseline characteristics by TG
Status
LOW TG HIGH TG
TREAT HP HC HP HC
Mean Mean Mean Mean
AGE 49.3 45.5 49.8 52.4
BMI 32.8 33.0 31.8 32.9
WEIGHT 89.1 86.5 85.2 86.4
Triglycerides 0.9 0.9 1.9 2.0
 Weight Loss by Triglyceride Status
•Sig diet effect
P=0.023

9.0
8.0 7.8
8.0 7.3
weight loss (kg)

7.0 High
5.9 protein
6.0
High
5.0 carb
4.0

3.0
Low TG High TG
 Total And Midriff Fat Loss
Dexa Data

* Sig diet effect

8
6.28
5.86
6 5.13

3.63 High
4
kg

protein
High
2 carb

0
hig h p ro t e in - hig h hi g h p ro t e in - hig h
lo w TG c a rb o hy d ra t e - hig h TG c a rb o hy d ra t e -
lo w TG hig h TG
 Triglycerides by TG status
2.5
HIGH TG Diet X TG interaction P=0.01
2
10%
1.5
TGmmol/L

28%
1

LOW TG
0.5

0
week 0 week 4 week 8 week 12
 CRP

14
12
10
CRP mg/L

8 high protein
6 high carb
4
2
0
week 0 week 12

(Effect of diet P=0.07, with overall ANOVA p=0.019)

after accounting for TG status
 Total T3

2.6

2.4

2.2
pmol/L

high protein
2 high carb

1.8

1.6
week 0 week 12

Significant effect of diet P=0.005 suggesting

lower energy intake on high protein diet
 Calcium Excretion
4.6
Calcium excretion decreased on both diets
4.4

4.2
Mean Calcium excretion mmol/24hr

4.0

3.8

reference range
3.6
2.5-7.5mmol/24hr
3.4

3.2

3.0 week 0
calciumv1

2.8 week12
calciumv4
1 2
High protein High carb
TREAT
 CONCLUSION
There is emerging evidence that moderately
high protein lower carbohydrate diets for
weight loss are advantageous to conventional
high carbohydrate diets for:
Women with high TG
Women with type 2 diabetes
Subjects with elevated insulin levels
There is no evidence that such moderate dietary patterns pose
any health risks although more extended studies in overweight
subjects with mild renal impairment are necessary to confirm
safety in this group
Popular Weight Loss Diets: What’s the
Evidence?
Low-fat and Very Low-fat Diets
Keith-Thomas Ayoob, EdD, RD, FADA
Albert Einstein College of Medicine
July 17, 2003
Macronutrient distribution
• American Heart Association-style
– < 30% fat
– 15% protein
– 55% CHO
• Very low-fat diet
– 10-15% fat
– 12-15% protein
– 72-75% CHO
Typical dinner--AHA-style
3-oz. Broiled salmon
2/3 cup rice
1/2 cup mixed vegetables
1 cup salad greens, 2 tsp.
Vinaigrette
1 slice whole wheat bread
1 tsp. butter
1/4 cantaloupe
Typical dinner--very low-fat
• 3-oz. baked whitefish
• 1-1/2 cups sauteed
collard greens
• baked sweet potato
• 1/2 cup wild rice pilaf
• baked apple
 Low-fat Diets: Weight Loss
• 2003 review by Pirozzo, et al
• Criteria for inclusion
– Randomized Ccontrolled Trial (RCT) of
Low- Fat vs. other wt loss diets
– primary purpose was wt. Loss
– Age > 18 years & BMI > 25 at baseline
– Followed for at least 6 months
– “Low fat diet” = < 30% of calories from fat
Pirozzo, S. et al. Cochrane Database of Systematic Reviews. 1, 2003.
Low-fat Diets: Weight Loss
• Four 6-months studies
• Five 12-month studies
• Three 18-months studies

Result:
• No significant differences compared to
other weight loss diets
 Low-fat Diets: Weight Loss
• Meta-analysis of 34 studies,
• 2-12 months duration
• ad lib low-fat diets vs. controls
• Result:
– 3.3 kg loss in low-fat groups (sig.)
• Main factors:
– degree of dietary fat reduction
– pre-treatment body weight
Astrup A, et al. Int J Obes Relat Met Dis. 2000 Dec; 24(12):1545
 Low-fat Diets:
Long-term Success
• Review by Ayyad & Anderson, 2000
• Criteria for inclusion:
– adults
– Follow up of > 3 years
– Follow up of > 50% of original study group

• Criteria for long-term success:

– Maintain > 9-11 kg or all weight initially lost
Ayyad C and Anderson T. Obesity Reviews, 2000; 1:113-9
 Low-fat Diets:
Long-term Success
• Overall, 15% of followed-up subjects met
criteria for success
– Diet alone = 15% met follow-up success
– Diet + group therapy = 27% met success
– Diet & behavior mod. = 14% met success
• Active follow-up better than passive for
long term success (19% vs. 10%)
 Low-fat diets vs. moderate/high-fat
diets
• 12 month duration
• 4 DIETS:
– low-fat(LF)
– moderate fat(MF)
– mod fat-isocaloric(MF-iso)
– high fat(HF)
• weight, lipids, CV risk
Fleming, RM. Preventive Cardiology. 5(3):110, 2002
RESULTS: Low-fat vs moderate
& high-fat diets
% body weight lost at 12 months
0%
-2%
-4%
NS
-6% LF
-8% MF
-10%
HF
-12%
MF-iso
-14%
-16%
-18%
1st Qtr
Wt loss associated with calories ONLY
 Low-fat diets and satiety
• Review of low-energy diets, glycemic index
(GI), and obesity
– Low-GI foods = Increased satiety
– low-fat, low-GI-based diets = promote satiety
and may therefore promote weight loss
– high-GI diets = weight gain and lower satiety
• Similar in animal studies

Brand-Miller et al. Am J Clin Nutr. 76(1):281S, 2002 July

Long-term Compliance/Maintenance

• McGuire et al
– random digit phone survey
– maintainers = lost > 10% of wt. , maintained for
> 1 yr.
• 69 Maintainers, 56 regainers, 113 wt-stable
controls

McGuire et al. Int J of Ob & Rel Metab Disord:

J of the Int Assoc Study of Ob. 23(12):1314, 1999 Dec
Long-term Compliance/Maintenance
Common factors for success:

• lower-fat intake than

regainers/controls
• behavior strategies used
more than regainers
• more physically active,
esp. strenuous activity
National Weight Control Registry

• Maintained a loss of >

25 lb. for > 1 year
• lost weight by
different methods
• Maintainers:
– low fat, modest calorie
intake
– Physically active 4-5
days/week
 LF Diets and Drop-outs
• Problem in many studies
– Most studies are 6 months or less
– Only count those who completed study
• Longer studies self-select
• Several studies do show compliance after 1
year
• Drop-out rate: 11-40% (Pirozzi, et al)
 Lipids & low-fat diets
• Low calorie AND low/moderate fat:
– significantly lower cholesterol, LDL-chol, TGs,
TC/HDL ratios
• Low-calorie, high-fat diets, OR mod fat,
isocaloric diets
– no significant changes:TG, TC, HDL, LDL-
chol
Fleming, RM. Preventive Cardiology. 5(3):110, 2002
 C-reactive protein (CRP) & low-
fat diets
• CRP strongly associated with BMI
• LF diet (15% fat, 62% CHO) -- 12 wks
• CRP significantly  with Low-Fat/low-
calorie diet and wt loss
• Other dietary composition not investigated
Heilbron et al. Arterioscl, Thromb & Vasc Biol. 21(6):968, 2001 June
 LDL & low-fat diets
• 2001 review by Krauss
• LDL-A = desirable large, fluffy LDL
• LDL-B = small, dense LDL (atherogenic)
• Conclusion:
– genes play a role
– LDL-B: benefits from a low-fat diet
– LDL-A: no benefits from low-fat diet IF isocaloric, &
may convert to LDL-B profile
– CAVEAT: studies are often short-term, not all hi-CHO
diets are alike

Krauss. J of Nutr. 131(2) 340S; 2001 Feb

 HDL & low-fat diets
• Meta-analysis of studies > 2months
• LF diet lowered HDL, less so when fed ad
lib than isoenergetically
• Epi studies consistently show populations
with LF diets and low-HDL have LOW
atherosclerosis
• Diets high in fruits, vegetables and grain
fiber promote less atherosclerosis, cancer
Jequier E and Bray G. Am J of Med. 113(9) supp. 2:41 2002 Dec
 Effect of type and amount of fat on
HDL/lipids
• Meta-analysis of 60 studies, Mensink et al
• Conclusion:
– Replacing fat with CHO isoenergetically may increase
total/HDL ratio
– Replacing fat with CHO and reducing calories could
have same effect as replacing SFA with MUFA and
PUFA
– Replacing trans-fat with MUFA/PUFA had best effect
on blood lipids
– Caveat: Results are general but the need is individual
Mensink, RP et al. AM J Clin Nutr. 77(5); 1146, 2003 May
 LF diets and insulin
• Review by Brand-Miller, et al
– Low-fat, hi-CHO, hi-GI diets may increase
insulin response
– BUT low-fat, low glycemic diets may promote
wt loss via increased satiety, lowered insulin
response
• Exercise improves insulin response
Brand-Miller et al. Am J Clin Nutr. 76(1):281S, 2002 July
CONCLUSIONS

• Calories count more

than fat: Low energy
diets are more effective at
producing weight loss
than low-fat diets

• BUT low-fat diets may

make it easier to reduce
total energy intake
 CONCLUSIONS
• Risks of Low-fat diets
– Difficult long-term compliance on very low-fat diet
• increases with close follow-up and group support
– type of CHO may affect outcome, possibly by altering
satiety
• Benefits
– Strong association with long-term maintenance
– LF, hi-CHO, hi-fiber diet reduces other health risks and
is usually rich in vitamins, minerals, and phytonutrients
 UNANSWERED QUESTIONS

• LONGER STUDIES NEEDED!

• Do metabolic parameters show adaptation
over time? Is long-term adaptation
different?
• Should physical activity and/or metabolic
profile affect recommendation of diet type?