Open access Viewpoint

Hiding unhealthy heart outcomes in a

Open Heart: first published as 10.1136/openhrt-2021-001680 on 21 July 2021. Downloaded from http://openheart.bmj.com/ on April 5, 2023 by guest. Protected by copyright.
low-­fat diet trial: the Women’s Health
Initiative Randomized Controlled
Dietary Modification Trial finds that
postmenopausal women with
established coronary heart disease were
at increased risk of an adverse outcome
if they consumed a low-­fat ‘heart-­
healthy’ diet
Timothy David Noakes

To cite: Noakes TD. Hiding ABSTRACT least in postmenopausal women with IR, especially T2DM.
unhealthy heart outcomes in a The Women’s Health Initiative Randomized Controlled According to the medical principle of ‘first do no harm’,
low-­fat diet trial: the Women’s this practice is now shown to be not evidence-­based,
Dietary Modification Trial (WHIRCDMT) was designed to
Health Initiative Randomized making it scientifically unjustifiable, perhaps unethical.
test whether the US Department of Agriculture’s 1977
Controlled Dietary Modification
Trial finds that postmenopausal Dietary Guidelines for Americans protects against coronary
women with established heart disease (CHD) and other chronic diseases. The
coronary heart disease were at only significant finding in the original 2006 WHIRCDMT
publication was that postmenopausal women with CHD
INTRODUCTION
increased risk of an adverse
outcome if they consumed a randomised to a low-­fat ‘heart-­healthy’ diet in 1993 The Women’s Health Initiative Randomized
low-f­ at ‘heart-­healthy’ diet. were at 26% greater risk of developing additional CHD Controlled Dietary Modification Trial
Open Heart 2021;8:e001680. events compared with women with CHD eating the (WHIRCDMT)1 is one of the most expensive
doi:10.1136/ control diet. A 2017 WHIRCDMT publication includes data long-­term dietary intervention trials yet under-
openhrt-2021-001680
for an additional 5 years of follow-­up. It finds that CHD taken. Beginning in 1993, the WHIRCDMT
risk in this subgroup of postmenopausal women had was designed to provide supporting evidence
Accepted 14 June 2021 increased further to 47%–61%. The authors present three for a single dietary pattern, consistent with
post-­hoc rationalisations to explain why this finding is the US Department of Agriculture’s 1977
‘inadmissible’: (1) only women in this subgroup were less Dietary Guidelines for Americans (DGA),
likely to adhere to the prescribed dietary intervention;
which encouraged North Americans to
(2) their failure to follow the intervention diet increased
their CHD risk; and (3) only these women were more
reduce their dietary, especially saturated, fat
likely to not have received cholesterol-­lowering drugs. intake.2 This specific dietary intervention had
These rationalisations appear spurious. Rather these yet to be evaluated with respect to its effects
findings are better explained as a direct consequence of on weight gain and the development of coro-
© Author(s) (or their postmenopausal women with features of insulin resistance nary heart disease (CHD), cancer and type 2
employer(s)) 2021. Re-­use (IR) eating a low-­fat high-­carbohydrate diet for 13 years. diabetes mellitus (T2DM).
permitted under CC BY-­NC. No All the worst clinical features of IR, including type 2 The goal of the intervention diet was to
commercial re-­use. See rights diabetes mellitus (T2DM) in some, can be ‘reversed’ replace especially saturated fat intake with
and permissions. Published by the prescription of a high-­fat low-­carbohydrate diet.
by BMJ. an increased intake of carbohydrates from
The Women’s Health Study has recently reported that grains, fruits and vegetables. The dietary
Applied Design, Cape Peninsula T2DM (10.71-­fold increased risk) and other markers of
University of Technology, intervention effectively lowered dietary fat
IR including metabolic syndrome (6.09-­fold increased
Bellville, South Africa risk) were the most powerful predictors of future CHD intake and was associated with a reduction
development in women; blood low-­density lipoprotein-­ in blood cholesterol concentrations. Impor-
Correspondence to
cholesterol concentration was a poor predictor (1.38-­fold tantly the goal of the dietary intervention was
Professor Timothy David
Noakes; ​timothy.​noakes@​uct.​ increased risk). These studies challenge the prescription not to replace dietary saturated fat intake
ac.z​ a of the low-­fat high-­carbohydrate heart-­healthy diet, at with an increased intake of polyunsaturated

Noakes TD. Open Heart 2021;8:e001680. doi:10.1136/openhrt-2021-001680   1

Open Heart

fats. This particular intervention had been evaluated in T2DM.14 A prior meta-­analysis found only a 9% increased

Open Heart: first published as 10.1136/openhrt-2021-001680 on 21 July 2021. Downloaded from http://openheart.bmj.com/ on April 5, 2023 by guest. Protected by copyright.
two other trials, the final results of which were ultimately risk of T2DM associated with statin use.15
published in 20133 and 2016.4 In contrast to these neutral or negative findings, the
During the WHIRCDMT, only postmenopausal most important discovery in the 2006 report was essen-
women randomised to the intervention group received tially dismissed as unreliable: postmenopausal women
the programme: an ‘intensive behavioral modification who entered the trial with established CHD and who
program involved 18 group sessions in the first year were randomised to the intervention diet in 1993 were
and quarterly maintenance sessions thereafter, led by at 26% increased risk of an adverse outcome compared
specially trained and certified nutritionists’ (p656).1 In with those women with CHD who continued eating their
addition, ‘group activities were supplemented during the usual ‘high’-fat control diet. This was the sole outcome
intervention period by individual interviews…targeted-­ that reached statistical significance.
message campaigns, and personalized feedback on fat I have argued7 16 that this finding may not have been
intake’ (p657).1 In contrast, ‘women in the comparison properly communicated.
group received a copy of the DGA, as well as other health-­ Here I review the two most recent 20175 and 20196
related materials, but had no contact with the nutrition publications from the WHIRCDMT which report addi-
interventionists’ (p657).1 tional findings from a further 5 years of follow-­up of the
In this article I review two more recent publications population originally described in 2006.1 These publica-
from the WHIRCDMT5 6 which show that compared with tions are important because they provide an analysis of
postmenopausal women who continued to eat the more the effects of a more prolonged duration of exposure to
usual, higher-­ fat, supposedly heart-­ unhealthy control the low-­fat heart-­healthy intervention diet.
diet, postmenopausal women randomised to the interven-
tion diet were at 47%–61% increased risk of developing
Findings of the 2017 WHIRCDMT publication
additional CHD complications during a further 5 years of
The 2017 report describing the 13-­year follow-­up data for
follow-­up. The authors provide three post-­hoc rationalisa-
the WHIRCDMT5 introduced a novel subgroup analysis
tions to explain why their latest finding is ‘inadmissible’. I
based on the health of the postmenopausal women on
argue that none of these rationalisations is valid. I further
admission to the WHIRCDMT in 1993. Study participants
propose that this iconic study definitively establishes that
were categorised into three subgroups based on their
the prescription of the low-­fat ‘heart-­healthy’ diet to post-
health status in 1993 when they entered the trial:
menopausal women with established CHD, because they
1. No CHD or hypertension (HTN).
are likely to be insulin-­resistant, is scientifically unjustifi-
2. HTN only.
able and potentially unethical.
3. Pre-­existing CHD.
This categorisation allows for the identification of
WHAT ARE THE TRUE FINDINGS OF THE WHICRCDMT? specific subgroups who may either benefit the most
Findings of the original 2006 publication from the WHIRCDMT or be exposed to the greatest harm from the dietary
As detailed previously,7 the original publication describing intervention.
the first 8 years of the WHIRCDMT failed to identify even This subgroup analysis confirmed that the risk of
one statistically significant beneficial outcome of this developing additional CHD complications during the
dietary intervention. The low-­ fat dietary intervention extended follow-­ up in the group of postmenopausal
failed to protect against invasive breast8 or colorectal9 women with pre-­existing CHD had increased from 26%
cancer and produced only a marginal (0.4 kg) weight in the first analysis1 to 47%–61% 5 years later (figure 1
loss over the first 8 years of the trial.10 In contrast, strict and 2 in reference5) if they were assigned to the inter-
adherence to low-­fat and DGA diets was associated with vention diet. Postmenopausal women with HTN in 1993
increased risk of weight gain, whereas strict compliance received neither overall benefit nor harm if they ate the
with a higher-­fat reduced-­carbohydrate diet ‘was associ- intervention diet (figure 1 and 2 in reference5). Healthy
ated with a sharply lower risk of weight gain in adjusted postmenopausal women with neither CHD nor HTN in
models…’ (p1191).11 ‘Our findings therefore challenge 1993 received some benefit in terms of a small reduction
prevailing dietary recommendations, suggesting instead in CHD risk but at the cost of an increased risk of stroke
that a low fat (diet) may promote rather than prevent (figure 1 and 2 in reference5).
weight gain after menopause’ (p1196).11 Accordingly these data indicate that the examined
While the heart-­healthy low-­fat dietary intervention did heart-­healthy intervention diet substantially worsened
not reduce the risk of developing T2DM during the first outcomes in postmenopausal women with established
8 years of the trial,12 already within the first year, women CHD while providing only a marginal benefit for those
who began the experiment with T2DM showed signifi- who are the most healthy because they had neither CHD
cantly worsened (p<0.001) blood glucose control.13 nor HTN when the trial began. While the authors of
An unexpected finding was that postmenopausal the 2017 publication5 acknowledge this increased risk
women prescribed cholesterol-­ lowering medications of adverse CHD outcomes in those with prior CHD, this
(statins) were at 49% increased risk of developing finding cannot be dismissed as simply due to chance.

2 Noakes TD. Open Heart 2021;8:e001680. doi:10.1136/openhrt-2021-001680

 Viewpoint

The conclusion in the abstract also fails to mention any to the original hypothesis being tested). The possibility

Open Heart: first published as 10.1136/openhrt-2021-001680 on 21 July 2021. Downloaded from http://openheart.bmj.com/ on April 5, 2023 by guest. Protected by copyright.
adverse outcomes for those with prior CHD eating the that the heart-­healthy intervention diet could be harmful
intervention diet for 13 years: rather than healthy was simply unimaginable when the
WHIRCDMT was planned. The decades-­long history of
Conclusions: CVD risk in postmenopausal women
how this came about has been detailed by Teicholz,17 as
appears to be sensitive to a change to a low-­fat dietary
also by Noakes and Sboros.18
pattern and among healthy women, including both
However, the authors’ rationalisation is moot since the
CHD benefit and stroke risk (p35).5
WHIRCDMT was designed and analysed as an intention-­
As a result and as was the case in the 2006 report,1 7 to-­treat trial. As the authors describe the WHIRCDMT
these conclusions fail to emphasise that women with CHD design: ‘Design: This randomized controlled trial was
at the start of the trial were at a substantially increased analyzed as intent to treat’ (p260).19
risk of additional cardiovascular events if they adopted The intention-­ to-­
treat analysis is defined as the
the heart-­healthy intervention diet. following: ‘A method for analyzing results in a prospec-
tive randomized study where all participants who are
Post-hoc rationalisations to explain why findings of harm randomized are included in the statistical analysis and
are‘uninterpretable’ analyzed according to the group (to which) they were
Rationalisation 1 originally assigned, regardless of what treatment (if any)
Women in the dietary intervention group failed to comply they received’.20
with the required dietary change to a low-­fat diet. In summary, since the WHIRCDMT was designed as an
The main explanation offered by the authors to de-­em- intention-­to-­treat trial, this attempted post-­hoc rationali-
phasise the importance of the statistically significant sation is itself inadmissible.
findings of potential harm of the intervention diet in
those with prior CHD appears to be based on concerns Rationalisation 2
regarding dietary compliance: The existence of what the authors describe as ‘other
contexts’, in particular the Estrogen Replacement and
We concluded that the trial results for CHD were un-
Atherosclerosis (ERA) trial of progression of coronary
interpretable in the prior CVD subjects (i.e., the find-
artery narrowing in postmenopausal women eating diets
ing showing increased CHD risk in women with established
with different macronutrient compositions.
CHD randomized to the “heart-­ healthy” intervention in
The authors submit an additional ‘other contexts’21–23
1993 – my addition). We were not able to rule out the
argument to support their post-­hoc rationalisation. Of
possibility that dietary changes in the intervention group
these three references, two refer to original data collected
participants (my emphasis) could have contributed to
as part of the ERA trial.21 22 The third23 is a character
their unfavorable CHD experience. Others have hy-
reference for the heart-­healthiness of the heart-­healthy
pothesized an unfavorable CHD effect based on stud-
intervention diet.
ies in other contexts.17–19 (p41)5
The ERA trial21 found that coronary atherosclerosis did
The logic of this post-­hoc rationalisation seems to be not progress in postmenopausal women who reported
the following: Women without CHD assigned to the heart-­ that they ate the most saturated fat during a 3-­ year
healthy low-­fat dietary intervention in 1993 had adhered observational trial to determine the effects of hormone
scrupulously to that diet. However, when randomised to replacement therapy on the progression of coronary
the identical diet, another group of women, differing only atherosclerosis (arrow 1 in figure 1), whereas those
because they started the trial with CHD, failed to comply women whose diets contained either more carbohydrates
with that same diet, producing results that are now ‘unin- or more polyunsaturated fats (and therefore less satu-
terpretable’ (p41).5 rated fat) showed progression of coronary atherosclerosis
While the harms of the intervention diet were apparent (arrows 2 and 3 in figure 1).
in the original 2006 publication,1 7 questions about Figure 1 shows that the highest rates of progression of
dietary compliance were not raised at that time. Rather coronary artery narrowing occurred in postmenopausal
the authors chose to dismiss the finding as most prob- women in the highest quartile of intake of either poly-
ably due to chance: ‘The intervention was associated with unsaturated fats (arrow 2) or carbohydrates (arrow 3).
increased risk in the 3.4% of women with baseline CVD; However, women in the highest quartile of saturated fat
this may be a chance observation, or rates in this small intake showed a modest regression of coronary artery
subset may be confounded by concurrent therapy or narrowing (arrow 1).
comorbid conditions’ (p663–664).1 Thus the findings of the ERA trial predict that those
This establishes that there was no hint in the original eating less fats and especially less saturated fats and more
article1 that some women with prior CHD assigned to the carbohydrates (and more polyunsaturated fats if they so
heart-­healthy intervention diet in 1993 had not complied chose) would experience a more rapid progression of
with the experimental diet, reverting rather to their coronary artery narrowing.
previous heart-­ unhealthy high-­ fat control diet and so So, correctly interpreted, the results of the ERA trial
increasing their risk for further CHD events (according indicate those postmenopausal women with prior CHD

Noakes TD. Open Heart 2021;8:e001680. doi:10.1136/openhrt-2021-001680 3

Open Heart

Open Heart: first published as 10.1136/openhrt-2021-001680 on 21 July 2021. Downloaded from http://openheart.bmj.com/ on April 5, 2023 by guest. Protected by copyright.
Figure 1  Changes in mean minimal coronary artery
diameter measured in postmenopausal women participating
in the Estrogen Replacement and Atherosclerosis trial.21
Note that increased rates of coronary artery narrowing were
associated with increasing intake of polyunsaturated fats Figure 2  Percentage of postmenopausal women in the
(arrow 2) and carbohydrates (arrow 3). The highest intake three different subgroups who were using statin drugs at the
of saturated fat was associated with a slight regression start (year 0) and end (year 6) of WHIRCDMT. Figure drawn
of coronary artery narrowing (arrow 1). Redrawn and from data in figure 3 in Prentice et al.5 CHD, coronary heart
reproduced from Mozaffarian et al21 with permission from the disease; WHIRCDMT, Women’s Health Initiative Randomized
American Journal of Clinical Nutrition. Controlled Dietary Modification Trial.

would be more likely to experience progression of their the dietary intervention group were confounded by
coronary artery narrowing when eating the intervention ‘postrandomization use of cholesterol-­lowering medi-
rather than the control diet. Importantly, this interpreta- cations’ (p35).5 The clear assumption is that statin use
tion is consistent with and not discordant from the find- in women is associated with a significant reduction
ings of the WHIRCDMT. in CHD risk,24 a claim that is contested especially in
The logic of Prentice et al’s5 explanation must be that
women.25
some women with prior CHD failed to comply with the
Prentice et al’s5 explanation appears to be that post-
heart-­healthy intervention diet. Instead they reverted to
menopausal women in the intervention group who failed
their usual heart-­unhealthy control diet containing too
to comply with the dietary advice and instead continued
much dietary fat, especially saturated fat. This change,
they must claim, would have caused CHD to progress in to consume higher levels of fat and saturated fat would
these women, whereas some other women in the same be more likely to be prescribed and to comply with statin
diet intervention group who faithfully followed the therapy. This assertion is however unsupported by base-
heart-­healthy intervention diet presumably remained line and follow-­up data from the WHIRCDMT, which in
disease-­free. fact demonstrate a higher level of statin use in women
This explanation is only logical if the ERA study found with prior CHD randomised to the dietary intervention
that diets high in carbohydrates and low in saturated fats (figure 2).
prevented progression of coronary artery narrowing. Thus figure 2 shows that the subgroup of 1656 post-
However, the ERA study found the opposite (figure 1). menopausal women with prior CHD in 1993 were ~15%–
20% more likely to be prescribed statin drugs during the
Rationalisation 3 trial and follow-­up than were postmenopausal women
Women with prior CHD randomised to the heart-­healthy in the other two subgroups. However, the percentage of
intervention diet in 1993 were less likely to be prescribed
statin users in women with prior CHD was essentially the
statin drugs during the trial and follow-­up than were
same at all times during the trial regardless of whether
women in the respective control group. Since, according
they were assigned to either the intervention or control
to this logic, statins reduce CHD events, CHD data from
the intervention group are inadmissible. diet (figure 2).
The third rationalisation used to explain why follow-­up This finding that postmenopausal women with CHD
data for postmenopausal women with CHD in 1993 in 1993 were equally likely to be taking statin drugs
randomised to the intervention diet are inadmissible is regardless of their dietary assignment seems to disprove
the claim that such women were less likely to be prescribed the authors’ claim that ‘postrandomization use of
cholesterol-­ lowering statin drugs that, it is claimed, cholesterol-­lowering medications’ invalidates the find-
protect against future CHD events. ings of increased CHD events in women with prior CHD
The claim is that the 13-­year follow-­up data of post- randomised to the low-­fat heart-­healthy intervention diet
menopausal women with prior CHD randomised to (p35).5

4 Noakes TD. Open Heart 2021;8:e001680. doi:10.1136/openhrt-2021-001680

 Viewpoint

Post-hoc rationalisations should not be used to explain data that 6. Percentage use of statins was equivalent in postmeno-

Open Heart: first published as 10.1136/openhrt-2021-001680 on 21 July 2021. Downloaded from http://openheart.bmj.com/ on April 5, 2023 by guest. Protected by copyright.
challenge a favoured hypothesis pausal women with CHD in 1993 randomised to ei-
These post-­hoc rationalisations introduced by Prentice et ther the control or intervention diet. In both groups
al5 arise perhaps from an abiding, largely unchallenged >40% of participants were prescribed statin drugs.
certainty in the safety and efficacy of the low-­fat heart-­ The finding that risk of future CHD events was great-
healthy diet.17 Thus, ‘the results from the WHI diet trial er in the group receiving the low-­fat dietary interven-
taken as a whole are consistent with our current under- tion, despite high rates of statin use, proves that statin
standing of the major dietary components that influence use did not eliminate and may not have lessened the
cardiovascular disease’ (p281).26 However, this statement increased risk of future CHD events associated with
is not supported by the extensive scientific evidence the eating the heart-­healthy low-­fat diet.
authors have provided. 7. Postmenopausal women in either dietary interven-
Rather the WHIRCDMT disproved the hypothesis it tion group who were prescribed statin drugs were at
was designed to test. Once disproven, the tested hypoth- 49% increased risk of developing T2DM.
esis must be summarily rejected; it cannot be rescued 8. At the end of the first year of the trial, postmenopaus-
by post-­hoc rationalisations of convenience. As Stephen al women with T2DM at the start of the trial in 1993
Hawking wrote, ‘if the observations don’t agree with the showed worsened glucose control if they were ran-
theory, one abandons the theory’ (p36).27 domised to the low-­fat dietary intervention.
9. Postmenopausal women who complied strictly with a
Findings of the 2019 WHIRCDMT publication personally chosen, higher-­fat, reduced-­carbohydrate
In their most recent6 2019 publication, the authors diet showed a ‘sharply lower’ risk of weight gain
chose to report the results only for postmenopausal during the trial.
women who entered the trial without CHD in 1993. By 10. The most important practical finding of the
excluding women for whom the diet has now been shown WHIRCDMT was that only those postmenopausal
to be harmful, the authors were able to conclude that women who are the healthiest because they have nei-
‘reduction in dietary fat with corresponding increase ther CHD nor HTN can be reassured that eating the
in vegetables, fruit, and grains led to benefits related to heart-­healthy DGA intervention diet will not cause
breast cancer, CHD and diabetes, without adverse effects, long-­term cardiovascular harm and may instead pro-
among healthy post-­menopausal US women’ (p1565).6 vide some benefit.
This statement fails to warn that this same diet produced
measurable harm in unhealthy women with established
CHD in 1993. FINDINGS FROM THE WOMEN’S HEALTH STUDY
T2DM and other markers of insulin resistance as predictors of
What are the current findings of the WHIRCDMT? future CHD risk
The reports of the findings of the WHIRCDMT could The Women’s Health Study (WHS), also established
and perhaps should have emphasised the following: between 1992 and 1995 at Harvard Medical School, was
1. Compared with the experience of women with CHD designed as a clinical trial to evaluate the effects of vitamin
in 1993 who ate a diet with more fats, including satu- E28 or low-­dose aspirin29 30 on the risk of developing CHD
rated fats, postmenopausal women eating the heart-­ or cancer in initially healthy women free from cardiovas-
healthy low-­fat intervention diet were at 47%–61% cular disease and cancer at baseline. These studies found
increased risk of developing additional CHD events no overall benefit for either intervention. A subsequent
during 13 years of follow-­up. 21.4-­year-­long, prospective follow-­up cohort study of 28
2. This represented a substantial increase in adverse 024 of these women31 has evaluated more than 50 clin-
events for women in the intervention group, com- ical, lipid, inflammatory and metabolic risk factors and
pared with the findings 5 years earlier, when the risk biomarkers for the subsequent development of CHD.
difference (26%) between the two study groups was T2DM has long been considered by some32 to be a
substantially less. more important risk factor in the pathogenesis of CHD
3. This risk difference occurred even though only the than elevated blood cholesterol concentrations,33 34 the
intervention group had received the intensive be- latter being the hypothesis tested by the WHIRCDMT.
havioural modification programme led by especially The results of the 21.4-­year-­long, prospective follow-­up
trained and certified nutritionists during the first 8 WHS31 confirm that T2DM, not an elevated blood choles-
years of the trial. terol concentration, is the key driver of future CHD
4. Postmenopausal women with HTN but without CHD development. figure 3 shows the most important results
in 1993 received neither benefit nor harm from eat- of this part of the WHS.
ing the heart-­healthy intervention diet for 13 years. Figure 3 shows that the strongest predictors of future
5. Postmenopausal women with neither HTN nor CHD CHD development in these postmenopausal women are
in 1993 received some benefits from eating the heart-­ all the classic clinical markers of insulin resistance (IR),
healthy intervention diet for 13 years but at the cost most especially T2DM (10.7-­fold increased risk), meta-
of an increased risk of stroke. bolic syndrome (6.09-­fold increased risk), HTN (4.58-­fold

Noakes TD. Open Heart 2021;8:e001680. doi:10.1136/openhrt-2021-001680 5

Open Heart

Open Heart: first published as 10.1136/openhrt-2021-001680 on 21 July 2021. Downloaded from http://openheart.bmj.com/ on April 5, 2023 by guest. Protected by copyright.
Figure 3  HR for the six most important risk factors and the six biochemical markers for the development of CHD in 28 024
postmenopausal women who were healthy on entry to the Women’s Health Study. Drawn from data from Dugani et al.31 CHD,
coronary heart disease; HDL, high-­density lipoprotein; LDL, low-­density lipoprotein.

increased risk) and obesity (4.33-­fold increased risk). All More evidence that abnormalities in carbohydrate
these markers of IR had greater predictive value for CHD rather than in fat metabolism drive coronary athero-
than did smoking. sclerosis comes from the Progression of Early Subclin-
In addition the most important metabolic risk marker ical Atherosclerosis study,47 which found an association
was the Lipoprotein Insulin Resistance (LPIR) score35 between haemoglobin A1c (HbA1c) values and subclin-
(6.40-­fold increased risk). The LPIR score is based on ical atherosclerosis in persons without T2DM. The rela-
lipoprotein subclass and size information measured tionship was present at all HbA1c values, even at values
with nuclear magnetic resonance (NMR) imaging. It has below 5.5%, the level at which pre-­diabetes is usually first
strong associations with multiple markers of IR and ‘may diagnosed.
represent a simple means to identify individuals with IR’ Indeed evidence disputing the traditional diet-­ heart
(p422).35 The score is based on studies showing changes and lipid hypotheses, now seemingly also disproven by
in lipoproteins in persons with IR. In particular, those the WHIRCDMT, continues to accumulate.48–59
with IR show the following characteristic NMR lipopro-
tein patterns35–38:
1. Greater number of the large subclass of very low-­ WHY DID ONLY THE MOST HEALTHY POSTMENOPAUSAL
density lipoprotein (VLDL) particles. WOMEN BENEFIT FROM EATING THE HEART-HEALTHY DGA
2. Greater number of the small subclass of low-­density li- DIET IN THE WHIRCDMT?
poprotein (LDL) particles. A fundamental question is: Why did only the healthiest
3. Lower number of the large subclass of high-­density li- postmenopausal women in the WHIRCDMT receive
poprotein (HDL) particles. some benefit from eating the heart-­healthy intervention
4. In addition, mean VLDL particle sizes are generally diet?
larger and mean LDL and HDL particle sizes usually One possibility is that persons with higher levels of IR
smaller in persons with IR36 37 or pre-­diabetes.39 expressed clinically as metabolic syndrome, HTN or T2DM
In contrast serum LDL-­cholesterol concentration—the may be more likely to show reversal of some or all of the
principal target of the low-­fat heart-­healthy intervention metabolic features of these conditions if they avoid the
diet in the WHIRCDMT1 because the Women’s Health heart-­healthy low-­fat intervention diet.40–46 Historically the
Initiative (WHI) principal investigators consider it to be association between IR and ‘essential’ HTN,60–66 obesity,67–69
the determinant of CHD risk33 34—was of little predictive T2DM,60 endothelial dysfunction70 and CHD60 63 65 71–76 is
value (1.38-­fold increased risk) (figure 3). well established. The clinical and metabolic characteristics
Accordingly a low-­ fat diet, which may indeed lower of all these conditions improve and can be ‘reversed’ in
blood LDL-­cholesterol concentrations but at the cost of some individuals when a low-­carbohydrate, high-­fat heart-­
an increasing atherogenic dyslipidaemia,40–42 especially unhealthy diet is eaten.43–46 77–92
in those with IR,43–46 would be expected to worsen CHD A reasonable suggestion might be that the postmeno-
outcomes, precisely as happened in the subgroup of post- pausal women in the WHIRCDMT who were not harmed
menopausal women with prior CHD in the WHIRCDMT. by eating the low-­ fat high-­carbohydrate intervention

6 Noakes TD. Open Heart 2021;8:e001680. doi:10.1136/openhrt-2021-001680

 Viewpoint

diet were insulin-­ sensitive in 1993 and remained so increased the intake of carbohydrates, not of fats which

Open Heart: first published as 10.1136/openhrt-2021-001680 on 21 July 2021. Downloaded from http://openheart.bmj.com/ on April 5, 2023 by guest. Protected by copyright.
during the trial and follow-­up. However, postmenopausal could have been contaminated with trans fats.
women with either HTN or CHD in 1993 likely had more In reviewing all the current evidence, Lawrence101
advanced IR, placing them at risk of worsening IR if they concludes that:
continued to eat the low-­fat high-­carbohydrate diet.
PUFAs are unstable to chemical oxidation and their
Interestingly a 2015 study of the WHI population93
oxidation products are harmful in a variety of ways.
does not strongly support this interpretation, as this study
PUFAs also form powerful signaling agents that can
found that, when corrected for blood HDL-­cholesterol
initiate inflammation which can have dire health
concentrations, markers of IR were not a significant
consequences…If saturated fats are replaced by car-
predictor of future CHD risk in postmenopausal women
bohydrates in the diet, there would be no significant
free of T2DM in 1993.
improvement in serum cholesterol, and it can result
However a low blood HDL-­cholesterol concentration is
in a more atherogenic lipoprotein profile. When
a key marker of IR,60 so that correcting for this biomarker
looking at much of the data in the context of known
removes, in part, the influence of IR as a contributor to
biochemical and physiological mechanisms, it ap-
CHD risk.
pears that saturated fats are less harmful than the
However, IR in this population was associated with
common alternatives .101
higher breast cancer incidence and all-­cause mortality
after breast cancer,94 as well as increased risk of cancer-­
specific and all-­cause mortality.95
DIETARY ADVICE FOR PERSONS WITH IR OR T2DM
This set of findings from four different studies effectively
EVIDENCE THAT REPLACING DIETARY SATURATED FAT WITH ends the debate about which diet should be eaten to
POLYUNSATURATED FATS ALSO CAUSES HARM lower the risk of CHD, especially in those with IR.
The significance of the Dugani et al31 study is to show, as The answer is that the prescribed diet must prevent the
Kraft first proposed,96 that T2DM/IR is the single most development of the clinical features of IR leading to T2DM.
important risk factor, by far, for future development of The two diets shown to achieve this are the restricted low-­
CHD. However, as he argued, it is often missed because calorie diet developed by Lim et al102 and the ad libitum low-­
of inappropriate testing to detect either condition.97–99 carbohydrate higher-­healthy-­fat ketogenic diet as reported
While these new data are specific to postmenopausal by a number of research teams.79–83 86–88 90
women, it is reasonable to assume that they also apply to According to the principle of first do no harm, it now
men (and women) of all ages. If correct, it follows that becomes the ethical responsibility of all those managing
advising persons with IR to replace dietary saturated fat persons with established CHD or at risk of its development
with heart-­healthy carbohydrates from fruits, vegetables because they have IR, especially if they have T2DM, not to
and grains will worsen blood glucose control especially prescribe the never-­proven17 18 103 and now-­disproven low-­
in those with IR and T2DM43–46 and produce a proath- fat heart-­healthy DGA diet.
erogenic dyslipidaemia,40–42 while increasing whole body
inflammation and IR.42 These changes would be expected Twitter Timothy David Noakes @ProfTimNoakes
to lead inexorably to progression of CHD. Acknowledgements  The Cape Peninsula University of Technology contributes to
Two other studies have also recently shown that the costs of the submission and publication of this article. The author covered all
replacing saturated fat with polyunsaturated fat according other expenses.
to the DGA dietary guidelines also worsens long-­term Contributors  TDN wrote the manuscript without assistance from anyone else.
outcomes. Funding  The Cape Peninsula University of Technology contributes to the costs of
The Recovered Minnesota Coronary Experiment the submission and publication of this article. TDN covered all other expenses.
(RMCE)3 found that persons randomised to the inter- Competing interests  TDN is the author of a number of books on low-­
vention diet which replaced saturated fat with the poly- carbohydrate diet, including The Real Meal Revolution, Super Food for
Superchildren, Lore of Nutrition, The Banting Pocket Guide, Real Food on Trial and
unsaturated fatty acid (PUFA), linoleic acid, were at 22% The Eat Right Revolution. TDN derives no personal income from the sale of these
higher risk of death for each 30 mg/dL (0.78 mmol/L) books. Instead all royalties are donated to the NGO The Noakes Foundation, of
reduction in blood cholesterol concentrations, an effect which TDN is the chairman. The money is used to fund the work of The Noakes
that was especially apparent in those over 65 years of Foundation, including the Eat Better South Africa Campaign.
age. The Recovered Sydney Diet Heart Study (RSDHS)4 Patient consent for publication  Not required.
also found that replacement of dietary saturated fat with Provenance and peer review  Not commissioned; externally peer reviewed.
linoleic acid was also associated with increased all-­cause Data availability statement  The publication makes use of data already published
mortality and with increased deaths from both cardiovas- in the scientific literature.
cular disease and CHD. Open access  This is an open access article distributed in accordance with the
Importantly one criticism of the RMCE and the RSDHS Creative Commons Attribution Non Commercial (CC BY-­NC 4.0) license, which
permits others to distribute, remix, adapt, build upon this work non-­commercially,
is that neither controlled for the intake of trans fats and license their derivative works on different terms, provided the original work is
considered to increase CHD risk.100 This criticism does properly cited, appropriate credit is given, any changes made indicated, and the use
not apply to the WHIRCDMT since the intervention is non-­commercial. See: http://​creativecommons.​org/​licenses/​by-​nc/​4.​0/.

Noakes TD. Open Heart 2021;8:e001680. doi:10.1136/openhrt-2021-001680 7

 Open Heart

REFERENCES 27 Hawking S. Black holes and baby universes and other essays. UK:

Open Heart: first published as 10.1136/openhrt-2021-001680 on 21 July 2021. Downloaded from http://openheart.bmj.com/ on April 5, 2023 by guest. Protected by copyright.
1 Howard BV, Van Horn L, Hsia J, et al. Low-­fat dietary pattern Bantam Books London, 1993.
and risk of cardiovascular disease: the women's health 28 Lee I-­M, Cook NR, Gaziano JM, et al. Vitamin E in the primary
Initiative randomized controlled dietary modification trial. JAMA prevention of cardiovascular disease and cancer: the women's
2006;295:655–66. health study: a randomized controlled trial. JAMA 2005;294:56–65.
2 Select Committee on Nutrition and Human Needs of the United 29 Cook NR, Lee I-­M, Gaziano JM, et al. Low-­dose aspirin in the
States Senate. Dietary goals for the United States. 2nd edn. primary prevention of cancer: the women's health study: a
Washington, DC: US Government Printing Office, 1977. randomized controlled trial. JAMA 2005;294:47–55.
3 Ramsden CE, Zamora D, Leelarthaepin B, et al. Use of dietary 30 Ridker PM, Cook NR, Lee I-­M, et al. A randomized trial of low-­dose
linoleic acid for secondary prevention of coronary heart disease aspirin in the primary prevention of cardiovascular disease in
and death: evaluation of recovered data from the Sydney diet heart women. N Engl J Med 2005;352:1293–305.
study and updated meta-­analysis. BMJ 2013;346:e8707. 31 Dugani SB, Moorthy MV, Li C, et al. Association of lipid, inflammatory,
4 Ramsden CE, Zamora D, Majchrzak-­Hong S, et al. Re-­evaluation and metabolic biomarkers with age at onset for incident coronary
of the traditional diet-­heart hypothesis: analysis of recovered heart disease in women. JAMA Cardiol 2021;6:437–47.
data from Minnesota coronary experiment (1968-73). BMJ 32 Noakes TD. The 2012 University of Cape Town Faculty of Health
2016;353:i1246. Sciences centenary debate. South African Journal of Clinical
5 Prentice RL, Aragaki AK, Van Horn L, et al. Low-­fat dietary Nutrition 2015;28:19–33. doi:10.1080/16070658.2015.11734522
pattern and cardiovascular disease: results from the women's 33 Rossouw J. Serum cholesterol as a risk factor for coronary heart
health initiative randomized controlled trial. Am J Clin Nutr disease revisited. S Afr J Clin Nutr 2015;28:34–7.
2017;106:35–43. 34 Rossouw J. The diet-­heart hypothesis, obesity and diabetes. S Afr
6 Prentice RL, Aragaki AK, Howard BV, et al. Low-­fat dietary J Clin Nutr 2015;28:38–43.
pattern among postmenopausal women influences long-­term 35 Shalaurova I, Connelly MA, Garvey WT, et al. Lipoprotein insulin
cancer, cardiovascular disease, and diabetes outcomes. J Nutr resistance index: a lipoprotein particle-­derived measure of insulin
2019;149:1565–74. resistance. Metab Syndr Relat Disord 2014;12:422–9.
7 Noakes TD. The women's health Initiative randomized controlled 36 Garvey WT, Kwon S, Zheng D, et al. Effects of insulin resistance
dietary modification trial: an inconvenient finding and the diet-­heart and type 2 diabetes on lipoprotein subclass particle size and
hypothesis. S Afr Med J 2013;103:824–5. concentration determined by nuclear magnetic resonance. Diabetes
8 Prentice RL, Caan B, Chlebowski RT, et al. Low–fat dietary 2003;52:453–62.
pattern and risk of invasive breast cancer: the women's health 37 Goff DC, D'Agostino RB, Haffner SM, et al. Insulin resistance and
Initiative randomized controlled dietary modification trial. JAMA adiposity influence lipoprotein size and subclass concentrations.
2006;295:629–42. results from the insulin resistance atherosclerosis study.
9 Beresford SAA, Johnson KC, Ritenbaugh C, et al. Low-­fat Metabolism 2005;54:264–70.
dietary pattern and risk of colorectal cancer: the women's health 38 Reaven GM, Chen YD, Jeppesen J, et al. Insulin resistance and
Initiative randomized controlled dietary modification trial. JAMA hyperinsulinemia in individuals with small, dense low density
2006;295:643–54. lipoprotein particles. J Clin Invest 1993;92:141–6.
10 Howard BV, Manson JE, Stefanick ML, et al. Low-­fat dietary pattern 39 Festa A, Williams K, Hanley AJG, et al. Nuclear magnetic resonance
and weight change over 7 years: the women's health Initiative lipoprotein abnormalities in prediabetic subjects in the insulin
dietary modification trial. JAMA 2006;295:39–49. resistance atherosclerosis study. Circulation 2005;111:3465–72.
11 Ford C, Chang S, Vitolins MZ, et al. Evaluation of diet pattern and 40 Siri-­Tarino PW, Sun Q, Hu FB, et al. Saturated fat, carbohydrate,
weight gain in postmenopausal women enrolled in the women's and cardiovascular disease. Am J Clin Nutr 2010;91:502–9.
health Initiative observational study. Br J Nutr 2017;117:1189–97. 41 Volk BM, Kunces LJ, Freidenreich DJ, et al. Effects of step-­wise
12 Tinker LF, Bonds DE, Margolis KL, et al. Low-­fat dietary pattern increases in dietary carbohydrate on circulating saturated fatty
and risk of treated diabetes mellitus in postmenopausal women: acids and palmitoleic acid in adults with metabolic syndrome. PLoS
the women's health Initiative randomized controlled dietary One 2014;9:e113605.
modification trial. Arch Intern Med 2008;168:1500–11. 42 Kuipers RS, de Graaf DJ, Luxwolda MF, et al. Saturated
13 Shikany JM, Margolis KL, Pettinger M, et al. Effects of a low-­fat fat, carbohydrates and cardiovascular disease. Neth J Med
dietary intervention on glucose, insulin, and insulin resistance in the 2011;69:372–8.
women's health Initiative (WHI) dietary modification trial. Am J Clin 43 Volek JS, Feinman RD. Carbohydrate restriction improves the features
Nutr 2011;94:75–85. of metabolic syndrome. metabolic syndrome may be defined by the
14 Culver AL, Ockene IS, Balasubramanian R, et al. Statin use and response to carbohydrate restriction. Nutr Metab 2005;2:31.
risk of diabetes mellitus in postmenopausal women in the women's 44 Volek JS, Fernandez ML, Feinman RD, et al. Dietary carbohydrate
health Initiative. Arch Intern Med 2012;172:144–52. restriction induces a unique metabolic state positively affecting
15 Sattar N, Preiss D, Murray HM, et al. Statins and risk of incident atherogenic dyslipidemia, fatty acid partitioning, and metabolic
diabetes: a collaborative meta-­analysis of randomised statin trials. syndrome. Prog Lipid Res 2008;47:307–18.
Lancet 2010;375:735–42. 45 Volek JS, Phinney SD, Forsythe CE, et al. Carbohydrate restriction
16 Noakes TD. WHIDMT: Rossouw and Howard blatantly miss the has a more favorable impact on the metabolic syndrome than a low
point. S Afr Med J 2014;104:262. fat diet. Lipids 2009;44:297–309.
17 Teicholz N. The big fat surprise. Why butter, meat and cheese 46 Hyde PN, Sapper TN, Crabtree CD, et al. Dietary carbohydrate
belong in a heathy diet. New York, NY: Simon and Schuster, 2014. restriction improves metabolic syndrome independent of weight
18 Noakes TD, Sboros M. Real food on trial. How the diet dictators loss. JCI Insight 2019;4:e128308.
tried to destroy a top scientist. UK: Columbus Publishing Ltd, 47 Rossello X, Raposeiras-­Roubin S, Oliva B, et al. Glycated
2019. hemoglobin and subclinical atherosclerosis in people without
19 Assaf AR, Beresford SAA, Risica PM, et al. Low-­fat dietary pattern diabetes. J Am Coll Cardiol 2021;77:2777–91.
intervention and health-­related quality of life: the women's health 48 Volek JS, Forsythe CE. The case for not restricting saturated fat on
Initiative randomized controlled dietary modification trial. J Acad a low-­carbohydrate diet. Nutr Metab 2005;2:21–2.
Nutr Diet 2016;116:259–71. 49 Demasi M, Lustig RH, Malhotra A. The cholesterol and calorie
20 McCoy CE. Understanding the intention-­to-­treat principle in hypotheses are both dead – it is time to focus on the real culprit:
randomized controlled trials. West J Emerg Med 2017;18:1075–8. insulin resistance. Clin Pharmacist 2017;9.
21 Mozaffarian D, Rimm EB, Herrington DM. Dietary fats, 50 DuBroff R. Cholesterol paradox: a correlate does not a surrogate
carbohydrate, and progression of coronary atherosclerosis in make. Evid Based Med 2017;22:15–19.
postmenopausal women. Am J Clin Nutr 2004;80:1175–84. 51 Ravnskov U, de Lorgeril M, Diamond DM, et al. LDL-­C does not
22 Mozaffarian D. Low-­fat diet and cardiovascular disease. JAMA cause cardiovascular disease: a comprehensive review of the
2006;296:279–80. current literature. Expert Rev Clin Pharmacol 2018;11:959–70.
23 Anderson CAM, Appel LJ. Dietary modification and CVD 52 Ludwig DS, Willett WC, Volek JS, et al. Dietary fat: from foe to
prevention: a matter of fat. JAMA 2006;295:693–5. friend? Science 2018;362:764–70.
24 Grundy SM. Should women be offered cholesterol lowering drugs 53 DuBroff R. A reappraisal of the lipid hypothesis. Am J Med
to prevent cardiovascular disease? Yes. BMJ 2007;334:982. 2018;131:993–7.
25 Kendrick M. Should women be offered cholesterol lowering drugs 54 Dehghan M, Mente A, Rangarajan S, et al. Association of dairy
to prevent cardiovascular disease? no. BMJ 2007;334:983. intake with cardiovascular disease and mortality in 21 countries
26 Aberegg SK, Majure DT, Howard BV. Low-­fat diet and from five continents (PURE): a prospective cohort study. Lancet
cardiovascular disease. JAMA 2006;296:279–1. 2018;392:2288–97.

8 Noakes TD. Open Heart 2021;8:e001680. doi:10.1136/openhrt-2021-001680

 Viewpoint

55 Astrup A, Geiker NRW, Magkos F. Effects of full-­fat and fermented 82 Unwin DJ, Tobin SD, Murray SW, et al. Substantial and sustained

Open Heart: first published as 10.1136/openhrt-2021-001680 on 21 July 2021. Downloaded from http://openheart.bmj.com/ on April 5, 2023 by guest. Protected by copyright.
dairy products on cardiometabolic disease: food is more than the improvements in blood pressure, weight and lipid profiles from
sum of its parts. Adv Nutr 2019;10:924S–30. a carbohydrate restricted diet: an observational study of insulin
56 Diamond DM, O'Neill BJ, Volek JS. Low carbohydrate diet: are resistant patients in primary care. Int J Environ Res Public Health
concerns with saturated fat, lipids, and cardiovascular disease risk 2019;16:2680.
justified? Curr Opin Endocrinol Diabetes Obes 2020;27:291–300. 83 Webster CC, Murphy TE, Larmuth KM, et al. Diet, diabetes status,
57 Diamond DM, Alabdulgader AA, de Lorgeril M, et al. Dietary and personal experiences of individuals with type 2 diabetes
recommendations for familial hypercholesterolaemia: an evidence-­ who self-­selected and followed a low carbohydrate high fat diet.
free zone. BMJ Evid Based Med 2020:bmjebm-2020-111412. Diabetes Metab Syndr Obes 2019;12:2567–82.
58 Astrup A, Magkos F, Bier DM, et al. Saturated fats and health: a 84 Skytte MJ, Samkani A, Petersen AD, et al. A carbohydrate-­reduced
reassessment and proposal for food-­based recommendations. J high-­protein diet improves HbA1c and liver fat content in weight
Am Coll Cardiol 2020;76:844–57. stable participants with type 2 diabetes: a randomised controlled
59 DuBroff R, de Lorgeril M. Fat or fiction: the diet-­heart hypothesis. trial. Diabetologia 2019;62:2066–78.
BMJ Evid Based Med 2021;26:3–7. 85 Choi YJ, Jeon S-­M, Shin S. Impact of a ketogenic diet on metabolic
60 Reaven GM, lecture B. Role of insulin resistance in human disease. parameters in patients with obesity or overweight and with or
Diabetes 1988;37:1595–607. without type 2 diabetes: a meta-­analysis of randomized controlled
61 Reaven GM, Hoffman BB. A role for insulin in the aetiology and trials. Nutrients 2005;12:2005.
course of hypertension? Lancet 1987;2:435–7. 86 Unwin D, Khalid AA, Unwin J, et al. Insights from a general practice
62 Ferrannini E, Buzzigoli G, Bonadonna R, et al. Insulin resistance in service evaluation supporting a lower carbohydrate diet in patients
essential hypertension. N Engl J Med 1987;317:350–7. with type 2 diabetes mellitus and prediabetes: a secondary analysis
63 DeFronzo RA, Ferrannini E. Insulin resistance. A multifaceted of routine clinic data including HbA1c, weight and prescribing over
syndrome responsible for NIDDM, obesity, hypertension, 6 years. BMJ Nutr Prev Health 2020;3:e000072.
dyslipidemia, and atherosclerotic cardiovascular disease. Diabetes 87 Ahmed SR, Bellamkonda S, Zilbermint M, et al. Effects of the low
Care 1991;14:173–94. carbohydrate, high fat diet on glycemic control and body weight in
64 Reaven GM, Lithell H, Landsberg L. Hypertension and associated patients with type 2 diabetes: experience from a community-­based
metabolic abnormalities--the role of insulin resistance and the cohort. BMJ Open Diabetes Res Care 2020;8:e000980.
sympathoadrenal system. N Engl J Med 1996;334:374–81. 88 McKenzie AL, Athinarayanan SJ, McCue JJ, et al. Type 2 diabetes
65 Reaven GM. Insulin resistance/compensatory hyperinsulinemia, prevention focused on normalization of glycemia: a two-­year pilot
essential hypertension, and cardiovascular disease. J Clin study. Nutrients 2021;13:749.
Endocrinol Metab 2003;88:2399–403. 89 Moriconi E, Camajani E, Fabbri A, et al. Very-­low-­calorie
66 Xun P, Wu Y, He Q, He K, et al. Fasting insulin concentrations and ketogenic diet as a safe and valuable tool for long-­term glycemic
incidence of hypertension, stroke, and coronary heart disease: management in patients with obesity and type 2 diabetes. Nutrients
a meta-­analysis of prospective cohort studies. Am J Clin Nutr 2021;13:758.
2013;98:1543–54. 90 Goldenberg JZ, Day A, Brinkworth GD, et al. Efficacy and safety of
67 Björntorp P. Adipose tissue distribution, plasma insulin, and low and very low carbohydrate diets for type 2 diabetes remission:
cardiovascular disease. Diabete Metab 1987;13:381–5. systematic review and meta-­analysis of published and unpublished
68 Wiebe N, Ye F, Crumley ET, et al. Temporal associations among
randomized trial data. BMJ 2021;372:m4743.
body mass index, fasting insulin, and systemic inflammation:
91 Skytte MJ, Samkani A, Astrup A, et al. Effects of carbohydrate
a systematic review and meta-­analysis. JAMA Netw Open
restriction on postprandial glucose metabolism, β-cell function, gut
2021;4:e211263.
hormone secretion, and satiety in patients with Type 2 diabetes. Am
69 Zhang AMY, Wellberg EA, Kopp JL, et al. Hyperinsulinemia
J Physiol Endocrinol Metab 2021;320:E7–18.
in obesity, inflammation, and cancer. Diabetes Metab J
92 Lennerz BS, Koutnik AP, Azova S, et al. Carbohydrate restriction
2021;45:285–311.
for diabetes: rediscovering centuries-­old wisdom. J Clin Invest
70 Steinberg HO, Chaker H, Leaming R, et al. Obesity/insulin
2021;131:e142246.
resistance is associated with endothelial dysfunction.
93 Schmiegelow MD, Hedlin H, Stefanick ML, et al. Insulin resistance
Implications for the syndrome of insulin resistance. J Clin Invest
and risk of cardiovascular disease in postmenopausal women: a
1996;97:2601–10.
71 Yip J, Facchini FS, Reaven GM. Resistance to insulin-­mediated cohort study from the women's health Initiative. Circ Cardiovasc
glucose disposal as a predictor of cardiovascular disease. J Clin Qual Outcomes 2015;8:309–16.
Endocrinol Metab 1998;83:2773–6. 94 Pan K, Chlebowski RT, Mortimer JE, et al. Insulin resistance and
72 Facchini FS, Hua N, Abbasi F, et al. Insulin resistance as a predictor breast cancer incidence and mortality in postmenopausal women in
of age-­related diseases. J Clin Endocrinol Metab 2001;86:3574–8. the Women’s Health Initiative. Cancer 2020;126:3638–47.
73 Eddy D, Schlessinger L, Kahn R, et al. Relationship of insulin 95 Pan K, Nelson RA, Wactawski-­Wende J. Insulin resistance and
resistance and related metabolic variables to coronary artery cancer-­specific and all-­cause mortality in post-­menopausal
disease: a mathematical analysis. Diabetes Care 2009;32:361–6. women: The Women’s Health Initiative. J Natl Cancer Inst
74 Ginsberg HN. Insulin resistance and cardiovascular disease. J Clin 2020;112:djz069.
Invest 2000;106:453–8. 96 Kraft JR. Diabetes epidemic and you. 2nd edn. Victoria, BC:
75 Reaven G. Insulin resistance and coronary heart disease Trafford, 2011.
in nondiabetic individuals. Arterioscler Thromb Vasc Biol 97 Kraft JR. Detection of diabetes mellitus in situ (occult diabetes). Lab
2012;32:1754–9. Med 1975;6:10–22.
76 Ormazabal V, Nair S, Elfeky O, et al. Association between insulin 98 Crofts C, Schofield G, Zinn C, et al. Identifying hyperinsulinaemia in
resistance and the development of cardiovascular disease. the absence of impaired glucose tolerance: an examination of the
Cardiovasc Diabetol 2018;17:122. Kraft database. Diabetes Res Clin Pract 2016;118:50–7.
77 Evans CE, Greenwood DC, Threapleton DE, et al. Glycemic 99 DiNicolantonio JJ, Bhutani J, OKeefe JH, et al. Postprandial insulin
index, glycemic load, and blood pressure: a systematic review assay as the earliest biomarker for diagnosing pre-­diabetes,
and meta-­analysis of randomized controlled trials. Am J Clin Nutr type 2 diabetes and increased cardiovascular risk. Open Heart
2017;105:1176–90. 2017;4:e000656.
78 Cucuzzella MT, Tondt T, Dockster NE. A low-­carbohydrate survey: 100 Mozaffarian D, Aro A, Willett WC. Health effects of trans-­fatty
evidence for sustainable metabolic syndrome reversal. J Ins Resist acids: experimental and observational evidence. Eur J Clin Nutr
2017;2:a30. 2009;63(Suppl 2):S5–21.
79 Hallberg SJ, McKenzie AL, Williams PT, et al. Effectiveness and 101 Lawrence GD. Perspective: the saturated fat-­unsaturated oil
safety of a novel care model for the management of type 2 diabetes dilemma: relations of dietary fatty acids and serum cholesterol,
at 1 year: an open-­label, non-­randomized, controlled study. atherosclerosis, inflammation, cancer, and all-­cause mortality. Adv
Diabetes Ther 2018;9:583–612. Nutr 2021;12:647–56.
80 Athinarayanan SJ, Adams RN, Hallberg SJ, et al. Long-­term effects 102 Lim EL, Hollingsworth KG, Aribisala BS, et al. Reversal of type
of a novel continuous remote care intervention including nutritional 2 diabetes: normalisation of beta cell function in association
ketosis for the management of type 2 diabetes: a 2-­year non-­ with decreased pancreas and liver triacylglycerol. Diabetologia
randomized clinical trial. Front Endocrinol 2019;10:348. 2011;54:2506–14.
81 Walton CM, Perry K, Hart RH, et al. Improvement in glycemic and 103 Hite AH, Feinman RD, Guzman GE, et al. In the face of
lipid profiles in type 2 diabetics with a 90-­day ketogenic diet. J contradictory evidence: report of the dietary guidelines for
Diabetes Res 2019;2019:8681959. Americans Committee. Nutrition 2010;26:915–24.

Noakes TD. Open Heart 2021;8:e001680. doi:10.1136/openhrt-2021-001680 9