The relationship of saturated fats and coronary heart disease: fa(c)t or fiction? A commentary

The risk of coronary heart disease (CHD) associated with saturated fatty acids (SFA) varies from no association to a significantly important risk.^1–52 CHD is the number one cause of death in the USA, accounts for approximately 17% of deaths, and is associated with over 1 million myocardial infarctions (MIs) each year.^1,22 Nutrition and modification of diet make up one of the primary lifestyle approaches to preventing CHD with a major focus on the reduction of dietary SFA intake.^2,22

The purpose of this commentary is to address the published scientific evidence in an attempt to answer the following questions.

1. Is there an association between SFA and CHD risk?

2. What is the comparator macronutrient that proves that SFA will increase CHD risk? In other words, there is no placebo to investigate their independent effects and so the effects must be expressed relative to a similar amount of energy from the other macronutrients that replace the SFA.

3. Is there a different CHD risk depending on the specific SFA?

4. What is the difference in CHD risk depending on the source of the SFA?

5. If the SFA is replaced with isocaloric energy at different levels, what is the change in CHD risk?

Present dietary guidelines recommend keeping SFA intake at 8–10% of total energy intake for the prevention of CHD.²² Clinical trials offer conflicting conclusions regarding the role of SFA and the risk of CHD and its clinical complications, such as MI, congestive heart failure, angina, and sudden death. The lay public is confused by media miscommunications and misinterpretation of published data, recently published national best sellers advising the public to eat more fat or wear shirts with similar labels, and the conflicting nutrition recommendations published by national and international committees.^1,22,27 The source of the confusion lies within the complexity, accuracy, and co-ordination of the results and conclusions in basic science, clinical epidemiology, and prospective clinical trials. Some of the misconceptions and improper interpretations of the association of CHD risk and SFA studied are related to the source of the SFA,^{1,13,14,34–37} the carbon length of the SFA,^13,15,34 which may exhibit unique biological properties (i.e. short-chain FA [SCFA], medium-chain FA [MCFA], or long-chain FA [LCFA]),^13,15,25,34 and whether it is an odd or even numbered carbon chain,^12,31 as well as the replacement macronutrient.^13,14 Animal products are rich in SFA and meat and dairy are the predominant sources in Western diets. For example, a previous analysis in the Nurses’ Health Study (NHS), reported that intake of major SFA (LCFA and MCFA), including lauric (12:0), myristic (14:0), palmitic (16:0), and stearic (18:0), had an increased risk of CHD, whereas the sum of SCFA such as butyric acid (4:0), caproic acid (6:0), caprylic acid (8:0), and capric acid (10:0) did not have an association with increased CHD risk.³⁴

It is difficult to rank definitively all the possible variables that relate to the discrepancies in the role of SFA and the risk of CHD, but many have been mentioned in the literature. However, the dietary source of the SFA (i.e. milk, dairy meat, cheese, butter, processed foods, eggs, coconut oil),^{1,13,14,34–37,45} the carbon chain length of the SFA,^13,15,25 and the replacement macronutrient^13,14,24 probably account for most of the discrepancies in the studies. There are many other issues that could increase the risk of CHD related to SFA. The reader is referred to these references for a more detailed review of the effects on lipids, inflammation, oxidative stress, thrombosis, and genetics.^1–3,18 The effects of specific SFA on blood lipids is reviewed in a comprehensive meta-analysis by Mensink.⁵³

There have only been a few observational studies that have investigated the relationship between specific SFA and the risk of CHD.^13–15 A meta-analysis of 32 trials that included approximately 600,000 participants, with 17 observational studies of FA biomarkers, 32 observational studies of FA intake as well as 27 randomized controlled clinical trials (RCCT) of FA supplementation evaluated CHD risk and types of FA.¹² Results were based on self-reported dietary FA intake from 32 prospective studies that included 512,420 participants with 15,945 cases of CHD. CHD was defined as fatal or nonfatal MI, CHD, coronary insufficiency, coronary death, angina, angiographic coronary stenosis, or sudden death.¹² This meta-analysis is at variance with other studies perhaps due to the heterogeneity of the populations, selection bias, quality of studies selected, self-reporting of diet, and other confounders due to unmeasured dietary factors and other lifestyle factors. Despite the size of this meta-analysis, the results and conclusions need to be interpreted with caution.

Another review and meta-analysis of observational studies by de Souza and colleagues¹⁷ suggested that SFA are not associated with all-cause mortality, cardiovascular disease (CVD), or CHD. The authors admit that the “certainty of associations between SFA and all outcomes was very low and the evidence is heterogeneous with methodological limitations and issues with evidence synthesis and quality”. They also delineate at least five major limitations to their study conclusions related to inherent analytic techniques of meta-analysis, lack of causality but only associations, measurement error with bias, dietary record capture of adequate SFA intake, utilization of a ‘most adjusted model’ as well as the small number of cohorts limiting dose-response relations or difference between SFA sources on cardiovascular outcomes.¹⁷

The 10-year Multi-Ethnic Study of Atherosclerosis (MESA)⁴¹ demonstrated that SFA from meat had a higher risk for CVD: hazard ratio (HR) (95% confidence interval [CI]) for +5 g/day (1.26 (1.02, 1.54) and a +5% of energy from meat SFA (1.48 (0.98, 2.23).

In the 24-year Cohort Health Professionals Follow-Up Study (HPFS) and the 22-year and 20- year cohort NHS I and II,¹⁴ dairy fat consumption was not associated with the risk of total CVD. Isocaloric replacement of 5% energy from dairy fat by polyunsaturated fats (PUFA) or vegetable fat was associated with 24% and 10% reductions in CHD risk, respectively.¹⁴

In a prospective longitudinal cohort study in NHS and HPFS,¹⁵ the individual SFA intakes related to CHD were 1.07 for 12:0, 1.13 (for 14:0, 1.18 for 16:0, 1.18 for 18:0), and 1.18 for all four SFA combined (12:0–18:0).¹⁵ HRs of CHD for isocaloric replacement of 1% energy from 12:0–18:0 were 0.92 (p < 0.001) for PUFA, 0.95 (p = 0.08) for monounsaturated fatty acids (MUFA), 0.94 (p < 0.001) for whole grain carbohydrates, and 0.93 (p = 0.001) for plant proteins. For individual SFA, the lowest risk of CHD was observed when the most abundant SFA, 16:0, was replaced, which was also noted in the Rotterdam study.³⁷ A case-control study of 933 Costa Rican people also showed a positive association of MI with MCFA and LCFA 12:0–18:0.³⁵ A Cochrane meta-analysis in 2012 found a significant 14% reduction in cardiovascular events by reducing dietary SFA (relative risk [RR]: 0.86, 95% CI: 0.77–0.96).²⁶

The NHS and HPFS studies evaluated incident cases of CHD and the association of SFA compared with unsaturated fats and different sources of carbohydrates.²⁴ Higher intakes of whole grain carbohydrates and PUFA were associated with a lower CHD risk based on quintiles. Refined starches and added sugars were associated with an increased risk of CHD (HR: 1.10, 95% CI: 1.00–1.21; p trend = 0.04).² Isocaloric replacement at 5% of energy from SFA with PUFA, MUFA, or whole grain carbohydrates showed a 25%, 15%, and 9% lower risk of CHD, respectively.

A study by Jakobsen and colleagues of 11 US and European cohort studies showed similar results by replacing SFA with PUFA and the risk of CHD.⁴⁰ During a 4–10 year follow up a 5% lower energy intake from SFA with an isocaloric intake from PUFA significantly reduced coronary events by 13% and coronary deaths by 26%.⁴⁰

The Cochrane database review in 2015 suggested that reducing dietary SFA lowered cardiovascular events by 17% (RR: 0.83, 95% CI: 0.72–0.96).⁴² The review incorporated 15 RCCTs (17 comparisons, 59,000 participants). The all-cause mortality (RR: 0.97, 95% CI: 0.90–1.05) and cardiovascular mortality (RR: 0.95, 95% CI: 0.80–1.12, 12 trials, 53,421 participants) were not significantly reduced. Reduction in SFA intake reduced the risk of MI (fatal and nonfatal) (RR: 0.90, 95% CI: 0.80–1.01), but did not reduce nonfatal MI (RR: 0.95, 95% CI: 0.80–1.13).

The PREvención con DIeta MEDiterránea (PREDIMED) was a 6-year prospective study of 7038 subjects with a high risk for CVD that included MI, CVA, or death from cardiovascular causes.¹⁹ The dietary consumption of SFA and trans fatty acids from the highest to the lowest quintiles increased overall CVD by 81% and 67%, respectively. The reduction in CVD in the highest quintile of total fat, MUFA, and PUFA compared with those in the lowest quintile was 42%, 50% and 32%, respectively. The isocaloric replacement of SFA with PUFA and MUFA reduced the risk of CVD and death. SFA from processed foods and pastries had the highest association with overall CVD risk and SFA from vegetable (vegetable oils, nuts, vegetables and margarine) and fish were associated with a lower risk of CVD.¹⁹

A total of 7447 patients were enrolled in the prospective RCCT PREDIMED diet study over a study period of 4.8 years.²⁰ The primary endpoint was the rate of major cardiovascular events (i.e. MI, stroke, or death from cardiovascular causes). The major cardiovascular events from MI, cerebrovascular accident (CVA), or total cardiovascular deaths decreased 28% with the consumption of nuts and 30% with extra-virgin olive oil (EVOO).²⁰ The group assigned to the Mediterranean diet with EVOO had an HR of 0.70 (95% CI: 0.54–0.92). The group assigned to a Mediterranean diet with nuts had a HR of 0.72 (95% CI: 0.54–0.96). The reduction in CVA was 39% (p < 0.003) (33% reduction from EVOO and a 46% reduction with nuts). The reduction in MI was 23% (p = 0.25) (20% reduction from EVOO and a 26% reduction with nuts). The total number of cardiovascular deaths was reduced by 17% (p = 0.8).²⁰

Another group analyzed the relationship between choices of dietary fats and overall mortality.³² The study included 126,233 participants from NHS and HPFS. Every 5% increase in SFA consumption was associated with an 8% higher risk of overall mortality. For those who replaced SFA with unsaturated fats, especially PUFA, there was a significantly lower risk of total death and mortality from CVD compared with those on chronic high intakes of SFA.

SFA are diverse compounds, are not created equal, and cannot be ‘lumped’ into a single category. It is prudent to replace LCFA with PUFA, MUFA, whole grains, and dairy and plant proteins. The recommended grams per day, or percentage of SFA relative to total fat or total calories, cannot be accurately determined nor recommended at this time, but it is suggested that the SFA dietary intake should be well below 10% of the total caloric intake.²² The overall relationship of the human diet to CHD should include the totality of our nutrition and avoid reductionist evaluations of single macronutrients. New nutritional guidelines should promote dietary patterns that improve CHD based on validated science.