-
This was a prospective, population-based research study in 1981 men using a 4-day food diary to estimated nutrient intakes; the primary endpoints were fatal and non-fatal coronary artery disease (CAD) events.
-
Polyunsaturated fatty acid intake correlated with a decreased risk of fatal CAD events; total fat intake correlated with an increased risk of non-fatal CAD events.
-
No change in CAD event risk correlated with saturated fat intake.
By David Kiefer, MD
A prospective, population-based study in middle-aged and older men found that polyunsaturated fatty acid intake correlated with a lower risk of coronary artery disease events.
Virtanen JK, et al. Dietary fatty acids and risk of coronary heart disease in men: The Kuopio Ischemic Heart Disease Risk Factor study. Arterioscler Thromb Vasc Biol 2014 Sep 25. [Epub ahead of print].
Following up on conflicting research results about the role of dietary saturated fatty acids (SFA) and other macronutrients in the development of coronary artery disease (CAD), the researchers in this study aimed to improve the research methodology and generalizability surrounding this important topic. For example, in the introduction, the authors mention early research pointing to a direct and inverse relationship between SFA and polyunsaturated fatty acid (PUFA) intakes, respectively, on total and low-density lipoprotein (LDL) cholesterol, leading to dietary recommendations to lower SFA intake. Follow-up studies found less of a connection between SFA intake and CAD, leading to the postulation that other risk factors besides total and LDL concentration exist, and/or that macronutrients other than SFA may have a connection to CAD. This study explored these "other" macronutrients.
The researchers used a 4-day food diary in a population with high SFA intakes and high rates of CAD, more akin to what is seen with a standard Western diet (hence, the increased generalizability). The food diary is considered the "gold standard," and an improvement over prior 24-hour food recalls or food frequency questionnaires.
The researchers tapped into data from the Kuopio Ischemic Heart Disease Risk Factor Study (KIHD) in Finland and used fatal and non-fatal CAD as the primary endpoints. Secondary analyses used carotid artery intima-media thickness measurements (when available) as a representation of the atherosclerotic disease process, though the subset of this cohort with this measurement available were younger and had lower systolic blood pressure, lower body mass index, lower baseline SFA and protein intakes, higher baseline PUFA intakes, and were more likely to smoke.
Data from 1981 men, free of CAD in the years 1984-1989, were analyzed. Table 1 lists notable baseline characteristics of this cohort. The average total cholesterol was 237.8 mg/dL, triglyceride 110.6 mg/dL, high-density lipoprotein 50.6 mg/dL, and LDL 154.4 mg/dL. The men in this cohort ate an average of 2391 kilocalories ("calories") daily, which was 38.9% fat, including 18.0% SFA, 1.1% trans-fatty acid (TFA), 11.0% monounsaturated fat (MUFA), 4.6% PUFA, 15.6% protein, and 43.5% carbohydrate. Of the PUFA intake, 3.5% were omega-6 fats and 0.7% were omega-3 fats.
The average follow-up for the 1981 men was 21.4 years. During this time, 183 fatal CAD events and 382 non-fatal CAD events were recorded in the KIHD database. PUFAs were the only fat to show a correlation with fatal CAD events: The hazard ratio (HR) was 0.46 when quartile 4 (highest PUFA intake) was compared to quartile 1 (lowest PUFA intake) (P = 0.03). Of note, this statistically significant trend was seen only when the analysis was adjusted for other baseline variables and the other fat types. The trend was not observed when omega-6 PUFAs and omega-3 PUFAs were analyzed separately over the quartiles (P = 0.08 and P = 0.11, respectively). PUFA intake also correlated with smaller carotid artery intima-media thickness (P < 0.05), whereas the other fats did not have statistically significant correlations with that measurement.
With respect to fatal CAD events and the other fats, no such statistically significant trend was documented for any of the other fat types.
Non-fatal CAD events increased proportionally with the quartiles of total fat intake (HR = 1.38, P = 0.03); a statistically significant trend was only observed, again, when the analysis was adjusted for the other baseline variables and other fat intake. No associations were found with the other fat types.
|
Table 1: Baseline Characteristics of the 1981 Men from the KIHD Study
|
|
Average age
|
52 years
|
|
Average number of years of education
|
9 years
|
|
Percent married
|
87%
|
|
Percent living in rural areas
|
27%
|
|
Average body mass index
|
26.7 kg/m2
|
|
Percent currently using tobacco
|
30%
|
|
Percent with diabetes mellitus
|
5%
|
|
Percent with hypertension
|
56%
|
The database allowed the researchers to do another interesting calculation: the change in CAD risk with a 1% energy substitution of one nutrient for another. When MUFAs (but, the researchers noted, the MUFAs in this cohort more likely came from animal products, such as meat and dairy, and not olive oil) replaced SFA, TFA, or carbohydrates, fatal (and, less so, non-fatal) CAD risk increased; the opposite occurred when PUFAs replaced one of those three nutrients. A replacement of SFA or TFA with carbohydrate did not change CAD risks.
COMMENTARY
The debate about healthy and unhealthy fats is far from over, especially in context of the doubts over the efficacy of the low-fat diet for cardioprotection,1 the resurgence of saturated fats (coconut oil and butter trumping margarine?), and the scientific "victories" of the MUFA-heavy Mediterranean diet for CAD and diabetes mellitus.2
This study adds to our knowledge with more detailed data about dietary correlates with CAD risk in a more typical Western diet-eating, albeit one gender, population. If we believe these results, then there is less concern than in the past to focus on SFA intake as opposed to total fat intake. Perhaps the inadequacies of the literature as pointed out by the authors of this study were enough to skew the data and create an anti-SFA frenzy in past decades. Less surprising were the findings that PUFAs correlated with some degree of cardioprotection. The PUFA results did not translate into a tangible omega-3 vs omega-6 benefit, different than most of the medical literature and possibly related to one key factor; the cohort’s baseline 5:1 ratio of omega-6 to omega-3 is closer to a historically healthy intake (1:1 or 2:1) than the oft-quoted 15:1 to 20:1 found in a typical, CAD-prone Western diet. This more optimal PUFA intake may have negated the ability of this study to statistically demonstrate an omega-3 or omega-6 effect.
The calorie substitution calculation, albeit a statistician’s trick, may be the most clinically intriguing aspect of this study. Clinicians often play the "substitution game," swapping out high-glycemic for low-glycemic index carbohydrates. The results here hint at what might happen if PUFAs (benefit) or animal-based MUFAs (harm) replace other fats and nutrients. Are we to stop the trend of inserting olive oil (a MUFA) in people’s diet? (Please don’t. The MUFA-effect here was likely due to MUFAs from animal products, not olive oil). Or, only substitute with PUFAs, such as seed oils, notoriously high in omega-6 fatty acids? For any of the detailed substitutions, it is difficult to come to a definitive conclusion just based on this trial. As the structure of the study only allows us to comment on correlations, follow-up clinical interventional research trials are necessary to determine cause-effect on CAD events. Until then, for this demographic, saturated fat was not as evil as we would have predicted, and PUFAs may indeed play an important role in the prevention of fatal and non-fatal CAD events.
References
-
Bazzano LA, et al. Effects of low-carbohydrate and low-fat diets. Ann Int Med 2014;161:309-318.
-
Estruch R, et al. Primary prevention of cardiovascular disease with a Mediterranean diet. N Engl J Med 2013;368:1279-1290.
