by Kenneth W. Krause.
Kenneth W. Krause is a contributing editor and “Science Watch” columnist for the Skeptical Inquirer. Formerly a contributing editor and books columnist for the Humanist, Kenneth contributes regularly to Skeptic as well. He may be contacted at email@example.com.
Was the Government Wrong?
Dietary recommendations were issued by the U.S. Select Committee on Nutrition and Human Needs and the UK National Advisory Committee on Nutritional Education in 1977 and 1983, respectively. In each case, the government focused heavily on reducing consumption of dietary fat to combat coronary heart disease. The official advice was, and remains into 2015, to restrict overall fat intake to thirty percent and saturated fat intake to ten percent of total calories.
One might assume that the national guidelines were based on a plethora of hard evidence. Not so, according to obesity researcher, Zoe Harcombe. Earlier this year, her team published a review and meta-analysis of randomized controlled trials accessible in 1983 to the U.S. and UK committee members. Their conclusion? The “best evidence” then available “did not support the contention that reducing dietary fat intake would contribute to a reduction in coronary heart disease or related mortality.”(1)
Harcombe et al. narrowed their collection of professional literature down to six relevant studies including a total of 740 deaths, 423 from coronary heart disease, among 2467 participants—all male. Every trial but one had examined only secondary prevention participants—in other words, those who had already experienced myocardial infarction. The remaining study was comprised of twenty percent secondary and eighty percent primary prevention subjects.
Somewhat interestingly, five of the six trials never examined the effects of dietary interventions set at levels matching the eventual recommendations. The one study that had examined the consequences of a diet limited to ten percent saturated fat, according to Harcombe, actually “reported a higher incidence of all-cause mortality and coronary heart disease in the intervention group.”
In fact, Harcombe et al.’s overall analysis revealed no statistically significant relationship between dietary intervention and either all-cause mortality or heart deaths, despite the fact that reductions in mean serum (blood) cholesterol levels were significantly greater among the intervention groups. This, they insist, “undermines the role of serum cholesterol levels as an intermediary to the development of coronary heart disease and contravenes the theory that reducing fat generally and saturated fat particularly potentiates a reduction in coronary heart disease.”
As such, Harcombe concludes, “it seems incomprehensible that dietary advice was introduced for 220 million Americans and 56 million UK citizens , given the contrary results from a small number of unhealthy men.” And because other reviews of more recent evidence have also questioned the same diet-heart hypothesis upon which these recommendations were based, her group contends, our governments’ “dietary advice not merely needs review; it should not have been introduced” in the first place.
Harcombe’s position on the diet-heat hypothesis is characteristic of an expanding and increasingly popular minority view that saturated dietary fats in particular have been wrongly demonized as killers. In this article, I’ll briefly discuss the debate’s necessary history and many of the most important studies and reviews published by nutrition researchers on each side of the scientific rift. In some cases—to reveal the debate’s curious ferocity on the one hand, and notable lack of scientific confidence on the other—I’ll also include certain experts’ personal opinions and criticisms. Finally, I’ll attempt to expose what the science really says, if anything, about the much-disputed diet-heart hypothesis and the foods we really ought to eat.
From Proposal to Paradigm to Policy.
In the January 4, 1985 issue of Science, journalist Gina Kolata covered the 47th consensus panel report from the National Institutes of Health, published some three weeks earlier. Since 1961, the American Heart Association had urged Americans to consume less saturated fat and cholesterol, and recommended its so-called “prudent diet” emphasizing fruits, vegetables, and vegetable oils.
The NIH had been hesitant to take a firm position on the controversial diet-heart hypothesis, according to Kolata, because the scientific literature focusing on the connection between dietary cholesterol and saturated fatty acids (SFA) on the one hand, and heart disease on the other, did “not show that lowering cholesterol makes a difference.”(2)
But the NIH’s reticence was finally overcome by the results of a then-new study conducted by the National Heart, Lung and Blood Institute on the effects of a cholesterol-reducing diet along with a drug, cholestyramine, on about 4000 middle-aged men with elevated serum cholesterol levels.(3) On average, the intervention group’s cholesterol had plunged by 13.4 percent since the investigation began in 1973, or 8.5 percent better than the average decrease found among placebo-treated controls.
According to the NHLBI, its researchers had provided not only “strong evidence” of a “causal role” for low-density lipoprotein cholesterol, or LDL-C, in the pathogenesis of coronary heart disease (CHD), but good reason as well to extend its findings to “other age groups and women” along with “others with more modest elevations of serum cholesterol.”
Others were less impressed. In an interview with Kolata, University of Chicago and frequent NIH statistician, Paul Meier, deemed the new study’s findings “weak,” referring to the disappointing and statistically insignificant distinction between the intervention and control groups in terms of deaths from all causes. And although incidences of angina, bypass surgery, and abnormal exercise electrocardiograms all dropped in the modified diet-drug group, Kolata judged that the new study had “failed, as every other trial did, to prove that lowering blood cholesterol saves lives.”
But perhaps most provocative was the panel’s recommendation that all Americans from the age of two should reduce their consumption of SFA and cholesterol. Also interviewed by Kolata, Thomas Chalmers of the Mt. Sinai Medical School argued that the report “made an unconscionable exaggeration of the data,” emphasizing that “there is absolutely no evidence that it’s safe for children to be on a cholesterol-lowering diet.”
In a subsequent letter to Science, Daniel Steinberg, who had chaired the NIH consensus development conference, criticized Kolata for devoting the lion’s share of her article to “no more than a handful among some 600 conferees.”(4) “The panel’s recommendation,” he countered, “is sound when all of the evidence is taken into account.” Such evidence also led quickly in 1985 to the National Cholesterol Education Program, a new NIH administration created to instruct physicians how to identify and treat “at-risk” patients.
In fact, there was much agreement among the conferees. But dissent in 1984 was both loud and determined, as it remains at the dawn of 2015. The differences between the debates then and now are as multifactorial as heart disease itself. While television continues to manipulate viewers’ emotions and bombard them with highly-varied and always simplistic scientific interpretations and dietary advice, internet bloggers now flood every recess of the popular consciousness with harsh, often tactically ruthless diatribes against their nutritional adversaries.
Journalists and popular authors publish infuriated or one-sided narratives alleging conspiracies between researchers, big food, and government agencies to intentionally mislead the public for personal gain. And while new studies and reviews have lately called the very foundations of the diet-heart hypothesis into question, some leading scientists have entrenched themselves as well, separating into mutually antagonistic nutritional camps.
It might sound too melodramatic to be true. But the most regrettable result is unmistakable—the American public is bewildered and incredulous. What kind of diet will help us to not only lose or maintain our weight, but remain healthy and safe as well? Should we cast our lots with the low-fat or low-carbohydrate paradigm? Do saturated fats really raise the risk of heart disease, and, if so, are the officially endorsed replacements perhaps even more dangerous? Worst of all, Americans wonder whether nutrition science has anything at all of substance to offer them.
The classic diet-heart hypothesis (D-Hh) posits simply enough that diets high in SFA and cholesterol (and low in polyunsaturated fatty acids [PUFA]) raise serum total and LDL cholesterol levels and lead to the accumulation of atheromatous plaques. These plaques gradually narrow coronary arteries, reduce blood flow to the heart, and can eventuate in myocardial infarction.
Early evidence linking heart disease to foods rich in cholesterol (and SFA), including red meat, eggs, and shellfish, derived from animal experiments. In 1913, for example, a Russian pathologist reported the ability to induce atherosclerotic-like lesions in rabbits by feeding them copious amounts of cholesterol.(5) Others soon replicated these results, mainly with other herbivorous animals. Many researchers objected, however, that such creatures were naturally ill-suited to metabolize cholesterol. And when similar experiments were carried out on non-herbivorous (and more human-like) dogs, they added, the animals appeared to tolerate the cholesterol much better.
But the D-Hh wasn’t formally articulated until University of Minnesota physiologist Ancel Keys presented the concept in 1952 at Mt. Sinai in New York. Also published in a famous paper the following year, Keys’ employed a simple yet powerful graph correlating in precise curvilinear fashion total fat intake as a percentage of all calories with death rates from heart disease among men in six countries—Japan, Italy, England and Wales, Australia, Canada, and the United States.(6)
Detractors claimed that Keys intentionally ignored data from twenty-two other countries. And when researchers scrutinized additional data from those nations, they found not only that Keys’ correlation was greatly diminished, but that no association whatsoever existed between dietary fats and death from all causes.(7) Nevertheless, according to James DiNicolantonio, cardiovascular researcher at St. Luke’s Mid America Heart Institute, Keys’ early data “seemingly led us down the wrong ‘dietary-road’ for decades to follow.”(8)
Nevertheless, the most convincing early evidence for the D-Hh may have originated from Keys’ Seven Countries Study of sixteen cohorts (12,763 rural males age 40-59) in Greece, Italy, the former Yugoslavia, the Netherlands, Finland, the U.S., and Japan. By this time, Keys had refined his initial proposal to impugn primarily SFA and animal products (dietary cholesterol not so much). Indeed, in this, the first multi-country epidemiological undertaking in history, coronary mortality and the five-year incidence of CHD was positively correlated with SFA but not with either total fat or PUFA intake.(9)
The cross-cultural results were striking. Upon 25-year follow-up, inter-population death rates from CHD had differed dramatically. In East Finland, for example, 268 per 1000 lumberjacks and farmers living on diets high in meat and dairy had died. By contrast, of the Greeks of Crete who in terms of dietary fats subsisted on olive oil and very little meat, only twenty-five per 1000 had perished. Perhaps most notably, however, SFA had accounted for twenty-two percent of the Finns’, yet only eight percent of the Cretans’, total calories.
But Keys was and continues to be criticized for having “cherry-picked” his Seven Countries data. Some argue that information from populations in countries like France, Switzerland, Germany, Norway, or Sweden, for example, might have challenged Keys’ hypothesis. Some, like journalist Nina Teicholz, have recently gone so far as to charge that he deliberately selected “only those nations … that seemed likely to confirm it.”(10)
Indeed, Keys’ data was not chosen randomly. But Henry Blackburn, Keys’ colleague in Seven Countries, sees no reason why the populations should have been selected entirely by chance. “Demonstrating a lack of understanding of how scientists approach new questions,” he explains, the critics ignore the fact that “any savvy scientist at an early phase of questioning knows to look first not randomly but across wide variations of the cause under consideration, in this case diet.”(11)
Others observed of Keys’ results that CHD mortality varied widely within certain countries. “Despite similar risk factors and diet,” Danish independent researcher Uffe Ravnskov found that “the 5-year incidence of fatal CHD in Crevalcore, Italy, was more than twice that in Montegiorgio, while in Karelia it was five times higher than in West Finland; and on Corfu, 6-7 times higher than on Crete.”(12) Thus, for these and other dissenters, the supposed correlation between diet and heart disease was little more than a well-staged illusion.
On May 28, 1980—only five years previous to the NIH’s consensus report on the same subject—the Food and Nutrition Board of the National Academy of Sciences issued a far more controversial paper, Toward Healthful Diets, finding no clear evidence that reducing serum cholesterol through dietary intervention could prevent CHD. Therein, the fifteen-member Board reproved those “who seek to change the national diet” for assuming a nominal risk in widespread dietary adjustment and for relying so heavily on epidemiological rather than experimental evidence.
Critics of the Board accused its members of maintaining inappropriately cozy relationships with big food organizations like the American Egg Board. In response to the report itself, Robert Levy, director of the NHLBI, offered the following faint-hearted guidance: “Existing information indicates that Americans should hedge their bets and seek a diet lower in saturated fats and cholesterol, at least until more evidence is available.”(13)
At that point, only the Academy and the American Medical Association stood in defiance of the U.S. government and at least eighteen distinguished health organizations. For better or worse, America had officially become a low-fat, low-cholesterol nation.
The Modern Macronutrient Wars Begin.
In her now-classic tome, Food Politics, Marion Nestle—NYU professor of nutrition and former nutrition science adviser to the DHHS, USDA, and FDA—affirmed that “scientists consistently have demonstrated the health benefits of diets rich in fruits and vegetables [and] limited in foods and fats of animal origin.”(14) She continued:
Decades ago, researchers discovered that high levels of cholesterol in the blood predispose individuals to coronary heart disease and that saturated fat (most prominent in meat and dairy products) raises blood cholesterol more than monounsaturated fats (typical of olive oil). They also observed that polyunsaturated fats (most prominent in vegetable seed oils) reduce blood cholesterol levels.
In fact, by the time Food Politics was first published in 2002, every major health organization, including the National Academy of Sciences, had agreed that saturated fatty acids (SFA) raise the risk of cardiovascular disease (CVD), including coronary heart disease (CHD), and that their replacement with either monounsaturated fatty acids (MUFA) or polyunsaturated fatty acids (PUFA)—including all Omega 3 and Omega 6 fatty acids—reduces that risk.
But when I contacted Nestle later, in the summer of 2014, her position seemed less definite. For example, when I inquired about the wisdom of current governmental and non-governmental guidelines relating to SFA intake, most if not all of which warn the entire public against consumption greater than ten percent—in some cases, more than six percent—of total calories, Nestle flatly replied, “I don’t think the jury is in yet on this one.”
So what’s the problem? The “jury” seemed to be “in” decades ago. Despite consistency among official guidelines, the research community now seems hopelessly—indeed, zealously—divided on the topic. Meanwhile, average Americans grow increasingly confused about dietary fats. Science is relentlessly progressive by nature, but its nutritional constituent appears paralyzed. Why?
Consider a recent wave of conflicting studies. In 2009, researchers led by Danish epidemiologist Marianne Jakobsen, pooled data from eleven cohort studies to examine the effects of replacing SFA with MUFA, PUFA, or carbohydrates (CHO) on CHD risk.(15) After four to ten years of follow-up, along with 5249 coronary events and 2155 coronary deaths among 344,696 participants, Jakobsen et al. concluded that MUFA were unassociated, CHO were modestly but directly associated, and PUFA were inversely associated with coronary events. The effect was not modified by either age or gender.
Thus, consistent with conventional wisdom, this group argued that “replacing SFA intake with PUFA intake rather than MUFA or carbohydrate intake prevents CHD over a wide range of intakes and among all middle-aged and older men and women.” Although such distinctions were not tested, Jakobsen also suggested that CHO quality—i.e., fiber content, extent of processing, and glycemic index—might alter the analysis.
But the modern-day macronutrient wars positively erupted a year later when four American researchers led by Patty Siri-Tarino published a meta-analysis of twenty-one studies finding “insufficient evidence … to conclude that dietary saturated fat is associated with an increased risk of CHD, stroke, or CVD.”(16) In other words, this group’s rousing judgment had threatened to undercut the very foundation upon which the diet-heart hypothesis (D-Hh) had been built.
Arguably, however, the Americans’ findings were not entirely hostile to those of Jakobsen. They acknowledged, for example, that a decreased risk often results when SFA are replaced with PUFA. But their interpretation of the data suggested not that SFA were independently problematic, but rather that the benefit of replacement might derive from either increased PUFA or the PUFA to SFA ratio.
In a companion opinion piece, the Americans commented further on the relationships between SFA, CHO, and CVD risk.(17) Therein, Siri-Tarino assailed not only existing and proposed guidelines, but the long-standing use of low-density lipoprotein-cholesterol (LDL-C) as the primary biomarker for CVD risk as well:
Recommendations for further reductions in saturated fat intake (e.g., to ≤ 7% of total energy) are based primarily on the prediction of a progressive reduction in CVD risk associated with greater reductions in LDL cholesterol. However, from the standpoint of implementation, further reductions in saturated fat intake usually involve … increased proportion of carbohydrate…. [But] the effect of higher carbohydrate diets, particularly those enriched in refined carbohydrates, coupled with the rising incidence of overweight and obesity, creates a metabolic state characterized by elevated triglycerides, reduced HDL cholesterol, and increased concentrations of small, dense LDL particles.
Similarly, this group noted that CHO restriction (or weight loss absent restriction) had been shown to yield reductions in the total-cholesterol to high-density lipoprotein-cholesterol (HDL-C) ratio, apolipoprotein B, and the number of small, dense LDL particles. In Siri-Tarino’s estimation, these markers were more closely associated with reduced CVD risk than LDL-C, which “appears to be specific to [less dangerous and] larger, more buoyant particles.”
The Americans finally emphasized the data’s failure to support recommendations for reductions in SFA below ten percent of total calories. While conceding that SFA might raise CVD risk by increasing inflammation and reducing insulin sensitivity, they insisted that their relative effect should be reevaluated “given the changing landscape of CVD risk factors.” Dietary efforts to curb CVD, they urged, “should primarily emphasize limitation of refined carbohydrate intakes and reduction in excess adiposity.”
But the professional response was both swift and severe.(18) Oxford University nutritionist Peter Scarborough denounced the Americans’ methodological design along with their failure to recognize the “well established” association between SFA, serum cholesterol, and CVD. Siri-Tarino countered that her “overall results” remained “robust” and unaffected by “different analytic strategies” even after adjustment for the alleged methodological weakness.
Dutch researcher Martijn Katan criticized the Siri-Tarino et al.’s reliance on the underlying studies’ use of single-day dietary assessments, as opposed to multiday diet records. He insisted as well that, since fat-reduction recommendations had been issued fifty years ago, falling LDL-C concentrations had resulted in conspicuous declines in CHD. Finally, he questioned the private interests of Siri-Tarino’s colleague, Ronald Krauss, relative to his “advisory activities for the dairy industry.” Siri-Tarino acknowledged the underlying studies’ limitations, but argued that all dietary assessments had been subjected to a “quality score” which confirmed her final results. She also informed Katan that Krauss had discontinued his dairy industry associations years prior to publication.
In a separate editorial, Northwestern University preventive-medicine specialist, Jeremiah Stamler—who had served with Ancel Keys on the AHA nutrition committee during the early 1960s—vigorously defended what he deemed an unwarranted attack on the D-Hh:(19)
Do they doubt the validity of the equations of Keys … which are based on dozens of metabolic ward-type feeding experiments, showing independent relations of dietary SFA and cholesterol (direct) and [PUFA] (inverse) to cholesterol …, findings that are repeatedly confirmed in observational and interventional studies in free-living people? … Do they bring into question the classical findings?
Stamler accused Siri-Tarino’s group of ignoring several important investigations, including the Seven Countries, Ni-Hon-San, and National Diet-Heart studies, along with the DASH/Omni-Heart and Multiple Risk Factor Intervention trials. He also challenged Siri-Tarino’s emphasis on CHO-induced dyslipidemia, wondering how she might explain why native Japanese have demonstrated favorable lipid profiles relative to American Japanese, despite their distinctly low-fat, high-CHO diets.
Finally, Stamler denied that the “limited data” supporting the differential effects of SFA and CHO on smaller or larger LDL particles could justify significant changes to the official guidelines. Indeed, nothing Siri-Tarino had to say, Stamler maintained, could “warrant modification of recommendations … beyond intensified emphasis on prevention and control of obesity.”
But the macronutrient guidelines would be challenged again in 2010 by NIH biochemist Christopher Ramsden. Although the AHA, for example, recommended substantial replacement of SFA with Omega-6 (n-6) PUFA-rich vegetable oils, Ramsden suspected that the underlying literature had failed to either, one, distinguish between interventions increasing n-6 PUFA specifically and those boosting both Omega-3 (n-3) and n-6 PUFA, or, two, compare the relative effects of these interventions on CHD outcomes.
Indeed, following their meta-analysis of “all randomized controlled trials that increased PUFA and reported relevant CHD outcomes,” Ramsden’s team calculated a twenty-two percent reduction in CHD risk for mixed n-3/n-6 PUFA diets, but a thirteen percent increase in risk for specific n-6 diets.(20) “These analyses were thus not appropriate,” Ramsden decided, “for formulating advice specific to n-6 PUFA.” Recommendations to increase n-6 PUFA “should be reconsidered,” he warned, “because there is no indication of benefit, and there is a possibility of harm.”
One year later—in apparent disregard of Ramsden’s findings—a diverse group consisting of members from both Jakobsen’s and Siri-Tarino’s 2010 teams and led by Danish nutritionist, Arne Astrup, decided to reevaluate SFA in light of recently published evidence.(21) While now confessing the over-simplicity of the D-Hh, this group predictably confirmed prior claims that SFA should be replaced with PUFA, but not refined CHO. The data regarding MUFA, they added, were too limited to be instructive.
Astrup conceded as well that biomarkers other than LDL-C—to include the total cholesterol-to-HDL-C ratio, non-HDL-C, and apolipoprotien B—can be more enlightening of CVD risk. He also acknowledged evidence indicating that distinct SFA have different physiological effects, depending on their complement of carbon atoms. For example, in terms of raising serum cholesterol levels, stearic acid (found in meat and cocoa butter) appears neutral, while lauric and palmitic acids (found, for example, in tropical oils and dairy, respectively) might be far more problematic.
So the consensus as of 2011, according to Astrup, was that “the effect of a specific food on risk of CVD cannot be determined simply on the basis of the fatty acid profile.” Indeed, “the total matrix of a food is more important.” Nevertheless, he advised, a “healthy dietary pattern is primarily plant-based and low in SFA.”
Perhaps, but 2011 was a lifetime ago in the increasingly volatile world of nutrition science. Important questions about fats, CHO, and various biomarkers and risk factors for CHD were raised and recognized. But ensuing research would reveal quite clearly that the macronutrient wars were just warming up.
The Macronutrient Wars Rage On.
Controversy has beset the diet-heart hypothesis (D-Hh) from its very inception. Throughout the 1950s and 1960s, the idea’s founder, Ancel Keys, was accused of “cherry picking” evidence to advance his career. If the Masai warriors of Kenya and Tanzania can subsist healthfully on raw milk, meat, and blood, the detractors prodded, why should Westerners shun butter, omelets, and cheeseburgers?(22) Since then, every official recommendation urging Americans to consume less cholesterol and saturated fatty acids (SFA) to reduce their risks of coronary heart disease (CHD) and cardiovascular disease (CVD) has suffered incessant criticism.
In recent months, the attacks have only intensified. Cardiologist Aseem Malhotra scolded researchers for their allegedly outdated obsession over the link between SFA and low-density lipoprotein cholesterol (LDL-C). Yes, consumption of SFA tends to elevate blood LDL-C levels, he conceded. But that increase “seems to be specific to large, buoyant (type A) LDL particles,” and not the “small, dense (type B) particles” implicated in CVD.(23) More dangerous type B particles are actually “responsive to carbohydrate intake,” Malhotra insisted. Indeed, “saturated fat has been found to be protective.”
So, while the majority continues to condemn the SFA typically found in animal products and tropical oils, the minority has instead begun to impugn carbohydrates (CHO), particularly the refined variety favored by Western palates. Most researchers agree that excess energy from whatever sources leads to obesity, diabetes, and CHD. But authors of the official guidelines spurn minority influences and continue to recommend replacement of SFA with their polyunsaturated counterparts (PUFA).
In response, indignant writers have pummeled the professional literature with papers hostile to official policies. For example, following a meta-analysis of twelve studies involving 7150 participants, Austrian nutritionist, Lukas Schwingshackl, argued that replacing SFA with PUFA “showed no significant benefit in the secondary prevention of coronary heart disease.”(24) NIH biochemist Christopher Ramsden raised the stakes after examining newly-recovered data from the Sydney Diet Heart Study. He decided not only that linoleic acid (LA), a common omega-6 (n-6) PUFA, “did not provide the intended benefits,” but also that its substitution for SFA “increased all-cause mortality, cardiovascular death, and death from coronary heart disease.”(25) Uffe Ravnskov concurred, emphasizing studies associating consumption of PUFA with inflammation, immune system suppression, decreased high-density lipoprotein cholesterol (HDL-C) levels, and an increased risk of many cancers.(26)
At that point, CVD researcher (and Ravnskov co-author) James DiNicolantonio stepped in to summarize the detractors’ position. First, he stressed that the current outbreak of Western diabetes and obesity derives from overconsumption of CHO, not SFA. Second, the replacement of SFA with CHO only increases small, dense LDL particles and shifts the overall lipid profile toward decreased HDL-C, elevated triglycerides, and an increase in the apolipoprotein B-to-apolipoprotein A-1 ratio (ApoB/ApoA-1). Third, the substitution of SFA with n-6 PUFA only reduces HDL-C and raises the risk of cancer, CHD, and overall mortality. Finally, he argued, the PREDIMED and Lyon Diet Heart studies had demonstrated that Mediterranean-style diets reduce cardiovascular events, cardiovascular mortality, and all-cause mortality relative to either low-fat diets or the AHA’s “prudent” diet. Thus, DiNicolantonio concluded, those responsible for the official guidelines “should assess the totality of evidence and strongly reconsider their recommendations.”(27)
When I contacted him, DiNicolantonio maintained that the D-Hh “has never been proven.” SFA acids might raise LDL-C levels, he advised, but an increased risk of CHD simply doesn’t follow. Small, dense LDL particles are more atherogenic, or “more capable of penetrating a damaged endothelium”—which likely results from a “high refined carb/sugar diet,” among other things. To the contrary, he argued, the SFA found in whole, unprocessed foods like red meat can provide immune-boosting properties and stability against oxidation.
Meanwhile, minority support continued to mount. In his very controversial meta-analysis of forty-nine observational studies and twenty-seven randomized controlled trials, Cambridge University epidemiologist, Rajiv Chowdhury, found null associations between coronary risk and both total SFA and monounsaturated fatty acids (MUFA), along with a statistically non-significant association between coronary risk and PUFA supplementation.(28) Likewise, Tulane University epidemiologist, Lydia Bazzano, conducted a twelve-month randomized, parallel-group diet intervention trial of 148 healthy men and women, and found that participants who completed the low-CHO regime lost more weight and presented with fewer CVD risk factors compared to subjects who completed the low-fat program.(29)
Predictably, the majority’s response was immediate. Harvard University nutritionist, Maryam Farvid, performed her own systematic review and meta-analysis of prospective cohort studies to examine the effect of increased LA consumption in healthy subjects. Contrary to Ramsden’s and Chowdhury’s findings, Farvid revealed a linear inverse association between the predominant n-6 PUFA and CHD. LA’s “cardio-protective effects,” she argued, included a “9% lower risk of total CHD and 13% lower risk of CHD deaths” with a “5% increase in energy from LA, replacing SFA.”(30)
While acknowledging Ramsden’s and Ravnskov’s concerns relating to LA, Farvid nevertheless resolved that such fears remained unsupported by both prospective studies and randomized controlled feeding trials. Even so, she conceded, “the effects of LA on heart disease risks are difficult to predict,” and diets high in LA may increase lipid oxidation and “play a role in the pathogenesis of cancer.”
Farvid also distinguished her analysis from those of her minority predecessors. Ramsden, she noted, based his results primarily on one short-term trial from the 1960s restricted to a small sample of unhealthy men. And because the partial hydrogenation of vegetable oils was then common, she suggested, Ramsden’s findings might have been confounded by the trans-fats found in margarines high in LA. Chowdhury, on the other hand, had based his analysis on a limited number of studies and was unable to compare LA with SFA or any other macronutrient. In a separate critique, Walter Willet—Farvid’s co-author and Harvard colleague—accused Chowdhury of committing “multiple serious errors and omissions” and creating unnecessary confusion. When SFA are replaced with PUFA or MUFA in the form of olive oil, nuts and other plant oils, he disputed, “we have much evidence that risk will be reduced.”(31)
The Bazzano study’s relevance to the D-Hh and the guidelines are less clear than many imagine. While certain popular media outlets seized on these results to extol the virtues of bacon and eggs breakfasts, for example, Bazzano had actually urged subjects in both the low-fat and low-CHO intervention groups to consume less SFA and more unsaturated fat.(32) Further, the low-fat cohort was advised merely to diminish overall fat intake to thirty percent of total calories which is “hardly low,” as NYU professor of nutrition, Marion Nestle, reminded me.
Finally, an interesting variation on the minority theme alleges first, that dietary SFA do not affect plasma SFA levels, and second, that excessive intake of refined CHO does raise either SFA or palmitoleic MUFA plasma levels via de novo lipogenesis in the liver, thereby increasing the risk of CHD.(33, 34) But James Kenny, nutrition researcher at the low-fat advocacy Pritikin Longevity Center judges this argument a “gross over-interpretation of the data.” Yes, the liver will convert sugar to both SFA and MUFA when its glycogen stores are maximized. Nonetheless, Kenney told me, even under these circumstances, mere trivial portions of ingested CHO will be converted and only when consumed far in excess of energy needs. Regardless, he added, “low-fat/high-CHO diets composed of vegetables, whole-grains, and fruit decrease total cholesterol and ApoB-containing lipoproteins, reduce inflammation, and may actually reverse atherosclerosis.”
A Crossroads; What Should We Eat?
So where does the D-Hh stand at the dawn of 2015? My conversations with the experts could not have yielded more starkly conflicting opinions. According to Ravnskov, for example, “the diet-heart idea is the greatest medical scandal in modern time.” By contrast, Tufts University researcher Alice Lichtenstein—instrumental in generating the AHA’s most recent guidelines—wouldn’t dignify minority rhetoric with a response. She maintains that “the observational and intervention data are entirely consistent” and “support substituting PUFA for saturated fat to decrease the risk of CVD.” Kenney was less reluctant to characterize minority views as “fringe” and “pseudoscientific.”
Regrettably, the members of neither faction are likely to abandon their positions soon. So the general public is left shrugging its collective shoulders. Nonetheless, important inferences can be drawn from a century’s worth of diet-heart literature. With respect to biomarkers, for example, we might soon reconsider our preoccupations with LDL-C and HDL-C per se. Non-HDL-C (total cholesterol minus HDL-C) and total ApoB containing lipoproteins likely provide more revealing indicators of cardiovascular risk and, correspondingly, particle quantity is more critical than size. Concern also grows that, while SFA and MUFA increase HDL-C, much of that HDL-C can become dysfunctional and actually pro-atherogenic.
As intense friction between research communities continues to mount, the USDA Dietary Guidelines Advisory Committee has proposed meaningful revisions for 2015.(35) Recommended ceilings for the consumption of dietary cholesterol—currently set at 300 milligrams per day—will likely be rescinded, overturning forty years of advisory precedent. Further restriction of refined sugars was also advised. On the other hand, the Committee apparently remains committed to the D-Hh and long-standing recommendations to reduce intake of SFA.
In an editorial response to Zoe Harcombe’s recent condemnation of the original (and persisting) government guidelines, British cardiologist Rahul Bahl reflected on their empirical support overall. Given the results of more recent analyses detailed here, he found Harcombe et al.’s results and conclusions “unsurprising” but not necessarily convincing.(36) “There remain reasons to postulate a causal connection” he argued, “between fat consumption and heart disease.” First, the epidemiological and ecological evidence suggest such a link and, second, we should expect certain randomized controlled trials to produce negative results given the capriciousness of human behavior. On the other hand, Bahl reasoned,
There is certainly a strong argument that an overreliance in public health on saturated fat as the main dietary villain for cardiovascular disease has distracted from the risks posed by other nutrients such as carbohydrates. Yet replacing one caricature with another does not feel like a solution. It is plausible that both can be harmful or indeed that the relationship between diet and cardiovascular risk is more complex than a series of simple relationships with the proportions of individual macronutrients.
Beyond that, the inherent limitations of nutrition science tend to frustrate the public’s demand for concrete conclusions. Confounding factors, including genetics, are difficult to both identify and account for, especially in cross-sectional observation studies. Assessing the significance of any single risk factor for a chronic disease of multifactorial etiology is a knotty problem at best. And because people consume foods rather than mere nutrients, intake itself complicates the issue. High-fat diets, for example, might be loaded with sugar too, and are often low in fiber, antioxidants, flavonoids, folate, and carotenes. Study methodologies vary considerably as well, and participant behaviors often render findings non-replicable and scientifically suspect. Finally, given the establishment of a paradigm like the D-Hh, ethical considerations often make meaningful test conditions impossible.
Even so, a sensible combination of nutrition science and sound reasoning proves very helpful. First, ignore the popular media. Some truths simply don’t sell advertising. Second, nutritionally irredeemable trans-fats should be eliminated from our diets unceremoniously. Otherwise, as Bahl and others suggest, focus on whole foods over fatty acids and remember that humans never evolved to consume processed foods, including refined fats and oils. Third, learn to prepare and spice lean, unprocessed meats and fresh vegetables to otherwise unembellished satisfaction. Add a few fruits and nuts and avoid everything else—especially refined sugars and starches. Finally, exercise regularly and vigorously, avoid excess adiposity, and don’t smoke. With that, what within our control could possibly go wrong?
(1)Harcombe, Z., J.S. Baker, S.M. Cooper, et al. 2015. Evidence from randomized controlled trials did not support the introduction of dietary fat guidelines in 1977 and 1983: a systematic review and meta-analysis. Open Heart 2015;2:e000196. DOI:10.1136/openhrt-2014-000196.
(2)Kolata, G. 1985. Heart panel’s conclusions questioned. Science 227:40-41.
(3)Lipid Research Clinics Program. 1984. The Lipid Research Clinics Coronary Primary Prevention Trial. 1. Reduction in incidence of coronary heart disease. Journal of the American Medical Association 251(3):351-364.
(4)Steinberg, D. 1985. Heart panel’s conclusions. Science 227:582.
(5)Anitschkow, N., S. Chalatov, C. Muller, et al. 1913. Uber experimentelle cholesterinsteatose: Ihre bedeutung fur die entehung einiger pathogischer prozessen. Zentralblatt fur Allgemeine Pathologie und Pathologiche Anatomie 24:1-9.
(6)Keys, A. 1953. Atherosclerosis: a problem in newer public health. Journal of Mt. Sinai Hospital, New York 20(2):118-139.
(7)Yerushalmy, J. and H. Hilleboe. 1957. Fat in the diet and mortality from heart disease. A methodological note. New York State Journal of Medicine 57:2343-54.
(8)DiNicolantonio, J. 2014. The cardiometabolic consequences of replacing saturated fats with carbohydrates or omega-6 polyunsaturated fats: Do the dietary guidelines have it wrong? Open Heart 2014;1:e000032. doi:10.1136/openhrt-2013-000032.
(9)Keys, A. 1970. Coronary heart disease in seven countries. Circulation 41(Suppl. 1):1-211.
(10)Teicholz, N. 2014. The big fat surprise: why butter, meat & cheese belong in a healthy diet. NY: Simon & Schuster.
(11)Blackburn, H. 2014. In defense of U research: The Ancel Keys legacy. The Star Tribune (July 17). Online at http://www.startribune.com/opinion/commentaries/267581481.html
(12)Ravnskov, U. 1998. The questionable role of saturated and polyunsaturated fatty acids in cardiovascular disease. Journal of Clinical Epidemiology 51(6):443-460.
(13)Broad, W.J. 1980. Academy says curb on cholesterol not needed. Science 208:1354-55.
(14)Nestle, M.. 2013. Food Politics: How the food industry influences nutrition and health. Berkeley: University of California Press.
(15)Jakobsen, M.U., E.J. O’Reilly, B.L. Heitmann, et al. 2009. Major types of dietary fat and risk of coronary heart disease: a pooled analysis of 11 cohort studies. American Journal of Clinical Nutrition 89:1425-32.
(16)Siri-Tarino, P.W., Qi Sun, F.B. Hu, and R.M. Krauss. 2010a. Meta-analysis of prospective cohort studies evaluating the association of saturated fat with cardiovascular disease. American Journal of Clinical Nutrition 91:535-46.
(17)Siri-Tarino, P.W., Qi Sun, F.B. Hu, and R.M. Krauss. 2010b. Saturated fat, carbohydrate, and cardiovascular disease. American Journal of Clinical Nutrition 91:502-09.
(18)Letters to the editor. 2010. American Journal of Clinical Nutrition. 92:458-61.
(19)Stamler, J. 2010. Diet-Heart: a problematic revisit. American Journal of Clinical Nutrition 91:497-99.
(20)Ramsden, C.E., J.R. Hibbeln, S.F. Majchrzak, and J.M. Davis. 2010. n-6 fatty acid-specific and mixed polyunsaturated dietary interventions have different effects on CHD risk: a meta-analysis of randomized controlled trials. British Journal of Nutrition 104:1586-1600.
(21)Astrup, A., J. Dyerberg, P. Elwood, et al. 2011. The role of reducing intakes of saturated fat in the prevention of cardiovascular disease: where does the evidence stand in 2010? American Journal of Clinical Nutrition 93:684-88.
(22) Indeed, Masai serum cholesterol levels were minimal. Their autopsies, however, later revealed severe atherosclerotic lesions.
(23)Malhotra, A. 2013. Saturated fat is not the major issue: Let’s bust the myth of its role in heart disease. British Medical Journal 347:f6340.
(24)Schwingshackl, L. and G. Hoffmann. 2014. Dietary fatty acids in the secondary prevention of coronary heart disease: A systematic review, meta-analysis and meta-regression. British Medical Journal Open 4:e004487.
(25)Ramsden, C.E., D. Zamora, L. Boonseng, et al. 2013. Use of dietary linoleic acid for secondary prevention of coronary heart disease and death: Evaluation of recovered data from the Sydney Diet Heart Study and updated meta-analysis. British Medical Journal 346:e8707.
(26)Ravnskov, U., J. DiNicolantonio, Z. Harcombe, et al. 2014. The questionable benefits of exchanging saturated fat with polyunsaturated fat. Mayo Clinic Proceedings 89(4):41-53.
(27)DiNicolantonio, J. 2014. The cardiometabolic consequences of replacing saturated fats with carbohydrates or Ω-6 polyunsaturated fats: Do the dietary guidelines have it wrong? Open Heart 1:e000032.
(28)Chowdhury, R., S. Warnakula, S. Kunutsor, et al. 2014. Association of dietary, circulating, and supplement fatty acids with coronary risk: A systematic review and meta-analysis. Annals of Internal Medicine 160:398-406.
(29)Bazzano, L.A., T. Hu, K. Reynolds, et al. 2014. Effects of low-carbohydrate and low-fat diets: A randomized trial. Annals of Internal Medicine 161(5):309-318.
(30)Farvid, M.S., D. Ming, P. An, et al. 2014. Dietary linoleic acid and risk of coronary heart disease: A systematic review and meta-analysis of prospective cohort studies. Circulation 130:1568-1578.
(31)Willet, W. 2014. Dietary fat and heart disease study is seriously misleading.
http://www.hsph.harvard.edu/nutritionsource/2014/03/19/dietary-fat-and-heart-disease-study-is-seriously-misleading/ (posted March 19, 2014).
(32)ABC News. 2014. Low-carb diet trumps low-fat diet in weight-loss study. http://abcnews.go.com/GMA/video/low-carb-diet-trumps-low-fat-weight-loss-25211617 (posted September 2, 2014).
(33)Kuipers, R.S., D.J. de Graff, M.F. Luxwolda et al. 2011. Saturated fat, carbohydrates and cardiovascular disease. Netherlands Journal of Medicine 69(9):372-378.
(34)Volk, B.M., L.J. Kunces, D.J. Freidenreich et al. 2014. Effects of step-wise increases in dietary carbohydrates on circulating saturated fatty acids and palmitoleic acid in adults with metabolic syndrome. PLOS ONE DOI:10.1371/journal.pone.0113605.
(35)Scientific Report of the 2015 DGAC, http://www.health.gov/dietaryguidelines/2015-scientific-report/PDFs/Scientific-Report-of-the-2015-Dietary-Guidelines-Advisory-Committee.pdf; O’Connor, A. 2015. Nutrition panel calls for less sugar and eases cholesterol and fat restrictions. New York Times. http://well.blogs.nytimes.com/2015/02/19/nutrition-panel-calls-for-less-sugar-and-eases-cholesterol-and-fat-restrictions/?_r=0.
(36)Bahl, R. 2015. The evidence base for fat guidelines: a balanced diet. Open Heart 2015;2:e000229. DOI:10.1136/openhrt-2014-000229.