The conventional theory of the cause of coronary artery disease has been driven by the so called “Diet Heart Hypothesis.” This was the idea that diets that were high in saturated fats raised blood cholesterol which caused atherosclerotic blockage of arteries. This was started by several flawed studies. The first was done by Nikolai Anichkov in 1913. He found that when he fed rabbits diets that were high in cholesterol, that they developed coronary atherosclerosis. This was and still is considered a groundbreaking study proving the role of cholesterol in heart disease. The problem was that rabbits are herbivores, not accustomed to eating cholesterol as there is no cholesterol in plant foods. Hence their bodies were not adapted to consuming dietary cholesterol. When the same experiments were done using dogs which are carnivores, no such coronary lesions developed. Furthermore, when it was eventually agreed that dietary cholesterol played no role in atherosclerosis in humans, the American Heart Association and USDA eventually dropped the 30 year old recommendation that all Americans should restrict dietary cholesterol. So why didn’t they drop the recommendations to limit saturated fat? Well, the other early study that was influential and that was also flawed was the Seven Countries Study by a physiologist named Ancel Keyes where he looked at dietary saturated fat intake in various countries and compared it with the incidence of heart disease in those countries. The problem with his study was that he hand picked the seven countries that fit his hypothesis and ignored those that didn’t, namely France and Switzerland which have among the highest saturated fat intake and the lowest incidence of heart disease. None the less, he was able to convince the scientific community that his hypothesis was correct and subsequently convinced the US government to formulate guidelines telling people to limit saturated fat intake. What happened in studies that tested the hypothesis that reduction in saturated fats reduced CAD? Several have been done. Two of the most notable studies where polyunsaturated fats were substituted for saturated fats were the Minnesota Coronary Experiment and the Sydney Diet Heart Study. The Minnesota Coronary Experiment was done by Ancel Keys’ own staff at the University of Minnesota. Here, long term residents in psychiatric hospitals were put on regular vs diets where saturated fats were replaced by polyunsaturated vegetable fats. The results were the opposite of what was expected. The experimental group had more deaths due to heart disease as well as total deaths. The authors waited 15 years to publish these results. When author of Good Calories Bad Calories, Gary Taubes asked the principle investigator why he waited to long, the response was that they were disappointed with the results. See related recent editorial about this (1). The Sydney Diet Heart Study showed similar results ie higher coronary and all cause mortality in the group that ate the diet where polyunsaturated fats were substituted for saturated fats. For every 30 point reduction in total cholesterol, there was a 20% increase in deaths. The results of this study also laid dormant in storage until they were recently discovered and published (2).
The problem with the diet heart hypothesis is that not only does intake of saturated fat, which can raise total cholesterol, but also raises HDL or good cholesterol, not correlate with heart disease, but that blood cholesterol itself doesn’t correlate with heart disease either. A landmark ongoing study of diet, cholesterol and heart disease that has been going on since 1948 is the Framingham Heart Study. There 5,200 residents of the town of Framingham in Massachusetts have been followed since 1948. The initial much heralded finding was that serum cholesterol correlated with heart disease. However as the study went on it later turned out that total cholesterol had weak correlation with heart disease in men only below the age of 50 and no correlation with heart disease after 50. In women there was no correlation with heart disease before 50 and a reverse correlation after 50. That is after the age of 50, the higher a woman’s cholesterol, the longer she lived. See Gary Taubes’ NYT article describing this (3). Several other studies have shown the same thing, ie basically, the higher your cholesterol, the longer you live. See Verner Wheelock’s excellent blog on this (4).
Another damning study is the “UCLA Study” where 137,000 patients who were admitted for heart disease from 541 hospitals were studied for 5 years (5). They found that over 75% had LDL cholesterol levels below 130, considered to be a good level, and 50% had levels below 100, a level considered to be ideal and well below normal. A meta analysis published in BMJ in 2016 looked at 19 studies involving 68,0000 patients, found an inverse relationship between LDL levels and mortality in patients over 60, similar to Framingham findings above (6).
So if there is no data that high total or in particular LDL cholesterol correlates with heart, disease why is there still belief of it’s role in causing heart disease? Enter statins and the medical industrial complex. When belief was high that blood cholesterol correlated with heart disease, the search was on to find classes of drugs that would lower blood cholesterol. The most powerful class discovered was the statin class. Statins inhibit production of cholesterol which causes and increase in liver LDL which increases removal of cholesterol particles. Several studies were done by the manufacturers that showed reduction in cardiac events, however reduction in total mortality was weak. And when analyses were done of long term benefits it was found that they might prolong live by only 3-4 days for every 5 years of use. However there is conflicting data on the benefit of statins as well. A recent study of 100,000 Medicare patients who had suffered a myocardial infarction showed that those that were intolerant of statins, ie could not and did not take them, had slightly higher incidences of heart events but lower overall mortality (7). There is also data that the benefit that has been shown with statins has more to do with anti inflammatory effects than cholesterol lowering. Numerous studies have shown no correlation with the degree of LDL lowering and degree of benefit in statin trials.
So when it became apparent that LDL itself had basically no correlation with heart disease, investigators started to look at subgroups of lipid particles such as small dense LDL and apoB which is a marker of total LDL particles. It turns out that increased levels of small dense LDL and apoB particles do correlate with incidence of CAD. So it was then believed that if a person has a high number of small dense LDL particles ie high apoB count, and since these particles do correlate with heart disease, that these particles must cause atherosclerosis by getting into the walls of arteries and triggering atherosclerosis. Ok, so, finally, a study of one of the new potent LDL lowering drugs, PCSK9 inhibitors, evolocumab, was recently published, the FOURIER TRIAL (8). This drug lowered LDL significantly to very low levels. It also raised HDL and lowered triglycerides. Basically, all the lipid parameters and particle numbers were “fixed” to all be in what would be considered ideal ranges. LDL cholesterol was reduced to 30. Apo B was reduced by 49%. Other lipid parameters were fixed as well. Lp(a) was reduced by 27%. HDL was increased by 8.4%. Apo A1 was increased by 6.5%. Triglycerides were reduced by 16%. Basically all the numbers were in as good a range as could be hoped for. They were basically “perfect.” So how did the patients do? In the patients on the drug, the incidence of events over the two year period was 9.8% compared to 11.3% in the control group. This is hardly earth shattering. So, in these patients who had as “good” a total lipid profile as could possibly be imagined, the incidence of events was 10% over two years, which is still considered extremely high as far as percent risk scores go. According to the Framingham risk calculator, a 20% ten year risk of coronary events is the highest category for cardiac risk. These patients had a 10% incidence of cardiac events in only two years. That is, these patients still suffered cardiac events at what is considered an extremely high rate. Cardiac events were not reduced anywhere close to what would be considered a low cardiac risk category. What’s more, you guessed it, the subjects on this powerful LDL lowering drug had higher total mortality. So, again, you can basically fix all the lipid particle numbers of at risk patients to what would be considered perfect lipid profiles, yet they still had a high incidence of cardiac events.
So when it became apparent that LDL was not such a good indicator of cardiac risk along came Ronald Kraus who showed us that there were two main types of LDL particles ie small dense or so called dangerous LDL particles that did have a correlation with heart disease and large “fluffy” LDL particles that supposedly did not. The small dense type was associated with diabetes and metabolic syndrome (more on that below) and was called pattern B. The large fluffy type was called pattern A. In that pattern B is found most often in patients with metabolic syndrome and diabetes, which does correlate with significantly higher risk of heart disease, this distinction made sense. This is also the basis for the now much touted Apo B to Apo A ratio which basically accounts for total number of LDL particles vs HDL particles which takes into effect the number of LDL particles ie the smaller LDL particles having the higher number. However, this relationship also does not hold up completely either. In a recent study (9), it was found that in patients who had suffered an MI, those with pattern B actually had a LOWER follow up mortality than those with pattern A! What the..?
Maybe lipid moieties or particles, ie LDL, apo B, apo A ratios, HDL have no causative role in coronary artery disease whatsoever, rather they are merely markers of a condition that is the real cause of coronary artery disease, ie insulin resistance and what is probably the real cause of atherosclerosis, high circulating insulin levels.
In 1988, at the Banting lecture series, Gerald Reavens presented his discovery of insulin resistance and metabolic syndrome which he called “syndrome X.” This syndrome consists of resistance of tissues, to effects of insulin, causing the pancreas to secrete more and more insulin to overcome this resistance in order to maintain glucose homeostasis ie normal glucose levels. It is characterized by high resting and food stimulated insulin levels. The insulin resistant condition or metabolic syndrome has other characteristic features, namely central obesity, hypertension, elevated blood sugar, lipid abnormalities, mainly elevated triglycerides and lowered HDL cholesterol and finally vascular disease. Insulin resistance is most likely caused by the consumption of large amounts of sugar over long periods of time. This sugar consumption has an affect on the liver whereby the liver can’t handle the large amounts of the fructose part of sugar and the liver converts it to fat which accumulates in the liver. This leads to initially insulin resistance within the liver. This role of sugar in causing insulin resistance which starts in the liver and the subsequent development of metabolic syndrome and the diseases associated with it is eloquently laid out by Gary Taubes in his outstanding book, The Case Against Sugar. As the pancreas secretes more and more insulin to overcome this liver resistance, other organs such as muscle and fat or adipocytes then become resistant as well, probably as a result of more and more exposure to higher and higher insulin levels. The pancreas compensates by secreting more and more insulin as the insulin resistance becomes worse and worse until eventually it can’t secrete enough insulin to keep glucose levels controlled and the blood glucose rises and a person is then identified as diabetic. The whole time insulin levels are rising due to this sugar induced insulin resistance, but while blood glucose levels are still normal, the individual has what is called prediabetic. As Gary Taubes describes in The Case Against sugar, consumption in the US and throughout the world has gone up over 20 fold since the beginning of the 19th century and diabetes has paralleled this going up incidence over 8 fold over the past 100 years. It is estimated that 50% of the US population has diabetes or prediabetes. 25% of Americans have fatty liver. Not only is insulin resistance associated with obesity, diabetes, and vascular disease, it is the cause of the vast majority of those conditions. It is felt that over 80% of hypertensive patients and 80% of cardiac patients are either diabetic or pre diabetic and if one uses the insulin tolerance test developed by Robert Kraft, one could identify nearly all patients with coronary artery disease as diabetic or pre diabetic. Recent studies have identified fatty liver as a stronger predictor of vascular disease than any other cardiac risk factor including total cholesterol, LDL, HDL levels (10, 11).
So what about lipids? Numerous studies have identified HDL and triglycerides as the strongest predictors of heart disease as far as lipid markers go. As stated above small dense LDL and high apoB to apoA ratios also correlate with heart disease. However, what all these so called atherogenic lipid marker patterns have in common is that all are characteristic of the metabolic syndrome. In fact the most “atherogenic” patterns of lipids ie high triglycerides, low HDL, high apoB, small dense LDL are all only seen in metabolic syndrome, strongly implicating metabolic syndrome in the cause of atherosclerosis. So is it these lipid particles that all the researchers and drug companies have focused on for so long that are the atherogenic trigger of vascular disease caused by metabolic syndrome? Or is it something else? As the FOURIER trial described above showed, when all these particles were normalized by drugs, there was still an extremely high level of cardiac events. You have all these lipid particles brought to ideal levels yet there was still a high level of cardiac events. So if it is not the lipid particles that are characteristic of cit metabolic syndrome that causes vascular disease, what is it? I would propose that there is a strong case that insulin itself and high insulin levels of the metabolic syndrome are the primary drivers of vascular atherosclerotic disease and that all the lipid particles and particle patterns associated with the metabolic syndrome are nothing more than markers of the syndrome ie high insulin levels, and have no causal roll in the promotion of atherosclerosis whatsoever. The true feature of metabolic syndrome that damages the lining of arteries leading to atherosclerosis is not the so called atherogenic lipid particles, but rather the effects of a high insulin level itself on the walls of arteries that causes them to develop atherosclerosis. Taubes discusses this possibility in Good Calories Bad Calories. On page 189, he cites Harvard researchers who in 2005 felt vascular disease was possibly due to direct action of insulin on vessel walls. Rabbits fed high cholesterol diets “developed plaques throughout their arteries.” But diabetic rabbits (type 1) did not until they were infused insulin along with the high cholesterol diets wherein “plaques and lesions will promptly blossom everywhere.” “This phenomenon was first reported in 1949 in rabbits, and then, a few years later, in chickens, and later in dogs, too.” “In the late 1960’s , Robert Stout of Queen’s University on Belfast published a series of studies reporting that insulin enhances the transport of cholesterol and fats into the cells of the arterial wall and stimulates the synthesis of cholesterol and fat in the arterial lining.” A paper published in 2016 did a genetic analysis and concluded, “These results support the causal role of insulin in the pathogenesis of CHD. Efficient treatment and prevention of insulin resistance is essential to prevent future CHD events ” (12). The following study among others showed that “although several old studies provided conflicting results, the majority of large observational studies show strong dose-dependent associations for injected insulin with increased CV risk and worsened mortality.” (13,14.)
So what is the cause of coronary artery disease? As I detailed above I don’t believe it has anything to do with so called “atherogenic” lipids. I believe the prime cause of the development of atherosclerotic vascular lesions is the effect of high insulin levels which are the prime characteristic of metabolic syndrome, insulin resistance and type 2 diabetes. The insulin causes the initial injury to the vessel wall, then the cascade of events ie macrophages ingesting LDL particles, foam cells etc are secondary reactions. Many, including myself, believe that the development of atherosclerotic plaques are part of the body’s reaction and attempted repair mechanism to the initial wall injury. Ie blaming the cholesterol deposits and plaque development is like blaming the fire department and fire trucks for the damage of the fire. I feel that in most cases the majority of the initial artery wall damage is done my insulin but other factors, mainly hypertension alone can cause the initial injury as well in some individuals. Most cases of hypertension are associated with metabolic syndrome, again the result of too much sugar consumption for too long a period of time. However not all hypertensives have metabolic syndrome and high insulin levels. Coarctation of the aorta is a congenital condition where the aorta has a narrowing in the arch. This causes elevated blood pressure in the head, arms and heart arteries, but not the lower extremities. These individuals obviously do not have metabolic syndrome. They develop severe atherosclerosis in the arteries of their heart and brain and usually die in their 30’s and 40’s of myocardial infarctions and strokes. So something has to cause the initial arterial wall injury be it high insulin levels or the effect of hypertension, and the atherosclerotic process is nothing more than the body’s reparative response to the initial injury and lipid particles and the plaque forming process are nothing more than the firemen and fire trucks trying to put out the fires.
To be continued…..