In the context of coronary artery disease (CAD) and in particular atherosclerosis, blood lipids and total cholesterol and LDL cholesterol in particular, have over the years achieved a near dominant position, and what was merely a hypothesis has progressed to a universally accepted truth on a par with the laws of Newton or thermodynamics. A corollary to this hypothesis is that LDL levels are directly related to the risk of developing atherosclerosis and its progression. But as Karl Popper pointed out, hypotheses are made to be falsified, scientific progress requires efforts in this direction and as well requires the recognition rather than the out-of-hand rejection or disregard of results that provide grounds for falsification. The hypothesis that total or LDL cholesterol is associated in general with the incidence or progression of atherosclerosis has over the years been confronted with a large number of inconvenient questions that relate to results which appear to provide falsification. What is important is that the failure to recognize these inconvenient questions or reject or ignore them has significantly delayed progress in developing alternative hypotheses, this to the detriment of patients and preventive medicine in this critical area.

In a debate on lipid lowering published in 2008 in the journal Circulation,1 Scott Grundy, a strong advocate of the cholesterol hypothesis and lipid lowering, called LDL the driving force of atherosclerosis. He first cites animal studies using as a reference a paper that provided justification for the 2002 National Cholesterol Education Program guidelines. If one looks at that paper for the animal studies, the rabbit is the poster child, but as critics have repeatedly pointed out in the literature and in books dealing with the cholesterol hypothesis, rabbits are rather different than humans in that they evolved to eat quite different diets, and they never ate cholesterol except when forced to in studies trying to tie cholesterol to atherosclerosis. This 2002 paper also points out that animals that do not develop atherosclerosis generally have LDL below 80 mg/dL, but no reference is given. But animals that have recently been held up as examples include elephants and rhinoceroses. Next, Grundy claims that epidemiological studies reveal a strong association between serum cholesterol levels and the prevalence of atherosclerotic cardiovascular disease. The one reference is to a paper published in 1994 by Law et al which deals with the relationship between total cholesterol levels, not LDL, and either fatal or non-fatal heart attacks. Some of the studies used by Law et al were not in fact epidemiological but clinical trials from the pre-statin era and used fibrates, niacin or both, but this therapy mainly addresses so-called atherogenic dyslipidemia, i.e. low HDL, high triglycerides, and the subclass of small atherogenic LDL particles. It is hard to see that while these drug regimes reduced the rate of events, this demonstrates that LDL cholesterol drives atherosclerosis when fibrates are now looked upon as one of the drugs of choice for raising HDL and lowering triglycerides. Even the broad generalization generated by the meta-analysis of Law et al seems inconsistent with a large body of literature concerning individuals who have no symptoms of heart disease. There now appears to be only a weak relationship between cholesterol and the risk of adverse coronary events in this category, and only for young men, and even this may be confounded by exaggerated hypertensive response to anger.2 In addition, epidemiological studies, which in the Law et al analysis, represented the vast majority of subjects, do not establish causality, a point made dozens of time each year in the medical literature. Yet the verb "drives" used by Grundy implies causality. The inadequacy of looking at acute events in order to judge the importance of cholesterol in the incidence and progression of atherosclerosis will be discussed below.

What is more important, Grundy is ignoring a vast body of literature covering several decades up to the present, that looked directly at the presence of atherosclerosis and its progression as a function of LDL or total cholesterol levels, i.e. the real question. As will be discussed, this literature contradicts the assertion that LDL is the driving force of atherosclerosis. Grundy also uses the lipid lowering argument. Problems with this argument are also discussed below. Nevertheless, paper after paper takes the same position as does Grundy and treats as an established fact that serum cholesterol drives atherosclerosis. The same message is delivered in TV ads.

Of particular significance has been the huge impact on the cholesterol-atherosclerosis hypothesis of the results of lipid lowering drug therapy on the risk of adverse events in individuals with established CHD. These results have been widely accepted as an absolute proof that there is a causative relationship to serum lipids and that this can be extrapolated to those who have not experienced an adverse CHD event, and to both genders and all ages including children as young as 8 years. This is in spite of the fact that it is acknowledged that the drugs in question exhibit non-lipid lowering (pleiotropic) effects which would be consistent with the treatment results without evoking lipid lowering in the mechanism, and that the pleiotropic mechanisms probably depend on the stage of acute coronary heart disease present.3 After all, beneficial results are sometimes seen in a few weeks, and even the most ardent believer in the cholesterol- atherosclerosis hypothesis and its "proof" through lipid lowering cannot attribute these rapid results to lipid lowering effects. These pleiotropic effects of statin drugs are currently the subject of a considerable research effort worldwide.

The "lowering' paradigm appears fundamentally flawed by virtue of the complexity of human biochemistry and microbiology and the inadequate understanding of this chemistry and microbiology when measured against what needs to be understood. To assert unequivocally and with absolute certainty that the introduction of a foreign chemical substance (e.g. a manmade drug) that is known to affect a certain pathway or process, does not also affect numerous other unrecognized processes and pathways requires understanding and knowledge that is not there and may never be there. To assume that one knows all that is necessary to make a causal connection by using a drug to lower some marker or substance appears to be a manifestation of unjustified arrogance. To quote Karl Popper again, "Our knowledge can only be finite, while our ignorance must necessarily be infinite."

This review will examine some important inconvenient questions in order to attempt to put the above-described problem in a more proper perspective. It is offered as an antidote to the constant barrage of statements in the medical literature, popular health magazines and TV ads for statin drugs, all of which assume the cholesterol-atherosclerosis hypothesis constitutes a fundamental truth. Believers in this hypothesis should find these questions inconvenient if not downright disturbing. The focus will be on the hypothesis that serum total cholesterol and LDL cholesterol are associated with the risk of developing coronary atherosclerosis and its progression The relationship to the risk of adverse events such a myocardial infarct or severe angina requiring emergency attention are regarded as a separate issue. That is, incidence and progression are to be considered as the primary issue, not events. The mechanisms, to the extent they are understood, appear different for initiation of atherosclerosis, its progression, and subsequent occurrence or recurrence of acute events. However, some studies of the presence or extent of coronary atherosclerosis, prompted by symptomatic indications of the presence of CAD, also provide insight. Fundamental to this discussion is the assumption that the development of coronary plaque is involved in the progression from being free of the risk of CAD to having risk, with the greater the extent of the plaque burden, the greater the risk of adverse aspects of this disease. There now appears to be general agreement about this.

INCONVENIENT QUESTIONS

The principal questions relate to the correlation of the extent of atherosclerosis and circulating cholesterol, especially total cholesterol (TC) and LDL cholesterol. These questions already surfaced many years ago and prompted a number of autopsy studies, an obvious approach. Questions have multiplied significantly with the advent of electron beam tomography (EBT) used to measure coronary plaque calcification, contrast enhanced CT which allows visualization of all plaque, and the widespread use of various other types of angiography which has also allowed the extent of CAD to be examined in the context of risk factors such as lipid levels. Thus the following inconvenient questions.

• Why do autopsy studies of the correlation between the extent of coronary atherosclerosis and serum cholesterol yield null results? The answer that the blood samples, mostly from accident or suicide victims, were obtained too long after death has been discredited by several studies. There appears to be no reason to suspect that these studies were carried out either incompetently and with selection bias.4
• Why did Hecht et al5 find that TC, LDL, and HDL did not correlate with either the extent or prematurely of calcified plaque burden in 1105 consecutive, asymptomatic individuals self- referred for EBT?
• Why did Hecht et al6 fail to find a correlation between LDL and the coronary calcium percentile (correlation coefficient 0.06 with a scatter plot showing no visible correlation) for 304 asymptomatic women?
• Why when type 2 diabetics without evidence of coronary heart disease were subjected to EBT, did Elkeles et al7 find that the progression of coronary calcium was not related to lipid risk factors in the PREDICT study?
• Why when 1653 men and women without a history of CHD were subjected to coronary CT angiography using contrast media did Johnson et al8 fail to find a correlation between total plaque burden (calcified and non-calcified) and total serum cholesterol, a result the authors indicate agreed with other studies? This result was apparently adjusted for statin use. This study, since it looked at total plaque, in fact removes a potential criticism of those studies listed above that the calcified plaque burden is only a fraction of the total burden. Also it was recently shown that 70% of individuals with zero calcium score had no plaque when examined by coronary CT angiography.9
• Why in a study by Kim et al10 of 544 Korean men and women considered at low risk for CAD who underwent coronary angiography was the apolipoprotein ratio Apo B/Apo A-1 the only variable that differentiated patients with from those without CAD when the criterion for having CAD was = 50% stenosis in at least one coronary artery?
• Why in a study by Horimoto et al11 involving 437 patients who underwent coronary angiography for suspected CAD, did a multivariate analysis find that only the Apo B/Apo A-1 ratio, among lipid parameters, was an independent predictor of coronary atherosclerosis?
• Given the last two questions, why was "no clinically useful relationship" found between LDL and plasma Apo B, the apparently "bad" apolipoprotein, and that none of the traditional lipids added significant predictive information to the Apo B/Apo A-1 ratio in the context of cardiovascular risk?12
• Why in a study of 182 men undergoing diagnostic coronary angiography motivated by chest pain syndromes or abnormal stress test results, was the degree of CAD related only to age, HDL and free testosterone? Why in addition, when the group was divided into two groups, those with and without conventional risk factors which included hypercholesterolemia, was the degree of CAD essentially the same in both groups and the increase of CAD with age not augmented by these risk factors?13
• Why in a study of type 2 diabetics using EBT to determine the extent of atherosclerosis, was the risk of a first CVD event strongly dependent on the calcium score, but the calcium score was independent of serum lipids over the full range of calcium scores (Agatston) from 0 to 10,000.14 A score above 400 reflects a high plaque burden.
• Why is coronary calcium progression "primarily related to baseline severity of coronary calcification with little relationship to standard cardiovascular factors or to modification of LDL cholesterol levels?"15 If one examines a large number of relevant EBT studies, among the standard risk factors, association of progression of atherosclerosis with total cholesterol or LDL is either nonexistent, very weak, or statistically insignificant. The few studies that show a significant association are thus inconsistent with the main body of evidence, suggesting no real association at all. HDL is frequently the only form of cholesterol with a significant association, but the results again are inconsistent.
• How does one explain the apparent paradox that if circulating cholesterol is a significant risk factor for the development and progression of atherosclerosis, even in young adults, cholesterol is also the only raw material for the endogenous synthesis of a variety of essential hormones? It is odd that this could happen during the evolution of our species?
• If circulating cholesterol is so dangerous, why does one find that once individuals with familial hypercholesterolemia (FH) approach middle age, they have life expectancies which converge on that found in the general population in spite of prolonged exposure to highly elevated total and LDL cholesterol?4 Furthermore, why when comparisons are made between those with FH who have cardiovascular disease (CVD) and those who do not, the most common result after correcting for confounding, is that there are no statistically significant differenced between the levels of total cholesterols and LDL in those with and without CVD, in spite of these individuals all having very high levels of both lipids.16 Why for individuals with FH, did Jensen et al17 find that age-adjusted coronary calcium scores were not associated with cholesterol as assessed by either plotting or correlation coefficients and that when those with and without coronary heart disease were compared, there was no significant difference in total untreated cholesterol? Finally, as Awan et al pointed out in 2008, the lack of correlation between total cholesterol and aortic calcium scores in FH patients raises the concern that the vascular calcification may progress independent of marked decrease in total and LDL cholesterol levels.18 FH is a complex disorder and its introduction into the cholesterol-atherosclerosis debate in the context of the general population is highly debatable.

DISCUSSION

These questions provide pieces or sets of evidence that, unless discredited or explained away in a satisfactory manner, falsify or seriously undermine the hypothesis that total or LDL cholesterol is associated with the development or progression of atherosclerosis. It does not suffice to present studies that reach a different conclusion—that simply indicates that the hypothesis rests on inconsistent evidence. Nor do studies which find that statins lower the progression of atherosclerosis help resolve the problem being discussed, since these studies suffer from the same potential confounding by pleiotropic effects discussed above. These studies also suffer from being inconsistent. For example, two primary prevention trials showing no effect on progression,19,20 suggesting the overall benefit is nil. Similar null results were obtained in a randomized placebo controlled clinical trial with intensive statin treatment where reducing systemic inflammation and halving LDL had no major effect of the rate of progression of coronary artery calcification in a group with calcific aortic stenosis and coronary artery calcification.21

Since it appears that there is considerable direct evidence that the total cholesterol or LDL serum levels are not associated with the development or progression of atherosclerosis, drugs that lower the levels and impact some measure of the extent or progression of atherosclerosis presumably are operating by a non-lipid lowering mechanism, and what ever it might be, it may not be the optimum approach. There is also the question of how progression is measured, since serious questions have been raised regarding the concordance of ultrasonic measurements on the carotid arteries and the extent of atherosclerosis in the coronary arteries.22 Using measures of atherosclerosis in the carotid arteries as a surrogate for coronary atherosclerosis is and has been popular as an endpoint in lipid lowering studies.

The inconvenient questions must be answered since they concern falsification, which is the tool of scientific progress. To ignore this situation simply delays a potentially more effective approach to the primary prevention of atherosclerosis and thus coronary heart disease, but also results in millions of individuals taking drugs for life on the basis of what may be a false hypothesis. This point is underscored by recent studies which find that significant numbers of patients are taking lipid lowering drugs based on current assessment protocols but have zero or near zero risk based on measured total or calcified plaque burden.4,8,10,12

Finally, there is the inconvenient result of a recent study which found that more than 50% of patients admitted to hospital for an adverse coronary event had low to very low levels of LDL, in spite of the central position LDL occupies in the assessment of risk. The mean level of LDL was only 104.9 mg/dL and almost half had admission levels of < 100 mg/dL.23

The probability that the cholesterol-atherosclerosis hypothesis is false appears high enough to justify a new approach to risk assessment and a focus on other risk factors which do not appear to have a history of constant challenges. These include smoking, hypertension, the metabolic syndrome and its associated insulin resistance, inflammation and dyslipidemia (high triglycerides and low HDL), a low vitamin D status, a low omega-3 index, and psychological stress.4 Furthermore, there is a need to clarify the different approaches that are appropriate for the prevention of the incidence and progression of atherosclerosis, the primary prevention of CVD events such as heart attacks or the emergency need for revascularization, and secondary prevention after these adverse events have occurred. Too often these are all bunched together under the heading of preventing heart disease. But the physiological aspects are presumably quite different. In addition, the above discussion calls into question the "truth in advertising" aspect of TV ads that imply elevated serum cholesterol is causing plaque to grow and that the problem is easily solved with a pill. Finally, there is need to give greater recognition to the differences between men and women and the young and the elderly. This is highlighted by the compelling data indicating that in general, women do not benefit at all from lipid lowering for primary prevention and that other approaches need to be emphasized.24,25 The time has come, it would seem, for those concerned with preventing coronary heart disease to stop barking up the wrong tree.

If one looks at the coronary calcification literature for guidance as to what to do to reduce the progression of atherosclerosis, which amounts to preventing symptomatic coronary heart disease, it is surprising what little guidance is forthcoming. Many studies have looked at TC, LDL, HDL, TGs, hypertension, BMI, smoking, diabetes, family history of CHD, and other less common factors. In general, either no factor turns out to consistently have significance, and if a factor is found to be predictive in one or two studies, other studies do not confirm its significance. Those that turn up most often are HDL, TGs, hypertension, diabetes and smoking, but there are as many or more studies find no association even with these factors than find a statistically significant link. That is, a very inconsistent picture emerges when the endpoint is the incidence or progression of coronary calcification and one tries to find risk factors that are consistently important. This level of inconsistency over a large number of studies suggests that researchers are looking at the wrong set of factors. Even when the factors are assembled, such as done in the Framingham risk score, the results are suggestive but far from definitive and again inconsistent. Even studies of the association between coronary calcification and C-reactive protein are inconsistent.

CONCLUSIONS

Thus there is considerable evidence available that appears to falsify the cholesterol- atherosclerosis hypothesis. Furthermore, measurement of the extent of atherosclerosis can lead to a better assessment of risk and prompt preventive action, research so far has failed to produce evidence-based guidance regarding the prevention of this disease, and especially prevention prompted by traditional CHD risk factors. One it seems is forced to go back to the broader generalizations regarding the prevention of heart disease. These involve lifestyle, weight control, the type of diet consumed, psychological stress, systemic inflammation, etc. While these more general actions impact the risk factors that are discussed above, it may well be that a more global view is necessary which does not isolate one or two risk factors, but instead attempts to optimize all of them through a broad approach to coronary artery health. This conclusion suggests the folly of the focus on LDL. Lifestyle and dietary interventions generally do not work when LDL target levels are used to judge success, and the net result is pharmaceutical intervention. In addition, there are other risk factors that are not generally included in the above-discussed studies which may be important. This leads one back to observational studies which have consistently indicated that the risk of heart disease and diabetes can be dramatically reduced by diet and lifestyle changes. But one must be careful to whom one listens in regard to diet. This will be the subject of future research reviews. In this global approach, one is concerned with eating a prudent diet, exercising regularly, bringing weight to a value corresponding to young adulthood (before the childhood obesity epidemic), making sure that the problem of belly fat is solved if it exists, and avoiding smoking or second-hand smoke. The reader is referred to the Research Report Coronary Heart Disease Risk and Its Reduction for a more detailed discussion of global approaches to prevention of heart disease and diabetes. The big mistake is to think that if this does not work after a few months, a cholesterol-lowering pill will. One should keep in mind the results of lipid lowering trials where for women with no evidence of heart disease, there is no statistically significant effect on the probability of adverse events, and for men with no evidence of heart disease, the benefit is small, the absolute risk reductions border on negligible, and the number needed to treat to prevent on adverse event is large. Furthermore, as has been discussed above, there is a large body of evidence to justify the view that LDL cholesterol does not drive atherosclerosis.

The decision to agree to take statin drugs for primary prevention should depend on weighing risks and benefits. The benefits judged by absolute risk reductions range from small to insignificant, especially in the primary prevention setting. This has been repeatedly pointed out in the literature but ignored in favour of the larger relative risk reductions. The central and very serious problem is that the risks can be viewed as unknown since the acknowledged numbers depend on official reporting which is widely described as picking up only a few percent of cases. It in fact appears that the side effect frequency and severity are substantially underappreciated. This is a serious problem when one considers that the guidelines based on the new JUPITER study discussed in the last Newsletter could lead to 80% of all U.S. men > 50 and women > 60 years of age being considered appropriate candidates for statin therapy (the heart.org, March 9, 2009).

The reader is referred to http://www.spacedoc.net which seems like one of the best sites for keeping up with the rapidly growing list of serious if not devastating statin side effects. This is the website of Dr. Duane Graveline, a retired NASA physician and astronaut with a masters in public health along with his MD. He has a free newsletter and has written two books on statin side effects and is about to come out with a third. He has personal experience with severe and progressing problems he attributes to statins and operates a clearinghouse for anecdotal reports. Dr. Graveline would appear to agree with the thesis of this review since in the newsletter describing his new book The Statin Damage Crisis he states "Cholesterol level appears irrelevant to the process of atherosclerosis."

NOTE: The quotes attributed to the Austrian-British philosopher Karl Popper (1902-1994) were obtained from Internet sources.

REFERENCES

1. Grundy SM. Promise of low-density lipoprotein-lowering therapy for primary and secondary prevention. Circulation. 2008;117:569-573.
2. Ware WR. High cholesterol and coronary heart disease in younger men: the potential role of stress induced exaggerated blood pressure response. Med Hypotheses. 2008;70:543-547.
3. Mays ME, Dujovne CA. Pleiotropic effects: should statins be considered an essential component in the treatment of dyslipidemia? Curr Atheroscler Rep. 2008;10:45-52.
4. Ware WR. Psychological stress, insulin resistance, inflammation and the assessment of heart disease risk. Time for a paradigm shift? Med Hypotheses. 2008;71:45-52.
5. Hecht HS, Superko HR, Smith LK, McColgan BP. Relation of coronary artery calcium identified by electron beam tomography to serum lipoprotein levels and implications for treatment. The American Journal of Cardiology. 2001;87:406-412.
6. Hecht HS, Superko HR. Electron beam tomography and national cholesterol education program guidelines in asymptomatic women. Journal of the American College of Cardiology. 2001;37:1506- 1511.
7. Elkeles RS, Godsland IF, Rubens MB, Feher MD, Nugara F, Flather MD. The progress of coronary heart disease in Type 2 diabetes as measured by coronary calcium score from electron beam computed tomography (EBCT): the PREDICT study. Atherosclerosis. 2008;197:777-783.
8. Johnson KM, Dowe DA, Brink JA. Traditional clinical risk assessment tools do not accurately predict coronary atherosclerotic plaque burden: a CT angiography study. AJR Am J Roentgenol. 2009;192:235-243.
9. Venkatesh V, Ellins ML, Yang S, Natarajan M, Amlani S, Sheth T. Incremental detection of coronary artery disease by assessment of non-calcified plaque on coronary CT angiography. Clinical Radiology. 2009;64:250-255.
10. Kim HK, Chang SA, Choi EK et al. Association between plasma lipids, and apolipoproteins and coronary artery disease: a cross-sectional study in a low-risk Korean population. Int J Cardiol. 2005;101:435-440.
11. Horimoto M, Hasegawa A, Ozaki T, Takenaka T, Igarashi K, Inoue H. Independent predictors of the severity of angiographic coronary atherosclerosis: the lack of association between impaired glucose tolerance and stenosis severity. Atherosclerosis. 2005;182:113-119.
12. Barter PJ, Ballantyne CM, Carmena R et al. Apo B versus cholesterol in estimating cardiovascular risk and in guiding therapy: report of the thirty-person/ten-country panel. J Intern Med. 2006;259:247-258.
13. Phillips GB, Pinkernell BH, Jing TY. Are major risk factors for myocardial infarction the major predictors of degree of coronary artery disease in men? Metabolism. 2004;53:324-329.
14. Elkeles RS, Godsland IF, Feher MD et al. Coronary calcium measurement improves prediction of cardiovascular events in asymptomatic patients with type 2 diabetes: the PREDICT study. Eur Heart J. 2008;29:2244-2251.
15. Taylor AJ, Bindeman J, Le TP et al. Progression of calcified coronary atherosclerosis: relationship to coronary risk factors and carotid intima-media thickness. Atherosclerosis. 2008;197:339-345.
16. Skoumas I, Masoura C, Pitsavos C et al. Evidence that non-lipid cardiovascular risk factors are associated with high prevalence of coronary artery disease in patients with heterozygous familial hypercholesterolemia or familial combined hyperlipidemia. International Journal of Cardiology. 2007;121:178-183.
17. Jensen JM, Gerdes LU, Jensen HK, Christiansen TM, Brorholt-Petersen JU, Faergeman O. Association of coronary heart disease with age-adjusted aortocoronary calcification in patients with familial hypercholesterolaemia. J Intern Med. 2000;247:479-484.
18. Awan Z, Alrasadi K, Francis GA et al. Vascular calcifications in homozygote familial hypercholesterolemia. Arterioscler Thromb Vasc Biol. 2008;28:777-785.
19. Raggi P, Davidson M, Callister TQ et al. Aggressive versus moderate lipid-lowering therapy in hypercholesterolemic postmenopausal women: Beyond Endorsed Lipid Lowering with EBT Scanning (BELLES). Circulation. 2005;112:563-571.
20. Schmermund A, Achenbach S, Budde T et al. Effect of intensive versus standard lipid-lowering treatment with atorvastatin on the progression of calcified coronary atherosclerosis over 12 months: a multicenter, randomized, double-blind trial. Circulation. 2006;113:427-437.
21. Houslay ES, Cowell SJ, Prescott RJ et al. Progressive coronary calcification despite intensive lipid- lowering treatment: a randomised controlled trial. Heart. 2006;92:1207-1212.
22. Bots ML, Baldassarre D, Simon A et al. Carotid intima-media thickness and coronary atherosclerosis: weak or strong relations? Eur Heart J. 2007;28:398-406.
23. Sachdeva A, Cannon CP, Deedwania PC et al. Lipid levels in patients hospitalized with coronary artery disease: An analysis of 136,905 hospitalizations in Get With The Guidelines. American Heart Journal. 2009;157:111-117.
24. Walsh JM, Pignone M. Drug treatment of hyperlipidemia in women. JAMA. 2004;291:2243-2252.
25. Eisenbertg T, Wells M. Statins and Adverse Cardiovascular Events in Moderate-risk Females: A Statistical and Legal Analysis with Implication for FDA Preemption Claims. J Emperical Legal Studies. 2008;5:507-550.