A Half Century of Controversy over Cholesterol in Britain

Edited transcript of a lecture given at the Centenary Meeting
of the British Atherosclerosis Society in Cambridge on September 11th, 2008

Gilbert Thompson

Gil Thompson is Emeritus Professor of Clinical Lipidology at Imperial College London and was the first Chair of the BAS, from 1997 to 1999.

If one searches Google for the phrase “cholesterol controversy” more than a million entries come up, one of which is the website of THINCS - The International Network of Cholesterol Skeptics. Their motto is a quotation from Karl Popper “The growth of knowledge depends entirely on disagreement” and it provides an appropriate theme for my talk.

The beginning of the controversy
It is hard to be certain when and where the cholesterol controversy originated but a case can be made for its starting at the Postgraduate Medical School of London at Hammersmith Hospital after the Second World War. It was an uncompromising place to work in at that time, full of ambitious and competitive individuals. One of these was Paul Wood, a brilliant but abrasive cardiologist from Melbourne. He left the Hammersmith in 1949 but much of his work there was published the following year in his classic textbook Diseases of the Heart and Circulation. According to Michael Oliver it contains the first reference in the British literature to a relationship between cholesterol and vascular disease, namely that “considerations should be given to the possibility that raised cholesterol might result from coronary heart disease” (my italics).

In 1953 Michael Oliver presented a paper at the British Cardiac Society meeting in Newcastle showing for the first time in Britain that patients with coronary heart disease had significantly higher serum cholesterol levels than controls. However, not a single question was asked during the discussion period and the silence was eventually broken by Paul Wood, who commented that these findings were irrelevant to cardiology.

Michael Oliver

It wasn’t just lipids but also epidemiology that got short shrift from cardiologists in those days. In that same year, 1953, Jerry Morris showed that fatal myocardial infarcts were less common in bus conductors than drivers and in postman than sedentary civil servants, suggesting that exercise had a protective effect. Before these findings were published Morris showed them to John McMichael, Director of the Dept of Medicine at Hammersmith Hospital and suggested to him that epidemiology could play a useful role in supplementing clinical research. McMichael listened until he had finished and then retorted “Rubbish”!

However, it wasn’t all doom and gloom at the Hammersmith because Nick Myant and his MRC Lipid Metabolism Unit were busily conducting research there into familial hypercholesterolaemia. Nick’s restrained attitude to the cholesterol controversy was summed up in the preface to his book The Biology of Cholesterol and Related Steroids where he states: “...it is a fact that the very word “cholesterol” tends to arouse emotional reactions that cloud judgement...The reason...is...the link...between cholesterol and a disease so serious and widespread that it is difficult not to feel strongly about it...It seems to me that those engaged in the current controversy should try to emulate Darwin, one of whose endearing traits was a horror of being unfair to his opponents...he kept a special notebook for facts that went against his theory because he found that he tended to forget them.”

Nick Myant (centre of 2nd row) and his MRC Lipid Metabolism Unit in 1983, including Anne Soutar (in front of NM) and the author (2nd from left in 3rd row)*

Origins of the Atherosclerosis Discussion Group
Let’s leave the Hammersmith for the time being and consider the origins of the Atherosclerosis Discussion Group (ADG). It was founded in 1958 and renamed the British Atherosclerosis Society in 1998. Sir Howard Florey, the Nobel Laureate, was the first chairman, Michael Oliver was the first secretary and the topic of the inaugural meeting was “Interrelationships of dietary fat, blood lipids and atherosclerosis”. Three of the founder members, namely Michael Oliver, Elspeth Smith and Jerry Morris, feature on the cover of the transcript of the Wellcome Witness Seminar “Cholesterol, Atherosclerosis and Coronary Heart Disease in the UK, 1950-2000”, which has been an invaluable source of information.

Three of the founder members of the Atherosclerosis Discussion Group: Michael Oliver (top), Elspeth Smith (centre) and Jerry Morris (bottom)

The original co-founders of the ADG were Michael Oliver and John French, the latter succeeding the former as secretary. He worked in Florey’s pathology dept at Oxford during the 1960s until his untimely death at the age of 50. His description of the pathological features of atherosclerosis in Florey’s General Pathology is hard to better even now and the vital role he played in the early days of the ADG is commemorated by the John French Lecture each spring.

Another prominent member of the ADG in Oxford was Hugh Sinclair, nutritionist and Fellow of Magdalen College. He was especially interested in the apparent protection against coronary heart disease among Eskimos, which he attributed to their high intake of omega-3 fatty acids. He once lived on seal meat for 3 months to study the effects of omega-3 fatty acids on his own lipids and clotting factors. He used to send his friends greetings in verse, such as the one shown below illustrating his concept that atherosclerosis was caused by essential fatty acid deficiency. His influence on the ADG is commemorated by the Hugh Sinclair Memorial Lecture each autumn.

Oh, infarcted English,
Whose blood is full of saturated fats,
Cholesterol whose esters with wrong lipids
Have in these parts accumulated thus
Insoluble for lack of double bonds.

Pure and white but not deadly
Initial trials of lipid-lowering diets for the secondary prevention of coronary heart disease in Britain were all disappointingly negative. This opened the door for alternative hypotheses, one of which was that coronary heart disease was caused by excessive consumption of sugar. The man behind this idea was John Yudkin, Emeritus Professor of Nutrition at Queen Elizabeth College, who publicised his findings in books and newspaper articles, including one in The Times in 1974 titled “Why suspicion falls on sugar as a major cause of heart disease”. I persuaded that newspaper to let me rebut his arguments in an article titled “Beware sweet reason in the search for causes of heart disease” in which I put forward the evidence behind the lipid hypothesis. The BBC spotted these articles and arranged for a televised debate between John Yudkin and Don Fredrickson, who was the foremost lipidologist in the USA at the time and soon to become the head of the National Institutes of Health. The topic was “The dietary cause of heart disease is sugar not fat” which was proposed by Yudkin and opposed by Fredrickson, assisted by Peter Taggart and myself. The debate was chaired by Sir George Porter, the Nobel Laureate, and took place at the Royal Institution. We pointed out that Yudkin’s evidence was mainly circumstantial but he persisted with his beliefs, despite the conclusions of an MRC working party that the evidence for sugar being a factor in coronary heart disease was extremely slender.

Programme of televised BBC debate on role of sugar in coronary disease

The Leader of the Opposition
The leader of the opposition to the lipid hypothesis in Britain was undoubtedly Sir John McMichael. Despite being retired his reputation ensured that he remained very influential in cardiological circles. During the latter half of the 1970s he sent a stream of letters to The Lancet criticising official efforts to prevent coronary heart disease by dietary means. One consequence of this was another debate, this time organised by the European Society for Clinical Investigation. It took place in Rotterdam in 1977, the motion being “That modification of serum lipids by dietary and/or other means will influence the incidence of, or mortality from, coronary heart disease”. The motion was proposed by myself, Lars Carlson and Shlomo Eisenberg and opposed by Sir John McMichael, Paul Astrup and Christian Crone. My colleagues and I put our case in an orthodox manner but our arguments were demolished by the debating skills of Crone who showed 2 slides, one of a sleek, glossy coated rat which he said had been fed butter from birth, the other of a mangy rodent with moulting fur which he said had been raised on corn oil. The audience dissolved into laughter and that was the end of the debate!

Sir John McMichael*

The following year, Oliver et al reported the results of the WHO trial of clofibrate, which showed that the drug reduced non-fatal myocardial infarcts but increased total mortality. McMichael seized upon these results and administered what he regarded as the coup de grāce to the lipid hypothesis in the form of an article titled “Fats and atheroma: an inquest”. These were testing times for lipidologists in Britain but they were thrown a life-line in 1984 with publication of the results of the Lipid Research Clinics Coronary Primary Prevention Trial, which showed that cholestyramine reduced myocardial infarcts without affecting total mortality.

The Abominable No Man
This is the title Tony Mitchell claimed was bestowed upon him by those he called Cholesterol Evangelists. The inspiration behind the success of the new medical school at Nottingham, he was also a past chairman of the ADG and an implacable critic of the lipid hypothesis. He was especially critical of proposals to reduce dietary saturated fats and the title of one of his lectures was “Should every cow carry a government health warning?” His articles had similarly pejorative titles and in one he listed the “cosy beliefs” which he considered to be untrue or irrelevant such as “atherosclerotic plaques are cholesterol deposits in artery walls”. With hindsight it is apparent that he was wrong in every instance. He liked to quote H.L. Mencken’s aphorism “The mostly costly of all follies is to believe passionately in the palpably-not-true”. Or is it even more costly to disbelieve passionately, as Mitchell did, what eventually proves to be true?

Tony Mitchell*

The discovery and development of statins
The discovery of statins was the result of the single-minded pursuit of an idea by the Japanese microbiologist, Akira Endo. Between 1971- 1973 he painstakingly screened 6000 moulds for their ability to inhibit cholesterol synthesis, before finally discovering compactin. This was the first HMG CoA reductase inhibitor to be shown to effectively lower serum cholesterol in man although it was never licensed. Endo and workers at Merck then discovered mevinolin simultaneously but from different moulds. This compound was renamed lovastatin and approved by the FDA in 1987, an event greeted on the cover of Readers’ Digest with the headline “Wonder Drug that Zaps Cholesterol”!

Many consider that Endo’s discovery of compactin is on a par with Fleming’s discovery of penicillin and believe he should be added to the select band of those who have received a Nobel Prize for their research on cholesterol, namely Bloch and Lynen (1964), Cornforth (1975) and Goldstein and Brown (1985). (Two days after this lecture was delivered it was announced that Akira Endo had won the De Bakey-Lasker Clinical Research Award for 2008, the US equivalent of the Nobel Prize for Medicine).

Regression Studies
The availability of statins in the 1980s provided the impetus for angiographic studies of the effects of LDL-lowering on the progression and regression of coronary atherosclerosis. In 1992 Alan Howard and I organised a joint symposium of the Atherosclerosis Discussion Group and European Atherosclerosis Society in Cambridge on “Regression of Atherosclerosis”. Participants included Michael Oliver, Neville Woolf, the late Michael Davies, Greg Brown and the late David Blankenhorn. One of the main conclusions of this meeting was that lesion composition was more important than lesion size in determining clinical outcome. This meeting illustrated the value of the ADG’s multi-disciplinary approach to research into atherosclerosis, with fruitful interactions between cardiologists, pathologists and lipidologists.

Joint Symposium of the Atherosclerosis Discussion Group and European Atherosclerosis Society

Participants of symposium on Regression of Atherosclerosis included Michael Oliver (extreme left), Michael Davies (2nd from left in 2nd row), Greg Brown (3rd from left in 2nd row) and David Blankenhorn (2nd from right in 2nd row)

Low cholesterol, murder and suicide
This brings us to consider the possible risks of lowering cholesterol for therapeutic purposes. Concerns over safety had been aroused in 1990 when Muldoon et al published a meta-analysis showing an increased risk of a violent death, including from murder and suicide, in patients on lipid-lowering diets or drugs. Michael Oliver had been concerned with safety ever since the clofibrate trial and his anxieties resurfaced in articles with titles like “Might treatment of hypercholesterolaemia increase non-cardiac mortality?” Two years later Davey Smith and Pekkanen published another meta-analysis showing increased non-cardiac mortality in patients on lipid-lowering drugs. Their paper was titled “Should there be a moratorium on the use of cholesterol lowering drugs?” and it resulted in a hyperbolical reaction in the newspapers, with headlines such as “Heart Pills May Kill You”. A much larger meta-analysis by Law et al subsequently showed that the association between low cholesterol and violent death was due to confounding, which helped allay anxiety among patients and doctors alike.

Newspaper headlines after publication of article by Davey Smith and Pekkanen*

The proving of the lipid hypothesis
Clear and unequivocal proof of the lipid hypothesis came in 1994 when the Scandinavian Simvastatin Survival Study (4S) conclusively demonstrated that lowering LDL cholesterol significantly reduced both total and coronary mortality without increasing non-cardiac mortality. A recent meta-analysis by the Cholesterol Treatment Trialists’ Collaboration of 14 statin trials between 1994-2004 showed a 20% decrease in coronary mortality for every 1 mmol/L decrease in LDL cholesterol.

Between 1981-2000 the death rate from cardiovascular disease in people below the age of 75 has fallen by 50% in England. It has been estimated that 10.5% of the decreased mortality was attributable to a fall in cholesterol, the remainder to decreases in other risk factors and in treatments such as angioplasty. Similar calculations suggest that in the USA almost 33% of the decrease in cardiovascular mortality was attributable to a fall in cholesterol, which presumably reflects the greater use of statins there during that period. In the light of all the evidence I think we can now give an affirmative response to Pollak’s assertion made in 1953 that “Some day, the question as to the value of prophylactic or therapeutic reduction of blood cholesterol will be answered”.

However, although the lipid hypothesis may have been proved several controversial issues remain concerning cholesterol-lowering. These include the validity or otherwise of CETP inhibition; the efficacy and safety of ezetimibe; the possible atherogenicity of dietary plant sterols; and the question of how much LDL cholesterol should be lowered to achieve maximal benefit.

Conclusions
I’d like to end by quoting the inimitable Donald Rumsfeld who said in 2002:

“That there are known knowns – things we know that we know”.
I consider that the lipid hypothesis comes under this heading.

“There are known unknowns – things we know we don’t know”.
Controversial issues still needing an answer fall into this category.

“But there are also unknown unknowns – things we do not know we don’t know.”
This last category presumably refers to questions of the future, which are impossible to predict. Who would have thought 50 years ago, when this Society was founded, that today it would be discussing nuclear receptors and atherosclerosis? As for 50 years hence, that’s an unknown unknown to which I shall never know the answer!

(Image reproduced with permission from Thompson G., The Cholesterol Controversy*, 2008, published by RSM Press.)