The Cholesterol Myth Part 2: The dangers of low blood cholesterol

by Barry Groves, PhD

"For every problem there is a solution, neat, plausible and wrong." H L Mencken

Till now, advertisers and news media have concentrated on the supposed danger from high levels of blood cholesterol. The dangers of low blood cholesterol levels have largely been ignored.

Countries with diets high in saturated fats also tend to have high levels of colon cancer. However, in 1974 surprising results came from a review of the Framingham data and those from Keys' 'Seven Countries' study (see part one of this article). The review was expected to show that the cancer could also be blamed on high blood cholesterol levels. The baffled researchers found the opposite: those with the cancer tended to have blood cholesterol levels that were lower than average.

Reports of more than twenty studies into the relation between blood cholesterol and cancer have been published since 1972. Most have reported an association between low blood cholesterol and cancer.

The authors of the Renfrew and Paisley Study conclude: "it may be a mistake to assume that dietary advice given to the general population to reduce the intake of saturated fat will necessarily reduce overall mortality."

In a study from the US published in 1990, changes in blood cholesterol over time were studied in patients with colon cancer. The doctors found that people diagnosed with colon cancer had an average thirteen percent decline in blood cholesterol levels in the ten years prior to diagnosis, compared with an average increase of two percent in the non-cancer control group. Both those with the cancer and those free from it had similar blood cholesterol levels initially.

It seemed possible that the decline in blood cholesterol levels was a result of the cancer, not the cause of it, but this was ruled out by the investigators. Comparing cases in which there were higher cholesterol readings prior to diagnosis with cases in which prior readings were lower, they concluded that it was a long term lowering of blood cholesterol levels that gave rise to the cancers.

Interestingly, the average blood cholesterol level of those who developed the cancers declined to an average 5.56 millimoles per liter (mmol/l), which in typical American readings would be 217 milligrams per deciliter (mg/dl). Yet the British government's Health of the Nation strategy still aims to reduce everyone's levels to below 5.2 mmol/l, which is almost the same as the US government's optimal level of less than 200 mg/dl for total cholesterol.

Low cholesterol means more strokes

Published at about the same time was a very large study in Japan, covering two decades, which concluded that low levels of blood cholesterol also increase the incidence of stroke.

Over the past few decades, Japan has experienced a rapid change in its living and eating patterns. The Japanese are eating more total fat, saturated fatty acids and cholesterol, animal fats and protein, and less rice and vegetables. This has provided a unique opportunity for a large-scale, natural experiment into the effects of those changes.

Investigators have shown that this change to Western and urban eating patterns, departing as it does from centuries old traditions, has been accompanied by a general lowering of blood pressure and a large decline in the incidence of stroke deaths and cerebral hemorrhage between the 1960s and the 1980s.

They attribute this decline to an increase in blood cholesterol levels over the period. Supporting their findings were the results of a follow-up of 350,000 men screened for the MRFIT in the United States that showed that the risk of death from cerebral hemorrhage in middle-aged men was six times greater if they had low blood cholesterol levels.

On Christmas Eve, 1997, yet one more study's results were headlined in the press. The Framingham researchers said that "Serum cholesterol level is not related to incidence of stroke" and showed that for every three percent more energy from fat eaten, strokes would be cut by fifteen percent. They conclude:

"Intakes of fat and type of fat were not related to the incidence of the combined outcome of all cardiovascular diseases or to total or cardiovascular mortality."

So after 49 years of research, they are still saying that there is no relation between a fatty diet and heart disease. The evidence now is clear and unequivocal: animal fats-and saturated fats in general-are not harmful.

Two more studies, which considered total blood cholesterol levels and mortality in the elderly, were published in the Lancet almost simultaneously in 1997.

In the first study, scientists working at the Leiden University Medical Center found that "each 1 mmol/l increase in total cholesterol corresponded to a 15% decrease in mortality."

Similarly, doctors at Reykjavik Hospital and Heart Preventive Clinic in Iceland noted that the major epidemiological studies had not included the elderly. They too studied total mortality and blood cholesterol in the over 80s to show that men with blood cholesterol levels over 6.5 had less than half the mortality of those whose cholesterol level was around the 5.2 we are told is "healthy".

Approximately half of the brain is made up of fats. Dr. F. M.Corrigan and colleagues, writing in 1991 about the relief of Alzheimer's Disease, ask that "strategies for increasing the delivery of cholesterol to the brain should be identified." In the fight against Alzheimer's disease, they recommend increasing fat intake.

Low cholesterol and child mortality

In 1991 the US National Cholesterol Education Program recommended that children over two years old should adopt a low-fat, low-cholesterol diet to prevent Coronary Heart Disease (CHD) in later life. A table showing a good correlation between fat and cholesterol intakes and blood cholesterol in seven to nine-year-old boys from six countries supported this advice.

What it did not show, however, was the strong correlation between blood cholesterol and childhood deaths in those countries. These are in Table IV. As is clearly demonstrated, the death rate rises dramatically as blood cholesterol levels fall. So for children too, there is evidence that low blood cholesterol is unhealthy.

Aggressive behavior and suicide

Since 1992, several observers have noted increases in suicides among those undertaking cholesterol-lowering dietary regimes. Decreases in blood cholesterol cause decreases in serotonin receptors leading to increased microviscosity and affecting the balance of cerebral lipid metabolism which could have profound effects on brain function.

In institutions, aggressive people and those with antisocial personality have been found to have lower blood cholesterol levels than normal: Typically 5.04mmol/l vs 6.02mmol/l. Mental patients with high blood cholesterol (7.55mmol/l) were less regressed and withdrawn than those with lower (4.80mmol/l).

Dr Matthew G Dunnigan of Stobhill General Hospital, Glasgow, concludes that: "Without definite data on all-cause mortality and with current unresolved concerns about excess deaths from non-cardiac causes in RCTs, decisions to embark on lifelong lipid lowering drug treatment in most patients with primary hypercholesterolemia depend on the doctor's interpretation of available evidence. As in other situations in which certainty is illusory, this varies from evangelical enthusiasm for lowering lipid concentrations to therapeutic nihilism."

Cholesterol-Lowering Drugs

Although it became clear that a change in diet had little effect on CHD, that did not end the scientists' efforts to demonstrate that CHD could be prevented. If diet couldn't do it, then intervention with drugs would provide the evidence. And since drugs could be controlled much more strictly, and used in conjunction with placebos, the findings would be more demonstrable. But the drugs used to reduce blood cholesterol have all proved to be something of a disaster.

Launched on the public in 1961, Triparanol causes the levels of blood cholesterol to fall by inhibiting the liver's ability to make cholesterol. Two years later it was withdrawn because of serious side effects. Luckily for triparanol's manufacturers, a public scandal was avoided as the media's attentions were focussed on another drug marketed at the same time and by the same company-thalidomide.

More recently, a number of other drugs have been the subject of extensive and expensive trials. First was Cholestyramine (Questran) which reduces cholesterol by interfering with digestion. The gall bladder manufactures bile acid from cholesterol, and the bile acid is used in the intestine to digest fats. But when the drug is present in the gut, it binds with the bile acid, removing it from its normal function. Because the drug is indigestible, it, together with the bile acid, is excreted and the gall bladder has to make more by drawing cholesterol from the bloodstream.

As the trial would be very expensive, the scientists examined 480,000 men over a period of three years to find suitable subjects. They had to be men in the coronary age group and with extremely high blood cholesterol levels. As such men are in the most vulnerable group, their chances of success were greatly increased.

The investigators confidently announced in advance that blood cholesterol levels would be lowered by an average of 28% and, after seven years, coronary heart disease would be reduced by 50% in the treatment group.

At the end of the trial, however, cholesterol levels had fallen by less than a quarter of that called for at the start and heart disease rates were hardly affected. The $142 million trial was a total flop. Even if it had proved a success, however, those participating were so unrepresentative of the population that the question of its efficacy for the typical adult would still have remained. Another flaw that became apparent was an increase in the incidence of oral-gastro-intestinal cancers which could not be dismissed as a random chance. In the Lipid Research Clinics trial there were 21 cases and 8 deaths from gastrointestinal cancer in those taking the drug, compared to 11 cases and just 1 death in the control group.

Other organizations tested other drugs. The World Health Organization sponsored its own trial with Clofibrate (Atromid). This too was targeted against cholesterol and was confidently expected to lower blood cholesterol levels by 30%.

As with cholstyramine, the levels were lowered by much less than the expected amount, and at the end of the trial it became clear that there had been many more deaths in the group taking clofibrate than in the control group-notably from gallstones, and cancer of the liver and digestive system. In the WHO clofibrate trial, as Table V demonstrates, the drug killed more than it saved.

Among other drugs to be tested were:

1. The female hormone estrogen on the theory that if premenopausal women did not get heart disease, perhaps estrogen would protect men. But the hormone caused heart attacks rather than preventing them.
2. The hormone Dextrothyroxine, which lowers cholesterol levels, abandoned quickly when an increase in mortality was noticed in the treatment group.
3. The vitamin Niacin, which looked promising, but although there appeared to be a reduction in non-fatal heart attacks, there were marked side effects: skin disorders such as darkening, itches and rashes, as well as digestive problems and gout.
4. Gemfibrozil (Lopid) was tested and again an increase in deaths was noticed in the treatment group although this time the numbers did not reach statistical significance.
5. Compactin, which worked in a similar way to triparanol was withdrawn hurriedly and in some secrecy. The reason this time appears to be connected with cancer in dogs.
6. Lastly, despite the previous experiences with triparanol and compactin, yet another inhibitor, Lovastatin, has been approved for lifetime use on the general public after tests of very short duration only. (Derivatives pravastatin and simvastatin are marketed as Lipostat and Zocor.)

A study of all trials into cholesterol lowering by drugs up to 1987 showed an increase in mortality of 13.6% in those treated with drugs.

In 1993 a meta-analysis of all randomized controlled trials of cholesterol-lowering treatments showed that only those with very high risk showed any evidence of benefit. In all others mortality was increased.

Its authors conclude that: "Currently evaluated cholesterol-lowering drugs seem to produce mortality benefits in only a small proportion of patients at very high risk of death from coronary heart disease . . . a cautious approach to the use of cholesterol lowering drugs should be advocated".

Despite this, nearly eight times as many prescriptions for cholesterol-lowering drugs were being issued just six years later!

Has Anyone Gained?

Trials of cholesterol lowering have shown impressive results in the reduction of non-fatal heart attacks and a consequent improvement in the quality of life. In the case of drugs, the reduction of heart attacks has been in the order of twenty-three percent. Many see this as proof that lowering cholesterol in the total population, by whatever means, is worth fighting for.

But those trials were conducted on men rather than women. They were also conducted on those who had hypercholesterolemia or, at least, very high blood cholesterol levels-not people with average levels. They totally overlook the now well-established, non-linear relation between blood cholesterol and heart disease that indicates that lowering blood cholesterol in the general population is not worthwhile. The widespread agreement that the mainstay of the campaign should be a change in diet and lifestyle for all also overlooks the complete lack of evidence that such a course would have any significant beneficial effect. It even overlooks the fact that the trials involving cholesterol lowering by dietary means did not show any significant reductions in blood cholesterol.

In 1992 a report of 19 major studies published over the past 20 years suggested that public policy for reducing blood cholesterol should be reviewed. The graph below plots the relative mortality risk from all causes associated with levels of blood cholesterol in men and women. In the case of women, you can see clearly that risk rises as blood cholesterol falls. The report's author, Dr. Hulley, states:

"We are coming to realize that the resulting cardiovascular research, which represents the great majority of the effort so far, may not apply to women."

With men, the situation is more complicated as the curve is U-shaped. However, it is still noticeable that the risk with low cholesterol is similar to the risk with high cholesterol. Dr. Hulley concludes that "the findings call into question policies built over several decades on evidence that focussed only on CHD as an outcome... it may be time to review national policies aimed at shifting the entire population distribution of blood cholesterol to the left."

Another analysis based on a number of American studies estimated that on a lifelong program of cholesterol reduction by diet, the gain in life expectancy for those at very high risk (that is the 1 in 500 with hypercholesterolemia) would be between 18 days and 12 months, and for those at low risk (that is the other 499) between 3 days and 3 months. That is not very much with which to tempt people to endure a lifetime of unpalatable diets. And these figures assumed that cholesterol lowering was both effective and safe: they didn't take into account the increased risk of other debilitating and fatal diseases. Once these are added to the equation, it becomes quite evident that the current campaign is certain to do more harm than good. A study of Maori in New Zealand showed that those with the lowest levels of blood cholesterol had the highest mortality. Similar findings were also borne out by the Framingham study.

What we have then is a number of very large-scale, long-term human intervention studies showing that lowering blood cholesterol is possible but that it has no beneficial effect on coronary heart disease in the general population, and other studies showing that a low blood cholesterol level, or the methods used to attain it, are increasing the incidence of other serious killer diseases.

Forty years after the Framingham Heart Study began, its researchers looked at total mortality and cholesterol. The evidence was that for those with low cholesterol levels, deaths from non-cardiac causes offset any reduced incidence of heart disease. There was "no increased overall mortality with either high or low serum cholesterol levels" among men over 47 years of age. There was no relationship with women older than 47 or younger than 40. The researchers also concluded that people whose cholesterol levels are falling may be at increased risk.

And ten years later the Framingham researchers said: "Intakes of fat and type of fat were not related to the incidence of the combined outcome of all cardiovascular diseases or to total or cardiovascular mortality." Thus we now have 50 years of studies all demonstrating that saturated fat-including animal fat-is harmless.

Next issue: part 3: The Dangers of a "Healthy" Diet

Barry Groves is an international author based in England. He has devoted himself to dietary research since his retirement from the Royal Air Force in 1982. Having a doctorate in nutritional science, he has written a number of popular and technical books which have been published in countries as far apart as Argentina and Russia. He is also a champion archer. Groves has two websites related to health and dieting: www.second-opinions.co.uk, and www.theperfectweight.com .