Slowly and hesitantly, leading heart disease researchers are concluding that there seems to be a dark side to having especially low cholesterol levels.
Although there is no dispute that low concentrations of cholesterol in the blood protect people from heart disease, there also seems to be a newly found, but sometimes grudging, agreement that very low cholesterol levels make death from other causes more likely.
Several studies pointing to a risk from low cholesterol levels have appeared this year, with more awaiting publication.
These studies, based on data from hundreds of thousands of people, show that people who have extra-low levels of cholesterol in their blood are more likely to die in later years from a variety of causes, including suicide, homicide, strokes, certain cancers, liver disease and lung disease.
In fact, the deaths from these other causes mount so quickly that
the mortality rate for those with low cholesterol levels equals the rate for people with very high cholesterol levels, who are likely to die from heart disease.
"What it comes down to is that there is an extraordinary set of observations that have emerged this year because for the first time we have large enough studies to really see them," said Dr. Stephen Hulley, a heart disease researcher at the University of California in San Francisco. It is, he added, "a very serious and disconcerting set of observations."
The new findings do not question the standard medical advice that people with high concentrations of cholesterol in their blood should go on diets or take drugs to reduce them.
They concern people who were found to have extremely low cholesterol levels, defined as 160 units or less; cholesterol is measured in terms of milligrams per deciliter of blood. The average cholesterol level in this country is about 200 units.
The findings raise a question of national policy because medical authorities want the nation as a whole to lower cholesterol levels. That policy would presumably push some individuals into the very low cholesterol range.
"This is a political as well as a scientific concern," said Dr. Michael H. Criqui, an epidemiologist at the University of California in San Diego and the editor of Circulation, a medical journal where an important study on low cholesterol levels is to appear next month. "Some people just don't want to talk about it," Dr. Criqui said. "They think it is going to impede public health measures."
But many others agreed with Dr. Criqui that the questions that have been raised deserve hard study.
The issue of cholesterol has long been clouded by debate because large-scale studies of people with high cholesterol levels have shown that lowering their cholesterol levels reduces their rate of fatal heart disease, but that, paradoxically, no overall reduction in mortality has been seen.
Despite this puzzle, doctors have decided that the clear reduction in heart disease was sufficient to warrant advising almost all Americans to lower their cholesterol.
Many researchers dismissed as a fluke the disquieting hints from several important studies that some people with low cholesterol levels had higher mortality rates.
Those people's low cholesterol levels were probably an effect of other diseases, not a cause of them, they said. But "continuing analyses have not made the association go away," said Dr. Paul Meier, a statistician at the University of Chicago. Instead, he said, "they have made people feel more ill at ease."
The new group of reports reinforces that unease. One, published in July in The Archives of Internal Medicine, involved some 350,000 middle-aged healthy men, including 6 percent with very low cholesterol levels.
According to Dr. James Neaton of the University of Minnesota and his colleagues, the men who made up that 6 percent had hardly a trace of heart disease 12 years later, and their death rate from heart attacks was about half that of men with cholesterol levels of 200 to 239. But they were more likely to die of other diseases.
They were twice as likely to have an intracranial hemorrhage. They were three times as likely to have liver cancer. They were twice as likely to die of lung disease, an association that was strongest among the men who smoked.
They were twice as likely to kill themselves. And they were five times as likely to die of alcoholism.
Similar findings come from a paper by Dr. David Jacobs, an epidemiologist at the University of Michigan, and his colleagues, who studied 170,000 men and 120,000 women in various countries and will publish their results in the September issue of Circulation.
Dr. Criqui said the Circulation paper shows that cholesterol levels of about 180 to 200 or 220 "seem to be ideal for total survival." Like previous studies, including Dr. Neaton's, he said, the new data analysis shows that the real problem occurs when cholesterol levels drop below 160.
But, Dr. Criqui said, the data cannot prove cause and effect. The pressing challenge, he said, "is to try to see if the causal arrow goes from low cholesterol levels to illness or from illness to low cholesterol levels."
Although the link between violent death and low cholesterol levels sounds problematical to some researchers, it receives a modicum of support from studies with monkeys.
Animals that were fed diets low in fats and cholesterol became more aggressive, Dr. Criqui said. But those monkey studies "are just a very small, tentative suggestion," he cautioned.