Low cholesterol linked to high suicide levels: Celia Hall reports on the findings of a scientific survey that involved 50,000 people over 20 years

MEN with low blood cholesterol levels are more than four times as likely to commit suicide as men with higher levels, and nearly three times as likely to die prematurely from accidents or injury, a Swedish survey claims.

Fifty thousand men and women took part in the 20-year study, making it the biggest into low cholesterol.

Results published tomorrow in the British Medical Journal give more weight to calls for a halt to the use of cholesterol-lowering drugs until more is known about the peculiar association between low cholesterol and violent death.

The surge in suicides and accidents appeared in the six years after the subjects were first screened, which suggests that lowering cholesterol levels by drugs or by diet - suddenly changing the levels - may be a factor.

Low blood cholesterol and diets low in saturated fats are recommended to reduce the risk of heart disease. Drugs to reduce the levels are given when diet does not work and when levels are very high.

The Swedish research also raises the possibility of behaviour changes and aggression in people with lower levels or who eat a low cholesterol diet.

The report mentions a study of monkeys which were given 'luxury' high or 'prudent' low cholesterol diets. The monkeys which ate diets low in saturated fats and cholesterol were more aggressive. Lower cholesterol levels have also been found in the mentally ill in tests on criminals.

Some British cholesterol research now includes questions about personality. Results of other studies which have looked at cholesterol lowering drugs have been inconclusive. Some have reported similar results to the Swedish survey; others have found no connection with suicide or death.

The Swedish survey adds a further confusion. It could find no increased risk in women.

The new information is based on an investigation of cause of death among men and women aged 45 to 74 from Varmland, in western Sweden, who took part in a general health survey and follow-up between 1964 and 1985. Dr Gunnar Lindberg, of the Centre for Public Health Research, Karlstad, says that a reason for the inconsistent findings is that in comparison other studies were very small.

Similary the researchers say that the lack of evidence concerning women may be because, normally, fewer women than men suffer violent deaths so that numbers are small. In the study period, 376 men and 139 women died from injuries. Of these, 146 men and 44 women committed suicide.

Other investigations have also shown a link between cancer and a low cholesterol level.

But Dr Lindberg's results did not change when they removed cancer patients from the statistics.

Professor Michael Oliver, director of the Wynn Institute for Metabolic Research, in London, said the question remained as to whether lowering cholesterol is harmful or not or whether naturally low cholesterol is always harmless.

'Some cholesterol is essential for healthy cells - and that will include brain cells,' he said.

'Too little is known about the effects of very low cholesterol. It may be that in a small number of people suddenly lowering it could cause a 'cell sickness' in a broad sense.'