A large study has failed to confirm a cherished nutritional belief: that the more fish you eat, the better you will be protected against heart disease.
The study, by Dr. Alberto Ascherio of the Harvard School of Public Health and his colleagues, followed 44,895 men for six years to determine whether those who ate the most fish would have the lowest incidence of heart disease.
To their surprise, they found that it did not matter whether the men ate fish once a month or six times a week -- the rate of heart disease was unaffected.
Given the results of the study, some people who do not like fish but eat it three or four times a week in the expectation that it will forestall coronary disease may feel tempted to abandon it.
But experts cautioned against dropping fish from the diet. One study does not necessarily prove that fish has no beneficial effects on the heart, they said, and in any case fish remains a wholesome food for a variety of reasons.
Dr. Ascherio said his study could not answer whether it was better to eat some fish than none. The reason is that so few of the men studied -- just 2,042 -- ate no fish at all that comparisons of their heart disease rate with that of the others were not statistically significant.
For example, although the men who did not eat fish had a 26 percent greater risk of dying of heart disease than those who did, that number could have occurred by chance, Dr. Ascherio said. In any case, he said, "the belief that eating fish helps your heart is not supported by this or other studies."
"Some things," he said, "are more complex than we believed them to be."
In an editorial accompanying Dr. Ascherio's study, which is being published today in the New England Journal of Medicine, Dr. Martijn B. Katan of Wageningen Agricultural University in the Netherlands said the results "should somewhat dampen enthusiasm for fish and fish oil as a panacea against coronary heart disease."
But he and others said the study did not firmly rule out any benefits that fish might have on the heart.
"It is premature to close the books on fish oil based on one study," said Dr. Neil J. Stone, chairman of the American Heart Association's nutrition committee and associate professor of medicine at the Northwestern University School of Medicine. But, he added, "it certainly means we have to re-examine what the benefits of fish are."
The idea that fish helps protect the heart has become entrenched in public perception even though it has never been proved. Fish oil capsules are a staple at health food stores, where they are promoted as protection from heart disease, and fish itself has taken on a magical aura.
"I've had people tell me they eat fish every day," Dr. Denke said.
The fish hypothesis began a decade ago with the observation that Eskimos in Greenland had fewer heart attacks than residents of Denmark, even though the Eskimo diet was very high in fats.
Perhaps, a number of researchers proposed, it was a special class of fats -- the omega-3 fatty acids, found in fish oils and seafoods -- that were protective. Cold-water fish such as salmon, bluefish and mackerel, which are high in the omega-3 fatty acids, were considered more healthful than low-fat fish such as flounder and orange roughy.
"This was a very hot story indeed," Dr. Stone said.
The study on Eskimos was followed by several others indicating that the death rate from heart disease was lower among men who ate fish than among those who did not.
These were population studies, Dr. Ascherio said, and they were not completely consistent.
For example, he said, "we also have populations who eat lots of fish and have high rates of heart disease." In eastern Finland, "they have the highest rate of heart disease in the world, and they eat lots of fish."
But the hypothesis gained credibility when a mechanism for the supposed effects was discovered. Taken in large quantities, fish oils inhibit the formation of blood clots, a weak variation of the effect that aspirin has on blood clots. And they lower triglycerides, fats in the blood that are linked to an increased risk of heart disease.