The work pinpoints a mutation on a gene responsible for the production of a type of cholesterol known as lipoprotein(a) or Lp(a).
The lead author, Dr. George Thanassoulis of McGill University in Montreal, says the mutation is found in about 13 per cent of people of European descent.
It occurs to a lesser degree in people of African-American and Hispanic ethnicity, but is barely seen in people of Chinese-American ancestry.
Thanassoulis says researchers were able to show that the mutation was the number one genetic risk factor for the disease, and that it was the circulating Lp(a) in the blood that was causing the valve disease in the people studied.
About a million people in North America suffer from aortic stenosis, which is the hardening of the valve through which blood leaves the heart.
People with the condition develop shortness of breath and angina, and can go on to have heart attacks.
There is currently no remedy, except surgery to replace the valve. But by the time the condition is spotted, patients are typically elderly and may be in too weak a state to undergo the operation.
Statin drugs used to lower bad cholesterol don't work against Lp(a), says Thanassoulis. And modifying one's diet or increasing one's level of exercise also don't help prevent development of aortic stenosis.
"We prevent heart attacks, we prevent other things in cardiovascular medicine but we don't prevent valve disease," he says.
Thanassoulis says there is reason to believe niacin, one of the B vitamins, might be useful in preventing development of the condition. Niacin is known to lower levels of Lp(a).
Thanassoulis says he and colleagues at McGill hope to conduct a clinical trial to see if giving niacin to people with this gene mutation could forestall development of aortic stenosis.
The condition affects between one to five per cent of people aged 65 and older, and the incidence rises markedly as people age beyond that point.
Thanassoulis and a large group of colleagues set out to see if there is a genetic basis to the condition, analysing the full genomes of roughly 7,000 people from a number of countries.
"If we could understand the biology behind this, perhaps we can start to think about treatments and what we need to modify to prevent it. That was our major thrust," he says.
The resulting study was published Wednesday in the New England Journal of Medicine.
Lp(a) cholesterol has been known about for decades, he says, but its function hasn't been known. It's not even measured in blood tests done to assess an individual's cholesterol levels.
"So clearly we need to rethink the importance of this cholesterol and then potentially start looking more closely at individuals and see whether we should be measuring it."