Doubt Cast on the ‘Good’ in ‘Good Cholesterol’
The name alone sounds so encouraging: HDL, the “good cholesterol.” The more of it in your blood, the lower your risk of heart disease. So bringing up HDL levels has got to be good for health.
Or so the theory went.
Now, a new study that makes use of powerful databases of genetic information has found that raising HDL levels may not make any difference to heart disease risk. People who inherit genes that give them naturally higher HDL levels throughout life have no less heart disease than those who inherit genes that give them slightly lower levels. If HDL were protective, those with genes causing higher levels should have had less heart disease.
Researchers not associated with the study, published online Wednesday in The Lancet, found the results compelling and disturbing. Companies are actively developing and testing drugs that raise HDL, although three recent studies of such treatments have failed. And patients with low HDL levels are often told to try to raise them by exercising or dieting or even by taking niacin, which raised HDL but failed to lower heart disease risk in a recent clinical trial.
“I’d say the HDL hypothesis is on the ropes right now,” said Dr. James A. de Lemos, a professor at the University of Texas Southwestern Medical Center, who was not involved in the study.
Dr. Michael Lauer, director of the division of cardiovascular sciences at the National Heart, Lung and Blood Institute, agreed.
“The current study tells us that when it comes to HDL we should seriously consider going back to the drawing board, in this case meaning back to the laboratory,” said Dr. Lauer, who also was not connected to the research. “We need to encourage basic laboratory scientists to figure out where HDL fits in the puzzle — just what exactly is it a marker for.”
But Dr. Steven Nissen, chairman of cardiovascular medicine at the Cleveland Clinic, who is helping conduct studies of HDL-raising drugs, said he remained hopeful. HDL is complex, he said, and it is possible that some types of HDL molecules might in fact protect against heart disease.
“I am an optimist,” Dr. Nissen said.
The study’s authors emphasize that they are not questioning the well-documented finding that higher HDL levels are associated with lower heart disease risk. But the relationship may not be causative. Many assumed it was because the association was so strong and consistent. Researchers also had a hypothesis to explain how HDL might work. From studies with mice and with cells grown in the laboratory, they proposed that HDL ferried cholesterol out of arteries where it did not belong.
Now it seems that instead of directly reducing heart disease risk, high HDL levels may be a sign that something else is going on that makes heart disease less likely. To investigate the relationship between HDL and cardiovascular risk, the researchers, led by Dr. Sekar Kathiresan, director of preventive cardiology at Massachusetts General Hospital and a geneticist at the Broad Institute of M.I.T. and Harvard, used a method known as Mendelian randomization. It is a study design that has recently become feasible with the advent of quick and lower-cost genetic analyses.
The idea is that people inherit any of a wide variety of genetic variations that determine how much HDL they produce. The result is that people are naturally and randomly assigned by these variations in their inherited genes to make more, or less, HDL, throughout their lives. If HDL reduces the risk of heart disease, then those who make more should be at lower risk.
For purposes of comparison, the researchers also examined inherited variations in 13 genes that determine levels of LDL, the so-called bad cholesterol. It is well known and widely accepted that lowering LDL levels by any means — diet and exercise, statin drugs — reduces risk. Clinical trials with statins established with certainty that reducing LDL levels is protective. So, the researchers asked, did people who inherited gene variations that affected their LDL levels, have correspondingly higher or lower heart disease risk?
The study found, as expected, that gene variations that raise LDL increase risk and those that lower LDL decrease risk. The gene effects often were tiny, altering LDL levels by only a few percent. But the data, involving tens of thousands of people, clearly showed effects on risk.
“That speaks to how powerful LDL is,” Dr. Kathiresan said.
But the HDL story was very different. First the investigators looked at variations in a well-known gene, endothelial lipase, that affects only HDL. About 2.6 percent of the population has a variation in that gene that raises their HDL levels by about 6 points. The investigators looked at 116,000 people, asking if they had the variant and if those who carried the HDL-raising variant had lower risk for heart disease.
“We found absolutely no association between the HDL-boosting variant and risk for heart disease,” Dr. Kathiresan said. “That was very surprising to us.”
Then they looked at a group of 14 gene variants that also affect HDL levels, asking if there was a relationship between these variants and risk for heart disease. The data included genetic data on 53,500 people. Once again, there was no association between having the variants that increased HDL and risk of heart disease.
Dr. Lauer explains what that means with an analogy.
“One might think of a highway accident that causes a massive traffic jam,” he said. “Stewing in the jam many miles away, I might be tempted to strike the sign that says ‘accident ahead,’ but that won’t do any good. The ‘accident-ahead’ sign is not the cause of the traffic jam — the accident is. Analogously, targeting HDL won’t help if it’s merely a sign.”
Dr. Kathiresan said there were many things HDL might indicate. “The number of factors that track with low HDL is a mile long,” he said. “Obesity, being sedentary, smoking, insulin resistance, having small LDL particles, having increased cholesterol in remnant particles, and having increased amounts of coagulation factors in the blood,” he said. “Our hypothesis is that much of the association may be due to these other factors.”
“I often see patients in the clinic with low HDL levels who ask how they can raise it,” Dr. Kathiresan said. “I tell them, ‘It means you are at increased risk, but I don’t know if raising it will affect your risk.’ ”
That often does not go over well, he added. The notion that HDL is protective is so entrenched that the study’s conclusions may prove hard to accept, he and other researchers said.
“When people see numbers in the abnormal range they want to do something about it,” Dr. Kathiresan said. “It is very hard to get across the concept that the safest thing might be to leave people alone.”