A Quandary in Good News

NY Times ^ | January 7, 2005 | GINA KOLATA

[neverdem]

Suppose there were an anti-inflammatory drug that sharply reduced the level of CRP, the protein that has proved to be as powerful an indicator of heart disease risk as high cholesterol.

A doctor might well prescribe such a drug for a patient with high levels of the protein. After all, CRP is linked to inflammation, and high levels of it are linked to heart attacks.

As it turns out, there are such drugs. But this may not be good news. The anti-inflammatory drugs that lower CRP levels are COX-2 inhibitors, the very drugs that were recently found to increase the risk of heart attacks.

For example, Vioxx, which was pulled from the market in late September because of those risks, had been found in a study just a month earlier to cut CRP levels in half, from the danger zone to a level considered excellent. Celebrex, still on the market but linked to an increased risk of heart attack, has also been found to lower CRP levels.

The seemingly contradictory properties of the COX-2 drugs are just one of the medical and ethical quandaries raised this week in the wake of two new studies finding that lowering CRP with statins, the cholesterol-lowering drug that also can drive CRP levels down, may be as beneficial as lowering cholesterol in patients with severe heart disease.

Should doctors start testing for CRP just as they now test for cholesterol? If so, then what? Should patients with high CRP levels get them down by any means possible? Should doctors encourage them to do whatever it takes?

Not necessarily, some say. As Dr. Daniel Rader, a heart disease researcher at the University of Pennsylvania, said, "We cannot assume that just because something lowers CRP it's a good thing." (Dr. Rader consults for and receives honorariums from the drug companies that make statins.)

It is known that CRP itself is a risk factor for heart disease: the higher the CRP level, the greater the risk of heart attacks. This effect is independent of other risk factors, like cholesterol and blood pressure. You almost cannot get CRP too low, said Dr. Paul Ridker of Brigham and Women's Hospital in Boston, the author of one of the studies, published yesterday in The New England Journal of Medicine. Dr. Ridker found that the 10 percent to 15 percent of the population with very low CRP levels, less than 0.5, has a risk that is "essentially zero." Even if their cholesterol levels are high, he said, they seem not to have heart attacks. (Dr. Ridker is an inventor of a test for CRP and profits from its use.)

Other researchers, like Dr. Ishwarlal Jialal of the University of California, Davis, have found evidence that CRP itself might cause heart disease. The protein, which is secreted in response to inflammation, is present in plaque, he said, and can create conditions for a heart attack. (Dr. Jialal is on advisory boards for Merck, which made Vioxx, and Pfizer, which makes Celebrex.)

Scientists have learned that the standard advice for the public - lose weight, stop smoking and get any diabetes under control - can also result in lower CRP levels.

But much remains uncertain. One study found that aspirin, an anti-inflammatory drug that can prevent heart attacks, lowered CRP levels in heart patients. Another found no such effect in healthy men. That second study also found strenuous exercise slightly increased CRP levels.

Some researchers, like Dr. Steven Nissen of the Cleveland Clinic, lead author of the second article in The New England Journal, say that while more research is needed, he knows enough to make some clinical decisions for patients who are in danger of a heart attack now. High CRP levels are dangerous, Dr. Nissen said, and it is imperative for many patients to drive them down.

"I measure CRP in everyone now," he said.

He knows his own CRP level and while it is fine, he said, he has decided that if it started to creep up he would make sure he got it down. If he gained weight, he would "exercise like crazy." If that did not work, he would take a higher dose of a statin. (Dr. Nissen has consulted for, received honorariums from, and done studies supported by drug companies, including the makers of statins. He is paid a salary and donates the honorariums to a charity he established at the American College of Cardiology.)

Nearly half of heart patients have normal cholesterol levels. Dr. Nissen said that when he saw such patients and their CRP levels were above 2 milligrams per liter of blood, he prescribed a statin to lower their CRP. "That's not such a leap of faith," he said.

And if a patient has done everything possible, dieted, exercised, taken statins, and still has a high CRP?

"If I'm desperate enough, I might give a triglidazone off label," he said, referring to a class of oral diabetes drugs that lower CRP levels.

Dr. Nissen and his colleagues are starting a study comparing a triglidazone to glitizide, a diabetes drug that does not reduce CRP, in heart patients to see whether the triglidazone is more effective in slowing heart disease. And Dr. Ridker and his colleagues are studying healthy people with normal cholesterol levels and high CRP levels to see whether reducing CRP levels with a statin will prevent heart disease.

Those are the sorts of studies that need to be done before doctors and patients leap to the conclusion that anything that lowers CRP must be good, Dr. Rader and others said.

Dr. James Cleeman of the National Cholesterol Education Program at the National Heart, Lung and Blood Institute, said he would be careful about going beyond the data at hand. (He does not receive support or honorariums from any drug companies.)

"There is a reason to be cautious," Dr. Cleeman said, "and I hope it will be heard."

[neverdem]

Drugs That Reduce One Danger Have Been Shown to Create Another. That was the original title on the front page.

[EdReform]

BTTT for later

[neverdem]

FReepmail me if you want on or off my health and science ping list.

[Ernest_at_the_Beach]

Just came across this:

Genetic HIV Resistance Deciphered

[neverdem]

Thanks for the link. Happy New Year!

[lainde]

bttt