CRP Heart Test May Lead to Overuse of Statins Like Crestor

At the center of the JUPITER trial results released this week is a modest elevation in the hot molecule C-reactive protein—a nonspecific marker for low-grade inflammation. Increases in CRP are now being widely hailed as justification for starting people who have no evidence of cardiovascular disease, and may not even have other risk factors for stroke or heart attack, on intensive therapy with the statin Crestor.

In response to my comments on the study yesterday, a dentist, David Jones of Texas, posted a comment noting that gingivitis causes elevated CRP. He wondered whether JUPITER patients had been screened for that common gum disease. The JUPITER report does not mention any evaluation of oral health. But Dr. Jones's concern is absolutely correct: CRP is a nonspecific finding and can be bumped up by gingivitis. Before committing patients to a lifetime of statins, doctors had better know they are treating the right condition.

It's not just a smoldering artery or a smoldering mouth that drives up inflammatory chemicals in the body. Cancer, too, has inflammatory properties that can elevate CRP. In fact, in the JUPITER trial, there were 612 new cancers and just 196 heart attacks and strokes combined. It's very likely that the participants who were diagnosed with cancer during the study already had malignant cells lurking in their bodies when they joined the trial. (Researchers excluded people from the study if they'd had cancer recently, but the study period was so short—less than two years for most patients—that a growing but yet-to-be-detected cancer would be as likely to have caused the elevated CRP as any artery in the heart or brain.)

Another issue that has gotten almost no attention is that some entirely healthy people are genetically programmed to run CRPs that are higher than others. And in a little-noted study that involved 50,000 patients and was published just two weeks ago in the New England Journal of Medicine, investigators in Denmark showed that even a lifetime of genetically elevated CRP was not predictive of heart disease. This is in contrast to LDL cholesterol, where hereditary elevations invariably put arteries in jeopardy. Though some patients with elevated CRP have cardiovascular disease risk, that does not mean all do.

So, if a doctor concludes from the new CRP blood test that seemingly healthy patients with normal cholesterol levels warrant a lifelong prescription for Crestor, many will inevitably be getting drugs they don't need. And the danger of a drug is magnified if it brings no benefit. As we teach our medical students early on, treat the whole patient, not the blood test.