Evolving evidence. In a massive study, C-reactive protein didn


Study Refutes Protein's Role in Heart Attacks

A new study may be the last word in a controversy that's plagued cardiovascular disease research for years: whether a marker of inflammation known as C-reactive protein (CRP) drives heart attacks and strokes. In a survey of more than 128,000 people, researchers have found that genes that raise CRP levels don't make cardiovascular disease more likely. Although the study arrives at the same conclusion as earlier work, its massive size makes it statistically the most powerful test yet of this question and tough to refute, say experts.

Produced by the liver, CRP has long been eyed as a suspect in heart disease. In part, that's because of observational studies, which regularly find that higher CRP levels are associated with later heart trouble. CRP is also a vague indicator of many health problems that hike the risk of heart attacks and strokes, including obesity and type 2 diabetes. But these kinds of associations don't mean that CRP is actually causing heart attacks. Indeed, last fall, Danish researchers reported that genes that raise CRP don't appear to cause cardiovascular disease (ScienceNOW, 29 October 2008).

Now, a team of three dozen researchers from the United Kingdom, Canada, Germany, and elsewhere have teamed up to examine the question again. They drew on numerous health studies, which have banked DNA from tens of thousands of participants. Like the earlier Danish group, this one, led by epidemiologist Paul Elliott of Imperial College London, began with a simple premise: If high CRP levels cause heart attacks, then genes that raise CRP levels should also raise the risk of heart attacks. The researchers studied three variants that each raised CRP by about 20%. Then they tested whether having at least one of these variants made cardiovascular disease more likely in more than 28,000 people with disease and in 100,000 people without. The result: The genes had no effect on heart disease, the group reports tomorrow in The Journal of the American Medical Association.

"It's fairly well nailed shut" now, says James de Lemos, a cardiologist at the University of Texas Southwestern Medical Center in Dallas. "It's hard for me to imagine CRP is causal." Instead, he and others believe, the protein may be linked to other molecules that are driving disease--or it may simply indicate inflammation of the arteries that's already present, not disease that's yet to come.

Given the new data, Mark Pepys, a CRP expert at University College London, says it doesn't make sense to look for a drug that targets CRP to prevent heart attacks. Whether CRP is a useful way to gauge the risk of later disease is up in the air, he says, because it's unclear whether CRP levels add helpful information beyond the usual measures, such as family history and obesity.

This latest CRP study also raises questions about a clinical trial published last year called JUPITER. In that study, people with normal cholesterol levels received medication that lowered cholesterol levels as well as CRP, and their hearts benefited (ScienceNOW, 10 November 2008). Some argued that the trial prevented disease because CRP levels dropped, but others said that the real benefit was due to lowering cholesterol in those whose levels are normal to begin with. Increasingly, the latter point appears to be the case, says Elliott.