C-reactive protein fails to predict rheumatoid arthritis

A measurement of C-reactive protein, an indicator of acute inflammation, does not forecast who will develop rheumatoid arthritis, according to a study published last month.

Researchers from Brigham and Women's Hospital and Harvard Medical School in Boston worked with blood samples provided by 27,939 participants in the Women's Health Study, a randomized trial exploring the impact of aspirin and vitamin E on cancer and cardiovascular disease risk. Of this group, 90 received a diagnosis of rheumatoid arthritis that fit American College of Rheumatology guidelines. But after adjusting for age, body mass index, smoking and impact of the intervention this trial was studying, the disease's occurrence did not correlate with CRP measures.

"A single CRP level did not predict increased risk of RA," wrote the authors. Physicians responded that the study, published in the Dec. 11/25, 2006, Archives of Internal Medicine, highlighted this disorder's complicated nature. Specifically, many suspect that it has a variety of underlying mechanisms that translate to the same autoimmune response that destroys healthy tissue.

"It's a very individualized illness," said Adam Rindfleisch, MD, assistant professor of family medicine at the University of Wisconsin, Madison. "There are different inflammatory factors that come into play, and I think we will start to discover that there are a lot of different things that can be going on biochemically."

Many experts also questioned the choice of CRP as a potential RA risk marker. Detection of this protein is commonly used to monitor the course of the disease and determine if a treatment is working. It is not used as a diagnostic. Some specialists had expected the findings would not support it as a predictor of this disease's development, mainly because CRP can be elevated by a wide array of conditions.

"Don't confuse a test which gives you information about how a patient is doing with a test for the disease itself," said Melvin C. Britton, MD, a rheumatologist from Palo Alto, Calif., who represents the American College of Rheumatology at American Medical Association meetings. "If someone has pneumonia, you take their temperature to monitor them, but you don't take their temperature to predict whether they're going to have pneumonia."

Some also wonder whether a test to detect RA risk is really necessary, because of the fact that there is little, if anything, that can be done with the resulting information.

"If you had a way to prevent rheumatoid arthritis from occurring, then it would be useful to predict its development," said Neal Birnbaum, MD, ACR president and director of the division of rheumatology at the California Pacific Medical Center in San Francisco. "At this point in time, we don't have that."

But while many experts say a predictive test might not have much utility, they also say that, with better treatments now available, a test to differentiate those who have joint pain caused by RA from those who have other diseases is needed.

"We have terrific new treatments for rheumatoid arthritis, but a lot of damage can occur early on. What we need is a specific test to sort out who is going to have damage and who will resolve and get better by themselves," pointed out C. Kent Kwoh, MD, a rheumatologist and professor of medicine and epidemiology at the University of Pittsburgh. "It's an interesting study, but I'm not sure screening those who are asymptomatic would necessarily be that useful."

The authors suggest that a lab test that could pick out who is at increased risk would allow for the creation of targeted primary prevention programs and make it more likely that physicians would initiate treatment early to reduce the chance of disability related to this condition. In addition, CRP is increasingly being used to assess cardiovascular risk. These researchers say the fact that this biomarker has not proven useful to detect an increased chance of RA and, in previous studies of cancer, makes it more valuable for this purpose.