New C-reactive protein (CRP) study
Source: New England Journal of Medicine 2002; 347:1557-1565, 1615-1616  (November 14, 2002)

Measuring levels of a protein linked to inflammation may identify people who have a high risk of heart disease but whose risk might otherwise go undetected, according to study findings published Thursday. In fact, in the study of nearly 28,000 women, levels of the protein, known as C-reactive protein, or CRP, were more accurate than levels of "bad" cholesterol at predicting the risk of heart attack and other cardiovascular problems.

There is not yet a consensus on whether doctors should routinely measure C-reactive protein levels, but the study's lead author, Dr. Paul M. Ridker, of Harvard Medical School in Boston, said that such testing may be useful.

"More than half of all heart attacks and strokes occur among healthy men and women who have normal or even low cholesterol levels," Ridker told Reuters Health. "A simple, inexpensive blood test for CRP can detect high-risk individuals, even when cholesterol levels are low," the results of the study show.

"In fact, those with high CRP but low cholesterol are actually at substantial risk and need to better listen to their doctors about diet, exercise and smoking cessation, all of which can reduce their risk of heart disease," Ridker said.

Several studies have associated C-reactive protein, which is a marker for inflammation, with an increased risk of several conditions, including heart attack, stroke, sudden cardiac death and artery disease. Even though high levels of the protein seem to indicate a higher risk of cardiovascular disease, the appropriate level of the protein is uncertain. In other words, doctors have not known how much C-reactive protein is too much, or whether reducing it can actually cut heart risk. In contrast, the cut-off points for high levels of LDL, or "bad," cholesterol are well determined.

Ridker and his colleagues compared the accuracy of LDL and C-reactive protein levels in predicting the risk of heart attack, stroke, death from cardiovascular disease or treatment to restore blood flow to the heart in a group of nearly 28,000 women who were enrolled in a large, ongoing health study. At the start of the study, LDL cholesterol and C-reactive protein levels were measured in blood samples taken from the women, whose average age was about 55. The women were then followed for an average of 8 years.

After researchers accounted for other risk factors, including age, smoking diabetes, high blood pressure and use of hormone replacement therapy, they found that women with the highest levels of the inflammatory marker were 2.3 times more likely to have a heart attack, stroke, or other cardiovascular event than women with the lowest levels of the protein. Levels of the protein were even more accurate at predicting cardiovascular risk than LDL levels, according to the report.

But the results do not mean that C-reactive protein should replace LDL for measuring the risk of heart disease. Ridker and his colleagues found that a combination of the two measurements was the best method of predicting heart risk.

"We believe these data have implications for the detection and prevention of cardiovascular disease," the authors state. They point out that in the study "large proportions" of first heart attacks and other cardiovascular events occurred in women whose cholesterol levels were below the recommended cut-off for treatment. The authors note that there is some evidence that cholesterol-lowering drugs called statins may prevent heart attacks in people who have high levels of CRP but low levels of LDL cholesterol.

In his comments to Reuters Health, Ridker noted that many doctors are beginning to measure C-reactive protein as well as cholesterol. According to the Harvard researcher, C-reactive protein levels below 1 milligram per liter (mg/L) of blood are considered low, 1 mg/L to 3 mg/L reflect moderate levels, and levels greater than 3 mg/L are considered high. Since the study found that a woman's use of hormone replacement therapy did not affect the predictive powers of C-reactive protein levels, a single cut-off can be used in all women, according to the report.

The results "add to the growing body of evidence that C-reactive protein is an independent predictor of cardiovascular disease," Dr. Lori Mosca, of Columbia University in New York, writes in an editorial that accompanies the study.

Until the relationship between C-reactive protein and cardiovascular risk is confirmed in future studies, however, "it may be premature to adopt widespread assessment of C-reactive protein," according to Mosca. Besides making sure that more people undergo screening for known risk factors, Mosca concludes that "scientists and policymakers should develop a systematic approach to testing and adopting screening guidelines for emerging risk factors."

Posted on the Healing Heart Foundation website ( November 15, 2002