|
Correction
to
Ridker et al.,
|
| |||||||||||||||||||||||||||||||||||
We were unable to calculate the conventional predictive values from the data supplied in the article. It would be instructive if the authors provided the predictive values so that readers could determine whether this new test is genuinely ready for "prime-time" screening.
Even though this test performed better than measurements of conventional lipid markers such as low-density lipoprotein (LDL) cholesterol in this population (at least in terms of relative risk), there are other relevant data about LDL cholesterol that are lacking for hs-CRP. For example, we know that lowering LDL cholesterol levels has beneficial effects,3 we have effective methods to lower LDL cholesterol levels, and we have data on the cost effectiveness of such strategies.4
Gary L. Horowitz, M.D.
Bruce A. Beckwith, M.D.
Beth Israel Deaconess Medical Center
Boston, MA 02215
References
To the Editor: In our study of inflammatory and lipid markers we used a matched, nested case–control analysis that allowed direct comparison of the magnitude of risk associated with various cardiovascular risk factors after age and smoking status were taken into account. Of the 12 factors evaluated — which included LDL cholesterol, high-density lipoprotein (HDL) cholesterol, Lp(a) lipoprotein, and homocysteine — hs-CRP was the strongest predictor of future cardiovascular events. Moreover, hs-CRP levels were predictive of the risk of cardiovascular events among study participants with low levels of LDL cholesterol; these data underscore the importance of the inflammatory process in atherothrombosis.
Our matched, nested case–control study was designed to maximize biologic validity. It is not, however, conducive to calculating absolute risks. We thus concur with Horowitz and Beckwith that generalizing our results to other populations must be done with caution and that studies addressing absolute risks are needed. We further concur that the reduction of lipid levels remains a fundamentally important method to reduce cardiovascular risk. At the same time, since half of all heart attacks and strokes occur among apparently healthy men and women without overt hyperlipidemia, we believe it important for clinicians to consider emerging biologic data that go beyond the use of cholesterol screening as the sole method of assessing cardiovascular risk. With regard to hs-CRP, several large-scale studies in the United States1,2,3 and Europe4,5 have now demonstrated the potential importance of this inflammatory marker in the detection of cardiovascular risk.
Finally, we wish to correct an error in the last sentence of the Results section of our abstract. As described in the text and in Table 4 of our article, our multivariate analysis was performed on a per-quartile basis. Thus, this sentence should read, "In multivariate analyses, the only plasma markers that independently predicted risk were hs-CRP (increase in relative risk per quartile, 1.5; 95 percent confidence interval, 1.1 to 2.1) and the ratio of total cholesterol to HDL cholesterol (increase in relative risk per quartile, 1.4; 95 percent confidence interval, 1.1 to 1.9)."
Paul M. Ridker, M.D.
Julie E. Buring, Sc.D.
Nader Rifai, Ph.D.
Brigham and Women's Hospital
Boston, MA 02115
References
| |||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
