Correspondence

C-Reactive Protein in the Prediction of Cardiovascular Events

N Engl J Med 2003; 348:1059-1061March 13, 2003

Article

To the Editor:

In the analysis by Ridker et al. (Nov. 14 issue),1 several methodologic issues merit attention before claims of superiority for C-reactive protein in the prediction of cardiovascular risk can be substantiated. First, the authors did not adjust for body-mass index, which, given its strong correlation with C-reactive protein, is a major potential confounder of the relation between C-reactive protein and cardiovascular outcomes. Second, because high levels of low-density lipoprotein (LDL) cholesterol were undoubtedly treated during follow-up, the relation between LDL cholesterol and cardiovascular events was attenuated; similar treatment of high levels of C-reactive protein did not occur.

Third, in examining whether C-reactive protein adds to the Framingham risk score, the authors substantially hinder the predictive ability of the Framingham coronary-risk function2 by using it to predict noncoronary as well as coronary events. Furthermore, in calculating Framingham risk scores, the authors used covariates (blood pressure, smoking, and diabetes) that were self-reported by the participants rather than measured systematically. Finally, the authors show that the addition of C-reactive protein to the Framingham risk score improves the prediction of modeled relative risk, rather than showing whether it improves the prediction of observed events. The model developed in this cohort must necessarily allow better prediction than the external Framingham model, particularly when the Framingham model is used to predict noncoronary events. It would be useful to know whether C-reactive protein adds to the Framingham risk score in predicting coronary events in a setting where its constituent risk factors are actually measured.

Donald M. Lloyd-Jones, M.D.
Daniel Levy, M.D.
National Heart, Lung, and Blood Institute Framingham Heart Study, Framingham, MA 01702

2 References

1. 1

   Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med 2002;347:1557-1565
   Full Text | Web of Science | Medline

2. 2

   Wilson PWF, D'Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation 1998;97:1837-1847
   Web of Science | Medline

To the Editor:

Ridker et al. show that increased levels of C-reactive protein predict coronary events. The authors carefully adjusted their model for several known risk factors (such as age, smoking, and hypertension) but did not adjust for physical activity. There is mounting evidence that exercise, which reduces the risk of cardiovascular events, is also associated with decreased levels of C-reactive protein.1-4 It would be important to know how much predictive power is left after adjustment for the level of physical activity. While we await more data, a good screening measure would be to ask patients about their level of physical activity.

Mayer Brezis, M.D., M.P.H.
Hadassah University Hospital, Jerusalem 91120, Israel
brezis@vms.huji.ac.il

4 References

1. 1

   Mattusch F, Dufaux B, Heine O, Mertens I, Rost R. Reduction of the plasma concentration of C-reactive protein following nine months of endurance training. Int J Sports Med 2000;21:21-24
   CrossRef | Web of Science | Medline

2. 2

   Smith JK, Dykes R, Douglas JE, Krishnaswamy G, Berk S. Long-term exercise and atherogenic activity of blood mononuclear cells in persons at risk of developing ischemic heart disease. JAMA 1999;281:1722-1727
   CrossRef | Web of Science | Medline

3. 3

   Ford ES. Does exercise reduce inflammation? Physical activity and C-reactive protein among U.S. adults. Epidemiology 2002;13:561-568
   CrossRef | Web of Science | Medline

4. 4

   LaMonte MJ, Durstine JL, Yanowitz FG, et al. Cardiorespiratory fitness and C-reactive protein among a tri-ethnic sample of women. Circulation 2002;106:403-406
   CrossRef | Web of Science | Medline

To the Editor:

Ridker et al. suggest that the C-reactive protein level is a stronger predictor of first cardiovascular events than the LDL cholesterol level. They also suggest that the predictive value of C-reactive protein is enhanced when it is used with the Framingham risk score. These data support the possibility that C-reactive protein has a role in predicting cardiovascular events in this cohort. However, it is unwise to focus on the potential value of C-reactive protein as a biologic marker without considering the limitations of this and other homogeneous cohorts used to support the hypothesis.1 The Women's Health Study is neither racially nor socioeconomically diverse. Although the study enrolled 39,876 professional women 45 and older, only 2.5 percent of the women in any treatment group were black.2 The rate of death due to cardiovascular disease among black women 45 to 54 years old is 126.3 per 100,000, as compared with 41.2 per 100,000 among white women.3 It is unclear whether the C-reactive protein level is more predictive of a first cardiovascular event than traditional cardiovascular risk factors on the basis of a study that excludes those at highest risk. Using the Framingham risk score with C-reactive protein may or may not augment the predictive value of C-reactive protein, because the Framingham risk score itself must be modified for use in blacks. Data from the National Health and Nutrition Examination Survey, a representative sample of Americans, suggest that C-reactive protein levels may be predictive among those with self-reported stroke, but the report concludes that further studies of C-reactive protein and cardiovascular disease in minority groups are needed.4

Michele K. Evans, M.D.
Alan B. Zonderman, Ph.D.
National Institutes of Health, Baltimore, MD 21224
me42v@nih.gov

Wallace R. Johnson, M.D.
University of Maryland School of Medicine, Baltimore, MD 21201

4 References

1. 1

   Ridker PM, Stampfer MJ, Rifai N. Novel risk factors for systemic atherosclerosis: a comparison of C-reactive protein, fibrinogen, homocysteine, lipoprotein(a), and standard cholesterol screening as predictors of peripheral arterial disease. JAMA 2001;285:2481-2485
   CrossRef | Web of Science | Medline

2. 2

   Rexrode KM, Lee IM, Cook NR, Hennekens CH, Buring JE. Baseline characteristics of participants in the Women's Health Study. J Womens Health Gend Based Med 2000;9:19-27
   CrossRef | Medline

3. 3

   Anderson RN. Deaths: leading causes for 2000. National vital statistics reports. Vol. 50. No. 16. Hyattsville, Md.: National Center for Health Statistics, September 16, 2002. (DHHS publication no. (PHS) 2002-1120 PRS 02-0522.)
   

4. 4

   Ford ES, Giles WH. Serum C-reactive protein and self-reported stroke: findings from the Third National Health and Nutrition Examination Survey. Arterioscler Thromb Vasc Biol 2000;20:1052-1056
   CrossRef | Web of Science | Medline

Author/Editor Response

Dr. Brezis correctly points out that physical activity is a critical component of any program aimed at the reduction of vascular risk and that available evidence suggests that exercise lowers C-reactive protein levels. According to our data, however, the effect of C-reactive protein on the risk of future vascular events is independent of activity level.

In contrast to the idea suggested by Drs. Lloyd-Jones and Levy, the use of lipid-lowering agents in the Women's Health Study was minimal during the first eight years of follow-up and therefore probably did not affect outcomes. Furthermore, as we have previously shown,1,2 statins lower both LDL cholesterol and C-reactive protein levels. Therefore, even if statins were being used, these agents would have lowered the levels of C-reactive protein as well as those of LDL cholesterol and thus altered the predictive value of both variables.

We concur with Drs. Lloyd-Jones and Levy that the C-reactive protein level correlates with body-mass index. Indeed, the C-reactive protein level also correlates moderately with levels of high-density lipoprotein cholesterol, triglycerides, blood pressure, and plasma glucose, and abnormalities in all four of these variables are components of the metabolic syndrome. In this regard, we recently showed that C-reactive protein adds prognostic information to that achieved by the simultaneous use of all these interrelated risk factors.3 Thus, C-reactive protein appears to add prognostic information at all levels of LDL cholesterol, at all levels of the Framingham risk score, and at all levels of the metabolic syndrome. Furthermore, as described in our article, analyses restricted to coronary events actually showed larger, not smaller, differences between the predictive value of C-reactive protein and that of LDL cholesterol. In those analyses, C-reactive protein values provided prognostic information at all levels of the Framingham 10-year risk (Figure 1Figure 1Relative Risk of Coronary Events at Various Levels of the Framingham Study Estimates of 10-Year Risk.). We note, however, that there is nothing unusual about our finding that C-reactive protein is an independent predictor of noncoronary vascular events as well as coronary events. Indeed, the Framingham investigators themselves have published results showing the value of C-reactive protein as an independent predictor of incident stroke,4 and we have published similar data with regard to peripheral arterial disease.5

Last, we concur with Dr. Evans and colleagues that further study of C-reactive protein in minority populations needs to be done. In the Women's Health Study, the base-line levels of C-reactive protein are higher among black women and lower among Asian women than among white women — a pattern that parallels almost exactly the rates of vascular disease in these three groups.

Paul M. Ridker, M.D.
Julie E. Buring, Sc.D.
Nancy R. Cook, Sc.D.
Brigham and Women's Hospital, Boston, MA 02115
pridker@partners.org

5 References

1. 1

   Ridker PM, Rifai N, Pfeffer M, Sacks F, Braunwald E. Long-term effects of pravastatin on plasma concentration of C-reactive protein. Circulation 1999;100:230-235
   Web of Science | Medline

2. 2

   Albert MA, Danielson E, Rifai N, Ridker PM. Effect of statin therapy on C-reactive protein levels: the Pravastatin Inflammation CRP Evaluation (PRINCE): a randomized trial and cohort study. JAMA 2001;286:64-70
   CrossRef | Web of Science | Medline

3. 3

   Ridker PM, Buring JE, Cook NR, Rifai N. C-reactive protein, the metabolic syndrome, and risk of incident cardiovascular events: an 8-year follow-up of 14 719 initially healthy American women. Circulation 2003;107:391-397
   CrossRef | Web of Science | Medline

4. 4

   Rost NS, Wolf PA, Kase CS, et al. Plasma concentration of C-reactive protein and risk of ischemic stroke and transient ischemic attack: the Framingham Study. Stroke 2001;32:2575-2579
   CrossRef | Web of Science | Medline

5. 5

   Ridker PM, Stampfer MJ, Rifai N. Novel risk factors for systemic atherosclerosis: a comparison of C-reactive protein, fibrinogen, homocysteine, lipoprotein(a), and standard cholesterol screening as predictors of peripheral arterial disease. JAMA 2001;285:2481-2485
   CrossRef | Web of Science | Medline

Citing Articles (8)

Citing Articles

1. 1

   Shu Meng, Lin Zhang, Lei Zhao, Yi-ru Fang, Tetsuya Fujimoto, Shun-ichiro Hirano, Hiroshi Inoue, Kenji Uchihashi, Yasuo Nishikawa, Ya-fei Wu. (2012) Effects of C-Reactive Protein on CC Chemokine Receptor 2-Mediated Chemotaxis of Monocytes. DNA and Cell Biology 31:1, 30-35
   CrossRef

2. 2

   Lavinia Forte, Giovanni Cimmino, Francesco Loffredo, Raffaele De Palma, Gianfranco Abbate, Paolo Calabrò, Diego Ingrosso, Patrizia Galletti, Ciro Carangio, Beniamino Casillo, Raffaele Calabrò, Paolo Golino. (2011) C-reactive protein is released in the coronary circulation and causes endothelial dysfunction in patients with acute coronary syndromes. International Journal of Cardiology 152:1, 7-12
   CrossRef

3. 3

   C. S. D. Roxburgh, J. J. Platt, E. F. Leitch, J. Kinsella, P. G. Horgan, D. C. McMillan. (2011) Relationship Between Preoperative Comorbidity, Systemic Inflammatory Response, and Survival in Patients Undergoing Curative Resection for Colorectal Cancer. Annals of Surgical Oncology 18:4, 997-1005
   CrossRef

4. 4

   Shujuan Guo, Shu Meng, Bin Chen, Juan Liu, Li Gao, Yafei Wu. (2011) C-Reactive Protein Can Influence the Proliferation, Apoptosis, and Monocyte Chemotactic Protein-1 Production of Human Umbilical Vein Endothelial Cells. DNA and Cell Biology 30:3, 157-162
   CrossRef

5. 5

   C H Richards, E F Leitch, P G Horgan, J H Anderson, R F McKee, D C McMillan. (2010) The relationship between patient physiology, the systemic inflammatory response and survival in patients undergoing curative resection of colorectal cancer. British Journal of Cancer 103:9, 1356-1361
   CrossRef

6. 6

   Julia F. Slejko, Robert L. Page, Patrick W. Sullivan. (2010) Cost-effectiveness of statin therapy for vascular event prevention in adults with elevated C-reactive protein: implications of JUPITER. Current Medical Research and Opinion 26:10, 2485-2497
   CrossRef

7. 7

   J T Jenkins, G O'Neill, C G Morran. (2007) The relationship between patient physiology and cancer-specific survival following curative resection of colorectal cancer. British Journal of Cancer 96:2, 213-217
   CrossRef

8. 8

   Katherine M. Field. (2005) Effect of 3-Hydroxy-3-Methylglutaryl Coenzyme A Reductase Inhibitors on High-Sensitivity C-Reactive Protein Levels. Pharmacotherapy 25:10, 1365-1377
   CrossRef