C-Reactive Protein: A New Tool for Cardiovascular Disease Risk Prediction

Recent research into the use of markers of inflammation to aid in cardiovascular risk assessment has focused on C-reactive protein, an acute-phase reactant found in atherosclerotic lesions. Serum levels of C-reactive protein are strongly associated with cardiovascular disease risk. Several clinical trials have shown that elevated C-reactive protein concentration predicts increased risk in persons with average or even below-normal levels of low-density lipoprotein cholesterol. When measured with the high-sensitivity assay, C-reactive protein has been shown to add to the predictive power of traditional cardiovascular risk factors and can enhance the 10-year Framingham Risk Score risk prediction. This article reviews the available evidence and addresses the potential for lowering C-reactive protein levels to reduce the risk of cardiovascular disease.

Benjamin J. Ansell, MD, FACC

UCLA Cholesterol, Hypertension, and Atherosclerosis Management Program

PRACTICE POINTS

• Measure CRP levels using the hs-CRP assay in patients at intermediate risk of a CV event based on global risk assessment.
• Elevated CRP levels in women increase their risk of MI and stroke by 7-fold.
• Statins have been shown to lower CRP levels, but this has not yet been correlated with reduced CVD risk.

Atherosclerotic lesions are formed by cellular and molecular responses that can best be characterized as inflammatory. These interactions are not unique to atherogenesis but rather typify processes in other chronic inflammatory diseases, such as rheumatoid arthritis, cirrhosis, or pulmonary fibrosis.¹ Arterial injury caused by cigarette smoking, elevated levels of low-density lipoprotein cholesterol (LDL-C), hypertension, or infectious microorganisms, to name a few factors, disrupts the normal functioning of the endothelium, making it more adhesive and permeable and inciting an inflammatory response.¹ Monocyte-derived macrophages are drawn to the injured area, where they multiply; cytokines and growth factors are released, and smooth-muscle cells proliferate. Successive cycles of accumulation, migration, and proliferation produce an advanced atherosclerotic lesion that may rupture and occlude the lumen, restricting the flow of blood, and thereby oxygen, to the heart.¹

A New Inflammatory Marker

The recognition that atherosclerosis is an inflammatory disease has prompted research into the potential use of inflammatory markers for cardiovascular (CV) risk assessment. Of these, the most intensively studied has been C-reactive protein (CRP), an acute-phase reactant and a marker for underlying infection, tissue injury, and systemic inflammation.^2-4 We now know that CRP also participates actively in atherogenesis through a variety of mechanisms, including stimulating the release of inflammatory cytokines by monocytes and decreasing synthesis of the important vasodilator nitric oxide.⁵

P

P

Strong evidence of a link between CRP levels and CV disease (CVD) was demonstrated in an observational study conducted by the Honolulu Heart Program,⁶ which followed 8006 men of Japanese ancestry living in Hawaii to track factors that might be associated with CVD. Measurements of CRP for 369 patients who had myocardial infarction (MI) and 1348 controls were available from frozen sera samples obtained between 1967 and 1970.⁶ Elevated levels of CRP at baseline were associated with coronary events occurring as many as 15 years later.⁶ Within each 5-year period up to 15 years, the percentage of men who had an MI was significantly associated with increasing levels of CRP (<.05). In the first 5 years, men in the top quartile of CRP levels at baseline had more than 3- fold excess risk of MI compared with men in the lowest quartile (<.01). The association was not statistically significant beyond 15 years, however.⁶

CRP Levels Predict CVD Risk: The Evidence

P

P

P

Ahost of studies have examined the role of CRP as a forecaster of CVD; the Physicians'Health Study (PHS) was one of the earliest.³ Of 1086 apparently healthy male physicians, those with baseline plasma CRP levels in the highest quartile had a 2.9 times greater risk of MI and a 1.9 times greater risk of ischemic stroke than those in the lowest quartile ( <.001 and = .02, respectively). In addition, smokers had significantly more elevated CRP levels than nonsmokers.³ And in another analysis of the PHS, median CRP levels at baseline were significantly higher among men who developed symptomatic peripheral vascular disease than among controls (= .04).⁷

The Monitoring Trends and Determinants in Cardiovascular Disease (MONICA) program included a large cohort of men aged 45 to 64 years who were selected at random from the population of Augsburg, Germany, between 1984 and 1985.⁴ Blood samples and other data were available for 936 men who were followed for 8 years. Again, participants in the highest CRP quintile had a 2.6-fold greater risk of a CV event. CRP levels were twice as high for men with a body mass index (BMI) of 30 kg/m² or more compared with a BMI of less than 25 kg/m². Although smoking and a history of type 2 diabetes were also significantly associated with elevated CRP levels, multivariate analysis confirmed that CRP was an independent predictor of future coronary events.⁴

P

P

P

In the Women's Health Study (WHS), healthy postmenopausal women who had a first CV event during 3 years of follow-up (n = 122) had significantly higher median levels of high-sensitivity (hs-) CRP at baseline than did controls (n = 244; <.001).⁸ Those in the highest hs-CRP quartile had a 7-fold increased risk of MI or stroke (relative risk, 7.3; 95% confidence interval [CI], 2.7-19.9; <.001). The power of baseline CRP levels to predict risk of vascular disease was particularly striking in women who appeared to be at low risk; nonsmoking women with elevated hs-CRP levels had a relative risk of 4.5 for subsequent CVD (= .001).

The relative risk among women with elevated hs- CRP was:

• 2.8 for those with no history of hypertension (P = .03)

P

• 4.9 for those with no evidence of diabetes (P = .001).

The prospective Women's Health Initiative (WHI) observational study assessed the relationship of CRP and the cytokine interleukin (IL)-6 to incident coronary artery disease (CAD) in an ethnically diverse population of 77,343 healthy postmenopausal women; 304 women subsequently developed CAD and were matched by age, smoking status, and ethnicity with 304 who did not.⁹ The most elevated baseline levels of either hs-CRP or IL-6 were independently associated with a doubling of the risk for CAD events compared with the lowest baseline levels.⁹ In analyses stratified by the use of hormone replacement therapy (HRT), increasing baseline hs-CRP levels retained a correlation with incident CAD risk in both HRT users as well as nonusers.

P

P

In the original Framingham Heart Study cohort, baseline CRP levels also predicted incidence of first ischemic stroke or transient ischemic attack (TIA). During 12 to 14 years of follow-up, 13.4% of 1462 men and women suffered a first ischemic stroke or TIA.¹⁰ Compared with the lowest quartile, the relative risk in the highest CRP quartile was twice as high in men (relative risk, 2.0; = .028) and almost 3 times as high in women (relative risk, 2.9; <.001). The strong linear relationship between CRP levels and risk for stroke/TIA was independent of variables such as systolic blood pressure, smoking, diabetes, or total and high-density lipoprotein cholesterol (HDL-C).¹⁰

CRP versus Traditional Risk Factors

P

The WHS compared 4 plasma inflammatory markers (including hs-CRP), 7 lipids and lipoproteins, and homocysteine for their ability to predict CVD in study participants; 122 women who had a CV event during 3 years of follow-up were matched with 244 controls. Of the 12 variables, hs-CRP was the strongest predictor of a CV event (Table), defined as CV death, nonfatal MI or stroke, or revascularization procedure. Each inflammatory marker significantly improved the predictive power of lipid screening alone (<.001).¹¹

P

A follow-up study compared the predictive powers of CRP and LDL-C in the entire WHS database of 27,939 women.¹² Levels of each variable were divided into quintiles, with the lowest quintiles used as references. The relative risk of CVD for women in the highest hs-CRP quintile was 2.3, compared with 1.5 in the highest LDL-C quintile (both <.001). The minimal correlation (r = 0.08) between hs-CRP and LDL-C observed in this study suggests that they are markers for different high-risk groups, and that screening for both would provide better prognostic information than screening for either alone.

CRP and Global Risk Assessment

The Framingham Risk Score estimates the risks of heart disease within 10 years, based on points assigned to specific risk factors. An analysis of the WHS database showed that adding elevated hs-CRP levels to Framingham risk scores increased the risk of CV events at each level of estimated 10-year risk (Figure 1).¹²

P

Framingham risk calculations were also applied to the MONICA study's Augsburg cohort, with hs-CRP added to the traditional risk factors for 3435 men. Framingham estimates of increasing 10-year CVD risk were divided into 5 categories, each stratified by low, medium, and high hs-CRP levels.¹³ The hs-CRP concentrations modified all Framingham risk categories, particularly in individuals with a 10-year risk of 15% to 19%, an intermediate-risk category in which elevated hs-CRP levels (>3 mg/L) dramatically increased CVD risk (= .02).

CRP in AHA/CDC Guidelines

In 2003, the American Heart Association (AHA) and the Centers for Disease Control and Prevention (CDC) designated CRP as the most viable inflammatory marker to measure in clinical practice, based largely on the commercial availability, standardization, and precision of the relatively new hs-CRP assay.¹⁴ The AHA/CDC statement regarding inflammatory markers does not recommend screening the entire population for hs-CRP levels. It recommends, instead, that as an independent marker of CVD risk, hs-CRP may help direct evaluation and therapy for primary prevention in patients at intermediate risk based on global risk assessment (those with a 10%- 20% risk for CVD within 10 years).¹⁴ Measuring hs-CRP may also help predict recurrent CV events.

The AHA/CDC guideline classifies hs-CRP into 3 risk categories and recommends measuring hs-CRP, either fasting or nonfasting, twice?2 weeks apart? for accuracy, and averaging the results. If the hs-CRP level is higher than 10 mg/L, the patient probably has an acute infection or another inflammatory process, and the test should be repeated.^14,15

A New Therapeutic Target for Statins?

Landmark clinical trials have shown that statin therapy is associated with, on average, a 31% reduction in major coronary events, presumably because of their cholesterol-lowering effects.¹⁶ The benefits of statins have also been attributed, in part, to anti-inflammatory mechanisms.¹⁵

P

P

In the Cholesterol and Recurrent Events (CARE) trial, pravastatin sodium (Pravachol) therapy reduced recurrent coronary events in MI survivors by 24% (95% CI, 9%-36%; = .003) compared with placebo during a 5-year period.¹⁷ Among 472 randomly selected participants who did not have a recurrent coronary event, pravastatin decreased hs-CRP levels by 17.4% (= .004), compared with a 4.2% increase with standard therapy plus placebo.¹⁸ Reductions in hs-CRP levels with statin therapy were independent of any changes in lipid levels.

P

P

Similar findings were reported in the Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS), in which lovastatin (Mevacor) therapy decreased the relative risk of a first acute CV event by 37% (95% CI, 21%-50%; <.001) compared with placebo in 6605 men and women with average total cholesterol and LDL-C levels.¹⁹ Lovastatin also reduced median hs-CRP level by 14.8% after 1 year ( <.001) and effectively decreased the risk of a first CV event in nonhyperlipidemic individuals with elevated baseline CRP levels.²⁰

Two recent trials have confirmed that aggressive statin therapy lowers hs-CRP levels in survivors of MI and acute coronary syndromes. Atorvastatin calcium (Lipitor, 80 mg) reduced hs-CRP to a median level that was 34% less than with placebo in the Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIRACL) study²¹ and 38% lower than with pravastatin, 40 mg, in the Pravastatin or Atorvastatin Evaluation and Infection Therapy?Thrombolysis in Myocardial Infarction 22 (PROVE IT-TIMI 22) trial.²²

A new study currently underway, the Justification for the Use of Statins in Primary Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER) is designed to answer the question of the clinical significance of lowering hs-CRP in persons without heart disease.²³ The trial will enroll persons who have relatively low LDL-C (<130 mg/dL) but elevated hs-CRP (>2 mg/L) levels, excluding those who have 1 or more coronary risk equivalents.²³ It is the first CV prevention trial that has defined the population at risk primarily on the basis of baseline hs-CRP levels.

The JUPITER cohort has considerable clinical relevance, since about half of all CV events occur in persons with normal or low LDL-C levels.²³ In the WHS, 46% of first CV events (ie, MI, ischemic stroke, coronary revascularization, CV death) occurred in women with an LDL-C level below 130 mg/dL.¹² A survival analysis showed that a subgroup of women with high CRP and low LDL-C levels were at greater absolute risk than those with low CRP and high LDL-C levels (Figure 2).¹² It is estimated that 1 in 4 people have the pattern of high CRP and low LDL-C levels.¹⁵ Under current cholesterol guidelines, this sizable population would not be considered candidates for treatment to lower their CVD risk.

The choice of rosuvastatin calcium (Crestor) as the study drug (20 mg/day dose) will also enable the JUPITER investigators to determine whether aggressive lowering of LDL-C levels is effective in primary prevention.²³ Since participants'baseline LDL-C in JUPITER will be approximately 115 to 120 mg/dL by design, and since rosuvastatin, 20 mg, has been associated with an approximate 55% reduction in LDL-C, the treatment-associated reduction in LDL-C is expected to be around 57 mg/dL. JUPITER, which began enrolling participants in 2003 and is intended to conclude in 2009, should provide much-needed data on the role of hs-CRP as a clinical tool for the prediction of CVD risk as well as whether lowering hs-CRP concentration with statins will also lower CVD risk.

Illustrative Case

The following case illustrates how hs-CRP measurement may be used to guide treatment decision making in the primary care setting.

At his annual physical examination, a 51-year-old Hispanic man expressed concern about his risk of developing heart disease. He is asymptomatic but says that his 75-year-old father has angina and recently underwent angioplasty and stent placement. The patient is 5 ft 9 in tall and weighs 185 lb (BMI, 27.3). He does not smoke and does not engage in physical activity. Physical examination and laboratory testing revealed: blood pressure, 140/80 mm Hg; fasting blood glucose, 98 mg/dL; total cholesterol, 233 mg/dL; HDL-C, 38 mg/dL; triglycerides, 135 mg/dL; LDL-C, 168 mg/dL; hs-CRP, 3.8 mg/L. His Framingham Risk Score was calculated at 12, for a 10-year CVD risk of 10%. When remeasured 2 weeks later, his hs-CRP level was 4.1 mg/L.

The National Cholesterol Education Program Adult Treatment Panel III guidelines set an LDL-C goal of less than 130 mg/dL for a patient who has a 10-year CVD risk of 10% to 20%. Because of the increased risk for CV events conferred by this man's dyslipidemia, he was given instructions to begin regular physical activity and reduce his intake of saturated fat and cholesterol. In addition, since his hs-CRP level was more than 3 mg/L, he was at high CV risk and was therefore prescribed low-dose statin therapy.

Conclusion

CRP, a central component of the immune response, has been shown to be a strong, independent predictor of first and recurrent CV events. It enhances and complements other markers of increased CVD risk and has the added benefit of being stable and easy to measure. Although statins have been shown to lower CRP levels substantially, that effect has not yet been prospectively correlated with a reduced CV risk. Results from an ongoing trial are expected to help resolve the question of whether lowering CRP can reduce CV risk and improve outcomes.

Disclosure Statement

Dr Ansell is a consultant for Pfizer and AstraZeneca. He serves on the Speaker's Bureau of and receives honoraria from AstraZeneca, Pfizer, and KOS.

SELF-ASSESSMENT TEST

1. What is the most likely significance of a hs-CRP level of more than 10 mg/L?

• The patient may have a paraproteinemia
• The patient probably has an acute infection, and the test should be repeated

2. Which of the following statements about CRP is NOT true?

• It causes vasoconstriction of vascular tissue by stimulating the release of nitric oxide
• It inhibits the synthesis of nitric oxide

3. Which statement does NOT reflect reported links between CRP and CVD?

• Lowering CRP levels has not yet been demonstrated to reduce CV risk
• CRP levels are strongly associated with CVD risk

4. All the following statements are correct, except:

• Statin therapy has been shown to reduce CRP levels in both men and women
• The clinical significance of lowering CRP levels has yet to be determined

5. The AHA/CDC recommendations include all the following suggestions for the use of CRP levels, except:

• Predicting recurrent CV events
• Screening for CRP in persons 55 years and older with LDL-C levels more than 130 mg/dL

N Engl J Med

1. Ross R. Atherosclerosis?an inflammatory disease. . 1999;340:115-126.

Circulation.

N

Engl J Med

2. Libby P, Ridker PM. Novel inflammatory markers of coronary risk: theory versus practice [editorial]. 1999;100:1148-1150. v3. Ridker PM, Cushman M, Stampfer MJ, et al. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. . 1997;336:973-979.

Circulation

4. Koenig W, Sund M, Fr?hlich M, et al. C-reactive protein, a sensitive marker of inflammation, predicts future risk of coronary heart disease in initially healthy middle-aged men: results from the MONICA (Monitoring Trends and Determinants in Cardiovascular Disease) Augsburg Cohort Study, 1984 to 1992. . 1999;99:237-242.

Circ Res

5. Blake GJ, Ridker PM. Novel clinical markers of vascular wall inflammation. . 2001;89:763-771.

J Clin Epidemiol.

6. Sakkinen P, Abbott RD, Curb JD, et al. C-reactive protein and myocardial infarction. 2002;55:445-451.

Circulation

7. Ridker PM, Cushman M, Stampfer MJ, et al. Plasma concentration of C-reactive protein and risk of developing peripheral vascular disease. . 1998;97:425-428.

Circulation

8. Ridker PM, Buring JE, Shih J, et al. Prospective study of C-reactive protein and the risk of future cardiovascular events among apparently healthy women. . 1998;98:731-733.

JAMA

9. Pradhan AD, Manson JE, Rossouw JE, et al. Inflammatory biomarkers, hormone replacement therapy, and incident coronary heart disease: prospective analysis from the Women's Health Initiative observational study. . 2002;288:980-987.

Stroke

10. 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. . 2001;32:2575-2579.

N Engl J Med

11. Ridker PM, Hennekens CH, Buring JE, et al. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. . 2000;342:836-843.

N Engl J Med

12. Ridker PM, Rifai N, Rose L, et al. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. . 2002;347:1557-1565.

Circulation.

13. Koenig W, L?wel H, Baumert J, et al. C-reactive protein modulates risk prediction based on the Framingham score: implications for future risk assessment: results from a large cohort study in southern Germany. 2004;109:1349-1353.

Circulation

14. Pearson TA, Mensah GA, Alexander RW, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. . 2003;107:499-511.

Circulation

15. Ridker PM. Cardiology patient page. C-reactive protein: a simple test to help predict risk of heart attack and stroke. . 2003;108:e81-e85.

JAMA

16. LaRosa JC, He J, Vupputuri S. Effect of statins on risk of coronary disease: a meta-analysis of randomized controlled trials. . 1999;282:2340-2346.

N Engl J Med.

17. Sacks FM, Pfeffer MA, Moye LA, et al, for the Cholesterol and Recurrent Events Trial Investigators. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. 1996;335:1001-1009.

Circulation

18. Ridker PM, Rifai N, Pfeffer MA, et al, for the Cholesterol and Recurrent Events (CARE) Investigators. Long-term effects of pravastatin on plasma concentration of C-reactive protein. . 1999;100:230-235.

JAMA

19. Downs JR, Clearfield M, Weis S, et al. Primary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels: results of AFCAPS/TexCAPS. . 1998;279:1615-1622.

N Engl J Med.

20. Ridker PM, Rifai N, Clearfield M, et al, for the Air Force/Texas Coronary Atherosclerosis Prevention Study Investigators. Measurement of C-reactive protein for the targeting of statin therapy in the primary prevention of acute coronary events. 2001;344:1959-1965.

Circulation.

21. Kinlay S, Schwartz GG, Olsson AG, et al, for the Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIRACL) Study Investigators. High-dose atorvastatin enhances the decline in inflammatory markers in patients with acute coronary syndromes in the MIRACL study. 2003;108:1560-1566.

N Engl J

Med

22. Cannon CP, Braunwald E, McCabe CH, et al, for the Pravastatin or Atorvastatin Evaluation and Infection Therapy?Thrombolysis in Myocardial Infarction 22 Investigators. Intensive versus moderate lipid lowering with statins after acute coronary syndromes. . 2004;350:1495-1504.

Circulation

23. Ridker PM, on behalf of the JUPITER Study Group. Rosuvastatin in the primary prevention of cardiovascular disease among patients with low levels of low-density lipoprotein cholesterol and elevated high-sensitivity C-reactive protein: rationale and design of the JUPITER Trial. . 2003;108:2292-2297.

Answers:

1. D; 2. B; 3. C; 4. C; 5. D