Proposed cardiovascular risk assessment algorithm using high-sensitivity C-reactive protein and lipid screening.

Several prospective epidemiologic studies from the United States and Europe have demonstrated that high-sensitivity C-reactive protein (hs-CRP) is a predictor of future coronary events among apparently healthy men and women. For example, findings from the Multiple Risk Factors Intervention Trial demonstrated a correlation between hs-CRP and coronary heart disease mortality among male smokers followed over a 17-year period [relative risk (RR) 5 2.8; 95% confidence interval (CI), 1.4–5.4] (1). A similar positive association between hs-CRP and future coronary events was noted in the Cardiovascular Health Study and Rural Health Promotion Project, which included elderly men and women with subclinical cardiovascular disease (2). A direct positive association between hs-CRP and future coronary events was also reported in apparently healthy men from the Physician’s Health Study (PHS); those in the highest quartile of hs-CRP had twice the risk of future stroke (RR 5 1.9; 95% CI, 1.1–3.3), three times the risk of future myocardial infarction (RR 5 2.9; 95% CI, 1.8–4.6), and four times the risk of future peripheral vascular disease (RR 5 4.1; 95% CI, 1.2–6.0) (3, 4). Furthermore, both the MONICA-Augsburg cohort (5) and the Helsinki Heart Study (6) showed that compared with those with low hs-CRP, individuals with the highest hs-CRP concentrations were at approximately three times the risk of developing future coronary events. Finally, two reports from the Women’s Health Study (WHS) showed that hs-CRP is also a strong predictor of future cardiovascular events in women (RR 5 4.4; 95% CI, 2.2–8.9) (7, 8). In fact, in that study, which directly compared several novel risk factors to standard lipid screening, hs-CRP was the single strongest predictor of future vascular risk (8). Specifically, as shown in Fig. 1, the relative risk associated with being in the top vs the bottom quartile for hs-CRP in the WHS was substantially greater than that associated with other “novel” risk factors such as homocysteine and lipoprotein(a) [Lp(a)], and in fact was greater than that associated with the usual lipid markers of risk. In almost all of these prospective studies, risk estimates associated with hs-CRP were independent of other recognized cardiovascular risk factors. Data from both the PHS and WHS demonstrate that the joint effects of hs-CRP and lipid screening are greater than the product of the individual effects of each risk factor considered alone (8, 9). Furthermore, when study participants were stratified according to the quintile of hs-CRP and the quintile of the ratio of total cholesterol to HDL-cholesterol (TC:HDL-C ratio), the relative risk of first coronary events in those in the highest quintiles of both hs-CRP and TC:HDL-C ratio was approximately eightto ninefold higher than that of those in the lowest quintiles of these analytes. In all of these analyses, risk prediction models that incorporated TC:HDL-C ratio were significantly better (P ,0.001) than those based on hs-CRP alone (8, 9). For the purpose of assessing risk of first coronary events, hs-CRP concentration should be interpreted using cut points established by prospective clinical studies. As the distribution of hs-CRP concentrations is (rightward) skewed, each patient should be classified into a quintile (or quartile) based on the measured hs-CRP concentra-