C-reactive protein: potential use for the future.

Studying the effect that a difficult-to-quantify lifestyle factor such as alcohol consumption has on a general inflammatory marker such as C-reactive protein is far from a simple task. Methodologic problems are inherent in design, implementation, and analysis of any such study. Not surprisingly, then, these concerns can be raised with the study by Stewart et al that appears in this issue of the Journal. Initial sample size (and therefore the response rate) of the population receiving the survey questionnaire is not specified, nor are criteria used for the variables required to qualify for receiving the survey. Selection of a value of 0.30 mg/dL as indicative of the upper limit of normal for the Creactive protein reading in this study appears to be retrospectively determined and therefore not necessarily reflective of an abnormal value in populations differing in any way from specifically surveyed study group respondents. Use of a C-reactive protein assay with limited range is also potentially problematic (a lower cutoff of 0.21 mg/dL was reported in this study for anyone with a C-reactive protein value of 0.21 mg/dL or below; C-reactive protein values less than this lower cutoff value have been associated with increased cardiovascular risk). Proof of a causeand-effect relation between alcohol consumption and reduced likelihood of C-reactive protein elevation as a mechanism for coronary protection cannot be established from retrospective analysis. Finally, the fundamental parameter being looked at in this study (alcohol consumption) is not optimally assessed by self-report, with implications based on drinking frequency but without attention to quantity not necessarily reflecting the relation between alcohol use and any effect it might have on an inflammatory marker such as C-reactive protein. That said, the evolving concept of using a highly sensitive assay for C-reactive protein (hs-CRP) as a marker of coronary artery disease risk is enticing, and the message suggested by results of this study is both provocative and in line with the increasing body of literature rapidly accumulating in the exciting field of preventive cardiology. Hardly a week passes without report of a new potential association between C-reactive protein values and some commonly encountered medical condition, physiologic state, or cardiac risk factor. Examples include cardiac arrhythmias (C-reactive protein tends to increase with atrial fibrillation, especially when the rhythm is persistent rather than paroxysmal); exercise (more frequent physical activity appears to correlate with a lesser chance of C-reactive protein elevation); estrogen use (hormone replacement is associated with an increase in C-reactive protein regardless of whether estrogen is given alone or in combination with progesterone); and obesity and smoking (both associated with increased C-reactive protein values). Accumulating data suggest that even modest hsCRP elevation is associated with increased risk of future cardiovascular events. The theory behind this association is fascinating: coronary artery disease is now thought to be the manifestation of an inflammatory process. Patients most vulnerable to acute ischemic events are more likely to have unstable atherosclerotic plaques that are at increased risk of developing fissure, rupture, and thrombogenesis, which at any time can lead to acute thrombotic occlusion of a major coronary vessel. Patients with high-risk lesions might be expected to manifest signs of increased inflammatory activity. Such inflammation appears to occur not only locally (in the affected vessel wall), but also systemically, as suggested by increased circulating levels of inflammatory markers, such as cytokines and C-reactive protein. In fact, many patients at highest risk, including those who have acute coronary syndromes, manifest evidence of surprisingly widespread rather than localized inflammation in several areas of the coronary vasculature. Submitted, revised 16 August 2002. From the Department of Family Medicine, University of Florida, Gainesville. Reprints are not available.