BACKGROUND The association between C-reactive protein (CRP) levels and risk of cardiovascular (CV) events has been reported in several studies. However, it is unclear whether a reduction in CRP is associated with a reduction in risk of clinical events. Therefore we sought to investigate, in a meta-regression analysis of randomized studies enrolling patients treated by statins, whether changes in CRP are associated with changes in risk of CV events or overall survival. METHODS Randomized trials enrolling patients treated by statins, reporting CRP at baseline and at end of follow-up, CV events [myocardial infarction (MI) and stroke], CV and all-cause mortality were selected. RESULTS Twenty-two trials enrolling 54,213 participants were included in the analysis. Meta-analysis showed that active treatment significantly reduced risk of all-cause death by 8%, myocardial infarction by 11%, stroke by 10.3% and the composite outcome (including CV death, MI and stroke) by 8%, whereas risks of CV mortality was not significantly reduced. Meta-regression analysis revealed that reduction in CRP levels was significantly associated only with the reduction of MI, whereas no relationship was identified between changes in CRP and risk of stroke, CV and all-cause mortality, and the composite outcome. CONCLUSIONS These findings demonstrate that statin-induced changes in CRP do not correlate with major CV events apart from the risk of MI nor with overall survival in high-risk patients. These data suggest that although CRP may be a surrogate marker for coronary risk, it should not be used for predicting the effectiveness of statin therapy.