Association between C-reactive protein and angiographic restenosis after bare metal stents: an updated and comprehensive meta-analysis of 2747 patients

BACKGROUND: Previous studies have shown conflicting results about the relationship between baseline C-reactive protein (CRP) and restenosis after stenting with bare metal stent (BMS). METHODS: We assessed the association between serum CRP and angiographic restenosis after BMS by meta-analysis. Studies that reported basal serum CRP levels, above a prespecified cutoff value, before BMS deployment were included. An inverse random weighted meta-analysis was performed by entering the logarithm of the odds ratio (OR) of angiographic restenosis with its standard error for each study. RESULTS: Nine studies enrolling 2747 patients were selected. CRP threshold value was around 3 mg/l in three studies, 5 mg/l in four studies, and 6.98 and 10 mg/l in one study. Follow-up duration was 6.2+/-3.0 (mean+/-S.D.) months. Higher preprocedural CRP levels were a significant predictor of angiographic restenosis: OR 1.59, 95% confidence interval 1.21-2.07, P=.001. Heterogeneity was found: chi2 14.47, P=.07; I2=44.7%. Publication bias was also detected (P=.01, Egger's test). A sensitivity analysis, after excluding each study in turn, confirmed the predictive value of higher CRP levels, in agreement with the results of the main analysis. CONCLUSIONS: Among patients with coronary artery disease, undergoing percutaneous coronary intervention with BMS, higher baseline CRP levels are associated with higher risk of angiographic restenosis. A targeted therapeutic approach to patients with high baseline CRP, based on statins, oral corticosteroids, or PPAR gamma agonists, or selective use of drug-eluting stents, aiming at abating the higher risk of in-stent restenosis should be considered