Prognostic significance of high on-clopidogrel platelet reactivity after percutaneous coronary intervention: Systematic review and meta-analysis

BackgroundA growing number of observational studies suggest that high on-clopidogrel platelet reactivity (HPR) is associated with recurrent thrombotic events after percutaneous coronary intervention. We aimed to perform an updated systematic review and meta-analysis on the clinical relevance of HPR to summarize the available evidence and to define more precise effect estimates.MethodsRelevant observational studies published between January 2003 and February 2010 were searched that presented intent-to-treat analyses on the clinical relevance of HPR measured with an adenosine diphosphate (ADP)–specific platelet function assay. The main outcome measures were cardiovascular (CV) death, definite/probable stent thrombosis (ST), nonfatal myocardial infarction (MI), and a composite end point of reported ischemic events. The outcome parameters were pooled with the random-effect model via generic inverse variance weighting.ResultsTwenty studies comprising a total number of 9,187 patients qualified. High on-clopidogrel platelet reactivity appeared to be a strong predictor of MI, ST, and the composite end point of reported ischemic events (odds ratios [95% CI]: 3.00 [2.26-3.99], 4.14 [2.74-6.25], and 4.95 [3.34-7.34], respectively; P < .00001 for all cases). According to the meta-analysis, patients with HPR had a 3.4-fold higher risk for CV death compared with patients with normal ADP reactivity (odds ratio 3.35, 95% CI 2.39-4.70, P < .00001). Although there were large differences in the methodology as well as in the definition of HPR between studies, the predicted risk for CV death, MI, or ST was not heterogeneous (I2: 0%, 0%, and 12%, respectively; P = not significant for all cases).ConclusionsHigh on-clopidogrel platelet reactivity, measured by an ADP-specific platelet function assay, is a strong predictor of CV death, MI, and ST in patients after percutaneous coronary intervention.