Background Restenosis after percutaneous coronary intervention (PCI) constitutes a serious complication in the treatment of cardiovascular disease, but known risk factors do not fully account for the observed restenosis risk. Preliminary studies of infection or inflammation in restenosis report varied results. We tested whether C-reactive protein (CRP) or pathogen burden (seropositivity to 0, 1, 2, or 3 pathogens, of Chlamydia pneumoniae [Cpn], cytomegalovirus [CMV], or Helicobacter pylori [Hpy]) predict clinical restenosis after percutaneous coronary intervention (PCI). Methods Blood samples were collected from 415 patients undergoing PCI, and levels of plasma CRP and antibodies to Cpn, CMV, and Hpy were measured. The patient's medical history, demographics, and procedural data were recorded. Patient end points were determined for as long as 6 months as a means of evaluating the incidence of clinical restenosis and major adverse cardiac events. Results The average patient age was 62 years, and 80% of patients were male. Fifty-eight patients (14%) experienced clinical restenosis, whereas 17 patients (4%) died or had an acute myocardial infarction. After adjusting for 19 possible predictors, we found the pathogen burden (P-trend =.04, adjusted odds ratio [OR] 1.5 per number of pathogens) and minimum luminal diameter (P =.003, OR 1.8 per mm decrease) to be significant predictors of clinical restenosis. Male sex was a nonsignificant predictor of restenosis (P =.06, OR 2.2), but CRP was not significant after adjustment (P-trend =.10, OR 0.73 per tertile). Conclusion Pathogen burden was associated with clinical coronary restenosis, an association that deserves further exploration and evaluation. CRP, a marker of inflammation, was not associated with an increased risk of restenosis. (Am Heart J 2002;144:491-500.)