Background: In the percutaneous coronary intervention (PCI) era, little evidence exists regarding the incidence, predictors and long-term mortality of recurrent myocardial infarction (Re-MI) following discharge for acute myocardial infarction (AMI).
Methods and results: A total of 7,870 patients who survived AMI were studied with a median follow-up period of 3.9 years: 353 patients (4.5%) experienced Re-MI, with 7 of those dying within 30 days, which was classified as fatal Re-MI. The incidence of Re-MI per year was 2.65% for the first year, and 0.91-1.42% thereafter up to 5 years. Multivariate Cox regression analyses revealed that predictors of Re-MI were diabetes mellitus (hazard ratio (HR): 2.079, P<0.001), history of MI (HR: 1.767, P=0.001), and advanced age (HR: 1.021, P=0.001). These 3 predictors remained significant when angiographic and procedural parameters were incorporated into the analyses. The incidence and adjusted risk of Re-MI increased when these variables were clustered (P<0.001). The all-cause mortality rate was significantly higher in patients with Re-MI than in those without (HR: 2.206, P<0.001).
Conclusions: In post-AMI patients treated in the PCI era, the incidence of Re-MI is low compared with that reported during the past 30 years. Patients' clinical factors of diabetes mellitus, history of MI, and advanced age appear to affect the occurrence of Re-MI after hospital discharge, and Re-MI still carries a risk for subsequent mortality.