BACKGROUND: The present study was designed to investigate the long-term prognostic impact of procedure-related and spontaneous myocardial infarction (MI) on cardiovascular mortality in patients with non-ST-elevation acute coronary syndrome. METHODS AND RESULTS: Five-year follow-up after procedure-related or spontaneous MI was investigated in the individual patient pooled data set of the FRISC-II (Fast Revascularization During Instability in Coronary Artery Disease), ICTUS (Invasive Versus Conservative Treatment in Unstable Coronary Syndromes), and RITA-3 (Randomized Intervention Trial of Unstable Angina 3) non-ST-elevation acute coronary syndrome trials. The principal outcome was cardiovascular death up to 5 years of follow-up. Cumulative event rates were estimated by the Kaplan-Meier method; hazard ratios were calculated with time-dependent Cox proportional hazards models. Adjustments were made for the variables associated with long-term outcomes. Among the 5467 patients, 212 experienced a procedure-related MI within 6 months after enrollment. A spontaneous MI occurred in 236 patients within 6 months. The cumulative cardiovascular death rate was 5.2% in patients who had a procedure-related MI, comparable to that for patients without a procedure-related MI (hazard ratio 0.66; 95% confidence interval, 0.36-1.20, P=0.17). In patients who had a spontaneous MI within 6 months, the cumulative cardiovascular death rate was 22.2%, higher than for patients without a spontaneous MI (hazard ratio 4.52; 95% confidence interval, 3.37-6.06, P<0.001). These hazard ratios did not change materially after risk adjustments. CONCLUSIONS: Five-year follow-up of patients with non-ST-elevation acute coronary syndrome from the 3 trials showed no association between a procedure-related MI and long-term cardiovascular mortality. In contrast, there was a substantial increase in long-term mortality after a spontaneous MI.