Background Early resolution of ST-segment deviation (ST recovery) on the postthrombolytic electrocardiograms and restoration of “normal” blood flow in the infarct-related artery are associated with improved outcomes after myocardial infarction (MI). Methods and Results To evaluate the relationships between ST recovery, infarct-related artery flow, and late survival we studied 766 patients with electrocardiograms recorded at a median of 167 minutes after thrombolytic therapy. Angiography was performed at 3 weeks, and follow-up was done at a median of 6.3 years (interquartile range [IQR] 5.0-8.4). At 10 years, the survival rates were 55% (95% CI 43-70) in patients with <30% ST recovery in the single lead with maximum ST elevation, 71% (95% CI 64-79) in those with 30% to 70% ST recovery, and 74% (95% CI 68-82) in those with >70% ST recovery (P = .0005), whereas ST recovery measured as the sum of voltage changes of either ST deviation (elevation or depression) or ST elevation was not associated with 10-year survival (log-rank test, P = .06 and P = .34, respectively). In patients with Thrombolysis In Myocardial Infarction (TIMI) grade 3 flow, ST recovery of >70% (vs <30% and 30% to 70%) in the lead with maximum ST elevation was associated with increased late survival (P = .04). On multivariate analysis, the predictors, at admission, of 5-year survival were age (P < .001), ST recovery (measured as a continuous variable, P = .001), diabetes (P = .003) and female gender (P = .02). When the ejection fraction (P = .003) and TIMI flow grade (P = .02) at 3 weeks were included in the analysis, the P value for ST recovery was .08. Conclusions ST recovery measured in the single lead with maximum ST elevation was a predictor of late survival, even in patients with TIMI grade 3 flow but ST recovery measured as the sum of voltage changes in all leads with ST deviation was not. This simple electrocardiographic parameter can identify patients with a reduced chance of survival who might benefit from additional therapies. (Am Heart J 2002;143:265-71.)