Determinates of infarct size in patients with acute myocardial infarction (AMI) undergoing percutaneous coronary intervention (PCI) have been incompletely characterized, in part because of the limited sample size of previous studies. Databases therefore were pooled from 4 contemporary trials of primary or rescue PCI (EMERALD, COOL-MI, AMIHOT, and ICE-IT), in which the primary end point was infarct size assessed using technetium-99m sestamibi single-photon emission computed tomographic imaging, measured at the same core laboratory. Of 1,355 patients, infarct size was determined using technetium-99m sestamibi imaging in 1,199 patients (88.5%), at a mean time of 23 +/- 15 days. Median infarct size of the study population was 10% (interquartile range 0% to 23%; mean 14.9 +/- 16.1%). Using multiple linear regression analysis of 18 variables, left anterior descending infarct artery, baseline Thrombolysis In Myocardial Infarction grade 0/1 flow, male gender, and prolonged door-to-balloon time were powerful independent predictors of infarct size (all p <0.0001). Other independent correlates of infarct size were final Thrombolysis In Myocardial Infarction grade <3 flow (p = 0.0001), previous AMI (p = 0.005), symptom-onset-to-door time (p = 0.021), and rescue angioplasty (p = 0.026). In conclusion, anterior infarction, time to reperfusion, epicardial infarct artery patency before and after reperfusion, male gender, previous AMI, and failed thrombolytic therapy were important predictors of infarct size after angioplasty in patients with AMI assessed using technetium-99m sestamibi imaging and should be considered when planning future trials of investigational drugs or devices designed to enhance myocardial recovery.