BACKGROUND Differences in the management of acute myocardial infarction have been reported among countries, but few studies have investigated this issue in regions of the United States. METHODS We compared the management of acute myocardial infarction among census regions across the United States, using data from the first Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries trial (GUSTO-1) comprising 21,772 patients, and from the American Hospital Association. RESULTS We found substantial regional variation in the management of acute myocardial infarction in the United States. Beta-blockers (prescribed for a range of 55 to 81 percent of patients in the various regions), nitrates (prescribed for 61 to 77 percent), and angiotensin-converting-enzyme inhibitors (prescribed for 18 to 23 percent) were used most often in New England, whereas calcium-channel blockers (31 to 42 percent) and lidocaine (14 to 43 percent) were used least often there. Similarly, the proportion of patients undergoing various cardiac procedures differed among regions (range for angiography, 52 to 81 percent of patients; angioplasty, 22 to 35 percent; and coronary-artery bypass surgery, 9 to 17 percent) and was lowest in New England. The regional use of cardiac procedures was closely related to their availability, except in New England. After the analysis was adjusted for clinical and hospital variables, patients in New England were found to be less likely to undergo angiography than patients in the other regions (odds ratio, 0.37; 95 percent confidence interval, 0.32 to 0.42). There was no apparent relation between the use of cardiac procedures and rates of recurrent infarction or death at 30 days or 1 year. CONCLUSIONS There is substantial regional variation in the use of cardiac medications and procedures to manage acute myocardial infarction in the United States. The use and availability of cardiac procedures are closely related. The management of acute myocardial infarction in New England is atypical in that the relatively limited availability of cardiac procedures does not account for their relatively low use in that region.