Coronary bypass surgery was performed prior to hospital discharge in 303 (22%) of 1387 consecutive patients enrolled in the TAMI 1 to 3 and 5 trials of intravenous thrombolytic therapy for acute myocardial infarction. Bypass surgery was of emergency nature (less than 24 hours from treatment with intravenous thrombolytic therapy) in 36 (2.6%) and was deferred (greater than 24 hours) in 267 (19.3%) patients. The indications for bypass surgery included failed angioplasty (12%); left main or equivalent coronary disease (9%); complex or multivessel coronary disease (62%); recurrent postinfarction angina (13%); and refractory pump dysfunction, mitral regurgitation, ventricular septal rupture or abnormal predischarge functional test (1% each). Although patients having bypass surgery were older (59.5 +/- 9.8 versus 56.0 +/- 10.2 years, (p less than 0.0001), had more extensive coronary artery disease (46% with three-vessel disease versus 11%, (p less than 0.0001), had more frequent diabetes mellitus (19% versus 15%, (p = 0.048), had more prior infarctions (p less than 0.0001), had more severe initial depression in global left ventricular ejection fraction (48.0 +/- 11.9% versus 51.8 +/- 11.9%, p = 0.0002), and regional infarct zone (-2.7 +/- 0.94 versus -2.5 +/- 1.1 SD/chord, p = 0.02) and noninfarct zone function (-0.36 +/- 1.8 versus 0.43 +/- 1.6 SD/chord, p less than 0.0001) than patients not having coronary bypass surgery, no difference in the incidence of death in hospital (7% surgical versus 6% nonsurgical) or death at long-term follow-up of hospital survivors (7% surgical versus 6% nonsurgical) was noted between groups. Surgical patients demonstrated a greater degree of recovery in left ventricular ejection fraction (3.4 +/- 9.8% versus 0.16 +/- 8.5%, p = 0.036) and infarct zone regional function (0.71 +/- 1.1 versus 0.34 +/- 0.99 SD/chord, p = 0.001) when immediate (90 minutes following initiation of thrombolytic therapy) and predischarge (7 to 14 days after treatment) contrast left ventriculograms were compared than did patients who received only intravenous thrombolytic therapy with or without coronary angioplasty. These data suggest a beneficial influence of coronary bypass surgery on left ventricular function and possibly on the clinical outcome of patients initially treated with intravenous thrombolytic therapy for acute myocardial infarction.