Revascularization of the extremity was performed upon 110 patients after preoperative radionuclide ventriculography (RNVG). Mean ejection fraction (EF) was 50 +/- 13 per cent. Ventricular wall motion abnormalities were present in 46 per cent. Revascularization included inflow procedures, such as aortofemoral (n = 25) or extraanatomic bypass (axillofemoral or femorofemoral, n = 11); infrainguinal reconstruction, including femoropopliteal or distal bypass (n = 43), and other procedures to improve perfusion of the limb or correct complications after previous vascular reconstruction upon the extremity (n = 31). Perioperative (30 days) mortality rate was 0.9 per cent and 97.0 per cent of the patients were discharged alive from the hospital. Myocardial infarction (MI) occurred in 3.6 per cent, new ventricular arrhythmia in 1.8 per cent and congestive heart failure in 6.4 per cent of the patients during the perioperative period. During follow-up study (607 +/- 363 days), 7.3 per cent required major amputation, ipsilateral to reconstruction, 5.5 per cent required surgical or angiographic revision for hemodynamic failure of the reconstruction prior to thrombosis and 12.7 per cent thrombosed part or all of the reconstruction. Revascularization failure did not appear to be related to the level of cardiac function. Those with normal (greater than 50 per cent) EF had greater over-all survival by life table analysis than those with EF less than or equal to 50 per cent (p = 0.0006, Mantel-Cox test). Ventricular wall motion abnormalities were associated with reduced over-all survival (p = 0.008, Mantel-Cox test). The presence of angina or previous MI, singularly or in combination, did not have an adverse effect on over-all survival, whereas diabetes (p = 0.0058, Mantel-Cox test) and cigarette smoking (p = 0.0137, Breslow test) were associated with significantly diminished over-all survival. Preoperative RNVG can identify subgroups at a survival disadvantage after revascularization of the extremity in a population in which the presence of angina or previous MI does not predict survival.