Simultaneous carotid disease was documented in 275 (2.8%) of 9714 patients scheduled for coronary artery bypass (CAB), including 80 (29%) who had had previous neurologic events and 195 with severe (greater than or equal to 70% diameter), asymptomatic carotid stenosis. Preliminary carotid endarterectomy (CE) was feasible before CAB in only 24 patients with stable cardiac disease (group I). Another 129 patients with unstable disease (group II) had unilateral, asymptomatic carotid lesions and were prospectively randomized to receive either combined operations (IIA; n = 71) or CAB followed by delayed CE (IIB; n = 58). The remaining 122 patients (group III) had symptomatic or bilateral carotid stenosis and were managed on a selective basis without randomization. The operative mortality rate ranged from 4.2% to 5.2%, and the early stroke rates were 4.2% in group I, 7.8% in group II, and 11% in group III. Postoperative strokes occurred after CAB in nine (4.7%) of the 193 patients protected by preliminary or simultaneous CE, compared with six (7.4%) of the 81 who received only delayed CE. Nevertheless, the composite stroke risk for "reverse-staged" procedures in group IIB (14%) exceeded that for combined operations (2.8%) in group IIA (p = 0.045). The stroke rate was 11% (7/61) when delayed CE was performed within 2 weeks after CAB compared with 2.2% (1/46) with longer staging intervals.