Patients with severe left ventricular systolic dysfunction (LVSD) undergoing coronary artery bypass grafting (CABG) have an increased risk for morbidity and mortality. The purpose of this study was to assess the association of pre-CABG right ventricular (RV) function with outcome for patients with severe LVSD. We performed a retrospective evaluation of 41 patients with severe LVSD (left ventricular ejection fraction [LVEF] ≤25%) scheduled for nonemergent CABG. Data were obtained from review of medical records, transesophageal echocardiography tapes, and phone interview. The pre- and post-cardiopulmonary bypass (CPB) LVEF and the RV fractional area of contraction (RVFAC) were calculated by using intraoperative transesophageal echocardiography. Group 1 patients had an RVFAC ≤35% (n = 7), whereas Group 2 patients had RVFAC >35% (n = 34). The durations of mechanical ventilation and of intensive care unit and hospital stays are presented as the median. Pre-CABG LVEF was similar between Groups 1 and 2 (15.8% ± 3.3% versus 17.8% ± 3.9%). Compared with Group 2, Group 1 patients required greater duration of mechanical ventilation (12 days versus 1 day;P < 0.01), longer intensive care unit (14 versus 2 days;P < 0.01) and hospital (14 versus 7 days;P = 0.02) stays, had a more frequent incidence and severity of LV diastolic dysfunction, and had a smaller change in LVEF immediately after CPB (4.1% ± 8.3% versus 12.5% ± 9.2%;P = 0.03). All Group 1 patients died of cardiac causes within 2 yr of surgery; five died during the same hospital admission. Three Group 2 patients died: one of colon cancer at 18 mo after CABG and two of cardiac causes 24 and 48 mo after surgery. A fourth patient was awaiting cardiac transplantation 4 yr after surgery. The remaining Group 2 patients were New York Heart Association Classification I or II. For patients with severe LVSD undergoing CABG, pre-CPB RV dysfunction was associated with poor outcome. Patients with RVFAC >35% had a relatively uneventful perioperative course and good long-term survival, whereas patients with RVFAC ≤35% had a poor early and late outcome. Assessment of RV function is useful to further assess the risk of CABG.