OBJECTIVE Off-pump coronary artery bypass grafting (CABG) [OPCAB] is preferred to conventional CABG with cardiopulmonary bypass (CPB) for many specific subgroups, such as elderly patients and high-risk patients. Whether OPCAB should be the first choice of surgical procedure for a wider range of coronary artery disease patient subgroups remains controversial. The purpose of this study was to compare the clinical results of OPCAB and CPB in our hospital. METHODS We retrospectively analyzed the results for 404 patients who received OPCAB or CPB between March 2000 and December 2001. Surgical methods adopted were at the discretion of the attending surgeon. In addition, subgroups of patients were analyzed to assess whether or not OPCAB was more successful in patients with specific characteristics. RESULTS There were no significant differences in the demographic data between the 2 groups. The perioperative mortality rate was similar for the CPB and OPCAB groups (2.5% and 2.0%, respectively; p = 1.00.) OPCAB was superior with respect to inotropic use, postoperative bleeding amount, transfusion amount, and length of intensive care unit and hospital stay (p < 0.05 for all), and also had a lower incidence of re-entry due to bleeding (6.7% vs 13.6%, p = 0.04) and prolonged hospitalization > 30 days (1.5% vs 10.8%, p < 0.01). At 12 months' follow-up, the OPCAB group had a lower rate of mortality (1.5% vs 5.6%, p = 0.03) and a comparable rate of readmission for cardiac reasons (6.6% vs 9.6%, p = 0.28). The rate of poor in-hospital outcome was lower in the OPCAB than in the CPB group (3.0% vs 13.3%, p </= 0.01), while the percentage with poor results on follow-up was comparable between the 2 groups (7.1% vs 12.1%, p = 0.09). In addition, subgroups defined as: female, age > 65 years, age < 65 years, diabetes, peripheral arterial occlusive disease, end-stage renal disease, left ventricular ejection fraction < 50%, preoperative intra-aortic balloon pump use, and left-main disease had better in-hospital outcome in the OPCAB group compared with the CPB group, while no subgroups had worse in-hospital outcomes with OPCAB. The 12-month follow-up outcome was similar with the 2 techniques. CONCLUSIONS Use of OPCAB in CABG operation was associated with improved in-hospital outcome and similar follow-up outcome compared to CPB. No subgroup had a worse outcome with OPCAB.