Introduction: A number of large-scale retrospective studies revealed that off-pump coronary artery bypass (OPCAB) was superior to on-pump coronary artery bypass (ONCAB). The aim of the study was to investigate risk factors for mortality when OPCAB is converted to ONCAB. Methods: Patients who underwent OPCAB conversion to ONCAB at the Beijing Anzhen Hospital between January 2003 and January 2013 were assigned to the non-survivor and survivor groups. Background demographics, illness history and preoperative, intraoperative and postoperative variables were compared. Results: Of the 247 cases, 15.4% of the patients died. Patients in the non-survivor group were older and more frequently had diabetes mellitus (DM), arrhythmia, myocardial infarction (MI) in the past 30 days (all p<0.05) and MI combined with mitral regurgitation (p<0.0001); they more frequently had bigger left ventricular end-diastolic dimension (p=0.0019), greater fall in blood pressure, ventricular fibrillation for longer periods, longer conversion time and bypass graft occlusion. All patients in the non-survivor group received intra-aortic balloon pump compared to 89.5% in the survivor group and extracorporeal membrane oxygenation was more common. Left main coronary artery disease (OR=4.431, 95%CI: 2.440-8.048, p<0.0001), blood pressure decline 40 mmHg (OR=0.509, 95%CI: 0.447-0.580, p<0.0001) and time for conversion to ONCAB 20 min were independently associated with mortality. Rates of postoperative complications, such as renal failure, cerebral infarction or hemorrhage, MI and redo sternotomy, were higher in the non-survivor group. Conclusions: Conversion from OPCAB to ONCAB is associated with high mortality. Risk factors include left main artery disease and duration of blood pressure decline >40 min.