Objective: To evaluate the impacts of an on-pump beating-heart versus an off-pump coronary artery bypass grafting (CABG) technique for surgical revascularization on the early clinical outcomes in patients with a left ventricular ejection fraction (EF) of 35% or less. Methods: A total of 216 consecutive patients with an echocardiographic estimated EF of 35% or less who underwent non-emergency, primary, isolated CABG from January 2010 to December 2014 at Department of Cardiovascular Surgery, Zhongshan Hospital, Fudan University were included in this study and were divided into either an OBCAB group (patients who received on-pump beating-heart CABG surgery, n=88) or an OPCAB group (patients who received off-pump CABG surgery, n=128). The early clinical outcomes were investigated and compared. The outcomes were compared between groups by t-test, χ2 test or Fisher's exact test, when appropriate. Results: No significant differences emerged between the two groups in baseline characteristics of the entire cohort except for more patients with diabetes and a larger left ventricular endo-diastolic diameter in the OBCAB group. Patients in the OBCAB group compared to the OPCAB group had a similar in-hospital mortality (3.4% vs. 4.7%, P= 0.741). Mean EF, as measured preoperatively and early postoperatively (before discharge), significantly improved from (31.0±2.8)% to (35.6±2.9)% (t=10.61, P=0.000) in the OBCAB group and from (31.0±2.9)% to (34.8±3.3)% (t=9.68, P=0.000) in the OPCAB group, respectively. The improvement of mean LVEF in the OBCAB group was significantly higher than that in the OPCAB group ((4.7±0.2)% vs. (3.6±0.3)%, t=29.53, P=0.000). Patients in the OBCAB group compared to the OPCAB group had a significant higher early postoperative EF ((35.6±2.9)% vs.(34.8±3.3)%, t=1.892, P=0.034) but shared a similar baseline EF ((31.0±2.8)% vs. (31.0±2.9)%, t=0.012, P=0.930). Patients in the OBCAB group compared to the OPCAB group received a greater number of grafts and an increased amount of drainage during the first 24 h (3.7±0.8 vs. 2.8±0.6, t=9.442, P=0.000; (715±187) ml vs. (520±148) ml, t=8.544, P=0.000, respectively), without evidence of worse in-hospital mortality or major postoperative morbidity. Conclusion: The on-pump beating-heart technique may be an acceptable alternative to the off-pump technique for surgical revascularization in patients with an estimated EF of 35% or less.