Abstract: Objective To investigate the effect of intravenous lidocaine on postoperative atrial fibrillation, IL-6 and IL-10 levels in patients undergoing cardiac surgery. Methods Sixty patients scheduled for cardiac surgery requiring cardiopulmonary bypass, were randomly divided into two groups based on computer-generated randomized assignments： lidocaine group （Group L, n = 30） and control group （Group C, n = 30）. Patients in Group L received three-phase intravenous lidocaine infusion regimen as follows： upon induction of general anesthesia, a bolus of intravenous lidocaine 1.5 mg/kg was given over a minimum of 5 min; subsequently, lidocaine 2.3 mg/（kg·h） was infused for the first 3 h; lidocaine was then reduced to 0.8 mg/（kg·h） and discontinued 24 h postoperatively. Patients in Group C received intravenous saline infusion. The bolus amounts and infusion rates were identical to those in Group L. Blood samples were collected before, at the end of surgery and 24 h after surgery for measurement of IL-6 and IL-10 levels. The episode of new onset of atrial fibrillation and the development of cardiovascular events after surgery were recorded. Results Compared with preoperative levels, serum IL-6 and IL-10 levels at the end of surgery and 24 h after surgery were significantly higher （all P < 0.01）. Compared with Group C, serum IL-6 and IL-10 levels at the end of surgery and 24 h after surgery were significantly lower in Group L （P < 0.05 or P < 0.01）. In Group C, postoperative extubation time was （15.9±3.6） h, significantly longer than （12.2±4.7） h in Group L （P < 0.01）. Compared with Group C, the incidence of postoperative atrial fibrillation tended to decline in Group L（10% vs. 24%, P > 0.05）. Conclusions Perioperative intravenous infusion of lidocaine could decrease postoperative serum IL-6 and IL-10 levels, shorten tracheal extubation time and lower the risk of postoperative atrial fibrillation.