Abstract: AIM To study the safety and efficacy of improved carbon dioxide (CO2) technique in the prevention of gas embolism in children with congenital heart disease (CHD). METHODS Ninety patients with CHD in our hospital from January 2014 to June 2015 were selected for this study and randomly divided into three groups (n=30 each): group CO2 10 L/min (CO2 flow 10 L/min), CO2 5 L/min group (CO2 flow 5 L/min) and control group (with no carbon dioxide). Head low, heart bottom massage and other mechanical methods were used for gas exhaust in all three groups. The effects of gas exhaust, blood gas analysis, blood pressure (PaCO2), surgical time, aortic clamping time and cardiopulmonary bypass time were compared among the three groups. RESULTS Surgical procedure was successfully performed in all patients. No statistically significant difference was found in surgical time, aortic blocking and cardiopulmonary bypass time among the three groups. When the aorta was opened for 10 min, the number of bubbles in the meningeal artery in CO2 10 L/min group and CO2 5 L/min group was less than that in the control group ［(16±3) times/min and (23±4) times/min vs.(50±7) times/min］ and the number of bubbles in the CO2 10 L/min group was lower than in the CO2 5 L/min group, with a significant difference (all P<0.05). After stopping cardiopulmonary bypass, no statistically significant difference in the number of bubbles in the meningeal artery was found among the three groups. After opening the aorta and stopping the cardiopulmonary bypass, there was no statistically significant difference in PaCO2 among the three groups. CONCLUSION The improved CO2 technique is efficacious and safe for pediatric patients with CHD because it can significantly reduce the number of gas emboli and increase the air flow.