ObjectiveTo summarize clinical experience and outcomes of modified Bentall procedure for the treatment of Stanford type A3 aortic dissection (AD). MethodsFifty-four patients with Stanford type A3 AD underwent aortic root replacement in Nanjing Hospital Affiliated to Nanjing Medical University from January 2004 to June 2013. There were 41 male and 13 female patients with their age of 21-73 years. According to different surgical methods, all the 54 patients were divided into 2 groups. In group A, there were 36 patients who received conventional Bentall procedure. In group B, there were 18 patients who received modified Bentall procedure ('gate' anastomosis of the coronary ostia for patients with aortic root less than 45 mm and nonsignificant displacement of the coronary ostia). Postoperative outcomes were compared between the 2 groups. ResultsThere was no statistical difference in age or gender between the 2 groups. Mean diameter of the aortic root of group A was significantly larger than that of group B (52.11±3.62 mm vs. 40.72±2.67 mm, P=0.000). There was no statistical difference in operation time, cardiopulmonary bypass time, intraoperative circulation arrest time, postoperative thoracic drainage or length of ICU stay between the 2 groups (P > 0.05). Four patients died postoperatively including 2 patients with uncontrollable bleeding, 1 patient with abdominal AD rupture and 1 patient with acute pulmonary embolism. There was no statistical difference in in-hospital mortality between group A and group B[5.56% (2/36)vs. 11.11% (2/18), P=0.462]. Forty-eight patients were followed up for 3 months, and 2 patients were lost during follow-up. Forty-eight patients received computed tomography angiography without false aneurysm formation in the aortic root, coronary ostial aneurysm or stenosis. ConclusionModified Bentall procedure ('gate' anastomosis of the coronary ostia)is simple and effective for patients with aortic root less than 45 mm and nonsignificant displacement of the coronary ostia.