AIM:To reduce the incidence of postoperative anastomoticleak,stenosis,gastroesophageal reflux(GER)for patientswith esophageal carcinoma,and to evaluate the conventionalmethod of esophagectomy and esophagogastroplastymodified by a new three-layer-funnel-shaped(TLF)esophagogastric anastomotic suturing technique.METHODS:From January 1997 to October 1999,patientswith clinical stage Ⅰ and Ⅱ(Ⅱa and Ⅱb)esophagealcarcinoma,which met the enrollment criteria,were surgicallytreated by the new method(Group A)and by conventionaloperation(Group B).All the patients were followed at leastfor 6 months.Postoperative outcomes and complicationswere recorded and compared with the conventional methodin the same hospitals and with that reported previously byMcLarty et al in 1997(Group C).RESULTS:58 cases with stage Ⅰ and Ⅱ(Ⅱa and Ⅱb)esophageal carcinoma,including 38 males and 20 femalesaged from 34 to 78(mean age:57),were surgically treatedby the TLF anastomosis and 64 by conventional method inour hospitals from January 1997 to October 1999.The qualityof swallowing was improved significantly(Wilcoxon W=2 142,P=0.0 001)2 to 3 months after the new operation in GroupA.Only one patient had a blind anastomatic fistula diagnosedby barium swallow test 2 months but healed up 3 weekslater.Postoperative complications occurred in 25(43%)patients,anastomotic stenosis in 8(14%),and GER in 13(22%).The incidences of postoperative anastomotic leak,stenosis and GER were significantly decreased by the TLFanastomosis method compared with that of conventionalmethods(x~2=6.566,P=0.038;x~2=10.214,P=0.006;x~2=21.265,P=0.000).CONCLUSION:The new three-layer-funnel-shaped esophagogastric anastomosis(TLFEGA)has more advantagesto reduce postoperative complications of anastomotic leak,stricture and GER. AIM: To reduce the incidence of postoperative anastomoticleak, stenosis, gastroesophageal reflux (GER) for patients with esophageal carcinoma, and to evaluate the conventional method of esophagectomy and esophagogastroplasty modified by a new three-layer-funnel-shaped (TLF) esophagogastric anastomotic suturing technique. : From January 1997 to October 1999, patientswith clinical stage Ⅰ and Ⅱ (Ⅱa and Ⅱb) esophagealcarcinoma, which met the enrollment criteria, were surgically treated by the new method (Group A) and by conventional operation (Group B) .All the patients were followed at least for 6 months. Postoperative outcomes and complications were recorded and compared with the conventional method in the same hospitals and with that previously reported byMcLarty et al in 1997 (Group C) .RESULTS: 58 cases with stage Ⅰ and Ⅱ (Ⅱa and Ⅱb) esophageal carcinoma , including 38 males and 20 femalesaged from 34 to 78 (mean age: 57), were surgically treated by the TLF anastomosis and 64 by conventional method inour hospitals from January 1997 to October 1999. The qualityof swallowing was improved significantly (Wilcoxon W = 2 142, P = 0.0 001) 2 to 3 months after the new operation in GroupA. One patient had a blind anastomatic fistula diagnosed by barium swallow test The incidences of postoperative anastomotic leak, stenosis and GER were 8 (14%), and GER in 13 (22%). 2 months but healed up 3 weekslater. by the TLFanastomosis method compared to that of conventionalmethods .CONCLUSION: The new three-layer-funnel (x = 2 = 6.566, P = 0.0038; -shaped esophagogastric anastomosis (TLFEGA) has more advantagesto reduce postoperative complications of anastomotic leak, stricture and GER.