Predictors And Prevention Of Complicating Anastomotic Leak After Esophagectomy

Objective:To analysis the predictors of complicating anastomotic leak after esophagectomy, and to find out preventive measures of cervical and thoracic anastomotic leak, improving the incidence and outcome of anastomotic leak.Methods:To perform a retrospective analysis of the patients suffering from esophageal or gastric cardia carcinoma, admitting in the department of Thoracic Surgery, Tongji Hospital, Wuhan, from Jan.2012to Aug.2013. The patients were divided into2groups depending on suffering from anastomotic leak after esophagectomy or not. To find out predictors of complicating anastomotic leak after esophagectomy and preventive measures by applying single factor analysis and multiple factors analysis. All analysis were performed by using SPSS software (version13.0; SPSS Inc., Chicago, IL, USA). Continuous variables were analyzed by2-tailed t tests and categorical data were evaluated by Chi-square test as appropriate. Logistic regression analysis was used to analyze the association between the risk factors and anastomotic leak. P<0.05was considered statistically significant.Results:1. There were122cervical anastomoses and262thoracic anastomoses. There were55patients complicating anastomotic leak after esophagectomy, including32cervical anastomotic leak and23thoracic anastomotic leak. Single factor analysis showed7factors, including sexuality (male:15.9%, female6.3%), anastomoses location (cervical:26.2%, thoracic:8.8%), anastomotic method (manual:75.0%, stapler:13.7%), embedding method (wrapping of the omental pedicle flap:12.0%, relaxation suture:25.0%), respirator time, ICU stay time, complicating pulmonary infection or not (complicated:33.6%, non-complicated5.7%), related with anastomotic leak. Multiple factors analysis showed sexuality (OR0.21,95%CI0.05-0.83), anastomoses location (OR4.06,95%CI2.06-7.99), anastomotic method (OR0.02,95%CI0.00-0.28), and complicating pulmonary infection or not (OR8.19,95%CI3.99-16.80) associated with anastomotic leak.2. For cervical anastomoses, single factor analysis showed10factors, including sexuality (male:32.2%, female9.4%), BMI, preoperative serum protein level (total protein, albumin), carcinoma location (mid-upper:46.2%, mid-lower:21.1%), lifting route (retrostemal:46.4%, posterior mediastinal:20.2%), operative time, blood loss during operation, prophylactic tracheotomy or not (yes:8.7%, no:30.3%), respirator time, ICU stay time, complicating pulmonary infection or not (complicated:51.1%, non-complicated10.7%), related with cervical anastomotic leak. Multiple factors analysis showed carcinoma location (OR5.81,95%CI1.41-23.94), lifting route (OR0.03,95%CI0.00-0.31), and complicating pulmonary infection or not (OR66.17,95%CI5.11-857.69) associated with cervical anastomotic leak.3. For thoracic anastomoses, single factor analysis showed4factors, including embedding method (wrapping of the omental pedicle flap:7.2%, relaxation suture:20.0%), respirator time, ICU stay time, complicating pulmonary infection or not (complicated:22.2%, non-complicated3.7%), related with cervical anastomotic leak. Multiple factors analysis showed complicating pulmonary infection or not (OR4.24,95%CI1.84-12.1) associated with cervical anastomotic leak..4. Wrapping of the omental pedicle flap around esophagogastric anastomoses could ease the symptoms associated with anastomotic leak and prevent anastomotic stricture forming (wrapping of the omental pedicle flap:18.8%, relaxation suture:20.0%).Conclusions:1. Predictors of complicating anastomotic leak after esophagectomy included sexuality, anastomoses location, anastomotic method and complicating pulmonary infection or not. To treat middle and lowe esophageal carcinoma by thoracic anastomoses and to prevent pulmonary infection could decreasing risk of anastomotic leak.2. Predictors of complicating cervical anastomotic leak after esophagectomy included carcinoma location, retrosternal or posterior mediastinal route and and complicating pulmonary infection or not. Prophylactic tracheotomy could reduce risk of cervical anastomotic leak.3. Wrapping of the omental pedicle flap around thoracic esophagogastric anastomoses could decrease risk of thoracic anastomotic leak. Applying common stapler for thoracic anastomoses with wrapping of the omental pedicle flap around thoracic esophagogastric anastomoses in thoracoscopic esophagectomy was safe and effective.4. Wrapping of the omental pedicle flap around esophagogastric anastomoses could ease the symptoms associated with anastomotic leak and prevent anastomotic stricture forming.