Mortality and morbidity after resection for adenocarcinoma of the gastroesophageal junction: predictive factors

BACKGROUND: Resection for adenocarcinoma of the gastroesophageal junction (AGEJ) is associated with severe mortality and morbidity. This retrospective study aimed to evaluate mortality and morbidity after resection for AGEJ and to determine their predictive factors. STUDY DESIGN: Data from 1,192 patients (mean age 65 +/- 11 years) who underwent resection for AGEJ by members of French Association of Surgery from 1985 to 2000 were collected. A stepwise logistic regression model was built to identify by multivariate analysis the variables independently associated with mortality, morbidity, anastomotic leakage, and major pulmonary complications. RESULTS: Distribution of Siewert's type was: I = 480 (40%), II = 500 (42%), and III = 212 (18%). Most type I and II tumors were treated by esophagectomy and proximal gastrectomy (93% and 58%, respectively), using an approach including a thoracotomy (82% and 64%, respectively); type III tumors were treated mainly by total gastrectomy and distal esophagectomy (83%), through an exclusive transabdominal approach (69%). Seventy-six (6%) patients died postoperatively. Only American Society of Anesthesiologists (ASA) scores III and IV (p < 0.001) and period of study (p = 0.025) were predictive of mortality. Predictive factors of overall morbidity (overall rate = 35%) were high ASA score (p < 0.001), age more than 60 years (p = 0.020), male gender (p = 0.039), and cervical anastomosis (p = 0.001). Factors predictive of anastomotic leakage (overall rate = 9%) were high ASA score (p = 0.006) and manual anastomosis (p = 0.010). Factors predictive of major pulmonary complications (overall rate = 23%) were high ASA score (p = 0.015), age more than 60 years (p < 0.001), anastomotic leakage (p < 0.001), and abdominal complications (p = 0.003). CONCLUSIONS: ASA score is a reliable predictive factor of operative mortality and morbidity after resection of AGEJ.