Factors associated with postoperative pulmonary morbidity after esophagectomy for cancer

Abstract

Background: Most studies analyzing risk factors for pulmonary morbidity date from the early 1990s. Changes in technology and treatment such as minimally invasive esophagectomy (MIE) and neoadjuvant treatment mandate analysis of more contemporary cohorts. Methods: Predictive factors for overall and specific pulmonary morbidity in 858 patients undergoing esophagectomy between 1998 and 2008 in five Australian university hospitals were analyzed by logistic regression models. Results: A total of 394 patients underwent open esophagectomy, and 464 patients underwent MIE. A total of 259 patients received neoadjuvant chemoradiotherapy, 139 preoperative chemotherapy alone, and 2 preoperative radiotherapy alone. In-hospital mortality was 3.5%. Smoking and the number of comorbidities were risk factors for overall pulmonary morbidity (odds ratio [OR] 1.47, P = 0.016; OR 1.35, P = 0.001) and pneumonia (OR 2.29, P = 0.002; 1.56, P = 0.005). The risk of respiratory failure was higher in patients with more comorbidities (OR 1.4, P = 0.035). Respiratory comorbidities (OR 3.81, P = 0.017) were strongly predictive of postoperative acute respiratory distress syndrome (ARDS). ARDS (4.51, P = 0.032) or respiratory failure (OR 8.7, P < 0.001), but not anastomotic leak (OR 2.22, P = 0.074), were independent risk factors for death. MIE (OR 0.11, P < 0.001) and thoracic epidural analgesia (OR 0.12, P = 0.003) decreased the risk of respiratory failure. Neoadjuvant treatment was not associated with an increased risk of pulmonary complications. Conclusions: Preoperative comorbidity and smoking were risk factors for respiratory complications, whereas neoadjuvant treatment was not. MIE and the use of thoracic epidural analgesia decreased the risk of respiratory failure. Respiratory failure and ARDS were the only independent factors associated with an increased risk of in-hospital death, whereas anastomotic leakage was not. © 2010 Society of Surgical Oncology.