Prediction of major pulmonary complications after esophagectomy

Abstract

Background: Pulmonary complications are the most frequent morbid event after esophagectomy. Understanding factors that are associated with pulmonary complications may help in patient selection and postoperative management.

Methods: We performed a retrospective review of patients who underwent esophagectomy between 1980 and 2009. Univariate analysis was used to identify potential covariates for the development of major pulmonary complications. Multivariable logistic regression analysis was used to identify predictors of complications. A scoring system was developed, and its ability to predict complications was assessed.

Results: A total of 516 patients (382 men [74%]) with a mean age of 59.0±12.5 years underwent esophagectomy for cancer (398 [77%]) or benign disease. Major pulmonary complications occurred in 197 patients (38%) and were associated with a 10-fold increase in operative mortality (2.5% vs 28%; p<0.001). Independent predictors included patient age, forced expiratory volume in 1 second (% predicted), diffusion capacity of the lung for carbon monoxide (% predicted), performance status, serum creatinine, current cigarette use, and transthoracic resection. The scoring system (based on weighted scores for the first 4 covariates listed above) predicted pulmonary complications with an accuracy of 70.8% (p<0.001). Score groups identified an incremental risk of complications of 0 to 2, 12%; 3 to 4, 18%; 5 to 6, 46%; 7 to 8, 52%; and 9 to 13, 60% (p<0.001).

Conclusions: Major pulmonary complications are frequent after esophagectomy and can be predicted using commonly available clinical information. A scoring system identifying the risk of such complications may assist in patient selection and in providing appropriate resources for postoperative management of higher-risk patients.