Will synchronous esophageal and lung resection increase the incidence of anastomotic leaks? A multicenter retrospective study

Abstract

Background: Reports on combined resection for synchronous lung lesions and esophageal cancer (CRLE) cases are rare and mostly individual cases. Furthermore, the feasibility of CRLE has always been a controversial topic. In the current study, the authors retrospectively analyzed the feasibility of CRLE and established an individualized prediction model for esophageal anastomotic leaks after CRLE by performing a multicenter retrospective study. Methods: Patients who underwent esophagectomy between January 2009 and June 2021 were extracted from a four-center prospectively maintained database, and those with CRLE at the same setting were matched in a 1:2 propensity score-matched (PSM) ratio to esophagectomy alone (EA) patients. A nomogram was then established based on the variables involved in multivariate logistic regression analysis. Internal validation of the nomogram was conducted utilizing Bootstrap resampling. Decision and clinical impact curve analysis were computed to assess the practical clinical utility of the nomogram. A prognosis analysis for CRLE and EA patients by Kaplan–Meier curves was conducted. Results: Of the 7152 esophagectomies, 216 cases of CRLE were eligible, and 1:2 ratio propensity score-matched EA patients were matched. The incidence of anastomotic leaks following CRLE increased significantly (P=0.035). The results of the multivariate analysis indicated the leaks varied according to the type of lung resection (anatomic>wedge resection, P=0.016) and site of resected lobe (upper>middle/low lobe; P=0.027), and a nomogram was established to predict the occurrence of leaks accurately (area under the curve=0.786). Although no statistically significant difference in overall survival (OS) was observed in the CRLE group (P=0.070), a trend toward lower survival rates was noted. Further analysis revealed that combined upper lobe anatomic resection was significantly associated with reduced OS (P=0.027). Conclusion: Our study confirms that CRLE is feasible but comes with a significantly increased risk of anastomotic leaks and a concerning trend of reduced survival, particularly when upper lobe anatomic resections are performed. These findings highlight the need for careful patient selection and surgical planning when considering CRLE.