Effect of thoracic versus cervical anastomosis on anastomotic leak among patients who undergo esophagectomy after neoadjuvant chemoradiation

Objective: To assess the rate and effect of anastomotic leak among patients who undergo esophagectomy with either thoracic or cervical anastomosis after neoadjuvant chemoradiation. Methods: We conducted a retrospective cohort study using data from the National Surgical Quality Improvement Program Esophagectomy Data File. We included adult patients who underwent esophagectomy for esophageal cancer (2016-2017) after neoadjuvant chemoradiation. We used inverse probability of treatment weighted regression adjustment to compare 30-day anastomotic leak and mortality rates among patients with cervical or thoracic anastomoses. We accounted for confounding due to patient-, surgeon-, and procedure-related variables. Results: Of the 908 patients who met inclusion criteria, 528 (58%) had a thoracic anastomosis and 119 (13%) experienced anastomotic leak. There was no statistically significant difference in leak rate for patients who underwent thoracic (12%) compared with cervical anastomoses (14%) in the inverse probability of treatment weighted regression adjustment analysis (P = .09). Although overall 30-day mortality was low (2.3%), it was significantly higher among patients who had an anastomotic leak (8.4% vs 1.4%; P < .01). Among patients with a leak, there was no significant difference in length of stay, mortality, or type of required intervention for patients with cervical versus thoracic anastomoses. Conclusions: Anastomosis type does not affect leak rates or mortality after esophagectomy in patients who have undergone neoadjuvant chemoradiation. Patient risk factors and surgeon experience should determine the ideal surgical approach for each patient.