Transthoracic esophagectomy compared to transhiatal extended gastrectomy for adenocarcinoma of the esophagogastric junction: a multicenter retrospective cohort study.

Optimal surgical treatment for Siewert type II esophagogastric junction adenocarcinoma is debated. The aim of this study was to compare transhiatal extended gastrectomy (TEG) and transthoracic esophagectomy (TTE). Patients with Siewert type II tumors who underwent a resection by TEG or TTE in two centers (Erasmus University Medical Center, Rotterdam, and University of Verona) between 2014 and 2019 were identified. To limit selection bias, patients were matched for baseline characteristics and compared with a multivariable logistic regression model. Some 159 patients treated by TEG (60 patients, 37.7%) or TTE (99 patients, 62.3%) were included. Patients in the TEG group were older, had less tumor invasion of the esophagus, and were more often excluded from neoadjuvant therapy. Post-operative morbidity was comparable (P = 0.88), while 90-day mortality was higher after TEG (90-day mortality 10.0% in TEG group vs. 2.0% in TTE group P = 0.01). R0 resection was achieved in 83.3% of patients after TEG and in 97.9% after TTE (P < 0.01), with the proximal resection margin involved in 16.6% of patients after TEG versus 0 in TTE group (P < 0.01). The 3-year overall survival was comparable (TEG: 36.5%, TTE: 48.4%, P = 0.12). At multivariable analysis, (y)pT category was an independent risk factor for 3-year recurrence. After matching, TEG was still associated with an increased risk of incomplete tumor resection (P = 0.03) and proximal margin involvement (P < 0.01), while there were no differences in post-operative morbidity (P = 0.56) and mortality (P = 0.31). Our data suggest that patients with Siewert type II tumors treated by TEG are exposed to a higher risk of positive proximal resection margin compared to TTE.