Esophagectomy using a thoracoscopic approach with an open laparotomic or hand-assisted laparoscopic abdominal stage for esophageal cancer: analysis of survival and prognostic factors in 315 patients.

UNLABELLED: Survival and prognostic factors were analyzed in 315 patients with esophageal cancer undergoing thoracoscopic-assisted esophagectomy (TAE). The 5-year survival rate of 57.8% was satisfactory, indicating the oncological feasibility of TAE. Perioperative outcomes affected overall survival in the whole cohort but not in the subgroup treated with 2 endoscopic stages. OBJECTIVE: To estimate the oncological feasibility of thoracoscopic-assisted esophagectomy (TAE) for esophageal cancer and to clarify the prognostic impact of perioperative factors after TAE. BACKGROUND: Favorable perioperative outcomes of TAE versus open surgery have been demonstrated. However, survival data after TAE in a large cohort are limited, and no information on the prognostic influence of perioperative factors after TAE is available. METHODS: Prospectively collected data for 315 patients undergoing TAE for esophageal cancer were analyzed. Survival was compared with the Kaplan-Meier analysis and Cox regression analysis between 2 surgical approaches: thoracoscopic and hand-assisted laparoscopic esophagectomy (THLE) and thoracoscopic and open laparotomic esophagectomy (TOE). Factors affecting overall survival were identified with Cox multivariate regression analysis in the whole cohort and the THLE subgroup. RESULTS: THLE and TOE were performed in 153 and 162 patients, respectively. The overall 5-year survival of the whole cohort was 57.8%, with no difference between the THLE and the TOE group. Multivariate analysis of the 315 patients identified the following prognostic factors: blood loss, blood transfusion, intensive care unit stay, cardiovascular complications, pathological T and N stages, lymphatic invasion, intramural metastasis, and number of metastatic nodes. In the THLE subgroup, cerebral comorbidity, histological subtype, pathological T stage, and number of metastatic nodes were independent prognostic factors. CONCLUSIONS: TAE was oncologically feasible. Perioperative factors affected survival in the whole cohort, but did not in the THLE subgroup. However, the reduced perioperative factor effect in this subgroup would be small because the survival rates of the 2 surgical approaches were equal.