Esophagectomy versus endoscopic resection for patients with early-stage esophageal adenocarcinoma: A National Cancer Database propensity-matched study.

Abstract

Endoscopic resection has been rapidly adopted in the treatment of early-stage esophageal tumors. We compared the outcomes after esophagectomy or endoscopic resection for stage T1a adenocarcinoma. We queried the National Cancer Database for patients with T1a esophageal adenocarcinoma who underwent esophagectomy or endoscopic resection and generated a balanced cohort with 735 matched pairs using propensity-score matching. We then performed a multivariable Cox regression analysis on the matched and unmatched cohorts. We identified 2173 patients; 1317 (60.6%) underwent esophagectomy, and 856 (39.4%) underwent endoscopic resection. In the unmatched cohort, patients who underwent esophagectomy were younger, more often not treated in academic settings, and more likely to have comorbidities (30.4% vs 22.5%, P = .002). They had longer hospital stays and more readmissions than patients who underwent endoscopic resection. Factors positively affecting overall survival were younger age, resection at an academic medical center, and lower Charlson-Deyo comorbidity score. In the matched cohort, patients who underwent esophagectomy had longer hospital stays and were more likely to be readmitted within 30 days (7.0% vs 0.6%, P < .001). When a time period-specific partition was applied, endoscopic resection had a lower death hazard 0 to 90 days after resection (hazard ratio, 0.15; P = .003), but this was reversed for survival greater than 90 days (hazard ratio, 1.34; P = .02). In patients with early-stage esophageal adenocarcinoma, survival appears equivalent after endoscopic resection or esophagectomy, but endoscopic resection is associated with shorter hospital stays, fewer readmissions, and less 90-day mortality. In patients surviving more than 90 days, esophagectomy may provide better overall survival.