Outcomes of 350 Robotic-assisted Esophagectomies at a High-volume Cancer Center: A Contemporary Propensity-score Matched Analysis

Objective To evaluate perioperative and oncologic outcomes in our RAMIE cohort and compare outcomes with contemporary OE controls. Summary of background data RAMIE has emerged as an alternative to traditional open or laparoscopic approaches. Described in all esophagectomy techniques, rapid adoption has been attributed to both enhanced visualization and technical dexterity. Methods We retrospectively reviewed patients who underwent RAMIE for malignancy. Patient characteristics, perioperative outcomes, and survival were evaluated. For perioperative and oncologic outcome comparison, contemporary OE controls were propensity-score matched from NSQIP and NCDB databases. Results We identified 350 patients who underwent RAMIE between 2010 and 2019. Median body mass index was 27.4, 32% demonstrated a Charlson Comorbidity Index ≥4. Nodal disease was identified in 50% of patients and 74% received neoadjuvant chemoradiotherapy. Mean operative time and blood loss were 425 minutes and 232 mL, respectively. Anastomotic leak occurred in 16% of patients, 2% required reoperation. Median LOS was 9 days, and 30-day mortality was 3%. A median of 21 nodes were dissected with 96% achieving an R0 resection. Median survival was 67.4 months. 222 RAMIE were matched 1:1 to the NSQIP OE control. RAMIE demonstrated decreased LOS (9 vs 10 days, P = 0.010) and reoperative rates (2.3 vs 12.2%, P = 0.001), longer operative time (427 vs 311 minutes, P = 0.001), and increased rate of pulmonary embolism (5.4% vs 0.9%, P = 0.007) in comparison to NSQIP cohort. There was no difference in leak rate or mortality. Three hundred forty-three RAMIE were matched to OE cohort from NCDB with no difference in median overall survival (63 vs 53 months; P = 0.130). Conclusion In this largest reported institutional series, we demonstrate that RAMIE can be performed safely with excellent oncologic outcomes and decreased hospital stay when compared to the open approach.