Outcomes and Cost Analysis of Robotic Versus Laparoscopic Abdominoperineal Resection for Rectal Cancer: A Case-Matched Study

摘要

BACKGROUND: Although laparoscopy for abdominoperineal resection has been well defined, the literature lacks comparative studies on robotic abdominoperineal resection. Because robotic abdominoperineal resections typically do not require splenic mobilization or an anastomosis for reconstruction, the mean console time is expected to be shorter than low anterior resection. We hypothesized that robotic and laparoscopic abdominoperineal resection would provide similar oncologic and financial outcomes. OBJECTIVE: The study aimed to compare the perioperative, oncologic, and economic outcomes of the robotic and laparoscopic abdominoperineal resection. DESIGN: This was a retrospective, case-matched patient cohort. SETTINGS: This study was conducted at a tertiary referral center. PATIENTS: This study included all patients who underwent either laparoscopic or robotic abdominoperineal resections between January 2008 and April 2017; they were case-matched in a 1:1 ratio based on age ±5 years, BMI ±3 kg/m 2 , and sex criteria. MAIN OUTCOME MEASURES: Perioperative, oncologic, and economic (including survival) outcomes were compared. Because of institutional policy, actual cost values are presented as the lowest direct cost value as "100%," and other values are presented as proportional to the index value. RESULTS: We examined 68 patients (34 in each group). Both groups had similar preoperative characteristics, including preoperative chemoradiation rates. Operative time (319 vs 309 min), length of stay (7.2 vs 7.4 d), postoperative complications (38.2% vs 41.2%), conversion to open (5 vs 4), complete mesorectal excision (76.4% vs 79.4%), radial margin involvement (2.9% vs 8.9%), and direct hospital cost parameters (mean difference 26%, median difference 43%) were comparable between robotic and laparoscopic abdominoperineal resection groups, respectively (all p > 0.05). Local recurrence, disease-free survival, and overall survival rates (85.3% vs 76.5%) were also similar after 22 months of follow-up between the groups. LIMITATIONS: The main limitations of this study are its retrospective nature and the variety in concomitant procedures. CONCLUSIONS: Robotic abdominoperineal resections provided in carefully matched patients with rectal cancer showed similar perioperative and short-term oncologic outcomes compared to laparoscopic abdominoperineal resections. Our study was not powered to detect a significant increase in cost with robotic abdominoperineal resections. See Video Abstract at http://links.lww.com/DCR/B920 . RESULTADOS Y ANLISIS DE COSTO DE LA RESECCIN ABDOMINOPERINEAL LAPAROSCPICA VS LA ROBTICA EN CASOS DE CNCER DE RECTO ESTUDIO DE CASOS EMPAREJADOS: ANTECEDENTES:Si bien la resección abdominoperineal laparoscópica está bien definida, la literatura carece de estudios comparativos sobre la resección abdominoperineal robótica. Dado que las resecciones abdominoperineales robóticas generalmente no requieren movilización esplénica o una anastomosis en casos de reconstrucción, se supone que el tiempo medio en la consola sea más corto que durante una resección anterior baja. Hipotéticamente las resecciones abdominoperineales robóticas y laparoscópicas nos proporcionarían resultados oncológicos y económicos similares.OBJETIVO:Comparar los resultados perioperatorios, oncológicos y económicos de la resección abdominoperineal robótica y laparoscópica.DISEÑO:Esta fue una cohorte de pacientes retrospectiva, emparejada por casos.AJUSTE:Estudio realizado en un centro de referencia terciario.PACIENTES:Todos los pacientes que se sometieron a resecciones abdominoperineales LAParoscópicas o ROBóticas entre Enero de 2008 y Abril de 2017 fueron identificados y emparejados según la edad ±5, el IMC ±3 y los criterios de sexo en una proporción de 1:1.PRINCIPALES MEDIDAS DE RESULTADO:Se compararon los resultados perioperatorios, oncológicos y económicos (incluida la sobrevida). Debido a la política institucional, los valores de costos reales se presentan como e