Comparison of Perioperative Outcomes between Open, Laparoscopic, and Robotic Distal Pancreatectomy: an Analysis of 1815 Patients from the ACS-NSQIP Procedure-Targeted Pancreatectomy Database

Robotic surgery is gaining acceptance for distal pancreatectomy (DP). Nevertheless, no multi-institutional data exist to demonstrate the ideal clinical circumstances for use and the efficacy of the robot compared to the open or laparoscopic techniques, in terms of perioperative outcomes. The 2014 ACS-NSQIP procedure-targeted pancreatectomy data for patients undergoing DP were analyzed. Demographics and clinicopathological and perioperative variables were compared between the three approaches. Univariate and multivariable analyses were used to evaluate outcomes. One thousand eight hundred fifteen DPs comprised 921 open distal pancreatectomies (ODPs), 694 laparoscopic distal pancreatectomies (LDPs), and 200 robotic distal pancreatectomies (RDPs). The three groups were comparable with respect to demographics, ASA score, relevant comorbidities, and malignant histology subtype. Compared to the ODP group, patients undergoing RDP had lower T-stages of disease (P = 0.0192), longer operations (P = 0.0030), shorter hospital stays (P < 0.0001), and lower postoperative 30-day morbidity (P = 0.0476). Compared to the LDP group, RDPs were longer operations (P < 0.0001) but required fewer concomitant vascular resections (P = 0.0487) and conversions to open surgery (P = 0.0068). On multivariable analysis, neoadjuvant therapy (P = 0.0236), malignant histology (P = 0.0124), pancreatic reconstruction (P = 0.0006), and vascular resection (P = 0.0008) were the strongest predictors of performing an ODP. The open, laparoscopic, and robotic approaches to distal pancreatectomy offer particular advantages for well-selected patients and specific clinicopathological contexts; therefore, clearly demonstrating the most suitable use and superiority of one technique over another remains challenging.