Sungho Kim, MD, Kil Hwan Kim, MD, In Gun Hyun, MD, Yoo-Seok Yoon, MD, PhD, Jai Young Cho, MD, PhD, YoungRok Choi, MD, PhD. Seoul National University Bundang Hospital
Introduction: Recent advancements in minimally invasive techniques led to increased effort and interest in laparoscopic pancreatic surgery. Laparoscopic distal pancreatectomy is a widely accepted procedure for left-sided pancreatic lesions. In other cases, the adoption of laparoscopic pancreaticoduodenectomy has been hindered by the technical complexity of laparoscopic reconstruction. Hybrid laparoscopy-assisted pancreaticoduodenectomy (HLAPD) in which pancreaticoduodenal resection is performed laparoscopically, while reconstruction is completed via a small upper midline minilaparotomy, is combines the efficacy of open approach, and the benefits of laparoscopic approach. The purpose of this study is to report our experience of HLAPD and to define the learning curves.
Methods: 90 patients with benign and malignant periampullary lesion underwent HLAPD by a single surgeon between July 2007 and May 2017 were retrospectively reviewed. The clinicopathologic variables were prospectively collected and analyzed. The learning curve for HLAPD was assessed using cumulative sum (CUSUM) and risk-adjusted CUSUM (RA-CUSUM) methods.
Results: The most common histopathology was pancreatic ductal adenocarcinoma (n=27, 27.8%), followed by intraductal papillary mucinous neoplasms (n=16, 16.5%), ampulla of Vater cancer (n=16, 16.5%), and common bile duct cancer (n=15, 15.5%). The median operation time was 540 min (range, 300 – 865 min) and the median estimated blood loss was 550 ml. The mean hospital stay was 21.4 days. Complications developed in 25 patients (27.7%). Among them, 13 patients (14.4%) had significant pancreatic fistula [International Study Group on Pancreatic Fistula grade B and C]. Based on the CUSUM and the RA-CUSUM analyses, the learning curve for HLAPD was grouped into four phases: phase I was the initial learning period (cases 1 – 10), phase II was the technical stabilizing period (cases 11 – 37), phase III was the second learning period (cases 38 – 70) and phase IV represented the second stabilizing period (cases 71 – 90). There was a statistical difference in terms of surgical indication between phase II and III (p=0.002).
Conclusions: HLAPD is a technically feasible and safe procedure in selected patients. This procedure has benefits of both open and minimally invasive procedure, and could be a stepping-stone for transition from open to purely minimally invasive pancreaticoduodenectomy.
Presented at the SAGES 2017 Annual Meeting in Houston, TX.
Abstract ID: 87304
Program Number: P521
Presentation Session: iPoster Session (Non CME)
Presentation Type: Poster