Minimally Invasive Surgical Site Infection in Procedure-Targeted ACS NSQIP Pancreaticoduodenectomies

Background: Pancreaticoduodenectomy (PD) incurs a surgical site infection rate of up to 18%. Published rates after minimally invasive PD are comparable or superior to open, but data are limited to high-volume, single-institution series. This study aimed to determine national outcomes. We hypothesized nationwide infections would be reduced with a minimally invasive approach. Materials and methods: Using the newly available pancreatectomy-specific outcomes in National Surgical Quality Improvement Project, data on surgical site infection in PD were extracted from the procedure-targeted participant user file from 2014 to 2015. chi(2) test determined correlation of infection with approach. Linear regression determined correlation of known parameters with infection rate. Results: Overall infection rate was 24%. Compared with open, laparoscopic rates were lower (P = 0.001), but robotic rates were comparable with open. Stenting, longer operative times, and soft gland texture were associated with increased infection rates, whereas larger duct size and drains were associated with decreased rates (all P < 0.01). Conclusions: Laparoscopic PD is associated with decreased surgical site infection on a national level. This represents the first procedure-targeted National Surgical Quality Improvement Project report on this endpoint. Despite greater infection rates than previously reported, these data support previous institutional reports of decreased infection rates with laparoscopic approach. (C) 2018 Elsevier Inc. All rights reserved.