Establishing an open and robotic pancreatic surgery program in a level 1 trauma center community teaching hospital and comparing its outcomes to high-volume academic center outcomes: a retrospective review

Abstract

Background The debate of whether to centralize hepato-pancreato-biliary surgery has been ongoing. The principal objective was to compare outcomes of a community pancreatic surgical program with those of high-volume academic centers. Methods The current pancreatic surgical study occurred in an environment where (1) a certified abdominal transplant surgeon performed all surgeries; (2) complementary quality enhancement programs had been developed; (3) the hospital’s trauma center had been verified; and (4) the hospital’s surgical training had been accredited. Pancreatic surgical outcomes at high-volume academic centers were obtained through PubMed literature searches. Articles were selected if they described diverse surgical procedures. Two-tailed Fisher exact and mid-P tests were used to perform 2 × 2 contingency analyses. Results The study patients consisted of 64 consecutive pancreatic surgical patients. The study patients had a similar pancreaticoduodenectomy proportion (59.4%) when compared to literature patients (66.8%; P = 0.227). The study patients also had a similar distal pancreatectomy proportion (25.0%) when compared to literature patients (31.9%; P = 0.276). The study patients had a significantly higher American Society of Anesthesiologists physical status ≥ 3 proportion (100%) than literature patients (28.1%; P < 0.001). The 90-day study mortality proportion (0%) was similar to the literature proportion (2.3%; P = 0.397). The study postoperative pancreatic fistula proportion was lower (3.2%), when compared to the literature proportion (18.4%; P < 0.001; risk ratio = 5.8). The study patients had a lower reoperation proportion (3.1%) than the literature proportion (8.7%; mid-P = 0.051; risk ratio = 2.8). The study patients had a lower surgical site infection proportion (3.1%) than those in the literature (21.1%; P < 0.001; risk ratio = 6.8). The study patients had equivalent delayed gastric emptying (15.6%) when compared to literature patients (10.6%; P = 0.216). The study patients had decreased Clavien–Dindo grades III–IV complications (10.9%) compared to the literature patients (21.8%; mid-P = 0.018). Lastly, the study patients had a similar readmission proportion (20.3%) compared to literature patients (18.4%; P = 0.732). Conclusion Despite pancreatic surgical patients having greater preoperative medical comorbidities, the current community study outcomes were comparable to or better than high-volume academic center results.