Perioperative outcome after pancreatic head resection: a 10-year series of a specialized surgeon in a university hospital and a community hospital

Abstract

Introduction: Hospital and surgeon volume are potential factors influencing postoperative mortality and morbidity after pancreatic resection. Data on perioperative outcome of individual surgeons in different institutions, however, are scarce. We evaluated the perioperative outcome after pancreatic head resections (PHR) performed by a high-volume pancreatic surgeon in a high-volume university department and (later) in a community hospital with low prior experience in major pancreatic surgery.

Methods: We compared the results after PHR were performed by a single experienced surgeon between 2001 and October 2006 in a specialized unit of a German university hospital (n = 83; group A) with the results after PHR were performed in a community hospital between November 2006 and 2011 (n = 145; group B). Before the study period (-2001), the surgeon already had a personal caseload of >200 PHR. In addition to the 228 PHR analyzed here, the surgeon also had taught further >150 PHR to residents and consulting surgeons. Comparable surgical and perioperative techniques were applied in both series (e.g., types of resection and reconstruction, abdominal drains, early enteral feeding). The data of both series were prospectively recorded in SPSS databases.

Results: The median age of the patients was lower in group A (58 vs. 66 years in B; p < 0.01). Indications for PHR were pancreatic cancer (A 39 % vs. B 45 %), other periampullary cancer (A 18 % vs. B 12 %), chronic pancreatitis (A 33 % vs. B 28 %), and others (A 10 % vs. B 15 %). Most PHR were pylorus preserving (64 vs. 75 %), with oncologically indicated portal vein resections in 24 % (A) or 33 % (B). The percentage of duodenum-preserving PHR was lower in group B (14 vs. 26 % in A). Mortality of PHR was 3.6 % in group A and 2.8 % in B (p = 0.72). Overall morbidity rate was 49 % (A) or 57 % (B; p = 0.25). Using the expanded Accordion classification, complications classified as grade 4 or higher occurred in 9 % (A) and 11 % (B; p = 0.74). Postoperative pancreatic leak (any grade) was documented in 26 % (A) and 25 % (B; p = 0.87).

Conclusions: Surgeon volume and a high individual experience, respectively, contribute to acceptable complication rates and low mortality rates after pancreatic head resection. An experienced surgeon can provide a good perioperative outcome after pancreatic resection even after a change of hospital or medical staff.