Early recognition of clinically relevant postoperative pancreatic fistula: a systematic review

HPB (Oxford). 2020 Jan;22(1):1-11. doi: 10.1016/j.hpb.2019.07.005. Epub 2019 Aug 21.


Background: Early recognition of postoperative pancreatic fistula might decrease the risk of subsequent life threatening complications. The aim of this review was to systematically evaluate the accuracy of postoperative clinical, biochemical and radiologic variables for early recognition of clinically relevant postoperative pancreatic fistula.

Methods: A systematic literature search was performed up to August 2018. Clinical studies reporting on the association between postoperative variables and clinically relevant postoperative pancreatic fistula were included. Variables were stratified: early prediction (postoperative day 1-2) versus early diagnosis (day 3) and had to be reported in 2 cohorts.

Results: Overall, 37 included studies reported on 17 different diagnostic variables after 8701 pancreatic resections. Clinically relevant postoperative pancreatic fistula occurred in 1532/8701 patients (18%). Early prediction variables included elevated serum and drain amylase (day 1). Identified variables for early diagnosis were: non-serous drain efflux (day 3); positive drain culture (day 3); elevated temperature (any day); elevated C-Reactive Protein (CRP; day 4); elevated white blood cell count (day 4) and peripancreatic collections on computed tomography (CT; day 5-10).

Conclusion: This review provides a comprehensive overview of postoperative variables associated with clinically relevant pancreatic fistula. Incorporation of variables in future algorithms could potentially mitigate the clinical impact of postoperative pancreatic fistula.

Publication types

  • Research Support, Non-U.S. Gov't
  • Systematic Review

MeSH terms

  • Early Diagnosis
  • Humans
  • Pancreatectomy / adverse effects*
  • Pancreatic Fistula / diagnosis*
  • Pancreatic Fistula / etiology
  • Postoperative Complications / diagnosis*
  • Postoperative Complications / etiology
  • Predictive Value of Tests
  • Risk Factors