Abstract
PURPOSE: Drains' role after pancreaticoduodenectomy (PD) is debated by proponents of no drain, draining selected cases, and early drain removal. The aim of the study was to assess the effect of "standard" and "draining-tract-targeted" management of abdominal drains still in situ after diagnosing a postoperative pancreatic fistula (POPF). METHODS: PubMed and Scopus were searched for "pancreaticoduodenectomy or pancreatoduodenectomy or duodenopancreatectomy," "Whipple," "proximal pancreatectomy," "pylorus-preserving pancreatectomy," and "postoperative pancreatic fistula or POPF.". Main outcomes included clinically relevant (CR) POPF, grade-C POPF, overall mortality, POPF-related mortality, and CR-POPF-related mortality. Secondary outcomes were incidence of radiological and/or endoscopic interventions, reoperations, and completion pancreatectomies. RESULTS: Overall, 12,089 studies were retrieved by the search of the English literature (01/01/1990-31/12/2018). Three hundred and twenty-six studies (90,321 patients) reporting ≥ 100 PDs and ≥ 10 PD/year were finally included into the study. Average incidences were obtained by averaging the incidence rates reported in the single articles. Pooled incidences were calculated by combining the number of events and the total number of patients considered in the various studies. These were then meta-analyzed using DerSimonian and Laird's (1986) method. Pearson's chi-squared test was used to compare pooled incidences between groups. Post hoc testing was used to see which groups differed. The meta-analyzed incidences were compared using a fixed effect for moderators. "Draining-tract-targeted" management showed a significant advantage over "standard" management in four clinically relevant outcomes out of eight according to pool analysis and in one of them according to meta-analysis. CONCLUSION: Clinically, "draining-targeted" management of POPF should be preferred to "standard" management.