Postpancreatectomy Hemorrhage—Incidence, Treatment, and Risk Factors in Over 1,000 Pancreatic Resections

• Published in: Journal of gastrointestinal surgery 2014-03, Vol.18 (3), p.464-475
• Main Authors: Wellner, U. F., Kulemann, B., Lapshyn, H., Hoeppner, J., Sick, O., Makowiec, F., Bausch, D., Hopt, Ulrich Theodor, Keck, T.
• Format: Article
• Language: English
• Subjects: 2013 SSAT Poster Presentation; Adolescent; Adult; Age Factors; Aged; Aged, 80 and over; Angiography; Blood Transfusion; Body Mass Index; Child; Clinical Competence; Female; Fistula; Gastroenterology; Hemorrhage; Hemostasis, Endoscopic; Humans; Incidence; Male; Medicine; Medicine & Public Health; Middle Aged; Mortality; Pancreatectomy - adverse effects; Pancreatectomy - mortality; Pancreatic Diseases - pathology; Pancreatic Diseases - surgery; Pancreatic Fistula - etiology; Pancreatic Fistula - mortality; Pancreaticoduodenectomy; Pancreaticoduodenectomy - adverse effects; Pancreaticoduodenectomy - mortality; Pancreaticojejunostomy - adverse effects; Pancreaticojejunostomy - mortality; Patients; Postoperative Hemorrhage - epidemiology; Postoperative Hemorrhage - etiology; Postoperative Hemorrhage - mortality; Postoperative Hemorrhage - therapy; Regression analysis; Reoperation; Retrospective Studies; Risk Factors; Severity of Illness Index; Sex Factors; Small intestine; Surgery; Veins & arteries; Young Adult
• ISSN: 1091-255X; 1873-4626
• DOI: 10.1007/s11605-013-2437-5

Background Postpancreatectomy hemorrhage is a rare but often severe complication after pancreatic resection. The aim of this retrospective study was to define incidence and risk factors of postpancreatectomy hemorrhage and to evaluate treatment options and outcome. Patients and Methods Clinical data was extracted from a prospectively maintained database. Descriptive statistics, univariate and multivariate risk factor analysis by binary logistic regression were performed with SPSS software at a significance level of p  = 0.05. Results N  = 1,082 patients with pancreatic resections between 1994 and 2012 were included. Interventional angiography was successful in about half of extraluminal bleeding. A total of 78 patients (7.2 %) had postpancreatectomy hemorrhage (PPH), and 29 (2.7 %) were grade C PPH. Multivariate modeling disclosed a learning effect, age, BMI, male sex, intraoperative transfusion, portal venous and multivisceral resection, pancreatic fistula and preoperative biliary drainage as independent predictors of severe postpancreatectomy hemorrhage. High-risk histopathology, age, transfusion, pancreatic fistula, postpancreatectomy hemorrhage and pancreatojejunostomy in pancreatoduodenectomies were independent predictors of mortality. Conclusions Our study identifies clinically relevant risk factors for postpancreatectomy hemorrhage and mortality. Interventional treatment of extraluminal hemorrhage is successful in about half of the cases and if unsuccessful constitutes a valuable adjunct to operative hemostasis. Based on our observations, we propose a treatment scheme for PPH. Risk factor analysis suggests appropriate patient selection especially for extended resections and pancreatogastrostomy for reconstruction in pancreatoduodenectomy.