Superselective embolization for the management of postpancreatectomy hemorrhage: a single-center experience in 25 patients.

Purpose: To evaluate the efficacy of superselective embolization in patients with late postpancreatectomy hemorrhage (PPH).

Materials And Methods: Between January 2002 and July 2010, 25 patients (19 men) with suspected late PPH (> 24 hours after the operation) were evaluated. The primary study endpoint was technical success, defined as complete angiographic occlusion of the site of hemorrhage. Secondary study endpoints were multidetector computed tomography (CT) and angiographic findings regarding accurate detection of the site of hemorrhage, persistence of hemorrhage, or occurrence of rebleeding during clinical follow-up.

Results: Multidetector CT was performed before the intervention in 17 (68%) patients with detection of hemorrhage in 15 (88%) patients. The site of hemorrhage was detected in 23 (92%) of 25 patients by angiography. Four (17%) patients required a superselective catheter position. Embolization was attempted in all 23 patients with angiographically visible hemorrhage. In three (13%) patients, embolization could not be performed because a superselective catheter position was not achievable. Technical success was 83% (19 patients). In one patient, hemostasis was not achieved by embolization. Minor complications occurred in three (13%) patients. No major complications occurred. Three patients with technically successful embolization (16%) had a second episode of bleeding during follow-up and required repeat embolization 5-23 days after the procedure. The 30-day mortality rate was 20%.

Conclusions: Superselective embolization is a technically and clinically effective procedure in patients with late PPH. Diagnostic angiography should be performed with a superselective microcatheter position to detect the bleeding site effectively.