Author(s): Nagesh Nayakarahalli Swamigowda, Shilpa Mariappa Casaba, Ashok Kumar K. V, Aravinda Kotresh Nirmala, Shreedevi Kshetrapaliah Nagadas

Peripancreatic haemorrhage though rare, has lethal outcomes. This rare scenario is a consequence of pancreatic diseases per se as well as surgical complication of pancreatobiliary surgery. The aim of this study is to present the retrospective analysis of our experience on the common aetiologies, clinical presentations and various approaches for the management of Peripancreatic haemorrhage.
We present our retrospectively analysed data from prospectively maintained database during the period 2014 -2018. In the 4 year period, 29 patients were diagnosed with Peripancreatic haemorrhage. 25 of these 29 patients were diagnosed with pancreatitis being the cause. 2 patients in this pancreatitis group presented with Peripancreatic haemorrhage during the post-operative course of pancreatic necrosectomy. 3 patients presented with early Grade C post-pancreatoduodenectomy haemorrhage, 1 patient presented with peripancreatic haemorrhage 1 month following pancreatoduodenectomy procedure. Peripancreatic haemostasis was attempted with endovascular angioembolisation in 26 cases, 3 cases underwent surgical haemostasis and 1 patient was attempted surgical haemostasis after failed endovascular haemostasis. The decision whether the patient first undergoes endovascular or surgical haemostasis was based on the patient’s clinical condition and on the availability of interventional radiologist during the emergency. The outcomes measured were clinical successful haemostasis and post interventional complications.
Conventional CT Angiogram could localise the site of bleed in 23 of 26 (88.46%) cases. Endovascular angiography was successful in localising the site of bleed in 24 of 26 (92.31%) cases. Primary endovascular angioembolisation was successful in controlling the Peripancreatic bleed in 22 cases (91.66%). Repeat endovascular angiography and blind angioembolisation was done in 2 cases based on the CT and endoscopic assessment of the possible site of bleed of which it was successful in both. In 1 patient, there was failure to cannulate the splenic artery. 1 patient suffered superior mesenteric arterial dissection and 1 patients suffered right groin hematoma at the femoral artery puncture site. All 3 early post – pancreatoduodenectomy haemorrhage patients underwent emergency re-exploration and surgical haemostasis was attained in 2 cases. Identified source of Peripancreatic haemorrhage include gastroduodenal artery in (7), splenic artery (15), inferior pancreatoduodenal artery in (3), superior mesenteric artery branches (2) and unnamed tuft of vessels in peri pancreatic region (2). Clinical success with endovascular angioembolisation alone (92.31 %) and attempted surgical haemostasis (50%). Overall mortality was 6.89% (2/29).
Peripancreatic haemorrhage though uncommon is known for high morbidity and mortality. The goal of management after initial resuscitation in a hemodynamically stable patients should be conventional angiography and angioembolisation. Surgical haemostasis is not to be ignored for its complimentary role and should be reserved for haemodynamically unstable patients and those who fail endovascular haemostasis.