RESUMO
OBJECTIVES: The frequency, risk factors, and impact on survival of hemorrhage into (peri)pancreatic collections in patients with acute pancreatitis (AP) has not been well studied. The study was designed to evaluate the risk factors for hemorrhage, successful hemostasis and its effect on in-hospital mortality. METHODS: In a prospective cohort study for prediction of severity of AP, the incidence, risk factors, and outcomes of pancreatic hemorrhage were analyzed. Patients with significant hemorrhage were managed according to a predefined protocol including endovascular intervention. RESULTS: Out of 363 patients hospitalized during the study-period, 33(9%) patients developed hemorrhage. Median time from onset of AP to hemorrhage was 59(45-68) days. The cause of hemorrhage was arterial in 19(57.5%) patients and unlocalized in 14(42.5%) patients. Hemorrhage was managed by conservative approach in 7 (21.2%), radiographic angioembolisation in 16 (48.5%), radiographic angioembolisation followed by surgery in 3 (9.1%), and surgery in 7 (21.2%) patients. Persistent organ failure [aHR 2.3 (1.1-5.1), p = 0.03], use of large bore (>20 Fr) catheter for initial drainage [aHR 3.9 (1.7-9.1), p = 0.001] and extensive (>50%) necrosis [aHR 3.1 (1.4-6.9), p = 0.005] were significant risk factors for hemorrhage. Hemorrhage was an independent predictor of mortality [aHR 2.0 (1.2-3.4), p = 0.008] in addition to persistent organ failure (aHR 12.1 (5.7-25.8), p < 0.001). In-hospital mortality in patients with hemorrhage was 22/33 (66.7%) vs. 81/330 (25%) in no hemorrhage group [p <0.001]. CONCLUSION: Pancreatic hemorrhage occurs later in the course of acute pancreatitis in relatively sicker group of patients with organ failure and extensive necrosis, and is independently associated with a higher risk of in-hospital mortality.