Chemoprophylaxis to Prevent Deep Venous Thrombosis in Patients Hospitalized for Pancreatitis: Beneficial or Harmful?

Vascular complications in pancreatitis generally occur in the form of hemorrhage or thrombosis. Pancreatitis resulting in splanchnic thrombosis has been well studied, but the cause of this correlation has not been studied in the current era of increasing anticoagulant use for deep venous thrombosis (DVT) prophylaxis. Hemorrhagic pancreatitis and peri-pancreatic bleeding are also known phenomena encountered in relation to pancreatitis, but these risks are not well established in the setting of chemical prophylaxis for DVT.

Our objective was to identify whether chemical DVT prophylaxis in pancreatitis harms the patient by increasing the risk of hemorrhagic conversion of pancreatitis or peri-pancreatic hemorrhage or if it is beneficial by preventing splanchnic venous thrombosis in the abdominal vasculature that surrounds the pancreas.

We undertook a retrospective chart review with approval from the Institutional Review Board on patients who were hospitalized for or developed pancreatitis during their hospital stay from April 2014 to July 2015. We reviewed the charts for imaging suggestive of venous thrombosis or the development of intra-abdominal hemorrhage at admission during hospitalization and within 30 days after hospitalization. We also reviewed the methods of DVT prophylaxis to identify any correlation with the risk of hemorrhage or thrombosis. A bedside index of severity in acute pancreatitis score was used within 24 hours of admission to calculate the severity of the patients' pancreatitis. The data collected were analyzed for descriptive statistics, correlation using Pearson's coefficient, and multivariate regression analysis using Microsoft Excel and SPSS Inc. Released 2009. PASW Statistics for Windows, Version 18.0. Chicago: SPSS Inc.

This study included 389 patients who met the inclusion criteria. Of these, 74.6% of patients received chemical prophylaxis, mostly low molecular weight heparin, and 18.5% of patients were not on chemical or mechanical means of DVT prophylaxis. Only 12 patients (3%) had complications related to thrombosis and hemorrhage. Seven patients had splanchnic venous thrombosis, one had a hemorrhagic conversion of pancreatitis, three had a peri-pancreatic hemorrhage, and one had both the hemorrhagic conversion of pancreatitis and peri-pancreatic hemorrhage. Ten patients out of 12 patients had complications before admission, and nine of the 12 patients were on chemical prophylaxis. Pearson's coefficient showed no statistically significant correlation between the incidence of complications and the use of chemical DVT prophylaxis. Multivariate analysis showed no specific variable that increased the risk of complications.

Our study showed that chemoprophylaxis for DVT in patients hospitalized for acute pancreatitis is neither harmful by causing hemorrhagic conversion of pancreatitis, peri-pancreatic hemorrhage nor beneficial by preventing splanchnic venous thrombosis.