Increased Blood Transfusion Requirements with Pharmacologic Venous Thromboembolism Prophylaxis During IBD Exacerbation: 2016 ACG Category Award (IBD)

Abstract

Introduction: In-patients with an inflammatory bowel disease (IBD) exacerbation are at increased risk for venous thromboembolism (VTE). Guidelines recommend pharmacologic VTE prophylaxis (PVTEP) with anticoagulants. Guideline compliance is compromised by fear that PVTEP may worsen bleeding or anemia, and by limited observational data. Our goal was to define the bleeding risks of PVTEP of IBD in-patients. Methods: We conducted a retrospective chart review of 4 hospitals of the Northshore-LIJ Health System from 01/2011 and 06/2015. Those admitted with IBD as their primary diagnosis were included, as were those with bleeding, abdominal pain, or diarrhea as the primary diagnosis and IBD as a secondary diagnosis. Demographic data, IBD type Crohn's disease (CD) or Ulcerative Colitis (UC), admission hemoglobin (Hgb), and use of anti-platelet therapy were recorded. Inpatient use of warfarin, enoxaparin, heparin, or xarelto qualified as PVTEP. Severe disease activity was defined by any inpatient use of corticosteroid or biologic therapy. Our primary outcome was packed red blood cell (PRBC) transfusion events, used as a surrogate marker for bleeding. Separate generalized linear mixed models for binary clustered data were used to examine the probability of transfusion related to PVTEP. Results: 717 patients, (49.5% UC, 53.4% female), met inclusion criteria totaling 891 admissions. Overall 60% of admissions involved PVTEP, and 46.5% of admissions met criteria for severe disease activity. The unadjusted risk of any PRBC transfusion with severe disease was greater with PVTEP vs. no PVTEP, 18.6% vs. 11.2%, OR 1.82 (95% CI 1.04-3.17), but not for non-severe disease, 7.4% vs. 9.42%, OR 0.77 (95% CI 0.38-1.54).In the multivariable model, the adjusted risk of transfusion was increased by PVTEP, OR 2.80 (95% CI 1.61-4.88, p < 0.0003), severe disease activity, OR 2.63 (95% CI 1.57-4.41, p < 0.0002), decreased admission Hgb, OR 1.96 (95% CI 1.72-2.27, p < 0.0001), and anti-platelet therapy OR 2.36 (95% CI 1.23-4.52, p < 0.0098) but not disease type (CD vs UC), OR 0.78 (95%CI 0.46-1.30). Conclusion: In a large inpatient IBD population PVTEP increased transfusion requirement, especially among those with severe IBD activity. While bleeding and transfusion are not desirable outcomes, the aim of PVTEP is to prevent the morbidity and mortality of VTE in this high risk population. Our data provides an important guide to inform patient management, as is available for other therapies used for active IBD.