Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

VA Health Systems Research

Go to the VA ORD website
Go to the QUERI website
2015 Conference Logo



2015 HSR&D/QUERI National Conference Abstract


1116 — Gastrointestinal Bleeding with Dabigatran and Warfarin in Patients with Impaired Renal Function

Kalil R, Iowa City VA Medical Center; Kaboli P, Iowa City VA Medical Center; Vaughan Sarrazin M, Iowa City VA Medical Center;

Objectives:
Studies demonstrate a higher risk of gastrointestinal bleeding (GIB) for patients taking dabigatran for stroke prevention in atrial fibrillation (AF), compared to warfarin. Because of renal clearance of dabigatran, bleeding is of particular concern in patients taking dabigatran with impaired renal function. This study evaluated relative risks of GIB in patients taking warfarin or dabigatran with low, moderate, and high renal function, as measured by the Glomerular Filtration Rate (GFR) ml/min/1.732.

Methods:
Using National VA files, we identified 46,166 VA patients age 65 and older with AF who received warfarin for a minimum of 6 consecutive months during June 2011 through December 2012. Of those, 1,630 subsequently converted to dabigatran use. Each patient who converted to dabigatran was matched to two patients in the same facility who continued to receive warfarin. GIB hospitalizations were identified in VA and Medicare claims and the likelihood of GIB for dabigatran relative to warfarin was examined for patients with low ( < 50), moderate (50 to 80), or high ( > 80) GFR. Analysis censored patients for death or end of follow-up period.

Results:
1,313 patients who switched to dabigatran after a minimum of 6 months on warfarin were matched to 2,601 patients who continued taking warfarin. Overall 11%, 53%, and 36% of matched patients had low, moderate, and high GFR. The rates of GIB per 100 person-years of follow-up were higher for patients with low GFR compared to patients with moderate or high GFR (27, 19, and 12 per 100 patient-years for warfarin, and 37, 29, and 26 for dabigatran). The relative hazard of GIB was 1.45, 1.53, and 2.09 for patients with low, medium, and high GFR. Differences between dabigatran and warfarin within GFR strata were not statistically significant, although this may partly reflect the small sample.

Implications:
While the overall rate of GIB was higher for dabigatran relative to warfarin, the relative hazard within GFR category did not differ significantly for the two drugs.

Impacts:
Health care providers should exercise caution and close monitoring for bleeding complications, even in patients with renal function in the range considered safe for dabigatran use per current guidelines.