Resumption of Antithrombotic Therapy after Gastrointestinal Bleeding in Patients with Atrial Fibrillation

EBM Focus - Volume 10, Issue 48

Read the full EBM Focus and earn CME credit.

Reference - BMJ 2015 Nov 16;351:h5876 (level 2 [mid-level] evidence)

  • Patients with atrial fibrillation experiencing a gastrointestinal bleed while on antithrombotic therapy and their care providers face 2 challenging clinical decisions: whether to restart antithrombotic therapy and which therapy to choose.
  • This study found that resumption of a single oral anticoagulant, single antiplatelet agent, or dual anticoagulant/antiplatelet therapy was associated with reduced risk of all-cause mortality and thromboembolism compared to not restarting antithrombotic therapy.
  • In addition, there were no significant differences in the risk of major bleeding with single antiplatelet therapy or either dual therapy regimen compared to no therapy but oral anticoagulation therapy was associated with a small increased risk of major bleeding.

In patients with atrial fibrillation experiencing a gastrointestinal bleed while on antithrombotic therapy, the decision of whether or not to resume antithrombotic therapy can be difficult. The risk of thromboembolism needs to be weighed against the risk of recurrent bleeding to determine if thromboembolic prophylaxis should be continued and if continued, what antithrombotic therapy to use (oral anticoagulation, antiplatelet agent, dual therapy, or triple therapy). Evidence to help guide decision making is limited, however. A recent retrospective cohort study evaluated 3,409 patients ≥ 30 years old (mean age 78 years) with atrial fibrillation who were hospitalized for a first gastrointestinal bleed while on single or combined antithrombotic therapy. Patients were followed for up to 5 years (median follow-up 2 years) and follow-up began 90 days after hospital discharge. Patients with thromboembolic events, major bleeding, recurrent gastrointestinal bleed, or who died within 90 days of hospital discharge were excluded from the study. Major bleeding was defined as intracranial bleeding or severe respiratory, gastrointestinal, or urinary tract bleeding. Baseline comorbidities included ischemic heart disease in 38%, heart failure in 30.7%, and stroke or thromboembolism in 22.5%.

By day 90 post-discharge, 72.9% of patients had resumed antithrombotic therapy including: single therapy with oral anticoagulation in 21.3% or antiplatelet agent in 38.5% and dual therapy with oral anticoagulation plus antiplatelet agent in 11.3% or aspirin plus adenosine diphosphate receptor antagonist in 1.5%. At 2 years, the cumulative incidence of all-cause mortality was 39.9%, thromboembolism was 12%, major bleeding was 17.7%, and recurrent gastrointestinal bleeding was 12.1%. Resumption of a single oral anticoagulant, single antiplatelet agent, or dual anticoagulant/antiplatelet therapy were all associated with reduced risk of all-cause mortality and thromboembolism compared to not restarting antithrombotic therapy (see table below), but there were no significant differences in either outcome with dual aspirin plus adenosine diphosphate receptor antagonist therapy. Also, while oral anticoagulation therapy was associated with a small increased risk of major bleeding compared to no therapy, there were no significant differences in the risk of major bleeding with single antiplatelet therapy or either dual therapy regimen. There were also no significant differences in recurrent gastrointestinal bleed with any antithrombotic regimen.

Outcomes Compared to no Antithrombotic Therapy

The results of this study are consistent with a smaller retrospective cohort study finding that patients who restarted warfarin had a reduced risk of thromboembolism or death, but no significant differences in recurrent gastrointestinal bleeding compared to patients not restarting therapy (Am J Cardiol 2014 Feb 15;113(4):662). In the current study, single therapy with antiplatelet agents decreased mortality compared to not restarting therapy, but this regimen was not as effective as therapy with an oral anticoagulant. Dual therapy with oral anticoagulation plus an antiplatelet agent was also no more effective than oral anticoagulation alone, further suggesting that antiplatelet agents may not be effective for reducing mortality or the risk of thromboembolism. These results were consistent across subgroup analyses of patients also taking proton pump inhibitors and in sensitivity analyses using 30 days after hospital discharge as the start of follow-up. However, the effects of novel oral anticoagulants (NOACs) could not be determined due to the low number of patients taking these medications at the time of the study. Overall, this study suggests that resuming antithrombotic therapy with a single oral anticoagulant in patients with atrial fibrillation and a serious gastrointestinal bleed results in better overall mortality outcomes despite potentially increasing the risk of major bleeding.

For more information, see the Thromboembolic prophylaxis in atrial fibrillation in DynaMed.


Other EBSCO Sites +