Restarting Anticoagulation and Outcomes After Major Gastrointestinal Bleeding in AFib
In patients with atrial fibrillation (AF) on warfarin who develop gastrointestinal bleeding (GIB), there is limited evidence to guide the decision to restart anticoagulation. Prior retrospective data has suggested that patients in whom warfarin is not restarted suffer increased risks of thrombosis and death.1, 2 This study by Qureshi et al. was designed to investigate the hypothesis that the benefits of restarting anticoagulation in these patients outweigh the risks.3
In this retrospective cohort study, the investigators reviewed data on patients with AF on warfarin at the Henry Ford Health System from 2005-2010 who developed major GIB, defined as a decrease in hemoglobin by 2 grams or by visible or endoscopically confirmed bleeding, followed by resolution of GIB, defined by <1 gram decrease in hemoglobin for 48 hours. The associations between restarting warfarin and recurrent GIB, atrial thromboembolism, and mortality were explored using time-to-event adjusted analyses. The analyses were then stratified by the duration of warfarin interruption using groups of less than seven days, seven to 30 days, and 30-60 days.
There were 1,329 patients with AF on warfarin who developed major GIB. The 653 patients (49.1%) who were restarted on warfarin did not experience an overall increase in recurrent GIB (hazard ratio [HR] 1.18, 95% confidence interval [CI] 0.94-1.10, p=0.47). However, restarting warfarin was associated with decreased thromboembolism (HR 0.71, 95% CI 0.54-0.93, p=0.01) and reduced mortality (HR 0.67, 95% CI 0.56-0.81, p <0.0001). When stratified by the duration of warfarin interruption, when compared to restarting warfarin between 30 and 60 days after GIB, restarting within seven days was associated with an increase in recurrent GIB (HR 3.27, 95% CI 1.82-5.91, p=0.002), a non-significant decrease in thromboembolism (HR 0.76, 95% CI 0.37-1.59; p=0.47), and an overall decrease in mortality (HR 0.56, 95% CI 0.33-0.93; p=0.03). Restarting warfarin between seven and 30 days of GIB was associated with significant decreases in thromboembolism and mortality without a significant increase in recurrent GIB.
In patients with AF on warfarin who develop GIB, restarting warfarin was associated with an overall decreased risk of thromboembolism and mortality without a significant increased risk of recurrent GIB. Among those in whom anticoagulation was resumed, restarting warfarin between seven and 30 days of GIB was associated with a decreased risk of thromboembolism and mortality without a significant increased risk of recurrent GIB when compared to restarting warfarin after 30 days.
This study by Qureshi et al. was designed to investigate whether patients with AF on warfarin benefit from the early reintroduction of anticoagulation after GIB. The investigators concluded that restarting warfarin was associated with decreases in thromboembolism and mortality without a significant increase in recurrent GIB when compared to restarting after 60 days or not at all.
The authors identified and discussed the study’s limitations, including its retrospective design, relatively small study population within one hospital system, and potential for detection and survivorship biases. Importantly, the patients restarted on warfarin had fewer comorbidities, which may have contributed to their improved outcomes. Given the difficulty discerning exactly why clinicians opted against restarting warfarin, with the most common reasons being perceived patient inability to follow up with the anticoagulation clinic (19%) and physician preference (18%), there may also have been other unidentified patient factors that caused clinicians to predict a higher risk of recurrent GIB in those patients, thus further contributing to the improved outcomes associated with those who were restarted on warfarin.
In patients with AF on warfarin who develop GIB, the existing data has been insufficient for clinicians to make evidenced-based decisions regarding if and when to restart anticoagulation. Despite this study’s limitations, it provides novel insight and should prompt further investigation into this challenging patient population. In the meantime, since a GIB generally carries a better prognosis than stroke, this study moves the pendulum in the direction of resuming anticoagulation in most patients within the first month after a GIB.
- Witt DM, Delate T, Garcia DA, Clark NP, Hylek EM, Ageno W, Dentali F and Crowther MA. Risk of thromboembolism, recurrent hemorrhage, and death after warfarin therapy interruption for gastrointestinal tract bleeding. Arch Intern Med 2012;172:1484-91.
- Lee JK, Kang HW, Kim SG, Kim JS and Jung HC. Risks related with withholding and resuming anticoagulation in patients with non-variceal upper gastrointestinal bleeding while on warfarin therapy. Int J Clin Pract 2012;66:64-8.
- Qureshi W, Mittal C, Patsias I, et al. Restarting Anticoagulation and Outcomes After Major Gastrointestinal Bleeding in Atrial Fibrillation. Am J Cardiol2014;113:662-8.
< Back to Listings