Real-World Studies Highlight GI Bleeding Concerns With Newer Anticoagulants

Two retrospective, population-based studies published online April 24, 2015, ahead of print in the BMJ are shedding more light on the risk of GI bleeding associated with newer anticoagulants. One found that bleeding was more common with dabigatran and rivaroxaban than with warfarin among older patients; the second detected no clear age-related differences between the newer agents and warfarin. 

For the first study, Neena S. Abraham, MD, MS, of the Mayo Clinic (Scottsdale, AZ), and colleagues looked at the incidence of GI bleeding in 92,816 patients from the Optum Labs Data Warehouse, a large database that holds administrative claims data on patients with private insurance and Medicare Advantage. The cohort included patients with or without A-fib who filled a new prescription for dabigatran (n = 8,578; Pradaxa; Boehringer Ingelheim), rivaroxaban (n = 16,253; Xarelto; Janssen Pharmaceuticals), or warfarin (n = 67,985) between November 2010 and September 2013. 

Propensity matching for A-fib status and use of a newer anticoagulant vs warfarin resulted in 7,749 and 732 matched pairs, respectively. 

Risk Increases With Age

The overall risks of GI bleeding (primary endpoint) with dabigatran and rivaroxaban were similar to that with warfarin regardless of A-fib status. However, there was a pattern of increasing risk with advancing age, such that by age 76, dabigatran and rivaroxaban both carried higher bleeding risks than warfarin.

Dr. Abraham and colleagues say that despite the rapid uptake of dabigatran and rivaroxaban in routine practice, there has been little real-world assessment of their safety because most trials have focused on the Medicare population.

“Caution should be used when prescribing novel oral anticoagulants to older people, particularly those over 75 years of age,” they observe. 

No Difference Between Agents, but Concerns Remain

For the second study, investigators led by Hsien-Yen Chang, PhD, MHA, MSc, of Johns Hopkins Bloomberg School of Public Health (Baltimore, MD), examined data on 46,163 commercially insured patients who received an initial prescription of dabigatran (n = 4,907), rivaroxaban (n = 1,649) or warfarin (n = 39,607) between October 2010 and March 2012. 

Dabigatran users tended to be older relative to rivaroxaban or warfarin users (62.0 vs 57.6 vs 57.4 years, respectively) and were more likely to be men (69% vs 49% vs 53%, respectively). 

In propensity-weighted analysis, dabigatran patients had the highest rate of GI bleeding and rivaroxaban the lowest compared with warfarin (9.01 vs 3.41 vs 7.02 per 100 person-years). However, after adjustment for potential confounders, bleeding risk was similar between users of dabigatran and warfarin (adjusted HR 1.21; 95% CI 0.96-1.53) as well as between users of rivaroxaban and warfarin (adjusted HR 0.98; 95% CI 0.36-2.69). 

Additionally, no differences were seen in adjusted risk when comparing the drugs in patients under or over age 65. Between dabigatran and warfarin users under age 65, however, the difference was “borderline significant…, suggesting a potentially increased risk of GI bleeding with dabigatran,” the authors write. 

This is “noteworthy,” they add, “given that several studies examining this outcome have identified such an association.” 

Dr. Chang and colleagues observe that their study participants were relatively young (mean age 57.6 years) compared with the Medicare cohorts assessed in many of the large trials of the newer anticoagulants. Also, they say, the excess hazard of dabigatran relative to warfarin “may be age-dependent and concentrated among very elderly people, a population that was not well represented in this sample of commercially insured patients.” 

The authors say their results are similar to a recent US observational trial showing no difference in real-world rates of GI bleeding between rivaroxaban and warfarin in patients with nonvalvular A-fib (HR 1.06; 95% CI 0.71-1.64). However, despite the suggestion of lower risk in younger, insured patients in the current study, the researchers do not “rule out a greater than 50% increased risk with dabigatran and more than twofold increased risk with rivaroxaban.” 

Unclear How Age Influences Risk

In an accompanying editorial, Mary S. Vaughan-Sarrazin, PhD, MA, of the Iowa City VA Medical Center (Iowa City, IA), and Adam J. Rose, MD, MSc, of Edith Nourse Rogers Memorial Veterans Hospital (Bedford, MA), say the studies “add to a complex picture of real-world risk associated with the newer drugs relative to warfarin.” 

However, they note, it is not yet clear exactly how age increases the propensity for GI bleeding in patients on anticoagulants. 

“We need better ways to predict which patients are at highest risk of gastrointestinal bleeding, especially during treatment with newer oral anticoagulants,” they write. “Monitoring of drug concentrations in patients taking newer agents, combined with a range of possible dose options, may hold the key to optimizing the safety and effectiveness of these unfamiliar drugs.” 


1. Abraham NS, Singh S, Alexander GC, et al. Comparative risk of gastrointestinal bleeding with dabigatran, rivaroxaban, and warfarin: population based cohort study. BMJ. 2015;Epub ahead of print. 

2. Chang H-Y, Zhou M, Tang W, et al. Risk of gastrointestinal bleeding associated with oral anticoagulants: population based retrospective cohort study. BMJ. 2015;Epub ahead of print.

3. Vaughan-Sarrazin MS, Rose A. Safety of new oral anticoagulants: we need reliable tools to predict risk of gastrointestinal bleeding [editorial]. BMJ. 2015;Epub ahead of print.

  • Dr. Abraham’s study was funded by the Mayo Clinic Robert D and Patricia E Kern Center for the Science of Health Care Delivery.
  • Drs. Chang, Abraham, Vaughan-Sarrazin, and Rose report no relevant conflicts of interest.