Benefits Could Outweigh Risks When Restarting Oral Anticoagulants in Patients with Major Bleeding

Patients who have experienced major bleeding while taking direct oral anticoagulants (DOAC) had better survival rates and better outcomes if oral anticoagulation was restarted, compared with patients who discontinued use of the medications, according to the results of a prospective study presented at the 2015 ASH Annual Meeting.

In this study, Sandra Marten, MD, from the Dresden University Hospital in Germany, and colleagues evaluated the outcomes and management after major bleeding complications of patients being treated with DOACs. While these newer agents reduce the risk of major bleeding compared with warfarin, Dr. Marten and co-authors wrote, “little is known about the management and outcome of survivors of major DOAC bleeding.”

Using data from 2,771 patients enrolled in the Dresden Novel Oral Anticoagulant Registry, the researchers assessed survival rates among patients who developed a major DOAC bleeding event (per International Society for Thrombosis and Hemostasis criteria), as well as whether restarting oral anticoagulation (including DOAC or vitamin K antagonists) 30 days after the bleeding event put patients at risk for a composite endpoint of recurrent major bleeding, stroke, transient ischemic attack, systemic embolism, and/or venous thromboembolism.

Patients were treated with the following agents: rivaroxaban (n=1,898), apixaban (n=525), and dabigatran (n=348).

During a mean follow-up of 23.6 months, 127 patients (mean age = 77±11 years; range = 37-94) developed a total of 170 major bleeding events while taking DOACs.

About two-thirds (64.6%) of the bleeding events occurred spontaneously; 21.3 percent occurred after surgery or an interventional procedure and another 14.2 percent were related to bleeding after trauma. Bleeding events resulted in a ≥2 g/l drop of hemoglobin in 106 (62.4%) cases, transfusion of ≥2 units of red blood cells in 105 (61.8%) cases, critical site bleeding in 43 (25.3%) cases, and/or fatal outcome in 9 cases (5.3%).

“Even in cases with major DOAC bleeding, acute mortality is low,” Dr. Marten and colleagues reported, with a case-fatality rate of 5.3 percent.

The most common sites of bleeding were:

  • the gastrointestinal tract (37%)
  • diffuse bleeding during or after surgery (15.7%)
  • intracranial bleeding (11%)

Bleeding incidents required hospitalization in 85 cases, while 31 bleeding events took place during a hospital stay. Eleven cases were handled in an outpatient setting.

Once major bleeding occurred, patients were managed in a variety of ways, including:

  • Red blood cell transfusion (57.1%)
  • Surgical or interventional treatment (55.9%)
  • Use of prothrombin complex concentrate (11.8%)
  • Use of fresh frozen plasma (8.7%)
  • Use of fibrinogen (3.1%)

Eighty patients (63%) restarted oral anticoagulation at 30 days after major bleeding. Though characteristics were similar between patients who did and did not restart oral anticoagulation, subsequent outcomes and survival rates varied significantly.

During a mean follow-up of 15.2 months following the bleeding event, the rate of the combined endpoint of recurrent major bleed and thromboembolism was significantly lower in patients who restarted oral anticoagulation compared with those who did not (TABLE).

Mortality also differed between the two groups, with those restarting oral anticoagulation experiencing a significantly lower mortality rate than those who did not restart. The most common causes of death were a cardiovascular event (31.6%) and bleeding (23.7%).

Overall, Dr. Marten and co-authors concluded, “the benefits of oral anticoagulation continuation may outweigh the risks even in patients with major DOAC-related bleeding.”


Marten S, Tittl L, Daschkow K, Beyer-Westendorf J. Pattern and management of ISTH major bleeding complications with direct oral anticoagulants – Results of the prospective Dresden Noac Registry (NCT01588119). Abstract #892. Presented at the American Society of Hematology Annual Meeting, December 7, 2015; Orlando, FL.

TABLE. Rates of Endpoints in Patients Who Did or Did Not Restart Oral Anticoagulation
Oral Anticoagulant Restarted (n=80) Oral Anticoagulant Not Restarted (n=47) p Value
Combined endpoint 14.7/100 patient-years (95% CI 8.0-24.7) 38.6/100 patient-years (95% CI 21.1-64.7) 0.0342
Mortality 16.4/100 patient-years (95% CI 9.7-25.9) 40.6/100 patient-years (95% 24.8-62) p=0.0099