Bleeding associated with the newer non-vitamin K oral anticoagulants (NOACs) is not more dangerous than that associated with warfarin, according to a large retrospective study of claims data.

“There is a lot of concern about the possibility of bleeding on NOACs given that it is difficult to measure the degree of anticoagulation on NOACs and the ability to reverse NOACs is limited,” said Dr. Blake Charlton from the University of California, San Francisco.

“In my discussions with patients, many wonder, ‘If I do bleed on a NOAC how bad will it be? Is it worse to bleed on a NOAC or on warfarin?’ In this study, we tried to answer that question,” he told Reuters Health by email. “The most important finding of this study is that there was no difference in mortality among patients admitted for bleeding on a NOAC vs. warfarin.”

Data are lacking on the outcomes of bleeding complications during anticoagulant therapy with warfarin in comparison to the newer NOACs, dabigatran and rivaroxaban, the authors note in PLoS ONE.

“The original NOAC trials and the subsequent NOAC research focused the incidence of bleeding. However, we went a step farther and examined those patients who experience bleeding,” Dr. Charlton said.

He and his colleagues at UCSF and HealthCore, an outcomes research company, identified 3,144 patients with atrial fibrillation in the HealthCore database who were hospitalized for bleeding after starting warfarin (2,446 patients), dabigatran (442 patients) or rivaroxaban (256 patients).

“Interestingly,” said Dr. Charlton, NOAC-related bleeding was associated with shorter hospitalizations than warfarin-related bleed. Warfarin users, who were older on average than the other patients and had more co-morbidities, were hospitalized an average of two days longer than dabigatran users and 2.6 days longer than rivaroxaban users.

There were no marked differences in ICU admissions among the groups, but warfarin users stayed three days longer in the ICU, on average, than dabigatran users and 2.4 days longer than rivaroxaban users.

There were no significant between-group differences in 30-day and 90-day all-cause mortality. “It is important to state that this study is subject to the limitations of retrospective data and claims data; however, the lack of a clear mortality signal is reassuring,” said Dr. Charlton.

“We found it surprising,” he added, “that warfarin was not associated with better outcomes given that warfarin can be fairly easily reversed. This suggests that reversal of anticoagulation may not play a significant role in improving outcomes among patients admitted for non-traumatic bleeding due to anticoagulation, which is particularly important now that reversal agents for NOACs are being evaluated by the FDA.”