Peer Reviewed

Controversies in Clinical Care

Direct Oral Anticoagulants and the Risk of Intracranial Bleeding and Death

Eric A. Dietrich, PharmD, BCPS, and Kyle Davis, PharmD, BCPS

Dietrich EA, Davis K. Direct oral anticoagulants and the risk of intracranial bleeding and death. Consultant. 2018;58(4):153-154.


Atrial fibrillation (AF) is the most common cardiac arrhythmia. One of the most devastating complications of AF is cardioembolic stroke. Patients with AF are 5 times more likely to experience a cardioembolic stroke, and many patients would prefer death over a disability from stroke.1 In an effort to combat this crippling complication, the use of antithrombotic therapy has become the standard of care in at-risk patients with AF.2 More recently, the armamentarium of antithrombotic agents has grown significantly with the development of direct oral anticoagulants (DOACs) in addition to vitamin K antagonists and aspirin. While many of these agents offer efficacy that is similar to or even superior to warfarin, many clinicians are drawn away from prescribing these agents because of the perceived bleeding risk and the lack of a reversal agent. However, recent research has attempted to evaluate the safety of these newer antithrombotic agents in the real-world setting.

Patient Case

WH is a 67-year-old man who recently received a diagnosis of AF, which had been detected on an electrocardiogram during a yearly physical examination. He was referred to a cardiologist who conducted an event monitor study, the results of which confirmed the diagnosis of AF. WH has a history of hypertension and stable angina, giving him a CHA2DS2-VASc stroke risk score of 3 of 9 (age 65-74, 1 point; hypertension, 1 point; and vascular disease, 1 point).

He is recommended to begin anticoagulation, and the differences between a DOAC and warfarin are discussed with him. WH is particularly concerned about the lack of a clear reversal mechanism for DOACs. He is worried that if he bleeds into his brain, the lack of reversal agent will increase his risk of death if he takes a DOAC instead of warfarin. Is there evidence to support this concern?

The Evidence

Four DOACs are currently approved by the Food and Drug Administration (FDA) for the prevention of stroke and systemic embolism in patients with AF: apixaban, dabigatran, edoxaban, and rivaroxaban. In premarketing clinical trials, the rate of intracranial hemorrhage (ICH) was significantly lower with DOAC therapy relative to warfarin therapy.3 Because of this important difference in bleeding risk, national guidelines consider DOACs as a first-line anticoagulant option for stroke prevention in AF. However, concern remains that the lack of true reversal agent for all DOACs will lead to higher mortality rates if a bleeding event does occur. (Note: The FDA approved idarucizumab in October 2015 for the rapid reversal of dabigatran).

Recently, a retrospective cohort study compared the in-hospital mortality rates of patients exposed to a DOAC or warfarin in the 7 days prior to an admission for ICH.4 A total of 141,311 patients admitted over 3 years (October 2013 to December 2016) to 1622 hospitals were identified in the analysis; the patients had a mean age of 68.3 years, and 48.1% were women. Before developing ICH, 15,036 patients had been taking warfarin, and 4918 had been taking a DOAC (apixaban, 34.9%; dabigatran, 11%; edoxaban, 0.1%; rivaroxaban, 54%). Compared with patients exposed to warfarin, patients exposed to a DOAC had a lower risk of in-hospital mortality (adjusted risk difference, −5.7% [97.5% CI, −7.3% to −4.2%]; adjusted odds ratio, 0.75 [97.5% CI, 0.69-0.81]). Of the warfarin-treated patients, 39.4% had a subtherapeutic international normalized ratio (INR) (< 2) at admission, 37.3% had a therapeutic INR (2-3), and 23.3% had a supratherapeutic INR (> 3). A dose-dependent effect on unadjusted mortality rates was observed, such that the mortality rate increased from 25% for patients with a subtherapeutic INR, 32.4% for patients with a therapeutic INR, and 37.9% for patients with a supratherapeutic INR.


NEXT: Clinical Application

Clinical Application

This large retrospective study corroborates previous reports that the rate of ICH with DOACs is lower than the rate with warfarin and, importantly, shows that the risk of mortality following ICH is lower with DOAC use compared with warfarin use. While the number of DOAC-treated patients experiencing an ICH was relatively small (n = 4918), the investigators were able to statistically adjust the analysis to account for and control for differences related to the larger warfarin-treated cohort (n = 15,036).

An additional consideration about this study, which concluded in December 2016, is that more than half of the admissions (63%) took place prior to FDA approval of idarucizumab for dabigatran reversal in October 2015. Nevertheless, it is unlikely that idarucizumab was widely available for use until many months later, and only 11% of the patients admitted for a DOAC-related ICH had been taking dabigatran. Therefore, the availability of idarucizumab likely did not significantly alter the results, but the results may suggest that the current hospital mortality rate for dabigatran-related ICH may be lower than the rate observed in the study, given the more widespread availability of idarucizumab now.

An important limitation of this study was the inability to directly determine the timing of patients’ most recent dose of a DOAC or warfarin and whether patient adherence or drug-drug interactions led to supratherapeutic or subtherapeutic levels of the DOAC. Additionally, anticoagulation-reversal strategies were not analyzed, and protocols for the management of bleeding may have differed among hospitals, contributing differences in mortality rates. Lastly, the study evaluated only in-hospital mortality, so it is unclear whether postdischarge mortality differed between the groups.

Separate analyses have confirmed the reduced rate of major bleeding with DOACs relative to warfarin. A 2015 meta-analysis5 that included 11 studies comparing a DOAC with warfarin (5 studies on AF, 6 studies on venous thromboembolism; N = 100,324) found that DOAC-treated patients had a 47% lower rate of fatal bleeding than did warfarin-treated patients (3 events avoided per 1000 patients) and that the case fatality rate following a major bleed was lower for DOAC-treated patients than for those receiving warfarin for AF (odds ratio, 0.68; 1 death avoided per 39 major bleeding events). The studies included in the meta-analysis were largely premarketing clinical trials, which had an inherently high level of oversight and monitoring of patients; thus, the results might not be directly comparable with current monitoring practices in real-world clinical practice.

Given the available evidence, especially in the context of a lack of a reversal agent being widely available during the study period, the safety profile of DOACs related to major and fatal bleeding is superior to that of warfarin. The approval and widespread availability of reversal agents specifically for DOACs may lead to a widening of the safety profile in the future. However, this is not to suggest that DOACs are the drug of choice for all patients requiring an anticoagulant and will always provide a more favorable risk profile. Kidney disease and issues related to adherence and appropriateness of the DOAC dose may negatively alter the risk profile of DOACs.6

Lastly, practical differences exist between DOACs and warfarin that patients may value differently. An ability to quantify the degree of anticoagulation with warfarin may represent a benefit to certain patients. Additionally, the requirement for more frequent follow-up with warfarin provides additional opportunities to evaluate and monitor comorbid disease states; in patients with more comorbidities, this may ensure the patients’ other medications and disease states remain stable and well controlled. But fear of increased bleeding and of increased risk of mortality with DOACs relative to warfarin should not be a factor driving a patient to warfarin, assuming that the patient is well educated and monitored appropriately to ensure the DOAC is dosed and administered properly. To achieve this, patients taking a DOAC likely require multiple dedicated in-clinic visits per year, but these visits can ensure that the safety margin is maintained.

Outcome of the Case

WH is counseled that although the DOACs other than dabigatran do not have a specific reversal agent, the rates of major bleeding, fatal major bleeding, and fatal ICH are lower on average with DOACs compared with warfarin. Combined with at least similar efficacy to warfarin, and potentially improved efficacy for certain agents, the overall risk-benefit profile of DOACs appears favorable.

Should WH elect to begin a DOAC, he would be counseled extensively on the proper administration and timing of the DOAC, would be screened for potential interacting medications such as antiplatelet agents and nonsteroidal anti-inflammatory drugs, and would undergo recommended laboratory monitoring, in addition to DOAC-specific monitoring visits 2 to 4 times per year based on other factors, to ensure the DOAC is optimally used to take advantage of the favorable safety profile. 

Eric A. Dietrich, PharmD, BCPS, is a graduate of the University of Florida College of Pharmacy and completed a 2-year fellowship in family medicine where he was in charge of an anticoagulation clinic. He works for the College of Pharmacy and the College of Medicine at the University of Florida in Gainesville.

Kyle Davis, PharmD, BCPS, is a graduate of the University of Florida College of Pharmacy in Gainesville and completed a 2-year residency in internal medicine at Indiana University in Indianapolis. He is an internal medicine specialist at Wake Forest Baptist Medical Center in Winston-Salem, North Carolina.


  1. Hanger HC, Fogarty B, Wilkinson TJ, Sainsbury R. Stroke patients’ views on stroke outcomes: death versus disability. Clin Rehabil. 2000;14(4):417-424.
  2. You JJ, Singer DE, Howard PA, et al. Antithrombotic therapy for atrial fibrillation: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2 suppl):e531S-e575S.
  3. Ruff CT, Giugliano RP, Braunwald E, et al. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials. Lancet. 2014;383(9921):955-962.
  4. Inohara T, Xian Y, Liang L, et al. Association of intracerebral hemorrhage among patients taking non–vitamin K antagonist vs vitamin K antagonist oral anticoagulants with in-hospital mortality. JAMA. 2018;319(5):463-473.
  5. Caldeira D, Rodrigues FB, Barra M, et al. Non-vitamin K antagonist oral anticoagulants and major bleeding-related fatality in patients with atrial fibrillation and venous thromboembolism: a systematic review and meta-analysis. Heart. 2015;101(15):1204-1211.
  6. Steinberg BA, Shrader P, Thomas L, et al; ORBIT-AF Investigators and Patients. Off-label dosing of non-vitamin K antagonist oral anticoagulants and adverse outcomes: the ORBIT-AF II Registry. J Am Coll Cardiol. 2016;​68(24):2597-2604.