Early Phase of NOAC, Vitamin K Antagonist Therapy: Similar Bleeding, Stroke Risk

Marlene Busko

August 06, 2015

PARIS and SAINT-DENIS, FRANCE — In a national cohort study of patients in France with newly diagnosed nonvalvular atrial fibrillation (AF), those who received a novel oral anticoagulant (NOAC)—dabigatran (Pradaxa, Boehringer Ingelheim) or rivaroxaban (Xarelto, Bayer/Janssen Pharmaceuticals)—had similar 90-day risks of an arterial thromboembolic event or major bleeding as those who received a vitamin K antagonist (VKA)[1].

The study, with co–first authors Dr Géric Maura (National Health Insurance, Paris, France) and Pierre-Olivier Blotière (French National Agency for Medicines and Health Products Safety, Saint-Denis, France) and colleagues, was published July 29, 2015 in Circulation.

The findings show that "although the NOACs (only dabigatran and rivaroxaban here) have provided AF patients with a convenient, fixed-dose alternative to VKA and they were proven to be superior (dabigatran) or noninferior in safety and efficacy end points vs warfarin in large randomized trials, [they] cannot be considered to be safer and more efficacious than VKA during the early phase of treatment," Maura told heartwire from Medscape. These findings are consistent with those of Sørensen R et al[2] in a Danish population, he noted.

Although it is reassuring that there was no marked early excess hospitalization in patients who were started on NOACs, the study suggests that "physicians must be just as cautious when initiating a NOAC as when initiating VKA, particularly in view of the absence of an antidote and objective monitoring of the extent of anticoagulation," the researchers write.

Safety and Efficacy of NOACs in First 3 Months of Use

Oral anticoagulants are recommended for long-term prophylaxis to prevent stroke in AF, but studies have also shown that stroke and bleeding rates are higher in the initial phase of warfarin treatment—which has been less well studied with NOACs, the researchers write.

They performed a propensity-matched cohort study using data from two French national linked databases (for health insurance and hospital discharge) to assess major bleeding and thromboembolic events in the initial phase of therapy with dabigatran and rivaroxaban, each vs a VKA, in patients with AF.

They identified patients with no prior reimbursement for an oral anticoagulant (new users) who were prescribed dabigatran or rivaroxaban between July 20, 2012 (when these agents entered the French marketplace) and November 30, 2012. They also identified patients who were prescribed a VKA for during this time in 2011, and they excluded patients with contraindications, lower-limb orthopedic procedures, deep vein thrombosis, or pulmonary embolism.

They classified 75-mg or 110-mg dabigatran and 10-mg or 15-mg rivaroxaban as low-dose NOACs, and 150-mg dabigatran and 20-mg rivaroxaban as high-dose NOACs.

The cohort consisted of 8443 patients who were prescribed dabigatran (69.8% received a low dose), 4651 patients who were prescribed rivaroxaban (38.5% received a low dose), and 19,713 patients who were prescribed a VKA (83.7% received fluindione and 11.8% received warfarin).

Compared with patients who were treated with VKAs, those treated with high-dose dabigatran or rivaroxaban were more frequently male and younger, with lower mean HAS-BLED and CHA2DS2-VASc scores and fewer comorbidities. Patients started on low-dose NOACs were more frequently female and older. Each NOAC-treated patient was matched with two VKA-treated patients.

During the first 90 days of therapy, there were 55 and 122 bleeding events in dabigatran- and VKA-treated patients, and 31 and 68 bleeding events in rivaroxaban- and VKA-treated patients, respectively.

There were also 33 and 58 arterial thromboembolic events in dabigatran- and VKA-treated patients, and 12 and 28 such events in rivaroxaban- and VKA-treated patients, respectively.

In this early phase of treatment, there was no significant difference in major bleeding or a composite of major bleeding or death in patients who received low-dose or high-dose dabigatran or rivaroxaban compared with matched patients who received VKA.

Similarly, there was no significant difference in ischemic stroke or systemic embolism or a composite of these end points plus death in patients who received low-dose or high-dose NOACs and patients who received VKA.

Early Bleeding and Stroke Risk, NOAC vs VKA, Hazard Ratio (95% CI)*

Group Major bleeding Stroke or systemic embolism
Dabigatran vs VKA 0.88 (0.64–1.21) 1.10 (0.72–1.69)
Rivaroxaban vs VKA 0.98 (0.64–1.51) 0.93 (0.47–1.85)
*Within 90 days of treatment initiation
NOAC=novel oral anticoagulant
VKA=vitamin K antagonist

French prescribing practices appear to be strongly guided by bleeding risk, since many patients were prescribed low doses—notably dabigatran 75 mg and rivaroxaban 10 mg, which have not been approved for AF in the European Union (unlike in the US), Maura noted. Possibly this was related to the fact that more than a third of the dabigatran- or rivaroxaban-treated patients were 80 and or older, a population that was underrepresented in pivotal clinical trials.

"Similar analyses should be extended to other NOACs such as apixaban [Eliquis, Bristol-Myers Squibb/Pfizer], and observational studies should now focus on NOAC head-to-head comparison in a noninferiority design," the researchers conclude.

The authors have no relevant financial relationships.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.