This article is more than 5 years old. Information may no longer be current.
Thromboembolic risk greater with direct oral anticoagulants vs. warfarin when adherence low
Click Here to Manage Email Alerts
BOSTON — Among patients with nonvalvular atrial fibrillation prescribed an oral anticoagulant who had low adherence to their medication regimen, those assigned warfarin had lower risk for thromboembolic events than those assigned a direct oral anticoagulant, according to findings presented at the Heart Rhythm Society Annual Scientific Sessions.
Dhanunjaya Lakkireddy, MBBS, FACC, FHRS, executive medical director of The Kansas City Heart Rhythm Institute & Research Foundation, and colleagues conducted a retrospective cohort study of 83,168 patients with nonvalvular AF and a CHA2DS2-VASc score of at least 2 from administrative claims databases who were prescribed an oral anticoagulant for the first time in 2014 or 2015.
“The whole world has moved toward using [direct oral anticoagulants] in a big way,” Lakkireddy said at a press conference. “Adherence is a big problem with any medication, but how does it impact medications that are supposed to improve adherence? One reason we switch people from warfarin is that it’s a pain to take it regularly, people don’t want to take their INRs and they can’t eat certain foods or take certain drugs when on warfarin. But does that translate into the outcomes we want to have?”
The researchers compared 42,639 patients (mean age, 76 years; 43% women) prescribed warfarin with 40,529 patients (mean age, 72 years; 44% women) prescribed a direct oral anticoagulant — dabigatran (Pradaxa, Boehringer Ingelheim), rivaroxaban (Xarelto, Janssen), apixaban (Bristol-Myers Squibb/Pfizer) and edoxaban (Savaysa, Daiichi Sankyo). Patients were stratified into high-adherence (adherent more than 80% of days) or low-adherence (adherent 40% to 80% of days) groups. Patients adherent to their medication regimen less than 40% of days were excluded.
The objectives of the study were to determine real-world adherence to the medications, to identify differences in the rates of thromboembolic and bleeding events by type of anticoagulant and level of adherence and to study the impact of adherence on clinical outcomes, Lakkireddy said during a presentation.
At 2.5 years, in the overall cohort, survival free from any thromboembolic events (P < .001) and from any stroke (P = .002) was higher in the warfarin cohort, whereas survival free from hemorrhagic stroke (P < .001) and from bleeding (P < .001) was higher in the direct oral anticoagulant cohort, Lakkireddy said.
Hospitalizations and ED visits were higher in the direct oral anticoagulant group vs. the warfarin group for thromboembolic events (3.37 per 100 patient-years vs. 3.16 per 100 patient-years; P = .011) but lower in the direct oral anticoagulant group for hemorrhagic stroke (0.54 per 100 patient-years vs. 0.88 per 100 patient-years; P < .001) and bleeding (4.9 per 100 patient-years vs. 5.92 per 100 patient-years; P < .001), whereas there was no difference in stroke hospitalizations (P = .96), according to the researchers.
In the warfarin group, 63.1% had high adherence, whereas in the direct oral anticoagulant group, 73.5% had high adherence, Lakkireddy said.
Among those with high adherence, event-free survival was similar to the overall cohort, with survival free from any thromboembolic events (P < .001) and any stroke (P < .032) being better in the warfarin group and survival free from bleeding and hemorrhagic stroke (P < .001 for both) being better in the direct oral anticoagulant group, according to the researchers.
Among those with low adherence, survival free from any thromboembolic events and any stroke also favored the warfarin group (P < .001 for both), but to a greater degree than in the high-adherence group, whereas survival free from hemorrhagic stroke favored the direct oral anticoagulant group (P = .006) and there was no difference in survival free from bleeding (P = .883), Lakkireddy said.
“These findings were profound,” Lakkireddy said at the press conference. “In low adherence, the warfarin arm did much better than the [direct oral anticoagulant] arm in terms of thromboembolic events.”
The rates of hospitalizations or ED visits for all outcomes were higher for patients with low adherence than for those with high adherence, regardless of anticoagulant choice, he said.
Lakkireddy said a propensity-matched analysis is planned in hopes of better eliminating confounders.
“There are a lot of other confounding variables that could be adding on to this mix,” he said at the press conference. “We are going to do a propensity-matched analysis and take a deeper dive into the differences between the individual drugs; for example, rivaroxaban is once-a-day dosing and apixaban is twice-a-day dosing, does that lead to differences in outcomes?” – by Erik Swain
Reference:
Lakkireddy D, et al. LBCT02-03. Presented at: Heart Rhythm Society Annual Scientific Sessions; May 9-12, 2018; Boston.
Disclosure: Lakkireddy reports he has financial ties with Biosense Webster, Boehringer Ingelheim, Bristol-Myers Squibb, Estech, Janssen, Pfizer, SentreHeart and St. Jude Medical.
Perspective
Back to Top
Andrew D. Krahn, MD, FHRS
These data provoke us to ask two questions. No. 1, are were sure about the therapeutic effects of the two types of agents? We still believe the highest form of evidence is properly conducted randomized controlled trials, and I don’t think these findings undermine those data. The adjusted ORs are aligned with the clinical trial results. But, it’s good to pose that question to make certain.
No. 2, are effective treatments are being adhered to? Clearly not. We need to get with the message of delivering health care effectively and not focus solely on identifying beneficial treatments.
Click Here to Manage Email Alerts