Atrial fibrillation (AF) is present in approximately 40% of patients with cardiac amyloidosis. AF in the setting of cardiac amyloidosis is associated with a significant risk of stroke and systemic thromboembolism regardless of the patient’s CHA₂DS₂-VASc (congestive heart failure, hypertension, age ≥ 75 years, diabetes mellitus, stroke or transient ischemic attack, vascular disease, age 65 to 74 years, sex category) score. Although current guidelines recommend anticoagulation therapy in all patients with cardiac amyloidosis and AF, data about the comparative effectiveness and safety of vitamin K antagonists (VKAs) and direct oral anticoagulants (DOACs) in this patient population are scarce.
Using the methods previously described by Gilstrap et al, we identified patients from the 5% enhanced sample of Medicare beneficiaries who were admitted to the hospital with a diagnosis of amyloidosis and heart failure from January 2015 to November 2019, then had a new diagnosis of AF in the follow-up, and were initiated on oral anticoagulation. We excluded patients who were not enrolled in Medicare Part D coverage during the year of AF diagnosis. We extracted all pharmacy claims for the study cohort, and patients were followed until they stopped anticoagulation, as evidenced by a gap of >90 days without a new prescription or refill. The primary exposure of interest was DOACs (rivaroxaban, dabigatran, and apixaban) versus VKAs. We used the inverse probability weighting method to adjust for differences between both groups. The study primary outcome was all-cause mortality and its secondary outcome was admission with a primary diagnosis of stroke or major bleeding as defined in previous studies. Data for mortality were available through September 2020 and for other outcomes through December 2019. Patients were censored when they stopped the medication, had an event, or reached the end of the study period.
The final study cohort included 551 patients, 213 on warfarin (38.7%) and 338 on DOACs (61.3% [8.9% on dabigatran, 20.7% on rivaroxaban, and 70.4% on apixaban]). Before adjustment, there was no significant difference between DOAC and VKA groups in age (mean 77.6 years [SD 9.1] vs 77.0 years [SD 9.7], p = 0.5), gender (women 36.4% vs 39.4%, p = 0.5), CHA₂DS₂-VASc score (median 5, interquartile range [IQR] 4 to 6] vs median 5, IQR 4 to 6]), or prevalence of most co-morbidities (coronary artery disease 59.5% vs 56.35%, anemia 40.2% vs 35.2%, liver disease 6.5% vs 10.3%, kidney disease 36.4% vs 39.4%, diabetes mellitus 32.5% vs 33.8%, previous stroke 9.1% vs 11.7%; p >0.1 for all). There was an adequate balance between the 2 groups after inverse probability weighting matching. After a median follow-up of 444 days (IQR 170 to 823), DOACs were associated with a lower risk of mortality (adjusted hazard ratio [HR] 0.71, 95% confidence interval [CI] 0.59 to 0.85, p <0.001), a lower risk of stroke (adjusted HR 0.44, 95% CI 0.28 to 0.70, p <0.001), and major bleeding (adjusted HR 0.55, 95% CI 0.36 to 0.84, p <0.001) ( Figure 1 ). The lower risk of stroke was driven mainly by ischemic stroke (HR 0.49, 95% CI 0.29 to 0.81, p <0.01), not hemorrhagic stroke (HR 0.71, 95% CI 0.33 to 1.50, p = 0.3). DOACs, compared with VKAs, were also associated with lower risk of 1-year stroke (4.3% vs 9.4%, p <0.01) and 1-year mortality (34.4% vs 43.5%, p <0.01).
