Author + information
- Peter A. Noseworthy, MD∗ (, )
- Xiaoxi Yao, PhD,
- Nilay D. Shah, PhD and
- Bernard J. Gersh, MBChB, DPhil
- ↵∗Division of Cardiovascular Diseases, Mayo Clinic College of Medicine, 200 First Street SW, Rochester, Minnesota 55905
Hypertrophic cardiomyopathy (HCM) affects >600,000 patients in the United States, and approximately 1 in 5 of them have atrial fibrillation (AF) (1). The incidence of stroke in patients with HCM is markedly increased when complicated by AF, with an annualized risk of nearly 4% (2). AF is an indication for warfarin in nearly all patients with HCM (3,4). Non–vitamin K antagonist oral anticoagulants (NOACs) may be reasonable alternatives to warfarin, but there are no explicit data to support their use (3,4).
Although not rare, HCM is not common enough to motivate a dedicated trial comparing NOACs with warfarin. Few patients with HCM were included in existing NOACs trials so subgroup analysis of trial data is not feasible. Large observational datasets, however, may offer an opportunity to examine outcomes in these patients. We aimed to compare the risks of stroke and major bleeding in patients with HCM and AF treated with NOACs or warfarin.
Using a large U.S. commercial insurance database, OptumLabs Data Warehouse, we identified 2,198 privately insured and Medicare Advantage patients with HCM and AF who were users of NOACs (n = 579) or warfarin (n = 1,619) between October 1, 2010 and April 30, 2015. Patients who had dialysis, kidney transplant, mitral stenosis, prosthetic heart valves, or mitral valve repair were excluded. The primary effectiveness outcome was stroke or systemic embolism, including ischemic stroke, hemorrhagic stroke, and systemic embolism. The primary safety outcome was major bleeding, including intracranial hemorrhage gastrointestinal bleeding, and bleeding from other sites. We created a propensity score–matched warfarin (n = 859) versus NOACs (n = 568; n = 235 dabigatran, n = 231 rivaroxaban, and n = 102 apixaban) cohort, using 3:1 propensity-score matching with replacement and with a caliper of 0.01, based on baseline characteristics, including age, gender, race, annual household income, residence region, Charlson-Deyo comorbidity index, CHA2DS2-VASc score, HAS-BLED score, individual components of the CHA2DS2-VASc and HAS-BLED scores, and previous warfarin experience. We used Cox proportional hazards models to compare time to first inpatient admission for a clinical endpoint occurring on therapy.
The median age was 67 years (interquartile range: 58 to 76 years), 65.4% were male, and the median CHA2DS2-VASc score was 4 (interquartile range: 2 to 5). A total of 15.7% of the NOAC-treated patients received a reduced dose of dabigatran, rivaroxaban, or apixaban. Over an average of 0.56 years of follow-up, the incidence rates for stroke or systemic embolism were similar between NOAC- and warfarin-treated patients (1.93 and 2.03 events per 100 person-years for NOACs and warfarin, respectively) (Table 1). NOACs showed a nonsignificant trend toward lower major bleeding rates (4.18 and 5.38 per 100 person-years, for NOACs and warfarin, respectively). Furthermore, we did not find any significant difference for any of the secondary effectiveness or safety outcomes, likely because of the small sample size. Although not statistically significant, the rates of intracranial bleeding and hemorrhagic stroke were lower in NOAC-treated patients compared with warfarin. These data support the use of NOAC for stroke prevention as an alternative to warfarin in patients with AF and HCM. In a sensitivity test performed in patients who received reduced dose NOAC, there were no observed strokes or systemic embolic events in NOAC-treated patients and the bleeding risk was comparable with warfarin.
The principle limitation of our study pertains to the observational nature of this analysis, which limits our ability to draw any causal inferences because of residual unmeasured confounding. In addition, we rely on billing codes for patient characteristics and outcome adjudication. As such, we lack clinical characteristics, such as mitral valve dysfunction, left atrial enlargement, left ventricular dysfunction, and other factors that may predispose to cardioembolism. Our data should be taken in this context and should be seen as one piece of the available literature to guide treatment decisions in this group of patients.
Nonetheless, these observational data provide a glimpse at real-world clinical outcomes associated with NOAC use for patients with HCM and AF, a group not well studied in existing or ongoing clinical trials. Our data suggest that patients with HCM and AF can be safely and effectively treated with the NOACs.
Please note: Dr. Gersh has consulted for Janssen Pharmaceuticals. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Siontis K.C.,
- Geske J.B.,
- Ong K.,
- et al.
- Guttmann O.P.,
- Rahman M.S.,
- O'Mahony C.,
- et al.
- Gersh B.J.,
- Maron B.J.,
- Bonow R.O.,
- et al.
- Elliott P.M.,
- Anastasakis A.,
- Borger M.A.,
- et al.