Characteristics associated with oral anticoagulant prescription after multivariable adjustment in patients with atrial fibrillation and low thromboembolic risk, as defined by a CHADS2 (defined as 1 point each for congestive heart failure, hypertension, age 75 years or older, and diabetes mellitus, and 2 points for prior stroke, transient ischemic attack, or thromboembolism) score of 0 (A) and those with a CHA2DS2-VASc (defined as 1 point each for congestive heart failure, hypertension, age 65 to 74 years, diabetes mellitus, vascular disease, and female sex, and 2 points each for prior stroke, transient ischemic attack, or thromboembolism and age 75 years or older) score of 0 (B). Error bars denote 95% CIs. Body mass index is calculated as weight in kilograms divided by height in meters squared. RR indicates relative risk.
Hsu JC, Chan PS, Tang F, Maddox TM, Marcus GM. Oral Anticoagulant Prescription in Patients With Atrial Fibrillation and a Low Risk of ThromboembolismInsights From the NCDR PINNACLE Registry. JAMA Intern Med. 2015;175(6):1062-1065. doi:10.1001/jamainternmed.2015.0920
In patients with atrial fibrillation (AF) who are at risk for thromboembolism, anticoagulation therapy with warfarin or the newer novel anticoagulants reduces morbidity and mortality.1,2 Because oral anticoagulant use carries a risk of bleeding, the drugs are not recommended in patients with AF who are at a particularly low risk for stroke. Specifically, previous AF guidelines recommend against the use of oral anticoagulation in patients younger than 60 years without heart disease or other known risk factors for thromboembolism,3 and more recently updated guidelines do not recommend the use of oral anticoagulation in patients with AF without any established risk factor for stroke.4 We sought to examine the prevalence of oral anticoagulant prescription that does not adhere to the guidelines in young and healthy patients with AF who were at the lowest risk for thromboembolism, as well as the clinical predictors of this practice.
Of 1 711 326 patients enrolled in the National Cardiovascular Data Registry’s PINNACLE (Practice Innovation and Clinical Excellence) Registry between January 1, 2008, and December 30, 2012, a total of 359 315 (21.0%) had received a diagnosis of AF. Our final study cohort, derived from 76 cardiology practices from 287 different geographic office sites in 33 states, comprised 10 995 young (<60 years) and healthy patients with AF and no structural heart disease who were at low risk for thromboembolism. While all these patients by definition had a CHADS2 (defined as 1 point each for congestive heart failure, hypertension, age ≥75 years, and diabetes mellitus, and 2 points for prior stroke, transient ischemic attack, or thromboembolism) score5 of 0, we also performed a secondary analysis restricted to those with a CHA2DS2-VASc (defined as 1 point each for congestive heart failure, hypertension, age 65-74 years, diabetes mellitus, vascular disease, and female sex, and 2 points each for prior stroke, transient ischemic attack, or thromboembolism and age ≥75 years) score6 of 0. To investigate the independent associations of various characteristics with the outcome of oral anticoagulant prescription, we constructed hierarchical modified Poisson regression models adjusted for patient demographics and clinical characteristics. Certification to use these deidentified data was obtained from the University of California, San Francisco, Committee on Human Research.
In the cohort of patients with a CHADS2 score of 0 and the cohort of those with a CHA2DS2-VASc score of 0, a total of 2561 (23.3%) and 1787 (26.6%) patients with AF, respectively, were prescribed an oral anticoagulant. Demographics and clinical characteristics of patients in each cohort stratified by prescription of oral anticoagulants are shown in the Table. In both the cohort of patients with a CHADS2 score of 0 and the cohort of those with a CHA2DS2-VASc score of 0, patients with AF who were prescribed oral anticoagulants were older and more frequently insured by Medicare or were uninsured. Patients in both cohorts who were prescribed oral anticoagulation had a higher body mass index, and a greater proportion of patients received anticoagulation than not in the Northeast and West. Patients prescribed oral anticoagulants were less likely to have paroxysmal AF or to be current smokers. In only the cohort of patients with a CHADS2 score of 0, patients who were prescribed oral anticoagulants were more likely to be male and have dyslipidemia. In multivariable analysis of the cohort of patients with a CHADS2 score of 0 that assessed clinical predictors of oral anticoagulant prescription adjusted for clustering of patients within sites, older age (adjusted relative risk [RR], 1.48 per 10 years; 95% CI, 1.41-1.56, P < .001), male sex (adjusted RR, 1.34; 95% CI, 1.22-1.46, P < .001), higher body mass index (adjusted RR, 1.18 per 5 kg/m2; 95% CI, 1.14-1.22, P < .001), and Medicare compared with private insurance (adjusted RR, 1.32; 95% CI, 1.17-1.49, P < .001) were associated with a higher likelihood of being prescribed oral anticoagulants, whereas treatment in the South compared with the Northeast of the United States was associated with a lower likelihood of being prescribed oral anticoagulants (adjusted RR, 0.69; 95% CI, 0.49-0.98, P = .04) (Figure, A).
In multivariable analysis of the cohort of patients with a CHA2DS2-VASc score of 0, older age (adjusted RR, 1.44 per 10 years; 95% CI, 1.36-1.54, P < .001), higher body mass index (adjusted RR, 1.19 per 5 kg/m2; 95% CI, 1.15-1.23, P < .001), Medicare compared with private insurance (adjusted RR, 1.29; 95% CI, 1.13-1.47, P < .001), and no insurance compared with private insurance (adjusted RR, 1.19; 95% CI, 1.03-1.37, P = .02) were associated with a higher likelihood of being prescribed oral anticoagulants, whereas treatment in the South compared with the Northeast of the United States was associated with a lower likelihood of being prescribed oral anticoagulants (adjusted RR, 0.67; 95% CI, 0.47-0.96, P = .03) (Figure, B).
In a large, nationally representative sample of young (<60 years) and healthy outpatients with AF who were at the lowest risk of stroke treated by cardiovascular specialists, approximately 25% of patients were prescribed oral anticoagulant therapy, contrary to contemporary guideline recommendations. Specific patient characteristics predicted an increased likelihood of oral anticoagulant prescription. These findings may have important public health implications since young and healthy patients with AF who are at the lowest risk of stroke have an unfavorable risk-benefit profile when prescribed oral anticoagulants.
Our study has limitations. First, the PINNACLE Registry did not capture data on certain diagnoses, such as previous pulmonary embolism or deep vein thrombosis, which may have warranted use of oral anticoagulation independent of AF. Second, the PINNACLE Registry did not include procedural data regarding electrical cardioversion or catheter ablation for AF. Since oral anticoagulation is often administered for 1 to 2 months after these procedures, patients who recently underwent these procedures may have been categorized as taking oral anticoagulants despite the actual intention for the treatment to be only transient.
Because oral anticoagulants have the potential for substantial benefit and harm, decision making for health care professionals regarding stroke prophylaxis in patients with AF presents a unique clinical challenge. Prescription of oral anticoagulants by cardiovascular specialists in a significant proportion of patients at the lowest thrombotic risk suggests that these health care professionals may not be fully aware of the potential risks associated with oral anticoagulation or the particularly low risk of stroke in this population.
Corresponding Author: Gregory M. Marcus, MD, MAS, Cardiac Electrophysiology Section, Division of Cardiology, Department of Medicine, University of California, San Francisco, 505 Parnassus Ave, Room M1180B, San Francisco, CA 94143 (firstname.lastname@example.org).
Published Online: April 13, 2015. doi:10.1001/jamainternmed.2015.0920.
Author Contributions: Drs Hsu and Marcus had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Hsu, Marcus.
Acquisition, analysis, or interpretation of data: Hsu, Chan, Tang, Maddox.
Drafting of the manuscript: Hsu, Marcus.
Critical revision of the manuscript for important intellectual content: Hsu, Chan, Tang, Maddox.
Statistical analysis: Tang.
Study supervision: Hsu, Marcus.
Conflict of Interest Disclosures: Dr Hsu reported receiving honoraria from Medtronic. Dr Chan reported receiving a Career Development Grant Award K23HL102224 from the National Heart, Lung, and Blood Institute. Dr Maddox reported receiving a Health Services Research and Development Career Development Award from the US Department of Veterans Affairs. Dr Marcus reported receiving research support from Medtronic and SentreHeart Inc. No other disclosures were reported.
Funding/Support: This research was supported by the American College of Cardiology Foundation’s National Cardiovascular Data Registry (NCDR).
Role of the Funder/Sponsor: The PINNACLE (Practice Innovation and Clinical Excellence ) Registry is an initiative of the American College of Cardiology Foundation. Bristol-Myers Squibb and Pfizer Inc are founding sponsors of the PINNACLE Registry. The PINNACLE Registry and the NCDR had no role in the design and conduct of the study and management, analysis, or interpretation of the data. The manuscript was approved with minor editorial suggestions by the PINNACLE Registry Research and Publications Committee prior to submission.
Disclaimer: The views expressed in this article represent those of the authors and do not necessarily represent the official views of the NCDR or its associated professional societies identified at http://www.ncdr.com.