Author + information
- Sean D. Pokorney, MD, MBA∗ (, )
- Eric D. Peterson, MD, MPH and
- Jonathan P. Piccini, MD, MHS
- Division of Cardiology, Duke Clinical Research Institute, Duke University Medical Center, Durham, North Carolina
- ↵∗Address for correspondence:
Dr. Sean D. Pokorney, Division of Cardiology, Duke Clinical Research Institute, Duke University Medical Center, 2301 Erwin Road, DUMC 3845, Durham, North Carolina 27710.
- atrial fibrillation
- non–vitamin K antagonist oral anticoagulant
- renal function
Patients with atrial fibrillation (AF) have a 5-fold higher risk of stroke than patients without AF (1). Fortunately, oral anticoagulation reduces the risk of stroke in patients with AF by more than two-thirds and reduces all-cause mortality (2,3). Nearly 72,000 patients with AF have been evaluated for stroke prevention in randomized clinical trials comparing non–vitamin K antagonist oral anticoagulants (NOACs) with vitamin K antagonists, demonstrating that NOACs are safer and, potentially, more effective than vitamin K antagonists at decreasing stroke or systemic embolism (3).
There is concern that in clinical practice neither warfarin nor NOACs are dosed optimally, as specified by the product labels. Approximately one-third of patients on warfarin have international normalized ratio values outside of the therapeutic range, with subtherapeutic international normalized ratio values being more common than supratherapeutic values (4). Warfarin underdosing has also been associated with a higher overall risk for stroke and, in particular, more severe debilitating stroke (5). In a recent large series, 14% of those with AF and a stroke were on warfarin yet had a subtherapeutic international normalized ratio at the time of their stroke (6).
Unlike warfarin, NOAC therapy does not require drug-level monitoring; however, the medications still need to be adjusted for certain clinical factors, such as age, weight, and renal function. Data from the ORBIT II (Outcomes Registry for Better Informed Treatment of Atrial Fibrillation phase II) U.S.-based registry found that about 13% of NOAC patients were incorrectly dosed, with most receiving underdosing (7). Additionally, 25% of global prescriptions for apixaban were for the low (2.5 mg) dose (8), whereas in the ARISTOTLE (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation) trial, fewer than 5% of patients met criteria for low-dose apixaban (8).
In this issue of the Journal, Yao et al. (9) further demonstrated that a substantial proportion of patients with AF who were treated with NOAC therapy were dosed inconsistently with product labeling. Using claims and laboratory data from OptumLabs Data Warehouse, Yao et al. (9) identified nearly 15,000 patients with AF who started apixaban, dabigatran, or rivaroxaban for stroke prevention. Similar to the ORBIT study, they found that 12% of all NOAC-treated patients were inappropriately treated with the lower-dose NOAC, whereas another 4% were inappropriately overdosed. Within a propensity-matched analysis of patients with criteria for renal adjustment of NOAC therapy, those who were overdosed had increased risk for major bleeding (hazard ratio: 2.19; 95% confidence interval: 1.07 to 4.46) and similar risks for stroke or systemic embolism (hazard ratio: 1.66; 95% confidence interval: 0.40 to 6.88) relative to appropriately dosed patients. In contrast, among patients without criteria for renal adjusted dosing, overall event rates for major bleeding and stroke or systemic embolism were similar for appropriately and underdosed patients; yet, apixaban-treated patients who received underdosing had higher rates of stroke or systemic embolism (hazard ratio: 4.87; 95% confidence interval: 1.30 to 18.26) relative to those who received appropriate dosing. Finally, multivariable modeling showed that older age was associated with inappropriate underdosing in patients without renal dosing indications.
This important and well done study by Yao et al. (9) had many strengths, including the large number of patients with AF recently started on a NOAC. Moreover, the population was relatively unselected, although all patients had UnitedHealthcare insurance. Also, all medication data were accumulated from insurance claims for medication prescription fills, so outcomes data were only evaluated for time periods covered by a medication fill.
The study did had some limitations. Weight was not available in the claims data, so the weight-based criteria for apixaban dosing could not be accounted for (in which weight ≤60 kg is 1 of 3 dose reduction criteria for apixaban). Also, because weight was not available, the authors used the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation to calculate estimated glomerular filtration rates instead of the Cockroft-Gault equation for estimated creatinine clearance, which was the formula used in the clinical trials and the product labels for dabigatran and rivaroxaban. With regard to this, the authors noted that CKD-EPI has been shown to be more accurate than Cockroft-Gault in assessing renal function; however, data from ORBIT found significant differences between CKD-EPI and Cockroft-Gault estimates of renal function (10). Although CKD-EPI and Cockroft-Gault found estimated that one-quarter of patients with AF in ORBIT had estimated renal function <50 ml/min, approximately 40% of patients were classified differently relative to the 50 ml/min cutoff, depending on whether CKD-EPI or Cockroft-Gault was used (10).
The authors also noted that the number of adverse events seen was low, which may have limited their power to detect an association between dosing and risks for bleeding or embolic events. For example, there were only 9 and 30 strokes or thromboembolic events in the patients with and without indications for renal dosing, respectively. Similarly, the analysis showing that underdosed apixaban was associated with higher rates of stroke was based on 7 events. The short follow-up (median, <4.0 months) contributed to the small number of events and highlighted the low rates of NOAC persistence seen in clinical practice (11).
These limitations aside, the current analysis also highlighted potential reasons for inappropriate dosing of NOACs. They found that older-age patients were particularly prone to being underdosed with NOACs. These data stand in contrast to the large NOAC trials that found age alone was not a reason to dose-reduce therapy (12).
Medicine is both a science and an art. Guidelines and prescribing recommendations often have to be tailored to the needs of an individual patient. The data from Yao et al. (9) suggest that both overdosing and underdosing of NOACs may be associated with some risks. Although the desire to avoid bleeding is a natural inclination, this study demonstrated that reduced dosing without a formal indication does not lead to improved outcomes and may be hazardous. Although less may be more in various aspects of life, less NOAC does not provide more to patients who should receive a standard dose, and underdosing may predispose patients to higher and fully avoidable risks.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Dr. Pokorney has received modest consulting support from Bristol-Myers Squibb, Boston Scientific, and Medtronic; modest research support from Gilead, Boston Scientific, Bristol-Myers Squibb, and Janssen Pharmaceuticals; and significant research support from the Food and Drug Administration. Dr. Peterson has received significant research grant support from Eli Lilly & Company, Genentech Inc., Janssen Pharmaceuticals, Inc., and the American Heart Association; and modest consultant/advisory board support from Boehringer Ingelheim, Bristol-Myers Squibb, Janssen Pharmaceuticals, Inc., Pfizer, and Genentech Inc. Dr. Piccini has received grants for clinical research from ARCA biopharma, Boston Scientific, Gilead Sciences, Johnson and Johnson, Spectranetics, and St. Jude Medical; and serves as a consultant to Amgen, Allergan, Johnson and Johnson, Parexel, Medtronic, and Spectranetics.
- 2017 American College of Cardiology Foundation
- Xian Y.,
- O'Brien E.C.,
- Liang L.,
- et al.
- Steinberg B.A.,
- Shrader P.,
- Thomas L.,
- et al.
- Weitz J.I.,
- Eikelboom J.W.
- Yao X.,
- Shah N.D.,
- Sangaralingham L.R.,
- Gersh B.J.,
- Noseworthy P.A.
- Pokorney S.D.,
- Shrader P.,
- Thomas L.,
- et al.
- Jackevicius C.A.,
- Tsadok M.A.,
- Essebag V.,
- et al.
- Alexander J.H.,
- Andersson U.,
- Lopes R.D.,
- et al.