Author + information
- Richard W. Asinger, MD∗ ( and )
- Gautam R. Shroff, MBBS
- Division of Cardiology, Department of Medicine, Hennepin County Medical Center and the University of Minnesota, Minneapolis, Minnesota
- ↵∗Reprint requests and correspondence:
Dr. Richard W. Asinger, Hennepin County Medical Center, Division of Cardiology, 701 Park Avenue South, Orange 5, Minneapolis, Minnesota 55415.
For over a half century, vitamin K antagonists, chiefly warfarin, were the exclusive oral anticoagulants available for long-term anticoagulation. Being “the only game in town,” the emphasis of the accompanying clinical research was focused on determining the most appropriate method to measure anticoagulant effects, define the most efficacious and safe target range for anticoagulation, and identify strategies to maintain and reverse therapeutic anticoagulation. This emphasis came at the expense of turning a blind eye to rare concerns raised about the potential for warfarin to cause or worsen renal dysfunction (1,2). The approval of dabigatran for prevention of stroke and systemic embolism in patients with nonvalvular atrial fibrillation (NVAF) ushered in a new era in long-term oral anticoagulation (3). Progress in this era has been unprecedented in recent years, and there are now several newer oral anticoagulant (NOAC) options approved by the U.S. Food and Drug Administration for clinical use (4). Interestingly, clinicians now find themselves in the relatively uncomfortable position of having to choose the “best” option among anticoagulants for their patients. The selection of the most appropriate oral anticoagulant can be a complex, multistep process that is based on the consideration of several clinical variables including, importantly, the assessment of renal function (4).
In this issue of the Journal, the study by Böhm et al. (5) is, therefore, timely and renews previously raised concerns about the detrimental effects of vitamin K antagonists on renal function. In this unique post -hoc analysis of the RE-LY trial, the authors longitudinally assessed estimated glomerular filtration rate (eGFR) among 16,490 patients assigned to therapy with dabigatran 110 mg twice daily, dabigatran 150 mg twice daily, or adjusted-dose warfarin. Although the reduction in eGFR was numerically higher among the dabigatran arms in the short term, after an average follow-up of 30 months, there was a modest but statistically significant reduction in eGFR in the warfarin group compared with the dabigatran arms. The authors appropriately acknowledge that the decline in eGFR with warfarin in comparison with dabigatran was small despite the long follow-up period in this large trial. Additional sensitivity analysis, based on time to deterioration of eGFR >25% (especially during the later phases of the trial), also supported the primary analysis. This study has several strengths including robust, albeit complex and exhaustive, statistical analyses, a thoughtful and deliberate choice of tools (Chronic Kidney Disease Epidemiology Collaboration equation for eGFR estimation), and clinical parameters (25% reduction in eGFR) that are meaningful to the clinician.
Importantly, the findings reported are directionally concordant with previous reports of nephropathy described among patients receiving warfarin therapy, hypothesized to occur from the effects of inhibition of the vitamin K–dependent protein gamma carboxyglutamic acid (Gla/MGP) and/or a specific nephropathy from episodic glomerular hemorrhage (1,2). The observations pertaining to greater reductions in eGFR among patients with previous warfarin use and supra-therapeutic international normalized ratio values could suggest a “dose” effect, lending credence to the latter theory. It is, therefore worth noting that the decline in eGFR in this study was similar in both dabigatran arms, even though the overall incidence of major hemorrhage in RE-LY was higher in the dabigatran 150 mg twice-daily arm (3). If glomerular hemorrhage was indeed the sole mechanism responsible for decline in eGFR with oral anticoagulation, one might have expected a similar trend with the higher dabigatran dose, but this was not observed. Thus, an alternative mechanism responsible for the reduction in eGFR in the warfarin group, such as renal vascular calcification (2), might be hypothesized as the predominant mechanism, although this is purely speculative. Owing to the retrospective nature of the analysis, it is not feasible to account for all relevant clinical variables, such as concomitant medications or procedural interventions (such as coronary angiography) that may contribute to renal dysfunction. Overall, the authors are to be congratulated for the innovative use of data from a large, randomized trial on long-term oral anticoagulation to address a theoretical and occasionally raised question with important clinical implications. Although it would be desirable to confirm these findings in a prospective fashion, the extraordinarily large number of patients from the RE-LY trial needed to demonstrate these findings highlight the fact that any absolute reduction in eGFR with warfarin is really quite modest compared with both doses of dabigatran. Thus, even though these observations probably are not a “game changer” for clinicians, fine-tuning of clinical practice would be appropriate with attention to these findings.
From a somewhat broader perspective, this study brings into light the observation of a continuous decline in eGFR among all three arms in this elderly cohort of patients with NVAF receiving therapeutic anticoagulation. Although this message is not necessarily novel, it is extremely pertinent in the context of escalating use of NOACs worldwide and implications for dose adjustments to avoid drug accumulation and excessive hemorrhage. In the absence of a placebo arm, it is not possible to determine if a decline in eGFR occurs among all patients with NVAF, simply among patients receiving therapeutic anticoagulation, or only among patients being anticoagulated with specific agents. Although the clinical significance of this reduction in eGFR (approximately 1 ml/min per year) can be debated, the overall message is compelling and deserving of careful thought. Deterioration of renal function among patients on warfarin does not have direct implications for dose adjustments, but it certainly would have implications for patients beginning therapy with or transitioning therapy to NOACs, with important attendant implications for bleeding risk from accumulating levels of these agents in the case of deteriorating eGFR.
In summary, for many decades, the medical community chose to ignore some of the pathophysiological concerns pertaining to the effect of warfarin on renal vascularity and function because warfarin was the only viable option. By virtue of the availability of several options for therapeutic anticoagulation now, we can finally permit ourselves to take those self-imposed blinders off and take a more critical look at the clinical effects of these drugs. The study by Böhm et al. (5) raises provocative questions about the effects of pharmacotherapy for atrial fibrillation on renal function but also serves to reassure that the observed decline in eGFR with warfarin does not appear to be a very heavy price to pay for its proven benefits. Other oral anticoagulant agents also should be evaluated for any association with a similar phenomenon of eGFR reduction because the results could influence dosing. For now, warfarin will continue to have an important role in contemporary therapeutic anticoagulation for NVAF for the simple reason that its anticoagulant effect can be easily measured clinically, and this may be especially desirable for select patient populations, including, ironically, patients with fluctuating renal dysfunction. We hope that studies like this provide an impetus to the cardiology/nephrology and pharmaceutical communities to design comparative effectiveness studies among the NOAC agents, particularly among select subsets such as patients with chronic kidney disease and diabetes. On the basis of the burgeoning epidemiology and clinical importance of renal dysfunction among patients with NVAF, it is clear that longitudinal assessment of eGFR would be an important endpoint in such studies.
↵∗ 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. Asinger is a member of the Data and Safety Monitor Board for the Watchman Trials, Boston Scientific, Plymouth, Minnesota. Dr. Shroff has reported that he has no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Kovacs R.J.,
- Flaker G.C.,
- Saxonhouse S.J.,
- et al.
- Böhm M.,
- Ezekowitz M.D.,
- Connolly S.J.,
- et al.