In dialysis patients, anticoagulation for AF may be cause for clinical equipoise, regardless of bleeding risk score. New research, reported here, may begin to shift the balance.
The CHADS2 or CHA2DS2-VASc risk scores can be used in most clinical settings to determine whether systemic anticoagulation for atrial fibrillation (AF) is indicated or not. However, patients with end-stage renal disease who are dialysis-dependent are particularly challenging to study and to treat. There are 3 primary reasons for this.
First, these patients are usually systematically excluded from clinical trials because of their dialysis dependence and resulting variability in the pharmacokinetics/pharmacodynamics of drug metabolism. Therefore, other than data from observational studies, little is known about the risk-benefit ratio of anticoagulation in this population. Second, the presence of renal disease and various comorbidities increases susceptibility to bleeding among these patients. Third, patients with chronic kidney disease and AF have a 3-fold higher risk for stroke. For these reasons, results of recent observational studies have been conflicting on whether the risk-benefit ratio of anticoagulation in end-stage renal patients with AF favors anticoagulation or not.
A provocative paper, published in Circulation this week, reported on a retrospective study of 1626 dialysis patients and 204,210 controls that aimed to determine the association between warfarin use and risk for stroke vs bleeding in these two patient populations over 9 years. Warfarin use was ascertained by prescriptions filled within 30 days of hospital discharge. Among the dialysis patients, 46% (756) filled prescriptions for warfarin within 30 days of hospital discharge and 54% were not receiving warfarin. Those receiving warfarin had less prior bleeding but more congestive heart failure. Patients were compared using adjusted Cox proportional hazards (adjusted for demographic factors and specific components of CHADS2 score [congestive heart failure, hypertension, diabetes, and history of stroke/TIA]).
The efficacy results were very surprising. In dialysis patients, warfarin use was not associated with a lower rate of stroke after adjustment for confounders (adjusted HR = 1.14; 95% CI, 0.78-1.67). With respect to bleeding, there was a 44% higher adjusted hazard for bleeding (95% CI, 1.13-1.85) in dialysis patients.
So, this well-conducted study provides somewhat convincing evidence that systemic anticoagulation may not prove to be beneficial in dialysis patients and is associated with increased harm. The study’s primary strength was its rigorous attempt to control for confounding variables and reproducible results, regardless of the method used to adjust for confounders. Although a randomized controlled trial would be the only way to definitively establish this risk-benefit relationship, this study provides convincing evidence that in dialysis patients, anticoagulation for AF may be cause for clinical equipoise, regardless of the CHADS2 score.
Shah M, Tsadok MA, Jackevicius CA, et al. Warfarin use and the risk for stroke and bleeding in patients with atrial fibrillation undergoing dialysis. Circulation. 2014 Jan 22. doi: 10.1161/ CIRCULATIONAHA.113.004777.