By Michael H. Crawford, MD, Editor
SOURCES: Böhm M, et al. Changes in renal function in patients with atrial fibrillation: An analysis from the RE-LY trial. J Am Coll Cardiol 2015;65:2481-2493.
Asinger RW, Shroff GR. Atrial fibrillation and renal function: How high is the price of anticoagulation? J Am Coll Cardiol 2015;65:2494-2495.
Vitamin K antagonists (VKAs) have been shown to increase vascular calcification. Thus, VKAs may accelerate vascular disease in organs such as the kidneys. These effects are not seen with the new oral anticoagulants. These investigators analyzed the Randomized Evaluation of Long Term Anticoagulation Therapy (RE-LY) trial database to assess for changes in renal function during treatment with VKAs or the direct thrombin inhibitor dabigatran in patients with atrial fibrillation (AF). The RE-LY trial evaluated two doses of dabigatran and warfarin therapy in AF patients with at least one other stroke risk factor and followed them for a median of 2 years. Patients with an estimated glomerular filtration rate (GFR) < 30 mL/min were excluded. More than 18,000 patients were randomized to the three treatments in this trial. Creatinine was measured at baseline, 3, 6, and 12 months and then yearly. GFR was estimated for up to 30 months. Baseline clinical characteristics did not differ significantly between groups, and 19% of the patients had GFRs between 30 to 50 mL/min. A gradual decline in renal function was observed over time in the total patient population. At an average follow-up of 30 months, the mean decline in GFR was significantly greater with warfarin (-3.68 mL/min) vs dabigatran at 110 mg/day (-2.57, P = 0.0009) and dabigatran 150 mg/day (-2.46, P = 0.0002). A > 25% decline in GFR was less likely with both dabigatran doses (hazard ratio 0.79; 95% confidence interval [CI], 0.68-0.93, with 150 mg and 0.81, CI 0.69-0.96 with 110 mg). A more rapid decline in GFR was associated with prior warfarin use, poor INR control (in therapeutic range < 65%), and with diabetes. The authors concluded that patients with AF taking oral anticoagulants exhibited a decline in renal function that was greatest with warfarin use vs dabigatran use.
This post-hoc analysis of the RE-LY trial focuses attention on an often ignored observation that patients with AF on warfarin can experience a progressive decline in renal function. Prior studies have shown that this “warfarin nephropathy” paradoxically increases mortality and stroke risk. The RE-LY trial presented a unique opportunity to look at the renal effect of warfarin vs dabigatran. The study showed a modest but statistically significant decrease in renal function over an average follow-up of 24 months, with the 30-month data in a large subgroup that showed it was worse with warfarin. The accompanying editorial emphasizes that renal function needs to be monitored periodically in AF patients on anticoagulants, and in those on the new oral agents, doses may need to be adjusted accordingly.
Without a placebo group, it is difficult to know if older patients with AF and not on oral anticoagulants would also experience a decline in renal function. If the renal decline is not related to anticoagulants then at the least warfarin seems to accelerate or worsen it. Several mechanisms have been proposed to explain this potential adverse effect of long-term warfarin use. VKAs block vitamin K-dependent protein gamma carboxyl-glutamic acid formation, which inhibits bone morphogenetic protein. This effect of VKAs leads to vascular calcification and increased plaques. In this study, the adverse effects of warfarin on renal function were greater in diabetics and in those with INRs above the therapeutic range, which would support this mechanism. In addition, factor VII is antagonized by warfarin, which can lead to plaque destabilization. On the other hand, dabigatran, which inhibits factor Xa, can reduce vascular inflammation. Also, vitamin K antagonism affects the clotting cascade higher up than dabigatran. This could lead to renal hemorrhage, which has been observed on renal biopsy. However, the high-dose dabigatran group did not show worse renal function, which argues against the renal hemorrhage theory.
Although this is a post-hoc analysis of a trial done for different reasons and the 30-month data is a subset of the study population, the large size of the study is a strength. There are several weaknesses. There are no data on other drug therapy, such as angiotensin-converting enzyme inhibitors, which could affect renal function. Also, the study is of a relatively short duration, and the effects on renal function are modest. Should we quit using warfarin? Probably not yet, since we have decades of experience with this drug, and it does prevent strokes. Also, it does not need to be adjusted for renal function and can be used in chronic kidney disease patients. Finally, there are effective antidotes to warfarin effects for episodes of major bleeding.