Comparing the Efficacy and Safety of Apixaban and Warfarin in Patients with Atrial Fibrillation and Advanced Chronic Kidney Disease

Introduction

A key part of the management of atrial fibrillation is anticoagulation to reduce the risk of stroke¹. Chronic kidney disease (CKD) is known to promote atrial fibrillation, possibly through an inflammatory pathway². While anticoagulation for atrial fibrillation has traditionally been done with warfarin, a vitamin K antagonist, it can also be accomplished via a direct oral anticoagulant (DOAC), including the factor Xa inhibitor apixaban. Recent studies have shown that a DOAC is noninferior to warfarin³. However, most studies have excluded people with advanced CKD. This retrospective observational study is, to our knowledge, the largest yet to compare effects of apixaban and warfarin use in patients with atrial fibrillation and advanced chronic kidney disease.

Methods

Admission encounters from a comprehensive, multi-institutional database were examined for patients diagnosed with atrial fibrillation and advanced CKD, defined as chronic kidney disease stage 3, 4, or 5, as well as end-stage renal disease. These patients were separated into cohorts taking either apixaban or warfarin. The rate of endpoint diagnoses, identified by International Classification of Diseases (ICD) codes, were compared between the two groups.

Results

The records of 14998 patients were collected people with a mean age of 76.9, including 11177 (74.5%) patients using apixaban and 3821 (25.4%) using warfarin. The two groups were not identical; the warfarin group was more likely to be in stage 3 chronic kidney disease compared to the apixaban group, which was more likely to be in end stage renal disease. The primary endpoint was the rate of stroke, and the secondary endpoints were rate of GI bleed, ocular hemorrhage, and intracranial hemorrhage. The apixaban cohort was associated with a higher rate of stroke (odds ratio 1.25, 95% confidence interval 1.05 - 1.4) than the warfarin group. Meanwhile, the warfarin cohort was more likely to develop a GI bleed compared to apixaban, as well as having a higher rate of all-cause mortality. Both groups were equally likely to develop intracranial or ocular hemorrhage.

Conclusions

In patients with chronic kidney disease who received warfarin instead of apixaban for atrial fibrillation, the risk of stroke was found to be lower but that of GI bleed to be higher. This may be in part because warfarin is renally excreted, while apixaban is excreted both by the kidney and hepatobiliary route. The higher likelihood of GI bleed and all cause mortality is concerning, possibly caused by the narrow therapeutic window of warfarin. These findings suggest that warfarin and apixaban have key differences in their risk profiles, and that treatment in patients with chronic kidney disease should be personalized. Patients with additional risk factors for stroke may consider warfarin. This is especially relevant because of the morbid nature of stroke relative to GI bleed. Further research should be done to determine if other stroke risk factors may be comorbid with CKD and be better treated by warfarin.

No relevant conflicts of interest to declare.