Managing Warfarin Therapy in Hemodialysis Patients with Atrial Fibrillation

The mainstay therapy to reduce the risk of stroke in patients with atrial fibrillation is warfarin. Thromboembolic risks are reduced with the use of target international normalized ratio (INR) between two and three without a significant increase in the risk of major bleeding.

Compared with the general population, the risk of thromboembolic events is higher in patients with advanced chronic kidney disease. There is no clear consensus on the benefit versus harm for use of anticoagulation in patients on hemodialysis. Some studies have shown a decrease in mortality and risk of stroke with warfarin; others have reported an increased risk of bleeding with no significant impact on risk of ischemic stroke or mortality. Further, guidelines from various national and international societies offer conflicting recommendations: some call for the use of warfarin in hemodialysis patients with atrial fibrillation, while others advise against using warfarin for primary thromboembolic prophylaxis in patients with atrial fibrillation on hemodialysis. In guidelines from the Kidney Disease Improving Global Outcomes, routine anticoagulation of patients with stage 5 CKD with atrial fibrillation for primary prevention of stroke is not indicated.

There is a correlation between time in therapeutic range (TTR), measured by the Rosendaal method, and bleeding and thromboembolic complications; there is also an association between a TTR of <60% and major bleeding, increased mortality, and systemic embolism. Limited data have shown that hemodialysis patients on warfarin have lower TTR than the general population.

Researchers in Canada, led by Hamad Bahbahani, MD, recently conducted a retrospective cohort study designed to compare nephrologist-led management of warfarin in terms of achieved TTR and frequency of INR testing with specialized thrombosis clinic-led management in a population of patients on hemodialysis. Study results were reported online in BMC Nephrology [doi:10.1186/s12882-017-0809-x].

The study included patients from the McGill University Health Centre, where anticoagulation with warfarin is managed by nephrologists (institution A), and the Jewish General Hospital, where warfarin-based anticoagulation is managed through an anticoagulation clinic led by hematologists (institution B). Both institutions are in Montreal, Quebéc, Canada. Primary end points were: (1) mean TTR, based on the Rosendaal method; (2) proportion of patients achieving TTR ≥60%; and (3) mean frequency of INR testing.

Between January 1, 2015, and November 26, 2016, the researchers identified 341 patients in institution A and 300 in institution B who were undergoing hemodialysis for at least 3 months. At institution A, 16.7% (n=57/341) had documented history of atrial fibrillation; 36.8% (n=21/300) of them were on warfarin at the end of the study period. At institution B, 54 patients (18%) had history of atrial fibrillation and 55.5% of those (n=30) were on warfarin. Following application of inclusion and exclusion criteria, the final TTR analysis included 20 patients in institution A and 30 patients in institution B.

Mean age was 75.6 years in the institution A group and 79.3 years in the institution B group; the cohorts were 60% male in both institutions. Distribution of comorbidities associated with increased risk of stroke such as previous cerebrovascular accident, type 2 diabetes mellitus, hypertension, congestive heart failure, and valvular atrial fibrillation were similar in the two groups, as was hemodialysis vintage. Duration of hospitalization of patients at institution A was shorter than that of patients at institution B (mean 10.9 days vs mean 20 days); the difference was not statistically significant.

At institution A, mean TTR based on the Rosendaal method was 61.8% compared with 60.5% at institution B. Mean frequency of INR testing was every 6 days at institution A compared with every 13.9 days at institution B.

At institution A, 65% of hemodialysis patients on warfarin achieved TTR ≥60% compared with 43.3% of patients at institution B who achieved that target level of TTR. Following adjustment for hemodialysis vintage, total duration of hospitalization, and total duration of outpatient antibiotic use, the odds ratio of patients achieving TTR ≥60% was 2.22 (95% confidence interval, 0.65-7.63).

At institution A, the average annual number of INR tests was 60, compared with 26 at institution B. At a total cost of 0.50$ (Canadian), the average annual cost of INR testing per patient at institution A was approximately $30.00 (Canadian), compared with a total per test cost of $12.50 (Canadian) for an average annual cost of INR testing at institution B, for an average annual cost of INR testing per patient of $338.00 (Canadian). Nephrologist-led warfarin management resulted in significant overall cost savings despite more frequent INR testing. (The cost analysis was based on laboratory and professional fees specific to the Province of Quebéc, Canada.)

The study had some limitations, including the relatively small sample size and the inclusion of only two dialysis centers, limiting the statistical power of the outcome measures and the generalizability to all hemodialysis patients; and the observational design of the study that may have allowed for the possibility of unmeasured bias.

In conclusion, the researchers said, “There was no statistical difference in mean TTR between nephrologist-led management of warfarin to that of anticoagulation clinic-led management. However, the former achieved a trend of higher proportion of patients with optimal TTR which was associated with more frequent monitoring and greater cost-effectiveness.”

Takeaway Points

  1. There is no clear consensus on benefit versus harm for use of warfarin in patients on hemodialysis with atrial fibrillation. Researchers in Canada conducted a retrospective cohort study to compare nephrologist-led management of warfarin therapy with that led by a specialized anticoagulation clinic.
  2. The study was conducted at two institutions in Montreal, Quebéc, Canada. The final analysis included 20 patients with nephrologist-led management and 30 with anticoagulation clinic-led management.
  3. There was no statistical difference between the two groups in time in therapeutic range (TTR); however, patients in the nephrologist-led group achieved a trend toward a higher proportion of patients with optimal TTR. There was also a significant overall cost savings associated with nephrologist-led warfarin management.