Management of Atrial Fibrillation in Chronic Kidney Disease Patients

Abstract & Commentary

By John P. DiMarco, MD, PhD

Professor of Medicine, Division of Cardiology, University of Virginia, Charlottesville

Dr. DiMarco does research for Medtronic, is a consultant for Medtronic, Novartis, and St. Jude, and is a speaker for Boston Scientific.

Source: Olesen JB, et al. Stroke and bleeding in atrial fibrillation with chronic kidney disease. N Engl J Med 2012;367:625-635.

Patients with chronic kidney disease or who have had renal replacement therapy have often been excluded from trials of anticoagulation in atrial fibrillation (AF). In this study, the authors analyzed data from hospitalized patients with nonvalvular AF in Denmark from 1997 through 2008. Patients with chronic kidney disease with and without renal replacement therapy were identified from a national registry. Pharmacotherapy was determined based on prescription data in the national patient registry. Stroke and bleeding risks were estimated using the CHA2DS2-VASC score and the HAS-BLED score, respectively. The primary outcomes in the study were hospitalization or death from stroke or systemic thromboembolism, major bleeding, myocardial infarction, and death from any cause. The analysis used a Cox proportional-hazards model with adjustment for changes in renal status or antithrombotic treatment during follow-up. The analysis was time dependent in that the renal status of patients could change during the course of follow-up if they developed chronic renal disease or began renal replacement therapy during follow-up.

During the course of the study, 146,251 patients were discharged from Danish Hospitals with nonvalvular AF. Of these, 13,879 were excluded for various reasons. In the final cohort of 132,372 patients, 96.6% had no renal disease at baseline, 2.7% had non-end-stage prior kidney disease, and 901 patients were receiving renal replacement therapy. During the course of the trial, an additional 4538 patients received a diagnosis of non-end-stage chronic kidney disease and renal replacement therapy was begun in an additional 477 patients. Among the patients with non-end-stage chronic kidney disease at baseline, 228 went on to receive renal replacement therapy during follow-up. Most patients on renal replacement therapy were receiving hemodialysis (77.9%), while 15.4% received peritoneal dialysis and 6.7% underwent kidney transplantation.

The rates of stroke or thromboembolism were increased in patients with chronic kidney disease. Among the patients with no renal disease, the event rate per 100 patient years was 3.61 compared to a rate of 6.44 in those with non-end-stage chronic kidney disease and 5.61 in those on renal replacement therapy. Bleeding rates were also increased. These were 3.54 per 100 patient years in those with no renal disease, 8.77 in patients with non-end-stage chronic kidney disease, and 8.89 in patients on renal replacement therapy. The rates for myocardial infarction and all-cause mortality were increased about three-fold. When the rates for strokes or systemic thromboembolism were adjusted for baseline risk factors, chronic kidney disease was still independently associated with an increase in risk. The adjusted hazard ratios for patients with non-end-stage chronic kidney disease was 1.49 (95% confidence interval [CI], 1.38 to 1.59; P < 0.001) and 1.83 (95% CI, 1.57 to 2.14; P < 0.001) for those on renal replacement therapy.

Warfarin decreased the risk of stroke in all groups. The hazard ratio for warfarin therapy compared to no anticoagulant therapy was 0.59 in those with no renal disease, 0.84 in those with non-end-stage chronic kidney disease, and 0.44 in those on renal replacement therapy. Similar reductions in stroke risks were seen when both warfarin and aspirin were prescribed. Aspirin alone did not decrease, but slightly increased the risk of stroke in patients with no renal disease or non-end-stage chronic kidney disease.

Warfarin increased the bleeding risks in all three groups. In the proportional hazard model, the hazard ratios for bleeding were 1.28 in those with no renal disease, 1.36 in those with non-end-stage chronic kidney disease, and 1.27 in those requiring renal replacement therapy. Bleeding rates were also increased when aspirin alone was used and almost doubled in patients who received both warfarin and aspirin.

The authors note that chronic kidney disease in patients with AF is associated with increased risks for stroke, systemic thromboembolism, and bleeding. Warfarin significantly reduces the risk of systemic thromboembolism at the cost of increasing bleeding risks. They conclude that initiation of warfarin therapy requires careful thought in patients with chronic kidney disease, since the risks and benefits of therapy may well be equal.

Commentary

In this paper, we see that in renal disease patients with nonvalvular AF, the rates for thromboembolic events off therapy and bleeding events on therapy are essentially equal. When this situation occurs, it’s hard to write a firm guideline that will apply to all patients in a class. We therefore need to look at each patient’s individual risk scores before making a decision. I also usually factor in that a stroke more often results in death or permanent disability than does non-CNS bleeding. Therefore, my usual first presumption is to start anticoagulation while monitoring closely for bleeding. Only in patients with a very high bleeding risk will I not make at least one attempt at oral anticoagulant therapy.