Anticoagulation in patients with atrial fibrillation (AF) who are on dialysis may cause harm, according to a systematic review and meta-analysis published in the American Journal of Cardiology.1
“In this systematic review and meta-analysis of 20 observational studies involving 31,321 patients with AF and end-stage renal disease on dialysis, we found unexpected associations of anticoagulation use with outcomes,” wrote first author Christopher Wong, MBBS, MSc, PhD, of the University of Oxford, and colleagues.
“[T]here were unfavorable or nonsignificant associations of anticoagulation use with any stroke, ischemic stroke, cardiovascular death, and all-cause mortality,” they continued, “Pooled event rates also revealed a substantial incidence of these outcomes in patients with AF on dialysis, and these were also inconsistent with any clear benefit of anticoagulation in this population.”
The results run counter to evidence from randomized trials in general populations supporting the benefits of anticoagulation in patients with AF. Those studies have suggested that the benefits of anticoagulation for decreasing the risk of ischemic stroke outweigh the risks of hemorrhagic stroke.2
While reviews on this subject do exist, until now no quantitative meta-analysis has been done. Moreover, randomized trials usually do not include dialysis patients, even though AF is not uncommon in such patients.
However, some observational studies have suggested that the benefits of anticoagulation regarding stroke risk may not extend to dialysis patients with AF. The large, prospective Danish Dialysis Outcomes and Practice Patterns Study (DOPPS) found increased risk of stroke in dialysis patients with AF on warfarin, especially elderly patients.3 That study did not include data on ischemic stroke, though retrospective analyses have not supported reductions in ischemic stroke with anticoagulation in this population, according to background information in the article.
The meta-analysis included twenty observational studies (3 prospective cohort studies, 17 retrospective cohort studies) of patients with AF on dialysis published up to June 2015. The analysis covered 529,741 participants, of whom 31,321 had AF and were on dialysis.
Key results for the use of anticoagulation in this population:
• Any type of stroke: 45% (95% CI 13% to 88%) increased risk
♦ Ischemic stroke: nonsignificant 13% (−4% to 34%) increased risk
♦ Hemorrhagic stroke: significant 38% (3% to 85%) increased risk
• Thromboembolism: 44% (95% CI 38% to 56%) lower risk
• Any bleeding: 31% (95% CI 12% to 53%) increased risk
• Cardiovascular death: no clear correlation (relative risk 0.99, 95% CI 0.86 to 1.15)
• All-cause mortality: no clear correlation (relative risk 0.97, 95% CI 0.90 to 1.04)
The small size, relatively short duration, and differences in methodology of included studies may have limited the meta-analysis, and raises the possibility of confounding. Significant between-study heterogeneity existed for analyses linking anticoagulation with ischemic stroke (P = 0.021) and hemorrhagic stroke (P = 0.008). Finally, most patients in the analysis were on hemodialysis, and the results may not extend to those on peritoneal dialysis.
The authors also stressed that the observational nature and lack of randomization in the included studies may further limit the analysis. However, given the clear benefits of anticoagulation in general populations, a plausible assumption would be that such benefits extend to patients on dialysis, they reasoned.
“Our findings support emerging concerns regarding the potentially unfavorable risk-benefit of anticoagulation for patients with AF on dialysis; the totality of observational evidence does not support the routine use of anticoagulation in this setting,” they concluded, “Until direct randomized evidence is available, individualized clinical decision-making must proceed on a case-by-case basis after a robust discussion of the potential risks and benefits. More importantly, however, there is a pressing need to generate reliable randomized data to resolve this area of uncertainty.”
• Meta-analysis of observational studies of patients with AF on dialysis failed to find any clear benefit of anticoagulation in this population.
• Results showed significant increases of any type of stroke, hemorrhagic stroke, and any bleeding.
• Risk of ischemic stroke was increased, though the results were not significant.
• A randomized trial is needed to confirm the link between anticoagulation and stroke in patients with AF on dialysis.
The authors report no conflicts of interest.
1. Wong CX, et al. Meta-analysis of anticoagulation use, stroke, thromboembolism, bleeding, and mortality in patients with atrial fibrillation on dialysis. Am J Cardiol. 2016 Jun 15;117(12):1934-9141.
2. Hart RG, et al. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med. 2007 Jun 19;146(12):857-867.
3. Wizemann V, et al. Atrial fibrillation in hemodialysis patients: clinical features and associations with anticoagulant therapy. Kidney Int. 2010 Jun;77(12):1098-1106.