Coumarins and survival in incident dialysis patients.

BACKGROUND The benefit and risk of oral anticoagulation in dialysis patients are debated controversially. METHODS We prospectively followed 235 dialysis patients of the INVOR Study (Study of Incident Dialysis Patients in Vorarlberg) for up to 7 years and analysed the prevalence and incidence of atrial fibrillation (AF) and the impact of coumarin therapy on survival. Oral anticoagulation was monitored frequently. RESULTS A total of 748 person-years were recorded with a median follow-up of 2.84 years. Twelve patients (5.1%) had AF at the start of dialysis. During follow-up, 40 patients (17.0%) developed AF, representing an incidence of 5.85 per 100 person-years. AF was positively associated with mortality (P = 0.004). Forty-six (19.6%) of the 235 patients were treated with coumarins. The majority (93.7%) had a clear indication for oral anticoagulation. In 65% of our patients, AF was the indication for coumarins. Patients without coumarins and without AF represented our reference group. The mortality risk of the coumarin-treated patients with AF or an alternative indication for coumarins was slightly lower compared to the reference group [hazard ratio (HR) 95% confidence interval (CI): 0.80 (0.28-2.29), P = 0.679 and 0.42 (0.16-1.10), P = 0.078, respectively]. No patient under sufficient oral anticoagulation experienced a stroke or a fatal bleeding event. Patients with AF and a contraindication for coumarins had a significantly higher mortality risk compared to the reference group [HR (95% CI): 3.90 (2.16-7.04), P < 0.001]. CONCLUSIONS Our data suggest that coumarins might be less harmful than previously anticipated when clearly indicated and closely monitored.