Atrial fibrillation and anticoagulation are associated with hospitalisations in patients with end-stage kidney disease on haemodialysis: a prospective population-based cohort study

BACKGROUND: Patients with end-stage kidney disease on haemodialysis suffer from frequent complications requiring hospitalisation. Atrial fibrillation is a burdensome comorbidity amongst patients on haemodialysis. We aimed to assess frequency, reasons, and duration of hospitalisations in haemodialysis patients and their association with atrial fibrillation and anticoagulation. METHODS: Prevalent patients with end-stage kidney disease on haemodialysis were recruited into a prospective cohort study and observed for a median observation time of 3.4 years. Hospitalisations were recorded from discharge letters, medical records, and patient interviews. The association of atrial fibrillation, anticoagulation, and time-in-therapeutic range of vitamin K antagonist treatment with hospitalisations was analysed using negative binomial regression. RESULTS: Out of 625 patients, 238 (38.1%) had atrial fibrillation. Median number of hospitalisations per patient was 3.0 (1.0–5.0). Incidence rate of hospitalisation was 1.7 per patient-year in all and 1.9 in atrial fibrillation patients, median duration per hospitalisation was 7.9 (4.8–12.9) and 8.8 (5.7–13.3) days, respectively. Most frequent reasons for hospitalisation were vascular access complication/intervention (11.7%) and infection/fever (11.4%), while bleeding events comprised 6.0% of all hospitalisations. Atrial fibrillation patients had 27% higher risk of hospitalisation than patients without atrial fibrillation (incidence rate ratio [IRR] 1.27, 95% confidence interval [CI] 1.10–1.47). In atrial fibrillation patients, anticoagulation (enoxaparin or phenprocoumon, 41.6% of AF patients) was associated with increased risk of all-cause (IRR 1.38, 95%CI 1.14–1.69) and bleeding-related hospitalisation (IRR 1.96, 95%CI 1.06–3.63). There was no association between anticoagulation and stroke-related hospitalisation. In atrial fibrillation patients on phenprocoumon, increasing time-in-therapeutic range was associated with decreased risk of all-cause (IRR 0.35, 95%CI 0.14–0.87), but not bleeding-related hospitalisation (IRR 0.13, 95%CI 0.01–1.38). CONCLUSION: In haemodialysis patients, presence of atrial fibrillation and, among those with atrial fibrillation, anticoagulation were associated with higher risk of all-cause hospitalisation, including bleeding-related hospitalisation in the latter. Increasing time-in-therapeutic range in patients on vitamin K antagonist treatment was associated with decreased risk of all-cause, but not bleeding-related hospitalisation.