Estimated glomerular filtration rate within the normal or mildly impaired range and incident non-valvular atrial fibrillation: Results from a population-based cohort study.
European Journal of Preventive Cardiology, 2017. 24(2): pp. 213-222.
Eisen A, Haim M, Hoshen M, Balicer RD, Reges O, Leibowitz M, Iakobishvili Z, Hasdai D.
Background: Lower estimated glomerular filtration rate, in particular in the significant renal impairment range (estimated glomerular filtration rate <60 ml/min/1.73 m2), is associated with incident atrial fibrillation. This association is less established within the normal or mildly impaired estimated glomerular filtration rate range.
Methods: Using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) estimated glomerular filtration rate formula, we identified ambulatory adults (>22 years old) without rheumatic heart disease or prosthetic valves and with 60 ml/min/1.73 m2<estimated glomerular filtration rate<130 ml/min/1.73 m2 in their index visit, for incident, newly-diagnosed atrial fibrillation. We analyzed cohorts with and without prior cardiovascular disease.
Results: Over a mean follow-up of 104 months and >10 m patient-years of follow-up (∼75% <60 years old, ∼57% females), >65,000 individuals had ≥1 atrial fibrillation event (incident atrial fibrillation rate 5.1% and 5.8% excluding or including prior cardiovascular disease, or 49 and 55 per 10,000 patient-years, respectively). In both cohorts, individuals with versus without incident atrial fibrillation had lower mean estimated glomerular filtration rate (∼83 versus 95 ml/min/1.73 m2). Adjusting for age, gender, hypertension, and diabetes mellitus, overall a 10 ml/min/1.73 m2 decrease in estimated glomerular filtration rate was independently associated with a mean increase in incident atrial fibrillation of 1.5% and 2.4% in the cohorts excluding or including prior cardiovascular disease, respectively ( p < 0.001 for both). However, a graded association between lower estimated glomerular filtration rate and atrial fibrillation was observed in the 90-130 ml/min/1.73 m2 range, whereas a blunted association was observed in the 60-90 ml/min/1.73 m2 range.
Conclusion: Within the 60 ml/min/1.73 m2 < estimated glomerular filtration rate < 130 ml/min/1.73 m2 range, lower estimated glomerular filtration rate is independently associated with incident non-valvular atrial fibrillation in adults without prior atrial fibrillation, mainly attributed to a graded association within the 90-130 ml/min/1.73 m2 range.