Abstract
I read with great interest the article by Marcos et al. investigating observational associations of urine albumin measures either by a simple urine albumin test or a more complex 24-h urine collection test with incident atrial fibrillation (AF) in an apparently healthy general population cohort enriched with individuals having an urine albumin concentration >10 mg/L at baseline.1 In line with previously published epidemiological studies, the authors report a positive association between elevated urine albumin and an increased risk of incident AF. While the associations were independent of common and novel risk factors in prospective observations, but a possibility of unmeasured confounding or reverse causality cannot be ruled out. As the authors and others noted a complementary strand of evidence from a randomized clinical trial or its analog, called...