Supplementary Material for: QTc interval prolongation independently associates with FGF23 and predicts mortality in predialysis CKD

Introduction: QTc interval prolongation is increasingly frequent as CKD advances and predicts death in dialysis. However, predictors and mortality-risk in predialysis CKD are understudied. FGF23 induces left ventricular hypertrophy (LVH) which is associated with QTc interval prolongation and death, suggesting a possible pathway from FGF23 to death that entails LVH and QTc prolongation. We looked for links between FGF23 and prolonged QTc intervals mediated by LVH, and for deaths associated with QTc prolongation in a prospective observational cohort of patients with predialysis CKD. Methods: Participants underwent protocolized baseline and semi-annual FGF23 testing, baseline and study end echocardiograms, and baseline and annual electrocardiograms over three years. Results: 2,254 participants (34.1% female; mean age 68.7 years; mean glomerular filtration rate 41.4 ml/min/m2) enrolled. Baseline LVH (left ventricular mass index >131 g/m2 (>100 g/m2 if female)) was present in 10.8% and prolonged QTc intervals (>=500 ms) in 1.5%. One hundred thirty-eight (6.1%) participants died during the study. In generalized mixed effects regression, each unit increase in the natural log of FGF23 - but not LVH - predicted an odds ratio of 1.76 (1.15, 2.70, p=0.009) for prolonged QTc intervals independently of 15 other covariates. Mediation analysis showed that only 13% of FGF23's total effect on prolonged QTc intervals was mediated by LVH. Patients with prolonged QTc intervals had higher unadjusted (log rank p