BACKGROUND AND OBJECTIVES: Hyperphosphatemia, vitamin D deficiency, hyperparathyroidism, and high serum fibroblast growth factor 23 (FGF23) levels, when studied separately, were found to predict the progression of CKD. However, studies with simultaneous measurement of mineral bone disorder (MBD)-related factors were scarce. This study aimed to identify factors predicting renal outcome independent of other factors. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: This was a prospective cohort study of 738 Japanese predialysis outpatients in the nephrology departments of two hospitals. The outcome was defined as a doubling of serum creatinine or initiation of dialysis. RESULTS: Mean estimated GFR (eGFR) was 35 ml/min per 1.73 m(2). At enrollment, the increase in intact FGF23 with decreasing eGFR was the earliest among changes in MBD-related factors, followed by 1,25-dihydroxyvitamin D decrease, parathyroid hormone increase, and phosphate increase. During a median duration of 4.4 years, 213 patients reached the endpoint. In a multivariable Cox model, high FGF23 and low 25-hydroxyvitamin D (25D) levels were the only MBD-related factors associated with a higher risk of renal endpoint (adjusted hazard ratio [95% confidence interval] per unit change of log FGF23 and 10 ng/ml of 25D: 1.83 [1.28-2.61] and 0.61 [0.41-0.90], respectively). There was no significant interaction between 25D and FGF23 (P=0.11). Active vitamin D therapy, serum phosphate, 1,25-dihydroxyvitamin D, and parathyroid hormone levels were not related to the renal endpoint. Treating death as a competing risk or multiple imputation for missing values yielded similar results. CONCLUSIONS: Combined use of two markers is useful for the risk stratification of renal outcome.
BACKGROUND AND OBJECTIVES:Hyperphosphatemia, vitamin D deficiency, hyperparathyroidism, and high serum fibroblast growth factor 23 (FGF23) levels, when studied separately, were found to predict the progression of CKD. However, studies with simultaneous measurement of mineral bone disorder (MBD)-related factors were scarce. This study aimed to identify factors predicting renal outcome independent of other factors. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: This was a prospective cohort study of 738 Japanese predialysis outpatients in the nephrology departments of two hospitals. The outcome was defined as a doubling of serum creatinine or initiation of dialysis. RESULTS: Mean estimated GFR (eGFR) was 35 ml/min per 1.73 m(2). At enrollment, the increase in intact FGF23 with decreasing eGFR was the earliest among changes in MBD-related factors, followed by 1,25-dihydroxyvitamin D decrease, parathyroid hormone increase, and phosphate increase. During a median duration of 4.4 years, 213 patients reached the endpoint. In a multivariable Cox model, high FGF23 and low 25-hydroxyvitamin D (25D) levels were the only MBD-related factors associated with a higher risk of renal endpoint (adjusted hazard ratio [95% confidence interval] per unit change of log FGF23 and 10 ng/ml of 25D: 1.83 [1.28-2.61] and 0.61 [0.41-0.90], respectively). There was no significant interaction between 25D and FGF23 (P=0.11). Active vitamin D therapy, serum phosphate, 1,25-dihydroxyvitamin D, and parathyroid hormone levels were not related to the renal endpoint. Treating death as a competing risk or multiple imputation for missing values yielded similar results. CONCLUSIONS: Combined use of two markers is useful for the risk stratification of renal outcome.