Background:
The relationship of fibroblast growth factor 23 (FGF23) with incident chronic kidney disease (CKD) has been examined in older but not younger populations.
Methods:
Linear regression models were used to examine the associations of c-terminal FGF23 (cFGF23) and intact FGF23 (iFGF23) with 10-year change (1995-96 to 2005-06) in estimated glomerular filtration rate (eGFR) and urine albumin to creatinine ratio (UACR) in the Coronary Artery Risk Development in Young Adults (CARDIA) cohort. Cox proportional hazards models were used to assess the association of cFGF23 with incident CKD, defined as eGFR < 60 ml/min/1.73 m2 or UACR ≥ 30 mg/g. Multivariable models adjusted for age, sex, race, education, field center, physical activity, body mass index, diabetes, smoking, and systolic blood pressure.
Results:
Among 2511 participants, the mean age was 45 ± 3.6 years, mean eGFR was 96.5 ± 14.0 ml/min/1.73 m2, median UACR was 4.3 (IQR, 3.0 – 6.7) mg/g. Most (62.6%) participants were non-smokers, the prevalence of diabetes was low (6.6%), and median values for 10-year changes in eGFR and UACR were modest (-5.50 ml/min/1.73 m2 and 0.70 mg/g, respectively). No consistent associations between cFGF23 and 10-year change in eGFR and UACR were observed. During a median follow up of 9.98 years, incident CKD developed in 258 participants. There was a non-linear association of cFGF23 with incident CKD, and relative to lowest quartile of cFGF23, significant relationship detected only among participants in the highest quartile (hazard ratio 1.58, 95% confidence interval [1.09, 2.27]). Similar findings were observed for iFGF23.
Conclusion:
Among middle-aged adults in the CARDIA cohort, median eGFR and UACR changes were modest and cFGF23 and iFGF23 were not consistently associated with 10-year change in eGFR or UACR. A non-linear relationship was observed between cFGF23 and incident CKD, with individuals with highest cFGF23 levels being at risk for developing CKD.