Serum 25-hydroxyvitamin D as an independent determinant of 1-84 PTH and bone mineral density in non-diabetic predialysis CKD patients.

The role of 25-hydroxyvitamin D [25(OH)D] and fibroblast growth factor-23 (FGF-23) in chronic kidney disease-mineral and bone disorder (CKD-MBD) remains elusive in predialysis CKD patients. From the fact that FGF-23 suppresses bone mineralization in vitro and that 1alpha-hydroxylase is present in parathyroid cells and osteoblasts, they may be associated with bone mass or serum parathyroid hormone (PTH) level. In this cross-sectional observational study, we investigated the potential associations of 25(OH)D or FGF-23 with 1-84 PTH and bone mineral density (BMD) in the femoral neck (FN) and lumbar spine (LS) of 325 non-diabetic patients. All patients had stages 3-5 CKD and had never been treated with bisphosphonate, estrogen, or vitamin D. We measured bone-specific alkaline phosphatase (bone ALP), intact FGF-23 and 1-84 PTH in a third generation assay, and performed a multiple regression analysis for 1-84 PTH and BMD Z-score. In our cohort, 80.1% had 25(OH)D levels less than 30 ng/mL, and 4.1% had levels less than 15 ng/mL. A univariate analysis indicated a negative association for 25(OH)D with 1-84 PTH and bone ALP. A multivariate analysis showed that the significant determinants for 1-84 PTH were 25(OH)D, estimated glomerular filtration rate (eGFR), corrected calcium, serum calcitriol and phosphate. Intriguingly, the three former parameters had negative associations with 1-84 PTH while calcitriol had a positive association. While further adjustment of FGF-23 extinguished the positive association of phosphate and 1-84 PTH, there was absolutely no increase in the R2. With regard to the BMD Z-score, 25(OH)D and the body mass index were the significant common independent positive determinants for both FN and LS, whereas bone ALP was the negative determinant even though there was no correlation noted for 1-84 PTH, calcitriol, or FGF-23 with BMD. In addition, eGFR positively contributed to the Z-score only in FN. Therefore, despite a positive correlation between 25(OH)D and calcitriol, their contribution to the CKD-MBD appears to be different. Since the significant associations for 25(OH)D with 1-84 PTH and BMD were independent of serum calcitriol and bone ALP, this might imply that 25(OH)D has a direct effect on the parathyroid gland and bone.