Biomarkers of phosphate homeostasis and metabolic bone disease of prematurity: impact of early nutrition and regulation by vitamin D, iPTH, PTHrP, and FGF23

Global practices for supplementing calcium, phosphate, and vitamin D in early life, as well as for screening for metabolic bone disease of prematurity (MBDP), remain inconsistent. Furthermore, researchers have not yet established gestational age (GA)-specific reference intervals for phosphate homeostasis biomarkers. This study aimed to assess phosphate homeostasis parameters and MBDP risk factors under local neonatal nutritional practices and to characterize the serial levels of key hormones involved in phosphate metabolism, including 1,25-dihydroxyvitamin D (1,25(OH)2D), intact parathyroid hormone (iPTH), parathyroid hormone-related protein (PTHrP), and fibroblast growth factor-23 (FGF23). We conducted a prospective cohort study in 657 preterm infants with GAs of 26–36+ 6 weeks. We analysed serum phosphate (sP), calcium (sCa), and alkaline phosphatase (ALP) levels across different GAs and postnatal ages, and we identified risk factors for MBDP using univariate and multivariate logistic regression analyses. In a separate, in-depth cohort of 63 preterm infants, we measured and analysed urinary phosphate (uP), calcium (uCa), tubular reabsorption of phosphate (TRP), renal phosphate threshold concentration (Tp/GFR), and these hormones, and we assessed correlations using Pearson’s test. Extremely preterm infants (EPIs) and very preterm infants (VPIs) developed low sP (median nadir: 1.54 mmol/L; interquartile range, IQR: 1.21–2.03), low sCa (nadir: 2.15 mmol/L; IQR: 2.07–2.22), and hypophosphaturia (uP nadir: 0.71 mmol/L; IQR: 0.31–1.66), along with elevated uCa (peak: 1.46 mmol/L; IQR: 1.18–1.70) and secondary hyperparathyroidism (peak iPTH: 29.2 pmol/L), which reflects insufficient early-life calcium-phosphate (Ca-P) intake. We observed lower sP levels (1.27 ± 0.20 mmol/L) in infants who started extra vitamin D supplementation after 14 postnatal days and in those with prolonged parenteral nutrition (PN) exposure. Our analysis showed a positive correlation between Tp/GFR and GA. We found no GA-based differences in the measured hormone levels. PTHrP levels progressively declined from umbilical cord blood (10.9 ± 2.4 pmol/L) to postnatal day 14 (9.1 ± 2.0 pmol/L). In contrast, initially very low iPTH levels (median 1.9 pmol/L) increased sharply after birth (median 28.3 pmol/L). FGF23 levels remained elevated throughout the study period. Our study reveals suboptimal Ca intake from PN, insufficient Ca-P supply via enteral nutrition (EN), and delayed vitamin D supplementation in the current local early-life nutrition protocol. The high prevalence of maternal vitamin D deficiency likely contributed to the vitamin D insufficiency that we observed in most preterm infants. Clinicians should utilize biomarkers including sP, sCa, uP, uCa, and iPTH to assess phosphate homeostasis and guide mineral supplementation in early life to reduce the incidence of MBDP.