Analysis of the distribution characteristics of serum phosphorus levels in neonates in the early neonatal period based on gestational age stratification and their association with perinatal factors

Objective To describe the distribution of serum phosphorus, 25-hydroxyvitamin D (25-OHD), and alkaline phosphatase (ALP) levels in neonates of different gestational ages in the early neonatal period, analyze the association between perinatal factors and serum phosphorus, and explore whether there is a non-linear relationship between gestational age and serum phosphorus.Methods A prospective cohort design was used to include 360 singleton neonates born in our hospital and transferred to the neonatal department between August 2019 and June 2022, with gestational ages ranging from 27+0 to 41+6 weeks. They were divided into four groups: 27–31 weeks (42 cases), 32–33 weeks (48 cases), 34–36 weeks (110 cases), and 37–41 weeks (160 cases). Venous blood was collected within 2–4 h after birth to measure serum phosphorus (phosphomolybdate method), 25-OHD (enzyme-linked immunosorbent assay [ELISA]), and ALP (rate method). Maternal pregnancy and delivery-related factors, neonatal birth weight, and Apgar scores were collected. Data analysis was performed using univariate linear regression, Spearman correlation analysis, restricted cubic splines (RCS), and piecewise regression. The <34 weeks subgroup was used as the primary analysis for the preterm turning point, while the full-cohort model was retained as an exploratory/descriptive reference. An unadjusted model was fitted as a sensitivity analysis.Results Serum phosphorus, 25-OHD, and ALP differed significantly across gestational age groups (all p < 0.001). Serum phosphorus was highest in the 27–31 weeks group (2.12 ± 0.44 mmol/L), lowest in the 32–33 weeks group (1.78 ± 0.35 mmol/L), and rebounded thereafter. Post-hoc Tukey’s tests showed that serum phosphorus was significantly lower in the 32–33 weeks group and the 34–36 weeks group compared with the 27–31 weeks group (both Tukey-adjusted p < 0.05). Birth weight, Apgar score, and maternal vitamin D/calcium supplementation were positively correlated with serum phosphorus (all p < 0.001). In the primary analysis restricted to infants <34 weeks (n = 90), both the adjusted and unadjusted models converged at a breakpoint of approximately 32.2 weeks (bootstrap 95% CI for the piecewise breakpoint: 30.8–33.6 weeks).Conclusions Gestational age and serum phosphorus show a U-shaped non-linear relationship. In preterm infants (<34 weeks), the turning point is consistently estimated at approximately 32 weeks, with the most rapid decline occurring before this age. Preterm infants <32 weeks may benefit from enhanced phosphorus monitoring; these exploratory, single-center findings require validation in larger cohorts.