Biochemical metabolic bone disease in preterm infants: clinical determinants, markers and risk of fracture in a cohort study

Biochemical metabolic bone disease (MBD) is a frequent complication in very preterm infants, yet its independent risk factors and relationship with fracture have not been fully elucidated. This study aimed to investigate the clinical characteristics, laboratory parameters, independent predictors, and fracture risk associated with biochemical MBD in very low birth weight (VLBW) infants. Between January 2020 and December 2024, 4,555 neonates admitted to the Neonatology Department of Bursa Yüksek İhtisas Training and Research Hospital were screened. Of these, 915 were preterm, and 413 infants with a gestational age ≤ 32 weeks and/or birth weight ≤ 1500 g met the inclusion criteria and were retrospectively analyzed. Biochemical MBD was defined as serum ALP ≥ 500 IU/L, serum phosphorus < 4.5 mg/dL, and tubular phosphate reabsorption > 95% at 4–6 weeks of life. Fractures were confirmed by X-ray. Multivariable logistic regression and ROC curve analyses were used to identify independent predictors and assess discriminative performance. Among the 413 very low birth weight infants included in the study, biochemical MBD was identified in 82 (19.8%); the overall fracture rate was 5.3% (22/413), whereas it was 26.8% (22/82) among infants with biochemical MBD. Male sex, lower 5-minute Apgar scores, need for resuscitation, caffeine administration, bronchopulmonary dysplasia, patent ductus arteriosus, postnatal steroid exposure, cholestasis, delayed attainment of full enteral feeding, prolonged total parenteral nutrition, and ≥ 3 red blood cell transfusions were significantly associated with biochemical MBD. Multivariate analysis identified male sex (OR 2.29; 95% CI: 1.317–3.970; p = 0.003), parenteral nutrition (PN) duration (OR 1.019 per day; 95% CI: 1.007–1.032; p = 0.003), and ≥ 3 red blood cell transfusions (OR 3.96; 95% CI: 2.193–7.149; p < 0.001) as independent predictors of biochemical MBD. The regression model demonstrated good calibration (Hosmer-Lemeshow p = 0.790) and acceptable discriminative ability (AUC 0.788; 95% CI: 0.733–0.844). Infants with biochemical MBD had significantly elevated alkaline phosphatase (ALP) levels and lower serum phosphate concentrations (698 vs. 392 U/L and 3.63 vs. 5.4 mg/dL, respectively; p < 0.001), while serum calcium levels were similar. ROC analysis indicated that a PN duration exceeding 19.5 days (AUC 0.730; 95% CI: 0.671–0.789; p < 0.001) was a strong predictor of biochemical MBD. Comparisons between infants with and without fractures revealed no significant differences in ALP, phosphate, or tubular reabsorption of phosphate levels; however, serum calcium levels were significantly lower in infants who developed fractures (9.1 vs. 9.5 mg/dL; p = 0.022). A PN duration ≥ 19.5 days is an early, clinically useful predictor of biochemical MBD in preterm infants and is associated with fracture risk. Male sex, prolonged PN, and multiple erythrocyte transfusions are independent predictors of biochemical MBD in VLBW infants. Lower serum calcium in infants with fractures suggests a more severe disease course. Early identification of high-risk infants may allow timely preventive interventions.