Establishment of a nomogram model for predicting metabolic bone disease in preterm infants: A case‒control study.

Abstract

To investigate risk factors for metabolic bone disease (MBD) in preterm infants and establish a nomogram model for predicting MBD risk. A total of 1104 preterm infants were enrolled, among whom 809 were included in the modelling set and 295 were included in the validation set. The modelling set was divided into MBD (n = 185) and non-MBD (n = 624) groups. A multivariate logistic regression analysis was used to investigate the independent risk factors for MBD. R software was used to plot the nomogram model, which was then validated by the data of the validation set. Receiver operating characteristic (ROC) and calibration curves were used to evaluate the nomogram model's performance, and the clinical decision curve was used to assess the clinical practicability of the model. Gestational age, time of trophic feeding initiation, parenteral nutrition duration, necrotizing enterocolitis, bronchopulmonary dysplasia, cholestasis and sepsis were independent risk factors for MBD in preterm infants (P < 0.05). The ROC curve of the modelling set had an area under the curve (AUC) of 0.801; the risk prediction value of 0.196 corresponding to the maximum Youden index was the best value, and the prediction critical value was 125 points. The ROC curve of the validation set had an AUC of 0.854. The calibration curve analysis showed good accuracy and consistency between the model's predicted and actual values. The nomogram model provides an efficient tool for the early assessment of MBD risk. Preterm infants with scores ≥ 125 should receive close attention and interventions in the early stage. • The incidence and severity of MBD are inversely proportional to gestational age and birth weight. Bone loss can lead to prolonged hospital stay, ventilator dependence, pathological fractures and short stature. • Gestational age, time of trophic feeding initiation, parenteral nutrition duration, necrotizing enterocolitis, bronchopulmonary dysplasia, cholestasis and sepsis were independent risk factors for MBD in preterm infants. The nomogram model provides an efficient tool for the early assessment of MBD risk.