Associations between body weight trajectories and neurodevelopment outcomes at 24 months corrected age in very-low-birth-weight preterm infants: a group-based trajectory modelling study.

Introduction: This study aimed to explore the relationship between the trajectories of body weight (BW) z-scores at birth, discharge, and 6 months corrected age (CA) and neurodevelopmental outcomes at 24 months CA. Methods: Conducted as a population-based retrospective cohort study across 21 hospitals in Taiwan, we recruited 3,334 very-low-birth-weight (VLBW) infants born between 2012 and 2017 at 23-32 weeks of gestation. Neurodevelopmental outcomes were assessed at 24 months CA. Instances of neurodevelopmental impairment (NDI) were defined by the presence of at least one of the following criteria: cerebral palsy, severe hearing loss, profound vision impairment, or cognitive impairment. Group-based trajectory modeling was employed to identify distinct BW z-score trajectory groups. Multivariable logistic regression was used to assess the associations between these trajectories, postnatal comorbidity, and neurodevelopmental impairments. Results: The analysis identified three distinct trajectory groups: high-climbing, mid-declining, and low-declining. Significant associations were found between neurodevelopmental impairments and both cystic periventricular leukomalacia (cPVL) [with an adjusted odds ratio (aOR) of 3.59; p < 0.001] and belonging to the low-declining group (aOR: 2.59; p < 0.001). Discussion: The study demonstrated that a low-declining pattern in body weight trajectory from birth to 6 months CA, along with cPVL, was associated with neurodevelopmental impairments at 24 months CA. These findings highlight the importance of early weight trajectory and specific health conditions in predicting later neurodevelopmental outcomes in VLBW infants.

Predicting early mortality and severe intraventricular hemorrhage in very-low birth weight preterm infants: a nationwide, multicenter study using machine learning.

Our aim was to develop a machine learning-based predictor for early mortality and severe intraventricular hemorrhage (IVH) in very-low birth weight (VLBW) preterm infants in Taiwan. We collected retrospective data from VLBW infants, dividing them into two cohorts: one for model development and internal validation (Cohort 1, 2016-2021), and another for external validation (Cohort 2, 2022). Primary outcomes included early mortality, severe IVH, and early poor outcomes (a combination of both). Data preprocessing involved 23 variables, with the top four predictors identified as gestational age, birth body weight, 5-min Apgar score, and endotracheal tube ventilation. Six machine learning algorithms were employed. Among 7471 infants analyzed, the selected predictors consistently performed well across all outcomes. Logistic regression and neural network models showed the highest predictive performance (AUC 0.81-0.90 in both internal and external validation) and were well-calibrated, confirmed by calibration plots and the lowest two mean Brier scores (0.0685 and 0.0691). We developed a robust machine learning-based outcome predictor using only four accessible variables, offering valuable prognostic information for parents and aiding healthcare providers in decision-making.