The association between bronchopulmonary dysplasia grade and risks of adverse neurodevelopmental outcomes among preterm infants born at less than 30 weeks of gestation.

BACKGROUND: Bronchopulmonary dysplasia (BPD) is a multifactorial disease with neurodevelopmental implications. This study aims to quantify the risks of adverse neurodevelopmental outcomes for each BPD grade among preterm infants born at less than 30 weeks' gestation. METHODS: We retrospectively studied infants who received care in our institution until at least 36 weeks postmenstrual age and had a formal neurodevelopmental assessment in our infant follow-up clinic using the Bayley Scales for Infant and Toddler Development (BSID). We assessed the association between BPD grade and adverse neurodevelopmental outcomes using descriptive statistics and regression models. RESULTS: Two hundred and fifty infants, including 89 (35.6%), 87 (34.8%), 65 (20.6%), and 9 (3.6%) with No BPD, Grade 1, Grade 2, and Grade 3 BPD, were included in the study. Small for gestational age, late pulmonary hypertension, dexamethasone administration, and adverse neurodevelopmental outcomes were more common as BPD grade increased. In a logistic regression analysis, Grades 2 and 3, but not Grade 1, BPD were associated with increased odds of a composite adverse neurodevelopmental outcome by 2.7 and 7.2 folds, respectively. A BSID domain-specific analysis showed that higher grades were associated with lower scores in the cognitive, gross motor, and fine motor domains. CONCLUSIONS: Grades 2 and 3 BPD, but not Grade 1, correlate with risks of adverse neurodevelopmental outcomes at a grade-dependent manner in our single-center cohort retrospective study. Further validation using a multi-center large cohort is warranted.

Oxygenation index in the first three weeks of life is a predictor of bronchopulmonary dysplasia grade in very preterm infants.

BACKGROUND: The new bronchopulmonary dysplasia (BPD) grading system was developed based on its correlation with long-term respiratory and neurodevelopmental outcomes and may provide better personalized prognostication. Identifying early-life predictors for accurate BPD grade prediction may allow interventions to be tailored to individual needs. This study aimed to assess whether oxygenation index (OI) dynamics in the first three weeks of life are a predictor of BPD grade. METHODS: A single-center retrospective study was performed. Generalized additive mixed modeling was used to model OI trajectories for each BPD grade subgroup. A multinomial regression model was then developed to quantify the association between OI dynamics and BPD grade. RESULTS: Two hundred fifty-four infants were identified for inclusion in the trajectory modeling. A total of 6,243 OI data points were available for modeling. OI trajectory estimates showed distinct patterns in the three groups, most prominent during the third week of life. The average daily OI change was -0.33 ± 0.52 (n = 85) in the No-BPD group, -0.04 ± 0.75 (n = 82) in the Low-Grade BPD group, and 0.22 ± 0.65 (n = 75) in the High-Grade BPD group (p < 0.001). A multinomial regression analysis showed the initial OI value and the average daily OI change both independently correlated with BPD grade outcomes after adjusting for birth gestation, birth weight z-score, sex, and the duration of invasive ventilation. CONCLUSION: Early-life OI dynamics may be a useful independent marker for BPD grade prediction. Prospective studies may be warranted to further validate the findings.

Machine learning for prediction of bronchopulmonary dysplasia-free survival among very preterm infants.

BACKGROUND: Bronchopulmonary dysplasia (BPD) is one of the most common and serious sequelae of prematurity. Prompt diagnosis using prediction tools is crucial for early intervention and prevention of further adverse effects. This study aims to develop a BPD-free survival prediction tool based on the concept of the developmental origin of BPD with machine learning. METHODS: Datasets comprising perinatal factors and early postnatal respiratory support were used for initial model development, followed by combining the two models into a final ensemble model using logistic regression. Simulation of clinical scenarios was performed. RESULTS: Data from 689 infants were included in the study. We randomly selected data from 80% of infants for model development and used the remaining 20% for validation. The performance of the final model was assessed by receiver operating characteristics which showed 0.921 (95% CI: 0.899-0.943) and 0.899 (95% CI: 0.848-0.949) for the training and the validation datasets, respectively. Simulation data suggests that extubating to CPAP is superior to NIPPV in BPD-free survival. Additionally, successful extubation may be defined as no reintubation for 9 days following initial extubation. CONCLUSIONS: Machine learning-based BPD prediction based on perinatal features and respiratory data may have clinical applicability to promote early targeted intervention in high-risk infants.