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.

Use of Esophageal Hemoximetry to Assess the Effect of Packed Red Blood Cell Transfusion on Gastrointestinal Oxygenation in Newborn Infants.

Objectives There are no widely accepted methods of continuously monitoring gut oxygenation in the newborn during packed red blood cell transfusion. We investigated the use of an orally inserted light spectroscopy probe to measure lower esophageal oxyhemoglobin saturations (eStO2) before, during, and after transfusion and made comparisons with abdominal near-infrared spectroscopy (NIRS) and superior mesenteric artery (SMA) flow. Study Design Thirteen neonates with corrected gestational ages ranging from 22 weeks, 0 day to 37 weeks, 5 days were enrolled. eStO2 and NIRS measurements were recorded continuously for a 25-hour period starting 1 hour prior to starting the 4-hour transfusion. Transabdominal ultrasound was used to measure SMA flow prior to, upon completion, and 20 hours after the transfusion. Results Twelve infants completed the study. eStO2 was well-tolerated and was weakly (r = 0.06) correlated (p < 0.001) with NIRS. Compared with NIRS, eStO2 demonstrated a markedly greater variation in oxyhemoglobin values. NIRS and SMA flow measurements did not change, while eStO2 increased from 48 ± 5% and 45 ± 5% in the pre- and intratransfusion periods to 57 ± 4% in the posttransfusion period (p = 0.03). Conclusion Measurement of eStO2 is feasible in neonates and may provide a continuous and sensitive index of rapid changes in mesenteric oxygenation in this patient population.