Premium
Secondary Analysis of Premature Respiratory Outcomes Program (PROP) Clinical Trial to Identify Correlations of Risk Factors for Respiratory Morbidity and Mortality in Premature Infants
Author(s) -
Torres Juliet,
Otero Jose Javier
Publication year - 2020
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.2020.34.s1.04620
Subject(s) - medicine , bronchopulmonary dysplasia , gestational age , pediatrics , logistic regression , birth weight , premature birth , apgar score , pregnancy , genetics , biology
Background The Premature Respiratory Outcomes Program (PROP) was a collaborative, multi‐disciplinary clinical study that sought to analyze the factors that could predict future risk for respiratory morbidity and mortality for 835 premature infants with gestational ages 23 0/7 to 28 6/7 weeks. PROP investigators conducted physiologic challenges called Room Air Challenges (RAC) and Hypoxic Challenges Tests (HCT) to determine whether the premature infants could maintain their SpO 2 in the target range of ≥90% and to assess pulmonary reserve in former premature infants without Bronchopulmonary Dysplasia(BPD) at 36 weeks’ premenstrual age (PMA). This project was funded by the National Institutes of Health (NIH) and the National Heart, Lung, and Blood Institute (NHLBI) for 5 years through a $18.5 million research initiative. Objective We are interested in understanding the relationship between brain imaging data and respiration outcomes, and infection or chorioamnionitis and respiration outcomes. In addition, the RAC and HCT provides the opportunity to analyze the outcomes of physiologic challenges in preterm infants compared to term infants undergoing the same challenges in studies that conducted preflight HCT. Statement of the methods We utilized machine learning tools such as random forest, support vector machines, gradient boosted machines, and logistic regression to generate predictive models of the aforementioned features. Summary of results Preliminary data analysis of key variables to be included in our future data analytics includes standard deviations for birth weight (gm) (235.98), the baby’s head circumference (cm) (83.05), APGAR for one minute (scored 1–10) (2.47), and APGAR 5 minutes (2.10). Statement of conclusions This PROP clinical study encapsulates an immense amount of multidisciplinary data from 835 premature infants, which is an extremely special population to include in research as approximately 1 out of every 9 births in the United States occurs prematurely. Support or Funding Information R01HL132355