Quantification of Early Neonatal Oxygen Exposure as a Risk Factor for Retinopathy of Prematurity Requiring Treatment

ABSTRACT Objective/Purpose Retinopathy of prematurity (ROP) is a leading cause of childhood blindness related to oxygen exposure to the developing retina in premature infants. Since oxygen monitoring protocols have reduced the incidence of treatment-requiring ROP (TR-ROP), it remains unclear whether oxygen exposure remains a relevant risk factor for incident TR-ROP and aggressive ROP (A-ROP), a severe, rapidly progressing form of ROP. The purpose of this proof-of-concept study was to use electronic health record (EHR) data to evaluate early neonatal oxygen exposure as a predictive variable for developing TR-ROP and A-ROP. Design Retrospective cohort study. Subjects, Participants, and/or Controls 244 infants screened for ROP at a single academic center. Methods Intervention, or Testing: For each infant, oxygen saturations and fraction of inspired oxygen (FiO₂) were manually extracted from the EHR until 31 weeks PMA. Cumulative minimum, maximum, mean oxygen saturation and FiO₂ were calculated on a weekly basis. Random forest models were trained with 5-fold cross-validation using GA and cumulative minimum FiO₂ at 30 weeks PMA to identify infants who developed TR-ROP. Secondary ROC analysis on infants with/without A-ROP was performed without cross-validation due to small numbers. Main Outcomes/Measures For each model, cross-validation performance for incident TR-ROP was assessed using area under the receiver operating curve (AUROC) and precision-recall curve (AURPC) scores. For A-ROP, we calculated AUROC, and evaluated sensitivity and specificity at a high-sensitivity operating point. Results Of the 244 infants included, 33 developed TR-ROP and 5 were diagnosed with A-ROP. For incident TR-ROP, random forest models trained on GA plus cumulative minimum FiO₂ (AUROC=0.93±0.06, AUPRC=0.70±0.08) were not significantly better those trained on GA alone (AUROC=0.92±0.06, p=0.59; AUPRC=0.74±0.12, p=0.32). However, models using oxygen alone had an AUROC of 0.80±0.09. Secondary ROC analysis for A-ROP found an AUROC=0.92 [95% CI 0.87-0.96]. Conclusions Oxygen exposure can be extracted from the EHR and quantified as a risk factor for incident TR-ROP and A-ROP. Extracting quantifiable clinical features from the EHR may be useful in the future for building integrative risk models for multiple diseases, and evaluating the complex relationships between oxygen exposure, ROP, and other sequelae of prematurity.