Early Prediction Model for Retinopathy of Prematurity Using Placental and Neonatal Risk Factors.

Area of Science:

• Neonatal Medicine
• Ophthalmology
• Pathology

Background:

• Existing retinopathy of prematurity (ROP) prediction models focus on screening starting 5-7 weeks post-birth.
• Early identification of high-risk neonates is crucial for timely intervention.
• Placental and early postnatal factors may offer predictive value before standard ROP screening.

Purpose of the Study:

• To develop and validate a model incorporating placental and early postnatal risk factors for earlier ROP prediction.
• To identify high-risk neonates for retinopathy of prematurity (ROP) before conventional screening.
• To enable earlier consideration of neonatal treatment for ROP prevention.

Main Methods:

• Retrospective analysis of 591 neonates born ≤32 weeks gestational age and/or ≤1500 grams birthweight.
• Placental examination for histological abnormalities.
• Development of the Placenta as an Additional Predictor for ROP (PAPROP) model including gestational age, birthweight, ventilation, postnatal corticosteroids, and placental pathologies (chorioamnionitis, villous hypoplasia).
• Internal validation using five-fold cross-validation and comparison with a reference model (GA & BW only).

Main Results:

• The PAPROP model demonstrated a higher discriminatory ability (0.81) compared to the reference model (0.78).
• The model achieved a sensitivity of 0.97 and specificity of 0.44 at a 10% threshold.
• Implementing PAPROP in the second postnatal week could reduce ROP screenings by 25% without missing severe ROP stages.

Conclusions:

• The PAPROP model accurately predicts ROP development by the second postnatal week.
• This model shows potential for high clinical utility and cost-effectiveness in ROP management.
• Further external validation may support personalized neonatal treatment for ROP prevention in high-risk infants.