Development of a scoring model integrating inflammatory markers for predicting ROP in preterm neonates.

Abstract

Purpose: ROP is a leading cause of blindness in preterm infants worldwide. ROP diagnosis is made through a detailed eye examination and supportive methods are required, especially in limited centers. ROP diagnosis primarily relies on frequent ophthalmological examinations; however, complementary diagnostic tools may significantly aid clinicians, particularly in centers with limited resources. This study introduces an innovative scoring model integrating systemic inflammatory markers for its early prediction of ROP.

Methods: A retrospective case-control study involving 120 preterm infants (≤ 32 weeks, ≤ 1500 g) was conducted. Hemogram-based inflammatory indices, including the Systemic Inflammatory Response Index (SIRI), Platelet-to- Lymphocyte Ratio (PLR), and Pan-Immun Inflammatory Value (PIV), were calculated from blood samples obtained within the first 24 h of life and before ROP diagnosis. Logistic regression and ROC analyses informed the scoring model.

Results: Infants with ROP showed lower gestational ages and birth weights (p < 0.001, p = 0.02). SIRI-2 displayed the highest diagnostic accuracy (AUC = 0.704). A combined model using SIRI-2, PLR-2, and PIV-2 achieved superior performance (AUC = 0.802). Logistic regression identified SIRI-2 and PLR-2 as independent predictors, enhancing risk stratification.

Conclusion: This scoring model not only enhances early risk stratification but also holds potential for widespread implementation in neonatal intensive care units, particularly in resource-limited settings. Future multicenter trials will further establish its role in optimizing neonatal outcomes globally.