Prediction of Retinopathy of Prematurity Using the WINROP (Weight, IGF-1, Neonatal Retinopathy of Prematurity) Algorithm in a South African Population

Abstract

Aim: This study aimed to assess the efficacy of the WINROP (Weight, IGF-1, Neonatal Retinopathy of Prematurity) screening algorithm in a South African population. Methods: A retrospective record review included infants born between 1 January 2013 and 1 December 2014 who underwent ROP (retinopathy of prematurity) screening. Outcomes of ophthalmology examinations were compared to alarms triggered on WINROP after gestational age, date of birth, and weekly weights were entered. Sensitivity, specificity, positive predictive, and negative predictive values and mean time of alarm were calculated. Results: Rates of ROP were 5.9% for all stages of ROP and 2.3% for severe ROP in the 220 infants included. Mean gestation age was 29.1 ± 1.3 weeks and mean birth weight 1,115.5 ± 201 g. WINROP triggered high-risk alarms in 70.5% of infants at a mean of 30.7 weeks of gestational age. Sensitivity for severe ROP was 100 and 76.9% for all stages of ROP. Specificity was low for both severe ROP and all stages of ROP at 30.2 and 30.0%, respectively. Conclusion: Rates of ROP are low in this population. The high number of alarms with a low negative predictive value would reduce the number of screens by 29.5%. Alarms were triggered before scheduled screening, possibly helpful in planning discharges and follow-up visits.