Evaluation of Modified Screening Criteria for Retinopathy of Prematurity in India

Abstract

Purpose : Retinopathy of Prematurity (ROP) screening criteria vary around the world based on the population at risk. Screening criteria based on birthweight (BW) and gestational age (GA) can be more broad in low- and middle-income countries like India, where ROP can be more commonly seen in less premature babies than in the US. The purpose of this project was to evaluate the effect of various screening criteria on a large population of premature babies in South India.

Methods : We analyzed two years of data from the Aravind Eye Hospital ROP telescreening program (1/1/19-12/31/20). During that period, participating neonatal care units referred babies for screening who were less than 36 weeks postmenstrual age (PMA), or less than 2000 grams, or who had other risk factors for ROP. We evaluated 3 different screening criteria for sensitivity and specificity for detecting treatment-requiring (TR) ROP: US screening criteria (<31 weeks[w] PMA or < 1500 grams[g] BW), Indian criteria (<36w or <2000g), and intermediate criteria (<34w or <1750g).

Results : The sensitivity / specificity of the 3 sets of guidelines (India, Intermediate, and US) were 100% / 13%, 100% / 40%, and 85% / 67%, respectively, as seen in the table. Of the entire screened population (N = 2714 babies), use of intermediate criteria would reduce screening burden by 38% without missing a case of TR-ROP in this population.

Conclusions : ROP screening criteria evolve as demographics and practices in a region change. Current Indian screening criteria are 100% sensitive, but nonspecific, and US screening criteria are insufficiently sensitive for implementation in this Indian population. An intermediate cutoff produced 100% sensitivity with 38% reduced screening burden and can be considered in the future if 100% sensitivity can be ensured.

This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.

 

Table: Sensitivity and Specificity For Different Criteria

 

Figure: Distribution of Babies With and Without Treatment Requiring ROP According to GA and BW, Shaded Portion Representing Babies Not Screened by the Different Criteria