Screening for retinopathy of prematurity in South Africa: are those developing severe ROP screened on time? Data from a prospective register

TL;DR

This study examines if retinopathy of prematurity screening in South Africa is timely, finding that most infants are screened before severe ROP develops.

Contribution

The study evaluates screening adherence to guidelines in South Africa and proposes recommendations based on local data.

Findings

Screening was initiated on time in 78.9% of infants, preventing detection of severe ROP at first screening.

Most infants were screened using postnatal age alone, and many missed follow-up exams.

The low incidence of severe ROP in the study limits generalizability to regions with higher disease rates.

Abstract

To determine whether retinopathy of prematurity (ROP) screening is initiated on time according to current South African (SA) guidelines, that is, before the onset of stage 3 and type 1 ROP.

A prospective study of preterm infants screened at five neonatal units between 1 May 2022 and 31 January 2023 in Cape Town, SA. Data on all infants screened with a birth weight <1250 g or gestational age (GA) <32 weeks were extracted from the ROP South African (ROPSA) register, including postnatal age (PNA) and postmenstrual age (PMA) at first screening.

A total of 696 infants were included, 58.9% (n=410) of whom had an early ultrasound (EUS) for GA estimation. Overall, 220 (31.6%) infants developed ROP, 20 (2.9%) had stage 3 or type 1 and 7 (1.0%) required treatment. Screening was initiated on time according to SA criteria in 549 (78.9%) infants, none of whom had stage 3 or type 1 ROP at first screening. Stage 3 and type 1 ROP were first detected at PNA and PMA of 6.3 and 33.1 and 8.9 and 35.9 weeks, respectively. Most infants (319, 45.8%) were screened according to PNA only, and 78.9% of the 185 infants screened only once did not attend subsequent examinations.

Screening started on time in most infants and prior to the development of severe ROP. Due to the limited availability of EUS in our region and to promote complete screening, we recommend that screening be initiated using PNA alone at 4–6 weeks or prior to discharge, whichever is earliest. The low proportion of infants with stage 3 and type 1 ROP is a limitation in our study. Therefore, recommendations may not be generalisable to South African regions where neonatal care results in a higher proportion of infants developing type 1 ROP.

Figures