Red reflex test (RRT) screening is yet to be a part of the neonate's normal examination before discharge from hospital in a majority of low- and middle-income countries. The purpose was this review was to systematically evaluate the diagnostic accuracy of RRT for the detection of ocular abnormalities in newborns. PubMed, EMBASE, Scopus, Web of Science, and Cochrane database of systematic reviews were the data sources. Quality of Diagnostic Accuracy Studies-2 (QUADAS-2) was utilized for quality assessment of bias and applicability. Random effects models were used to summarize sensitivities, specificities, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and respective confidence intervals (CI). The pooled sensitivity, calculated from the meta analysis of 11 studies, was 23% (95% CI: 21-24%) and pooled specificity was 98% (95% CI: 98-98%). The PLR was 32.52 (95% CI: 7.89-134.15), NLR was less than 1 (0.69 [95% CI: 0.55-0.88]), and DOR calculated was 138.48 (95% CI: 23.85-803.97). The area under the curve (AUC) and Q* index for RRT were 0.98 ± 0.02 and 0.95 ± 0.045, respectively. The results of our study justify the conclusion that RRT is a highly sensitive and specific test for the detection of anterior segment abnormalities.
Keywords: Congenital ocular diseases; neonatal screening; neonates; red reflex testing.