Application of a quantitative vascular severity score in retinopathy of prematurity in the United States and India: new insights into disease epidemiology and pathophysiology

PURPOSE: To compare the epidemiology and phenotypic characteristics of two populations of patients with retinopathy of prematurity (ROP) using a quantitative artificial-intelligence (AI) derived vascular severity scale (VSS) METHODS: We retrospectively analyzed two datasets of ROP images: 1) The iROP dataset including 1019 babies and 6552 eye-level exams developed as part of a large US multicenter cohort study from 2012-2020. 2) 8281 consecutive ROP telemedicine examinations from 2363 babies in the Indian Aravind Eye Care System (AECS), from 2019-2020. We analyzed all eye exams using the iROP DL system that outputs a VSS from 1-9. The main outcome was the relationship between the VSS and the zone, stage, and extent of stage 3 in each dataset, assessed using generalized estimating equations to account for correlation from bilateral and repeated exams. RESULTS: In both datasets, higher VSS was associated with higher stage (P<.001 for all stages in both datasets), higher extent of stage 3 (i-ROP P=.004 and AECS P<.001), and more posterior zone for a given stage of ROP (i-ROP stage 1 P=0.04, 2 P<.001, and 3 P<.001 and AECS stage 1 P<.001 and 3 P<.001. For any given stage, the VSS was lower in the AECS dataset versus i-ROP (all P<.001). CONCLUSIONS: In both populations, the VSS was associated with more posterior zone and higher ROP stage/extent, though on average the VSS was lower in the AECS population. A correlation between measured VSS and intraocular vascular endothelial growth factor concentration might explain both findings.