The dilemma of digital imaging in retinopathy of prematurity

Abstract

Listen

Translate article

lmology a Two articles, one in the August issue of th AAPOS byWallace et al and one in this Journal of AAPOS by Gelman et al, hi most vexing dilemma facing the ophthalmologis prematurely born babies who develop “serious” of prematurity (ROP): Are the posterior pole ve eye sufficiently abnormal to qualify as having plu The answer to this question largely determines not an eye undergoes peripheral retinal ablation, b clinical algorithm developed by investigators o Treatment for ROP Trial that established new indications for serious ROP in late 2003. Ans critical question is not straightforward, relying o judgment in which the examined baby’s fundus ap compared with a 20-year-old fundus photograph ROP experts as demonstrating the “minimum” d tortuosity sufficient to qualify as having plus Therein lies the dilemma—there is no quantitati for the diagnosis of plus disease—and yet the ter clinically useful in describing an important ocu that indicates potentially sight-threatening retino The two groups of investigators have take and creative approaches to addressing the prob disease diagnosis in ROP. Wallace and cowor developed “ROPtool,” a semi-automated pro takes high-quality digital images and develops m vascular tortuosity and dilation by comparing those of vessels in the reference image of plus d in clinical trials. First, each quadrant of the 2 pole images was judged by the two clinician have vascular abnormalities consistent with p preplus disease (a term that indicates the prese cular abnormalities of the posterior pole insu qualify as plus disease), or no abnormal d tortuosity (normal). Then images were analyze dently by the same authors using ROPtool to d tortuosity index and dilation index for vessels t fied in each quadrant. Though certainly open to of recall bias, the results showed very good eye-