Purchase this article with an account.
R. V. Chan, S. L. Williams, Y. Yonekawa, L. Wang, T. C. Lee, M. F. Chiang; Accuracy of Retinopathy of Prematurity Diagnosis by Retinal Fellows Compared to Experts. Invest. Ophthalmol. Vis. Sci. 2008;49(13):1402. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To measure the accuracy of retinopathy of prematurity (ROP) diagnosis by retinal fellows compared to that of an expert pediatric retinal specialist.
An atlas of 804 retinal images was captured from 248 eyes of 67 premature infants with a wide-angle camera (RetCam-II; Clarity Medical Systems, Pleasanton, CA). Images were uploaded to a study website, from which an expert pediatric retinal specialist and seven retinal fellows independently reviewed and provided a diagnosis (no ROP, mild ROP, type-2 ROP, or treatment-requiring ROP) for each eye. Sensitivity and specificity of each retinal fellow was calculated, compared to a reference standard of diagnosis by the expert specialist. Images were divided into those captured at 31-33 weeks post menstrual age (PMA) and those captured at 35-37 weeks PMA.
Six fellows had no ROP screening experience, and one fellow had a year of experience. For fellows at 31-33 weeks PMA: (a) sensitivity and specificity of fellows for identifying mild or worse ROP ranged from 0.684-0.921 (mean 0.813) and 0.700-0.965 (mean 0.835) respectively; and (b) sensitivity and specificity for identifying type-2 or worse ROP ranged from 0.000-0.889 (mean 0.569) and 0.780- 0.990 (mean 0.927) respectively. For fellows at 35-37 weeks PMA: (a) sensitivity and specificity for identifying mild ROP or worse were 0.909-1.000 (mean 0.967) and 0.774-0.980 (mean 0.887) respectively; (b) sensitivity and specificity for identifying type 2 or worse ROP were 0.576-1.000 (mean 0.809) and 0.655- 0.976 (mean 0.856) respectively; and (c) sensitivity and specificity for identifying treatment-requiring ROP were 0.769-1.000 (mean 0.947) and 0.745-0.971 (mean 0.874) respectively.
Performance of fellows for ROP diagnosis was less than that of the expert grader, particularly for detection of clinically-significant disease. This raises potential concerns about the quality of ROP screening performed by inexperienced examiners, and may indicate that formal training in ROP screening is required to provide appropriate care for this devastating condition.
This PDF is available to Subscribers Only