Examiner Level of Experience on Assessing Optic Discs in a Telemedicine System

Abstract

Purpose : The objective of our study is to investigate the level of agreement between trained non-medical personnel and glaucoma specialists for establishing cup-to-disc ratio and possible presence of glaucoma after digital photos assessment sent to the examiners from a teleophthalmology center.

Methods : During this prospective, cross-sectional study, 64 optic disc digital photographs from one of the eyes of 64 subjects (including either healthy persons, glaucoma suspects or glaucoma patients) were sent to four masked examiners (two trained non-medical personnel or NMP and two experienced glaucoma specialists or EGS). Examiners scored quality, vertical cup-to-disc ratio and likelihood of glaucoma presence. Photographs were taken at a resolution of 1624x1232 pixels using a non-mydriatic camera Cobra HD (CSO; Florence, Italy), saved in uncompressed TIFF format, and compressed 55x their original in a JPEG format. The reviewers, located in three different countries, rated photographs in two sessions of 120 minutes (two weeks between sessions and a different
photo arrangement order in each one). Student's t test, chi square test, ANOVA, kappa statistics and coefficient of variation were used; specificity and sensitivity were also estimated.

Results : Examiners rated all photos as readable with acceptable to good quality. Mean c/d ratio were similar between sessions (0.68 ± 0.2 vs. 0.7± 0.2) but were rated significantly larger by NMP (0.76 ± 0.3 vs. 0.61 ± 0.1, P = 0.04). The level of agreement for determining c/d ratio was good between NMP (r = 0.58, P = 0.01) and very good between EGS (r = 0.71, P = 0.001), and good between NMP and EGS (r = 0.56, P = 0.01). Specificity mean values increased significantly in NMP raters (P = 0.03) in the second session. EGS outperformed NMP regarding specificity / sensitivity values in both sessions (85% /78% vs. 79% / 70%, first session; 83% / 80% vs. 75% / 79%, second session) as shown in the table

Conclusions : We have confirmed during this study that subjective optic disc assessment is a feasible action under a teleophthalmology approach and influenced by the examiner level of experience. Although reports have pointed out that glaucoma specialists outperformed non-specialized personnel to detect glaucoma in different settings, our findings support the fact that NMP participation could play a useful role during the process of optic disc assessment on teleophthalmology initiatives.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.