The Influence of Clinical Data on Evaluation of the Optic Nerve

Abstract

Purpose: : To assess the influence of clinical data on the interpretation of glaucomatous optic nerve damage using the cup to disc ratio (CDR) and Disc Damage Likelihood Scale (DDLS).

Methods: : Forty-five pairs of stereo optic disc photographs from 45 Wills Eye Institute glaucoma patients were assembled. Four clinical graders (2 clinical glaucoma fellows, 1 glaucoma research fellow and 1 medical student) reviewed and interpreted the 45 stereo photo pairs a total of four times. For each of the four reviews, the graders evaluated the optic discs and assigned both a CDR and a DDLS score. Interpretations 1 and 2 were done without providing any ancillary clinical information. Interpretations 3 and 4 were done with accompanying clinical information that was either associated with that patient or unrelated to that patient. Each of the reviews was separated by a minimum one week interval and the order of the photographs was shuffled between each review. The interpretations made during the first two reviews were considered the gold standard. Interpretations 3 and 4 were then compared to the gold standard.

Results: : An ANOVA test for repeated measures revealed that there was no disagreement between any of the interpretations (CDR p=0.59; DDLS p=0.96). Intra-rater reliability revealed different levels of agreement between observers. More experienced graders had higher levels of agreement.

Conclusions: : The presence of ancillary clinical information did not appear to influence the interpretations of optic nerve damage by this group of observers. Individual reliability measurements were variable perhaps related to the extent of clinical training.

Keywords: optic nerve • optic disc