This study evaluated the detection of glaucoma using 42 non-stereoscopic optic disc photographs and estimated the optic disc characteristics that influence detection. The 197 participants used an online website to grade the photographs for features of glaucoma including disc size, shape, tilt, peripapillary atrophy, vertical cup-to-disc ratio, cup shape, cup depth, retinal nerve fiber layer defects, and presence of disc hemorrhage. They used a four-point scale from unlikely (score 1) to certain (score 4) to estimate the probability of glaucoma.
The authors used the grades from 37 glaucoma specialists as the reference standard and defined misclassification when the participant's score differed by > one point. A participant score higher than the reference score was overestimation of glaucoma, and a lower score was underestimation. Ophthalmology trainees and comprehensive ophthalmologists misclassified optic discs at a rate of 35% and 33%, respectively with similar rates of underestimation (~23%) and overestimation (~10%) of glaucoma. Features associated with misclassification included detection of nerve fiber loss, disc hemorrhage, rim loss (or thinning), and vertical cup-to-disc ratio. Large discs were associated with an overestimation of glaucoma likelihood.
Eye care providers need better case finding of glaucoma
The study limitations are clearly outlined in the manuscript. The study results may be different from the true clinical situation because participants did not have access to stereoscopic photographs. Also, the study did not externally validate the results to a dilated eye exam in a medical clinic, which is most likely the true clinical situation for case finding of glaucoma. It is not clear how the participants would be able to estimate the size of the optic discs considering this feature was most common with overestimation of glaucoma.
This study provides important information regarding the features of glaucoma that result in misclassification. Evaluation of the optic disc remains a key part of the ophthalmic examination, and for case finding of glaucoma. Several population-based studies suggest that approximately 40% of patients with glaucoma were unaware they had glaucoma despite a recent eye exam. Overall, this suggests that eye care providers need better case finding of glaucoma and the manuscript suggest targets of teaching for trainees and clinicians.