Helen HL Chan, Dai Ni Ong, Yu Xiang G Kong, Evelyn C O’Neill, Michael A Coote, Jonathan G Crowston

Purpose:
To compare monoscopic to stereoscopic conditions in evaluation of glaucomatous optic nerve damage.

Methods:
Prospective internet-based observational study. Twenty monoscopic disc photographs, and twenty stereoscopic photographs from the same sub-jects, demonstrating a range of features, were pre-sented to 16 glaucoma subspecialists in randomized order via website (http://stereo.gone-project.com/). Participants were asked to assess 9 topographic fea-tures and glaucoma likelihood.

Results:
For each disc, there was moderate-substan-tial agreement between monoscopic and stereosco-pic estimates of glaucoma likelihood (kw = 0.56, CI[0.52,0.61]), and of all disc parameters apart from haemorrhage (vertical CDR k = 0.58, disc size kw = 0.46, disc shape k = 0.57, disc tilt k = 0.49, PPA kw = 0.65, cup shape k = 0.65, cup depth kw = 0.47, nerve fibre layer loss k = 0.69, haemorrhage k = 0.23).Interobserver agreement for monoscopic assessments of glaucoma likelihood (kw = 0.61, CI[0.55, 0.67]) was not weaker than stereoscopic assessments (kw = 0.59, CI[0.54, 0.65]). Monoscopic viewing did not lead to significantly lower interobserver agreement compared to stereoscopic viewing in assessing any of the nine disc morphological features (P > 0.05), in fact, was higher for disc size (monokw = 0.65, stereokw = 0.52), disc tilt (monok = 0.49, stereok = 0.20) and CDR (monokw = 0.72, stereokw = 0.62).Discs were diagnosed as glaucomatous or non-glauco-matous based on visual field testing and other infor-mation. The use of monoscopic or stereoscopic views did not make any statistical difference to the spread of answers for estimated glaucoma likelihood in separate analysis of glaucomatous (ANOVA p = 0.688) and non-glaucomatous discs (p = 0.502).

Conclusion:
Monoscopic disc photographs do not represent a significant disadvantage compared to stere-oscopic photographs in the evaluation of glaucoma-tous optic nerve damage among expert observers.