The Ability of Scanning Laser Ophthalmoscopy to Identify Various Glaucomatous Optic Disk Appearances

Purpose: To compare four distinct glaucomatous disk appearances using scanning laser ophthalmoscopy and determine the ability of a previously described discriminant analysis to detect glaucomatous damage for each group. Methods: Two thousand three hundred eighty-eight stereoscopic optic disk photographs of 2,388 subjects with glaucoma or suspicion of glaucoma were reviewed to select four pure glaucomatous optic disk types. Twenty-three topographic, volumetric, and morphometric scanning laser ophthalmoscopic variables from one optic disk of each selected subject were compared. Differences between the groups were evaluated using an analysis of variance and the Pearson chi-square test. Results: Forty-four focal ischemic disks, 38 myopic glaucomatous disks, 30 senile sclerotic disks, and 28 disks characterized by generalized cup enlargement were studied. Significant intergroup differences were found for the majority of the evaluated optic disk variables. The accuracy with which the discriminant analysis model was able to detect disks with glaucomatous visual field damage was 93.2% for focal ischemic disks, 81.6% for myopic disks, 66.7% for senile sclerotic disks, and 78.6% for disks with generalized cup enlargement. Conclusions: By comparing mean values for certain optic disk variables, scanning laser ophthalmoscopic optic disk assessment can distinguish groups of disks with specific appearances. However, the ability of presently available software to detect glaucomatous damage varied considerably with disk appearance. Expert clinical optic disk evaluation remains an important part of the assessment of patients suspected of having glaucoma, although refinement of computerized scanning laser ophthalmoscopic disk assessment may eventually make this easier.