Discriminant analysis formulas of optic nerve head parameters measured by confocal scanning laser tomography

To evaluate whether discriminant analysis formulas of optic disc variables measured by confocal laser scanning tomography can detect glaucomatous visual field defects, to compare the diagnostic precision of these formulas to detect glaucomatous visual field defects in different types of chronic open-angle glaucoma, and to assess whether gender or refractive error influence the results obtained by the formulas.

One hundred and sixty-one patients with perimetrically defined glaucomatous optic nerve damage and 194 healthy subjects were recruited. All patients underwent confocal laser scanning tomography of the optic disc. The data were analyzed with three linear discriminant analysis formulas (sectorial, Bathija, and Mikelberg) obtained in sets of data different from those used in the present study.

The areas under the receiver-operating characteristic curves of the three formulas and of the cup shape measure as a single parameter ranged from 0.649 to 0.81 in the entire group, and the results did not change when age-matched eyes were considered (0.618-0.812). In each of the glaucoma subgroups with primary open-angle, pseudoexfoliative, and normal-pressure glaucoma, and additionally in the hyperopic, myopic, female, and male subgroups, the sectorial formula had the highest diagnostic precision and the highest correlation coefficients with the visual field indices, followed by the Bathija and Mikelberg formulas, without major differences between the subgroups. All three formulas were more effective than the cup shape measure as a single parameter.

In the various chronic open-angle glaucomas, the sectorial and Bathija formulas tended to have higher diagnostic precision than the Mikelberg formula and the cup shape measure. Gender and refractive error do not markedly influence the diagnostic precision of the formulas tested. The scores of the formulas are mild indicators of the amount of glaucomatous visual field loss. All three formulas are superior to the single cup shape measure in the detection of glaucomatous optic nerve damage.