[Measuring the optic papilla with planimetry and the optic nerve head analyzer in glaucoma and suspected glaucoma. II. Correlations of the results of both methods with changes in the visual field, studied with the Octopus automatic perimeter].

Sixty-seven optic nerve heads of 40 patients with proven or suspected glaucoma were measured by planimetry and with the Rodenstock Optic Nerve Head Analyser (ONHA). The results were compared to visual field indices obtained with the Octopus program G-1. Good correlation of the results obtained by the two measurement procedures has been shown (Stürmer et al. 1989), between values for the disk area, the excavation area, and the cup/disk ratio. However, there is only weak correlation of values for the neuroretinal rim. The planimetrically measured neuroretinal rim area in the total population examined proved to be significantly correlated only with the visual field index for mean retinal sensitivity (MS; r2 = 0.106; P = 0.007) and short-term fluctuations (SF; r2 = 0.066; P = 0.036). After division of the population examined into different diagnostic groups, further statistically significant correlations between optic nerve head parameters and the various visual field indices were shown; here, the cup/disk ratio of both measurement procedures in two subgroups showed the best correlation with the visual field indices MS and mean defect (MD). Neither in the total population nor in any of the subgroups was a statistically significant correlation found between the volume of the excavation and one of the visual field indices. Comparing only the data for the temporal quadrant of the optic nerve head with the visual field did not improve the correlations. The best, i.e. highly significant, correlations were between optic nerve head parameters and the age of the patient. The correlation factors are much lower than other published data. A variety of factors could be responsible for these weak correlations: different optic nerve head configurations on the one hand, and localized or diffuse visual field defects on the other. In unselected cases it appears impossible to predict the visual field of a given optic nerve head. Both methods are suitable for follow-up, but not all anatomical configuration of the optic nerve head permit this.