A statistical analysis of normal visual fields and hemianopsias recorded by the computerized perimeter 'Competer'

Publisher Website
Google Scholar
Abstract
Seventy-five patients with CNS disorders and normal visual fields or hemianopsia were examined on one to 11 occasions by the modified computerized perimeter 'Competer'. The aim of the work was to evaluate the previously introduced statistical parameters. Of these, the performance values of the central (Pc) and mid-peripheral field (Pp) express the total ability and the difference values (Dc and Dp) the size of a hemianopsia. In patients with normal fields gradual changes of Pc seemed to reflect changes of the general condition. For detecting small hemianopsias Dc was of great help, but for following up hemianopsias both Dc and Pc were indispensable. As a rule, a series of examinations demonstrating a trend was required for deciding whether the involvement of the visual pathway had changed or not. There were co-variations of the Pc-values of the right and left fields and also of the Dc-values in patients with homonymous hemianopsia, in whom the defects were usually congruous, but not in patients with bitemporal hemianopsia, in whom the defects were usually asymmetrical. A negative correlation and linear regression between D-and P-values was found in hemianopsias, but the calculated regression lines were displaced from the expected ones because of additional defects in the non-hemianopic hemifields. In fatigued patients false positive defects were common, particularly in the mid-peripheral area. Consequently. Pp and Dp were less reliable than Pc and Dc, and this area should be examined by the 'Competer' only in alert patients. To sum up. the elaborated statistical parameters were suitable for quantitative studies of visual fields in well-cooperating patients.

This publication has 3 references indexed in Scilit: