Perimetry update.

The possible applications of computer-assisted static perimetry are examined after five years of personal controlled studies of different types of computerized perimeters. The common advantage of all computer-assisted perimeters is the elimination of the influence of the perimetrist on the results. However, some perimeters are fully computer-assisted and some are only partially so. Even after complete elimination of the perimetrist's influence, some physiological and psychological influences remain due to the patient and will cause fluctuations of the results. In perimeters, the density of the grid and the adaptive strategy of exact threshold measurement are important in obtaining reproducible results. A compromise between duration of the test and exactness has to be found. The most acceptable compromise seems to be an uneven distribution of stimuli, which form a denser grid in areas of special interest and a wider grid in areas less likely to be involved, combined with exact threshold measurements only in suspicious areas. Multiple stimulus presentation is not adequate. High sensitivity of screening is not a great advantage, unless combined with a high specificity. We have shown that using the same stimulus luminosity for center and periphery (a nonadaptive strategy) produces nonspecific results. Only an adaptive strategy can result in high sensitivity and specificity. Adaptive strategy means that the luminosity of the stimulus is adapted to the individual threshold curve of the visual field. In addition, the exact individual thresholds are bracketed by small up and down steps of variation in luminosity. In some cases, scanning programs with two levels of adaptation can be sufficient. The user of modern perimeters must understand such terms as: asb, dB, presentation time, and diameter of stimuli. Projection of stimuli is preferred to light emitting diodes or glass fiber optics. The programs (software) of the modern instruments are of the greatest importance, because the clinical experience that the perimetrist had to acquire in previous manual perimetry is incorporated in these programs. In the Octopus perimeter a delta program is available that differentiates patient fluctuations that may be insignificant from directed significant alterations of the field which might require alteration of therapy. The programs are listed for different computer-assisted perimeters, and their choice is described. The costs of the perimeters are also given. Many controlled clinical studies are quoted briefly where they are useful for understanding the discussion. A brief chapter deals with the reliability of the perimetric test.(ABSTRACT TRUNCATED AT 400 WORDS)