Variability of quantitative automated perimetry in normal observers

The test/retest variability in different regions of the visual field were evaluated in six commercially available automated threshold static perimeters using normal volunteers. Three of the devices were projection perimeters (Squid®, Octopus 500®, Humphrey Field Analyzer®) and three of them used light-emitting diode stimuli (Dicon 2000®, Fieldmaster 50®, Digilab 350®). The authors' results show that, although some devices were better than others, all have some degree of variability. In addition, they found that there is increased variability from the central 10° out to 30° eccentricity for each perimeter.