Abstract
Purpose: :
Conventional static perimetry employs a constant stimulus size at all test locations, most often the Goldmann size III. When Differential Light Sensitivity (DLS) is recorded in linear (L–1, 1/Lambert) units, the relationship between DLS and ganglion cell number is approximately linear. The use of a fixed stimulus size gives rise to varying structure / function slopes across stimulus locations: the Goldmann III stimulus covers approximately 70 ganglion cells at 10 degrees and 15 ganglion cells at 30 degrees eccentricity. We tested the hypothesis that scaling stimulus diameter would reduce the normal variation of DLS with stimulus location (eccentricity).
Methods: :
We tested one eye each of 33 perimetrically–experienced healthy volunteers (as part of a continuing study, n=96). Primary testing was on a custom–built CRT display system using luminance increments and a circular stimulus scaled in size by estimates of location–specific normal ganglion cell density derived from histological data (Curcio CA, Allen KA. Topography of ganglion cells in human retina. J Comp Neurol. 1990;300:5–25).The stimulus size varied from 0.17 degrees in the macula to 0.84 degrees at an eccentricity of 30 degrees. DLS, in L–1, was measured for 24–2 locations, with a strategy similar to Full–Threshold. All subjects also underwent testing with the Goldmann size III stimulus and the SITA Standard Strategy on the Humphrey Field Analyzer. The change in DLS with eccentricity for the scaled stimulus was analysed and compared to the Goldmann size III stimulus.
Results: :
The preliminary analyses for the first 33 eyes are presented. The mean age of subjects was 65 years (SD 9 years, range 47–79 years). With the Goldmann size III stimulus, mean DLS varied from 1500 L–1 (31.8dB) at 10 degrees eccentricity to 540 L–1 (27.3dB) at 30 degrees. With the scaled stimulus, corresponding DLS values were 570 L–1(27.6dB) and 730 L–1 (28.6dB), respectively. There was a linear variation of mean DLS with eccentricity for both the Goldmann size III stimulus (r=–0.99, p<0.001) and the scaled stimulus (r=0.80, p=0.02) but the dependence of DLS on eccentricity was greater (Z=7.75, p<0.001) for the Goldmann size III (slope = –61 L–1/ deg) compared to the scaled stimulus (slope = 18 L–1/ deg).
Conclusions: :
Early indications suggests that the use of a scaled stimulus gives rise to less sensitivity variation across eccentricities, potentially creating more uniform structure / function slopes across locations.