Abstract:
The aim of this study was to determine, in stable glaucoma, the medium-term fluctuation (MF) of the perimetric threshold estimate, i.e. the variability occurring between a given examination and an immediate confirmatory examination, undertaken after a rest period. Thereby separation of field loss from fluctuation in sensitivity is facilitated.
Patients and methods: Forty-one eyes of 41 stable glaucoma patients underwent automated threshold perimetry using the Tübingen Computer Campimeter (TCC). Each visit consisted of two sub-examinations, undertaken after a rest period, each with a separate stimulus grid within 30° eccentricity. Nine stimulus locations were common to both grids. The procedure was repeated after 224 ± 80 days (mean ± SD) and again after a further 228 ± 62 days. MF was calculated for the nine common reference locations from each pair of sub-examinations using a multiple-factor analysis of variance (ANOVA). Additionally, the short-term fluctuation and the homogenous (LF[Ho]) and heterogeneous (LF[He]) long-term fluctuation were determined.
Results: The Standard Deviations of the effects in the ANOVA were: MF 0.72dB, SF 1.35dB, LF(Ho) 0.74dB, and LF(He) 1.34dB. The MF contributed 1.1% of the total variation whilst the SF, LF(Ho) and LF(He) contributed 3.8%, 1.2% and 3.8%, respectively. Systematic influences (eccentricity and position of the reference locations in the visual field) accounted for 61.7%. Individual influences (age, gender, etc.) contributed 28.5%. The medians of the frequency distributions over all three visits were: MF 0.47 dB (90% Reference Interval: 0.19 dB - 1.23 dB), SF 1.01 dB (0.55 dB - 2.15 dB), LF(Ho) 0.57 dB (0.14 - 1.44 dB), LF(He) 0.93 dB (0.49 - 2.20 dB).
Conclusion: The MF is small compared to the other components of variance. It serves as a reliability index between two visual field examinations, separated by a rest period, undertaken on the same eye to enhance the detection of field loss and/or of progressive loss.