Abstract
Purpose: :
To evaluate the inter–eye spatial relationship of standard automated perimetry (SAP) defects over time in glaucoma patients.
Methods: :
282 eyes of 141 patients with GON and at least 5 reliable SAP visual fields in both eyes were included. The eye with the greater number of defective (p<5%) pattern deviation (PD) locations at baseline was designated the "worse" eye. The number of overlapping and non–overlapping defective locations on the PD plot were calculated by comparing the baseline worse eye visual field to each "better" eye field. Concordance was the ratio of overlapping defective locations to the total number of defective locations. Slopes from linear regression of variables by years of follow–up estimated change over time. Longitudinal inter–eye concordance slopes were compared to longitudinal intra–eye and randomly–matched longitudinal concordance slopes. The effects of 3 baseline variables (number of defective locations, inter–eye concordance and inter–eye defect disparity) and change in inter–eye defect disparity on longitudinal inter–eye concordance slopes were evaluated in a multiple regression model. Inter–eye disparity was the difference in the number of defective locations between right and left eyes.
Results: :
Patients had an average+/–SD of 7.6+/–2.3 fields over 4.8+/–1.3 years. Longitudinal inter–eye concordance slopes (0.007+/–0.04) were significantly more positive (i.e., concordance increased over time) than longitudinal intra–eye concordances (–0.01+/–0.07) or randomly–matched longitudinal concordance (–0.007+/–0.04). Assuming that an inter–eye concordance slope of 0.01 represents change, concordance was relatively stable in 42% (60 patients), increased in 35% (49 patients), and decreased in 23% (32 patients) of patients. Baseline inter–eye concordance (0.2+/–0.2), baseline number of defective PD locations (19.6+/–14.9), and inter–eye defect disparity slope (0.3+/–1.4) significantly affected inter–eye longitudinal concordance rates.
Conclusions: :
Concordance in the spatial location of defects between the worse eye at baseline and the other eye follow–up tests was relatively stable in the majority of patients. Predicting the location of visual field loss in a better eye of a patient based on the worse eye defect is complicated by baseline factors and the change in the original disparity defect size between the eyes.
Keywords: perimetry • visual fields