Abstract
Abstract: :
Purpose:We previously demonstrated at ARVO the effects thatpointwise linear regression (PLR) criteria had on detectionof change in a visual field series. Here we use validated criteriato measure change over a longer field series.Methods:Visualfields from 127 patients, undergoing trabeculectomy, and followedfor 40 months were analysed using PLR. Change was assessed over6, 8, 10, or 12 fields, and over the first and last 6 fields.The criterion for change was 2 points changing in a perimetricnerve fibre bundle cluster with significance ranging from P<0.001to 0.05 and a slope= 1dB/yr.Results:Mean deviation at surgerywas -9.9dB and mean duration from diagnosis to surgery was 55months. At all significance levels, detection of progressionwas greater than improvement and increased at a greater ratewith increased follow up (table 1).
Table 1 Analysis over the firstand second 6 fields detected similar levels of improvement.Numbers of eyes showing progression were higher over the first6 fields.Progression and improvement occurred at similar levelsover the second 6 fields (Table 2).Table 2 Conclusion:Following trabeculectomy visual field progression was greaterthan improvement and increased at a greater rate than improvementwith time. However much of this difference is attributable toearly change over the first 6 months. If change occurs later,different methods will be required to assess it.
Keywords: 624 visual fields • 359 clinical research methodology