Effect of treatment on the rate of visual field change in the ocular hypertension treatment study observation group

Abstract

Purpose: The goal in this study was to compare rates of visual field (VF) change before and after the initiation of treatment in participants originally randomized to the observation arm of the Ocular Hypertension Treatment Study (OHTS).

Methods: We included OHTS participants originally randomized to observation and excluded those who reached non-POAG endpoints. VF progression was determined using trend analysis. Global and localized rates of VF change were calculated based on linear regression over time of mean deviation (MD) and threshold sensitivity values for each test location. MD rates (MDR) and pointwise linear regression (PLR) analysis were also assessed using six VF tests before and after the initiation of treatment. A PLR endpoint was defined as a VF test location progressing faster than -0.5 dB/year at P < 0.01.

Results: We included 780 eyes from 432 OHTS participants. Following the initiation of treatment, the mean MDR decreased from -0.23 ± 0.6 to -0.06 ± 0.5 dB/year (P < 0.01) and the number of VF locations reaching a PLR endpoint decreased from 2.13 ± 6.0 to 1.00 ± 4.0 (P < 0.01). The benefit of treatment was significant both among participants who did not convert (-0.17 ± 0.6 vs. -0.01 ± 0.5 dB/year, P < 0.01) and among those who converted to glaucoma (-0.51 ± 0.8 vs. -0.27 ± 0.7 dB/year, P < 0.01) based on the OHTS event-based endpoint.

Conclusions: The initiation of ocular hypotensive medication among OHTS participants originally randomized to observation significantly reduced the velocity of VF progression.

Trial registration: ClinicalTrials.gov NCT00000125.

Figures

Figure 1.

Comparison of changes in MDRs before (x-axis) and after (y-axis) treatment. (A) All participants are included. (B) Only participants who did not reach a POAG endpoint are included. (C) Only participants who reached a POAG endpoint are included. The dark grey lines show flat slopes (0.0 dB/year) in both axes, and the light grey lines show the average change in MDR before and after treatment.

Figure 2.

Comparison between changes in absolute IOP (x-axis) and changes in MDRs of decline (y-axis) following treatment. The black line corresponds to the best-fit linear regression line. (A) All participants are included (P = 0.005). (B) Only participants who reached a POAG endpoint are included (P < 0.0001).