Abstract
Purpose:
To describe the extent to which an established target intraocular pressure (IOP) was met during treatment, and to test the relationship between meeting target and visual field outcomes in the Collaborative Initial Glaucoma Treatment Study (CIGTS) participants.
Methods:
607 subjects with newly diagnosed open-angle glaucoma were enrolled in the CIGTS. A study eye was selected prior to treatment initiation for participants with two treatment-eligible eyes, and a target IOP was set based on a formula that accounted for pre-treatment IOP and visual field (VF) severity as measured by the mean deviation (MD). Measures of meeting the target IOP were tested for associations with VF loss over time. Using a previously published model for predicting VF loss, the predictive value of adding measures of target IOP adherence were assessed by means of repeated measures linear regression.
Results:
Initial target IOPs ranged from 13 to 25 mmHg. During follow-up, the measured IOP was within ± 2 mmHg of the target IOP in 42% of visits, >2 mmHg under target IOP in 44% of visits, and >2 mmHg over target in 14% of visits. IOP ≥5 mmHg over target IOP was found in 4% of visits. The visit-specific difference between the measured IOP and the target IOP, whether evaluated as a continuous or categorical variable, was not a significant predictor of MD over time when added to the previously published model. The cumulative percentage of visits in which the measured IOP was greater than target IOP was also not a significant predictor of MD. However, indices of variability in the difference between measured and target IOP over time (standard deviation, maximum, and range) were significant predictors of MD. Larger variation was associated with worse VF. Even so, these variables were no better at predicting VF than measured IOP and its variation.
Conclusions:
The extent to which measured and target IOP differed was not predictive of VF progression. Variability in this difference over time was predictive of VF progression, but not more predictive than variability around each patient’s mean. While setting and meeting a specific target IOP was not associated with further VF loss, quite probably the fact that patients were given a target forced CIGTS ophthalmologists to treat as aggressively as necessary to meet it, thereby reducing overall VF loss in the Study.