Confirmation Of "Likely Progression" In Standard Automated Perimetry Using Guided Progression Analysis

Abstract

Purpose: : To determine the ability of standard automated perimetry (SAP) guided progression analysis (GPA) to confirm "likely progression" in subsequent tests.

Methods: : 85 eyes from 69 glaucoma patients were included in this observational cohort study. Patients were recruited from the Diagnostic Innovations in Glaucoma Study (DIGS) and from the African Descent and Glaucoma Study (ADAGES). To be included, patients were required to have a minimum of 3 years of follow-up and at least 8 reliable SAP tests. All eyes had "likely progression" as defined by the GPA, which was based on the presence of 3 or more points with significant change compared to the baseline visual fields and repeatable in 3 tests. We then evaluated all available visual fields after the initial diagnosis of "likely progression" by the GPA, in order to assess the proportion of fields in which this diagnosis was subsequently confirmed or not. Univariable Cox models were built to evaluate the predictive ability of age, baseline mean deviation, baseline pattern standard deviation (PSD), central corneal thickness and baseline intraocular pressure in predicting those eyes that had confirmation of "likely progression" in all subsequent fields.

Results: : The average follow-up time was 10 years and the average number of subsequent visual fields performed after the initial diagnosis of "likely progression" was 6 (range 3-11). 36 of 85 eyes (42.9%) confirmed "likely progression" on all subsequent SAP tests while 41 of 85 eyes (48.2%) confirmed "likely progression" on at least 75% of subsequent SAP tests and 54 of 85 eyes (63.5%) confirmed "likely progression" on at least 50% of subsequent SAP tests. 10 of 85 eyes (11.8%) had no confirmation of "likely progression" in any of the subsequent visual fields. The only baseline characteristic that was predictive of confirmed "likely progression" in all subsequent visual fields was the SAP PSD (hazard ratio = 1.24, 95% CI (1.02-1.50) per 1dB).

Conclusions: : "Likely progression" as determined by SAP GPA is not always confirmed on subsequent visual field testing. However, patients with a worse baseline SAP PSD were more likely to have confirmed "likely progression".

Clinical Trial: : http://www.clinicaltrials.gov NCT00221897

Keywords: visual fields • perimetry