Andrew Turpin, School of Computer Science and Information Technology, RMIT University, Melbourne, Australia
Allison McKendrick, Department of Optometry and Vision Science, University of Melbourne, Melbourne, Australia
Balwantray Chauhan, Department of Ophthalmology Dalhousie University, Canada
This was our first application of the theoretical derivation of individual confidence limits presented in Vision Research, 2005.
AIM: To investigate if information from individual patient's psychometric functions (PFs) can be used to improve the detection of progression of visual field deficits in individuals with glaucoma.
METHODS: In this study we used data from 9 primary open-angle glaucoma patients that had undergone Standard Automated Perimetry (30-2, Full Threshold, Humphrey Field Analyzer) twice a year for 5-12 years (mean 8.3 years). These patients also had PFs measured for the same perimetry stimuli using a short MOCS procedure in six locations of their fields at their first visit. This PF information was used to derive false response rates (averaged over all six locations) and slope as a function of threshold (linear regression on the six available slope-threshold pairs) for each patient. In turn, this information was used to derive individual probability of change (IPoC) maps [1] and Glaucoma Change Probability (GCP) maps like those produced by some commercial perimeters. A patient was determined as progressing using both a conservative criteria (CONS): 2 locations flagged 2 times in a row for IPoC, 4 locations flagged 3 times in a row for GCP; and a liberal criteria (LIB): 1 location flagged 2 times in a row for IPoC, 4 locations flagged 2 times out of 3 consecutive visits for GCP.
RESULTS: CONS-GCP and CONS-IPoC agreed in 7 of 9 patients, with CONS-IPOC making the determination at least 2 visits before CONS-GCP. LIB-IPoC and LIB-GCP agreed on 8 of 9 patients, with LIB-IPoC calling progression earlier than LIB-GCP for 2, and vice-versa in 3. In 6 patients IPoC flagged significantly less points in each field (p<0.05), and flagged the same number in the other 3 patients.
CONCLUSION: IPoC gives the same results as GCP using fewer field measurements. IPoC flags less points than GCP, making clinical interpretation less arduous. PF information measured at baseline can assist in determining change in visual fields over time.