Detecting glaucomatous progression using GDx with variable and enhanced corneal compensation using Guided Progression Analysis.
OBJECTIVE: To compare detection of glaucoma progression with scanning laser polarimetry using two methods for corneal compensation. METHODS: Normal, glaucoma suspects and glaucoma patients with 36 months' follow-up meeting the eligibility criteria were prospectively enrolled. All subjects underwent complete eye exam, standard automated perimetry (SAP) and scanning laser polarimetry with variable and enhanced corneal compensation (GDxVCC, GDxECC). SAP progression was determined using the visual-field index (VFI). GDx progression was determined using Guided Progression Analysis software (GDxGPA) and was defined as a repeatable change on two consecutive scans compared with two baseline images using any of three strategies: ≥ 150 contiguous pixels on the image progression map (A), four or more adjacent segments on the Temporal Superior Nasal Inferior Temporal graph (B) or a significant change in slope of the summary parameter chart (C). Kappa statistics and logistic regression were used for the analysis. RESULTS: Thirteen normal, 30 glaucoma suspect and 25 glaucomatous eyes participating in the Advanced Imaging in Glaucoma Study were included. Progression was identified in six eyes (8.8%) using GDxVCC and in eight eyes (11.8%) using GDxECC. SAP progression was detected in seven (10.3%) eyes. Agreement among progression methods using GDxVCC and GDxECC was strongest for method C (kappa=0.57, p=0.002) compared with methods A (kappa=0.41, p=0.01) and B (kappa=0.41, p=0.01). The association between typical scan score (TSS) and overall or individual methods of progression was not significant using VCC or ECC (p>0.05). CONCLUSIONS: GDxGPA represents a novel approach for detection of glaucomatous progression. GDxVCC and GDxECC demonstrate moderate agreement.
Grewal, DS; Sehi, M; Greenfield, DS
Volume / Issue
Start / End Page
Pubmed Central ID
Electronic International Standard Serial Number (EISSN)
Digital Object Identifier (DOI)