Glaucoma diagnostic performance of humphrey matrix and standard automated perimetry

Abstract

To evaluate and compare the performance of Humphrey Matrix perimetry (Matrix) and standard automated perimetry (SAP) for glaucoma discrimination. Forty-seven healthy and 68 glaucomatous subjects were included in this study. Glaucoma was defined as having a glaucomatous optic disc and a nerve fiber indicator (NFI) GDx VCC scanning laser polarimetry score of >40. Glaucomatous subjects were classified into two groups, early (40 < NFI < or = 50) and moderate to advanced (NFI > 50). The number of clusters on Matrix and SAP pattern deviation maps were estimated and scored. Areas under the receiver operating characteristic curves (AUCs) of the cluster scores were calculated for early and moderate to advanced stages of glaucoma and compared between Matrix and SAP. Among 68 glaucomatous subjects, 24 eyes were classified as having early and 44 eyes as having moderate to advanced glaucoma. The overall AUC of the Matrix cluster score was comparable to that of SAP (0.857, 0.881, P = 0.538). The AUC of the cluster score did not reveal statistically significant differences between Matrix and SAP for either early or moderate to advanced stages of glaucoma (P = 0.831, 0.237). Both Matrix and SAP showed good diagnostic performance with glaucoma defined as structural loss. Matrix and SAP data showed similar discrimination capability for different stages of glaucoma determined by cluster analysis.