PURPOSE: To evaluate and compare the diagnostic performance of Humphrey Matrix frequency doubling technology perimetry (Matrix) global indices with standard automated perimetry (SAP) for glaucoma discrimination. METHODS: Forty-seven healthy and 63 glaucomatous subjects were included in this study. Glaucoma was defined as having glaucomatous optic disc and glaucomatous visual field defect. Correlations of mean deviation (MD) and pattern standard deviation (PSD) between Matrix and SAP were evaluated. Areas under receiver operating characteristic curves (AUCs) for discriminating healthy from glaucoma, sensitivity, and cut-off value at fixed specificity of MD and PSD were determined in Matrix and SAP. RESULTS: MD and PSD from Matrix were highly correlated with SAP data in glaucomatous eyes (r=0.80, 0.69 p<0.001, <0.001). The AUCs of MD and PSD from Matrix (0.941, 0.921) were of comparable diagnostic capability to SAP data (0.876, 0.923, p=0.068, 0.927). The sensitivity at 90% specificity of MD was 67.9% in SAP, 76.4% in Matrix, with the cut off value of MD at 90% specificity at -3.10dB in SAP and -3.72dB in Matrix. CONCLUSIONS: MD and PSD data from Matrix and SAP significantly correlated in glaucomatous eyes and showed similar diagnostic performance for discriminating healthy from glaucoma however, both MD and PSD are scaled differently on SAP and Matrix, which suggests that application of these parameters in a manner similar to that used in SAP should be employed with caution.
Area Under Curve ; Eye ; Glaucoma ; ROC Curve ; Sensitivity and Specificity ; Visual Field Tests ; Visual Fields