Interpretation of the Humphrey Matrix 24-2 test in the diagnosis of preperimetric glaucoma

Purpose: To define the diagnostic precision of the Humphrey Matrix 24-2 test and to determine a clinically applicable cutoff criterion for the inclusion of preperimetric glaucoma. Methods: Eyes with preperimetric glaucoma (99 eyes) were defined as having normal standard automated perimetry (SAP) results and a glaucomatous optic nerve head, along with one or more optical coherence tomography (OCT) parameters flagged as <0.05 or a retinal nerve fiber layer (RNFL) defect on red-free fundus photographs. Normal controls (122 eyes) were defined as having a normal optic disc and RNFL with normal SAP. Receiver-operating characteristic (ROC) curves were constructed to determine the best-discriminating parameter of the frequency-doubling technology (FDT) Matrix. The sensitivity, specificity, and hit ratio of all available criteria were calculated from the FDT Matrix. Results: Pattern standard deviation (PSD) showed a larger area under the ROC curve (AUC) (0.767) than mean deviation (MD) (0.619). A pattern deviation plot (PDP) analysis by modified Anderson criteria showed the highest discriminating power, with a sensitivity of 75.76%, followed by a PSD cutoff value of 3.14 dB (sensitivity, 73.74%). Conclusions: FDT Matrix seems to be a valuable clinical tool in the detection of preperimetric glaucoma. Among the parameters of the FDT Matrix, PDP analysis by modified Anderson criteria may provide a reliable cutoff criterion.