Purpose:
To compare standard automated perimetry (SAP), short-wavelength automated perimetry (SWAP) and frequency-doubling technology (FDT) using 3 definitions of visual field (VF) abnormality: unconfirmed abnormality, abnormality confirmed by the same test, and abnormality confirmed by a different test.
Methods:
One eye of each of 174 glaucoma patients and 164 age-matched healthy controls from the Diagnostic Innovations in Glaucoma Study were included (N=338). Evidence of glaucomatous optic neuropathy on stereophotographs was used to classify the eyes. Each participant had 2 reliable 24-2 tests on SAP-SITA, SWAP-SITA and Matrix FDT performed within 3 months. Receiver operating characteristic curves were generated for the PSD of each SAP, SWAP and FDT test and used to identify the VF abnormality criteria yielding 80% specificity. Three analyses were performed. 1. Percentage of confirmation of abnormality was defined as the % of abnormal VFs on the initial test that were also abnormal on a confirmatory test. 2. Agreement between the initial and confirmatory tests was assessed with the Kappa statistic. 3. Specificity and sensitivity were obtained by comparing the instrument-derived PSD result (p<5%) to the stereophotograph classification.
Results:
1. Of the 338 VFs, the initial test was abnormal for 121 SAPs, 127 SWAPs and 112 FDTs. Confirmation of abnormality rates ranged from 64-75%. 2. The agreement was substantial when the same test was used but moderate when a different test was used. 3. Specificity for unconfirmed tests ranged from 60-76% and improved with confirmation, ranging from 74-87%.
Conclusions:
1. Confirmation of abnormality is similar whether the same test or a different test is used to confirm. 2. Better agreement is achieved when the same test is performed twice, compared to when different test types are used. 3. Confirming VF results improves specificity, which is important for glaucoma diagnosis. Confirming with FDT produced the highest specificities.
Keywords: visual fields • perimetry