Abstract
Purpose: :
To assess how well SWAP-SITA, SWAP-FT and SAP-SITA outcomes match optic disc stereophotograph classifications of glaucoma, and to evaluate agreement across the visual function tests.
Methods: :
One eye of 306 glaucoma patients and 307 age-matched healthy participants from the Diagnostic Innovations in Glaucoma Study (DIGS) were included. Masked grading of optic disc stereophotographs classified the eyes as glaucomatous or normal. Stereophotographs were taken within 6 mos of the visual function tests, which were conducted within 3mos. For descriptive purposes, SAP-SITA mean deviation ±SD and pattern standard deviation was -4.40 ±6.26 and -1.00 ±1.89, and 4.33 ±3.72 and 2.06 ±1.05 for glaucomatous and normal eyes, respectively. Several parameters were assessed for each test; reported here are pattern standard deviation (PSD) and the number of pattern deviation points (PDP) triggered at <1%. Receiver operating characteristic curves were generated and the area under the curves (AUC) computed. Sensitivity at 80% and 90% specificity was calculated and agreement between visual function tests was evaluated using Kappa statistics (k).
Results: :
AUCs ranged from 0.695 ±0.020 - 0.721 ±0.020 and were not significantly different across all tests (p > 0.05). Sensitivity at 80% and 90% specificity ranged from 52.9% - 55.6% and 38.2% - 45.4%, respectively. At 80% specificity, there was substantial agreement between SWAP-SITA and SWAP-FT PSD,k=0.619, and moderate agreement between all other comparisons, k=0.517 - 0.583. At 90% specificity, there was moderate agreement between SWAP-SITA and SWAP-FT PDP, k=0.601, and substantial agreement between all other comparisons, k=0.619 - 0.703.