Reproducibility of Viual Field End Point Criteria for tandard Automated Perimetry, Full-Threhold, and wedih Interactive Threholding Algorithm trategie: Diagnotic Innovation in Glaucoma tudy

Purpoe To compare the interthrehold and intrathrehold trategy agreement of viual field end point criteria for tandard automated perimetry (AP) with the full-threhold (FT) algorithm and the wedih interactive threhold algorithm (ITA). Deign Propective, longitudinal cohort tudy. Method The intertrategy group included a randomly elected eye of 173 participant in the Diagnotic Innovation in Glaucoma tudy who had undergone FT algorithm and ITA analyi within three month (equence 1, FT + ITA). Intratrategy agreement for the FT algorithm (equence 2, FT + FT) wa teted for 44 (25.4%) participant who had undergone FT analyi within one year of the FT ued in the intertrategy pairing, and for 89 patient (51.4%) who had undergone ITA analyi within one year before (equence 3, ITA + ITA). Four different end point criteria uing tatpac II indice were teted. Intertrategy agreement wa compared with intratrategy agreement uing κ tatitic. Reult FT + ITA agreement (κ) for pattern tandard deviation (PD) < 1% wa 0.82; for PD < 5%, the κ value wa 0.64; and for four or more pattern deviation plot point, the κ value wa 0.43. Agreement with glaucoma hemifield tet (GHT) reult wa ignificantly higher (P < .01) for FT + FT (κ = 0.94) than FT + ITA (κ = 0.67), and approached ignificance (P = .07) when comparing FT + FT with ITA + ITA (κ = 0.77). GHT reult were more likely to be abnormal on the ITA analyi than on the FT analyi. No other ignificant difference were found. Concluion To minimize miinterpreting abnormal GHT reult on ITA a evidence of change when witching trategie, both ITA and FT hould be performed and compared within a hort period. Other indice are comparable between the two trategie.