The ability of eye tracking perimetry to detect glaucoma

Abstract

AbstractPurposeTo examine the diagnostic ability of saccadic vector optokinetic perimetry (SVOP), a new eye tracking perimeter, by assessing its capacity to differentiate healthy and glaucomatous participants, and to compare performance to standard automated perimetry (SAP).MethodsA prospective case‐control study including patients with glaucoma and healthy volunteers recruited from the glaucoma clinic at a university hospital and from the Scottish Health Research Register (SHARE). Participants underwent a comprehensive ophthalmic examination and performed SAP and SVOP in both eyes with the order of testing randomised.Results61 eyes of 31 participants with glaucoma and 47 eyes of 24 healthy participants were included in the study. Average SAP mean deviation (MD) in glaucomatous eyes was ‐8.72 ± 7.37 dB, with average SAP and SVOP sensitivities of 23.28 ± 0.89 dB and 22.1 ± 4.25 dB respectively. There was strong correlation between average SAP and SVOP sensitivities (r = 0.951 , P<0.001). On average, SAP sensitivity was 1.16 ± 1.43 dB higher than SVOP sensitivity, with 95% limits of agreement of ‐1.64 to 3.96 dB. SVOP average sensitivity had excellent ability to differentiate healthy and glaucomatous eyes with an AUC of 0.90 (95% CI 0.84 to 0.96), which was not significantly different to the performance of SAP average sensitivity (AUC = 0.92, 95% CI .0.86 to 0.97, P = 0.60) (Figure 6). SVOP had 85.2% sensitivity for 80% specificity, compared to 86.9% sensitivity for 80% specificity for SAP.ConclusionsThe results of this study suggest that eye tracking perimetry using SVOP could offer a feasible alternative to SAP for detection of glaucoma. Overall, there was good agreement between differential light sensitivity values measured using SVOP and those measured using SAP, and average SVOP and SAP sensitivities had similarly good ability to differentiate glaucomatous and healthy eyes.