Comparison of a Novel Head-mounted Objective Auto-perimetry (Gaze Analyzing Perimeter) and Humphrey Field Analyzer.

ABSTRACT

PURPOSE: To evaluate the agreement between 24-2 visual field (VF) test results obtained using the gaze analyzing perimeter (GAP; FINDEX, Tokyo, Japan) and the Humphrey field analyzer (HFA; Carl Zeiss Meditec, Dublin, CA, USA). DESIGN: Cross-sectional study PARTICIPANTS: Patients who underwent HFA 24-2 for suspected or confirmed VF loss and were treated at the Kyoto University Hospital between December 2022 and July 2023. METHODS: Patients underwent consecutive VF tests on the same eye using HFA and GAP 24-2 tests. Bland-Altman analysis was used to compare GAP and HFA results. Examination points where the sensitivity measured using GAP was ≥10 dB higher than that measured using HFA were reevaluated by referring back to the original gaze data; two ophthalmologists assessed whether the gaze moved linearly toward the new test target. MAIN OUTCOME MEASURES: Mean deviation (MD) and elapsed time on an individual basis and sensitivity on an examination point basis. RESULTS: Forty-seven eyes of 47 patients were analyzed. The correlation coefficient of the MD using HFA and GAP was 0.811 (95% confidence interval [CI] 0.683-0.891). Bland-Altman analysis showed good agreement between HFA and GAP tests. The mean difference (95% limits of agreement [LOA]) in MD between HFA and GAP results was -0.63 dB (-5.81 to 4.54 dB). Although no statistically significant differences were observed in the elapsed time (P = 0.99), measurements completed within 200 s were observed only in the GAP group (11 cases, 23.4%), who had significantly better HFA MD value than others (P=0.001). On an examination point basis for sensitivity, the correlation coefficient between HFA and GAP was 0.691 (95% LOA, 0.670-0.711). Original gaze data assessment revealed that the gaze moved linearly toward the new test target for 70.2% of the examination points with a sensitivity discrepancy. CONCLUSIONS: The results indicate that the GAP provides VF assessment outcomes comparable to those of the HFA. The GAP exhibited advantages in terms of testing time, particularly in patients with minimal VF impairment. Furthermore, the GAP records all eye movements, enabling the objective determination of VF abnormalities based on gaze patterns and facilitating easy post-hoc verification.