Authors: Herman Stubeda, Jack Quach, Jennifer Gao, Marcelo T. Nicolela, Balwantray C. Chauhan, Jayme R. Vianna

Abstract Body:

Purpose: There is currently no consensus on visual field (VF) criteria that define glaucoma. The purpose of this study was to compare the hit rates and overlap of 5 VF criteria for glaucoma: Glaucoma Hemifield Test (GHT), Hoddap-Anderson-Parrish 2 (HAP2), Foster, United Kingdom Glaucoma Treatment Study (UKGTS), and Low-pressure Glaucoma Treatment Study (LoGTS).

Study Design: Cross-sectional comparison.

Methods: We retrospectively included VF (Humphrey 24-2 SITA) and OCT (Spectralis) examinations from a 2-year period in the Glaucoma Clinic of the Queen Elizabeth II Health Science Centre, Halifax. VF and OCT exams were paired when performed within 4 months. One VF-OCT pair per randomised eye per patient was included. Eye pathologies other than glaucoma were excluded. Global and sectoral averages of optic nerve head minimum rim width, circumpapillary retinal nerve fiber layer thickness, and macular ganglion cell layer thickness were measured with OCT and considered abnormal if thinner than the 1% percentile of an age-adjusted normative value. An OCT abnormality score was created ranging from 0 (no average abnormal in any structure) to 6 (all three structures with 2 or more averages abnormal). Criteria hit rates and overlap were determined at each score. With the assumption that patients with OCT score of 0 do not have glaucoma, the criteria hit rates in this population could be an indicator of criteria specificity. The criteria hit rates in patients with higher OCT scores could be an indicator of criteria sensitivity.

Results: We included 1230 eyes, which had a mean (± standard deviation) age of 67.1 ± 12.28 years, and a mean VF mean deviation of -3.34 ± 4.78 dB.In patients with OCT score of 0 (n = 490), HAP2 (64%) and UKGTS (52%) had higher hit rates, while GHT (29%), Foster (29%), and LOGTS (16%) had lower hit rates. In patients with OCT score of 6 (n = 138), HAP2 (100%) and UKGTS (99%) had higher hit rates, while GHT (93%), Foster (83%), and LOGTS (85%) had lower hit rates. The hit rates on patients with other OCT scores followed the same trend. Of the 1030 patients flagged by at least one criterion, 428 (41%) were flagged by all the criteria.

Conclusions: Among the 5 criteria, HAP2 and UKGTS had low specificity and high sensitivity. LOGTS had a high specificity but low sensitivity. GHT had intermediary specificity and sensitivity, while Foster had intermediary specificity but low sensitivity. These results can inform the choice of VF criteria for glaucoma studies.