Variability of scan quality and perfusion density in longitudinal optical coherence tomography angiography imaging

• Published in: British journal of ophthalmology Vol. 108; no. 7; pp. 978 - 983
• Main Authors: Smith, Corey A, Josey, Vanessa L, West, Michael E, Dyachok, Oksana M, Sharpe, Glen P, Vianna, Jayme R, Rafuse, Paul E, Shuba, Lesya M, Nicolela, Marcelo T, Chauhan, Balwantray C
• Format: Journal Article
• Language: English
• Published: BMA House, Tavistock Square, London, WC1H 9JR BMJ Publishing Group Ltd 01.07.2024; BMJ Publishing Group LTD
• Subjects: Aged; Algorithms; Astigmatism; Female; Fluorescein Angiography - methods; Follow-Up Studies; Glaucoma; Glaucoma, Open-Angle - diagnosis; Glaucoma, Open-Angle - diagnostic imaging; Glaucoma, Open-Angle - physiopathology; Humans; Imaging; Intraocular Pressure - physiology; Longitudinal Studies; Male; Medical imaging; Middle Aged; Optic Disk - blood supply; Optic Disk - diagnostic imaging; Reproducibility; Reproducibility of Results; Retina; Retinal Ganglion Cells - pathology; Retinal Vessels - diagnostic imaging; Software; Systemic diseases; Tomography; Tomography, Optical Coherence - methods; Visual acuity; Visual Fields - physiology; Canada; United States > US; Massachusetts; Halifax Nova Scotia Canada; Glaucoma; Imaging
• ISSN: 0007-1161; 1468-2079; 1468-2079
• DOI: 10.1136/bjo-2022-322979

Background/aimsOptical coherence tomography angiography (OCT-A) images are subject to variability, but the extent to which learning impacts OCT-A measurements is unknown. We determined whether there is a learning effect in glaucoma patients and healthy controls imaged with OCT-A.MethodsNinety-one open-angle glaucoma patients and 54 healthy controls were imaged every 4 months over a period of approximately 1 year in this longitudinal cohort study. We analysed 15°×15° scans, centred on the fovea, in one eye of each participant. Two-dimensional projection images for the superficial, intermediate and deep vascular plexuses were exported and binarised after which perfusion density was calculated. Linear mixed-effects models were used to investigate the association between perfusion density and follow-up time.ResultsThe mean (SD) age of glaucoma patients and healthy controls was 67.3 (8.1) years and 62.1 (9.0) years, respectively. There was a significant correlation between perfusion density and scan quality in both glaucoma patients (r=0.50 (95% CI 0.42 to 0.58); p<0.05) and healthy controls (r=0.41 (95% CI 0.29 to 0.52); p<0.05). An increase in perfusion density occurred over time and persisted, even after adjustment for scan quality (1.75% per year (95% CI 1.14 to 2.37), p<0.01).ConclusionsPerfusion density measurements are subject to increasing experience of either the operator or participant, or a combination of both. These findings have implications for the interpretation of longitudinal measurements with OCT-A.