Diagnostic ability of spectral‐domain optical coherence tomography peripapillary retinal nerve fiber layer thickness to discriminate glaucoma patients from controls in an elderly population (The MONTRACHET study)

Abstract

Purpose To evaluate the capacity of retinal nerve fibre layer (RNFL) thickness measured by SD‐OCT to discriminate glaucoma patients from controls in an elderly population. Methods The MONTRACHET (Maculopathy, Optic Nerve, nuTRition, neurovAsCular and HEarT diseases) Study is a population‐based study including participants aged 75 years and over. All participants underwent a complete eye examination with optic nerve photographs, visual field testing and OCT peripapillary RNFL thickness measurement. Glaucoma was defined according to the ISGEO (International Society for Epidemiologic and Geographical Ophthalmology) classification. Performance indicators were calculated including area under the receiver operating characteristics curves (AUC), likelihood ratios (LR) and diagnostic odds ratios (DOR). Results In total, 1061 participants were included in the study, of whom 89 were classified as having glaucoma and 972 were classified as normal. The mean (SD) age of the population was 82.3 (3.7) years. The average RNFL thickness was significantly lower in the glaucoma group than in controls 64.0 (14.9) µm versus 88.9 (12.4) µm, respectively, p < 0.001) and in all sectors compared with controls. The average RNFL thickness had the highest AUC (0.901) followed by the temporal‐inferior (0.879) and temporal‐superior sectors (0.862). When RNFL thickness was classified as abnormal by SD‐OCT, the average RNFL thickness had the best sensitivity (83.75%) followed by the temporal‐inferior sector (75.64%). The specificity for these two parameters was 87.34% and 91.08%, respectively. The highest DOR was 28.70 for average RNFL thickness and reached 34.84 when using the reference database of the OCT manufacturer. Conclusion This study confirms that SD‐OCT could be useful as an additional test to discriminate glaucoma patients from controls in an elderly population.