Diagnostic accuracy of optic disc, macular and retinal nerve fiber layer analysis by means of SD-OCT at different stages of the glaucoma disease: Multicentric Italian Glaucoma Imaging Study (MIGIS).

Abstract

Purpose : To evaluate the diagnostic ability of the optic disc (ONH), peripapillary retinal nerve fiber layer (pRNFL) and macular analysis with Spectral Domain-OCT, in discriminating healthy from glaucoma at different stages of the disease.

Methods : Controls and glaucoma were enrolled from the Multicentric Italian Glaucoma Imaging Study involving 6 italian glaucoma clinics. All patients underwent a full eye examination, standard perimetry and imaging of ONH, pRNFL, and Posterior Pole (PP) with Spectralis SD-OCT (Heidelberg Engineering, Heidelberg, Germany). Abnormal and repeatable visual field test were used to classify glaucomatous eyes. For PP analysis we considered the thickness of the whole retina and the thickness of the first three innermost layers, mRFNL, mGCL and mIPL. The analysis was conducted on the whole sample and after stratification of the patients according to the disease severity (Hodapp classification). Area under the ROC curve (AROC) and sensitivity (Sn) at ≥90% fixed specificity were calculated for each parameter. The Henley-McNeil test was used to compare AROC of different parameters. P<0.05 was considered statistically significant.

Results : 182 glaucoma (mean age 65±10 years) and 89 healthy eyes (mean age 64±7 years ) were enrolled. Early glaucoma were 68 (MD, -3.03±1.58dB), moderate were 58 (MD, -9.2±1.9dB), advanced were 56 (MD, -15.5±2.9dB). In the overall analysis pRNFL, minimum rim width (MRW) and mRNFL showed similar AROCs (0.95, 0.94, 0.94, p=0.4). MRW showed the highest Sn in the early stage (89%) followed by mRNFL (82%). mRNFL and pRNFL showed the highest Sn in the moderate stage (96% and 94%). mRNFL showed the highest Sn in the advanced stage (100%) followed by pRNFL (94%) and MRW (94%). MRW showed the most stable Sn across glaucoma stages (89%, 91% and 94% in the early, moderate and advanced stage).

Conclusions : RNFL, ONH and Macular analysis as performed by Spectralis OCT showed overall high diagnostic accuracy in detecting glaucoma.
mRNFL provided the highest accuracy among macular layers. ONH analysis by MRW was the most accurate parameter in the early glaucoma and showed the most stable diagnostic performance across glaucoma stages.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.