Purchase this article with an account.
Hongli Yang, Haomin Luo, Christy Hardin, Cindy Albert, Danielle Millay, Jayme R Vianna, Glen Sharpe, Galen Williams, Shaban Demirel, Brad Fortune, Stuart Keith Gardiner, Balwantray C Chauhan, Claude F Burgoyne; The Relative Ability of Optical Coherence Tomography (OCT) Structural Parameters to Detect Structural Abnormality in Glaucoma Suspect and Glaucoma Eyes. Invest. Ophthalmol. Vis. Sci. 2018;59(9):4063.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To compare the sensitivity of OCT measurements of minimum rim width (MRW), retinal nerve fiber layer thickness (RNFLT), and central laminar depth (LD) relative to Bruch’s membrane opening (LD-BMO), anterior scleral canal opening (LD-ASCO), peripapillary Bruch’s membrane (LD-BM) and sclera (LD-SL), to detect structural abnormality in glaucoma suspect (GLS) and glaucoma (GL) eyes.
147 GL eyes with glaucomatous visual field (VF; standard automated perimetry pattern standard deviation or glaucoma hemifield test outside normal limits on the last three visits), 97 GLS eyes with normal VF, and 362 Normal eyes, underwent enhanced depth OCT imaging of the optic nerve head and peripapillary RNFL (Spectralis, Heidelberg Engineering). For each eye, manual segmentation of the internal limiting membrane, anterior laminar surface, Bruch’s membrane, BMO, peripapillary scleral surface and ASCO were performed using custom software, and the above parameters were quantified. MRW and RNFLT were corrected to age 70 using data from the Normal eyes. For each parameter, receiver operating characteristic (ROC) curves were plotted and compared.
The mean (range) mean deviation was 0.4 dB (−3.8 to +2.9) for GLS eyes and -4.2 dB (−17.5 to 1.3) for GL eyes. At least one structural parameter was abnormal (p<5%) in 17% of normal eyes. In GLS eyes (Fig. 1), at criteria that achieved a specificity of 94%, structural abnormality was detected by MRW (29.9% of eyes), LD-SL (25.8%) (the best-performing LD measure), RNFLT (17.5%) and “any 2 parameters abnormal” (24.7%). In GL eyes (Fig. 2), sensitivities at criteria that achieved 94% specificity were 77.6% for RNFLT, 67.3% for MRW, 34.0% for LD-SL, and 67.3% for “any 2 parameters abnormal”. The partial areas under the ROC curve for 100%-80% specificity in GLS eyes was 0.07 for MRW, 0.05 for RNFLT and 0.06 for LD-SL, (not significantly different; p>0.05). In GL eyes, these were 0.14 for MRW, 0.16 for RNFLT and 0.08 for LD-SL (smaller than for MRW or RNFLT; p<0.001).
In glaucoma suspect eyes, LD-SL and MRW may detect structural abnormality more frequently than RFNLT. In glaucoma eyes, RNFLT and MRW both detected structural abnormality more frequently than lamina depth. Among laminar depth parameters, LD-SL may outperform all other LD parameters.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.
This PDF is available to Subscribers Only