Purchase this article with an account.
Katherine Makedonsky, Michael H. Chen, Patricia Sha, Gary C Lee, Mary K Durbin; Comparison in optic nerve cupping between optical coherence tomography and widefield slit scanning ophthalmoscopy. Invest. Ophthalmol. Vis. Sci. 2018;59(9):2088.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Cup to disc ratio (CDR) measurements are important for monitoring glaucoma and interclinician variability in grading cupping has been noted in the past. The best clinical examination of the optic nerve is with a dilated stereoscopic exam on a slit lamp using a 60, 78 or 90 diopter lens. However this does not provide a baseline for future reference; thus OCT and fundus photographs are a useful adjunct. The purpose of this study is to compare the clinical evaluation of CDR on true color wide field CLARUS™ 500 (ZEISS, Dublin, CA) fundus images with the CDR obtained with the CIRRUS™ HD-OCT 5000 (ZEISS, Dublin, CA).
True color wide field images were obtained with CLARUS 500 and optic nerve cube 200x200 scans were obtained with CIRRUS HD-OCT 5000. Three optometrists used the caliper tool to measure the vertical length of the cup and vertical length of the disc on the CLARUS. The vertical cup to disc ratio (vCDR) was compared with the vCDR calculated by the CIRRUS device. Starting in CIRRUS SW version 5.0, disc measurement on CIRRUS is determined by defining the disc margin in reference to the Bruch’s membrane opening (BMO). The cup boundary is determined by finding the optimal points on the ILM that minimize the cross-sectional neuroretinal rim area, as described previously [Mwanza AJO 2017].
VCDR for both methods as well as interclinician variability is shown in Figure 1. ANOVA test was run to determine interclinician variability in CDR grading and showed good agreement (ICC=0.70). It was found that interclinician variability was present, however was not statistically significant (p= 0.108). More interclinician variability was found with cups that were shallow and small compared to those that were deep and large. A two tail paired t-test was run to determine the difference between average cupping determined by graders compared to CIRRUS (p=0.003).
When looking at the sample size as a whole, vCDR estimates differed between the CLARUS and the CIRRUS. However, glaucoma results in large cupping and both CIRRUS and graded CLARUS images were in strong agreement in judging the CDR for larger CDR values. For accurate monitoring, both CLARUS and CIRRUS images should be obtained for small or shallow cups. However, for larger cups, the clinician may decide which method they prefer as CIRRUS and CLARUS are comparable.
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