Purchase this article with an account.
Sasan Moghimi, Kouros Nouri-Mahdavi, Laurie Rahardjanoto, Gina Lee, Elena Bitrian, Jay Riddle, Naveed Nilforushan, Josepe Caprioli; Spectral-Domain OCT Underestimates Disc and Rim Areas Compared to Confocal Laser Ophthalmoscopy. Invest. Ophthalmol. Vis. Sci. 2011;52(14):3062.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To compare optic disc parameters measured with confocal laser ophthalmoscopy (HRT) to those from a spectral-domain optical coherence tomography (SD-OCT).
Sixty-six eyes of 37 glaucoma patients and glaucoma suspects from the UCLA Imaging Study were included. Eyes with reliable fields (MD >-15.0 dB), spherical refractive error <8 D and astigmatism <3 D, and no prior glaucoma surgery were prospectively enrolled. Eligible patients had a full eye exam including biometry with IOLMaster, and disc and retinal nerve fiber layer (RNFL) SD-OCT imaging (Optic disc Cube 200x200, Cirrus OCT) and imaging with Heidelberg Retina Tomograph (HRT, software version 3.01). HRT global stereometric parameters and rim area measurements for twelve 30-degree sectors were exported after correcting for corneal base curve. Disc parameters and rim area in matching 30-degree sectors were also obtained for SD-OCT with the Cirrus research browser version 22.214.171.1246. Disc area, rim area, cup to disc area ratio, vertical cup to disc ratio, cup volume, and the 12 rim-area sectors were compared between the two devices.
The average (± SD) age and mean deviation for the study eyes were 65.2 (± 8.1) years and -3.1 (± 3.3) dB, respectively. Average disc size and rim area from SD-OCT were significantly lower compared to HRT (1.88 ± 0.55 vs. 2.08 ± 0.62 mm and 0.78 ± 0.24 vs. 1.06 ± 0.40 mm, respectively, p <0.001 for both). Conversely, the cup-related parameters from SD-OCT were significantly larger (p <0.001). The highest correlations between the two devices were observed for disc area (r2 = 0.64, p <0.001) and cup volume (r2 = 0.65, p <0.001). The correlations for vertical cup to disc ratio, cup to disc area ratio, and rim area were weaker (r2 = 0.388, 0.375, and 0.28, respectively, p <0.001 for all). The correlation between rim-area sectors was highest in sectors 5, 6, 7, and 8 (nasal and inferonasal regions; r2 = 0.26-0.30). Bland-Altman plots demonstrated no significant trend (p = 0.09) for the difference in the disc area as a function of disc size, whereas the difference in the rim area increased with increasing rim area (p <0.001).
SD-OCT underestimates optic disc size and rim area and overestimates cup parameters compared with HRT. Disc parameters from the two devices are not interchangeable.
This PDF is available to Subscribers Only