Purchase this article with an account.
S. Aliyeva, F. Berisha, T. Roshdy, N. Pfeiffer, E. M. Hoffmann; Interobserver Reproducibility of Fourier Domain Optical Coherence Tomograph and Heidelberg Retina Tomograph Parameters. Invest. Ophthalmol. Vis. Sci. 2008;49(13):4621. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To compare the interobserver reproducibility of the optic disc and retinal nerve fiber layer (RNFL) measurements provided by Fourier-domain optical coherence tomography (FD-OCT) and the Heidelberg Retina Tomograph (HRT).
Eighty-nine eyes of 53 glaucoma subjects and healthy controls underwent confocal scanning laser ophthalmoscopy (HRT III; Heidelberg Engineering, Heidelberg, Germany) and Fourier-domain OCT (RTVue; Optovue Inc., Fremont, USA). The contour lines were traced by two glaucoma specialists and one observer with no previous HRT or OCT experience. The agreement between observers was assessed for OCT and HRT disc area, cup area, rim area, cup volume, rim volume, cup-to-disc ratio and RNFL measurements using ANOVA based intraclass correlation coefficient (ICC) and coefficient of variation (CV).
The mean ICC for FD-OCT measurements was almost perfect to perfect (between 0.95 and 1.0) for all measured parameters, except for the rim volume, that had a substantial agreement (ICC = 0.74). The mean OCT RNFL thickness showed best agreement between observers (ICC = 1.0). The CV values for FD-OCT parameters ranged between 0.3% and 12.2%. All HRT parameters demonstrated almost perfect interobserver agreement (between 0.90 and 0.98). The corresponding CV values ranged between 6.1% and 10.4%. The CV interobserver agreement for RNFL was significantly better for FD-OCT compared to HRT measurements (p<0.01), while HRT showed a better reproducibility for rim volume than FD-OCT (p<0.05). The mean values of FD-OCT and HRT parameters for each observer are presented in table 1.
A newly available Fourier-domain optic coherence tomography enables highly reproducible measurements of optic disc and RNFL parameters. Both FD-OCT and HRT showed a minor influence of operator-dependent contour line placement. The observers’ experience was found to be an unsubstantial factor to influence the results of the measurements.
This PDF is available to Subscribers Only