Abstract
Purpose:
To evaluate the results of retinal thickness and volume measured with the enhanced depth imaging (EDI) method compared to those measured with the conventional method in spectral domain optical coherence tomography (SD OCT).
Methods:
Clinical records of 20 healthy subjects, and consecutive 35 chorioretinopathy patients (central serous chorioretinopathy, polypoidal choroidal vasculopathy, Vogt-Koyanagi-Harada disease, and reticular pseudodrusen) were retrospectively analyzed. All subjects underwent SD OCT (Heidelberg Spectralis HRA+OCT) using both the conventional and the EDI OCT raster scan protocols. The raster scan was composed of 31 B-scans centered on fovea, and each B-scan was 9.0 mm in length, and 240 μm apart. Retinal thickness and volume of nine ETDRS subfields were investigated in two protocols by both the autosegmentation and the manually corrected segmentation. Intraclass correlation coefficient (ICC) and Bland-Altman plots were used to evaluate the degree of agreement between the two protocols. The Wilcoxon signed rank tests were used to compare the frequency of retinal boundary segmentation errors.
Results:
40 healthy eyes, 28 unaffected fellow eyes of chorioretinopathy patients, and 37 eyes with chorioretinopathy were evaluated. The automatically measured mean retinal thickness of 9 ETDRS subfields with conventional and EDI raster scan showed the ICC of 0.861 - 0.995. The automatically measured retinal volume of 9 ETDRS subfields with two protocols showed ICC of 0.873 - 0.995. The 95% limits of agreement between two protocols in the measurement of central subfield were from -14.52 to +12.88 μm in the retinal thickness, and from -0.014 to +0.013 mm3 in the retinal volume. The mean retinal thickness and volume measured with manually corrected segmentation showed ICC more than 0.960, and much decreased 95% confidence interval in Bland-Altman plot analysis. The differences of segmentation error between the two protocols were statistically insignificant (P>0.05), except in eyes with reticular pseudodrusen in the subgroup analysis (P=0.005).
Conclusions:
The retinal thickness and volume measured by EDI OCT raster scan showed a high agreement with those measured by conventional OCT. These data suggest that the EDI OCT raster scan could be used to evaluate both retina and choroid in retinochoroidal disorders.
Keywords: 552 imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) •
452 choroid •
550 imaging/image analysis: clinical