Abstract
Purpose :
To compare two models of Spectralis SD-OCT (Heidelberg Engineering Inc., Carlsbad, CA), in their image quality, segmentation accuracy and differences in measuring the thickness of outer retina macula layers.
Methods :
Posterior pole SD-OCT scans (8 x 8 grid) centered on the fovea were performed on normal volunteers. Three consecutive images of each eye were acquired on two models of Spectralis SD-OCT: an older model, built in 2011 and operating at 40 KHz scanning rate (OCT) and a newer, model built in 2017, operating at 85 KHz scanning rate (OCT2). The same procedure was repeated in 2-3 weeks. Four retinal layers: inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL) and retinal pigment epithelial layer (RPE) were segmented automatically by Spectralis software. For each grid cell the root mean square (RMS) deviation from the mean of each set of three measurements. Differences between the two models were evaluated by Dunn’s multiple comparisons test for RMS error and Sidak’s multiple comparisons test for layer thickness or algorithm failure.
Results :
Both eyes of 14 volunteers (4 men and 10 women) aged 35.1 +/- 12.3 years were imaged. The average image quality was significantly better for the OCT2 vs. OCT model: 35.7 ± 0.9 vs. 27.2 ± 0.7 (or ~31% improvement in quality, p<0.0001). Without any algorithm correction, there was no significant difference for either the absolute RMS error or the RMS error as percent of retinal layer thickness between the two models for any of the four layers (p>0.05). Similarly, algorithm failure, defined as variability of more than 10% of RMS error in retinal thickness between the 3 measurements, was no significantly different between the instruments, either as an absolute count or as a percent of areas with failure (p>0.05). When algorithm failure was accounted for, for two of the retinal layers (INL and ONL), OCT2 measured slightly higher thickness compared to OCT: 0.3 to 1 µm and 1.0 to 1.2 µm, respectively (p<0.05).
Conclusions :
These findings indicate that, despite the improvement in image quality, both OCT models show similar variability. Furthermore, small (2-3%), but significant systematic differences in retinal thickness may be expected for some outer retina layers when using the two models.
This is a 2020 ARVO Annual Meeting abstract.