Abstract
Purpose: :
To report normal macular thickness measurements and assess reproducibility of retinal thickness measurements acquired by time-domain optical coherence tomography (Stratus OCT) and three commercial Fourier-domain OCT instruments (Zeiss Cirrus HD-OCT, Optovue RTVue, Topcon 3D OCT-1000).
Methods: :
Forty randomly selected eyes of 40 normal, healthy volunteers were scanned using the macular thickness protocol for Stratus OCT (6x512 images in 6 mm diameter region); 5-line raster (5x4096 in 6x1 mm2) and macular cube (128x512 in 6x6 mm2) protocols for Cirrus; MM5 (22x668 in 5x5 mm2 + 12x400 in 3x3 mm2), MM6 (12x1024 in 6 mm diameter), and 3d macular (101x512 in 4x4 mm2) protocols for RTVue; radial (6x1024 in 6 mm diameter) and 3d scan (128x512 in 6x6 mm2) protocols for 3D OCT-1000. Subjects were scanned twice during one visit and again within one month. Retinal thickness measurements were automatically generated by OCT software. The results were displayed for each instrument as the mean and standard deviation for each of the 9 regions defined in the Early Treatment Diabetic Retinopathy Study. Reproducibility was analyzed by using intraclass correlation coefficients and coefficient of variation.
Results: :
Based on preliminary data, mean foveal thicknesses were 214±11 on Stratus OCT, 267±14 on Cirrus, 249±13 (MM5) and 238±12 (MM6) on RTVue, 200±26 (radial) and 208±24 µm (3d) on 3D OCT-1000 respectively. The Cirrus was the only instrument that included the retinal pigment epithelium, while the others measured from vitreoretinal interface to anterior photoreceptor inner / outer segment junction. Macular thickness measurements were thinnest at the center of the fovea and thickest within 3 mm of the center across machines. Quantitative analysis using the latest software was unavailable for raster scans on Cirrus and 3d macular scans on RTVue.
Conclusions: :
Commercial Fourier-domain OCT instruments vary greatly in scanning protocols and are limited in analysis functions, although they uniformly possess higher speed and higher axial image resolution over time-domain OCT. As the software continues to evolve, retinal layer segmentation errors in the newer models may be minimized further, resulting in greater accuracy and precision. In the meantime, clinicians should be aware of these discrepancies in measurement when interpreting scans from different OCT models.
Keywords: imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • imaging/image analysis: clinical • retina