Comparison of the Accuracy of Choroidal Thickness Measurements Using SD-OCT With and Without EDI and Swept Source OCT and Volumetric Analysis of the Choroid in Healthy Eyes

Abstract

Purpose: To compare the accuracy of choroidal thickness measurements in healthy eyes using spectral domain optical coherence tomography (SD-OCT) with and without enhanced depth imaging (EDI) and a novel prototype swept source OCT (SS-OCT). To analyze the choroidal volume in healthy eyes using SS-OCT.

Methods: Nineteen healthy subjects (19 eyes) were prospectively recruited and two consecutive imaging sessions within 10 minutes on the same randomly selected eye were performed. A Cirrus HD-OCT, and a prototype SS-OCT operating at 1050nm (100k A-scans/sec) developed at the Massachusetts Institute of Technology and deployed at the New England Eye Center, Boston were used. All subjects had a BCVA of ≥20/20, and underwent fundus examination to exclude retinal or choroidal pathology. A 1-line raster with and without EDI, and a macular cube protocol (512x128 A-scans within 6x6mm) were obtained with the Cirrus. The SS-OCT scan pattern was a 1-line raster and macular cube protocol (350x350 A-scans within 6x6mm). Choroidal thickness measurements were performed manually on all images. Manual segmentations using Amira software from images obtained on SS-OCT for volumetric analysis of choroid were also performed. Two independent raters performed all the measurements.

Results: The choroid-scleral interface was clearly visualized in 19/19 (100%), 13/19 (68.4%) and 14/19 (73.6%) of eyes imaged with SS-OCT, non-EDI SD-OCT and EDI SD-OCT respectively. There was no significant difference among choroidal thickness measurements on non-EDI SD-OCT, EDI SD-OCT and SS-OCT (p=0.10). The 95% limits of agreement for choroidal thickness measurements were -11.53% to 9.39% (bias -1.07) between non-EDI and EDI SD-OCT, -22.8% to 20.6% (bias -1.06) between non-EDI SD-OCT and SS-OCT and -15.9% to 15.9% (bias -0.007) between EDI SD-OCT and SS-OCT. Mean choroidal volume was 11.77±3.13mm³ (6.43mm³-17.15mm³).

Conclusions: The prototype SS-OCT dramatically improves visualization of the choroid-scleral interface, when compared to both EDI and non-EDI SD-OCT. Choroidal thickness measurements between SD-OCT and SS-OCT are reproducible. Deeper penetration, and high speed SS-OCT enables volumetric analysis of choroid. Further studies in diseased eyes are expected to help recognize the clinical utility of SS-OCT in determining the choroidal volume in various retinal diseases.

Keywords: 452 choroid • 550 imaging/image analysis: clinical • 552 imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)