Choroidal thickness and choroidal vascularity index (CVI) measurement for myopia research

Axial length is the endpoint that matters in myopia control, but it moves slowly. A trial usually needs a year or more before treated and untreated eyes separate clearly. The choroid, the vascular layer behind the retina, responds much faster. That makes choroidal thickness and the choroidal vascularity index (CVI) useful early endpoints for sponsors who want a biological signal from a short study.

What the two measures are

Choroidal thickness is the distance from the retinal pigment epithelium to the choroid-sclera boundary, measured on enhanced depth OCT scans, usually beneath the fovea and across the macula.

Choroidal vascularity index is the proportion of the choroid occupied by blood vessel lumen. The OCT image is binarised into luminal and stromal areas, and CVI is luminal area divided by total choroidal area. In a population study of 345 healthy eyes, Agrawal and colleagues reported a mean subfoveal CVI of 65.6% and found CVI far less variable than subfoveal choroidal thickness (coefficient of variation 3.55 against 40.30), which is why it is attractive as a trial measure.

What the evidence shows

• The eye responds to defocus within an hour. Read, Collins and Sander showed that 60 minutes of imposed myopic defocus shortened optical axial length by a mean of 13 µm and hyperopic defocus lengthened it by 8 µm in young adults, with accompanying choroidal thickness changes.
• Myopia control spectacles change the choroid. In a two-year randomised trial, macular choroidal thickness thinned steadily in children wearing single-vision lenses, while lenses with highly aspherical lenslets reduced or abolished that thinning (Huang et al., 2023).
• Early choroidal change predicts later control. In a multicentre red-light trial, choroidal thickness rose in treated children and fell in controls, and the change at three months alone predicted satisfactory 12-month myopia control with areas under the curve of 0.71 to 0.79 (Xiong et al., 2023).

An honest read of the evidence

Choroidal measures are promising surrogate endpoints, not validated replacements for axial length. Thickness varies with time of day, accommodation and recent near work, and the changes of interest are often 10 to 20 µm, which is small relative to measurement variability. A credible study therefore needs fixed measurement times, a standardised pre-scan routine, masked graders and repeat scans. CVI depends on the binarisation method, so the method has to be fixed in the protocol.

How NZERC measures the choroid

The New Zealand Eye Research Centre (NZERC) measures choroidal thickness and choroidal vascularity index as part of its myopia management research, using the Netrimind platform, with a full Heidelberg Spectralis OCT suite also on site. Axial length can be measured at the same visit with optical biometry, so choroidal and axial changes can be read together. For sponsored studies, time of day and pre-scan activity are fixed in the protocol.

Adding a choroidal endpoint to your study

Device, lens and pharmaceutical sponsors who want choroidal thickness or CVI as an exploratory or early endpoint in a myopia study are invited to contact NZERC at research@nzerc.com or through the sponsor enquiry form. See also early feasibility and pilot studies in New Zealand and our equipment and endpoints.

References

• Agrawal R, Gupta P, Tan KA, et al. Choroidal vascularity index as a measure of vascular status of the choroid. Sci Rep 2016;6:21090. doi:10.1038/srep21090
• Read SA, Collins MJ, Sander BP. Human optical axial length and defocus. Invest Ophthalmol Vis Sci 2010;51(12):6262-9. doi:10.1167/iovs.10-5457
• Huang Y, et al. Effect of spectacle lenses with aspherical lenslets on choroidal thickness in myopic children: a 2-year randomised clinical trial. Br J Ophthalmol 2023;107(12):1806-1811. doi:10.1136/bjo-2022-321815
• Xiong R, et al. Longitudinal changes and predictive value of choroidal thickness for myopia control after repeated low-level red-light therapy. Ophthalmology 2023;130(3):286-296. doi:10.1016/j.ophtha.2022.10.002