Abstract
Purpose :
Progressive retinal degeneration has been demonstrated in classic CLN2 disease, particularly loss of central retinal thickness with symmetrical bilateral onset and progression. However, changes in the outer retina, especially the ellipsoid zone (EZ), have not been described in detail. This retrospective, observational clinical study was therefore undertaken to examine whether longitudinal observations are possible between low- and high-resolution SD-OCT scans in children with classic CLN2 disease.
Methods :
SD-OCT macular cubes (8.8 x 7.8 mm, 61 B-scans, low resolution) were collected in 23 classic CLN2 patients, ages 24 to 122 months. In a subset of subjects, both low- and high-resolution macular cubes consisting of 241 B-scans were also collected on the same day. The contour of EZ loss was marked manually where the EZ band became indistinguishable from retinal pigment epithelium in cross-sectional B-scans and EZ area loss was measured using Heidelberg Eye Explorer (V1.9.10). In subjects with both low- and high-resolution scans acquired, EZ area loss was measured in both resolutions from the same eye for analysis of agreement. EZ loss from the same scans were assessed again after a 2 day washout period for estimation of repeatability. Bland-Altman analyses were used to determine the cutoffs in agreement and precision. Outer nuclear layer (ONL) thickness was measured using the standard ETDRS grid. Longitudinal data were analyzed when available.
Results :
Repeatability between EZ loss measurement by the same reader in the same scan was -0.28 to 0.25 mm2. EZ area loss measured with both low- and high-resolution scans showed strong correlation (Pearson’s correlation coefficient = 0.999, p < 0.01, with limits of agreement of -0.43 to 0.22 mm2). The area of EZ loss increased with disease progression and was associated with the magnitude of ONL reduction.
Conclusions :
Strong correlation and agreement in measurement of EZ loss between low- and high-resolution scans indicated the method is reliable, while low test-retest variability suggests it is repeatable. This allows comparison of EZ area loss between scans of different resolutions. Additional longitudinal data are needed to understand the pattern of EZ loss in CLN2 disease and its relationship with degeneration of other photoreceptor layers.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.