Abstract
Purpose :
Vision loss is a prominent feature of neuronal ceroid lipofuscinosis Type 2 (CLN2) Batten disease, yet cognitive, motor, and language impairments prevent reliable measurement of visual acuity (VA) using standard methods. We previously described a novel VA test using optokinetic nystagmus (OKN) detection technology (Threshold Visual Acuity Test, University of Auckland, NZ), demonstrating high correlation between central retinal thickness (CRT) and VA-OKN as well as symmetrical VA-OKN between eyes in children with CLN2 disease. Here, we report longitudinal data for the same children, as well as test-retest variability.
Methods :
Six to 10 months after initial VA-OKN measurement, binocular and monocular VA-OKN testing was performed at two consecutive visits two weeks apart in 23 children ages 3 to 9 years with classic CLN2 disease.
Results :
Longitudinal data were available for binocular VA-OKN (OU) in 14 children, and for monocular VA-OKN in the right eye (OD) in 11 children and the left eye (OS) in 10 children. Eight of nine children (89%) who initially cooperated with monocular testing in both eyes did so at the second and third visits. Binocular VA-OKN ranged from 0.3 to 1.4 logMAR. Monocular VA-OKN remained highly symmetric between right and left eyes. Mean longitudinal change in VA-OKN over the 6 to 10-month testing period was 0.06 logMAR OU, 0.03 OD, and -0.05 OS. The 95% CI for test-retest limits of agreement on Bland-Altman analysis were 0.24 to -0.26 (mean difference = -0.01) OU, 0.30 to -0.30 (-0.03) OD, and 0.13 to -0.31 (-0.09) OS, with Pearson correlation coefficients of 0.93 OU, 0.88 OD, and 0.91 OS.
Conclusions :
In children with the profound neurocognitive impairments associated with CLN2 disease that prevent use of standard VA methods, the novel, OKN detection-based Threshold Visual Acuity Test appears to be a valid measure, demonstrating satisfactory test-retest variability. More longitudinal data are needed to characterize the pattern and progression of VA loss in CLN2 disease.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.