Abstract
Purpose:
Low contrast visual function has been validated as an important clinical outcome measure in both observational and therapeutic investigational studies in multiple sclerosis. Computerized tests of visual performance provide convenient measures of visual function in the clinic. Subjects with multiple sclerosis frequently have structural injury to the retina that can be assessed via spectral-domain optical coherence tomography. Recent advances in software algorithms allow segmentation and quantification of individual retinal layers.
Methods:
We performed Rabin cone contrast testing (CCT), computerized low contrast threshold testing (LCT), and Heidelberg Spectralis spectral-domain optical coherence tomography (SD-OCT) in 52 people with multiple sclerosis (MS) (102 eyes, 71% female, 18 with a history of a prior optic neuritis). Linear mixed-models were used to analyze the association between cone-contrast and low-contrast visual function and retinal layer thickness, adjusting for age, sex, history of optic neuritis, and accounting for intra-subject inter-eye correlations.
Results:
Red, Blue, and Green CCT scores were positively correlated with macular ganglion cell / inner plexiform layer (GCIPL) thickness (p<0.001), but were not significantly associated with the thickness of other retinal layers. Linear mixed-effect regression models predicting GCIPL thickness showed significant improvement with the addition of LCT scores (change in deviance of 79.0 and 81.0 in models with and without CCT, respectively). The addition of CCT had a lower magnitude but significant effect on model deviance even when low contrast vision was included as a predictor in the model (change in deviance of 33.6 and 27.4 on models with and without LCT, respectively).
Conclusions:
Color cone contrast testing is correlated with GCLIP thickness in MS and provides added predictive value of visual pathway injury in combination with low-contrast testing in MS patients.