Optimal Inter-Eye Difference Thresholds by OCT in MS: An International Study

2019 
OBJECTIVE: To determine the optimal thresholds for inter-eye differences in retinal nerve fiber and ganglion cell+inner plexiform layer thicknesses for identifying unilateral optic nerve lesions in multiple sclerosis. BACKGROUND: Current international diagnostic criteria for multiple sclerosis do not include the optic nerve as a lesion site despite frequent involvement. Optical coherence tomography detects retinal thinning associated with optic nerve lesions. METHODS: In this multi-center international study at 11 sites, optical coherence tomography was measured for patients and healthy controls as part of the International Multiple Sclerosis Visual System Consortium. High- and low-contrast acuity were also collected in a subset of participants. Presence of an optic nerve lesion for this study was defined as history of acute unilateral optic neuritis. RESULTS: Among patients (n=1,530), receiver operating characteristic curve analysis demonstrated an optimal peripapillary retinal nerve fiber layer inter-eye difference threshold of 5 microns and ganglion cell+inner plexiform layer threshold of 4 microns for identifying unilateral optic neuritis (n=477). Greater inter-eye differences in acuities were associated with greater inter-eye retinal layer thickness differences (p≤0.001). INTERPRETATION: Inter-eye differences of 5 microns for retinal nerve fiber layer and 4 microns for macular ganglion cell+inner plexiform layer are robust thresholds for identifying unilateral optic nerve lesions. These thresholds may be useful to establish the presence of asymptomatic and symptomatic optic nerve lesions in multiple sclerosis and could be useful in a new version of the diagnostic criteria. Our findings lend further validation for utilizing the visual system in a multiple sclerosis clinical trial setting. This article is protected by copyright. All rights reserved.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    7
    Citations
    NaN
    KQI
    []
    Baidu
    map