Optical coherence tomography angiography indicates subclinical retinal disease in neuromyelitis optica spectrum disorders

Abstract

Background: Neuromyelitis optica spectrum disorders (NMOSD) are neuroinflammatory diseases of the central nervous system. Patients suffer from recurring relapses and it is unclear whether relapse-independent disease activity occurs and whether this is of clinical relevance.

Objective: To detect disease-specific alterations of the retinal vasculature that reflect disease activity during NMOSD.

Methods: Cross-sectional analysis of 16 patients with NMOSD, 21 patients with relapsing-remitting multiple sclerosis, and 21 healthy controls using retinal optical coherence tomography (OCT), optical coherence tomography angiography (OCT-A), measurement of glial fibrillary acidic protein (GFAP) serum levels, and assessment of visual acuity.

Results: Patients with NMOSD but not multiple sclerosis revealed lower foveal thickness (FT) (p = 0.02) measures and an increase of the foveal avascular zone (FAZ) (p = 0.02) compared to healthy controls independent to optic neuritis. Reduced FT (p = 0.01), enlarged FAZ areas (p = 0.0001), and vessel loss of the superficial vascular complex (p = 0.01) were linked to higher serum GFAP levels and superficial vessel loss was associated with worse visual performance in patients with NMOSD irrespective of optic neuritis.

Conclusion: Subclinical parafoveal retinal vessel loss might occur during NMOSD and might be linked to astrocyte damage and poor visual performance. OCT-A may be a tool to study subclinical disease activity during NMOSD.

Keywords: Neuromyelitis optica spectrum disorders; astrocytes; biomarker; disease activity; optical coherence tomography angiography.

Figure 1.

Results of optical coherence tomography angiography analysis. (a) Parafoveal vessel densities of the superficial (SVC) and deep vascular complex (DVC) and size of the foveal avascular zone (FAZ) in healthy individuals (HC, n = 21) and eyes with former optic neuritis (ON) of patients with relapsing-remitting MS (MS, n = 14 patients) and neuromyelitis optica spectrum disorders (NMOSD, n = 8 patients); one eye of one NMOSD patient was excluded due to poor OCT quality. (b) Parafoveal vessel densities of the SVC and DVC and size of the FAZ in HC (n = 21), MS (n = 18 patients), and NMOSD (n = 15 patients) in eyes without a history of optic neuritis (no ON); one MS patient underwent unilateral enucleation in the past. (c) Parafoveal vessel densities of the SVC and DVC and size of the FAZ in HC (n = 21), MS (n = 18 patients), and NMOSD with antibodies against aquaporin-4 (AQP-4 + NMOSD) (n = 9 patients) in eyes without a history of optic neuritis (no ON). (a–c) Mean ± standard deviation (SD); symbols depict single patient values; one-way ANOVA; n.s.: not significant; the statistical significance threshold was p<0.05.

Figure 2.

Retinal architecture, vasculature, and surrogate markers of disease activity in neuromyelitis optica spectrum disorders. (a) Association of the Expanded Disability Status Scale (EDSS) and the foveal avascular zone (FAZ) areas during neuromyelitis optica spectrum disorders (NMOSD) (n = 15); β regression estimate and 95% confidence interval (CI); symbols depict single patients; multiple linear regression model corrected for age and sex. (b) Serum levels of neurofilament light chain (sNfL) and glial fibrillary acidic protein (sGFAP) in healthy individuals (HC, n = 21), patients with relapsing-remitting MS (MS, n = 21) and NMOSD (n = 15); mean ± standard deviation; symbols depict single patients; Kruskal–Wallis test (sNfL), one-way ANOVA (GFAP). (c–e) Association of sGFAP levels and foveal thickness (FT) (c), vessel densities of the superficial vascular complex (SVC) (d), and the foveal avascular zone (FAZ) areas (e) in eyes without former optic neuritis in patients with NMOSD (n = 15); β regression estimates and 95% CI; symbols depict single patients; multiple linear regression models corrected for age, sex, and EDSS.

Figure 3.

Retinal architecture, vasculature, and visual function in neuromyelitis optica spectrum disorders. (a) Association of high-contrast visual acuity (HCVA) and thickness of the peripapillary retinal nerve fiber layer (pRNFL, left) or volumes of the common ganglion cell and inner plexiform layer (GCIP, right) in eyes without former optic neuritis (ON) in patients with neuromyelitis optica spectrum disorders (NMOSD) (n = 14). (b) Association of HCVA and vessel densities of the superficial vascular complex (SVC) in eyes of NMOSD patients without (n = 15, left) or with a history of optic neuritis (n = 8, right). (c) Association of low-contrast visual acuity (LCVA) and vessel densities of SVC in eyes of NMOSD patients without (n = 15, left) or with a history of optic neuritis (n = 8, right). (a–c) β regression estimates and 95% CI; symbols depict single patients; multiple linear regression model corrected for age and sex.