Longitudinal Changes in Retinal Nerve Fibre Layer Thickness after an Isolated Unilateral Retrobulbar Optic Neuritis: 1-Year Results

Machine learning classification of multiple sclerosis in children using optical coherence tomography

BACKGROUND

In children, multiple sclerosis (MS) is the ultimate diagnosis in only 1/5 to 1/3 of cases after a first episode of central nervous system (CNS) demyelination. As the visual pathway is frequently affected in MS and other CNS demyelinating disorders (DDs), structural retinal imaging such as optical coherence tomography (OCT) can be used to differentiate MS.

OBJECTIVE

This study aimed to investigate the utility of machine learning (ML) based on OCT features to identify distinct structural retinal features in children with DDs.

METHODS

This study included 512 eyes from 187 (n_eyes = 374) children with demyelinating diseases and 69 (n_eyes = 138) controls. Input features of the analysis comprised of 24 auto-segmented OCT features.

RESULTS

Random Forest classifier with recursive feature elimination yielded the highest predictive values and identified DDs with 75% and MS with 80% accuracy, while multiclass distinction between MS and monophasic DD was performed with 64% accuracy. A set of eight retinal features were identified as the most important features in this classification.

CONCLUSION

This study demonstrates that ML based on OCT features can be used to support a diagnosis of MS in children.

Dynamics of Retinal Vessel Loss After Acute Optic Neuritis in Patients With Relapsing Multiple Sclerosis

Background and Objectives

Rarefication of the retinal vasculature as measured by optical coherence tomography angiography (OCT-A) is a novel finding in patients with multiple sclerosis (MS). This study aimed to analyze longitudinal dynamics of the retinal vasculature following an acute inflammatory relapse including acute optic neuritis (ON) and to search for associations with alterations of the retinal architecture and visual function.

Methods

This prospective longitudinal cohort study included patients with relapsing-remitting MS or clinically isolated syndrome having an acute ON (n = 20) or a non-ON relapse (n = 33). Patients underwent examinations at baseline and after 7, 14, 28, 90, and 180 days with OCT, OCT-A, and assessment of the high- (HCVA) and low-contrast visual acuity (LCVA).

Results

Retinal vessel loss of the superficial vascular complex (SVC) evolves early after ON and reaches a plateau between 90 and 180 days (relative vessel loss 15% ± 8% [mean ± SD]). In addition, an 18% ± 18% intraindividual increase of the foveal avascular zone (FAZ) is evident within 180 days after acute ON. Both SVC thinning and FAZ enlargement were associated with worse HCVA and LCVA. Rarefication of the SVC evolved simultaneously to thinning of the common ganglion cell and inner plexiform layer (GCIP) after ON. No alterations of the deep vascular complex were seen in eyes with ON, and no alterations of the retinal vasculature were recognized in patients having acute non-ON relapses.

Discussion

Rarefication of the SVC and growing of the FAZ evolve rapidly after ON and are linked to persistent visual disability. ON-related SVC thinning might be closely linked to GCIP atrophy and might occur due to an altered local metabolic activity within inner retinal layers.

Peripapillary and parafoveal microvascular changes in eyes with optic neuritis and their fellow eyes measured by optical coherence tomography angiography: an Exploratory Study

PURPOSE

This study aimed to evaluate parafoveal and peripapillary microvascular alterations in eyes with optic neuritis (ON) along with their fellow eyes compared to healthy control eyes using optical coherence tomography angiography (OCT-A).

METHODS

We included 31 ON-affected eyes and 31 fellow eyes of 31 patients who had experienced unilateral ON and 33 eyes of 33 healthy controls in this exploratory retrospective cross-sectional study. Optical coherence tomography angiography (OCT-A) was used to generate microvascular structural images and quantify the vessel density of the superficial retinal capillary plexus (SRCP), the deep retinal capillary plexus (DRCP) and radial peripapillary capillary (RPC) segments. We used the Kruskal-Wallis test for the comparison of OCT-A results between the three groups and generalized estimating equation models for the pairwise comparisons.

RESULTS

There were significant differences of SRCP (p = 0.0003) and RPC segment (p < 0.0001) vessel densities between the three groups. Specifically, there was a reduction in parafoveal and peripapillary vessel density in the ON-affected eyes compared to fellow eyes (SRCP, estimates, -1.97, 95% confidence interval [CI], -3.07, -0.87; RPC, -6.95, 95% CI, -8.70, -5.19) and controls (SRCP, -3.15, 95% CI, -4.61, -1.69; RPC, -8.66, 95% CI, -10.55, -6.76). The superior sector of the RPC segments vessel density in the fellow eyes was decreased compared to the controls (-4.93, 95% CI, -8.07, -1.80).

CONCLUSIONS

The results of this study suggest that microvascular changes occur in both the affected eye and unaffected fellow eye after a unilateral ON episode. Future studies are needed to clarify the clinical implications of these findings.

The application of RNFL thickness detection in early differential diagnosis among various types of idiopathic optic neuritis

In this study, we attempted to investigate the application of RNFL thickness detection in the early differential diagnosis among various types of idiopathic optic neuritis (ION). In comparison with 19 healthy controls (HC), retrospective analysis of quadrant RNFL thickness in 83 patients with ION was performed, including eighteen multiple sclerosis (MS), forty-five neuromyelitis optica spectrum disorder (NMOSD), twenty patients with other idiopathic optic neuritis (O-ION). Our results showed that mean and every quadrant RNFL thickness of MS, NMOSD and O-ION were thinner than those of HC (P < 0.05). In comparison with MS and O-ION, NMOSD group had thinner RNFL thickness in nasal quadrant (P < 0.05). No significant difference in each quadrant RNFL thickness between MS-ON and O-ION was suggested (P > 0.05). Mean and every quadrant RNFL thickness of unaffected eyes of MS-ON were not different from those of HC (P > 0.05). We concluded that for patients only with simple optic neuritis, NMOSD has a significantly thinner RNFL thickness in nasal quadrant than MS and O-ION, which may contribute to the clinical differentiation and therapy among various types of idiopathic optic neuritis.