Retinal microvascular differences between multiple sclerosis and neuromyelitis optica spectrum disorder: a...

Area of Science:

• Ophthalmology
• Neurology
• Medical Imaging

Background:

• Retinal microvascular changes in multiple sclerosis (MS) and neuromyelitis optica spectrum disorder (NMOSD) are not fully understood.
• Specific patterns and extent of microvascular alterations require detailed investigation for differential diagnosis.

Purpose of the Study:

• To compare retinal microvascular and structural alterations between MS and NMOSD.
• To develop logistic regression and machine learning models for differentiating MS from NMOSD using combined retinal metrics.

Main Methods:

• Swept-source optical coherence tomography (OCT) and OCT angiography (OCTA) were performed on patients with MS or NMOSD.
• Quantified OCTA metrics (vascular density, perfusion) and structural OCT metrics (RNFL, GCIPL thickness) were analyzed.
• Logistic regression and machine learning models were developed for disease classification.

Main Results:

• Significant differences in superficial vascular complex (SVC) metrics were observed between MS and NMOSD, varying with optic neuritis (ON) history.
• MS showed more severe microvascular loss in non-ON eyes, while NMOSD had steeper SVC decline in ON eyes.
• Machine learning models, particularly support vector machine, achieved high accuracy (84.5%) and AUC (0.912) in differentiating the diseases.

Conclusions:

• Distinct retinal microvascular patterns identified by OCTA can differentiate NMOSD from MS.
• These patterns correlate with disability severity and ON history.
• OCTA-based models offer accurate, non-invasive tools for differential diagnosis, with microvascular integrity as a key biomarker.