Skip to content
Erfan Bashar

Multiple Sclerosis Neuroimaging

Updated:
~4 min read
Last medically reviewed:
On this pageTable of contents

Educational scope notice: This is a study note for medical students, not medical advice, diagnosis, or treatment guidance. Clinical management should follow local protocols and current guidelines.

MS cannot be reliably diagnosed without MRI. Imaging confirms that lesions are scattered in space and separated in time, and it excludes mimics. Recent diagnostic revisions give increasing weight to lesion characteristics visible on MRI, particularly the central vein sign and paramagnetic rim lesions.

Protocol and sequences

The standard protocol covers brain and spinal cord. Brain imaging uses 3D T2-FLAIR with T1 sequences before and after contrast, plus susceptibility-weighted imaging at higher field strengths. Spinal cord imaging uses T2 and STIR (short tau inversion recovery) with post-contrast T1.

Each sequence solves a visibility problem. FLAIR suppresses cerebrospinal fluid signal so periventricular lesions stand out. DIR (double inversion recovery) suppresses both cerebrospinal fluid and white matter signal, isolating the cortex so small cortical lesions become visible. PSIR (phase-sensitive inversion recovery) is another high-sensitivity cortical sequence. STIR suppresses fat, which matters for the optic nerve because the intraorbital segment is surrounded by orbital fat that would otherwise obscure swelling or hyperintensity on standard T2 images.

The five topographies

The 2024 criteria recognize five locations for dissemination in space, with two required.

Periventricular white matter is the most recognizable site. Lesions are typically round to oval and perpendicular to the lateral ventricles along medullary veins, known as Dawson fingers, and cluster at the calloso-septal interface. They are relatively specific for MS.

Cortical and juxtacortical lesions sit at the grey-white junction or within the cortical ribbon and usually need DIR or PSIR for reliable detection.

Infratentorial lesions affect brainstem and cerebellum, often near cerebrospinal fluid spaces, and their presence strongly supports the diagnosis.

Spinal cord lesions favor the cervical segment. They are typically short, spanning fewer than 2 vertebral body segments, occupy less than half the axial diameter, and sit in the lateral or posterior columns where the white matter tracts run. This peripheral, short pattern separates MS from neuromyelitis optica spectrum disorder, where lesions are central and longitudinally extensive.

Optic nerve lesions are typically unilateral, short-segment, and preferentially intraorbital. The optic nerve only joined the dissemination-in-space locations in 2024.

Acute versus chronic lesions

Acute lesions enhance because active inflammation disrupts the blood-brain barrier and gadolinium leaks into tissue. The typical pattern is nodular or ring enhancement, and an open ring facing the cortex or lesion periphery is characteristic of demyelination. In a large mass-like lesion, an open ring favors inflammation over tumor. Chronic lesions do not enhance because the barrier has resealed. Severe chronic lesions appear as T1-hypointense black holes, marking irreversible demyelination and axonal loss; roughly 20 to 40% of T2 lesions eventually become black holes. Finding enhancing and non-enhancing lesions together on one scan demonstrates dissemination in time.

Central vein sign and paramagnetic rim lesions

The central vein sign is a vein running through the center of a lesion on susceptibility-sensitive images, reflecting the perivenular origin of MS plaques. It is not sensitive enough to diagnose MS alone but is highly specific. Under the 2024 criteria, more than 6 lesions with a central vein sign (or more than half when fewer than 10 lesions exist) add diagnostic confidence, and in single-region disease they can support the diagnosis alongside dissemination in time or positive cerebrospinal fluid.

Paramagnetic rim lesions show a dark rim along more than two-thirds of the lesion margin on susceptibility-weighted imaging. The rim marks iron-laden macrophages at the plaque edge, a sign of smoldering chronic inflammation, and these lesions predict a worse prognosis. Like the central vein sign, one or more such lesions can support the diagnosis in single-region disease with dissemination in time or positive cerebrospinal fluid.

Large demyelinating lesions with mass effect, called tumefactive lesions, can mimic tumors; the open-ring pattern and an acute or subacute deficit rather than months of progression point toward demyelination.

Unfavorable MRI predictors include multiple spinal cord lesions, infratentorial lesions, black holes, and early atrophy.

Progressive multifocal leukoencephalopathy

Progressive multifocal leukoencephalopathy (PML) is a treatment complication from JC virus reactivation, most strongly associated with natalizumab and less often with fingolimod and dimethyl fumarate. It is rapidly progressive and often fatal. On MRI it produces T2 and FLAIR hyperintense lesions involving subcortical U-fibers with sharp borders toward the cortex and ill-defined borders toward white matter, sometimes with a milky-way pattern of adjacent punctate lesions. About 30 to 40% show contrast enhancement. The frontal lobes are most commonly affected, with posterior fossa involvement coming later.

Evidence anchors

  • Wattjes MP, Ciccarelli O, Reich DS, et al. 2021 MAGNIMS-CMSC-NAIMS consensus recommendations on the use of MRI in patients with multiple sclerosis. Lancet Neurol. 2021;20(8):653-670. doi:10.1016/S1474-4422(21)00095-8
  • Filippi M, Preziosa P, Banwell BL, et al. Assessment of lesions on magnetic resonance imaging in multiple sclerosis: practical guidelines. Brain. 2016;139(Pt 7):1858-1879. doi:10.1093/brain/aww144
  • Wijburg MT, Witte BI, Vennegoor A, et al. MRI criteria differentiating asymptomatic PML from new MS lesions during natalizumab pharmacovigilance. J Neurol Neurosurg Psychiatry. 2016;87(10):1138-1145. doi:10.1136/jnnp-2016-313772
  • Carroll WM, Barkhof F, Calabresi P, et al. New McDonald criteria for the diagnosis of multiple sclerosis. JAMA Neurol. 2025;82(9):875-885. doi:10.1001/jamaneurol.2025.1992
Suggest a correction