Demyelinating Cord Disease
- Demyelinating cord disease is the spinal-cord version of the same trouble that causes brain plaques: the insulation around nerve fibers gets stripped off, and the cord lights up on MRI.
- The classic multiple sclerosis (MS) cord lesion is short — under two vertebral body segments long — and sits off to one side, usually touching the back or side of the cord.
- On axial images it grabs less than half the cord's cross-section; that "partial" pattern is a big clue it's MS and not something nastier.
- The cord is rarely the whole story — go look at the brain, because most cord demyelination travels with brain plaques.
- The main impostors are a long, swollen, central lesion (think NMO) and a wedge of cord infarct. Telling them apart changes the whole treatment plan.
Think of every nerve fiber in your spinal cord as an extension cord, and the myelin sheath as the rubber insulation wrapped around it. Demyelinating disease is what happens when something gnaws that rubber off in patches. The wire still runs, but the signal leaks, slows, and garbles — which is why a patient might show up with a numb leg, a clumsy hand, or a weird band of tightness around the trunk. Our job on MRI is to find the chewed-up patch.
What it actually looks like
The cord version of demyelination has a signature, and once you've seen it a few times it's hard to unsee. On the sagittal T2 image — the long side-view down the length of the cord — you're hunting for a bright (high-signal) patch. The MS-flavored one is short, spanning less than two vertebral body segments, and it loves the cervical cord.
Now flip to the axial view, looking at the cord in cross-section like a sausage slice. The MS lesion is eccentric (off to one side, not dead center) and peripheral — it tends to hug the back columns or a lateral edge. Crucially, it takes up less than half the cord. That "partial cross-section" look is one of your best friends here.
The enhancement clue (how fresh is it?)
Reach for gadolinium contrast to ask one question: is this plaque angry right now? An actively demyelinating lesion breaks down its local blood-brain barrier, so it enhances — often as a faint patch or an incomplete, open ring that doesn't quite close. An old, burned-out plaque just sits there bright on T2 and ignores the contrast entirely. So enhancement roughly sorts "this is happening now" from "this happened a while ago."
A short, eccentric, partial-cross-section cord lesion that enhances during an attack and fades to a quiet T2 scar afterward is the textbook MS plaque. The pattern matters more than any single image.
Always check the brain
Here's the rule I wish someone had tattooed on me early: the spinal cord is a satellite office, not headquarters. If you find a demyelinating cord lesion, the brain is very often hiding plaques too — the classic ovoid white-matter lesions pointing out from the ventricles. Finding them turns a vague cord finding into a much more confident diagnosis, and it's central to how the brain demyelination picture and MS get worked up. Looking at the cord alone is like reviewing one chapter and guessing the plot.
The impostors (this is the whole game)
Cord demyelination shares a neighborhood with a few other diagnoses, and they don't all get treated the same way. The differences are mostly about how long the lesion is and how much of the cord it eats.
| Diagnosis | What tips you off | Why it matters |
|---|---|---|
| MS plaque (the thing itself) | Short (< 2 segments), eccentric, < half the cord, often brain plaques too | Treated as MS; usually the answer |
| NMO / longitudinally extensive lesion | Long (≥ 3 segments), central, swells the cord, fills most of the cross-section | Different antibody, different drugs — getting this wrong matters |
| Cord infarct | Sudden onset, anterior cord, "owl-eye" gray-matter pattern, vascular story | It's a stroke, not inflammation |
| Generic transverse myelitis | An umbrella term for cord inflammation; demyelination is one cause under it | Demyelination is a subtype, not a separate planet |
A long lesion spanning three or more vertebral segments and ballooning the cord from the center is not the typical MS look — that pattern points toward neuromyelitis optica (NMO) and friends. Calling a fat, central, longitudinally extensive lesion "just MS" is the classic miss, and it sends the patient down the wrong treatment road.
Why you don't just eyeball one slice
One more trap worth naming: cord lesions are small, and the cord is a moving, pulsating little structure. A single slightly-off slice can fake a lesion, and a real plaque can hide if you only skim one sequence. Always confirm a bright spot on both the sagittal and the axial — a true lesion shows up in two planes. If it only exists in one, be suspicious it's an artifact before you write the scary word.
Demyelination and cord compression can look superficially similar at a glance — both can brighten the cord on T2 — but they're solved completely differently. Before settling on inflammation, make sure nothing is physically squeezing the cord; see cord compression.
The one thing to remember
If you take a single image home: the MS cord lesion is the shy one — short, off to the side, partial. When a cord lesion instead grows long, fat, and central, your alarm should go off, because you're probably not looking at run-of-the-mill MS anymore, and the treatment changes with the diagnosis.