Wernicke Encephalopathy
- Wernicke encephalopathy is the brain running out of thiamine (vitamin B1) — an emergency that's treated, not just imaged.
- It loves symmetric, midline real estate: the mammillary bodies, the tissue around the third ventricle, the periaqueductal gray, and the medial thalami.
- The classic finding is symmetric T2/FLAIR brightness in those spots, often with restricted diffusion; the mammillary bodies may enhance.
- It is not just an "alcoholic's disease" — anyone who can't absorb or hasn't received thiamine can get it.
- A normal MRI does not rule it out. If the clinical picture fits, you give thiamine first and admire the pictures later.
Imagine the brain as a city that runs on a single, very specific fuel delivered fresh every day, because it can't keep a stockpile. That fuel is thiamine — vitamin B1 — and the most energy-hungry neighborhoods burn through it fastest. Cut off the deliveries for a couple of weeks and those neighborhoods are the first to brown out. That brownout, in brain form, is Wernicke encephalopathy (WE): an acute, treatable, genuinely dangerous problem that radiology gets to flag before things go irreversible.
Why the brain cares so much about one vitamin
Thiamine is a cofactor your cells need to turn sugar into usable energy. The brain is a fuel hog and stores almost no thiamine, so it depends on a constant supply. Run low — through poor intake, poor absorption, or burning through stores faster than they're replaced — and the most metabolically active gray matter starts to fail. Cells can't keep their pumps running, water sneaks into places it shouldn't, and tissue swells. That's why the imaging looks the way it does.
The cruel part is the timeline: this can develop over days to weeks, and the window to fix it without permanent damage is narrow. Treatment is simple and cheap — give thiamine — which is exactly why missing it stings.
This is a clinical diagnosis you treat empirically. If the picture fits, thiamine is given before glucose, because pushing glucose into a thiamine-depleted brain can tip a borderline patient into full-blown Wernicke. Imaging supports the diagnosis; it does not gatekeep treatment.
Who gets it (it's not just alcohol)
Yes, chronic alcohol use is the headline cause — alcohol wrecks both thiamine intake and absorption. But anyone whose thiamine supply chain breaks can land here: prolonged vomiting (including hyperemesis in pregnancy), bariatric surgery, malnutrition, prolonged IV feeding without vitamins, or a malignancy eating up reserves. If you anchor only on "alcoholic," you will miss the postoperative or pregnant patient, and that's a classic, painful miss.
What it looks like on imaging
Here's the satisfying part: WE is one of the most pattern-recognizable conditions in the brain, because it picks the same handful of midline structures almost every time. Think of them as the city's central power district, all clustered around the ventricles.
The hallmark on MRI is symmetric high T2/FLAIR signal — that bright, swollen-tissue look — in:
| Location | Why it matters |
|---|---|
| Medial thalami | Paired nuclei hugging the third ventricle; bright and symmetric is the giveaway. |
| Tissue around the third ventricle | The midline walls and floor of the third ventricle, including the hypothalamus. |
| Periaqueductal gray | The rim of tissue around the aqueduct in the midbrain. |
| Mammillary bodies | The signature; may swell acutely and enhance after contrast. |
Restricted diffusion is often present in these areas, reflecting the same fluid-trapped, energy-starved tissue you see in ischemic stroke — except here it's bilateral, symmetric, and in a distribution no single blood vessel would ever produce. That symmetry is your friend: strokes follow plumbing; Wernicke follows metabolism.
CT is mostly unhelpful here; it can be normal or show only vague low density in these regions. MRI is the test that earns its keep, ideally with diffusion and post-contrast sequences. If you want a refresher on which sequence shows what, the approach to brain MRI page lays out the toolkit.
Mammillary body enhancement and symmetric medial thalamic signal in the right clinical setting is about as close to a "say it's Wernicke" sign as neuroimaging gives you. Atypical, non-alcoholic cases more often show involvement of other gray matter regions too — keep that in mind for the bariatric or pediatric patient.
The look-alikes
The differential is short but worth knowing, because the treatment couldn't be more different.
| Mimic | How it differs |
|---|---|
| Top-of-basilar / deep venous infarct | Follows a vascular territory; usually not this clean midline symmetry. |
| Osmotic demyelination | Centers on the central pons (and sometimes basal ganglia), not the thalami/mammillary bodies. |
| PRES | Favors posterior, subcortical white matter; tied to blood-pressure swings. |
| Deep gliomatosis/neoplasm | Asymmetric, mass effect, no thiamine story. |
A normal-looking MRI does not exclude Wernicke. Imaging sensitivity is imperfect, and early disease can be subtle or invisible. If the clinical story fits — confusion, eye-movement problems, unsteady gait, in someone at risk — the answer is thiamine, not "MRI was clean, move on."
The one thing to carry out the door
When you see symmetric, midline brightness around the third ventricle and aqueduct, especially with mammillary body enhancement, in a patient who could possibly be thiamine-depleted, say the word Wernicke out loud and say it fast. This is one of the rare brain findings where a quick read, a phone call, and a cheap vitamin can flip the outcome from permanent to fully reversible.