Necrotizing Otitis Externa
- Necrotizing (a.k.a. malignant) otitis externa is an aggressive bacterial infection of the external ear canal that crawls into the skull base. It is an infection, not a cancer — the scary old name "malignant" just means it behaves badly.
- Think of it in the elderly diabetic or immunocompromised patient with relentless ear pain and drainage that won't quit despite ear drops. Granulation tissue at the bony-cartilaginous junction of the canal is the classic exam clue.
- The villain is almost always Pseudomonas aeruginosa, and the target it's heading for is the skull base and the cranial nerves running through it.
- On CT you're hunting for erosion of the skull base; on MRI you're tracking soft-tissue and marrow spread. Loss of the normal fatty signal in the marrow beneath the skull base is an early red flag.
- Miss it and the infection marches into cranial nerves, the jugular foramen, and the meninges. This is a "name it today" diagnosis.
Most ear infections are annoying. This one is trying to eat your skull. Necrotizing otitis externa (NOE) starts as what looks like a stubborn case of swimmer's ear and, if nobody catches it, tunnels straight down into the bone at the base of the head where all the important wiring lives. The good news: once you know who gets it and where to look, it's hard to miss. The bad news: it's quiet enough early on that plenty of people do miss it.
Who gets it, and why you should be suspicious
The patient is almost a stereotype: older, diabetic, with ear pain that is wildly out of proportion to how the ear looks, plus drainage (otorrhea) that has shrugged off a course of topical antibiotics. Immunocompromised patients of any age qualify too.
Why diabetics? Picture the small blood vessels of the ear canal as garden hoses that have gone stiff and narrow with age and high blood sugar. The immune system's delivery trucks can't get down those hoses efficiently, and the local environment is friendly to Pseudomonas aeruginosa — the bacterial culprit in the overwhelming majority of cases. It sets up shop at the bony-cartilaginous junction of the canal, the spot where the floppy cartilage tube meets the hard bony tube, and clinicians famously find beefy granulation tissue sitting right there.
"Malignant" otitis externa is a confusing antique of a name — there's no tumor here. It was coined because the infection acts as relentlessly as a cancer, eroding bone and killing patients. Modern radiology prefers "necrotizing." Same disease, less misleading.
What it actually is: skull base osteomyelitis
Here's the concept that unlocks everything. NOE isn't really an ear-canal disease for long — it's skull base osteomyelitis (a bone infection of the floor of the skull) that happens to start in the ear canal. Once you frame it that way, the imaging makes sense. You're not admiring the ear; you're chasing an infection along the bone.
The infection spreads through tiny channels and along tissue planes from the canal into the temporal bone, then medially toward the skull base and the soft tissue underneath it. Follow that path and you run into the cranial nerves — which is exactly why a facial droop (cranial nerve VII) in this setting is a four-alarm fire.
How to image it
This is a two-modality problem, and the modalities answer different questions.
| Modality | What it's best at | What you're looking for |
|---|---|---|
| CT (temporal bone) | Bone detail | Erosion of the bony ear canal and skull base; opacification of mastoid air cells. |
| MRI | Soft tissue & marrow | Replacement of normal fatty marrow signal in the skull base; spread into soft tissue, meninges, and around nerves. |
| Nuclear medicine | Bone activity | Adjunct for confirming osteomyelitis and following response over time. |
CT is usually the first stop because it shows bone erosion crisply — but here's the catch: by the time CT shows obvious bone destruction, the infection has already been busy for a while. CT is specific but not the earliest alarm.
MRI is the more sensitive early detective. The trick is the fatty marrow at the skull base, which normally glows bright on T1-weighted images, the way butter looks pale and rich. When infection moves in, that bright fat gets replaced by infected tissue and the signal goes dark. Spotting that "the butter is gone" on T1 can flag disease before CT shows a single eroded pixel.
The two big traps. First, don't anchor on a normal-looking eardrum — NOE lives in the canal and bone, not necessarily the middle ear, so the otoscope can look reassuring while the skull base is on fire. Second, don't mistake it for a skull base tumor or for routine otomastoiditis — the clinical context (elderly diabetic, Pseudomonas, granulation tissue) plus marrow signal loss across the skull base points to infection, but biopsy is sometimes needed to be sure.
Why missing it is a disaster
If NOE is ignored, the infection keeps tunneling medially toward the jugular foramen and the canals carrying cranial nerves VII, IX, X, XI, and XII. Knock those out and you get facial weakness, hoarseness, and trouble swallowing. Push further and you reach the meninges, the venous sinuses, and the brain — territory where infection becomes life-threatening.
A new cranial nerve palsy in a diabetic with chronic, treatment-resistant ear pain is necrotizing otitis externa until proven otherwise. Facial nerve involvement signals the infection has reached the skull base and is a poor prognostic sign. This is a same-day workup, not a "let's recheck in a week" problem.
The one-line takeaway
If you see an older diabetic with brutal ear pain, gunky drainage that laughs at ear drops, and granulation tissue in the canal, look past the ear and put your eyes on the skull base — on CT for bone erosion, on MRI for the missing marrow fat. Naming it early is the difference between IV antibiotics and a cranial nerve catastrophe.