UIP/IPF
- UIP is a pattern of lung scarring; IPF is the disease of having that pattern with no other cause found.
- The signature on CT is honeycombing: clustered, stacked little cysts that look like a bath sponge or a slab of honeycomb.
- The scarring loves the bottom and the outside of the lungs — basal and subpleural, getting worse the lower you go.
- Traction bronchiectasis (airways yanked open by surrounding scar) is your tip-off that this is real fibrosis, not just haze.
- Calling a "definite UIP" pattern matters because, in the right patient, it can let doctors diagnose IPF without a surgical biopsy.
Some diseases are a bang — a clot lands, a lung drops, alarms go off. This one is more like rust. Idiopathic pulmonary fibrosis is the lung quietly turning to scar tissue over years, from the bottom up and the outside in, until the once-springy lung handles air about as well as a stale loaf of bread. Your job on the scan is to recognize the texture of that rust and decide how confident you are that it's the real thing.
The two words people keep mixing up
Let's untangle the alphabet soup first, because it trips up everyone.
UIP stands for usual interstitial pneumonia. It's a pattern — a particular look the lung takes when it scars in this specific way. A pattern is descriptive; it doesn't name a culprit.
IPF stands for idiopathic pulmonary fibrosis. It's a disease — and "idiopathic" is doctor-speak for "we looked hard and found no cause." So IPF is what you call it when you see the UIP pattern and you've ruled out the usual suspects (rheumatoid arthritis and other connective tissue diseases, hypersensitivity pneumonitis from birds or mold, asbestos, certain drugs).
A UIP pattern can show up in several diseases. IPF is the one where no cause turns up. So every IPF looks like UIP, but not every UIP is IPF. Same shape, different backstory.
What it actually looks like
The whole diagnosis rides on a high-resolution CT (HRCT) — thin slices that show lung texture in fine detail. Three things, in order of importance:
Honeycombing. This is the money finding. Picture stacking a few rows of bubble wrap, or a wedge of actual honeycomb: clustered cystic air spaces, usually a few millimeters across, sharing walls, stacked in layers. They sit right up against the pleura (the lung's outer lining). When you see clustered, stacked subpleural cysts, your radiology brain should immediately whisper UIP.
Traction bronchiectasis. As the lung scars, it shrinks and stiffens, and it drags the little airways open like a fist closing around a drinking straw and warping it. Airways that should taper smoothly instead look pulled apart and irregular. This tells you the haze you're seeing is genuine fibrosis with tissue loss, not soft, reversible inflammation.
Reticulation. A fine net of crisscrossing white lines — the scaffolding of scar — usually mixed in with the above.
Where it lives: bottom and outside
Distribution is half the diagnosis. UIP is basal and subpleural — it piles up at the lung bases and hugs the outer rim, and it follows a gradient: worst at the bottom, fading as you climb toward the apices. On a coronal view it looks like the disease is creeping up from the floor.
So when you screen for it, always scroll to the very bottom of the lungs and look at the outer edge. That's where the rust starts.
Why the confident call matters
Here's the part that makes this more than a vocabulary quiz. Lung biopsies in fibrosis patients are no joke — these are fragile lungs and a surgical biopsy carries real risk. So when an HRCT shows the classic features in the right clinical setting, expert teams can diagnose IPF without cutting. Pin down the pattern confidently and you may spare someone an operation.
That's why radiologists hedge their language carefully — reporting how certain the pattern is rather than just "yes/no." A textbook basal, subpleural, honeycombing case sits at one end; a hazier, less typical case lands in "could be a few things, let's talk it over."
IPF is usually a multidisciplinary diagnosis — radiologist, pulmonologist, and pathologist comparing notes. The scan rarely gets the final word alone, and that's a feature, not a bug.
The traps
Honeycombing has impostors. Severe traction bronchiectasis seen end-on, or emphysema cysts, can fake the honeycomb look. True honeycombing is clustered, stacked, shares walls, and lives subpleurally. Walls too thin and scattered? Think emphysema. Single airway you can trace? Traction bronchiectasis.
Upper-zone or mid-lung predominance argues against UIP. Disease that favors the tops, or shows lots of ground-glass and air-trapping, points you toward mimics like NSIP or hypersensitivity pneumonitis instead. Distribution is a clue you ignore at your peril.
A quick side-by-side for the two you'll confuse most:
| Feature | UIP / IPF | NSIP |
|---|---|---|
| Dominant finding | Honeycombing | Ground-glass |
| Distribution | Basal, subpleural | Basal, often subpleural sparing |
| Reversibility | Mostly scar (fixed) | More inflammation (some reversible) |
| Typical cause | Often idiopathic (IPF) | Often connective tissue disease |
The one thing to keep
If you remember nothing else: clustered subpleural honeycombing at the lung bases, worst at the bottom, with airways dragged open by scar — that's UIP, and absent another cause, that's IPF. Find the rust at the floor and the outer wall, then decide out loud how sure you are.