NSIP
- NSIP (nonspecific interstitial pneumonia) is the "uniform and tidy" interstitial lung disease — one process, one timeline, spread evenly through the lungs.
- The classic CT look is ground-glass haze in the lower lungs, often hugging the back, frequently with subpleural sparing (a thin clean rim right against the chest wall).
- It comes in two flavors: cellular (mostly inflammation, looks better with treatment) and fibrotic (scar, with traction bronchiectasis and reticulation).
- Honeycombing is usually scarce or absent — that's a big part of how it separates from UIP, its meaner cousin.
- NSIP is strongly tied to connective tissue disease, so when you see this pattern, you go hunting for an underlying autoimmune cause.
Most interstitial lung diseases are chaos — a patchwork of old scar next to new inflammation next to weirdly normal lung, like a quilt sewn by three people who weren't speaking. NSIP is the polite exception. It's the one that does one thing, at one tempo, more or less everywhere at once. If the lung were a lawn, NSIP is an even haze of frost across the whole yard, not a few random scorched patches.
The name is almost an apology. Nonspecific interstitial pneumonia — radiology's way of admitting "we see a real, repeatable pattern, we just couldn't pin it to a single villain." Don't let the bland name fool you; the imaging is actually quite recognizable.
What NSIP looks like on CT
This is a HRCT diagnosis through and through. The dominant finding is ground-glass opacity — that hazy gray veil where you can still see the vessels through it, like looking at the lung through a foggy shower door. It's bilateral, it's fairly symmetric, and it loves the lower lobes and the back of the lungs (the posterobasal regions, where gravity has been quietly recruiting fluid and cells for years).
Mixed into the haze you'll often find fine reticulation (a net-like overlay of thin lines) and, in more advanced disease, traction bronchiectasis — airways yanked permanently open because the scar around them is shrinking and pulling, like guy-wires tugging a tent pole crooked.
The signature move: subpleural sparing
Here's the detail that makes attendings nod. In a sizable chunk of NSIP cases, the disease creeps right up toward the chest wall and then stops just short of it, leaving a thin clean rim of normal lung hugging the pleura. It's as if the haze respects a tiny "do not cross" border.
Subpleural sparing — a preserved sliver of normal lung directly against the chest wall — is suggestive of NSIP. It's not present in every case, but when you see it, NSIP should jump to the top of your list.
That single feature is genuinely useful, because most fibrosing lung disease does the opposite — it attacks the periphery first. A disease that politely leaves the edge alone is showing you its hand.
Cellular vs. fibrotic — two tempers
NSIP isn't one personality; it's two:
| Subtype | What's happening | CT clues | Why you care |
|---|---|---|---|
| Cellular | Mostly inflammation, little scar | Ground-glass dominates; minimal reticulation; little/no traction bronchiectasis | Tends to respond better to treatment |
| Fibrotic | Inflammation plus established scar | Reticulation, traction bronchiectasis, volume loss alongside the haze | More fixed; the scar part doesn't un-scar |
The cellular form is the optimist — a lot of what you see is angry, treatable inflammation. The fibrotic form has already laid down permanent scaffolding, and CT shows it: the lung volumes shrink and the airways get dragged open.
The relationship that defines it: connective tissue disease
If there's one reflex to build, it's this: NSIP pattern? Go looking for autoimmune disease. NSIP is the interstitial pattern most associated with connective tissue diseases — think scleroderma, polymyositis/dermatomyositis, and related conditions. Often the lungs raise the alarm before the rheumatologist has even been called.
A clean, symmetric, lower-lobe ground-glass pattern in a relatively young or middle-aged patient — especially a woman, especially a nonsmoker — should prompt the question: is there an undiagnosed connective tissue disease hiding here? The chest CT can be the first clue to a systemic problem.
Don't confuse it with UIP
This is the comparison every exam loves, because NSIP's optimism and UIP's grimness change everything downstream — UIP/IPF carries the worse prognosis.
| Feature | NSIP | UIP |
|---|---|---|
| Distribution | Lower lobes, often with subpleural sparing | Lower lobes, subpleural and peripheral predominant |
| Honeycombing | Scarce or absent | A defining feature |
| Ground glass | Prominent | Limited, usually overshadowed by reticulation/honeycomb |
| Uniformity | Temporally uniform, even | Heterogeneous — patches of every age side by side |
Fibrotic NSIP and UIP genuinely overlap and can be hard to separate on imaging alone — that's why these cases go to a multidisciplinary discussion with pathology and rheumatology. Don't over-commit to "NSIP" just because honeycombing is sparse; absence of honeycombing leans NSIP but doesn't clinch it. More on these traps in ILD pattern pitfalls.
It's also worth keeping the great ground-glass mimic in mind: organizing pneumonia, which prefers patchy peripheral and peribronchial consolidation rather than NSIP's even basal haze, and hypersensitivity pneumonitis, which tends to spread more evenly top-to-bottom and brings air-trapping along.
The one thing to remember
NSIP is the uniform, ground-glass-heavy, lower-lobe, honeycomb-light interstitial disease, and subpleural sparing is its calling card. See that pattern and your next two moves are reflexes: think connective tissue disease, and ask whether this is the gentler NSIP or its meaner lookalike, UIP.