Pulmonary Fibrosis Histology: Key Patterns & Diagnosis

Pulmonary fibrosis histology revolves around recognizing a handful of distinct scarring patterns under the microscope, each of which points toward a different underlying disease and a different prognosis. The pattern that matters most is called usual interstitial pneumonia, or UIP, which is the hallmark of idiopathic pulmonary fibrosis (IPF) and the one pathologists spend the most effort identifying or ruling out. But UIP is not the only game in town. Several other patterns, from nonspecific interstitial pneumonia to organizing pneumonia to the fibrosis seen in chronic hypersensitivity pneumonitis, can look deceptively similar, and telling them apart under a microscope changes treatment decisions in ways that imaging alone cannot always achieve.

The Usual Interstitial Pneumonia Pattern

UIP is the signature histological pattern behind IPF, and it has a few features that, taken together, are unlike anything else. The fibrosis starts at the edges of the lung lobules and creeps inward, producing peripheral rings of scarring in the subpleural and paraseptal regions.¹ At low magnification, a pathologist sees a patchwork: zones of dense scarring sitting right next to areas of relatively normal lung. The transition between the two is often strikingly sharp, as if someone drew a line between sick tissue and healthy tissue.

This patchiness is called spatial heterogeneity, and it is one of three pillars the diagnosis rests on. The second is architectural distortion, where honeycombing and dense scars obliterate normal alveolar structures. Honeycombing refers to enlarged airspaces embedded in fibrotic tissue, lined by bronchiolar-type epithelium, and frequently filled with mucin and inflammatory cells. The third pillar is temporal heterogeneity, meaning old scar tissue and fresh injury coexist in the same biopsy.² That fresh injury shows up as fibroblastic foci, small dome-shaped clusters of myofibroblasts and loose connective tissue that represent the active leading edge of the disease. These foci tend to sit at the border between established scar and uninvolved lung.¹

When all three features line up, the pathologist can call UIP with confidence. In early disease, honeycombing may be absent, which complicates matters. The term “probable UIP” exists precisely for cases where the overall architecture looks right but honeycombing has not yet developed. In a study of familial pulmonary fibrosis cases, radiologic imaging suggested UIP in about three-quarters of patients, but when biopsies were actually taken, nearly half were classified as indeterminate for UIP rather than definite UIP, underscoring how tricky early or borderline cases can be.³

Fibroblastic Foci and What They Tell Us

Fibroblastic foci deserve their own discussion because they sit at the center of an ongoing debate about prognosis. These tiny clusters of actively dividing fibroblasts are the biological engine of disease progression. Under the microscope, they stain strongly for collagen I and alpha-smooth muscle actin, both markers of myofibroblast activity.⁴ Within an individual focus, researchers have identified stages of maturation based on the ratio of glycosaminoglycans (early, gel-like matrix) to type I collagen (mature, stiff matrix), with the earliest foci rich in glycosaminoglycans and the latest dominated by dense collagen.⁵

The prognostic value of counting these foci is contested. One study using quantitative scoring found that fibroblastic foci density was a significant predictor of survival in UIP patients.⁶ Another, looking at 43 IPF patients, found no significant relationship between fibroblast focus profusion and survival.⁷ Part of the discrepancy likely comes from how foci are counted. Traditional two-dimensional tissue slices may underestimate the true volume of fibroblastic tissue. Three-dimensional micro-CT reconstructions have shown that what appear as isolated foci on a flat slide are often interconnected networks, and higher 3D focus density does correlate with faster lung-function decline.⁸ The clinical takeaway is that fibroblastic foci clearly represent active disease, but using them as a reliable prognostic yardstick depends heavily on how they are measured.

The NSIP Pattern and How It Differs from UIP

Nonspecific interstitial pneumonia, or NSIP, is the second most common pattern pathologists encounter in interstitial lung disease. Its defining trait is uniformity. Where UIP is patchy and heterogeneous, NSIP involves the lung in a more even, diffuse fashion. The hallmark architectural distortion of UIP, including dense scarring, honeycombing, and fibroblastic foci, is absent or minimal in NSIP.⁹ NSIP comes in two flavors: a cellular form dominated by inflammation in the alveolar walls, and a fibrotic form where collagen deposition takes over. Both tend to be temporally homogeneous, meaning the entire biopsy looks like it is at roughly the same stage of disease rather than showing old scars next to fresh injury.

The distinction matters enormously for patients. NSIP, especially the cellular form, responds better to anti-inflammatory treatment and carries a substantially better prognosis than IPF/UIP. Fibrotic NSIP occupies a middle ground. It can be difficult to distinguish from UIP on a small biopsy sample, which is one reason multidisciplinary discussion, combining the pathologist’s reading with clinical context and radiology, has become the standard approach to diagnosis.

Chronic Hypersensitivity Pneumonitis Under the Microscope

Hypersensitivity pneumonitis (HP) results from repeated inhalation of an offending antigen, anything from mold spores to bird proteins to certain chemicals. The chronic fibrotic form of HP is one of the most important mimics of UIP because it can produce a scarring pattern that looks almost identical at first glance. However, several clues help pathologists tell them apart.

The most important is the distribution of fibrosis. In UIP/IPF, fibrosis concentrates at the periphery of the lobule. In chronic HP, fibrosis is characteristically centrilobular, meaning it clusters around the small airways in the center of the lobule, and often shows bridging fibrosis connecting adjacent lobules.¹⁰ The second major clue is the presence of poorly formed granulomas or isolated multinucleated giant cells scattered among the fibrosis. These are not always easy to find, and the granulomas are typically small and loose rather than well-defined.¹¹ Cellular bronchiolitis, with lymphocytes cuffing the small airways, is another supporting feature.

The histological pattern in chronic HP is not always centrilobular. In a study of pathological findings across HP cases, fibrosis resembled UIP in 18 cases and NSIP in 11 others, with only 22 showing the classic centrilobular pattern.¹¹ This overlap is not trivial: the histological pattern affects prognosis. Patients whose chronic HP shows a UIP-like or fibrotic NSIP-like pattern have worse transplant-free survival than those with purely cellular inflammation or peribronchiolar granulomas.¹² Getting the exposure history right is often as important as anything the microscope reveals.

Smoking-Related Interstitial Lung Diseases

Smoking causes its own family of interstitial changes that overlap with other forms of pulmonary fibrosis. The two most relevant histological patterns are respiratory bronchiolitis-associated interstitial lung disease (RB-ILD) and desquamative interstitial pneumonia (DIP).¹³

Both involve accumulation of pigmented macrophages in the air spaces, but they differ in distribution. DIP fills alveoli diffusely with these tan-brown macrophages, giving large swaths of lung a distinctive look. RB-ILD shows the same cells in a patchy, peribronchiolar pattern, clustered around small airways.¹⁴ The fibrosis in both conditions tends to be mild compared to UIP, and the most effective treatment is smoking cessation. If a pathologist encounters large numbers of pigmented macrophages in a biopsy from a current smoker, these patterns should be considered before jumping to a diagnosis of IPF.

Organizing Pneumonia as a Histological Pattern

Organizing pneumonia, previously called bronchiolitis obliterans organizing pneumonia or BOOP, looks completely different from UIP. Rather than dense collagen scars, the pathologist sees buds of granulation tissue filling alveoli and alveolar ducts.¹⁵ These plugs of young connective tissue are loose and cellular, distinct from the dense, stiff collagen of UIP. The surrounding lung architecture is typically preserved, meaning the overall shape of the lobule has not been destroyed.

Organizing pneumonia can show up as a standalone condition (cryptogenic organizing pneumonia) or as a secondary reaction to infection, drug toxicity, connective tissue disease, or radiation. Its presence on a biopsy does not automatically mean a single diagnosis. Context is everything. When organizing pneumonia appears alongside established UIP, it often signals an acute exacerbation rather than a separate disease process.

What Acute Exacerbations Look Like on Biopsy

An acute exacerbation of IPF is a sudden, often devastating worsening of respiratory function. When tissue is available, the histological picture is typically diffuse alveolar damage, characterized by hyaline membranes lining the alveolar walls, superimposed on the background UIP pattern.¹⁶ In some cases, the acute injury takes the form of organizing pneumonia layered on top of the chronic scarring instead.¹⁷ Less commonly, pathologists have reported diffuse alveolar hemorrhage, pulmonary thromboembolism, or bronchopneumonia as the acute insult.¹⁶ Recognizing the layered nature of these biopsies, acute injury on top of chronic fibrosis, is critical. The acute component drives the immediate clinical crisis, while the chronic UIP beneath determines long-term outlook.

Occupational Dust Exposure and Asbestosis

Environmental and occupational exposures produce their own fibrotic patterns. Asbestosis, the fibrosis caused by asbestos inhalation, requires both diffuse interstitial fibrosis and the identification of asbestos bodies in the tissue to make a definitive pathological diagnosis.¹⁸ Asbestos bodies are fibers coated in iron-rich protein, giving them a golden-brown, dumbbell-shaped appearance under light microscopy. The trouble is that these bodies can be scarce and difficult to spot. When they cannot be identified by routine microscopy, the diagnosis is not necessarily excluded; the patient’s occupational history and imaging findings play a supporting role. This is one of those situations where the pathologist’s report alone may not settle the question.

How Biopsy Technique Affects the Diagnosis

The way tissue is obtained influences how much information the pathologist has to work with. Surgical lung biopsy, typically performed by video-assisted thoracoscopic surgery (VATS), retrieves larger specimens and has a pooled diagnostic yield of about 94%.¹⁹ Transbronchial lung cryobiopsy (TBLC), a less invasive alternative, achieves a pooled diagnostic yield of roughly 77%, rising to about 81% at high-volume centers.¹⁹ Older-style transbronchial forceps biopsies take even smaller pieces and are generally considered insufficient for diagnosing UIP, since the patchy nature of the disease means a tiny sample may miss key features entirely.

Beyond sample size, the mechanical forces involved in obtaining tissue can introduce artifacts that complicate interpretation. Crush artifact from forceps, telescoping of tissue layers, and processing problems during fixation can all mimic real pathology, potentially leading to an incorrect diagnosis.²⁰ Pathologists need to recognize these artifacts and avoid reading disease into mechanically damaged tissue. For diagnoses that hinge on architecture, like UIP, larger and better-preserved specimens substantially reduce the chance of error.

Immunohistochemistry in Fibrotic Lung Disease

Routine histological staining with hematoxylin and eosin does most of the diagnostic work, but immunohistochemistry (IHC) adds a layer of detail that can clarify ambiguous cases and help researchers understand disease biology. In IPF lungs, fibroblastic foci stain strongly for alpha-smooth muscle actin and collagen I, confirming myofibroblast activation. The alveolar epithelial cells overlying these foci do not stain for Ki-67, a marker of cell division, even though hyperplastic epithelial cells elsewhere in the same lung do. This suggests the cells immediately over the foci are not proliferating, a subtlety that may be relevant to understanding how the foci advance.⁴

At the tissue level, staining for immune cell markers like CD68 (macrophages), CD4 and CD8 (T-cell subtypes), and CD79a (B cells) helps characterize the inflammatory environment. In IPF lungs with more advanced fibrosis and honeycombing, CD79a-positive B cells and lymphoid follicles tend to be more prominent.²¹ Researchers have also explored serum biomarkers like CA-125 and CA19-9, whose cellular location in lung tissue was confirmed by IHC in IPF samples.²² These markers are not part of routine diagnostic workup, but they point toward a future where blood tests and tissue staining work together to track disease activity more precisely than pattern recognition alone.

How Antifibrotic Treatment Affects What the Pathologist Sees

The two approved antifibrotic drugs for IPF, pirfenidone and nintedanib, slow disease progression by clinical measures like lung-function decline, but what do they actually do to the tissue? A study of explanted lungs from IPF patients who underwent transplant found that the overall amounts of dense collagen fibrosis, fibroblastic foci, and alveolar macrophages were similar whether patients had been treated or not.²³ What did differ was the amount of acute lung injury at the time of transplant: pirfenidone-treated lungs showed less diffuse alveolar damage and organizing pneumonia, and nintedanib-treated lungs trended toward less lymphocytic inflammation.²³

In animal models, the picture is more encouraging. Mice given these drugs after bleomycin-induced fibrosis showed less alveolar structural damage, thinner alveolar walls, and reduced collagen deposition compared to untreated animals. IHC confirmed that alpha-smooth muscle actin and collagen I expression were significantly lower in the treated groups.²⁴ The gap between human and animal findings likely reflects the fact that patients who reach transplant represent the most advanced end of the disease spectrum, where established fibrosis is essentially irreversible. The drugs may prevent new fibrosis from forming without undoing what is already there.

When the Pattern Does Not Fit Any Category

Despite all of these defined patterns, a meaningful fraction of interstitial lung disease biopsies resist classification. The label “unclassifiable interstitial lung disease” exists for cases where the biopsy shows features of more than one pattern, or where sampling limitations prevent a confident call.²⁵ A small biopsy may show fibrosis that could be UIP or fibrotic NSIP depending on what the rest of the lung looks like. A patient with overlapping autoimmune features might have tissue that borrows from multiple patterns. In the familial pulmonary fibrosis cohort mentioned earlier, about 47% of biopsied patients fell into the “indeterminate for UIP” category, far outstripping those classified as definite UIP.³ This is not a failure of pathology. It reflects the biological reality that fibrosis in the lung does not always sort itself neatly into textbook categories.

Multidisciplinary team discussion, where pulmonologists, radiologists, and pathologists review each case together, has become the gold standard for navigating these gray zones. The pathologist’s report is one input, not the final word. When the clinical history, imaging pattern, and biopsy findings point in the same direction, confidence is high. When they conflict, the “unclassifiable” label is more honest and more useful than forcing the case into an ill-fitting box.

Artificial Intelligence in Lung Pathology

Human scoring of fibrosis severity is inherently subjective. Traditional grading systems rely on the pathologist’s visual impression, and interobserver variability is a well-known problem.²⁶ Artificial intelligence is beginning to change this. Deep learning algorithms trained on digitized lung tissue slides can quantify fibroblastic foci, interstitial inflammation, and intra-alveolar macrophages in ways that go beyond what the human eye can reliably measure, and these AI-derived measurements have shown promise as prognostic indicators in IPF.²⁷

The appeal is not just consistency. AI can process whole-slide images and generate continuous, granular measurements rather than the coarse 0-to-3 scales pathologists typically use. In preclinical research, automated quantification of fibrosis in animal models already outperforms manual scoring in reproducibility.²⁶ Whether this translates into better clinical decision-making for patients is still being tested. The classification framework itself, from the first coining of the term UIP in the 1960s to the guidelines in use today, has evolved through decades of refinement.²⁸ AI represents the next iteration of that evolution, not a replacement for pathologist expertise but a tool that could make the subjective parts of fibrosis diagnosis measurably more reliable.

1
Modern Pathology. The histologic diagnosis of usual interstitial pneumonia of idiopathic pulmonary fibrosis. Where we are and where we need to go
2
European Respiratory Review. Differential diagnosis of usual interstitial pneumonia: when is it truly idiopathic?
3
PubMed Central. Clinical, radiological and histopathological features of patients with familial pulmonary fibrosis
4
PubMed Central. Idiopathic pulmonary fibrosis: immunohistochemical analysis provides fresh insights into lung tissue remodelling with implications for novel prognostic markers
5
PubMed. The definition of fibrogenic processes in fibroblastic foci of idiopathic pulmonary fibrosis based on morphometric quantification of extracellular matrices
6
PubMed. Quantitative analysis of fibroblastic foci in usual interstitial pneumonia
7
PubMed. Profusion of fibroblast foci in patients with idiopathic pulmonary fibrosis does not predict outcome
8
The Journal of Clinical Investigation. Three-dimensional characterization of fibroblast foci in idiopathic pulmonary fibrosis
9
PubMed. Nonspecific interstitial pneumonia: pathologic features and clinical implications
10
PubMed. Pathology of hypersensitivity pneumonitis
11
PubMed. Granulomas and giant cells in hypersensitivity pneumonitis
12
PubMed. Pathologic Findings and Prognosis in a Large Prospective Cohort of Chronic Hypersensitivity Pneumonitis
13
Thorax. Challenges in pulmonary fibrosis · 4: Smoking-induced diffuse interstitial lung diseases
14
PubMed. Smoking-related interstitial lung diseases: radiologic-pathologic correlation
15
PubMed Central. Organizing Pneumonia as a Histopathological Term
16
Respiratory Medicine Case Reports. Diffuse alveolar hemorrhage complicating acute exacerbation of IPF
17
PubMed. Acute exacerbations of fibrotic interstitial lung disease
18
PubMed Central. Asbestos exposure diagnosis in pulmonary tissues
19
European Respiratory Review. Diagnostic yield and safety of transbronchial lung cryobiopsy and surgical lung biopsy in interstitial lung diseases: a systematic review and meta-analysis
20
PubMed. Interpretation of tissue artifacts in transbronchial lung biopsy specimens
21
PubMed. Pathology of Idiopathic Pulmonary Fibrosis Assessed by a Combination of Microcomputed Tomography, Histology, and Immunohistochemistry
22
The Lancet Respiratory Medicine. An epithelial biomarker signature for idiopathic pulmonary fibrosis: an analysis from the multicentre PROFILE cohort study
23
PubMed. Histopathological and molecular analysis of idiopathic pulmonary fibrosis lungs from patients treated with pirfenidone or nintedanib
24
PubMed Central. Pirfenidone and nintedanib attenuate pulmonary fibrosis in mice by inhibiting the expression of JAK2
25
PubMed Central. Unclassifiable interstitial lung disease: a pathologist’s perspective
26
PubMed Central. Machine learning-based automated fibrosis quantification in preclinical models of pulmonary fibrosis
27
PubMed. Artificial intelligence identifies inflammation and confirms fibroblast foci as prognostic tissue biomarkers in idiopathic pulmonary fibrosis
28
PubMed Central. Diagnosis of interstitial lung diseases: from Averill A. Liebow to artificial intelligence