Hypersensitivity pneumonitis (HP) is a lung disease that develops in susceptible individuals after inhalational exposure to an organic antigen or chemical compound. Pathogenesis is attributed to a combination of type III (immune complex–mediated) and type IV (delayed) hypersensitivity reactions to the inciting agent.
To provide an overview of the current status of the medical literature regarding hypersensitivity pneumonitis.
A literature search was performed using PubMed and Google search engines. The terms “hypersensitivity pneumonitis” and “extrinsic allergic alveolitis” were used, with the search starting on January 9, 2017, and concluding March 8, 2017.
As a pathologist, it is important to consider hypersensitivity pneumonitis when examining lung specimens because it is often clinically and pathologically overlooked. Recognizing the often subtle findings and correlating them with the patient's history or suggesting a thorough clinical investigation of potential exposures can be of help in identifying the underlying condition so that the patient can be appropriately managed.
Hypersensitivity pneumonitis (HP, also referred to as extrinsic allergic alveolitis in the medical literature) can develop after inhalational exposure to a particular antigen or chemical in a susceptible individual. Hypersensitivity pneumonitis is an inflammatory process involving the small airways and interstitium.1 A number of offending agents have been identified (Table 1), many of which are associated with a name referring to the resulting clinicopathologic condition (ie, farmer's lung, hot tub lung, byssinosis, etc).2,3 Hypersensitivity pneumonitis is often divided into acute, subacute, and chronic forms. Overlapping features and disagreement among physicians4,5 make the distinction between subacute and chronic forms particularly difficult from a clinical and radiographic standpoint. As such, some have proposed an alternative classification that includes acute or episodic HP, chronic HP with acute episodes, and chronic HP without acute episodes.5,6
Symptoms of acute (episodic) HP typically arise just hours after exposure and include malaise, fever, cough, and dyspnea (flulike symptoms). Symptom severity typically peaks around 18 to 24 hours after exposure, and resolution occurs, provided the offending agent is removed. For this reason, acute HP is rarely biopsied because the resulting symptoms are often attributed to a virallike illness or recognized as acute HP. When acute HP is recognized, the patient should be advised to refrain from subsequent exposure because symptoms will recur with subsequent exposure episodes.7 At face value, the treatment may seem relatively simple; however, in certain circumstances the treatment can result in a significant burden to the patient (for example, if the offending agent is linked to an occupation or hobby or the patient's home environment).
Subacute HP and chronic HP is associated with more subtle symptoms, and likely results from chronic, low-level exposure. As such, the correlation between developing symptoms and exposure to an offending agent often goes unrecognized.8 Chronic HP with acute episodes likely results from transient incidence of higher-level exposure. Chronic HP can be progressive, and without appropriate management, fulminant respiratory failure and death from end-stage fibrosis can result.4,5,9,10
Various diagnostic criteria have been proposed for HP11–13 ; however, their utility in clinical practice is often limited. One proposal is based on some or all of the following findings13: (1) known exposure to an offending agent; (2) clinical, radiographic, or physiologic findings; (3) bronchoalveolar lavage with lymphocytosis; (4) positive inhalational challenge testing; and (5) compatible histopathologic findings (Table 2).
Serologic panels for precipitating immunoglobulin G (IgG) antibodies against potential inciting agents have limited utility in clinical practice.14 A positive test result may simply mean exposure to an agent and not actual disease. For example, a large proportion of farmers and bird breeders will have positive serum test results but no clinical disease. Conversely, a negative result does not exclude disease, because the causative agent may not be included in the particular test (because there are multiple potential agents) or the test may simply be nonreactive despite active disease.15 The bottom line is that diagnosis is dependent on a high degree of suspicion and correlation of all clinical, radiographic, and pathologic findings.
PATHOGENESIS
Hypersensitivity pneumonitis pathogenicity results from a combination of both immune complex–mediated (type III) and delayed type (type IV) hypersensitivity reactions after a susceptible person is exposed to a provoking antigen or chemical. The dominant reaction early on and in acute HP seems to be immune complex mediated. With continual exposure (subacute/chronic forms), the predominant reaction shifts to a delayed hypersensitivity (type IV) reaction. A number of chemokines and proinflammatory cytokines mediate a sustained CD8 cytotoxic T-cell response, resulting in the tissue reaction response. Chemokines and cytokines also seem to play a role in macrophage activation, granuloma formation, and fibrosis. Individual susceptibility is poorly understood. Studies suggest a genetic link to the major histocompatibility complex because certain major histocompatibility complex haplotypes increase a person's susceptibility to developing HP.16–20 Cigarette smoking appears to reduce the risk of developing HP as well as the overall severity,21 by a currently unknown mechanism.22
RADIOGRAPHIC FINDINGS
High-resolution computed tomography can provide useful diagnostic information regarding HP.23 Findings are variable and depend on disease stage. The “classic” or characteristic findings include centrilobular ground-glass or nodular opacities with evidence of air-trapping in the mid to upper portion of the lung lobes.24,25 These described findings are more characteristic of subacute HP, which may also show mild fibrotic changes.26 As the disease progresses (chronic HP), emphysematous changes are more apparent and the fibrotic changes become more pronounced; overt honeycomb change often can be seen. The extent of fibrosis in chronic HP (particularly honeycomb change) tends to predict a worse outcome.27 Findings in acute HP are variable and dependent on when the imaging exam was conducted in relation to the onset or improvement of symptoms. Described findings range from essentially normal/unremarkable to ground-glass opacification.28
HISTOLOGIC FINDINGS
Because HP results from inhalational exposure to an offending agent, it is not surprising the classically described pathologic findings are predominantly airway centered. Subacute HP findings typically include peribronchiolar interstitial pneumonia, chronic bronchiolitis, and peribronchiolar giant cells with or without vague/poorly formed granulomas (Figures 1 through 3).11,29 This triad of findings is not seen in all cases of HP, and some cases may only have multinucleated giant cells/granuloma within airspaces.30 Cholesterol clefts31 and Schaumann bodies (nonspecific findings) are occasionally seen in association with the poorly formed granulomas, but necrosis is absent. Generally, eosinophils are few in number or entirely absent, correlating with the proposed pathogenicity of types III and IV hypersensitivity reactions (as opposed to a type I reaction, where eosinophils would be expected). An associated organizing pneumonia pattern (Figure 2), characterized by patchy airspace-filling fibroblastic plugs (Masson bodies), is often seen.14 As the disease progresses (chronic HP), fibrosis becomes increasingly pronounced (Figure 4). Described patterns include subpleural patchy fibrosis with architectural distortion, often with a fibroblastic foci “usual interstitial pneumonitis–like” (UIP-like) pattern; a fibrotic, nonspecific interstitial pneumonia pattern; and a peribronchial, accentuated irregular fibrotic pattern.32 Peribronchiolar fibrosis, bridging fibrosis between adjacent peribronchiolar areas, peribronchiolar metaplasia, and subpleural scarring are frequently encountered.33 With severe fibrosis (end-stage disease), only honeycomb (end-stage) fibrosis is present.
Subacute and chronic hypersensitivity pneumonitis with patchy, bronchiolocentric chronic interstitial pneumonia (hematoxylin-eosin, original magnification ×4).
Figure 2. Cellular bronchiolitis and organizing pneumonia (hematoxylin-eosin, original magnification ×20).
Figure 3. Vaguely formed nonnecrotizing granuloma in the interstitium (hematoxylin-eosin, original magnification ×20).
Figure 4Fibrosis with lymphoid aggregates (hematoxylin-eosin, original magnification ×2).
Subacute and chronic hypersensitivity pneumonitis with patchy, bronchiolocentric chronic interstitial pneumonia (hematoxylin-eosin, original magnification ×4).
Figure 2. Cellular bronchiolitis and organizing pneumonia (hematoxylin-eosin, original magnification ×20).
Figure 3. Vaguely formed nonnecrotizing granuloma in the interstitium (hematoxylin-eosin, original magnification ×20).
Figure 4Fibrosis with lymphoid aggregates (hematoxylin-eosin, original magnification ×2).
The histologic features of acute HP are poorly characterized because patients are rarely (if ever) biopsied as a result of their symptoms, which either are interpreted as a virallike illness or correctly recognized as acute HP. Published histologic findings include intra-alveolar fibrin, neutrophil accumulation (within interstitium, alveolar spaces, and bronchioles) with or without findings of acute lung injury, and possible small vessel vasculitis.34,35
Bronchoalveolar lavage specimens may provide some supportive evidence for the diagnosis of HP (pertinent bronchoalveolar lavage findings are considered in the diagnostic criteria for HP; Table 2). Bronchoalveolar lavage findings that support the diagnosis of HP include a lymphocytosis,36,37 particularly a lymphocytosis with a low CD4:CD8 ratio. It is important to keep in mind that lymphocytosis can be seen in a variety of other conditions (sarcoidosis, connective tissue diseases, drug effect, HIV-associated pneumonitis, etc) in bronchoalveolar lavage specimens, and is therefore not specific for HP. Additionally, although a low CD4:CD8 ratio generally favors a diagnosis of HP rather than sarcoidosis (which is often in the differential diagnosis); CD4:CD8 ratios in HP can be as high as they are in sarcoidosis.38,39 Therefore, a low ratio, although supportive of HP in the appropriate clinical context, is not specific or diagnostic for HP.
DIFFERENTIAL DIAGNOSIS
The lung, like other organs in the body, has a finite number of tissue reaction patterns that can occur. As such, evaluating nonneoplastic conditions in the lung often brings up a large differential diagnosis, particularly when clinical and radiographic information is not known. The light microscopic findings seen in HP are by no means specific, having overlapping features with numerous diseases, including (but not limited to) sarcoidosis (particularly early sarcoidosis), berylliosis, lymphoid interstitial pneumonia (LIP), and low-grade lymphoma (particularly extranodal marginal zone lymphoma of mucosa-associated lymphoid tissue lymphoma), drug reactions, infections, aspiration, and conditions that have a UIP pattern and nonspecific interstitial pneumonia pattern. Diagnostic considerations are often dependent on the phase of HP the patient is in (ie, acute, subacute, chronic, end-stage disease), the amount of tissue present for evaluation, and the availability of clinical and radiographic information. Although certain histologic features may suggest one etiology rather than another, clinical and radiographic correlation is essential in order to suggest one particular condition rather than others in consideration. In HP (and other disease states), clinical, radiographic, and pathologic correlation can help establish an etiology of a disease state and potentially have a positive impact on care. Ideally, the pathologist can help suggest an etiology as opposed to being purely descriptive in a diagnosis. Of course, limitations exist, and at times a descriptive diagnosis is all that can be rendered. When clinical information is incomplete and subtle findings of HP are present on pathologic specimens, suggesting a thorough clinical investigation (including exposure history) may help identify the underlying etiology of the disease process. That said, an inciting agent cannot be identified in 25% of HP cases.40
The vague, poorly formed granulomas seen in HP often lead one to consider sarcoidosis. Features of organizing pneumonia are not typically seen in sarcoidosis; as such, the combination of organizing pneumonia and poorly formed granulomas tends to suggest HP. Additionally, the granulomas seen in sarcoidosis are typically well formed, nonnecrotizing, and often associated with sclerotic fibrosis. Sarcoid granulomas are generally distributed along the lymphatic routes within the pleura, intralobular septa, and along the bronchovascular bundle. This pattern may be helpful in larger tissue specimens; however, a small transbronchial biopsy specimen will often only have tissue around and adjacent to the bronchovascular bundle; as such, other clues need to be considered (poorly formed versus well-formed granulomas, presence of organizing pneumonia).
Granulomas can be seen in a wide array of other conditions (not all of which are described in this review). For example, the histopathologic features of chronic berylliosis essentially mirror the histologic findings of sarcoidosis. Exposure history is therefore critical. Berylliosis occurs from inhalation exposure to beryllium metal or salts. Exposure is typically occupationally related (computer manufacturing or aerospace engineering)41–43 ; however, there are reports in the medical literature of berylliosis developing in those living near facilities using beryllium.44
Ill-defined granulomas and giant cells are also described in LIP, which is a form of lymphoid hyperplasia. Lymphoid hyperplasias (including follicular bronchiolitis, nodular lymphoid hyperplasia, and LIP or diffuse hyperplasia) can be seen concomitantly with each other or in isolation,45 and are associated with a variety of conditions, including autoimmune disorders, HIV (particularly in the pediatric population), common variable immunodeficiency syndrome, collagen vascular diseases, and Castleman disease46–50 . Histologic findings of LIP include expansion of the alveolar septa by lymphoplasmacytic inflammation, which may be bronchocentric. As such, lymphoplasmacytic inflammation in combination with vague granulomas can bring LIP into consideration on histologic grounds alone. It is important to also consider mucosa-associated lymphoid tissue lymphoma when lymphoplasmacytic inflammation is present. Features of mucosa-associated lymphoid tissue lymphoma include infiltration and destruction of the epithelium by small lymphocytes (lymphoepithelial lesions), colonization of germinal centers by marginal zone lymphocytes, and monoclonal B-cell proliferation with light chain restriction.51,52
Infections can also result in similar microscopic findings seen in HP. Essentially all of the classic features described in HP can be seen in infections (ie, peribronchiolar interstitial pneumonia, chronic bronchiolitis, peribronchiolar giant cells with or without vague/poorly formed granulomas,11,27 and possible organizing pneumonia14). A thorough search for potential causative agents is indicated in such cases. Special stains looking for acid-fast organisms, pneumocystis, and other fungal organisms should be considered, particularly when granulomas are present.
Hot tub lung (Table 1), which is caused by nontuberculous mycobacteria,17 has biopsy findings essentially identical to those seen in HP. Mycobacterium avium complex has been demonstrated by culture; therefore, there is some uncertainty as to whether hot tub lung represents an infection, hypersensitivity, or some combination of the two. Hot tub lung disease is placed in the category of hypersensitivity within the medical literature. Therapeutic approaches for hot tub lung are centered on exposure avoidance53,54 (as is HP, see Management and Prognosis below). Systemic corticosteroids are sometimes used in patients with more significant symptoms (again, similar to HP). Antimycobacterial agents have been used; however, their use did not seem to offer added benefit over simple exposure avoidance.55,56 Similar disease has been described with exposure to humidifiers and wind instruments (Table 1).57,58
Like other lung diseases, aspiration can be associated with a variety of patterns,59 which often correlate with the timing and severity of the aspiration event. Chronic aspiration can include bronchiolar chronic inflammation, interstitial fibrosis, and giant cell formation with foreign body material. When lung injury or infection coexists, acute lung injury and/or organizing pneumonia can be present. Given the overlapping histologic features with HP, the latter often can be overlooked, particularly in HP cases where only individual giant cells with Schaumann bodies or cholesterol clefts are seen. In this scenario, the intracellular material can potentially be confused with and misinterpreted as foreign body material. As such, clinical and radiographic correlation should be sought. Aspiration pneumonia results from inhalation of oral or gastric contents. The classic clinical findings include dyspnea, hypoxia, and fever following an aspiration event, and treatment is centered on supportive care and antibiotics if necessary.60 Chronic microaspiration is often clinically silent, signs and symptoms include slowly progressive cough and shortness of breath,60,61 and high-resolution computed tomography findings include tree-in-bud pattern with airway and interstitial thickening and/or basilar centrilobular nodules.62,63
As HP progresses and increasing fibrosis develops, other conditions that can have UIP and/or nonspecific interstitial pneumonia patterns of fibrosis often enter the differential diagnosis. Fortunately, many patients with a clinical HP will have at least some supporting findings histologically,64 and certain findings should prompt one to at least consider chronic HP,32,65 particularly a centrilobular accentuation of fibrosis.66 Centrilobular accentuation of fibrosis is not diagnostic of HP because it can be seen in other conditions,67 but its presence should prompt one to thoroughly look for poorly formed granuloma and giant cells (in the interstitium and associated with small airways) and peribronchiolar metaplasia, all of which suggest prior small airway injury.66 Of course, the presence of one or more of these findings is dependent on the amount and location of tissue sampling; as such, small specimens may yield limited diagnostic information in HP.
A subset of chronic HP patients will have histologic findings indistinguishable from UIP injury pattern secondary to other causes (particularly idiopathic pulmonary fibrosis). This is a particular issue because biopsies are often taken when patients are suspected of having idiopathic pulmonary fibrosis; however, the classic peripheral, lower lobe–predominant, nonsegmental honeycomb fibrosis68 is absent or the radiographic findings overlap with other processes causing fibrosis. Radiographic features supporting chronic HP include fibrosis and honeycomb change involving the lower and upper (in 80% of cases) lobes.69 Some cases have an upper lobe accentuation, a finding characteristic of chronic HP, particularly in cases with a histologic UIP pattern of fibrosis.70 Chronic HP patients with a UIP histologic pattern of fibrosis tend to have a worse prognosis.71 Correlating with radiographic findings is critical, particularly with small biopsy specimens or when only lower lobes have been biopsied. Another radiographic clue favoring chronic HP over other UIP (and nonspecific interstitial pneumonia) conditions is the presence of air trapping, which is exaggerated on expiration secondary to the small airway disease present in HP. This airway trapping is often referred to as “mosaic perfusion or attenuation.”70,72
Drug reactions (which can have a variety of microscopic patterns) can also have overlapping findings with HP. Again, because HP has overlapping and nonspecific histologic features, the importance of clinical and radiographic correlation cannot be overstated.
MANAGEMENT AND PROGNOSIS
The most effective therapy for HP is complete avoidance from the offending agent. Of course, appropriate therapy is dependent on recognition of the disease, and HP is often underrecognized. Additionally, the therapy of choice (exposure avoidance) can result in social and/or economic hardship, particularly if the offending agent is associated with an occupation, beloved hobby, or the patient's home or environment. If avoidance to the offending agents is not feasible or tolerable, protective measures can be attempted (which include the use of respirators, air purification, use of fungicides, dehumidification, mold removal, or other efforts to remove the offending agent).73 However, if the disease persists or progresses, complete avoidance is strongly recommended15 and necessary. Use of corticosteroids seems to accelerate initial recovery; however, their use has no effect on long-term outcome,74,75 and so they are often reserved for patients with more severe symptoms.76,77 With appropriate treatment, most patients will eventually experience near complete recovery, of lung function, providing irreversible fibrosis has not yet developed. Without appropriate management, the natural course is disease progression culminating in end-stage lung fibrosis, likely requiring transplantation.15 Late-state fibrosis in HP is prognostically similar to idiopathic pulmonary fibrosis.24,78
CONCLUSIONS
Careful microscopic scrutiny of lung specimens and correlation with clinical and radiographic findings are essential to recognize or to at least suggest HP (Table 3). Recognizing or suggesting HP can have a great impact on patient care. With appropriate treatment, many patients will have near complete resolution of symptoms and recovery of lung function when irreversible fibrosis is absent.
