Definition

Hypersensitivity pneumonitis (HP), also known as extrinsic allergic alveolitis, is a complex condition of varying intensity, clinical presentation, and natural history. It is the result of an immunologically induced inflammation of the lung parenchyma in response to inhalation exposure to a large variety of antigens.¹ These are primarily organic antigens to which patients have been previously sensitized and are hyperresponsive.^2,3

Back to Top

Etiology

Numerous inciting agents have been described, including micro-organisms, agricultural dusts, bioaerosols, and certain reactive chemical species. In order to facilitate learning, they can be organized into microbial agents, animal proteins, and low-molecular-weight chemicals. The bacteria species thermophilic actinomycetes are some of the causative agents of farmer’s lung disease, the first HP described in 1932. They can be found within moldy hay, grain, or silage. Animal proteins are an important cause of HP, exemplified by the disease called bird fancier’s lung. The avian antigens are found typically in the droppings, feathers, or serum of pigeons, parakeets, budgerigars, canaries, chickens, ducks, and turkeys. Less common causative antigens like low-molecular-weight chemicals have been increasingly recognized as a cause of HP. As an example, workers exposed to isocyanates during the production of polyurethane foam, elastomers, adhesives, and paints have developed HP. A list of the most common antigens with their respective sources and diseases can be found on Table 1.^4-6

Back to Top

Prevalence and Risk Factors

The prevalence and incidence of HP in the general population are low. In a population-based study in New Mexico from 1988 to 1990, the incidence of HP in the general population was found to be 0.3 cases per 100,000/year.⁷ A more recent study in the United Kingdom found a similar incidence of HP in the general population: 0.9 cases per 100,000/year.⁸

More useful information to clinicians is the prevalence and incidence of HP in populations exposed to potential antigens. Much of this data has been derived from studies of farmers and bird fanciers. In the medical literature, the prevalence of farmer’s lung among farmers varies from 0.4% to 4.3%.^9-13 The prevalence of bird fancier’s lung among bird fanciers varies from 1.6% to 21%.^11,14,15

Many risk factors have been associated with HP. All ages are susceptible, but working-age adults are most often affected. Men and women, in the other hand, are equally affected.^8,10 An increasingly recognized form of HP develops in patients exposed to mycobacterial antigens present in hot tub (hot tub lung) and contaminated metal-working fluid (machine operator’s lung). Individuals with other professions and hobbies are at increased risk of developing HP: woodworkers, office workers working under forced-air systems, cheese workers, plastic industry workers, painters/refinishers, lifeguards, among others.^4,5,16

Back to Top

Pathophysiology

The pathogenesis of HP starts with repeated antigen exposure. The antigens are generally derived from organic material (microbial agents, animal proteins), but low-molecular-weight chemicals have been linked to HP as well. Their small sizes (about one micron in diameter) allow them to reach the small airways and alveoli, where most of the disease is evident.¹⁷

The inflammatory reaction that occurs after antigen exposure has been widely studied in bronchoalveolar lavage (BAL) of patients with HP. Antibodies specific to the offending agent are increased in both serum and BAL of patients with HP, and the resulting antigen-antibody complex may play a role in the disease process. Following antigen sensitization, inflammatory cells accumulate in the lungs of patients with HP.¹⁸ Neutrophils predominate in the BAL fluid in the first 48 hours, followed by lymphocytes and macrophages.¹⁹ Lymphocytes are the main cells involved in the pathophysiology of HP, and their levels can remain elevated for years after cessation of antigen exposure.¹⁸ The numerous cytokines released by those inflammatory cells (including interferon-y and tumor necrosis factor-alpha) contribute to granuloma formation, an important histopathological feature of HP.

Back to Top

Signs and Symptoms

Three clinical presentations are recognized: acute, subacute, and chronic HP. Acute HP is marked by abrupt onset of cough, dyspnea, fever, and chest pain following heavy exposure to the offending antigen. Headache and malaise are also common. Symptoms begin 4 to 6 hours after exposure and commonly begin subsiding within 24 hours of removal from exposure. Physical exam can disclose inspiratory “pops and squeaks,” usually in the lower lung fields. Subacute HP is characterized by gradual development of the above symptoms with less severe intensity or repeated infrequent acute HP episodes.

Chronic HP involves insidious progressive dyspnea, cough, weight loss, and fatigue. Increased respiratory rate, muscle wasting, use of accessory breathing muscles, and lower lung fields’ inspiratory crackles can be found on physical exam. Digital clubbing is seen in up to half of the cases. The symptoms are progressive and can persist for months or years before the diagnosis is made. Most cases progress towards pulmonary fibrosis with chronic hypoxemic respiratory failure.

The differential diagnosis in patients presenting with the above symptoms is wide. In acute HP, viral or bacterial pneumonia should be considered due to the similarities in symptoms and signs. Organic dust toxic syndrome (ODTS) can also mimic acute HP, especially in farmers presenting with typical symptoms. It is the result of an acute heavy exposure to bioaerosols contaminated with toxin-producing fungi. Differently than HP, no specific antibodies are found in ODTS. Other diseases to think about when suspecting acute HP are toxic fume bronchiolitis (caused by nitrogen dioxide, sulfur dioxide, among other agents), psittacosis (caused by Chlamydophila psittaci), and Q fever (caused by Coxiella burnetii).^4,16

In chronic HP, the differential diagnosis is different but it remains broad. A prospective cohort from the HP Study Group demonstrated this clinical challenge. After analyzing 661 consecutive patients with HP in the differential diagnosis, idiopathic interstitial pneumonias (IIP) and sarcoidosis were the top two alternative diseases diagnosed.² Among the IIP, the most common and also most challenging to differentiate from HP is idiopathic interstitial fibrosis (IPF). It is a chronic, progressive fibrotic disorder of the pulmonary parenchyma that occurs as an aberrant response to lung injury. It has a clinical picture similar to HP, but without a history of exposure. Part of the challenge is that often in HP a significant exposure is not identified either. Another IIP that can be a confounder in the clinical scenario is nonspecific interstitial pneumonitis (NSIP). A recent case series of six patients with HP demonstrated that NSIP can be the sole finding on lung biopsies, without evidence of granulomas.²⁰The inability to distinguish HP from NSIP can result in failure to avoid the provocative exposure, resulting in progressive pulmonary fibrosis. Regarding sarcoidosis, many clinical characteristics distinguished HP from this disease in the HP Study Group. The two main ones were inspiratory crackles (87% in HP and 15% in sarcoidosis) and hilar and/or mediastinal lymphadenopathy (2% in HP and 46% in sarcoidosis).3 Other conditions that should be kept in mind when suspecting chronic HP are: coal worker’s pneumoconiosis (caused by coal dust), berylliosis (caused by beryllium), silicosis (caused by inhaled crystalline silica dust), asbestosis (caused by inhalation of asbestos particles), and various drug reactions.

Back to Top

Diagnosis

Many diagnostic criteria for HP have been proposed, but they all have limitations due to the lack of consensus on disease definition and prospective data validating them.^1,3,21,22 The one proposed by Schuyler suggests that diagnosis of HP requires fulfillment of four or more major criteria and two or more minor criteria.²¹ Major criteria are: (1) symptoms compatible with HP; (2) evidence of exposure to appropriate antigen by history or detection of antibody in serum and/or BAL fluid; (3) findings compatible with HP on chest radiograph or high resolution CT scan (HRCT); (4) BAL lymphocytosis (if BAL performed); (5) pulmonary histologic changes compatible with HP (if lung biopsy performed); and (6) positive “natural challenge” (reproduction of symptoms and laboratory abnormalities after exposure to the suspected environment). Minor criteria are: (1) bibasilar rales; (2) decreased diffusing capacity of the lung for carbon monoxide (DLCO); and (3) arterial hypoxemia, either at rest or during exercise. Conditions with similar symptoms (sarcoidosis or IPF) must be ruled out to confirm diagnosis.

In order to validate clinical diagnostic criteria for HP, the HP Study Group recently undertook a prospective cohort of consecutive patients with HP in the differential diagnosis.² Six significant predictors of HP were identified: (1) exposure to a known offending antigen, (2) positive precipitating antibodies to the offending antigen, (3) recurrent episodes of symptoms, (4) inspiratory crackles on physical examination, (5) symptoms occurring 4 to 8 hours after exposure, (6) and weight loss. This study added to the clinician’s apparatus the possibility of moving from a pre-test probability to a post-test probability based on how many predictors the patients have. The main limitation of the study relies on its applicability: most patients had acute/subacute HP and were from areas with known offending antigens.

Laboratory Tests

Laboratory tests are neither sensitive nor specific, but in acute disease complete blood count may reveal a mild leukocytosis with a left shift several hours after the exposure. Lymphocytosis and eosinophilia may be present.^2,23

Specific serum antibodies may be identified by washing serum over an antigen-impregnated gel. Precipitation of antigen-antibody complex will occur if the patient has significant levels of antibody corresponding to the antigen used. The use of those circulating precipitins has been debatable due to variable sensitivities and specificities in the literature.^16,24 This is reflected in the clinical setting where we may see patients with HP and no antibodies detected, or individuals with positive precipitins (meaning they were exposed to the tested antigen) that do not have the disease. However, if used in the right patient (appropriate exposure, symptoms and signs present), positive circulating antibodies increase the likelihood of HP.^2,24-27

Pulmonary Function Tests

Pulmonary function tests can be normal in acute HP.²⁷ Chronic cases typically demonstrate a restrictive ventilatory defect with small lung volumes, and decreased diffusing capacity (DLco).²⁸ Interestingly, up to half of the patients with HP exhibit increased nonspecific airway reactivity.²⁹ Importantly, none of the above features are specific to HP and can be present in other interstitial lung diseases.

Imaging Studies

Chest X-ray is a useful screening test. It is relatively cheap, noninvasive, and yields information regarding extent of lung involvement. Patients with acute or subacute disease may show bilateral interstitial and alveolar nodular infiltrates. The distribution could be patchy or homogeneous. In patients with chronic disease, diffuse reticulonodular infiltrates and fibrosis are evident, and honeycombing may occur. However, chest X-ray alone is not diagnostic, and a normal chest X-ray does not rule out the disease.

HRCT scan has become an essential tool in the diagnosis of HP, especially the subacute and chronic forms. The characteristic findings in subacute cases are: lobular areas of decreased attenuation or air-trapping, patchy or diffuse ground-glass opacities, and poorly defined small centrilobular nodules (Figure 1).³⁰ In a study by Morell et al, their respective prevalence were: 75%, 65%, and 35%.³¹ The characteristic findings in chronic cases were well demonstrated in a study by Tateishi et al where they evaluated 62 patients with chronic HP. The most common findings were: traction bronchiectasis (100%), interlobular septal thickening (96%), and intralobular reticulation (74%) (Figure 2). The distribution of the above findings was mainly peribronchovascular (90%) with no zonal predominance (87%) in nature.³²

Diagnostic Procedures

Analysis of BAL fluid may be helpful in the diagnostic evaluation of HP. A lymphocyte count greater than or equal to 25% suggests granulomatous disease such as sarcoidosis or HP. A lymphocyte count greater than 50% is particularly suggestive of HP, especially when associated with a mast cell count greater than 1% and a neutrophils count greater than 3%.³³

The utility of transbronchial biopsies (TBBx) was evaluated by Lacasse et al in a very elegant study. In this study, the TBBx yielded a nonspecific result in up to 48% of samples. However, some findings had very good test characteristics when present: diffuse lymphocytic infiltrates had a likelihood ratio (LR) varying from 9 to 12 depending on its severity, and diffuse lymphocytic infiltrates combined with granulomas had a LR of 11.³⁴ Surgical lung biopsy is a more sensitive test, and it should be pursued in the appropriate clinical scenario if the diagnosis is still uncertain.

Pathologically, subacute HP is defined by the triad of: (1) interstitial lymphoplasmocytic pneumonitis, (2) cellular bronchiolitis, (3) and giant cells or poorly formed non-necrotizing granulomas (Figures 3 and 4).³⁵ In the chronic stages of HP, a predominantly fibrotic pattern is seen (Figure 5). The fibrosis mimics other types of interstitial lung disease, particularly usual interstitial pneumonia (UIP). Because of that, the diagnosis of HP should always be made in combination with clinical and radiological findings.³⁶

Back to Top

Treatment

The first-choice treatment is avoidance of the inciting antigen. If avoidance begins during early stages, it can prevent progression towards lung fibrosis. However, because the inciting antigen is not always identified, its avoidance is not always possible. Even when the antigen is known, there are challenges in eliminating the exposure. Some patients may fear losing a job, or may hesitate to remove a pet bird or give up a hobby. In addition, antigens may persist in rooms where birds have been kept for a long time.⁶

Systemic corticosteroids are sometimes required to treat severe disease, but there is no formal evidence that such treatment is associated with improved long-term outcomes. The only randomized controlled trial available was published in 1992 by Kokkarinen et al. Thirty-six patients with acute HP (farmer’s lung) were randomly allocated to prednisolone treatment for 8 weeks (40 mg per day on week 1 followed by a taper) or placebo. After 1 month follow-up, the prednisolone group had a significantly higher DLco. However, after 5 years follow-up no significant differences were found between the groups in terms of lung function.³⁷

Back to Top

Prognosis

The mortality of patients with HP is probably higher than those with no HP, as suggested by a large epidemiologic study in the United Kingdom.⁸ In the United States, the total number of deaths from HP between 1980 and 2002 was 814, and the annual age-adjusted HP death rate seems to have increased during this period (from 0.09 per million in 1980 to 0.29 per million in 2002).³⁸ However, the prognosis differs among the different presentations (acute, subacute and chronic). For example, prognosis is excellent in acute disease for patients who avoid further exposures.³⁹

HRCT has been shown to be a very useful tool in predicting prognosis.^32,40 Hanak et al demonstrated this when they studied 69 consecutive patients with HP diagnosed between 1997 and 2002. Pulmonary fibrosis was seen in 26 patients. During a median follow up of almost 6 years, there were 11 deaths (42%) in the fibrotic group and 1 death (2%) in the non-fibrotic group. The age-adjusted hazard ratio for mortality in patients with HRCT fibrosis was 4.6. Furthermore, the study showed a relationship between the extension of fibrosis and mortality: 21% mortality for <10% of fibrosis, 50% mortality for 10% to 40% of fibrosis, and 83% mortality for >40% of fibrosis.⁴⁰ Pathology can also assist with prognostication, with the presence of histologic fibrosis being associated with decreased survival.^35,41

Patients with chronic HP and fibrosis are also susceptible to acute exacerbations (AE) of fibrotic lung disease. Miyazaki et al demonstrated that 14% of patients with bird fancier’s lung developed AE over a 13-year period. Their study showed that some factors present at diagnosis can predict the development of future AE: low total lung capacity and DLco, low lymphocyte levels in BAL, and UIP-like pattern in histology.⁴²

Back to Top

Prevention

HP is a fairly rare disorder that is linked to either chronic or intense exposures to dust containing organic antigens. Farmers and bird keepers are among those people who are most often affected. Contaminated heating, cooling, or ventilation systems are other potential causes of this condition. Possible preventive measures for individuals who have not developed the disorder are to limit exposures to organic dusts such as composts; to clean bird cages daily; and to ensure that water contained in ventilation systems is clean.

Back to Top

Conclusion

HP is caused by inhalational exposure to a large variety of antigens. Corticosteroids may improve symptoms in the short term, but additional well designed studies are needed to test the long-term effects of steroids and other immunosuppressants in the care for HP patients. Prognosis is good especially in those patients who avoid the inciting antigen, but in a minority of patients the disease will progress to pulmonary fibrosis and death.

Back to Top

Summary

• HP is caused by inhalational exposure to a large variety of antigens.
• The diagnosis of HP requires analysis of clinical, radiological, and pathological data together.
• When HP is suspected or diagnosed, an effort should be made to identify the inciting agent, mainly because the first and most effective treatment is antigen avoidance.
• Corticosteroids may improve symptoms in the short term, but additional well designed studies are needed to test the long term effects of steroids and other immunosuppressants in the care for HP patients.
• Prognosis is good especially in those patients who avoid the inciting antigen, but in a minority of patients the disease will progress to pulmonary fibrosis and death.

Back to Top