How to investigate a patient with suspected interstitial lung diseaseBMJ 2010; 340 doi: https://doi.org/10.1136/bmj.c2843 (Published 09 June 2010) Cite this as: BMJ 2010;340:c2843
- Owen J Dempsey, consultant chest physician1,
- Keith M Kerr, professor of pulmonary pathology2,
- Hardy Remmen, consultant cardiothoracic surgeon3,
- Alan R Denison, clinical senior lecturer and honorary consultant thoracic radiologist4
- 1Department of Respiratory Medicine, Aberdeen Royal Infirmary, Foresterhill, Aberdeen AB25 2ZN
- 2Aberdeen University Medical School, Foresterhill, Aberdeen AB25 2ZD
- 3Department of Cardiothoracic Surgery, Aberdeen Royal Infirmary
- 4Aberdeen Biomedical Imaging Centre, Aberdeen Royal Infirmary
- Correspondence to: O J Dempsey
Patients with interstitial lung disease often present with breathlessness, chronic cough, inspiratory crackles on auscultation, and abnormal spirometry.1 2 Over 200 different types of disease can cause thickening of the pulmonary interstitium, with the common final pathway for many of these being fibrosis (figs 1 and 2)⇓ ⇓.3 4 Some forms of interstitial lung disease can be life threatening, such as idiopathic pulmonary fibrosis, approximately 4000 new cases of which are diagnosed in England and Wales each year. Survival rates of 57% at three years and 43% at five years are worse outcomes than those for some cancers.5 About 3000 new cases of sarcoidosis are diagnosed each year in the UK.5 6
Early recognition and treatment of interstitial lung diseases may prevent progression to irreversible pulmonary fibrosis and respiratory failure. This review provides a guide to the initial investigation of patients with suspected interstitial lung disease and provides a summary of further investigations.
Sources and selection criteria
We sought papers that were fully published between 1966 and April 2010 using appropriate index terms (interstitial lung disease, diffuse parenchymal lung disease) the National Library of Medicine’s computerised search service (providing access to Medline, PreMedline and other related databases). We also consulted Cochrane database systematic reviews and a personal archive of references.
What are the most common interstitial lung diseases and how are they classified?
Some forms of interstitial lung disease are idiopathic, whereas others are associated with connective tissue diseases, commonly prescribed drugs, smoking, and exposures to environmental factors such as dusts.7 The histopathological classification of interstitial lung diseases is complex, simplified in figure 3⇓.3 The key clinical point is to try and identify a known cause if at all possible. The most commonly encountered conditions are:
Connective tissue disease associated pulmonary fibrosis—for example, rheumatoid or scleroderma lung10
What questions are important to ask when taking a history?
Box 1 lists medications that may be associated with adverse effects in the lung. Prescribed and “over the counter” medications, herbal therapies, and illicit drugs have been associated with interstitial lung diseases; www.pneumotox.com provides information about drug induced lung disease.12 13 Drug induced interstitial lung disease is straightforward to recognise if it develops within days of prescription, but presentation may be subacute (over months) or chronic (years) even with the same drug (for example, nitrofurantoin).14 15 Similarly, various pathological patterns of interstitial lung disease can occur for the same drug (for example, methotrexate) ranging from a relatively benign interstitial pneumonitis through to life threatening diffuse alveolar damage. If recognised early enough, most forms of drug induced interstitial lung disease are reversible12 13
Antibiotics—cephalosporin, minocycline, nitrofurantoin, quinine
Rheumatology treatments—gold, leflunomide, methotrexate, NSAIDs, penicillamine, sulphasalazine, tumour necrosis factor α blockers (infliximab, etanercept)
Cardiology treatments—amiodarone, angiotensin converting enzyme inhibitors, aspirin, atenolol, statins
Oncology treatments—bleomycin, busulfan, chlorambucil, melphalan, methotrexate, mitomycin C, tyrosine kinase inhibitors (dasatnib, imatinib)
Illicit drugs—cocaine, heroin, intravenous talc, methadone
Immunomodulators—azathioprine, cyclophosphamide, epidermal growth factor receptor inhibitors (gefitinib, erlotinib), interferons, methotrexate, rapamycin derivatives (sirolimus), tumour necrosis factor α blockers, rituxumab
Miscellaneous—high flow oxygen, inhalation or aspiration of fat-containing substances (drugs containing mineral oils, certain laxatives, petroleum jelly), paraquat, radiotherapy
History of a connective tissue disease
Any of the connective tissue diseases may be complicated by interstitial lung disease.10 16 17 Occasionally patients present with pulmonary symptoms before rheumatological manifestations become obvious. Clinical clues to the possibility of an underlying connective tissue disease are shown in figure 6.⇓
Hobbies, occupation, travel, and other possible environmental exposures should be taken into account. Taking a detailed current and previous occupational history can be time consuming;18 a practical solution is to ask patients in advance to prepare a chronological list of jobs and to bring this list to their initial assessment. Inorganic dusts linked with interstitial lung disease include asbestos,19 coal and silica,20 and beryllium.21 New industries have generated new forms of interstitial lung disease, such as flockworker’s lung in manufacturers of synthetic microfibres (flock) of nylon,22 and metalworker’s lung in those exposed to metalworking fluid aerosol.23 Hypersensitivity pneumonitis can be caused by antigens in organic dusts, such as Saccharopolyspora rectivirgula in hay, which causes farmer’s lung, and proteins in bird droppings and on feathers that cause bird or pigeon fancier’s lung.11 Case reports have associated feathers in duvets with hypersensitivity pneumonitis.24 25 and more than 20 case reports have linked domestic mould exposure with hypersensitivity pneumonitis.26 27
Questions about recreational activities and travel may reveal an important exposure—for example, Mycobacterium avium complex causes “hot tub lung” and Trichoderma koningii “woodworker’s lung.”28 A meta-analysis of six case-control studies conducted in the UK, United States, and Japan reported that smoking (odds ratio 1.58, 95% confidence interval 1.27 to 1.97), agriculture or farming (1.65, 1.20 to 2.26), livestock (2.17, 1.28 to 3.68), wood dust (1.94, 1.34 to 2.81), metal dust (2.44, 1.74 to 3.40), and stone or sand (1.97, 1.09 to 3.55) were associated with an increased risk of idiopathic pulmonary fibrosis.29
Cigarette smoking is associated with several interstitial lung diseases.30 31 A recent population based cohort study of 2563 adults in the United States suggested that smoking may cause subclinical parenchymal lung disease that is detectable by spirometry and computed tomography imaging.32 Curiously, some forms of interstitial lung disease, such as hypersensitivity pneumonitis and sarcoidosis, are less common in smokers.33 34
The nature of an observed link between chronic recurrent microaspiration of gastric acid and the development of chronic lung disease is not clear.38 The prevalence of reflux is increased in patients with idiopathic pulmonary fibrosis and scleroderma lung, although whether its development and course can be influenced by anti-reflux treatment is not known.39 40
What are the key features on examination?
Clinical examination does not always reveal signs of illness. However, inspiratory crackles may be heard on chest auscultation.2 Crackles are typically best heard over the lower lobes and axillary regions in patients with idiopathic pulmonary fibrosis. In those with sarcoidosis crackles may be absent, despite radiological abnormality. Mid-inspiratory “squeaks” are suggestive of bronchiolitis. Clubbing of the fingers occurs in about half of patients with idiopathic pulmonary fibrosis and may accompany other interstitial lung diseases. If interstitial lung disease is advanced patients may be cyanosed and show signs of secondary pulmonary arterial hypertension.41 Examples of extrapulmonary signs that may occur are shown in figure 6⇑.
What first line investigations should be considered before referral?
Figure 7⇓ outlines the initial assessment of patients with suspected interstitial lung disease. Full history and examination may identify known causes of such disease and will help direct subsequent investigations.
The chest radiograph can be normal in early interstitial lung disease.42 By the time the disease is clinically apparent chest radiographs will usually be abnormal. Abnormalities include decreased lung volumes, subpleural reticular opacities, peripheral migratory air space shadowing, and mediastinal or hilar lymph node enlargement. For example, bilateral hilar lymphadenopathy with upper lobe pulmonary infiltrates in a young adult suggests sarcoidosis; basal fibrosis with pleural thickening and pleural plaques suggests asbestosis. The chest radiograph may also help to identify lung cancer, infection, or pneumothorax, which may be associated with interstitial disease. Chest radiography has limited sensitivity and lacks specificity for diagnosing subtypes of interstitial lung disease; high resolution computed tomography is more useful in diagnosis and differentiation.1
An oxygen saturation of less than 94% at rest is abnormal and should prompt referral for more detailed assessment by a chest physician.
A full blood count, including eosinophils, and a serum biochemistry profile (including renal and liver function tests, calcium, glucose, thyroid function, C reactive protein) are potentially useful baseline investigations in combination with clinical and radiological findings.
Since many patients who present with a cough and breathlessness may also have a history of smoking and other cardiovascular risk factors, a baseline electrocardiogram is necessary. It may show signs of right ventricular strain, prompting further investigation of suspected secondary pulmonary vascular disease. Rarely, conduction defects and arrhythmias are found in patients with cardiac sarcoidosis.47
What are the second line investigations?
An algorithm for further investigation is presented in figure 7. Some approaches are discussed here.
Other blood tests
An arterial blood gas is needed to confirm respiratory failure in patients with a resting oxygen saturation of less than 94%, in those who are polycythaemic and those with signs suggestive of cor pulmonale. If a connective tissue disease is suspected then an immunology screen (antinuclear antibody, extractable nuclear antigens, rheumatoid factor, and anti-neutrophil cytoplasmic antibody) will be required. If an environmental factor is suspected, specific serum IgG antibody screens can be performed for relevant antigens (such as Saccharopolyspora rectivirgula, Thermactinomycetes vulgaris, Aspergillus for mouldy hay).
British Thoracic Society guidelines suggest that measurement of serum angiotensin converting enzyme has a limited role in the diagnosis of suspected sarcoidosis and does not contribute to monitoring patients when added to serial lung function and imaging.1 Furthermore, assays to detect the enzyme have limited sensitivity and specificity and values are affected by angiotensin converting enzyme gene polymorphisms.48 Levels of angiotensin converting enzyme are raised in many other conditions including other granulomatous diseases (leprosy, tuberculosis, histoplasmosis), Gaucher’s disease, and hyperthyroidism49 Results are invalidated if a patient is taking an angiotensin converting enzyme inhibitor.
Testing for HIV infection may be helpful, since some forms of interstitial lung disease are associated with immunodeficiency, such as lymphocytic interstitial pneumonia.50 Brain natriuretic peptide, secreted in response to ventricular stretch by cardiomyocytes, is a marker of right ventricular dysfunction. Although the test is not available in all centres, some studies in patients with interstitial lung disease suggest that elevated levels of the peptide can identify patients with pulmonary hypertension and poorer prognosis.51 52 53 54 A wide range of serum biomarkers (lung epithelium specific proteins, cytokines, and other biological markers of cellular activity) may also theoretically reflect more severe or widespread interstitial lung disease.55 However, as yet, no perfect biomarker has been identified, and since the tests are costly and of uncertain clinical usefulness their role is unclear.56
What is the role of high resolution computed tomography?
High resolution computed tomography is a specialised imaging test in which thin (<1.5 mm) axial sections of the lungs are acquired with 10 mm or 20 mm gaps between sections. No intravenous contrast is used. Image reconstruction with a high spatial frequency algorithm allows sub-millimetre structures to be visualised. A high resolution computed tomogram is a comparatively low dose test, with a radiation exposure six times less than that of a conventional thoracic computed tomogram, and equivalent to approximately 50 chest radiographs. It is accordingly a sampling examination and should be reserved for patients who have a raised clinical or radiographic suspicion of interstitial lung disease. Interpretation of high resolution computed tomography is best done by radiologists with an interest in thoracic imaging and with ready access to multidisciplinary input.1 Typical reasons for high resolution computed tomography in interstitial lung disease are:
To detect interstitial lung disease in patients with normal or equivocal chest radiograph findings
To focus the differential diagnosis in patients with obvious but non-specific chest radiograph abnormalities
To guide the type and site of lung biopsy
To evaluate disease reversibility.
Characteristic features of idiopathic pulmonary fibrosis on high resolution computed tomography (fig 4⇑) include bilateral and symmetrical bibasal pulmonary reticular shadowing with destroyed and fibrotic lung tissue that contains numerous cystic airspaces with thick fibrous walls (honeycombing). The reticular pattern reflects a combination of interlobular septal thickening and intralobular opacities. As the fibrosis progresses, traction bronchiectasis is often present. Ground glass opacity is often present in idiopathic pulmonary fibrosis and in the absence of adjacent fibrosis it is associated with response to treatment and longer survival.57 Other common features include reactive mediastinal lymphadenopathy and, since many patients are smokers, coexisting emphysema. Most patients with clinical evidence of interstitial lung disease will have abnormal high resolution computed tomography features. High resolution computed tomography has better sensitivity than chest radiography for the detection of interstitial lung disease.58 However, patients with very early disease may not have imaging abnormalities. Conversely, some patients (in particular older people and smokers) may have parenchymal changes in the absence of clinical or physiological abnormality.59 60
The main clinical value of high resolution computed tomography is the identification of interstitial lung disease and the most common variant, idiopathic pulmonary fibrosis. In an appropriate clinical context, the presence of classic features of idiopathic pulmonary fibrosis on high resolution computed tomography allow a confident non-invasive diagnosis without the need for lung biopsy.61 62 However, diagnostic doubt may often remain because of the considerable overlap between findings in “non-specific interstitial pneumonia” and variants of interstitial lung disease. In a well conducted retrospective study of 92 patients with biopsy proved interstitial lung disease, blinded observers made the correct diagnosis from high resolution computed tomography images in 79% of cases and identified some features that would favour idiopathic pulmonary fibrosis over non-specific interstitial pneumonia,63 suggesting that a diagnosis of non-specific interstitial pneumonia will usually require a lung biopsy.64 65 High resolution computed tomography at diagnosis is valuable in determining prognosis. Several recent studies have shown that patients with features of fibrosis at diagnosis (including honeycombing and reticular change) have a worse prognosis and that the extent of lung fibrosis is a predictor of death.64 66
What is the role of pulmonary function tests?
Pulmonary function tests offer useful information on the degree of physiological impairment and disease severity, and the results provide a baseline against which disease progression and response to treatment can be measured
Diseases of the lung parenchyma, such as interstitial lung diseases, typically display a “restrictive” pattern on spirometry, as the forced expiratory volume in one second (FEV1) and the forced vital capacity (FVC) are both reduced, which results in a preserved or sometimes increased FEV1:FVC ratio. However, this is not always the case, particularly in smokers who may have coexisting emphysema and have normal, or near normal, lung volumes resulting from the opposing effects of hyperinflation and fibrosis.67 68 69 Other interstitial lung diseases associated with relative preservation of lung volumes include respiratory bronchiolitis associated interstitial lung disease, Langerhans’ cell histiocytosis, and lymphangioleiomyomatosis (either sporadic or in association with tuberous sclerosis). Spirometry is usually done routinely at clinic visits.70 In patients with idiopathic pulmonary fibrosis, a fall in forced vital capacity of greater than 10% is considered clinically important, and is associated with a 2.4-fold increase in risk of mortality.1 71
Transfer factor (also known as diffusing capacity for carbon monoxide) is measured in pulmonary function laboratories, and allows a non-invasive assessment of gas exchange.72 Transfer factor is typically reduced in patients with interstitial lung disease, secondary to thickening of the alveolar-capillary barrier impairing gas exchange. A spuriously raised transfer factor measurement can occasionally result from minor pulmonary haemorrhage from a systemic vasculitis or, conversely, from anaemia in the context of an interstitial lung disease. In patients with idiopathic pulmonary fibrosis, serial measurement of transfer factor can help to clinicians to estimate prognosis and track progression of disease, and a change may prompt referral for transplant in suitable patients; a change in transfer factor of 15% or greater from baseline is considered clinically important.1 73 In a UK study of 115 patients with idiopathic pulmonary fibrosis referred to a transplant centre, a transfer factor under 40% (advanced disease) with high resolution computed tomography scoring predicted death within two years with a specificity of 84% and sensitivity of 82%.74 A disproportionately reduced transfer factor in a patient with relatively normal spirometry may suggest that the underlying disease process is vascular rather than interstitial lung disease and these patients should be screened for pulmonary hypertension.75
Exercise tests may help evaluation of disease severity.76 The six minute walk test records oxygen saturation before, during, and after exercise and measures the total distance walked. Some patients with interstitial lung disease may appear deceptively well at rest, with normal oximetry and spirometry, but become profoundly hypoxic when walking.76 Exercise associated oxygen desaturation significantly correlates with pulmonary hypertension75 and patients who have these features merit more detailed assessment for pulmonary artery hypertension including echocardiography.77 Substantial oxygen desaturation (<88%) during a six minute walk test has been associated with a poor prognosis.78 For example, in a retrospective study of 105 patients with biopsy proved interstitial lung disease (83 with the “usual interstitial pneumonia” histological pattern, typically associated with idiopathic pulmonary fibrosis), those with a usual interstitial pneumonia pattern who experienced desaturation to less than 88% for a minute or longer during a six minute walk test had significantly reduced survival (hazard ratio 4.47, 95% confidence interval 1.58-12.64).79 Thus, desaturation during exercise is a useful prognostic marker and identifies patients at high risk who should be screened for pulmonary hypertension and, if appropriate, considered for early transplant referral.
What is the role of echocardiography?
British Thoracic Society guidelines suggest that pulmonary hypertension should be considered in patients with interstitial lung disease who have either breathlessness or lung dysfunction (reduced transfer factor or desaturation on exercise).1 Pulmonary hypertension is defined as a mean pulmonary artery pressure of greater than 25 mm Hg at rest, or 35 mm Hg or greater during exercise, with normal left atrial pressure.75 Right heart catheterisation is the gold standard test for measuring pulmonary artery pressure but is invasive, and typically transthoracic echocardiography is performed first.80 81 Importantly an estimation of systolic pulmonary artery pressure can only be made in patients who have a tricuspid regurgitant jet (70-80%) evident at echocardiography. If so, the systolic pulmonary artery pressure can be estimated and used to calculate the mean pulmonary artery pressure as follows:82
Mean pulmonary artery pressure=(0.61×systolic pulmonary artery pressure)+2
For example, an echocardiography assessed systolic pulmonary artery pressure of 50 mm Hg equates to a mean pulmonary artery pressure of 32 mm Hg.
The prevalence of pulmonary hypertension varies widely according to the disease and severity of lung impairment, but is relatively high (30-40%); higher levels are associated with increased mortality.77 For example, in a single centre American study of 88 patients with idiopathic pulmonary fibrosis who had a trans-thoracic echocardiograph (allowing estimation of systolic pulmonary artery pressure) within three months of their initial assessment, the median survival for those with a systolic pulmonary artery pressure of ≥0 mm Hg to ≤35 mm Hg was 4.8 years, for those with a systolic pulmonary artery pressure >35 mm Hg to ≤50 mm Hg was 4.1 years, and for those with a systolic pulmonary artery pressure >50 mm Hg was 0.7 years.83 Patients with systemic sclerosis are particularly at increased risk of pulmonary hypertension (prevalence of 8-12%) and are particularly likely to develop pulmonary hypertension.84 85 86 British Thoracic Society guidelines advocate annual lung function testing for patients with systemic sclerosis, with echocardiography in patients who have a falling transfer factor or if the transfer factor is less than 50% of the predicted value.1 Further assessment and management of pulmonary hypertension is a complex topic that is well summarised elsewhere.87 88 89 Patients with interstitial lung disease who have pulmonary hypertension that is judged to be contributing to symptoms and is disproportionate to the extent of their lung disease, or is severe (systolic pulmonary artery pressure >50 mm Hg) should be considered for referral to the regional specialist pulmonary hypertension centre for assessment and recruitment to high quality trials.
Is lung biopsy necessary?
If key radiological features are present (for example, subpleural, basilar, and reticular abnormalities with traction bronchiectasis and honeycombing in a patient with clinical features consistent with idiopathic pulmonary fibrosis) the diagnostic accuracy of computed tomography approaches 90-100% and most clinicians would not opt for surgical lung biopsy.90 Such biopsies, which are done under general anaesthetic, are most helpful when clinical and radiological data result in an uncertain diagnosis or when patients are not thought to have idiopathic pulmonary fibrosis.61
The site of surgical biopsy is best informed by high resolution computed tomography images of the lung. Video assisted thoracoscopic surgery is now used routinely rather than open thoracotomy. Since a range of interstitial patterns may be seen within the same lung, best practice guidelines encourage sampling from at least two lobes1 91 and the biopsy with the pathological pattern associated with the poorest prognosis should dictate treatment and estimates of disease prognosis. The risks of video assisted thoracoscopic surgery include persistent air leak (approximately 10-15%), which results in delayed re-expansion of the lung following the procedure, and perioperative bleeding and sepsis. One retrospective cohort study in the United States of 68 patients with interstitial lung disease who had video assisted thoracoscopic surgery reported a mortality rate of 4% (95% confidence interval 1% to 12%) that was associated with the severity of disease, which underlies the importance of careful patient selection for this procedure.92
Fibreoptic bronchoscopy, which is typically performed as a day case procedure under local anaesthetic, allows visualisation of proximal airways and sampling of more distal lung using transbronchial biopsy, and bronchoalveolar lavage to exclude infectious causes.93 It is particularly useful in sarcoidosis, with diagnostic yields of up to 90%.94 Sampling of lymph nodes adjacent to the airways via endobronchial ultrasound guided aspiration is now also possible.95 96 Bronchoscopy with lavage and transbronchial biopsies may be diagnostic, or increase diagnostic confidence, in other conditions such as cancer, acute pulmonary eosinophilia, cryptogenic organising pneumonia, subacute hypersensitivity pneumonitis, and some rarer forms of interstitial lung disease such as alveolar proteinosis.1 However, in most other form of interstitial lung disease bronchoscopy is not particularly helpful since the biopsies obtained are too small to make a confident histopathological diagnosis.
Although lung biopsy will reveal a pathological pattern the final clinical diagnosis still requires integration of clinical, physiological, and radiological data, ideally by a multi-disciplinary team.
Who should manage these patients?
Interstitial lung disease clinics that provide multidisciplinary care are now common in the UK and elsewhere. This team approach improves inter-observer agreement, diagnostic confidence,97 98 and perhaps outcome,99 and this model of care is advocated in recent guidelines.1 2 Key individuals in the team typically include chest physicians, pathologists, radiologists, pulmonary physiologists, thoracic surgeons, rheumatologists, palliative care specialists, specialist nurses, and physiotherapists. For example, in an American study of 58 patients with suspected interstitial lung disease, information was provided sequentially to a group of three clinicians, two radiologists, and two pathologists, with each recording their diagnostic impression and diagnostic confidence at each stage.97 Interobserver agreement improved from beginning to end and after presentation of histopathological information, radiologists changed their diagnostic impressions more than clinicians. Of the 30 patients who had idiopathic pulmonary fibrosis, clinicians identified 75% and radiologists identified 48% before presentation of the histopathological data.97 Multidisciplinary interaction may prevent unnecessary surgical lung biopsies (when the clinical and radiological impression is consistent with a confident diagnosis of idiopathic pulmonary fibrosis) and identify those in whom a biopsy may be helpful. Links with regional transplant and pulmonary hypertension centres are important and, ideally, patients should have the opportunity to participate in well designed clinical studies where available.100 Regardless of the model of care available, the patient’s general practitioner and local chest physician remain integral to providing continuity and quality of ongoing care.
Additional educational resources
Resources for healthcare professionals
British Thoracic Society (www.brit-thoracic.org.uk)—Recent guidelines on the management of interstitial lung diseases, with useful patient information sheets and management algorithms
American Thoracic Society (www.thoracic.org/statements/index.php)—Includes consensus statements on classification of idiopathic forms of interstitial lung disease and disease-specific guidelines on sarcoidosis and idiopathic pulmonary fibrosis
Pneumotox Online (www.pneumotox.com)—Excellent site providing information about drug induced lung diseases
Ongoing clinical trials (www.clinicaltrials.gov)—Involvement in clinical trials will improve our understanding and management of interstitial lung diseases. This website provides a useful up to date summary of ongoing trials
Resources for patients
British Lung Foundation (www.lunguk.org)—Useful site with links to patient support groups (Breathe Easy groups)
Pulmonary Fibrosis UK (www.pulmonaryfibrosis.org.uk)—Excellent site including video about pulmonary fibrosis
Pulmonary Fibrosis Foundation (www.pulmonaryfibrosis.org)—Support group for patients with pulmonary fibrosis in United States
Persistent crackles on chest auscultation may be due to interstitial lung disease and should prompt a detailed history and some first line tests including a chest radiograph
Some commonly prescribed drugs, environmental dusts, smoking, and underlying connective tissue diseases can cause interstitial lung disease
Detailed pulmonary function tests and high resolution computed tomography are usually done in specialist care and some patients may need a lung biopsy
Refer to a chest physician with an interest in interstitial lung diseases for diagnosis and management by a multi-disciplinary team, and possible participation in a clinical trial.
Cite this as: BMJ 2010;340:c2843
Contributors: OJD conceived the idea for the article and all authors contributed to the first draft and subsequent revisions. OJD is guarantor.
Funding: No funding was received for this work.
Competing interests: All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare that they have (1) no support from companies for the submitted work; (2) no relationship with any companies that might have an interest in the submitted work in the previous 3 years; (3) no spouses, partners, or children with financial relationships that may be relevant to the submitted work; and (4) no non-financial interests that may be relevant to the submitted work.
Provenance and peer review: Not commissioned, externally peer reviewed.