Pulmonary

Sarcoidosis

Daniel A. Culver

Print this Content


 

Definition and causes

Sarcoidosis is a systemic disorder characterized by non-necrotizing granulomatous inflammation, with varying degrees of concomitant fibrosis. It predominantly affects the lungs, eyes, and skin. A transmissible cause has long been suspected based on epidemiologic studies, transmission via organ transplantation, and experimental data. 1,2 However, the causative agent(s) remain(s) unknown at present (Box 1).

Box 1: Suspected Causes of Sarcoidosis
Infectious
Mycobacteria
  1. Tuberculous
  2. Nontuberculous*
  3. Cell-wall deficient (L-forms)*
Bacteria
  1. Corynebacterium spp.
  2. Propionibacterium acnes*
  3. Tropheryma whippleii
  4. Others
Fungi
  1. Cryptococcus spp.
  2. Endemic fungi
Viruses
  1. Cytomegalovirus
  2. Epstein-Barr virus
  3. Herpes simplex virus
  4. Others
Noninfectious
Dusts
  1. Clay
  2. Pine
  3. Pollen
  4. Talc
  5. Mixed*
Metals
  1. Aluminum
  2. Beryllium
  3. Zirconium
Note: Many of these are no longer considered relevant suspects as triggers for sarcoidosis.

* These organisms have been the focus of most recent studies, but no single agent is confi rmed. It is very possible that several disparate agents induce similar reactions leading to sarcoidosis.

† Beryllium causes a histologically identical pulmonary reaction, but berylliosis can be differentiated from sarcoidosis by exposure history and lymphocyte proliferation testing.

Back to Top

Prevalence and risk factors

Epidemiologic characterization of sarcoidosis is problematic due to variability in case definitions, ascertainment bias, and lack of precise diagnostic methods. Population-based chest x-ray screening programs in Scandinavia and the United Kingdom have suggested that there are a sizeable number of asymptomatic patients whose disease never becomes overt. 3,4

Demographic factors, including race, ethnicity, age, and gender, markedly influence incidence. In the United States, the incidence in African Americans is 3.8-fold higher than in whites, conferring an overall lifetime risk of 2.4% versus 0.85%. 5 Most patients present between the ages of 20 and 40 years, although a number of studies have suggested a second peak after 50 years, especially in women. 6,7

Familial clustering of sarcoidosis was first recognized 80 years ago. In A Case-Control Etiologic Study of Sarcoidosis (ACCESS), a recently completed descriptive U.S. study, the familial relative risk of sarcoidosis was estimated using 736 patients, who were matched for age, gender, and geographic location with 10,862 first-degree and 17,047 second-degree relatives. 8 The relative risk for development of disease in a first- or second-degree relative was 4.7 after adjustments for age, gender, relative class, and shared environment.

Disease presentation and natural history are also influenced by epidemiologic factors. White patients tend to present more often without symptoms, whereas severe multisystem disease occurs more often in blacks. 9,10 Black race has been described as conferring added mortality; however, data from population-based settings have suggested that at least part of the differential outcome may be related more to access to medical care than to inherent differences in disease behavior. 11–13 The ACCESS study systematically characterized 736 newly diagnosed patients at ten U.S. centers. 14 Organ involvement was significantly influenced by demographic variables, including age, gender, and race (Fig. 1). 15

Back to Top

Pathophysiology and natural history

Sarcoid inflammation is characterized by non-necrotizing granulomas (Figs. 2 and 3). The granuloma is a compact mass of cells that walls off foreign antigens, typically microbes. Epithelioid histiocytes, together with a few multinucleated giant cells, compose the core, surrounded by an outer rim of T lymphocytes. The lymphocyte population is oligoclonal, with restricted T cell receptor repertoires, consistent with an antigen-driven process. Inflammation in sarcoidosis is dependent on persistent stimulation by CD4+ T cells. 16 Thus, polymorphisms of the major histocompatibility complex and the T cell receptor that modulate affinity for the antigen may be responsible for development or course of the disease. 17 Granulomatous inflammation that is widespread or at critical locations (e.g. A-V node) can lead to functional organ impairment. However, it is not uncommon for organ involvement to be clinically silent.

The search for a causative agent has spanned more than a century. It is still unknown whether a single agent triggers the disease or if sarcoidosis represents a stereotyped immune response to diverse etiologies. A large number of infectious and noninfectious agents have been proposed (see Box 1). Recent attention has focused most prominently on Mycobacteria spp. 18,19 and Propionibacterium acnes, 20 but the evidence to date has been inconclusive and sometimes contradictory.

The natural history of disease is variable. Spontaneous resolution occurs within 5 years in approximately two thirds of patients 10 ; over a 2-year follow-up period in ACCESS, 80% demonstrated either improvement or stability, with no requirement for treatment. 15 Features believed to predict poor prognosis vary among studies (Box 2); of all the putative risk factors, no study to date has comprehensively ascertained which are independently associated with disease chronicity or progression.

Box 2: Proposed Clinical Prognostic Factors for the Outcome of Sarcoidosis
Favorable
  • White race
  • Löfgren's syndrome
  • Scadding stage I chest radiograph
Unfavorable
  • Black race
  • Age >40 years
  • Organomegaly
  • Lupus pernio
  • Cardiac disease
  • Nephrocalcinosis
  • Sinus involvement
  • Bone involvement

Staging of pulmonary disease based on the chest x-ray (Fig. 4) allows a general prediction of outcome. For chronic sarcoidosis, there is a striking degree of heterogeneity between patients: Some have persistent inflammation, and others shift to a fibrotic phenotype. This shift is not seen in all patients, and the onset and pace of the fibrosis are likewise highly variable. Large-scale epidemiologic surveys have suggested that between 5.4% and 10% of cases are fibrotic at presentation alone. In the United States, pulmonary fibrosis is the leading cause of death in sarcoidosis, which carries a mortality rate of 1% to 5%. Other complications of pulmonary sarcoidosis include mycetomas, pleural effusions, bronchiectasis, pulmonary hypertension, endobronchial stenosis, and, rarely, bullous lung disease. Depression, chronic pain syndromes resembling fibromyalgia, and sleep apnea are prevalent in sarcoidosis patients.

Back to Top

Signs and symptoms

Sarcoidosis can affect any organ, although the lungs are involved in up to 95% of patients (Fig. 5). Other organ systems commonly affected include the skin, eyes, and lymphoreticular system. Estimates of organ involvement are confounded by the method of discovery, referral bias, and the sensitivity of diagnostic modalities. Why different organs are differentially affected in various patients is unclear. Manifesting symptoms are organ dependent, but the most common ones include fatigue, arthralgias, diffuse pain syndromes, cough, dyspnea, wheezing, chest discomfort, rash, photophobia, scleritis, decreased visual acuity, weight loss, and fever.

In the lungs, important differential diagnostic possibilities include granulomatous infections, idiopathic interstitial pneumonias, hypersensitivity pneumonitis, and asthma. Cutaneous manifestations are protean. 21 Ultimately, one quarter of patients develop at least one dermatologic feature. Erythema nodosum is associated with acute onset of disease and confers a good prognosis. Lupus pernio is a chronic plaquelike induration of the face, usually appearing with violaceous discoloration of the cheeks, lips, nose, and ears (Fig. 6). It can erode into cartilage or bone, causing permanent disfigurement. Lupus pernio generally portends chronic, multisystem sarcoidosis, is more common in older African-American and West Indian women, and is notoriously difficult to treat. Other skin lesions include plaques, maculopapular eruptions, hypo- or hyperpigmented patches, subcutaneous nodules, and alopecia (Fig. 7).

Any ocular structure can be affected. In acute disease, anterior uveitis (iritis) causes photophobia, conjunctivitis, tearing, pain, and blurred vision (Fig. 8). It is usually easily treated with topical therapy. Chronic inflammation is typically more insidious and can affect the anterior and posterior structures. Retinal involvement, with characteristic candle-wax dripping exudates (Fig. 9), multifocal choroiditis (Fig. 10), macular edema, and neovascularization may be asymptomatic but always requires treatment.

Other commonly affected organs include the liver and spleen, with elevated liver enzymes or radiologic organomegaly in approximately 50% of patients. 22,23 However, these rarely cause symptoms or functional impairment. Uncommon manifestations include neurologic, cardiac, bone marrow, renal, exocrine gland, sinonasal, and bone disease. Clinically relevant presence of these uncommon manifestations is often associated with chronic or progressive sarcoidosis. When they are discovered as incidental findings on imaging studies or biopsies, they do not usually portend significant organ dysfunction.

Back to Top

Diagnosis

Sarcoidosis is a diagnosis of exclusion. In an appropriate clinical setting, the presence of non-necrotizing granulomas, without evidence for infection, is the usual criterion to suggest the diagnosis. Sarcoidosis mimics and nonspecific granulomatous reactions should be excluded by a careful examination and by medical, occupational, and medication histories. Box 3 lists pulmonary diseases that are often confused with sarcoidosis. In practice, the disease is most often diagnosed by biopsy of accessible tissues, usually skin, lungs, or peripheral lymph nodes. We recommend a histologic diagnosis before commencing any treatment, in the presence of any atypical clinical features, or when there is suspicion that an alternative diagnosis (e.g., lymphoma) could be present. When sarcoidosis is suspected in extrapulmonary organs, chest imaging or random transbronchial biopsy may be useful to help confirm the diagnosis. 24

Box 3: Pulmonary Mimics of Sarcoidosis
Granulomatous diseases
Infections (fungal, mycobacterial, others)
Chronic beryllium disease
Hypersensitivity pneumonitis
Other exposures (methotrexate, metals)
Rheumatologic syndromes (Wegener's granulomatosis, Churg-Strauss syndrome)
Lymphoma
Tumor-associated granulomas
Other parenchymal lung diseases (e.g. pulmonary fibrosis)
Asthma

Bronchoscopy is a reliable, minimally invasive technique to diagnose the disease. Combination of transbronchial biopsy with endobronchial (mucosal) biopsy or transbronchial needle aspiration of enlarged lymph nodes increases the sensitivity of the technique to as much as 91%. Endobronchial ultrasonography has been used more recently to further improve diagnostic yield. Samples should be analyzed for infectious agents by appropriate stains, including stains for mycobacteria and fungus, as well as by culture.

The Scadding scale, a descriptive schema that allows a loose characterization of the chance for disease remission, is used widely to describe chest x-ray findings. 25 Use of the scale entails several caveats: The stages of the Scadding scale do not represent sequential or temporal disease states, the predictive ability of the scale allows an approximation of outcome only, there are no data correlating stages of the scale with treatment response, the scale was developed more than 50 years ago from only 136 British patients, and the usefulness of the scale has not been validated in comparison with chest computed tomography (CT). The most common radiologic findings include bilateral hilar and mediastinal lymphadenopathy and bilateral nodular or reticulonodular infiltrates. Features that should prompt consideration of an alternative diagnosis include pleural effusion, unilateral abnormalities, and the presence of calcification in the lymph nodes. Chest CT usually shows typical micronodular infiltrates distributed in a bronchovascular pattern, and often predominating in the mid to upper lung zones.

Other diagnostic testing may be appropriate as guided by the initial symptoms and examination findings. Routine testing for elevated liver enzymes, hypercalcemia, renal dysfunction, ophthalmologic involvement, and electrocardiogram (ECG) abnormalities is standard at baseline. Other common tests to assess for organ involvement demonstrate various sensitivity and specificity, depending on the population studied.

Magnetic resonance imaging (MRI) with gadolinium or gallium is useful for suggesting the diagnosis but nonspecific. 26 However, leptomeningeal enhancement after gadolinium has been shown to predict reversibility of sarcoid lesions. 27 Lumbar puncture should be performed in the appropriate clinical context to exclude mycobacterial or fungal infections. CSF analysis might reveal lymphocytosis, elevated protein, oligoclonal bands, and elevated angiotensin converting enzyme levels. 28 Thallium or sestamibi scintigraphy is useful to identify areas of active or inactive myocardial involvement. 29 It is more sensitive than gallium scan. In the presence of unremarkable coronary angiograms, abnormal thallium scans in patients with known sarcoidosis are highly suggestive of myocardial involvement. 30 However, asymptomatic patients with abnormal thallium scans are unlikely to have clinically bothersome long-term disease. 31 Newer imaging techniques, such as fluorodeoxyglucose positron emission tomography (FDG-PET) and gadolinium-enhanced MRI are promising modalities for diagnosis and for monitoring treatment response.

ECG, Holter monitoring, and event monitoring can identify patients with dysrhythmias, and a baseline ECG is recommended for all patients with newly diagnosed sarcoidosis. Endomyocardial biopsy is rarely useful, due to poor sensitivity. 32 Therefore, a nondiagnostic biopsy should not preclude treatment when clinical suspicion is high.

The initial evaluation should include an assessment accounting for the common manifestations. Other assays of disease activity have been the subject of intense investigation, including serum angiotensin-converting enzyme level, gallium-67 scan, bronchoalveolar lavage fluid characteristics (CD4/CD8 ratio, total lymphocyte count), andnumerous other biomarkers; unfortunately, none has consistently predicted disease course in multiple populations. Recent data have demonstrated that the presence of certain human leukocyte antigen (HLA) haplotypes (e.g., HLA-DR17 and HLA-DQB1) confer good prognosis in certain European populations. 33 However, the usefulness of HLA typing in other populations has not been confirmed. For now, chest x-ray stage, demographic features, clinical involvement, and serial pulmonary function testing remain the best indicators in assessing prognosis and therapeutic decisions.

Back to Top

Treatment

Several situations warrant aggressive systemic treatment. The most commonly accepted indications include active neurologic or cardiac involvement of any degree (except Bell's palsy), severe hypercalcemia, ocular disease refractory to topical therapy, lupus pernio, significant hepatic involvement, symptomatic splenic disease, and bulky lymphadenopathy resulting in symptomatic compression of surrounding structures. Other scenarios that require case-by-case decisions include sarcoid myositis; sinonasal, gastrointestinal, exocrine gland, or skin involvement; granulomatous nephritis; and bone disease. Constitutional symptoms, such as fatigue and achiness, are clinically bothersome for a large percentage of patients; they are best approached with nonsteroidal anti-inflammatory medications and careful assessment for the presence of depression, reserving the toxicities of steroids for patients with potentially life-threatening progressing disease. Neuropathic symptoms, often manifesting as paresthesias or pain syndromes, may be due to small-fiber neuropathy that can be present in some patients.

Several principles guide treatment strategy: Corticosteroids prevent granuloma formation in most patients, and they reverse immunologic derangements. Granuloma formation proceeds at varying rates between patients, necessitating individualized treatment strategies. A threshold dose of medication is usually required for effectiveness. Different tissues respond differently to various medications. 16 For example, antimalarial medications are effective for hypercalcemia and skin disease but have little effect on pulmonary disease.

Decisions to treat pulmonary sarcoidosis should be framed in the context of the lack of evidence that treatment significantly changes the natural history of the disease. The likelihood of spontaneous remission, degree of physiologic impairment, magnitude of symptoms and chronicity of disease must be considered. For patients with acute (<2 years) disease, systemic treatment is generally withheld unless there are significant symptoms, moderate to severe physiologic impairment, or rapid progression and chest imaging suggesting the presence of reversible disease (e.g., infiltrates that are not wholly fibrosis). For patients with chronic disease (≥2 years), the chances of spontaneous remission are low. The general goal of therapy in this group is to define the lowest dose of medication needed to maintain stability. Significant pulmonary symptoms might also be due to pulmonary vascular involvement, a diagnosis that should be considered in patients whose symptoms are disproportionate to the degree of parenchymal involvement.

Corticosteroids are the current first-line agents for treatment because clinical experience is greatest with them, they generally have acceptable (and reversible) side-effect profiles, and the time to onset of effect is quicker than for most alternatives. For pulmonary disease, initial doses of 30 to 40 mg/day of prednisone are probably sufficient. Although some clinicians have advocated higher doses (1 mg/kg per day), the ratio between benefit and toxicity of such doses is usually unfavorable. Improvements in symptoms, chest radiographs, and lung function tests are typical; failure to respond within 3 to 4 months suggests the presence of steroid-refractory disease. Although some clinicians have advocated every-other-day dosing regimens, this approach might not be sufficient to control active granuloma formation. In general, I suggest that systemic treatment of pulmonary disease should entail at least weeks to months of therapy.

Inhaled corticosteroids may be useful for symptoms of bronchospasm and cough in conjunction with bronchodilators. Numerous trials have assessed their usefulness in treating parenchymal disease, with the net result suggesting that they have very modest to no benefits. However, for patients with mild disease, a trial of inhaled corticosteroids is generally well tolerated and may be helpful.

Steroid-sparing agents are used when steroids are ineffective or cause unacceptable side effects. Methotrexate, a folic acid analogue with anti-inflammatory and antiproliferative properties, is generally used as a second-line agent, but it has not been subjected to rigorous controlled studies. In small trials, it has been effective for controlling disease and reducing corticosteroid dose for acute and chronic disease. 34,35 Doses of 10 to 20 mg per week are typically used, and treatment response can require 3 to 6 months of therapy. The purine analogue azathioprine is also effective as a steroid-sparing agent, but its use is limited due to its potential for substantial toxicity. 36 One series suggested that leflunomide might be useful in conjunction with methotrexate. 37 For severe or refractory disease, agents with greater potential for toxicity may be necessary, including cyclophosphamide and chlorambucil. Antimalarial agents are often efficacious for treating dermatologic disease, sinonasal sarcoid, osseous disease, and hypercalcemia.

Newer approaches include antagonists of tumor necrosis factor a (TNF-α), including thalidomide and monoclonal antibodies such as infliximab or adalimumab. 38,39 The largest randomized controlled double-blind study of therapy in sarcoidosis included 138 patients with pulmonary disease of more than 1 year's duration. 40 Use of infliximab resulted in a modest net gain of 2.5% from baseline in FVC at 24 weeks, compared with no change for placebo. The clinical importance of this finding is questionable; however, it is evident that a portion of patients treated with this medication experience substantial benefits after failing other medical approaches. A major challenge for treatment decision-making in sarcoidosis is the lack of head-to-head studies between agents. For now, toxicity profile, clinician comfort, and experience are likely to be important factors guiding therapeutic choices. As additional biologic agents become available for use in other diseases, their application to sarcoidosis will be best elucidated by development of controlled trials. Due to the heterogeneity of the disease, designing these trials will be challenging.

Summary

  • Sarcoidosis is a systemic granulomatous disease, of unknown etiology, which most often affects the lungs, skin, and eyes.
  • In the United States, sarcoidosis is more prevalent and more severe in blacks than in whites. The prevalence has been estimated at 35/100,000 in blacks and 11/100,000 in whites.
  • Approximately two thirds of patients experience resolution of their disease without specific therapy.
  • Treatment of sarcoidosis relies on a combination of immunosuppressive medications and realistic appraisal of therapeutic goals.

Back to Top

References

  1. Donor-acquired sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis. 19: 2002; 18-24.
  2. Epidemiology of sarcoidosis in the Isle of Man—2: Evidence for space-time clustering. Thorax. 42: 1987; 427-430.
  3. A worldwide review of sarcoidosis. Ann N Y Acad Sci. 278: 1976; 321-334.
  4. Epidemiology of sarcoidosis: New frontiers to explore. Curr Opin Pulm Med. 8: 2002; 424-428.
  5. Racial differences in sarcoidosis incidence: A 5-year study in a health maintenance organization. Am J Epidemiol. 145: 1997; 234-241.
  6. The epidemiology of sarcoidosis in Rochester, Minnesota: A population-based study of incidence and survival. Am J Epidemiol. 123: 1986; 840-845.
  7. Sarcoidosis: Epidemiology and prognosis. A 15-year European study. Am Rev Respir Dis. 130: 1984; 29-32.
  8. Familial aggregation of sarcoidosis. A Case-Control Etiologic Study of Sarcoidosis (ACCESS). Am J Respir Crit Care Med. 164: 2001; 2085-2091.
  9. Sarcoidosis in caucasians, blacks and Asians in London. Br J Dis Chest. 79: 1985; 27-36.
  10. ATS/ERS/WASOG statement on sarcoidosis. American Thoracic Society/European Respiratory Society/World Association of Sarcoidosis and other Granulomatous Disorders. Sarcoidosis Vasc Diffuse Lung Dis. 16: 1999; 149-173.
  11. Sarcoidosis mortality in the United States 1979– 1991: An analysis of multiple-cause mortality data. Am J Med. 100: 1996; 423-427.
  12. Hospital, age, racial, occupational, geographical, clinical and survivorship characteristics in the epidemiology of sarcoidosis. Am J Epidemiol. 94: 1971; 222-230.
  13. Mortality of intrathoracic sarcoidosis in referral vs population-based settings: Influence of stage, ethnicity, and corticosteroid therapy. Chest. 121: 2002; 32-39.
  14. Clinical characteristics of patients in a case control study of sarcoidosis. Am J Respir Crit Care Med. 164: 2001; 1885-1889.
  15. Two year prognosis of sarcoidosis: The ACCESS experience. Sarcoidosis Vasc Diffuse Lung Dis. 20: 2003; 204-211.
  16. Cells and cytokines involved in the pathogenesis of sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis. 16: 1999; 24-31.
  17. HLA-DR predicts the prognosis in Scandinavian patients with pulmonary sarcoidosis. Am J Respir Crit Care Med. 156: 1997; 1601-1605.
  18. Mycobacterial antigens may be important in sarcoidosis pathogenesis. Curr Opin Pulm Med. 12: 2006; 359-363.
  19. Mycobacterial catalase-peroxidase is a tissue antigen and target of the adaptive immune response in systemic sarcoidosis. J Exp Med. 201: 2005; 755-767.
  20. Quantitative PCR of mycobacterial and propionibacterial DNA in lymph nodes of Japanese patients with sarcoidosis. Lancet. 354: 1999; 120-123.
  21. Cutaneous sarcoidosis: clinical features and management. Chest. 61: 1972; 320-325.
  22. Hepatic sarcoidosis. Clinicopathologic features in 100 patients. Am J Surg Pathol. 17: 1993; 1272-1280.
  23. Splenomegaly in sarcoidosis. Arch Intern Med. 140: 1980; 35-37.
  24. Transbronchial lung biopsy in the diagnosis of suspected ocular sarcoidosis. Arch Ophthalmol. 111: 1993; 642-644.
  25. Prognosis of intrathoracic sarcoidosis in England. A review of 136 cases after five years' observation. BMJ. 5261: 1961; 1165-1172.
  26. Sarcoidosis and its neurological manifestations. Arch Neurol. 42: 1985; 909-917.
  27. Central nervous system sarcoidosis: Follow-up at MR imaging during steroid therapy. Radiology. 214: 2000; 411-420.
  28. Diagnosis and management of neurological sarcoidosis. Arch Intern Med. 157: 1997; 1864-1868.
  29. Cardiac involvement in sarcoidosis. Sem Respir Crit Care Med. 23: 2002; 513-527.
  30. Reversibility by dipyridamole of thallium-201 myocardial scan defects in patients with sarcoidosis. Am J Med. 85: 1988; 189-193.
  31. Do thallium myocardial perfusion scan abnormalities predict survival in sarcoid patients without cardiac symptoms?. Angiology. 41: 1990; 573-576.
  32. Histologic diagnostic rate of cardiac sarcoidosis: Evaluation of endomyocardial biopsies. Am Heart J. 138: 1999; 299-302.
  33. HLA-DQB1* 0201: A marker for good prognosis in British and Dutch patients with sarcoidosis. Am J Respir Cell Mol Biol. 27: 2002; 406-412.
  34. Methotrexate for sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis. 15: 1998; 147-149.
  35. Methotrexate is steroid sparing in acute sarcoidosis: Results of a double blind, randomized trial. Sarcoidosis Vasc Diffuse Lung Dis. 17: 2000; 60-66.
  36. Azathioprine treatment of chronic pulmonary sarcoidosis. Sarcoidosis. 2: 1985; 107-113.
  37. Leflunomide for chronic sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis. 21: 2004; 43-48.
  38. Thalidomide for chronic sarcoidosis. Chest. 122: 2002; 227-232.
  39. Infliximab for refractory sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis. 18: 2001; 70-74.
  40. Infliximab therapy in patients with chronic sarcoidosis and pulmonary involvement. Am J Respir Crit Care Med. 174: 2006; 795-802.

Back to Top

Suggested Readings

  • Infliximab therapy in patients with chronic sarcoidosis and pulmonary involvement. Am J Respir Crit Care Med. 174: 2006; 795-802.
  • Clinical characteristics of patients in a case control study of sarcoidosis. Am J Respir Crit Care Med. 164: 2001; 1885-1889.
  • Sarcoidosis: Epidemiology and prognosis. A 15-year European study. Am Rev Respir Dis. 130: 1984; 29-32.
  • ATS/ERS/WASOG statement on sarcoidosis. American Thoracic Society/European Respiratory Society/World Association of Sarcoidosis and other Granulomatous Disorders. Sarcoidosis Vasc Diffuse Lung Dis. 16: 1999; 149-173.
  • Two year prognosis of sarcoidosis: The ACCESS experience. Sarcoidosis Vasc Diffuse Lung Dis. 20: 2003; 204-211.
  • Treatment of sarcoidosis—from a basic science point of view. J Intern Med. 253: 2003; 31-40.
  • Familial aggregation of sarcoidosis. A Case-Control Etiologic Study of Sarcoidosis (ACCESS). Am J Respir Crit Care Med. 164: 2001; 2085-2091.
  • Racial differences in sarcoidosis incidence: A 5-year study in a health maintenance organization. Am J Epidemiol. 145: 1997; 234-241.
  • Mycobacterial catalase-peroxidase is a tissue antigen and target of the adaptive immune response in systemic sarcoidosis. J Exp Med. 201: 2005; 755-767.
  • The cytokine network in sarcoidosis and its clinical relevance. J Intern Med. 253: 2003; 18-30.