Introduction
Sarcoidosis has been defined as a multisystem granulomatous disorder of unknown cause. Sarcoidosis affects the lung in over 90% of cases but can affect any part of the body. For many patients the disease is self-limited and resolves within 2 to 5 years. However, a chronic form of the disease can lead to significant morbidity and some mortality. Glucocorticoids are usually effective for treating the disease; however, because long-term treatment may be required, steroid-sparing agents are often used.
Epidemiology
Sarcoidosis is a worldwide disease, but it is more common in some parts of the world and within certain ethnic groups. Figure 66-1 summarizes the reported rates of sarcoidosis around the world. In the United States the disease appears to be more common in the southeastern part of the country.
Sarcoidosis is rare before adulthood. Pediatric sarcoidosis is usually diagnosed in patients older than 10 years, with a peak in the 13- to 15-year-old age group. When it does present in childhood, there is a different clinical phenotype with mostly eye (uveitis), skin, and joint involvement, although lung involvement is commonly identified when lung imaging studies are performed. The clinical appearance of juvenile-onset sarcoidosis resembles that of the adult type of the disease. As in adults, the most common finding in childhood sarcoidosis is a chest radiograph with abnormal findings (>90% at onset, with stage I in two thirds of the subjects). Although previously felt to be a disease mostly of young adults, it has been increasingly diagnosed in older patients. In the United States, half of the patients are older than age 40. There appear to be two peaks of age of onset, 20 to 29 years and 60 to 65 years. Sarcoidosis is believed to be more common in women compared to men at a ratio of less than 2 : 1.
Etiology
Although sarcoidosis was first described as a distinct clinical entity more than 140 years ago, its cause remains unknown. It is rational to expect that the immunopathogenesis of sarcoidosis is similar to that of other granulomatous diseases. That is, some antigen is encountered and phagocytized by an antigen-presenting cell. The antigen-presenting cell then processes the antigen and presents it, via a human leukocyte antigen (HLA) class II molecule, to a restrictive set of T-cell receptors on a T lymphocyte, usually of the CD4 + class. This interaction results in a polarization of the T lymphocytes to a type 1 T-helper phenotype, which is followed by monocyte recruitment, T-cell proliferation, and differentiation leading to the development of the sarcoidosis granuloma. In this process a myriad of cytokines and chemokines is released, but the relative importance of most of them is unclear.
Potential Antigens
Because many infections induce a granulomatous response, infectious pathogens have been implicated as potential causes of sarcoidosis. However, because sarcoidosis responds to immunosuppressive therapy, it is unlikely that sarcoidosis represents an invasive infection; it is nonetheless possible that a host response to an infectious antigen, even if the infecting organism is dead, may induce a granulomatous response. For example, propionibacteria, including Propionibacterium acnes, the common acne bacterium, have also been implicated as a cause of sarcoidosis. Propionibacterial DNA has been detected within granulomatous lymph nodes of Japanese sarcoidosis patients. These findings have not been corroborated, however, in subsequent studies outside of Japan.
Mycobacteria have been implicated as a potential cause of sarcoidosis. More than two dozen studies have assessed the presence of mycobacterial DNA and RNA in sarcoidosis tissues. A meta-analysis has suggested that 26% of sarcoidosis tissues have evidence of mycobacterial DNA that is 9- to 19-fold greater than in nonsarcoidosis control tissues. However, quantitative polymerase chain reaction comparing lung tissue from sarcoidosis patients, controls, and those with tuberculosis in an area in which tuberculosis is endemic (China) found that the copies of mycobacteria DNA identified in the tissue from sarcoidosis patients were at levels similar to those of control tissues and 1000-fold less than in the tissues of patients infected with Mycobacterium tuberculosis . It is therefore not clear that mycobacterial DNA is consistently higher in the tissues of patients with sarcoidosis.
The protein mycobacterial catalase-peroxidase has been identified in human sarcoidosis tissue and has been shown to induce an immunoglobulin G antibody response in 12 of 25 sarcoidosis patients compared to purified protein derivative–negative controls. Subsequent studies have demonstrated a T-lymphocyte response to mycobacterial catalase-peroxidase both in the peripheral blood and bronchoalveolar lavage (BAL) of sarcoidosis patients.
Environmental and occupational exposures may also be associated with sarcoidosis. Combustible wood product exposure may be associated with sarcoidosis, because the incidence and prevalence of sarcoidosis in firefighters far exceeds that of emergency medical technicians who travel to the same fires. In addition, a relationship between wood stove/fireplace exposure and the development of sarcoidosis has been documented. An increased incidence of a sarcoidosis-like granulomatous lung process was found in firefighters and first responders involved in the World Trade Center disaster who were exposed to large amounts of debris during a prolonged period. It is not clear, however, whether patients exposed to World Trade Center debris exhibit the same clinical features as patients with non–World Trade Center–related sarcoidosis.
An increased incidence of sarcoidosis has been found in individuals exposed to metals, in particular, titanium, metalworking, metal machining, and photocopier toner (that contains silicates, iron, and copper). Analysis of lung biopsy specimens from pulmonary sarcoidosis patients has revealed various metals, including silicates, aluminum, and titanium. Sarcoidosis has also been associated with several additional occupations and exposures, including hairdressers, health care workers, agricultural employment, insecticide use at work, work environments with mold or mildew, industrial organic dust exposure, educators, and workers of suppliers of building materials, hardware, and gardening materials. Many of these exposures may, in and of themselves, cause granulomas. Many of the sarcoidosis cases associated with environmental exposure to respirable material are of isolated pulmonary sarcoidosis. It is not clear whether these are truly cases of sarcoidosis or of “sarcoid-like” conditions (e.g., hypersensitivity pneumonitis).
Genetic Aspects
There is compelling evidence that sarcoidosis is the result of environmental triggers acting upon an immunogenetically susceptible host. The importance of genetics in the development of sarcoidosis is further supported by evidence of familial clustering of the disease.
Because HLA class II molecules and T-cell receptors appear integral to the immunopathogenesis of sarcoidosis, various polymorphisms of these molecules have been examined for their association with sarcoidosis. Indeed, some HLA polymorphisms have been found to be associated with sarcoidosis. Most of these associations appear in specific ethnic groups and have not been universal. In addition, some HLA polymorphisms appear to protect against sarcoidosis, whereas others are associated with certain clinical phenotypes. The presence of HLA-DRB1*03 in a Swedish sarcoidosis cohort was strongly associated with a Löfgren syndrome phenotype and also with disease resolution. Thus the presence or absence of HLA-DRB1*03 could provide clinical prognostic information in the near future.
Although not studied as extensively as HLA molecules, specific arrangements of T-cell receptors have also been associated with sarcoidosis. A restricted use of T-cell receptor α and β chain variable gene segments on T cells in the lungs of sarcoidosis patients has been identified. Non–HLA class II genes have also been shown to be associated with sarcoidosis. HLA class I polymorphisms have been associated with sarcoidosis susceptibility. In a study of German families the butyrophilin-like 2 ( BTNL2 ) immunoregulatory gene explained 23% of the sarcoidosis risk in that population. Other genome-wide approaches have implicated regions associated with sarcoidosis in chromosome 5 in an African American population and mutations in the annexin1 gene in a German population. Gene-wide assays have been used to identify other potential associations. Bioinformatic analyses of global gene expression (“pathway analysis”) identified a dominant network regulated by signal transducer and activator of transcription-1 as the most significantly associated with sarcoidosis in a study of lung and lymph nodes and identified genes associated with type 1 T-helper cells, type 17 T-helper cells, signal transducer and activator of transcription-3, and interleukin-21 in sarcoidosis skin tissue.
Sarcoidosis as the Result of Immune System Exhaustion
Chen and colleagues demonstrated intense expression and wide distribution of serum amyloid A within the sarcoidosis granuloma that surpassed that found in all other granulomatous diseases examined. Serum amyloid A appeared to have originated from macrophages and giant cells within the sarcoid granuloma. These authors postulated that serum amyloid A could bind to matrix proteins and thereby consolidate a poorly soluble protein aggregate to form a nidus for granuloma formation. Serum amyloid A may disrupt the clearance of an antigen within the granuloma that allows for its persistence. Therefore sarcoidosis may relate not only to certain exposures or antigens but also to failure of effective antigen clearance. The granulomatous inflammation in sarcoidosis may result from a prolonged immunogenic response to a persistent antigen resulting in “immune system exhaustion.”
Evidence for a potential role of immune system exhaustion from chronic stimulation as an integral mechanism in the formation of the sarcoid granuloma is beginning to emerge. Invariant natural killer T cells have been found to be depleted in sarcoidosis, and this is postulated to be the result of functional exhaustion. T regulatory (Treg) cells have been found to be increased in the BAL of several granulomatous diseases, including sarcoidosis. Increased Treg cells in the BAL of sarcoidosis is associated with more active disease. Treg cells may induce anergy and mollify the immune response. However, recent data suggest that there is an anergic response in CD4 T cells in sarcoidosis that is not reversed by Treg cell depletion. This suggests that CD4 T-cell exhaustion may be a primary event in sarcoidosis.
Diagnostic Approach
Histologic evidence of granulomatous inflammation alone is inadequate for the diagnosis of sarcoidosis because alternative causes of granulomatous inflammation need to be excluded. Additionally, with rare exceptions (see later), clinical findings without histologic confirmation of granulomatous inflammation are inadequate to secure the diagnosis of sarcoidosis.
Figure 66-2 outlines the approach to the diagnosis of sarcoidosis. This process usually involves a review of clinical information, histologic examination of tissue for the presence of granulomatous inflammation, and exclusion of known causes of granulomatous inflammation.
Clinical Data Collection
One can never be certain of the diagnosis of sarcoidosis. Similar to other diseases, sarcoidosis may be considered as the probable diagnosis if clinical data exceed a certain “threshold” so that the diagnosis is plausible. Table 66-1 outlines clinical findings that are often used to gauge the likelihood of the diagnosis of sarcoidosis. For most patients the clinical information suggests the diagnosis, but a tissue biopsy is usually indicated to enhance the probability of sarcoidosis.
| Supports | Weakens | |
|---|---|---|
| Demographics |
|
|
| Medical history |
|
|
| Laboratory data |
| |
| Radiographic findings |
|
Patients with sarcoidosis may present with no symptoms. This is more common in white than black patients. Therefore sarcoidosis should be considered in asymptomatic patients with hilar adenopathy, mediastinal adenopathy, and/or diffuse parenchymal opacities on lung imaging. An inquiry should be made about a family history of sarcoidosis because the prevalence rate of sarcoidosis is much higher in first-degree relatives of sarcoidosis patients than in the general population. Sarcoidosis is more common in nonsmokers than in smokers in most studies.
In addition, patients should be questioned concerning potential exposures that may cause diseases that may mimic sarcoidosis. Specifically, a history of active tuberculosis, latent tuberculosis infection, and tuberculosis exposure should be obtained. The possibility of beryllium exposure should be explored because chronic beryllium disease can mimic sarcoidosis radiographically and histologically. Chronic beryllium disease has been misdiagnosed as sarcoidosis in up to 40% of cases. Because most patients are unaware of potential exposures to beryllium, it is important to ask about work industries where exposure to beryllium is plausible, including aerospace, nuclear weapons, electronics, jewelry, sporting goods, ceramics, and dental. Furthermore, minimal beryllium exposure may lead to significant disease. Hypersensitivity pneumonitis, which may mimic sarcoidosis, is a granulomatous pulmonary disease resulting from exposure to numerous agents (see Chapter 64 ).
Evidence of Extrapulmonary Involvement
At presentation, 95% of sarcoidosis patients have clinical evidence of pulmonary involvement, and more than 40% have evidence of involvement in the skin, liver, peripheral lymph node, or eye. Therefore evaluation for involvement of these organs should be performed in any patient being evaluated for possible sarcoidosis.
Radiographic Findings
Bilateral hilar adenopathy on chest radiograph suggests the diagnosis of sarcoidosis, especially if the patient has no fever, night sweats, or weight loss. The chest radiograph often demonstrates concomitant enlargement of the right paratracheal lymph nodes ( Fig. 66-3 ). Scadding defined four patterns of the chest radiograph findings: stage 1, with adenopathy alone (see Fig. 66-3 ); stage 2, adenopathy and parenchymal opacities ( Fig. 66-4 ); stage 3, opacities alone ( Fig. 66-5 ); and stage 4, fibrosis ( Fig. 66-6 ).
Findings on chest high-resolution computed tomography (HRCT) may be more specific for the diagnosis of sarcoidosis than those found on chest radiography (see Figs. 66-3 to 66-6 ). Typical HRCT findings that suggest sarcoidosis include parenchymal nodules and opacities that represent conglomerations of these nodules that have a perilymphatic distribution along the bronchovascular bundles as well as in subpleural locations.
Serum Markers for Disease
Angiotensin-converting enzyme (ACE) is produced in the epithelioid cell of the sarcoid granuloma, and serum ACE (SACE) levels may reflect the total body burden of sarcoidosis granulomas. Although SACE has been suggested as a diagnostic test for sarcoidosis, elevated SACE levels alone are inadequately sensitive or specific to diagnose or exclude the disease. In a review of 14 studies encompassing 4195 patients concerning the diagnostic accuracy of SACE for sarcoidosis, the sensitivity was 77% (range, 41% to 100%) and the specificity was 93% (range, 83% to 99%). The likelihood of sarcoidosis increases with higher SACE levels, and SACE levels greater than two times the upper limit of normal are rarely seen in other diseases and not seen in cancer or lymphoma. Polymorphisms of the ACE gene also influence the SACE level and likely alter the diagnostic accuracy of SACE measurements in individual patients. In addition, none of these polymorphisms have been shown to be associated with increased incidence or worsening of the disease. Furthermore, the differences of these polymorphisms between the white and African American populations suggest that the role of the polymorphism of the ACE gene is population dependent, and this explains the reported racial difference in the relationship between SACE levels and the polymorphism.
Other serum markers have been studied in sarcoidosis. These include serum chitotriosidase, which has been shown to be elevated in a small cohort of Italian sarcoidosis patients. Higher serum levels may be associated with a worse prognosis. The soluble interleukin-2 receptor is a marker of T-cell activation and found to be elevated in sarcoidosis patients. In small studies it was an effective measure of disease activity. The value of these biomarkers regarding diagnosis and prognosis needs to be verified by larger multinational trials.
Tissue Examination
With the exception of the rare instances in which the clinical findings are highly specific for sarcoidosis, the diagnosis requires a tissue biopsy (see Fig. 66-2 ). Our approach to the selection of a biopsy site is summarized in Figure 66-7 . It is in the patient’s best interest that the biopsy be minimally invasive and associated with the least morbidity. For these reasons, superficial biopsy sites are preferred compared to visceral organs. Even in patients suspected to have sarcoidosis on the basis of obvious thoracic or abdominal disease, a thorough skin, conjunctival, lacrimal gland, and peripheral lymph node examination should be performed. The patient should be questioned about the presence of scars or tattoos (see Fig. 66-14D ), because sarcoidosis skin nodules have a predilection to form in these areas.
When there is no clinical evidence that a superficial site is involved with sarcoidosis, a biopsy is usually attempted in an organ in which sarcoidosis involvement is suspected. This is very often the lung, because the lung is involved in more than 90% of sarcoidosis patients early in the course of the disease. Bronchoscopy is the most commonly used procedure to obtain tissue from the lung. More invasive techniques such as mediastinoscopy are reserved for cases in which bronchoscopy was not diagnostic. Video-assisted thoracoscopic biopsy is rarely needed to confirm the diagnosis of sarcoidosis.
Bronchoscopy
The bronchoscope allows for several different samples, including the transbronchial biopsy (TBB), endobronchial biopsy, and transbronchial needle aspiration (TBNA). The yield of TBB for the diagnosis of pulmonary sarcoidosis ranges from 60% to 97% depending on the number of biopsy specimens taken and the presence of parenchymal disease on chest imaging studies. Endobronchial biopsy results may be positive in up to 60% of patients with pulmonary sarcoidosis. Biopsy results are more frequently positive in individuals with abnormal-appearing airways but may provide the diagnosis even from normal-appearing airways. Furthermore, endobronchial biopsy can be performed with TBB and increases the yield for sarcoidosis above that using TBB alone.
TBNA has been extensively evaluated as a diagnostic approach for pulmonary sarcoidosis over the last decade. The diagnostic yield is in the range of 80%. The use of endobronchial ultrasonography has been shown to be superior to blind TBNA in two randomized trials. The yield for TBNA is much higher for patients with adenopathy on chest radiograph (stage 1 or 2), whereas the yield for TBB is higher for those with a stage 3 pattern on the chest radiograph. The use of on-site cytopathologic examination allows for rapid review of the TBNA specimen, and, if the TBNA is diagnostic, the bronchoscopist may not need to proceed to TBB.
Bronchoalveolar Lavage
Examination of inflammatory cells from BAL fluid is sometimes used as a complementary test for the diagnosis of pulmonary sarcoidosis. The diagnostic accuracy of the percentage of lymphocytes and the ratio of CD4/CD8 lymphocyte subpopulations in BAL has been assessed. In general, BAL lymphocytosis (>15% lymphocytes) has a 90% sensitivity for the diagnosis of sarcoidosis, although the specificity is lower. Other conditions leading to a BAL lymphocytosis must be excluded, including infections such as tuberculosis and fungal infections, lymphoma, and hypersensitivity pneumonitis. A BAL lymphocytosis of more than 60% and presence of mast cells is more common in hypersensitivity pneumonitis than in sarcoidosis. The BAL CD4/CD8 ratio is increased more than 3.5 times in 50% to 60% of pulmonary sarcoidosis patients. However, the specificity of the BAL CD4/CD8 ratio criterion has approached 95% in some but not all studies. Some have advocated that the BAL CD4/CD8 criterion is diagnostic of sarcoidosis when there are concomitant chest imaging findings compatible with sarcoidosis ; however, these criteria have not been formally tested. Like other studies, BAL provides supportive evidence for the diagnosis of sarcoidosis. In addition, BAL samples can be tested along with bronchial washings for evidence of fungal or mycobacterial infection.
Extrapulmonary Tissue Biopsy
Granulomas can be detected histologically in any organ that is involved with sarcoidosis. The biopsy of neural tissue and the heart are particularly problematic because of the potential morbidity associated with these procedures. Because patients with neurosarcoidosis will have extraneural sarcoidosis nearly 90% of the time, most patients have extraneural disease from which a biopsy specimen can be obtained. Although endomyocardial biopsy is a fairly specific test for cardiac sarcoidosis in the proper clinical setting, its sensitivity is very low. For this reason, endomyocardial biopsy is rarely performed for the diagnosis of sarcoidosis. Often, imaging studies are used as surrogate tests for the diagnosis of neurosarcoidosis and cardiac sarcoidosis. Such studies should be interpreted cautiously because their specificity depends upon associated clinical evidence for sarcoidosis, which should almost always include previous biopsy confirmation of granulomatous inflammation of unknown cause in another organ (see later). Also, the specificity of imaging studies for neurosarcoidosis and cardiac sarcoidosis is unknown.
Clinical Phenotypes Suggestive of Sarcoidosis
In some cases the clinical presentation of sarcoidosis is so specific that the diagnosis can be made without performing a confirmatory tissue biopsy. These presentations are listed in Table 66-2 . In patients with these presentations, one may consider performing a bronchoscopy to rule out other possible causes, including infection. In patients in whom a bronchoscopy is not diagnostic, the presence of factors in Table 66-2 may help bolster confidence for the diagnosis of sarcoidosis. Lupus pernio (see Fig. 66-14A ) is an indurated, raised skin lesion that is characteristically found on the ears, cheeks, and nose (see later); it is considered highly specific for sarcoidosis. Löfgren syndrome initially included only patients with erythema nodosum of lower legs or forearms ( Fig. 66-8 ). However, periarticular inflammation or arthritis of the ankles with or without erythema nodosum has been added to the expanded definition of Löfgren syndrome. Uveoparotid fever, also known as Heerfordt syndrome is unusual but specific for the diagnosis of sarcoidosis.
|
Other Diagnostic Approaches
On some occasions the diagnosis of sarcoidosis is suspected on clinical grounds although no specific organ can be found to biopsy. This would include patients presenting with disease of the eye, brain, or heart that is consistent with sarcoidosis but does not provide a safe site for biopsy. There is no established approach for this situation. Total body imaging such as 18 F-fluorodeoxyglucose positron emission tomography (PET) or gallium-67 scanning has been proposed in such cases to identify an organ for biopsy ( Fig. 66-9 ), although no rigorous analysis of this approach has been undertaken. Another suggested approach in this situation is to biopsy organs that are commonly affected, even in the absence of symptoms or other clinical findings suggestive of sarcoidosis involvement of that organ. Conjunctival biopsies have been performed in this situation, and the yield has ranged from 27% to 55% in patients without ocular symptoms. In vivo confocal microscopy of the conjunctiva can detect multinucleated giant cells without the need for a biopsy, a noninvasive approach that has been advocated to confirm granulomatous inflammation in sarcoidosis patients. Liver biopsy demonstrates granulomas in 24% to 78% of sarcoidosis patients, even when they have no symptoms attributable to the liver and have normal serum liver function test results. However, hepatic granulomas are not specific for sarcoidosis so that clinical evidence of extrahepatic sarcoidosis must be present for the diagnosis to be secure. Andonopoulos and colleagues found that gastrocnemius muscle biopsy revealed granulomas in 22 consecutive patients without muscle symptoms who had bilateral hilar adenopathy on chest radiograph. However, most of these patients had strong clinical evidence of sarcoidosis; furthermore, this procedure is fairly invasive.
Another test to consider when sarcoidosis is suspected on clinical grounds although no specific organ is identified to biopsy is the Kveim-Siltzbach test. This test is only available in selected centers and may be indicated when chest imaging studies are normal (e.g., in cases of uveitis of unknown origin, hypercalciuria, hepatic granulomatous disease, suspected neurosarcoidosis, or recurrent erythema nodosum). This test involves the intradermal inoculation of a suspension of splenic tissue from spleen that was involved with sarcoidosis. If a skin nodule develops at the inoculation site in 4 to 6 weeks, it is biopsied. A biopsy demonstrating noncaseating granulomatous inflammation is highly specific for the diagnosis of sarcoidosis. The Kveim-Siltzbach test has a false-negative rate of 20% to 40%. The sensitivity and specificity of the test vary depending on the spleen that is used to prepare the Kveim-Siltzbach reagent and the duration of disease. Transmission of infective agents is possible if the antigen is poorly prepared or controlled.
Pathologic Findings
Granulomatous inflammation is necessary to establish a diagnosis of sarcoidosis in most cases; however, the finding of granulomas is not sufficient for the diagnosis of sarcoidosis ( Fig. 66-10 ). Meticulous histologic examination with appropriate staining of all biopsy specimens should be performed to search for known causes of granulomatous inflammation, such as mycobacteria, fungi, parasites, and foreign material (e.g., talc).
Although there are no specific histologic features that are diagnostic of sarcoid granulomas, there are certain features that suggest this diagnosis. The sarcoid granuloma usually consists of a compact (organized) collection of mononuclear phagocytes (macrophages and epithelioid cells). Typically there is no necrosis within the sarcoid granuloma; however, on occasion, there is a small to moderate amount of necrosis. Usually, giant cells fuse within the sarcoid granuloma to form multinucleated giant cells. These granulomas are typically surrounded by lymphocytes in the periphery. A variety of inclusions may be present within the sarcoid granuloma, including asteroid bodies, Schaumann bodies, birefringent crystals, and Hamazaki-Wesenberg bodies; however, these inclusions are neither specific nor diagnostic for sarcoidosis ( Fig. 66-11 ). In particular, birefringent crystals within the sarcoid granuloma may lead to a misdiagnosis of talc granulomatosis. Care must be taken to ensure that the crystal morphologic features and size are compatible with intravenously injected talc to ensure the diagnosis of talc granulomatosis.
Sarcoidosis has been defined as a multisystem granulomatous disorder of unknown cause. The fact that the disease is “multisystem” implies that there should be evidence of granulomatous inflammation in at least two organs for the diagnosis of sarcoidosis to be secure. However, the biopsy of a second organ is not routinely performed. Of note, there are some conditions that appear to be distinct from sarcoidosis in which granulomatous disease seems to be limited to a single organ (e.g., idiopathic granulomatous hepatitis, and idiopathic panuveitis ).
Other Idiopathic Multiorgan Granulomatous Diseases
There are other multiorgan granulomatous syndromes that should be considered in the differential diagnosis of sarcoidosis. These include infections, such as tuberculosis.
Blau syndrome consists of granulomatous iritis, arthritis, and skin rash. The disease is a genetic disorder and has an autosomal dominant pattern of inheritance with variable penetrance. In contrast to sarcoidosis, most cases present before 12 years of age. Blau syndrome is considered a separate entity from childhood sarcoidosis on the basis of a lack of visceral (e.g., pulmonary) involvement, mode of inheritance, and absence of Kveim-Siltzbach skin test reactivity. Also, the Blau gene has not been found in sarcoidosis patients.
The granulomatous lesions of unknown significance (GLUS) syndrome consists of granulomatous inflammation in the liver, spleen, bone marrow, and lymph nodes, a benign course, and a tendency for recurrence. The GLUS syndrome is thought to be distinct from sarcoidosis because (1) elevated SACE levels are not found, (2) hypercalcemia is not found, (3) the Kveim-Siltzbach test results are negative, and (4) immunotyping of the T lymphocytes in the GLUS syndrome granulomas is different from that in sarcoidosis granulomas.
Necrotizing sarcoid granulomatosis is a systemic granulomatous vasculitis. Because of vascular involvement, necrosis is a prominent feature, unlike in most cases of sarcoidosis. Although the lung is commonly involved, extrapulmonary involvement is also common. It is debated whether necrotizing sarcoid granulomatosis is a distinct clinical entity or a form of systemic sarcoidosis.
Evaluation of Pulmonary Disease
The evaluation of pulmonary disease in sarcoidosis patients relies on three major determinants: pulmonary function, chest imaging, and symptoms. There have been several instruments developed in all three of these areas. The pulmonologist uses pulmonary function testing and chest imaging as a method to measure lung involvement. However, the patients’ major concerns are how they feel and how their lung disease affects their quality of life. Dyspnea and cough are the major reasons for treatment of pulmonary sarcoidosis.
Pulmonary Function
A significant proportion of sarcoidosis patients will have normal spirometry findings and lung volumes at the time of diagnosis. Over time, some of these patients will develop a restrictive pattern, with reduction of lung volumes. However, a significant proportion of sarcoidosis patients have obstructive lung disease. The diffusing capacity for carbon monoxide (D l CO ) is a more sensitive measure of early interstitial lung disease and often predicts a reduction in exercise capacity. A disproportionately reduced D l CO (compared to lung volumes) may also be an indicator of sarcoidosis-associated pulmonary hypertension.
Clinical pulmonary exercise testing may be useful in assessing dyspnea in sarcoidosis patients. Patients with normal lung function test results may still have abnormalities in exercise testing. However, exercise testing is not well standardized and relatively cumbersome to perform. The 6-minute walk distance (6MWD) test has been widely used to assess exercise capacity. In sarcoidosis a reduction in 6MWD has been found to correlate with reduced forced vital capacity (FVC), fatigue, and quality-of-life measures. A reduced 6MWD and oxygen desaturation have been found in patients with sarcoidosis-associated pulmonary hypertension. However, several factors besides cardiopulmonary function influence the 6MWD test, including muscle strength, joint disease, and neurologic symptoms.
Lung Imaging
Chest Radiograph
The findings on the routine chest radiograph have been classified into several stages, originally proposed by Scadding (see earlier). The Scadding chest radiographic stages correlate with prognosis. Those with stage 1 (see Fig. 66-3 ) have a more than 80% chance of resolution of hilar adenopathy 2 to 5 years after presentation, whereas those with stage 3 have less than a 30% chance of resolving to a normal chest radiograph. Fibrotic changes (stage 4) do not resolve. The limitation of chest radiographic staging is that it does not characterize extrapulmonary manifestations of sarcoidosis. In addition, there is significant variability in the classification of radiographic stages even among experienced radiologists.
An alternative to monitoring the chest radiograph by Scadding staging is to compare chest radiographs serially over time. This is what is commonly done in clinical practice. The change in the chest radiograph has been shown to have good reproducibility and correlates with changes in pulmonary function. In one study the kappa coefficient for comparison reading was much better than for the Scadding scoring system.
Another method for scoring the chest radiograph developed by Muers and coworkers is similar to a scoring system used for pneumoconiosis. It was shown that the portion of this score that assessed reticulation was able to detect changes after corticosteroid therapy and after treatment with infliximab. The scoring system was reproducible with a good kappa coefficient. However, the scoring is tedious and does not appear practical for routine clinical practice.
Chest Computed Tomography Scanning
As noted earlier, chest computed tomography (CT) imaging, including HRCT, has proved useful in the diagnosis of sarcoidosis (see Figs. 66-3 to 66-6 ). A scoring system for CT scans in sarcoidosis has been described and has been suggested to assess severity of disease. However, it is unclear what features of the CT scan are important for the management of the disease. Some HRCT scan features do correlate with physiologic impairment in sarcoidosis; for example, honeycombing is associated with a reduced D l CO and an increased alveolar-arterial P o 2 difference, and peribronchial thickening may lead to airway obstruction and air trapping. HRCT is also useful in defining the extent of bronchiectasis and the presence of complications, such as aspergilloma ( Fig. 66-12 ).
Radionuclide Scanning
Radionuclide scanning has been used as an aid in the diagnosis of sarcoidosis and the assessment of organ involvement. Early studies focused on the gallium scan as a method of identifying inflammation. The presence of extrapulmonary disease was often useful in identifying potential areas to biopsy. In addition, uptake in the parotid and lacrimal glands (“panda sign”) and/or right paratracheal and bilateral hilar areas (“lambda sign”) were highly supportive of the diagnosis of sarcoidosis. Octreotide scanning has also been used as a marker of lung inflammation in sarcoidosis.
The detection of radioactive glucose uptake by PET scan has been widely applied to malignant conditions, including lung cancer and lymphoma. PET scan activity can also be quite enhanced in sarcoidosis, demonstrating diffuse activity in the lungs, mediastinal lymph nodes, and other parts of the body (see Fig. 66-9 ). The PET scan has proved useful in identifying potential areas for biopsy in those suspected of sarcoidosis. It has also been found to demonstrate ongoing activity in patients who are receiving immunosuppressive therapy and may suggest that the disease will relapse when therapy is withdrawn. Patients with increased PET uptake in the pulmonary parenchyma have been shown to have active disease by virtue of changes in their pulmonary function. In this study, sarcoidosis patients with increased parenchymal PET uptake who were treated showed a significant improvement in their FVC and D l CO , and those who were not treated showed a significant fall in D l CO ; those without increased PET parenchymal activity had no change in FVC and D l CO without therapy, indicating inactive disease. In a study comparing PET to gallium scans, the PET scan was found to be more sensitive and reproducible. Because the PET scan is more widely available and can be performed in 1 day rather than the 2 days necessary for gallium scanning, the PET scan is preferred, provided that there are no issues with reimbursement. Imaging of the heart for detection of cardiac sarcoidosis can be performed with either PET or magnetic resonance imaging (MRI) scanning ; cardiac MRI appears to be more sensitive and specific. However, cardiac MRI requires special expertise to interpret and is usually contraindicated once a defibrillator or pacemaker has been placed.
Health-Related Quality of Life
Sarcoidosis has been associated with impaired health-related quality of life (HRQOL), and treatment of sarcoidosis has been associated with changes in HRQOL. However, the results of studies have been discordant. Corticosteroids have demonstrated improvement, no change, or worsening in the short form 36, a general quality-of-life instrument. The Saint George Respiratory Questionnaire, which was originally developed as a measure of HRQOL in chronic obstructive pulmonary disease, has also been used for several interstitial lung diseases, including idiopathic pulmonary fibrosis and sarcoidosis. Improvements in the Saint George Respiratory Questionnaire results have been reported for patients with sarcoidosis-associated pulmonary hypertension who have been treated with pulmonary hypertension therapy. In addition, two sarcoidosis-specific quality-of-life instruments have been developed: the Sarcoidosis Health Questionnaire and the King’s Sarcoidosis Questionnaire.
Fatigue is a common complaint in sarcoidosis patients. It has been reported by more than half of sarcoidosis patients in both Europe and the United States. Although fatigue is reported by patients with both pulmonary and extrapulmonary sarcoidosis, it is more common in the latter. Fatigue may persist long after other evidence of disease has regressed. There are several fatigue scales that are not specific for a particular disease ; however, the Fatigue Assessment Scale is a sarcoidosis-specific fatigue questionnaire. The questionnaire appears to have a good correlation with general fatigue questionnaires in some but not in all studies. The Fatigue Assessment Scale has been found to improve in sarcoidosis patients treated with infliximab.
Depression is a common underdiagnosed problem in sarcoidosis patients. The prevalence of depression ranged from 25% to 60% of the subjects and may contribute to a poorer quality of life. Consequently, a referral for a psychiatric or psychological evaluation and counseling should be considered for many sarcoidosis patients.
Sarcoidosis-associated pulmonary hypertension (SAPH) can develop from a variety of mechanisms, including left ventricular diastolic dysfunction, pulmonary arterial vasculitis, pulmonary veno-occlusive disease, pulmonary fibrosis, and hypoxia. The overall incidence of pulmonary hypertension in sarcoidosis appears to be 5% to 15%. In patients with moderate to severe symptoms, the prevalence of pulmonary hypertension has been reported to be 50% or greater. Precapillary pulmonary hypertension is the most common cause of SAPH, but left ventricular diastolic dysfunction is present in a significant proportion of cases. The survival of patients with precapillary SAPH is significantly worse than for those with diastolic dysfunction.
Extrapulmonary Sarcoidosis
Sarcoidosis is a multiorgan disease. Pulmonary involvement has been reported in over 90% of cases in most large series. However, several other organs are commonly affected with the disease. Table 66-3 lists organ involvement as detected by two large clinics in the United States with a total of more than 2700 patients. Both of these groups used the same standardized criteria for organ involvement and listed only those organs with definite or probable disease. Patients may have undetected organ involvement. For example, liver biopsies performed in patients without symptomatic liver disease may still demonstrate granulomatous inflammation in up to half of the cases.
| Charleston (%) | Cincinnati (%) | |
|---|---|---|
| Female | 66 | 71 |
| African American | 66 | 43 |
| SPECIFIC ORGAN INVOLVEMENT | ||
| Lungs | 89 | 88 |
| Eye | 23 | 33 |
| Skin | 32 | 33 |
| Liver | 20 | 13 |
| Neurologic | 9 | 14 |
| Cardiac | 4 | 5 |
Eyes
The eyes are involved in about a third of patient in the United States and Europe. Eye disease is more common in African Americans and in women. In Japan up to 80% of sarcoidosis patients have eye disease. Table 66-4 lists some of the more common ocular manifestations of sarcoidosis. It is recommended that all sarcoidosis patients undergo an ocular examination by a specialist at the time of initial evaluation. Uveitis is the most common manifestation of ocular sarcoidosis and at times can be clinically silent. Anterior uveitis typically presents acutely, with pain, photophobia, lacrimation, and redness, but may have a more chronic course. Posterior uveitis is typically gradual in onset and is more likely to affect vision. Conjunctival nodules are another common, although usually asymptomatic, presentation. In fact, the diagnosis of sarcoidosis can sometimes be made with conjunctival biopsy, even in patients without any ocular symptoms. Patients with ocular sarcoidosis may develop complications from eye involvement, such as sicca syndrome, glaucoma, and cataracts. Cataracts and glaucoma can also be a complication of either local or systemic corticosteroid treatment. Although less common, lacrimal gland enlargement is a characteristic finding. Vigilance for ocular involvement is extremely important, because approximately 10% of patients with sarcoid-associated uveitis develop blindness in at least one eye.
| Uveitis: anterior, intermediate, posterior |
| Optic neuropathy |
| Adnexal and orbital disease |
| Lacrimal glands: enlargement, sicca syndrome, dacryocystitis |
| Orbital mass |
| Scleral involvement |
| Glaucoma |
| Cataracts |
Skin
Cutaneous involvement is also encountered in about a third of sarcoidosis patients. Skin manifestations include maculopapular lesions, papules, hyperpigmentation, and hypopigmentation ( Fig. 66-13 ). Lupus pernio is an indurated facial lesion that is specific for sarcoidosis ( Fig. 66-14 ). It is more common in patients of African descent. It is usually chronic and can be resistant to usual therapy. Erythema nodosum, as noted earlier, is often, but not always, associated with a good prognosis. In particular, patients of African descent with erythema nodosum will often have chronic disease. Other cutaneous involvement may appear as patches, plaques, violaceous areas, localized alopecia, ichthyotic areas, subcutaneous nodules, ulcers, and pustules.
Nervous System
Neurosarcoidosis develops in less than 10% of patients. Neurologic disease can involve the spinal cord or just the cranial nerves. The most sensitive method of detecting disease is MRI with gadolinium because neurologic lesions will often enhance. Spinal fluid analysis can be helpful, because it may demonstrate a lymphocytosis and elevated protein level. An elevated level of ACE in spinal fluid has been reported to be specific for neurosarcoidosis, but a sensitivity as low as 50% has been reported. In patients with neurosarcoidosis, especially those who present as isolated neurosarcoidosis, one has to consider multiple sclerosis. Patients with sarcoidosis can develop optic neuritis, with or without uveitis. It is important to distinguish these patients from multiple sclerosis, because neurosarcoidosis patients will often improve with treatment, including with anti– tumor necrosis factor (TNF) treatments such as infliximab.
Heart
Sarcoidosis can directly cause an infiltrative cardiomyopathy with two major manifestations: arrhythmias and reduced ejection fraction. The detection of cardiac sarcoidosis can be difficult. The presence of cardiac symptoms, including palpitations and syncope, is a sensitive but not specific screening for sarcoidosis. Echocardiography and 24- to 48-hour cardiac monitoring are useful supplemental tests for screening. Table 66-5 summarizes the prevalence, sensitivity, and specificity for cardiac sarcoidosis of screening tests by themselves or in combination. Currently imaging with MRI and/or PET scanning is felt to be the most specific for diagnosing cardiac sarcoidosis. The arrhythmias, especially ventricular arrhythmias, remain a potential cause of death in cardiac sarcoidosis. However, implantable defibrillators have markedly reduced the mortality from this manifestation. Therefore patients with potential cardiac sarcoidosis should be screened for ventricular arrhythmias and evaluated by a specialist for the need for defibrillator implantation.
| Abnormalities on Baseline Testing | Prevalence | Sensitivity | Specificity |
|---|---|---|---|
| History of cardiac symptoms | 12 (19%) | 46% | 95% |
| ECG | 3 (5%) | 8% | 97% |
| Holter monitor | 13 (21%) | 50% | 97% |
| Echocardiogram | 8 (13%) | 25% | 95% |
| Any screening variable | 29 (47%) | 100% | 87% |
| Two or more variables | 7 (11%) | 25% | 97% |
| Three or more variables | 1 (2%) | 4% | 100% |
| All variables abnormal | 0 (0%) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
