Article Text

Download PDFPDF

Sarcoidosis: the links between epidemiology and aetiology
  1. Simon Dubrey1,
  2. Shreena Shah2,
  3. Timothy Hardman3,
  4. Rakesh Sharma4
  1. 1Department of Cardiology, Hillingdon Hospital, Uxbridge, Middlesex, UK
  2. 2The Medical Admissions Unit, Queens Hospital, Romford, Essex, UK
  3. 3Niche Science & Technology Ltd., Unit 26, Richmond-Upon-Thames, London, UK
  4. 4Department of Cardiology, The Royal Brompton Hospital, London, UK
  1. Correspondence to Dr Simon W Dubrey, Department of Cardiology, Hillingdon Hospital, Pield Heath Road, Uxbridge, Middlesex UB8 3NN, UK; simon.dubrey{at}


Sarcoidosis is a multisystem inflammatory disease, the aetiology of which has still to be resolved. The proposed mechanism is that a susceptible genotype is exposed to one or more potential antigens. A sustained inflammatory response follows, which ultimately results in pathognomonic granuloma formation. Various clinical phenotypes exist with specific genetic associations influencing disease susceptibility, protection, and clinical progression. Occupational and environmental factors, including microbial elements, may then effect the development of this disease. Sarcoidosis is a heterogeneous disease, showing geographic and racial variation in clinical presentation. It demonstrates a familial tendency and clear genotype associations. Additionally, it appears to cluster within closely associated populations (eg, work colleagues) and appears to be related to selected occupations and environmental exposures. Frequently occult, but occasionally fatal, this disease has a very variable prognosis. It is also unusual in having no specific biomarker. The epidemiology and multiple factors that appear to influence the aetiology of sarcoidosis illustrate why this disease state is frequently described as a clinical enigma.

View Full Text

Statistics from


Since sarcoidosis was first reported, the search continues for an aetiology, a biological marker and specific cure for this disease. Attention has shifted to a more genetic orientated approach, but researchers continue to argue the case for infective, environmental and occupational factors. Sarcoidosis is a multisystem inflammatory disease in which non-caseating granulomas form in almost any organ or tissue (figure 1). The exposures and occupations shown in figure 2 indicate the vast range of factors proposed as influencing the aetiology of this disease.

Figure 1

Summary of factors influencing the development and/or resolution of sarcoid disease.

Figure 2

Summary of features associated with the aetiology of sarcoid disease. HLA, human leucocyte antigen.

From the outset sarcoidosis is an unusual condition; many patients are unaware they have the disease with the majority recovering spontaneously and without treatment. For a smaller, but significant, proportion of patients there can be devastating and, on occasions, fatal consequences. Despite agreement that the condition is influenced by a variety of genetic and environmental interactions, many questions remain unanswered. These questions, with no specific ‘cure’, or biomarker of disease presence, means sarcoidosis has frequently been designated an ‘enigma’.1–4

We describe the epidemiology and aetiology of sarcoidosis and expand on reasons why this disease continues to confound the medical profession.


Human sarcoidosis is recognised to occur worldwide, although large regional variations in prevalence exist. In Europe, Caucasians are more commonly affected than other races and western Europeans more than eastern Europeans. Scandinavia, particularly Sweden, has a high prevalence of 64 per 100 000 population (incidence of 19 per 100 000/year).5 UK estimates, reported in 1992, are <1 per 100 000 of the population.6 Later estimates (1991–2003) indicate an annual incidence for the UK of 5 per 100 000.7 A study within London reported annual incidence rates per 100 000 population of 4 for British, 21 for Irish, 58 for West Indian, and 14 for Asian patients.8

Historically, Irish women in London have been reported as having a prevalence of 200 per 100 000. This alarming, and frequently cited, value appears to have been derived from documents reporting an incidence rate of 3 per 1000 in the years 1958–1960 that fell to 1.7 per 1000 in the period 1961–1963.9 More recently (1996–2005), the prevalence of sarcoidosis for the Irish republic was 28.13 per 100 000 of the population, compared with 11.16 per 100 000 for Northern Ireland. Within this period, distinct spatial clusters were detected, reaching 44.9 per 100 000 in the northwest of Ireland,10 which approach the rates reported for Scandinavia. A specific time–space cluster, between the years 1999 and 2000, produced a prevalence rate for the Galway/Mayo region of 96.55/100 000.10 These reports, and a further high prevalence rate of 85 cases/100 000, reported in 2011 from County Offaly indicate the Irish propensity for sarcoidosis.11

In the USA sarcoidosis is more common in African Americans, with rates as high as 40 per 100 000/year, compared to those of white European origin with a range of 5–11 per 100 000/year.12 The estimated lifetime risk of developing sarcoidosis among African Americans may be as high as 2% and reaching 3% in women.13 The effect of population migration has been suggested to explain the greater prevalence among those of African origin in the USA. This is supported by the one study from the African continent which found a similar disproportion of prevalence among blacks (23.2/100 000) and whites (3.7/100 000). In a South African study the prevalence among those of mixed race (‘coloureds’) was intermediate at 11.6/100 000.14 Disease severity also appears to be greater in those of African origin compared to those of white European origin.2 In western countries most fatalities are due to advanced pulmonary disease with lung fibrosis ultimately causing respiratory failure or pulmonary hypertension.15 However, variability in phenotype with race are evident, with 13–50% of deaths from sarcoidosis in Americans of African or European origin due to cardiac involvement, as compared to 77–85% in Japanese patients where this disease phenotype predominates.16 However, no races seem immune and there tends to a female preponderance throughout all populations17–19—a feature reminiscent of many autoimmune conditions. Women are also more likely to have systemic disease20 and more severe pulmonary involvement than men.15 Overall, the incidence of sarcoidosis is highest in African American women, followed in order by African American men, white women, and white men. Differences in prevalence and phenotype, between any broad racial groups, might also be influenced through socioeconomic disparity, influencing numerous environmental exposures.21

Sarcoidosis shows age clustering, with a predilection for the disease in early adulthood, peaking in those aged 20–29 years.15 In Scandinavian countries and Japan there is a second peak in women over age 50 years.22 It is unusual in children, teenagers, and those aged ≥70 years.23

Interestingly, sarcoidosis appears to be extremely unusual in other animal species.24 Rare reports of horses suffering from a moderate to severe wasting disease, fever, and lymphadenopathy with sarcoidal granulomatous inflammation of multiple organs has been reported.25 As in human sarcoidosis, the aetiology and pathogenesis of equine sarcoidosis remains obscure. Occasional cases of cutaneous ‘sarcoidosis’ in dogs have been described and there are sparse reports of possible sarcoidosis in cats.26 These examples are distinct from the localised sarcoid-like tissue reactions produced by inoculation of animals with sarcoid tissue. In addition, such inoculated animals do not develop the generalised systemic condition of sarcoidosis.

Clinical features and differing phenotypes in sarcoidosis

Many patients are asymptomatic, but when present features may be localised or systemic and can mimic several other conditions. Sarcoidosis most frequently involves the respiratory system,27 but virtually any organ system can be involved, with clinical consequences (table 1).

Table 1

Organ system involvement and clinical features of sarcoidosis

The diagnosis of sarcoidosis can be difficult as there is no unique identifying feature and no biomarker. The differential diagnosis will depend on which organs are involved and usually entails the exclusion of other granulomatous diseases, tuberculosis, immune deficiency syndrome, certain manifestations of cancers and lymphoma.15

Pathognomonic granulomas can form in any tissue but the spectrum appears divided into several distinct clinical forms with individualised outcomes. This raises the question of whether the term ‘sarcoidosis’ represents a spectrum of different diseases (box 1).

Box 1

Clinical forms of sarcoid disease

Disease types

  • Asymptomatic form—predominantly lung and lymph node involvement

  • Lofgren syndrome—hilar lymphadenopathy, arthritis, fever and erythema nodosum

  • Isolated uveitis

  • Lupus pernio—isolated skin involvement

  • Swelling of the parotid glands, fever and uveitis (Heerfordt syndrome or uveoparotid fever)

  • Acute onset form—with a short duration and early resolution

  • Systemic form—with a chronic and relapsing course

  • Localised scar and tattoo related sarcoidosis

Many of the individual features of sarcoidosis (patterns of organ involvement) also appear specific to individual geographic populations.28 Some of the clinical features associated with specific populations are shown in box 2.

Box 2

Clinical features predominating in specific populations

  • Erythema nodosum in young white females (Scandinavia, Ireland and Spain)

  • Lofgren syndrome in Scandinavians

  • Chronic dermatological lesions predominate in African Americans

  • Ocular and cardiac features in Japanese

  • Hypercalcaemia in Italians

  • Constitutional symptoms of fever, lymphadenopathy and weight loss in patients from the Indian sub-continent and of African ancestry

    Lupus pernio in Puerto Ricans

  • Chronic uveitis in African Americans

Despite these specific phenotypes, 30–50% of individuals are entirely asymptomatic and are usually identified by incidental chest radiography. Around a third of patients present with constitutional symptoms of weight loss, fatigue, fever and malaise—a presentation complex more prevalent in blacks and Asians.15 Racial differences also influence the disease course, with acute sarcoidosis more common in white Europeans than blacks and usually associated with remission within 2 years. In contrast, almost 80% of Japanese patients with sarcoidosis will die from cardiac related causes.29

Lofgren's syndrome, consisting of bilateral hilar lymphadenopathy, ankle arthritis, fever, myalgia, weight loss, and erythema nodosum, is also associated with spontaneous remission. The demographics, phenotype and genetics of Lofgren syndrome implicate this to be a distinct disease entity encompassed within the broader term of sarcoidosis.30

Heart involvement epitomises the conundrum of sarcoid disease with most patients unaware of its presence. Initial clinical presentation may be due to severe cardiac dysrhythmias, both in the form of tachycardia or heart block, both of which can prove fatal. Much less commonly, patients develop structural heart involvement and mechanical dysfunction. Clinical disease may be evident in only 2–7% of patients but in 20–76% at autopsy.16 Due to patchy involvement of the myocardium (usually the left ventricle) and an unexplained predilection for the basal interventricular septum, a right ventricular biopsy may provide a pathological diagnosis in only 25–50% of autopsy proven cases.

Genetic factors

An appreciation of both familial and racial influences18 ,31 point to a genetic element to the occurrence and prognosis of sarcoidosis. Twin studies have demonstrated an 80-fold risk of developing sarcoidosis in monozygotic twins (male or female) with a 7-fold increase in dizygotic twins, compared to the general population.32 The genetic effect is calculated as twice the difference between the correlation values for the presence of sarcoidosis in monozygotic and dizygotic twin pairs, which in this Scandinavian twin study gave a heritability value (genetic effect) of 0.66.

The genetic influence on the clinical presentation of this disease is illustrated by African Americans tending to have a preponderance of extra-thoracic involvement with a lower likelihood of spontaneous remission as compared to European Americans. The case–control aetiological study of sarcoidosis (ACCESS study) derived a relative risk of sarcoidosis developing in first or second degree relatives as 4.7.33 It is higher among siblings than between other relatives and familial risk was higher (19%) than for those of European origin (5%).33

In the 1970s investigators began searching for sarcoidosis susceptibility genes using case–control studies and by serologically testing for human leucocyte antigen (HLA) alleles, which occupy the major histocompatibility complex (MHC) region. HLA gene products are expressed at the cell surface where CD4+ T lymphocytes can recognise them and cascade an immune response.34 The HLA class I antigen HLA-B8 was soon found to be associated with acute sarcoidosis. Some evidence suggests that HLA class I and II genes may also be synergistically involved in the underlying pathophysiology.35 For HLA genes within class II, the HLA-DRB1 association predominates in the literature,36 ,37 with variations in this gene affecting both sarcoidosis susceptibility and prognosis. Some class II HLA genes (DRB1*11, DRB1*12, and DRB1*15) have been identified as being risk factors for the disease, while others (DQB1*0201, DRB1*01, and less so DRB1*04) appear protective against developing sarcoidosis.34 ,38 HLA DRB1*03 is associated with a particular acute form of sarcoidosis with a relatively good prognosis (Lofgren syndrome), while DRB1*1501 is associated with a more chronic disease course and severe pulmonary involvement.39 A full list of HLA associations with disease phenotypes, susceptibility, progression, protection, and prognosis is provided in recent reviews.36 ,38

Specific clinical phenotypes of the disease, including Lofgren syndrome, uveitis, hypercalcaemia, bone marrow, and cardiac or predominant pulmonary involvement, have now been identified as having specific HLA linkages.40 Such differences in clinical phenotype may explain why individual centres have reported different genotype and HLA associations. More encouraging is that some highly specialist units are now able to guide their management of different phenotypes, through an understanding of these associations.41

Several non-MHC candidate gene associations, (BTNL2, ANXA11, RAB23, and NOTCH4), chosen because of their known function in antigen presentation or T cell response, have been implicated in several studies.31 ,42 Genetic polymorphisms appear to affect each of various pathophysiological steps in overt sarcoidosis, starting with antigen presentation, through macrophage and T cell activation to granuloma accumulation. In some cases they may also influence tissue fibrosis. Many, including HLA associations, demonstrate ethnicity specific susceptibility and influence disease severity.31 ,36 ,37 However, unlike the HLA associations, non-MHC gene associations are not as consistently identified across populations or within family-based studies.38

An alternative approach is to identify candidates in chromosomally linked regions within families. Two affected sibling pair linkage studies in sarcoidosis have been reported, one in German families and the other in African American families. Their genome scans clearly suggest a heterogeneity of sarcoid risk between these two populations and the presence of multiple susceptibility genes.37 In the German population the linkage was found in the MHC region on the short arm of chromosome 6, with a suggestion of involvement of chromosomes 1, 3, 9 and X.43 Follow-up studies examining chromosome 6 identified a candidate gene, the butyrophilin-like 2 (BTNL2) gene—a single nucleotide polymorphism that is also located in the MHC class II region—to be consistently associated with sarcoidosis.44 ,45

In an African American study (the Sarcoidosis Genetic Analysis (SAGA) study), decreased allele sharing among discordantly affected pairs, as well as increased sharing among concordantly affected sib pairs, pointed to both ‘susceptibility’ and ‘protective’ alleles on the short arm of chromosome 5, but also on chromosomes 1, 2, 11, and 20.37 In a follow-up to this, microsatellite markers positioned this susceptibility allele on chromosome 5 at 5p15.2, with a protective allele at position 5q11.2.46 More recently, several studies have identified multiple gene loci associated with the risk of sarcoidosis, on chromosomes 1, 2, 5, 6, 9, 11, 12, and 20.33–36 Differences in disease susceptibility between ancestral populations have also been used to identify genetic loci associated with disease and ancestry.47 ,48 The genetic influence on the clinical presentation is illustrated by those of African origin having a preponderance for extra-thoracic involvement, more severe disease, and a lower likelihood of spontaneous remission.7 ,49

Thus far, researchers have demonstrated familial clustering, increased concordance in monozygotic twins, distinct variations among racial groups, HLA associated phenotypes and gene associations which support the concept of a genetic susceptibility, further influenced by environmental precipitants.

Environmental factors

As an overview, any environmental antigen must be capable of inducing features of a T-helper-1 lymphocyte mediated immune response, through the release of T cell cytokines and an oligoclonal expansion of T cells, which ultimately form granulomas. Sarcoidosis appears to resemble the disease reaction seen on exposure to beryllium, a widely used metal (in computers, cars electronics, ceramics, nuclear weapons, dental alloys) and by other metal-associated granulomatous diseases in an antigen-induced manner. An unusual type of sarcoidosis is that which occurs at the site of tattoos. Several case reports and small series describe sarcoid reactions developing in relation to individual pigment colours (red, green, or blue/black) that contain metallic elements. In some cases, local tattoo reactions have preceded the diagnosis of systemic disease, which in most cases were predominantly pulmonary.50

Increasingly, evidence suggests that sarcoidosis can occur in association with specific occupations where there is exposure to foreign antigens that promote an exuberant immune response.19 ,51 Proposed inducers of sarcoidosis fall into several antigenic categories (box 3).

Box 3

Proposed antigens in the aetiology of sarcoidosis

Environmental antigens

  • Metals

    • Aluminium, zirconium, titanium, beryllium

  • Organic dusts

    • Pine tree pollen, wood dusts

  • Inorganic dusts

    • Clay, soil, talc, glass fibres, starch, silicone, complex alkaline dust (World Trade Centre)

Infectious agents

  • Bacteria

    • Mycobacterium tuberculosis, Mycoplasma species, atypical mycobacteria

    • Corynebacteria species, spirochaetes, Propionibacterium acnes, Borrelia burgdorferi

  • Viruses

    • Herpes simplex, cytomegalovirus, hepatitis C, Epstein-Barr, coxsackie, rubella

  • Fungi

    • Histoplasma, cryptococcus, coccidioidomycosis, sporotrichosis


  • Heat shock protein

Other proposed antigens

  • Radon

Reports of cases clustered among individuals within certain occupations include nurses, US Navy personnel, and firefighters,22 ,52–54 although occupations in ‘hazardous’ environments may be biased by increased surveillance or screening due to the nature of their work. A US based study has identified agricultural employment, insecticides, and microbial bio-aerosols (moulds and mildews) as being associated with sarcoidosis risk.30 Further analysis of the ACCESS study data reported positive associations with employment in the retail of building materials, hardware, garden supplies, and mobile homes. Employees in these occupations were three times more likely to have sarcoidosis than were controls.55 A rural setting has been associated with development of the disease with a dose-response gradient demonstrated for the use of wood-burning stoves and fireplaces.56 This suggests the handling of wood is a culprit, and corroborates historical suggestions of exposure to pine pollen as a disease trigger.57 As recently as 2005, reports continued to support an association between pulmonary sarcoid and pinewood dust.58

Associations with specific geographic areas has created the hypothesis of a ‘sarcoid terroir’ with researchers considering soil, proximity to forests, and exposure to farm animals and pets among possible factors providing localised environmental conditions permissive to disease development. Within the UK, a study in the early 1960s demonstrated an increase in sarcoidosis from north to south for both sexes.59 This is an unusual feature for disease distribution within the UK, as most diseases follow an inverse geographic relationship to this. The difference was greatest for men and showed an incidence of 0.41/10 000/year in Cornwall and Plymouth, compared to 0.21 in north east England and east Scotland. Compared to later studies, all subjects were described as being of Caucasian stock. Despite the likelihood of greater variability between industrial and rural areas at this time (1961–1966), the authors were unable to identify any relationship to occupational factors or forestation. Later studies are divided about such a north–south trend in disease prevalence,8 ,60 possibly reflecting changes in occupational and socioeconomic environment or population genotype. In one study, no cases were reported from the west country (Devon and Cornwall) and only 10.4% (10 of 163) from north of Leicester. However, this study only examined cardiac cases and was probably biased by the collator, author and leading authority being located in Cambridge.60

Rather oddly, smoking at any time appears to be associated with a reduced risk.30 ,61 It has been suggested that foreign particulate matter, at a nanoparticle scale (<1 μm), may be the reason why agents cannot be visualised under microscopy.3 Sarcoidosis, while being a multisystem disease, will almost invariably involve the lungs, with the implication that the primary route of entry of any antigen might be by inhalation. However, lung involvement, although frequent, is by no means obligatory. Pulmonary sarcoidosis was noticed to be significantly higher in New York fire-fighters involved in the aftermath of the World trade Centre terrorist attack, suggesting an inhaled microbial or inanimate particulate agent.62 Earlier reports (1985–1998) had also indicated a higher annual incidence in New York fire-fighters (12.9/100 000) compared to a control group of emergency health care workers (0/100 000).53 ,54 Among US Navy personnel, reports suggest a statistical association of a diagnosis of sarcoidosis with assignment to aircraft carriers and ‘dirty ships’ (industrial type ships) and reduced risk in men working on ‘clean ships’.63 Aside from these relationships, and despite some fluctuations in the ratio over the last 40 years, the prevalence among African American enlisted Navy men has always surpassed that of white men. One of the problems in understanding occupational exposure as a causative factor is an unexplained overriding female preponderance.

Among these potential antigens, some have been specifically identified at the point source of exposure and then reclassified from an ‘idiopathic’ condition (ie, sarcoidosis) to a specific condition. Examples include chronic beryllium disease, metalworking-fluid hypersensitivity pneumonitis, and sick-building syndrome.51 ,64 The latter is likely related to fungal contamination coexisting with damp buildings and supports the role of moulds and musty environments as precipitants of ‘sarcoidosis’.30 Occupants of historically water-damaged premises have been shown to have a high prevalence of new-onset sarcoidosis.65 Metalworking-fluid associated ‘sarcoidosis’ might also be due to bacterial or fungal contamination of the liquids and generated aerosols. In support of the mold/fungal antigen theory, blood monocytes from patients with sarcoidosis show reactivity to fungal cell wall components.66

In addition, seasonal clustering has been described in several countries, including the UK, Greece, Spain, and Turkey, with the peak in cases occurring in the spring.17 ,67 Southern hemisphere countries show a clustering towards the spring months (August–October).68 Any relationship of this seasonal clustering to specific antigen sources, including confinement in buildings and homes in winter months, agriculture, forestry, insecticides, bio-aerosols and activity in the construction industry, is unclear.

Genetic and environmental overlap

A landmark case–control study performed on the Isle of Man illustrates this overlap between genetic, environmental, and occupational exposures. This study found that 40% of 96 patients with sarcoidosis had been exposed to at least one other person with the disease, compared with 1–2% of control subjects. Fourteen cases occurred within the same household, of whom nine were blood relatives. Nineteen pairs were associated through work, two were neighbours, and 14 were friends.69 ,70 A high proportion (18.8%) of those with sarcoidosis were healthcare workers (predominantly nurses), a rate considerably higher than that in the control subjects (4.2%).

Occupational exposure as a risk to development of sarcoidosis may also be influenced by racial subgroup. The ACCESS study concluded that genetic factors appear stronger predictors of sarcoidosis than environmental factors.23 Caucasians appear to show elevated risk with exposure to industrial organic dusts, but a negative relationship to jobs involving metal dust or metal fume exposure.55 The latter is surprising given the similarity of beryllium related lung disease and pulmonary sarcoidosis. Different phenotypes of the disease (ie, predominant systemic extrapulmonary involvement in black patients) suggest that these subgroups have individual environmental exposure associations with their genotype. In African Americans, occupations within the transport service industry or education sector are associated with increased risk, and, in contrast to Caucasians, also with metal working. Specific risk associations include titanium, vegetable dust, indoor high humidity, water damage or musty odours.19 One of the strongest gene–environment interactions found is between HLA DRB1*1101 and occupational insecticide exposure. The situation is further complicated by the fact that genes which influence the sarcoid phenotype may be separate from sarcoidosis susceptibility genes.37

Infection and sarcoidosis

Several infectious agents continue to be proposed inducers of sarcoidosis, not least the Mycobacterium species or antigenic portions thereof.71 However, mycobacteria have not been cultured or histologically stained in sarcoid tissue, with one author stating “that experienced infectious disease pathologists have not seen a micro-organism after a century of looking”.3 There is some evidence for mycobacterium-specific nucleic acid sequences in a significant proportion of sarcoid patients,72 and cell wall deficient remnants of mycobacterium have also been implicated,73 although the latter have been isolated as frequently in healthy controls.74 The examination of archival biopsies for bacterial rRNA found no association between Mycobacterium tuberculosis complex infection and sarcoidosis.75 The same research group cultured mediastinal lymph nodes from patients recently diagnosed with pulmonary sarcoidosis and found no evidence of mycobacteria after 12 months. Additionally, all bronchial washings, microscopy and analysis for M tuberculosis DNA were negative.76 Furthermore, if tuberculosis and sarcoidosis are related, a reduction in cases of sarcoidosis in parts of the world where tuberculosis has declined might be predicted but has not materialised.

Also proposed is Propionibacterium acnes, a common skin commensal that has been shown to be capable of inducing pulmonary granulomas in a mouse model.77 To date P acnes is the only microorganism that, while not proving causality, has been both isolated from human sarcoid granuloma and cultured from sarcoid lesions.78 In one multinational study, propionibacterial DNA was detected in 106 of 108 sarcoidosis lymph node samples, compared to 0–9% proving positive for M tuberculosis DNA. However, propionibacterial DNA was also found in 0–60% of control or tuberculous samples.79 Recent work suggests P acnes may persist in a latent cell wall deficient form, with granulomatous inflammation triggered by intracellular proliferation of the bacterium.80

The ACCESS trial also failed to identify an infectious agent using blood cultures for cell deficient microorganisms and DNA PCR tests.81 A recent review reiterates that there remains no evidence for sarcoidosis being an infectious disease but rather an exaggerated immune response to “molecular patterns of killed and partially degraded mycobacterium and propionibacteria”.27

Although not proof of an infective cause, the following two situations appear suggestive of such. The original Kveim test, using lymph node extract, and the modified Kveim test (Siltzbach test), using spleen extract, produced localised granuloma formation in a majority of patients with sarcoidosis; neither form of this test is now used, due to potential risks from ‘any transmissible process’.15 In transplant patients, donor-acquired sarcoidosis is reported in recipients of tissues or organs from known sufferers of sarcoidosis.82–85 Recurrence of sarcoid has also been seen in (non-affected) transplanted lungs of recipients with sarcoidosis.86 Many tissue transplants have now been described to cause this phenomena, but of note, not all recipients develop sarcoidosis.87 This suggests that host susceptibility, and any presumed immunosuppressive therapy, may influence the development of sarcoid in the recipient.


Despite many advances in determining the genetic and environmental influences on the development of sarcoidosis, it remains an unusual disorder. Few diseases states combine geographic, seasonal, gender, and race specific distributions, with ‘clustering’ and familial features. Sarcoidosis can also present with ‘isolated’ or ‘systemic features’, and can be ‘asymptomatic’, ‘spontaneously resolve’ or prove ‘fatal’. The failure to identify a specific cause of sarcoidosis, and the varied presentations in different regions of the world, raise the likelihood that sarcoidosis is in fact a spectrum of disorders,41 ,88 with a common histology.

Sarcoid epidemiology clearly suggests a genetic role while investigations into environmental factors seem to rule out any ‘purely infective’ role. Undoubtedly, genetic susceptibility is a host requirement, permitting one of many proposed antigens to elicit an aberrant immune response.

Self assessment questions

  • Please answer true (T) or false (F) to the below,

    • Serum angiotensin-converting enzyme (ACE) values reliably diagnose sarcoidosis

    • The Kveim test is no longer used as part of the diagnosis of sarcoidosis

    • Sarcoidosis appears more common in women than men

    • A cardiac biopsy is universally positive in the presence of sarcoid cardiomyopathy

    • Sarcoid disease spares the central nervous system

    • Peripheral anergy is a feature of sarcoidosis

    • Lofgren syndrome and Heerfordt syndrome are forms of sarcoid disease

    • The majority of patients with sarcoidosis are likely to be symptomatic

    • Some of the highest rates of prevalence of sarcoidosis have been reported in Wales

    • Sarcoid disease tends to be more severe in Afro-Caribbean patients compared to those of European origin

    • In the USA, sarcoidosis is more common in those of European origin than in African Americans

    • Hypocalcaemia is a feature of sarcoid disease

    • The strongest genetic associations with sarcoidosis are found for human leucocyte antigen (HLA) class I and II antigens within the major histocompatibility complex region of chromosome 6

    • Some forms of sarcoid (eg, Lofgren syndrome) are associated with spontaneous remission

    • Sarcoidosis has a peak prevalence in the teenage years

    • Certain HLA types are associated with disease severity

    • Sarcoidosis has been shown to cluster among individuals working together

    • Cell wall deficient mycoplasma are the cause of sarcoidosis

    • Sarcoidosis is unique to human beings

    • Positron emission tomography scanning represents an important advance in identifying active disease

    • Familial clustering is a feature of sarcoid disease

    • Sarcoidosis is likely due to an interaction between genotype and environmental exposure

    • Cigarette smoking appears to be associated with an increase risk of sarcoidosis

    • Seasonal clustering in several countries occurs in the autumn and winter

    • Sarcoidosis can be transmitted via transplanted tissues

Current research questions

  • The development of a reliable method of diagnosing sarcoidosis, including biomarkers to determine disease presence and activity.

  • The development of an effective steroid-sparing therapy to treat sarcoidosis.

  • Establishing whether sarcoidosis is a collection of different disease states, all responding to different environmental stimuli, with a common end-point.

  • A focus on specific phenotypic subgroups of sarcoidosis in the determination of individual environmental causes.

Key references

  • Rybicki BA, Iannuzzi MC, Frederick MM, et al. Familial aggregation of sarcoidosis. A case-control etiological study of sarcoidosis (ACCESS). Am J Resp Crit Care Med 2001;164:2085–91.

  • Valeyre D, Prasse A, Nunes H, et al. Sarcoidosis. Lancet 2014;383:1155–67.

  • American Thoracic Society. Statement on sarcoidosis. Joint Statement of the American Thoracic Society (ATS), the European Respiratory Society (ERS) and the World Association of Sarcoidosis and Other Granulomatous Disorders (WASOG) adopted by the ATS board of Directors and by the ERS Executive Committee February 1999. Am J Resp Crit Care 1999;160:736–55.

  • Rybicki BA, Hirst K, Iyengar SK, et al. A sarcoidosis genetic linkage consortium: the sarcoidosis genetic analysis (SAGA) study. Sarcoidosis Vasc Diffuse Lung Dis 2005;22:115–22.

  • Rossman MD, Kreider ME. Lesson learned from ACCESS (a case controlled etiologic study of sarcoidosis). Proc Am Thorac Soc 2007;4:453–6.

Main messages

  • Sarcoidosis appears to have multiple influences on the aetiology

  • Sarcoidosis has a well recognised genetic propensity

  • Sarcoidosis is likely to represent a spectrum of disease states

  • Sarcoid disease can have a very variable prognosis

  • Sarcoidosis remains something of a clinical conundrum

  • Sarcoidosis has no specific biomarkers of disease presence or severity



  1. A(F); B(T); C(T); D(F); E(F)

  2. A(T); B(T); C(F); D(F); E(T)

  3. A(F); B(F); C(T); D(T); E(F)

  4. A(T); B(T); C(F); D(F); E(T)

  5. A(T); B(T); C(F); D(F); E(T)


View Abstract


  • Contributors All authors contributed to the writing of this manuscript with SS and SD formatting the figures. SD is guarantor of this article.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.