ReviewThe spectrum of opportunistic diseases complicating sarcoidosis
Introduction
Sarcoidosis is a systemic disease of unknown cause that is characterized by the formation of immune granulomas in various organs, mainly the lungs and the lymphatic system [1], [2]. The persistence of an as-yet unidentified poorly degradable antigen in genetically susceptible hosts is thought to trigger the typical Th1 cellular immune response, leading to formation of granulomas. Sarcoidosis is characterized by a paradoxical immune status. Indeed, the exaggerated immune response within granulomas contrasts with various immune defects that manifest by anergy to tuberculin test and by the possible occurrence of opportunistic infections (OIs).
Several situations can be observed. First, OIs may occur in untreated patients or reveal sarcoidosis. This is mainly described for cryptococcosis and progressive multifocal leukoencephalopathy (PML). Second, corticosteroid (CS) therapy [3], [4], [5] and the use of immunosuppressive agents or anti-tumor necrosis factor (TNF)-α [6] in severe or refractory sarcoidosis may, like in other systemic diseases, predispose patients to OIs. Third, OIs in sarcoidosis patients may only depend on epidemiologic or geographic factors. This is mostly true for tuberculosis and histoplasmosis. Finally, anatomic changes, such as parenchymal fibrocystic lesions in advanced pulmonary sarcoidosis, can predispose to aspergillosis.
Seven studies have focused on the risk of OI in the setting of sarcoidosis and have reported an incidence of 0–10% (Table 1) [4], [7], [8], [9], [10], [11], [12]. These studies were heterogeneous in terms of purpose, studied populations, sarcoidosis definition, and follow-up periods. Thus, it is not possible to conclude in an increased risk of OI during sarcoidosis.
Although the medical literature contains a substantial number of reports of OIs during sarcoidosis, only one focused review has been published in French, ten years ago [13]. Girard et al. reported on 65 cases of OIs complicating sarcoidosis, between 1966 and 2004. Cryptococcosis was the most frequently reported infection (59%) followed by mycobacterial infections (13%), nocardiosis (11%), histoplasmosis, pneumocystosis (9% each), and aspergillosis (7%). However, no newer reviews have been undertaken despite several new case series and case reports becoming increasingly available. Notably, two recent studies of our group have reported on several cases of cryptococcosis or PML complicating sarcoidosis [14], [15]. Mycobacterial infections are also increasingly reported. Given these data, we performed an exhaustive literature review with the aim of determining how sarcoidosis predisposes to OIs.
Section snippets
Search strategy
Searches were conducted in PubMed database (including Medline, National Library of Medicine, and PubMed Central), for the time period between January, 1974 and March, 2014, using strategies recommended by the Cochrane handbook [16]. The review strategies consisted of a exhaustive search using the terms “sarcoidosis”, “opportunistic infection”, “tuberculosis”, “mycobacteria”, “aspergillosis”, “cryptococcosis”, “nocardiosis”, “toxoplasmosis”, “mucormycosis”, “zygomycosis”, “histoplasmosis”,
Sarcoidosis
Sarcoidosis affects people from all racial/ethnic origins and occurs at any time of life. Its incidence is estimated at between 4.7 and 64/100,000 and its prevalence varies from 1.0 to 35.5/100,000 per year, with a particular proclivity for young adults [1], [18], [19], [20]. Diagnosis is established when compatible clinico-radiological features are supported by histological evidence of a non-caseating epithelioid granuloma [21]. Histological confirmation should be obtained whenever possible,
Tuberculosis and other mycobacteria
The role of mycobacteria has been repeatedly investigated as an etiological agent for sarcoidosis but, so far, no study has given indisputable results [79], [80], [81]. A number of small trials using anti-tuberculous drugs to treat sarcoidosis have also been published with negative results [82]. It now seems clear that sarcoidosis-associated immunodepression is not a major risk factor for mycobacterial infection. Epidemiological factors and treatments, mainly CSs and immunosuppressants, are
Epidemiology
The annual incidence of invasive cryptococcosis in the USA has been estimated at 1700–6600/100,000 patients with AIDS and at 0.2–0.9/100,000 patients without HIV [91]. CS therapy is the main risk factor for non-HIV-associated cryptococcosis (28% of cases) [92]. Cryptococcosis is very rare in the setting of autoimmune diseases, except in patients taking CSs, but most series report that cryptococcosis occurs in up to 30% of cases with no discernible underlying condition [91].
Recently, we have
Epidemiology
While PML can occur in patients with minimal or occult immunodepression, it is more likely to occur in immunocompromised patients [106], [107]. In AIDS, the incidence of PML was estimated at 3.3/100,000 patients-years before the introduction of HAART whereas it is now estimated at 1.3/100,000 patients-years [108]. In a review of 61 patients with PML, ~ 80% had AIDS and the majority of patients who develop PML had a CD4 T-cell count of < 200/mm3 [109]. Progressive ML has also been associated with
Chronic pulmonary aspergillosis (CPA) and other Aspergillus-related disease
Sarcoidosis represents a risk factor in 7–17% of CPA cases [127], [128]. Conversely, CPA complicates around 2% (range 0–17%) of sarcoidosis, with geographic differences [4], [127], [128], [129], [130], [131], [132]. In > 90% of cases, it presents as aspergilloma simplex complicating fibrocystic lung cavities and is considered as a saprophytic colonization rather than OI. Overall, CPA mostly complicates advanced fibrocystic sarcoidosis [133], [134]. Long-term CS therapy is a major risk factor for
Limitations
As available data on OIs complicating sarcoidosis arise from case series it was not possible to determine their true frequency in sarcoidosis patients compared to other groups at risk. Therefore, it remains unclear if sarcoidosis is associated with a greater risk of OIs. A proper control group would have to be at the same risk in terms of immunosuppression and other risk factors (e.g. fibrocystic disease) to determine if sarcoidosis truly places the patients at higher risk. For example,
Conclusions
Opportunistic infections seem to be rare in the setting of sarcoidosis. Although, we have collected information on > 280 cases, large reference cohorts are needed to assess the exact risk for their occurrence during sarcoidosis. Sixty to 82% of sarcoidosis patients were receiving CS therapy, regardless of the OI. In contrast, the mean blood CD4 rate did not seem to influence the risk of OI.
A substantial number of OIs occurred while the patients were not receiving any treatment. Together, these
Disclosures
The authors declare that they have no conflict of interest and no source of funding for this review.
Take-home message
- •
Although rare, opportunistic infections can occur during sarcoidosis, even in untreated patients.
References (150)
- et al.
Sarcoidosis
Lancet
(2014) - et al.
Diagnostic criteria for sarcoidosis
Autoimmun Rev
(2014) - et al.
Glucocorticoids and invasive fungal infections
Lancet
(2003) - et al.
Course and prognosis of sarcoidosis around the world
Am J Med
(1974) - et al.
Aspergillomas complicating sarcoidosis. A prospective study in 100 patients
Chest
(1984) - et al.
Infections opportunistes et sarcoïdose
Rev Mal Respir
(2004) - et al.
Methotrexate vs azathioprine in second-line therapy of sarcoidosis
Chest
(2013) - et al.
Pharmacotherapy for pulmonary sarcoidosis: a Delphi consensus study
Respir Med
(2010) - et al.
Efficacy and safety of TNF antagonists in sarcoidosis: data from the Spanish registry of biologics BIOBADASER and a systematic review
Semin Arthritis Rheum
(2012) - et al.
Granulomatosis and primary immunodeficiency in adulthood
Rev Med Interne
(2013)