From mycobacteria to sarcoidosis--is the gate still open?

ciation between mycobacteria and sarcoidosis, and it has been hypothesized that sarcoidosis could be a separate manifestation of infection with mycobacteria. Therefore, most studies of a possible causal microoganism have focused on mycobacteria. Mycobacteria are ubiquitous all over the world with reservoirs in humans, animals, water and soil. Besides the usual forms, mycobacteria may exist in other disguises, such as cell wall-defective L forms. However, there is still no defi nite evidence for a causal relationship between mycobacterial infection and sarcoidosis. Acute/subacute pulmonary sarcoidosis has features in common with a disease caused by inhalation of a precipitating antigen. In the initial phase of the disease, pulmonary sarcoidosis probably spreads from the bronchioli/alveoli through the lymphatic vessels to the hilar and mediastinal lymph nodes. However, serum antibody profi les (IgG and IgA) against Mycobacterium tuberculosis in patients with sarcoidosis and healthy control subjects demonstrate no consistent differences [2] , and prolonged culture for mycobacteria from mediastinal lymph nodes in patients with sarcoidosis has been negative and did not support the hypothesis that mycobacteria could be directly involved in the pathogenesis of sarcoidosis [3] . Novel biomolecular techniques using identifi cation of mycobacterial DNA with PCR have reopened the gate for a mycobacterial pathogenesis. The paper by Fité et al. [4] in this issue of Respiration is the latest of more than 20 reports within the last 13 years on the prevalence of M. tuberculosis DNA in biopsy specimens from sarcoidosis patients. Most of these studies have been performed in Despite more than a century of searching, the aetiology of sarcoidosis is still hidden in the veils of mystery. Sarcoidosis most frequently involves the lung, so in the majority of patients, the lungs are the principal target organ. The pathogenetic hallmark of sarcoidosis is granulomatous infl ammation with the formation of non-caseating epithelioid cell granulomas, which may be triggered by persistent presentation of poorly degradable antigen(s). It appears that clinical, overt disease is the result of an exposition to the unknown antigen(s) together with a genetic predisposition [1] . Sarcoidosis has many features in common with infectious diseases, and for decades, researchers have been looking for microorganisms which might be involved in the pathogenesis. Many known infectious agents can induce granuloma formation, including bacteria, fungi, protozoans and helminths. In contrast, the antigens involved in the pathogenesis of sarcoidosis, Crohn’s disease, biliary cirrhosis and Wegener’s granulomatosis are unknown. For many years, sarcoidosis was thought to be tuberculosis, or a variant hereof. Even the ‘fathers’ of sarcoidosis, Boeck and Schaumann, held this view. Clearly, however, by virtue of its clinical presentation, the absence of a demonstrable infectious agent, spontaneous improvement and response to corticosteroid therapy, as well as the absence of a positive tuberculin skin test and response to antituberculous medications, sarcoidosis is not classic tuberculosis. The histological similarity between sarcoidosis and tuberculosis (with epithelioid cell granulomas as the typical common fi nding) has stimulated the search for an asso-