RESUMEN
Sarcoidosis is a multisystemic granulomatous disease of uncertain etiology. Recently, mycobacterial DNA especially Mycobacterium tuberculosis and Mycobacterium avium complex were detected in lung tissue and bronchial lavage fluid from patients with sarcoidosis by polymerase chain reaction (PCR) assays in 30% to 50% cases. Moreover, cell wall-defective form (CWDF) acid-fast bacteria have been isolated from skin lesions of patients with sarcoidosis which were later confirmed as M. avium complex by PCR assays. CWDF acid-fast bacteria were also found to grow from the blood of 95% patients with active sarcoidosis demonstrating a mycobacterial origin similar to M. tuberculosis. In view of these reports, we investigated 20 cases of cutaneous sarcoidosis using PCR/restriction enzyme pattern analysis (PCR/REPA) to detect mycobacterial DNA from paraffin-embedded skin biopsy samples. The method involves restriction enzyme analysis of nested PCR products obtained with primers encoding for the 65-KDa protein common to all mycobacteria. Using three restriction enzymes, the mycobacterial DNA from PCR product was differentiated to the species level. All the 20 cases had clinical and histologic evidence of sarcoidosis. Special stains for fungi (PAS) and mycobacteria (Fite) were negative and no foreign body was identified on polaroscopic examination in any of the cases. The cell lysates of M. tuberculosis, Mycobacterium bovis, Mycobacterium avium-intracellulare, Mycobacterium kansasii and Mycobacterium marinum from Centers for Disease Control (CDC) were used as standard control for PCR/REPA. Eight cases of foreign body granuloma, seven normal skin samples from the margin of surgical excisions and 5 cases of dermatitis were used as negative controls, and 4 cases of cutaneous tuberculosis were used as positive controls. Mycobacterial DNA was detected by PCR in 16 of the 20 cases of sarcoidosis. PCR/REPA subtyped 8 of these to M. tuberculosis complex (2 cases), M. avium-intracellulare (4 cases), M. kansasii (2 cases) while the other 8 cases were non-tuberculous mycobacteria. All four cases of cutaneous tuberculosis were positive by PCR and had a typical M. tuberculosis PCR/REPA pattern. Mycobacterial DNA was not detected in any of the negative controls. Our results demonstrated that mycobacterial DNA is present in 80% of cutaneous lesions of sarcoidosis and these mycobacteria may play a role in the pathogenesis of sarcoidosis.