This is the first identified case of Mycobacterium bovis bacillus Calmette-Guérin (BCG)-derived cutaneous tuberculosis that localizes at a place different from the vaccination site in hosts without immune deficiency. A healthy baby with a developing abscess is described. A multiplex PCR identified the abscess as originating from M. bovis BCG Tokyo 172.
CASE REPORTA 6-month-old girl with no previous significant medical problems developed an abscess in her left humerus. She had been vaccinated with Mycobacterium bovis bacillus CalmetteGuérin (BCG) Tokyo 172 at the age of 4 months. At 5 months, a red and swelling nodule appeared in her left humerus at a place different from the vaccination site. At 6 months, she visited our hospital. A red, fluctuating, and swollen abscess about 2 cm from the region of the vaccination was found. The abscess measured 10 by 15 mm. We found no swelling of the lymph nodes. The patient was afebrile and did not have respiratory symptoms. There was no abnormal chest radiographic finding. Skin testing with purified protein derivative tuberculin was positive with 10-by 7-mm erythema, without an induration or surrounding erythema. There was no family history of a similar reaction to BCG vaccination. The history of the girl's grandmother revealed pulmonary tuberculosis in her twenties. In addition, the possibility of close contact with other tuberculosis patients could not be excluded.We punctured the abscess and submitted the pus for conventional microbial culture, mycobacterial culture, amplification of mycobacterial nucleic acid using an AMPLICOR MTB microwell plate assay kit (Roche Diagnostic Systems, Branchburg, N.J.), and microscopic examination of a smear. Conventional microbial culture failed to detect any significant pathogens. The smear examination was negative, but amplification of nucleic acid using a primer for a Mycobacterium tuberculosis complex-specific region of the 16S rRNA gene (KY172-T3) was positive for M. tuberculosis complexes. A 4-week mycobacterial culture grew an M. tuberculosis complex colony. This colony showed sensitivity to isoniazid, rifampin, ethambutol, and streptomycin and resistance to pyrazinamide, suggesting that the colony may have derived from BCG Tokyo 172. However, we could not clearly distinguish the origin among M.
tuberculosis complex strains (M. tuberculosis, M. bovis BCG, or other strains of M. bovis).For the purpose of quick and accurate identification of the sample, we ran multiplex PCR on the sample by a simple modification of the method of Bedwell et al. (1). In 1999, a DNA microarray clarified that 16 regions in BCG strains, named RD1 to RD16, were different from those in M. tuberculosis H37Rv and that some regions differed among BCG strains (2). RD1 is not present in all M. bovis BCG strains, but it is present in other strains of the M. tuberculosis complex. For M. bovis BCG strains, Talbot et al. designed primers to amplify the complement sequence of RD1 (9). Another PCR procedure based on the intergenic region named SenX3-RegX3 was reporte...