[Pathogenicity of Mycobacterium kansasii].

BACKGROUND

In a recent prospective study on pulmonary infections with non-tuberculous mycobacteria (NTM) led by the WATL group, disease rates in patients with M. kansasii infection were found to be 100 %. In the present study we re-evaluated the pathogenicity of M. kansasii infections in a large lung diseases treatment center in Berlin (Lungenklinik Heckeshorn).

METHODS

All patients in whose respiratory specimen cultures M. kansasii was detected between January 2003 and June 2013 were included. The 2007 ATS diagnostic criteria were applied to differentiate disease from asymptomatic infection. The strains were further investigated by sequencing of the 16S-23S rDNA internal transcribed spacer (ITS) region.

RESULTS

We evaluated 43 consecutive cases. Complete patient data were available in 38 cases. In one patient, no culture results were obtained, in 37 patients M. kansasii was isolated and patient data could be retrieved. In 25/37 patients (68 %) clinical disease was present so that a specific treatment was initiated (underlying diseases were COPD in 8/25 (32 %), bronchiectasis in 5/25 (20 %), TB scar or scar due to prior chest surgery in 3/25 (12 %) and alcohol abuse in 4/25 (16 %)). Twelve out of 37 patients (32 %) were found to be colonized or asymptomatically infected (underlying diseases were COPD in 7/12 (58 %), bronchiectasis in 3/12 (25 %) and TB scar or scar due to prior chest surgery in 3/12 (25 %)). Sequencing results identified 30 strains as genotype I, and 2 strains as genotype II. In 22/30 cases (73 %) genotype I was considered pathogenic.

CONCLUSIONS

In our cohort, we could not confirm the high M. kansasii pathogenicity of 100 % found in a previous multi-center study; we therefore support the clinical and semiquantitative microbiologic diagnostic criteria also for infection with M. kansasii.