Pulmonary Mycobacterium kansasii Infection in Israel, 1999-2004: Materials and Methods
The study sample included 56 patients with a diagnosis of M kansasii lung infection who attended the tuberculosis centers in Tel Aviv and Rehovot, Israel, from April 1999 to April 2004. The diagnosis was based on the guidelines of the ATS, namely appropriate symptomatology, compatible radiographic abnormalities, and multiple culture-positive respiratory specimens for M kansasii.15 Patients in whom there was a high clinical suspicion of tuberculosis but negative sputum smear results underwent bronchoscopy with BAL and transbronchial biopsy to confirm the diagnosis.
Mycobacterial cultures were performed with standard meth-ods. Sputum smears were stained with auramine and examined using fluorescence microscopy, and the presence of acid-fast organisms was confirmed with Ziehl-Neelsen stain. Lowenstein-Jensen slopes were incubated for 12 weeks. All mycobacterial isolates were sent to the Public Health Laboratory Service, Mycobacterial Reference Unit in Tel Aviv for identification and sensitivity testing.
M kansasii strains were routinely tested for susceptibility to rifampicin, ethambutol, clarithromycin, ciprofloxacin, ofloxacin, capreomycin, ethionamide, and cycloserine. We did not test the M kansasii strains for susceptibility to isoniazid.
The resistance ratio method was used to test the susceptibility strains to all the drugs mentioned. This method is based on the growth of the tested strain relative to a standard sensitive (control) strain at five standard doubled drug concentrations. We calculated the resistance ratio of each test strain to each drug by dividing the minimum inhibitory concentration of the test by the model minimum inhibitory concentration. As doubling dilutions are used, the resistance ratio is 1 (or less), 2, 3, 4, or 8. Strains with a resistance ratio of 1 or 2 are reported as susceptible, those with a resistance ratio of 4 are resistant, and those with a resistance ratio of 8 are highly resistant.
Molecular characterization was carried out on 20 isolates. For genotypic identification of M kansasii, we used the Accuprobe test (Gen-Probe; San Diego, CA), based on the hybridization of a DNA sequence, labeled with chemoluminescent substance, with the ribosomal RNA of a grown mycobacteria. All our patients were treated with rifampicin (600 mg), ethambutol (25 mg/kg for the first 2 months, then 15 mg/kg), and clarithromycin (1,000 mg/d) administered daily for at least 12 months of negative sputum culture results.