This retrospective study included 41 patients with M. kansasii lung infection who attended two tertiary referral hospitals in Seoul, Korea, between January 1998 and December 2007: Samsung Medical Center and Asan Medical Center. All patients met the diagnostic criteria of NTM lung disease,1 and we reviewed their medical records and radiographs. This study was approved by the institutional review board of Samsung Medical Center and Asan Medical Center, both of which waived the requirement for informed consent of the individual patients due to the retrospective nature of the study.

Mycobacterial stains and cultures were performed using standard methods.18 The results of smear microscopy were reported semiquantitatively. A positive smear was defined as one with > 1 acid-fast bacilli (AFB) per 100 high-power fields. Respiratory specimens were decontaminated using N-acetyl-L-cysteine/2% NaOH, and processed specimens were plated onto 3% Ogawa medium. Inoculated tubes were incubated at 37℃ and then inspected weekly for 8 weeks. To distinguish between M. tuberculosis and NTM, all AFB isolates were assessed according to growth rates, colony morphology, and pigmentation, as well as with a commercial polymerase chain reaction-based assay system (MTB-ID; M&D, Wonju, Korea). NTM species were identified using a polymerase chain reaction and restriction fragment length polymorphism methods based on the rpoB gene, as described previously.14

The medical records of all study patients were reviewed, including information such as age, sex, body mass index (BMI), respiratory and constitutional symptoms, smoking history, underlying illness, pulmonary function tests, tuberculosis history, results for human immunodeficiency virus (HIV) antibody, and treatment history.

Chest radiographs and computed tomography (CT) images were those obtained closest in time to the date of the first positive respiratory culture result. The chest X-ray and CT images were evaluated in terms of the presence of nodules, consolidation, cavitation, and pleural effusion. In addition, hilar or mediastinal lymphadenopathy and bronchiectasis was evaluated on CT.

Nodules (≤ 3 cm in diameter) were considered present when a rounded opacity was observed, either well or poorly defined. Consolidation was defined as a homogeneous increase in pulmonary parenchymal attenuation that obscures the margins of vessels and airway walls. A cavity was regarded as present when a gas-filled space was seen within pulmonary consolidation, a mass, or a nodule. Bronchiectasis was defined as bronchial dilatation relative to the accompanying pulmonary artery, a lack of tapering of bronchi, and bronchi identified within 1 cm of the pleural surface.19

The distribution (upper vs. middle and lower lung zone) and laterality (unilateral vs. bilateral infiltrates) of the lung lesions were also analyzed. Lesions were considered to be in the upper zone of the lung if cephalad to the aortic arch, in the lower zone if caudad to the inferior pulmonary vein, and in the middle zone if between these two zones.20

Sputum conversion was defined as three consecutive negative cultures within 6 months, with the time of conversion defined as the date of the first negative culture. If the patient could not expectorate sputum during the treatment period, the sputum was considered to have converted to negative. Sputum relapse was defined as two consecutive positive cultures after sputum conversion.21

The baseline characteristics of the patients are shown in Table 1. The median age was 63 years [interquartile range (IQR) 51-75 years] and 33 (81%) were men. Twenty-three (56%) patients were current or ex-smokers. None of the patients were HIV positive. Comorbid conditions included previous pulmonary tuberculosis in 13 (32%) patients, chronic obstructive pulmonary disease in 13 (32%), underlying malignancy in nine (22%), and diabetes mellitus in two (5%). The most common presenting symptoms were cough (n = 27, 66%) and sputum (n = 26, 63%). Ten (24%) patients had dyspnea and four (10%) had hemoptysis.

The diagnosis was confirmed by obtaining an adequate number of positive sputum cultures (n = 37, 90%). Patients, who were either unable to produce sputum or had negative sputum cultures, were diagnosed using bronchial aspirates (n = 3, 7%) or lung tissues (n = 1, 2%). Sputum AFB smear examinations were positive in 11 patients (27%). Testing for rifampin susceptibility was performed on M. kansasii isolates recovered from 28 patients and 89% (25/28) were susceptible to rifampin.

All 41 patients had abnormal chest radiographic findings at the time of diagnosis (Table 2). The most common radiographic findings were nodules (n = 22, 54%) and consolidation (n = 22, 54%). Cavitation was present in 13 patients (32%) and pleural effusion in four (10%). Radiographic abnormalities were confined to the upper lung zones in 15 patients (37%), and 21 patients (51%) had bilateral disease.

A chest CT was available in 36 patients (88%). The most common CT finding was nodules (n = 25, 69%). Consolidation and cavitation were present in 16 patients each (44% each). Combined bronchiectasis was found in 12 patients (33%).

Ten (24%) patients did not receive anti-mycobacterial drugs after the diagnosis of M. kansasii disease. Of these, eight patients had mild symptoms and no significant change on their radiographs during the follow-up period. One patient was lost to the follow-up 1 month after the diagnosis. The upper lobar consolidation resolved spontaneously without antibiotic therapy in one patient (Table 3).

Thirty-one (76%) patients received antibiotic therapy including isoniazid, rifampin, and ethambutol. Of these, 16 patients completed the treatment and were followed for a median of 13 months (IQR 7-28 months) without relapse. Sputum relapse occurred after treatment completion in one patient who had no history of pulmonary tuberculosis and the isolate of M. kansasii was sensitive to rifampin. Four patients remained on the treatment as of the end of May 2009. Nine patients were lost to the follow-up after antibiotic therapy for a median of 6 months (IQR 2-10 months) and two patients died of causes unrelated to M. kansasii lung disease during antibiotic therapy (Table 3). Three patients had rifampin-resistant M. kansasii lung disease. One patient was lost to follow-up after 1.5 months of therapy with isoniazid, rifampin, and ethambutol before drug susceptibility results were available. Treatment regimens were changed to clarithromycin and levofloxacin in two patients after the confirmation of resistance to rifampin. However, one patient was lost to the follow-up and the other patient died of causes unrelated to M. kansasii lung disease during antibiotic therapy.

The median time for negative culture conversion was 1 month (IQR 1-3 months) in patients who achieved negative conversion. For 21 patients who received antibiotic therapy for more than 12 months, the median treatment duration was 16 months (IQR 9-18 months) and the 12-month culture conversion rate was 95% (20/21).