[A comparison of chest radiographs between patients with pulmonary Mycobacterium kansasii infection and those with Mycobacterium tuberculosis infection in the initial stage of disease].

OBJECTIVES: To elucidate the differences in affected lung segments between patients with pulmonary M. kansasii infection and those with M. tuberculosis infection in the initial stage of disease, we examined chest radiography images and CT scans. The initial stage of disease was defined as the period when less than one-sixth of the total lung area was affected by the infection, as visualized on chest radiography and CT. SUBJECTS AND METHODS: One hundred eighty-four patients were diagnosed with M.kansasii infection between 1996 and 2010 and 835 patients, with M.tuberculosis infection between 2008 and 2009 at our hospital. The diagnosis was made on the basis of the results of sputum culture and/or bronchial washing. After excluding the patients with underlying lung diseases such as chronic pulmonary emphysema, interstitial pneumonia, and old pulmonary tuberculosis as well as those in advanced stages, 24 patients with M. kansasii infection and 62 patients with M. tuberculosis infection were included in this study. The affected segments of the lungs and the rates of cavity development were determined by using CT scans. RESULTS: In patients with M.kansasii, 17 had an infected right lung, while 7 had an infected left lung. Additionally, in patients with M.tuberculosis, 58 had an infected right lung, 3 had an infected left lung, and 1 had a bilateral infection. In patients infected with M. kansasii, the upper lobes were affected in 22 cases and the lower lobes in 3 cases. In patients infected with M. tuberculosis, the upper, middle, and lower lobes and the lingular segment were affected in 41, 8, 24, and 1 cases, respectively. Upper lobe lesions were seen more frequently in patients with M. kansasii infection than in those with M. tuberculosis infection (p < 0.05). Cavity formation was identified more frequently in patients infected with M. kansasii (91.7%) than in those infected with M. tuberculosis (32.3%) (p < 0.001). Cavitary lesions were more frequently localized to the apical, posterior, and apico-posterior regions (S1, S2 or S1 +2) of the upper lobes in patients infected with M. kansasii (86.4%) than in those infected with M. tuberculosis (35%) (p < 0.001). A solitary lesion without endobronchial spread, which is characterized by centrilobular micronodules and tree-in-bud appearance, was more frequently demonstrated in patients infected with M.ka nsasii (45.8%) than in those infected with M. tuberculosis (6.5%) (p < 0.001). CONCLUSION: Our study revealed that the apical, posterior, and apico-posterior regions of the upper lobes are vulnerable to infection by not only M.tu berculosis, but also M.ka nsasii. It is likely that M.ka nsasii might gain access to these regions via the airways and that its weak virulence may lead to higher localization.

[A case of tuberculous pleurisy developing contralateral effusion during anti-tuberculosis chemotherapy].

A 55-year-old woman was admitted to our hospital because of chest pain, fever, and right pleural effusion that was exudative and lymphocyte-dominant with a high level of adenosine deaminase (ADA). Since her blood QuantiFERON-TB 3G test (QFT) was positive, she was diagnosed with tuberculous pleurisy. After initiation of anti-tuberculosis chemotherapy with isoniazid, rifampicin, ethambutol, and pyrazinamide, her symptoms improved. Later, liquid culture of the pleural effusion turned positive for Mycobacterium tuberculosis. On the 18th day of treatment, her chest X-ray and computed tomography exhibited pleural effusion in a moderate amount in the left thorax, with subsiding pleural effusion in the right thorax. Thoracocentesis demonstrated that the left thorax effusion was also exudative and lymphocyte-dominant, with elevated QFT response and high ADA concentration, suggesting tuberculous pleurisy. Mycobacterium tuberculosis was detected in the culture of a left pleural biopsy specimen obtained by thoracoscopy. We assumed that the left pleural effusion was due to paradoxical worsening because (1) on admission no effusion or lung parenchymal lesion was detected in the left hemithorax, (2) on the 14th day of treatment she was afebrile without pleural effusion on both sides, and (3) the bacilli were sensitive to the drugs she had been taking regularly. We performed drainage of the left effusion and continued the same anti-tuberculosis drugs, which led to the elimination of all her symptoms and of the pleural effusion on both sides. In conclusion, paradoxical worsening should be included in the differential diagnosis when contralateral pleural effusion is detected during the treatment of tuberculosis.