OBJECTIVE The excellent in vitro activity of fluoroquinolones against Mycobacterium tuberculosis has raised concerns about the delayed diagnosis of pulmonary tuberculosis (PTB) initially misdiagnosed as pneumonia. The aim of this study was to assess the effect of empiric fluoroquinolone therapy on delays in diagnosis in patients with PTB initially misdiagnosed as bacterial pneumonia.

DESIGN Patients with PTB initially treated as having community-acquired pneumonia and treated with fluoroquinolones for more than 5 consecutive days, from January 2000 to December 2004, were enrolled. As a control group, TB patients initially treated with nonfluoroquinolone antibiotics were enrolled. We reviewed the clinical data and compared treatment responses between the two groups.

RESULTS Nine patients in the fluoroquinolone group and 19 patients in the non-fluoroquinolone group were enrolled. In the fluoroquinolone group, eight patients (89%) improved clinically or radiographically, whereas only eight patients (42%) in the non-fluoroquinolone group improved (P = 0.04). The delay in initiation of anti-tuberculosis medication was longer in the fluoroquinolone group than in the non-fluoroquinolone group (43.1 - 40.0 vs. 18.7 +/- 16.9 days, P = 0.04).

CONCLUSION Delay in the initiation of anti-tuberculosis treatment is possible in patients administered fluoroquinolone and initially misdiagnosed as having bacterial pneumonia.