Abstract: Objective To observe the efficacy of antituberculosis regimens containing different dosage of levofloxacin in the treatment of multidrug-resistant tuberculosis(MDR-TB), and to guide to choose rational dosage of levofloxacin (Lfx) in clinical practice.  Methods One hundred and twenty-five inpatients and outpatients with MDR-TB treated from Jan. 2011 to Nov. 2012 in Beijing Chest Hospital, Capital Medical University were enrolled, and all patients were randomly divided into 3 groups with different dosage of Lfx (400 mg/d-control group, 600 mg/d and 800 mg/d) combined with other anti-tuberculosis agents. Plasma samples were taken to detect plasma drug concentrations at 2 hours after taking Lfx at 7 days’ treatment. The clinical efficacy (sputum negative conversion rate) and adverse effects of the three groups were observed for 12 months. Statistical analysis was performed using the Statistical Package for the Social Sciences(SPSS) for Windows(version 13.0)．Theχ² test was used in the comparison of categorical variables．The Kruskal-Wallis H test was applied in the mean comparison among many groups. P<0.05 was considered statistically significant. Results The rates of sputum conversion for the group of 600 mg/d Lfx at the end of 3, 6, 9 and 12 months were 53.8% (21/39), 56.4% (22/39), 59.0% (23/39) and 59.0% (23/39), which were significantly higher than those of the group of 400 mg/d Lfx (30.0% (12/40), 32.5% (13/40), 35.0% (14/40) and 35.0% (14/40)), respectively. The differences were significant statistically (χ²=4.60, 4.57, 4.95 and 4.95，P<0.05). The rates of sputum conversion of the group of 800 mg/d Lfx (60.0%(24/40), 65.0%(26/40), 65.0%(26/40) and 65.0% (26/40)) were also significant higher than those of the group of 400 mg/d Lfx at the different time of treatment. The differences were also significant statistically (χ²=7.27, 8.46, 7.20 and 7.20, P<0.05). However, the differences of the rates of sputum conversion between the group of 800 mg/d Lfx and 600 mg/d Lfx were not significant with statistics (χ²=0.31, 0.61, 0.21 and 0.21, P>0.05). Gastrointestinal reaction, muscle and joint pain and central nervous system abnormalities were the main performance as adverse reaction in the three groups. The rates of the main adverse reactions were 0.0%, 15.0% (6/40) and 0.0% for the group of 400 mg/d Lfx, 2.6% (1/39), 10.3% (4/39) and 2.6%(1/39) for the group of 600 mg/d Lfx, 5.0% (2/40), 17.5% (7/40) and 7.5%(3/40) for the group of 800 mg/d Lfx. There were no significant differences in adverse reactions among the three groups (χ²=2.04, 0.87 and 3.58, P>0.05).  Conclusion 600 mg/d and 800 mg/d Lfx were better for MDR-TB treatment, but the difference was not statistically significant, so 600 mg/d is recommended for MDR-TB treatment and 800 mg/d can be applied for some cases.

Key words: Tuberculosis, pulmonary/drug therapy, Tuberculosis, multidrug-resistant/drug therapy, Ofloxacin, Treatment outcome