Abstract: Objective To analyze the status and epidemic trend of anti-tuberculosis (TB) drug resistance in Shanghai from 2007—2012.  Methods 12221 new registered and sputum culture-positive cases in Shanghai during 2007—2012 were enrolled. Isolates obtained from each case were tested for susceptibility to first-line anti-TB drugs (INH, RFP, S and EMB) and species identification. The drug susceptibility testing result and demographic data of 10979 tuberculosis strains were used to analyze the status and trend of anti-TB drug resistance. Logistic regression model was used to determine the risk factors of drug resistance and multidrug-resistance.  Results The overall drug resistance rate was 23.0% (2522/10979), the total multidrug-resistance rate was 5.3% (587/10979), the initial drug and multidrug-resistance rate were 20.4% (1981/9702) and 3.4% (328/9702), and the acquired drug and multidrug-resistance rate were 42.4% (541/1277) and 20.3% (259/1277), respectively. There was no significant increase in total drug resistance (χ²=2.38, P>0.05), acquired drug resistance (χ²=0.23, P>0.05), total multidrug-resistance (χ²=0.33, P>0.05), initial multidrug-resistance (χ²=1.67, P>0.05) and acquired multidrug-resistance (χ²=0.48, P>0.05) in recent 6 years. The initial drug resistance rate significantly increased during 2007—2012 (χ²=9.02, P<0.05). The multivariate logistic regression analysis showed that the history of previous treatment for TB (OR=3.080, 95%CI＝2.717-3.492, P<0.05 and OR=7.781, 95%CI＝6.481-9.341, P<0.05) and age(41~≤60) (OR=1.270, 95%CI＝1.045-1.544, P<0.05 and OR=1.669, 95%CI＝1.064-2.620, P<0.05) were significantly associated with any drug and multidrug-resistance.  Conclusion There was no significant increase in total drug resistance rate and total multidrug-resistance rate. The history of previous treatment for TB is significantly associated with any drug and multidrug-resistance.

Key words: Tuberculosis, Drug resistance, bacterial, Shanghai city