Data of 198 MDR-TB patients diagnosed from August 2013 to December 2015 were retrospectively collected. Among them, 10 patients refused to accept treatment, 3 patients died and 2 patients were lost to follow-up before treatment, the chemotherapy regimens and treatment outcomes of 43 patients were unknown, the baseline strain of 1 patient was confirmed as sensitive by the drug sensitivity test three times, the baseline strains of 7 patients were identified as non-tuberculous mycobacteria by 16S rRNA and hsp65 gene, and 16 patients received non-standardized multidrug-resistant tuberculosis regimens. Finally 116 patients received standardized regimens for MDR-TB. We recorded their baseline clinical characteristics, baseline drug susceptibility results, sputum culture conversion, clinical symptom and imaging changes. The primary endpoint was cure, treatment completed, treatment failure, death, or default (lost to follow-up) at the end of the treatment period.

Of 116 patients enrolled into the standardized regimens for MDR-TB, the number of new patients and previously treated patients were 39 (33.6%) and 77 (66.4%), respectively. There were 83 male patients accounting for 71.6% and 33 female patients accounting for 28.4%. The average age was (48.7±12.6) years old and the age ranged from 15 to 77 years old. Of the 116 MDR-TB patients, 59 (50.9%) were cured or completed treatment, 19 (16.4%) experienced treatment failure, 5 (4.3%) died, and 33 (28.4%) were lost to follow-up. The success rate of new patients was higher than that of previously treated patients (69.2% vs 41.6%), and the difference was statistically significant (χ ²=7.932, P=0.005). The success rate of MDR-TB (67.7%, 21/31) was higher than that of pre-XDR-TB (42.6%, 20/47) and XDR-TB (15.4%, 2/13), the differences were statistically significant (χ ²=4.753,P=0.029 and χ ²=10.064,P=0.002). The loss rate of MDR-TB (25.8%, 8/31) was respectively compared with that of pre-XDR-TB (27.7%, 13/47) and XDR-TB (23.1%, 3/13), while there was no statistically significant difference among them (χ ²=0.033, P=0.857; χ ²=0.036, P=0.849). Of the 91 patients with baseline second-line drug susceptibility, 34.1% (31/91), 51.6% (47/91) and 14.3% (13/91), respectively, was simple MDR-TB, pre-XDR-TB, and XDR-TB. And the resistant proportion of ofloxacin accounted for 63.7% (58/91). Simple MDR-TB (87.1%, 27/31) was higher than the pre- XDR-TB (66.0%, 31/47) and XDR-TB (38.5%, 5/13) in terms of sputum culture conversion rates, the difference were statistically significant (log-rank trend test, χ ²=14.500, P<0.001).

The low success rates of MDR-TB treatment in Wenzhou may be related to the high proportion of Pre-XDR-TB and XDR-TB and the high default rates.