The four research sites, where a comprehensive China-Gates Collaborative TB Project was implemented, were involved in this study. During the project implementation period (from January 1, 2011 to December 31, 2011), a total of 50 patients, who were confirmed to have RMR or RPR pulmonary TB and met the recruitment criteria, were enrolled and the relevant information were collected from those patients. And then based on the collected information, 14 patients were excluded due to incompletely information on treatment, such as lost during treatment or unevaluated or failed on forming an effective treatment regimen, while the remaining 36 patients’ clinical data and information were analyzed. Among them, 11 patients (30.56%) were treated with isoniazid which was added in their treatment regimens after 3 or 4 months of treatment (3-6 Am-Lfx-H-PZA-EMB/9-12 Lfx-H-PZA-EMB), which was called “isoniazid group”; 25 cases (69.44%) were not supplemented with isoniazid (6 PZA-Am (Cm)-Lfx(Mfx)-PAS(EMB)-Pto/18 PZA-Lfx(Mfx)-PAS(EMB)-Pto), which was called “standard group”. Descriptive analysis was used for the patients’ basic information. The categorical variables were studied by using Fisher exact test, α=0.05.

The incidence of adverse drug reactions in the isoniazid group (54.55% (6/11)) was lower than that in the standard group (60.00% (15/25)), but there was no statistically significant difference between the two groups (Fisher exact test, P=1.000). The treatment success rate in the isoniazid group and the standard group was 90.91% (10/11) and 48.00% (12/25)) respectively; the treatment failure rate was 0.00% (0/11) and 44.00% (11/25) respectively in the isoniazid group and the standard group; the mortality was 9.09% (1/11) in the isoniazid group and 8.00% (2/25) in the standard group. The difference of treatment success rate and failure rate between the two groups was statistically significant (Fisher exact test, P=0.025, P=0.015).

By adding isoniazid into the patients’ treatment regimens, the number of adverse drug reactions of the patients does not increase, and the treatment outcomes of the patients with isoniazid is better than that of patients without isoniazid. In order to improve the treatment outcomes of the patients, it is suggested that isoniazid should be timely added into the patients’ treatment regimens if isoniazid resistance is not confirmed or if the patients are determined to have RMR-TB/RPR-TB.