Chitotriosidase, secreted by activated macrophages, is a biomarker of activated macrophages. In this study, we explored whether chitotriosidase could be adopted as a biomarker to evaluate the curative effect on tuberculosis (TB). Five counties were randomly selected out of 122 counties/cities/districts in Hunan Province, China. Our cases were all TB patients who were newly diagnosed or had been receiving treatment at the Centers for Disease Control (CDCs) of these five counties between April and August in 2009. Healthy controls were selected from a community health facility in the Kaifu district of Changsha City after frequency-matching of gender and age with the cases. Chitotriosidase activity was evaluated by a fluorometric assay. Categorical variables were analysed with the χ2 test. Measurement data in multiple groups were tested with analysis of variance and least significant difference (LSD). Correlation between chitotriosidase activity and the degree of radiological extent (DRE) was examined by Spearman's rank correlation test. The average chitotriosidase activity levels of new TB cases, TB cases with different periods of treatment (<3, 3–6, >6 months) and the control group were 54·47, 34·77, 21·54, 12·73 and 10·53 nmol/h.ml, respectively. Chitotriosidase activity in TB patients declined along with the continuity of treatment. The chitotriosidase activity of both smear-positive and the smear-negative pulmonary TB patients decreased after 6 months' treatment to normal levels (P < 0·05). Moreover, chitotriosidase activity was positively correlated with DRE (r = 0·607, P < 0·001). Our results indicate that chitotriosidase might be a marker of TB treatment effects. However, further follow-up study of TB patients is needed in the future.