Selection of cases

Sources of cases: Cases were randomly selected with a two-step stratified sampling method. First, five counties out of 122 counties/cities/districts in Hunan Province were randomly selected using a random number table (e.g. Qidong county, Yueyanglou district, Yueyang county, Zixing city, Hongjiang city). According to the TB management regulation, all suspected TB patients would be transferred to local Centers for Disease Control (CDCs) for confirmation based on WHO diagnostic criteria [15]. Once confirmed, the patients would receive treatment at the CDCs free of charge. Therefore, we identified cases from local CDCs. All patients who were newly diagnosed or had been receiving TB treatment at the CDCs of these five counties between April and August 2009 were selected. Patients with HIV/TB co-infection or other co-infections including fungi and nematodes were excluded because these co-infections could enhance serum chitotriosidase activity. Patients who stopped receiving medications or were previously treated were also excluded.

Selection of healthy controls

Controls were also selected using the stratified sampling strategy. First, one from the 14 community health service centres (e.g. that in Xingang community) in Kaifu district, Changsha city was selected using the random number table. Next, one from the six community health service stations (e.g. Xin'ansi) managed by Xingang Community Health Service Centre was randomly selected. Next, controls were matched with cases using a frequency-matching method. Because the ratio of male to female TB patients was about 2·5:1 in Hunan [Reference Chen16], the healthy controls were selected from permanent residents in Xin'ansi community by a gender-age frequency-matching method. All controls were confirmed free from active TB.

We also excluded participants (both cases and controls) who were not Han nationality.

Estimation of sample size

Sample size was calculated with:

${n_{ij}} = \displaystyle{{{{({Z_{a/2}} + {Z_{1 - \beta}} )}^2} \times (\sigma _2^2 + \sigma _2^2 )} \over {\delta _{ij}^2}}. $

The standard deviation (s.d.) σ in controls and the >6-month treatment group was estimated to be 4·2 [Reference Cakir14]; assuming a difference between these two groups of δ = 2, α = 0·05 (two-sided), and β = 0·10, the estimated sample size was 93, i.e. at least 93 participants of each group eligible for inclusion.

The protocol of our study was approved by the Ethics Review Committee of Central South University. A written informed consent form was signed by all participants. After that, 5 ml of blood was collected through venepuncture using an EDTA-containing tube. Plasma and packed cells were separated by centrifugation at 2500 g for 10 min and stored at −80 °C until used. In addition, all TB patients underwent a chest X-ray examination to determine the degree of radiological extent (DRE). The healthy controls received free physical examinations, including chest X-ray.

Determination of DRE

DRE was determined through chest X-rays by two chest physicians independently. In case of inconsistent interpretations, a final decision was made by consensus after the physicians evaluated the chest X-ray together. A previously described scoring system with modification was used [Reference Cakir14, Reference Deniz17]. Briefly, in each chest X-ray radiograph, each lung was divided into three zones: the middle zone was set between the levels of the second and fourth ribs at the costosternal junction, the upper zone was above the middle zone, and the lower zone was below the middle zone. A chest X-ray score was determined by visually estimating the extent of the opacities attributable to TB (e.g. nodular infiltration, consolidation, atelectasis, or linear density) and separately evaluating the existence of cavitation. The scoring was based on the percentage of opacities in each lung zone (PER) and the total diameter of the cavities (D). The scoring system was set as follows:

• Score 0: no abnormality;

• Score 1: PER of <25% without obvious cavitation;

• Score 2: PER of 25–50% without obvious cavitation, or PER <25% and D < 2 cm;

• Score 3: PER of 50–75% without obvious cavitation, or PER <50% and D = 2–4 cm;

• Score 4: PER >75%, or any cavitary disease with D > 4 cm.

Treatment regimen

The medications used for TB treatment included isoniazid (H), rifampicin (R), pyrazinamide (Z), ethambutol (E), thioacetazone (T), and streptomycin (S). The treatment regimens used in Hunan in 2009 were based on the WHO-recommended regimen [15]. Specifically, the smear-positive patients were treated with 3 times/week HRZE for the first 2 months and 3 times/week HR for 4 months (2H3R3Z3E3/4H3R3). The smear-negative patients were treated similarly except that E was not used (2H3R3Z3/4H3R3).

Detection of chitotriosidase activity

We adopted a published method for measurement of chitotriosidase activity [Reference Hollak18]. Specifically, 5 μl of EDTA plasma was incubated with 100 μl of 0·022 mm 4-methylumbelliferyl-fl-d-NN,N'-triacetylchitotriose (4 MU-chitotrioside; Sigma Chemical Co., USA) as substrate in a citrate/phosphate buffer (0·1/0·2 m), pH 5·2, at 37 °C. After 60 min, the reaction was stopped with addition of 0·3 m glycine/NaOH buffer (2 ml, pH 10·6). The fluorescence intensity was tested at an excitation wavelength of 360 nm and emission wavelength of 450 nm. The unit of chitotriosidase activity is ‘nmol/h.ml’.

Statistical analysis

Data were analysed with SPSS v. 19.0 (IBM Corporation, USA). Categorical variables were analysed by χ ² test. Measurement data in multiple groups were tested with analysis of variance (ANOVA) and least significant difference (LSD). Correlation between chitotriosidase activity and DRE was examined by Spearman's rank correlation test, with the correlation coefficient expressed as r. All tests of hypothesis were two-tailed with a type-1 error at 5%.

Ethical standards

The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.