Tuberculosis Transmission and Risk Factors in a Chinese Antimony Mining Community

INT J TUBERC LUNG DIS 20(1):57–62 Q 2016 The Union http://dx.doi.org/10.5588/ijtld.15.0215

Tuberculosis transmission and risk factors in a Chinese antimony mining community

K-S. Chen,*† T. Liu,† R-R. Lin,‡ Y-P. Peng,§ G-C. Xiong§ *Key Laboratory of Medical Molecular Virology, Fudan University, Shanghai, †Department of Clinical Laboratory, First Affiliated Hospital of Nanchang University, Nanchang, ‡Department of Respiration, Wuning County People’s Hospital, Wuning, §Department of Clinical Laboratory, Jiangxi Provincial Chest Hospital, Nanchang, China

SUMMARY

SETTING: An antimony mine in Jiangxi Province, into 35 clusters. Estimated recent transmission of TB China. was 53.2% within the community. Patients who failed OBJECTIVE: To investigate the incidence of tuberculosis treatment were more likely to be in clusters (adjusted (TB) transmission and associated risk factors in a odds ratio [aOR] 0.03, 95%CI 2.12–6.89). Patients with Chinese antimony mining community. multiresistant isolates were more likely to have failed DESIGN: Retrospective cohort study treatment and to be in a cluster than those carrying a METHODS: The 15-locus mycobacterial interspersed susceptible strain (aOR 0.001, 95%CI 4.89–29.7). repetitive unit-variable number of tandem repeats CONCLUSIONS: Individuals who fail treatment are an

(MIRU-VNTR15-China) method was used to determine important source of infection in TB transmission, and clustering of patients. A region of difference (RD105) multiresistant isolates are mostly responsible for this. TB deletion-targeted multiplex polymerase chain reaction control plans need to focus on treatment failure cases in was adopted to identify Beijing strains. Risk factors for the community. clustering were assessed. KEY WORDS: TB transmission; antimony; China RESULTS: Of 263 TB patients, 175 were distributed

IN 2013, AN ESTIMATED 9 MILLION new cases of Jiangxi, located in Jiujiang, covers approximately tuberculosis (TB) were reported globally, more than 30 km2 and employs several thousand workers. half of whom (56%) are in South-East Asia and the Nearby farmers comprise the majority of miners Western Pacific; China accounts for 11% of total engaged in long-term antimony mining. Antimony cases.1 Concentrated efforts in China to control TB ore is usually processed in crowded, enclosed spaces transmission resulted in a sharp decrease in TB with a high density of dust, creating conditions 2 incidence between 1990 and 2010. Anti-tuberculosis favourable to the spread of TB.6 However, there are treatment has been less successful, however, in no publicly available data on TB transmission in the individuals exposed to silica in their work environ- antimony mining community. ments.3,4 As a result, they pose a challenge to TB The use of mycobacterial interspersed repetitive control. Antimony mining constitutes a setting in unit-variable number of tandem repeats (MIRU- which conditions might be expected to facilitate the VNTR) techniques have allowed a better under- transmission of TB. standing of TB transmission. When patients share Antimony (Stibium) is a valuable metal, with wide military and material application. Metallic the same MIRU-VNTR profile it is assumed that antimony is extracted from antimony ore, and recent transmission has occurred, while those with 7,8 during the subsequent smelting, workers are ex- unique isolates represent reactivation. In this posed to significant amounts of silica. China study, we adopted the MIRU-VNTR15-China method possesses the world’s largest deposits of antimony to identify potential epidemiological links. We also ore, and is the largest producer of refined antimo- collected and identified risk factors associated with ny.5 Antimony mines are distributed throughout such transmission, which provide a theoretical Guizhou, Hunan and Jiangxi Provinces and employ basis for future TB control activities in this nearly 100 000 people. The antimony mine in community.

Correspondence to: Kaisen Chen, Department of Clinical Laboratory, First Affiliated Hospital of NanChang University, 17 Yongwaizhengjie, Nanchang 330006, China. Tel: (þ86) 189 7096 8159. Fax: (þ86) 791 8862 3153. e-mail: Chenks100@ 126.com Article submitted 7 March 2015. Final version accepted 22 July 2015. 58 The International Journal of Tuberculosis and Lung Disease

STUDY POPULATION AND METHODS Estimation of minimum recent tuberculosis Study setting transmission rate The study was conducted in the Antimony Mineral The minimum recent transmission (n-1 method) was 12 Resource Development Company in Jiangxi Prov- estimated according to the following formula: rate ince, China, between January 2011 and December ¼ (number of clustered patients-number of clusters)/ 3 2013. All employees were male miners living in total number of patients 100%. nearby villages, except for a few foreign migrant Risk factors of clustered patients workers living in private accommodation. Since A cluster was defined as a group of two or more 2011, the local government departments have patients who shared the same MIRU-VNTR profile. required mining companies to provide regular health Patients whose isolates had more than one MIRU- checkups for miners. The local Chinese Center for VNTR profile were defined as ‘mixed infections’. They Disease Control and Prevention (CDC) offers TB were categorised according to the most common screening to individuals with symptoms as well as isolate, if this existed, and were excluded if it did not. control measures for local residents, including TB category was judged by clinical and/or CXR miners. All first-line anti-tuberculosis drugs are findings and self-reported history of TB. Recurrence routinely tested against isolated strains, and chest was considered any time after the end of treatment of radiography (CXR) services are available at all the previous episode. Treatment failure implied that times. All patients are treated under the DOTS patients had failed to respond to treatment or had only strategy under the supervision of a community experienced a transient response to treatment, and health worker. symptoms did not disappear even if the patients had taken anti-tuberculosis drugs for .6 months. Individ- Study population uals self-reported personal information such as age, Patients with at least one episode of culture-con- educational level, marital status, smoking status and firmed TB between January 2011 and December duration of employment as part of the health checkup. 2013 were eligible for the study, and were offered an Information on TB, such as drug resistance, CXR and additional free health checkup by the company. sputum smear results, was obtained from the local Enrolment was restricted to culture-confirmed TB CDC. patients who were currently employed by the mining company. Patients were excluded if they had extra- Data management and analysis pulmonary TB or atypical mycobacterial infection Clustering was analysed using the unweighted pair- with no concomitant pulmonary disease. The screen- group method with arithmetic averages (UPGMA). ing process is outlined in the Figure. Discrimination of locus combination was calculated using the Hunter-Gaston Discriminatory Index Laboratory methods (HGDI).13 The v2 test was used to analyse the Mycobacterial 16S rRNA gene sequencing and the correlation between the Beijing genotype and clusters. National Center for Biotechnology Information’s Univariate logistic regression modelling was used to Basic Local Alignment Search Tool (NCBI BLAST) calculate odds ratios (ORs) for factors associated with were employed to identify Mycobacterium tubercu- clustered and non-clustered patients. A multivariate losis complex.9 Beijing strains were discriminated by logistic regression analysis was used to control possible detecting the genomic deletion region of difference confounding factors, and age was included in the model (RD105) using multiplex polymerase chain reaction a priori. All other risk factors with a level of significance 10 (PCR). The MIRU-VNTR15-China method was of P , 0.1 were originally included. A forward stepwise adopted to genotype these strains. Genotyping approach was used in the analysis, and the interaction markers included MIRU10, MIRU16, MIRU23, terms were retained if they were significant. P , 0.05 MIRU26, MIRU27, MIRU39, MIRU40, Mtub21, was considered statistically significant. Mtub30, Mtub39, ETR-A, ETR-B, ETR-C, ETR-D and ETR-E. Primer design, amplification conditions Ethical considerations and the calculation of the number of repeats were The study was approved by the ethics committee of performed as reported in the literature.11 Drug the First Affiliated Hospital of Nanchang University, susceptibility testing (DST) to detect resistance to Nanchang, China (approval number: 2014036). All rifampicin (RMP), isoniazid (INH), ethambutol patients agreed to participate in the study, and all (EMB) and streptomycin (SM) was carried out on personal data were kept confidential. all M. tuberculosis isolates using the proportion method on Lowenstein-Jensen¨ media. Quality con- RESULTS trol was routinely performed during DST using H37Rv strains (American Type Culture Collection Of the 3543 employees, not all attended a health 27294). checkup at the local hospital; between January 2011 Molecular epidemiology of TB in China 59

interval [CI] 1.09–4.83, P ¼ 0.019) (see Appendix Table A).* Risk factors responsible for clustering are shown in Tables 1 and 2. Patients who failed treatment were more prone to clustering (Table 1); the study results indicate that 84.8% (28/33) of such cases were clustered. Although recurrence was strongly associated with an increased rate of clustering in the initial analysis, its impact diminished with adjustment for age (adjusted OR [aOR] 2.32, 95%CI 1.17–4.06, P ¼ 0.07). More than 50% of patients (18/33) who failed treatment had multidrug-resistant isolates, and these were significantly more likely to be clustered than those infected with susceptible strains (aOR 12.1, 95%CI 4.86–29.7, P ¼ 0.0014). The longer they worked in the mines, the more likely they were to acquire TB (Ptrend ¼ 0.04), presumably because they had more opportunity to contact pathogens. There was no association between clustering and a wide range of other characteristics, including age, level of education, marital status and smoking.

DISCUSSION

Figure Study profile of antimony miners. TB ¼ tuberculosis. To the best of our knowledge, this is the first detailed NTM ¼ non-tuberculosis mycobacteria; MIRU ¼ mycobacterial study of the dynamics of TB transmission in an interspersed repetitive unit; VNTR ¼ variable number of tandem antimony mining community using a genotyping repeats. approach. A total of 263 isolates were divided into 35 clusters and 88 unique isolates using established methods.11 Consistent with the findings of others and December 2013, 2714 did so in 2011, 2712 in concerning TB in China,14,15 HGDI was 0.968, 2012 and 2652 in 2013, for a total of 8078 visits; confirming that the MIRU-VNTR15-China method 11.8% (418/3543) employees were confirmed to have was suitable for typing the isolates in our study. The pulmonary TB, and 273 had at least one isolate high incidence of TB in this community was believed sputum culture positive for M. tuberculosis. Howev- to be due to lack of health and safety awareness, er, we failed to recover strains for genotyping in five heavy physical labour and mine-related risk factors patients, and five strains had invalid MIRU-VNTR such as long-term exposure to silica. There are no results. Thus, a final 263 patients were included in the published data on the incidence of TB in an study (Figure). antimony mining community, although the current Results of genotyping showed that 175 patients incidence of TB is about 3955 per 100 000 popula- were distributed into 35 clusters, at a clustering rate tion per year, twice as high as the rate reported for of 66.5% (175/263). The largest cluster included 37 goldminersinSouthAfrica.6 We estimated that patients, accounting for 21.1% of the total. The other more than 10% of the miners in the community we clusters were composed of varying numbers of studied had silicosis, which, as an important patients (range 2–23). The minimum estimated rate predisposing factor for TB, may account for the of TB transmission in these miners was 53.2% ([175 high TB prevalence there.16 Unlike the gold mining 35]/263 3 100%). community in South Africa, whose sound manage- Among the 263 strains isolated, 66.5% (175/263) ment policies and good ventilation equipment have were susceptible to all first-line anti-tuberculosis been developed over a hundred years, antimony drugs, 26.2% (69/263) were resistant to a single mining in China is only 30 years old. Chinese mines drug and 7.2% (19/263) exhibited multidrug resis- operate with inadequate or no ventilation equip- tance. Although there was no correlation between ment, which may be partially responsible for the resistance and clustering, strains that were epidemic high TB incidence among its miners.17 TB preva- attheminingsiteweredrug-resistant.Of226 patients diagnosed as carrying the Beijing family * The appendix is available in the online version of this article, at genotype, 163 were clustered. The Beijing genotype http://www.ingentaconnect.com/content/iuatld/ijtld/2016/ exhibited a high rate of clustering (95% confidence 00000020/00000001/art00010 60 The International Journal of Tuberculosis and Lung Disease

Table 1 Univariate analysis of risk factors for clusters in 263 antimony miners

Clustered Not clustered (n ¼175) (n ¼ 88) Risk factor n (%) n (%) P value OR (95%CI)

Tuberculosis category New 124 (70.9) 78 (88.6) — 1.00 Recurrent 23 (13.1) 5 (5.7) 0.02 1.87 (1.15–6.07) Treatment failure 28 (16.0) 5 (5.7) 0.01 3.75 (1.34–13.0) Age, years ,30 13 (7.4) 8 (9.1) 0.42* 1.00 30~50 58 (33.1) 30 (34.1) — 1.15 (0.37–3.38) .50 104 (59.4) 50 (56.8) — 1.31 (0.49–3.66) Years of employment in mine ,5 20 (11.4) 18 (20.4) 0.04* 1.00 6~10 64 (36.6) 33 (37.5) — 1.74 (0.76–4.00) .10 91 (52.0) 37 (42.0) — 2.46 (1.09–5.48) Drug resistance Susceptible 111 (63.4) 64 (72.7) — 1.00 Monodrug-resistant 47 (26.8) 22 (25.0) 0.49 1.23 (0.66–2.35) Multidrug-resistant 17 (9.7) 2 (2.3) 0.02 4.90 (1.10–44.8) Educational level Pre-school 79 (45.1) 37 (42.0) — 1.00 Primary 43 (24.6) 13 (14.8) 0.68 0.88 (0.44–1.75) Secondary 53 (30.3) 38 (43.2) 0.14 0.65 (0.36–1.64) Marital status Unmarried 26 (14.8) 14 (15.9) — 1.00 Married 149 (85.2) 74 (84.1) 0.82 1.08 (0.49–2.30) Smoking Never 57 (21.5) 31 (35.2) — 1.00 Previous 82 (30.9) 40 (45.4) 0.83 1.11 (0.60–2.07) Current 36 (13.6) 17 (19.3) 0.42 1.15 (0.53–2.55) Chest radiograph No/minimal change 43 (24.6) 26 (22.7) — 1.00 Moderate change 79 (45.1) 38 (39.8) 0.89 1.26 (0.64–2.45) Extensive change 53 (30.3) 24 (37.5) 0.40 1.36 (0.64–2.81) Sputum smear Negative 47 (26.8) 51 (58.0) — 1.00 Positive 128 (73.1) 37 (42.0) 0.001 3.75 (2.11–6.67)

* v2 test for trend. OR ¼ odds ratio; CI ¼ confidence interval.

lence among the general population in Jiangxi Province was 463/100 000 based on the 2010 national TB epidemiology survey in China, far higher than the average level of TB in the world, Table 2 Multiple logistic regression of the association of risk but significantly lower than that of other mining factors with clustering in an antimony mining community communities.1,6,18,19 Treatment failure is an important risk factor for Risk factor Adjusted OR (95%CI)* P value clustering, probably because it can result in patients Tuberculosis remaining infectious for longer periods. As shown in New 1.00 the Figure, there were 15 clusters with patients who Recurrent 2.32 (1.17–4.06) 0.07 Treatment failure 3.03 (2.12–6.89) 0.03 had failed treatment; the largest cluster had 13 Years of employment in mine patients with treatment failure. Patients who fail ,5 1.00 treatment are more likely to develop drug resis- 6~10 3.57 (2.10–6.74) 0.01 20 21 .10 5.26 (3.06–9.58) 0.005 tance, which can impede TB control. We there- Drug resistance fore recommend periodic screening of TB patients to Susceptible 1.00 determine whether their sputum contains drug- Monodrug-resistant 5.59 (2.70–13.14) 0.005 resistant bacilli, and that treatment regimens be Multidrug-resistant 12.01 (4.86–29.7) 0.001 adjusted based on DST results. These measures could Sputum smear Negative 1.00 significantly reduce treatment failures. Most patients Positive 2.13 (0.92–4.18) 0.052 in our study with treatment failure had moderate or

* Adjusted for age group and the four other variables in this table. extensive changes on CXR. Although the extent of OR ¼ odds ratio; CI ¼ confidence interval. the disease seen on CXR was not associated with Molecular epidemiology of TB in China 61 clustering, these patients were more likely to have References positive sputum smear results, in accordance with 1 World Health Organization. Global tuberculosis report, 2014. previously reported data.22 WHO/HTM/TB/2014.08. Geneva, Switzerland: WHO, 2014. There was no association between smoking and 2 Wang L, Zhang H, Ruan Y, et al. Tuberculosis prevalence in clustering in our study. However, smoking is a risk China, 1990–2010; a longitudinal analysis of national survey data. Lancet 2014; 383: 2057–2064. factor for pulmonary TB,23 andanimportant 3 Wang L, Zhang H, Ruan Y, et al. Long-term exposure to silica 24 influence on TB transmission. Although the associ- dust and risk of total and cause-specific mortality in Chinese ation was not statistically significant, our data workers: a cohort study. PLOS MED 2012; 9: e1001206. suggest that smoking may weakly affect TB trans- 4 Zhang M, Zheng Y D, Du X Y, et al. Silicosis in automobile mission. We found no direct relationship between foundry workers: a 29-year cohort study. Biomed Environ Sci level of education and TB, although persons who 2010; 23: 121–129. 5 Ren J H, Ma L Q, Sun H J, et al. Antimony uptake, receive a good education might be expected to pursue translocation and speciation in rice plants exposed to a healthier lifestyle and so reduce their likelihood of antimonite and antimonate. Sci Total Environ 2014; 475: 83– contracting TB.25 One reason may be that the 89. constraints of lifestyle and environment common to 6 van Halsema C L, Fielding K L, Chihota V N, et al. Trends in miners offset the protective effect that a good drug-resistant tuberculosis in a gold-mining workforce in South Africa, 2002–2008. Int J Tuberc Lung Dis 2012; 16: education might otherwise confer. 967–973. The generalisation of our results may be limited 7 Anderson L F, Tamne S, Brown T, et al. Transmission of because the 15-locus MIRU-VNTR method does multidrug-resistant tuberculosis in the UK: a cross-sectional not provide enough resolution in genotyping.26 For molecular and epidemiological study of clustering and contact example, not all patients with a given genotype had tracing. Lancet Infect Dis 2014; 14: 406–415. 8 Zmak L, Obrovac M, Katalinic Jankovic V. First insights into an epidemiological connection to the same clusters. the molecular epidemiology of tuberculosis in Croatia during a As a result, TB transmission is overestimated in our three-year period, 2009 to 2011. Scand J Infect Dis 2014; 46: study. However, the high rate of TB transmission in 123–129. the antimony community is likely to be general- 9 Hall L, Doerr K A, Wohlfiel S L, et al. Evaluation of the isable. The generalisability of our findings concern- microSeq system for identification of mycobacteria by 16S ing some risk factors for TB transmission may be ribosomal DNA sequencing and its integration into a routine clinical mycobacteriology laboratory. J Clin Microbiol 2003; questioned, as we had no opportunity to overcome 41: 1447–1453. selection bias for incomplete data on TB in the local 10 Chen J, Tsolaki A G, Shen X, et al. Deletion-targeted multiplex population. However, the individuals in our study PCR (DTM-PCR) for identification of Beijing/W genotypes of were highly similar, as almost all were miners; our Mycobacterium tuberculosis. Tuberculosis (Edinb) 2007; 87: results therefore have a high degree of credibility. 446–449. 11 Wan K, Liu J, Hauck Y, et al. Investigation on Mycobacterium The generalisability of risk factors on TB transmis- tuberculosis diversity in China and the origin of the Beijing sion may be incomplete, as we were not able to clade. PLOS ONE 2011; 6: e29190. collect detailed information, and some important 12 Mandal S, Bradshaw L, Anderson L F, et al. Investigating measures might have been missed.7 Nevertheless, transmission of Mycobacterium bovis in the United Kingdom in measures similar to those that we recommend 2005 to 2008. J Clin Microbiol 2011; 49: 1943–1950. 13 Hunter P R, Gaston M A. Numerical index of the should be taken to control the spread of TB. discriminatory ability of typing systems: an application of Simpson’s index of diversity. J Clin Microbiol 1988; 26: 2465– CONCLUSION 2466. 14 Liu J, Tong C X, Liu J, et al. First insight into the genotypic The study describes TB transmission and associated diversity of clinical Mycobacterium tuberculosis isolates from Gansu Province, China. PLOS ONE 2014; 9: e99357. risk factors in a Chinese antimony mining communi- 15 Jiang Y, Liu H C, Zhang H, et al. 19-VNTR loci used in ty. Our results highlight the importance of chronic TB genotyping Chinese clinical Mycobacterium tuberculosis as a source of infection. This suggests that measures complex strains and in association with spoligotyping. J Basic need to be adopted to reduce TB transmission by Microbiol 2013; 53: 562–580. detecting the TB pathogen in a community as early as 16 Hnizdo E, Murray J. Risk of pulmonary tuberculosis relative to possible, improving supervision of treatment and silicosis exposure to silica dust in South African gold miners. Occup Environ Med 1998; 55: 496–502. reducing the rate of treatment failure. 17 Brouwer M, Coelho E, das Dores Mosse C, et al. Implementation of tuberculosis infection prevention and Acknowledgements control in Mozambican health care facilities. Int J Tuberc The authors thank the staff of the Antimony Mineral Resource Lung Dis 2015; 19: 44–49. Development Company miners for permission to publish these 18 Dharmadhikari A, Smith J, Nardell E, et al. Aspiring to zero data, and the local medical services for their assistance in tubetculosis deaths among southern Africa’s miners: is there a facilitating data collection. This project was partially funded by a way forward? Int J Health Serv 2013; 43: 651–664. grant from Nanchang University, Nanchang, Jiangxi Province, 19 South African Ministry of Health. Tuberculosis Strategic Plan China (grant number: YFY05). for South Africa, 2007–2011. Pretoria, South Africa: Ministry Conflicts of interest: none declared. of Health, 2007. 62 The International Journal of Tuberculosis and Lung Disease

20 Huang F L, Jin J L, Chen S, et al. MTBDRplus results correlate 24 Huang C C, Tchetgen E T, Becerra M C, et al. Cigarette with treatment outcome in previously treated tuberculosis smoking among tuberculosis patients increases risk of patients. Int J Tuberc Lung Dis 2015; 19: 319–325. transmission to child contacts. Int J Tuberc Lung Dis 2014; 21 Jenkins H E, Crudu V, Soltan V, et al. High risk and rapid 18: 1285–1291. appearance of multidrug resistance during tuberculosis 25 Li X X, Lu W, Zu R Q, et al. Comparing risk factors for primary multidrug-resistant tuberculosis and primary drug-susceptible treatment in Moldova. Eur Respir J 2014; 43: 1132–1141. tuberculosis in Jiangsu province, China: a matched-pairs case- 22 Mor Z, Goldblatt D, Kaidar-Shwartz H, et al. Drug-resistant control study. Am J Trop Med Hyg 2015; 92: 280–285. tuberculosis in Israel: risk factors and treatment outcomes. Int J 26 Luo T, Yang C, Gagneux S, et al. Combination of single Tuberc Lung Dis 2014; 18: 1195–1201. nucleotide polymorphism and variable-number tandem repeats 23 Levy M H, Connolly M A, O’Brien K J. Cigarette smoking as a for genotyping a homogenous population of Mycobacterium risk factor for tuberculosis in young adults: a case-control tuberculosis Beijing strains in China. J Clin Microbiol 2012; 50: study. Tuber Lung Dis 1996; 77: 570. 633–639. Molecular epidemiology of TB in China i

APPENDIX

Table A MIRU-VNTR typing results of Mycobacterium tuberculosis isolates from antimony miners

Key MIRU10 MIRU16 MIRU23 MIRU26 MIRU27 MIRU39 MIRU40 Mtub21 Mtub30 Mtub39 ETRA ETRB ETRC ETRD ETRE Beijing a1011 5 1 5 1 3 2 2 5 4 5 3 2 4 3 2 N a1050 3 3 5 8 3 3 3 6 4 4 4 2 4 1 5 Y a1112 3 3 5 7 3 3 3 4 4 5 4 2 4 3 2 Y a1130 1 3 5 7 3 3 3 4 4 4 4 2 4 3 4 Y a1154 2 3 5 7 3 3 3 5 4 4 4 2 4 3 5 Y a1172 3 2 5 7 3 3 3 5 4 4 4 2 4 3 4 Y a1174 3 3 5 7 3 3 3 4 4 4 4 2 4 3 5 Y a1201 2 3 5 6 3 3 3 5 4 4 4 2 4 3 4 N a1245 3 3 5 7 3 3 3 5 4 4 4 2 4 3 5 Y a1256 3 3 6 7 3 3 3 6 4 4 4 2 4 3 6 Y a1278 3 3 5 7 3 3 4 5 4 4 4 2 4 3 5 Y a1337 2 3 6 5 3 2 3 3 2 3 2 1 4 3 2 N a1365 3 3 5 5 3 3 2 3 4 4 3 2 4 3 6 N a1376 3 2 5 4 3 3 3 4 4 4 4 2 4 3 4 Y a1382 3 3 5 7 3 3 3 5 4 4 3 2 4 3 5 Y a1411 3 3 5 7 3 3 3 4 4 4 4 2 4 3 5 Y a1422 3 3 5 6 3 3 3 6 4 6 4 2 4 3 5 Y a1535 3 3 5 7 3 3 3 5 4 5 4 2 4 3 6 Y a1556 2 3 5 7 3 3 3 4 4 4 4 2 4 3 5 Y a1570 3 3 5 7 3 3 3 5 4 4 4 2 4 3 5 Y a1576 3 3 5 7 3 3 3 5 4 4 4 2 4 3 5 Y a1585 3 3 5 7 3 3 3 5 4 6 4 2 4 3 5 Y a1596 3 3 5 7 3 3 3 5 4 4 3 2 4 3 5 Y a1612 4 1 5 1 3 2 2 3 4 4 3 2 4 3 3 N a1627 2 3 5 7 3 3 3 5 4 4 3 2 4 3 5 Y a1668 1 3 5 7 4 3 3 4 2 4 4 2 4 3 5 Y a1687 3 3 5 4 3 3 3 4 4 4 4 2 4 3 5 Y a1702 1 3 4 7 3 3 3 4 4 4 4 2 4 3 5 Y a174 3 3 5 5 3 3 2 3 4 4 3 2 4 3 6 N a1761 2 3 5 7 3 2 3 3 2 4 3 1 4 3 3 N a1762 3 2 5 7 3 3 3 5 4 4 4 2 4 3 4 Y a1794 3 3 5 4 3 3 3 4 4 4 4 2 4 3 5 Y a2335832354442435Y a2086 3 3 5 7 3 3 3 5 4 4 4 2 4 3 5 Y a2121 3 3 5 7 2 2 3 5 4 4 4 2 4 3 5 Y a2141 3 3 5 4 3 3 3 4 4 5 4 2 4 3 4 Y a2152 3 3 5 7 3 3 3 5 4 4 4 2 4 3 5 Y a217 2 3 5 5 4 2 3 3 2 3 3 1 4 3 3 Y a2171 2 3 5 5 3 2 3 3 2 4 3 1 4 3 3 N a242 3 3 5 7 3 3 3 4 4 4 4 2 4 3 5 Y a250 3 3 5 7 2 3 3 5 4 4 4 2 4 3 5 Y a277 3 3 5 7 3 3 3 4 4 4 4 2 4 3 5 Y a295 3 3 5 7 3 3 3 4 4 5 4 2 4 3 2 Y a307 3 3 5 7 3 3 3 5 4 4 4 1 4 2 5 Y a343 3 4 5 6 3 3 2 4 4 4 4 2 4 3 6 Y a346 3 3 5 6 3 3 3 6 4 4 4 2 4 3 5 Y a358 3 3 5 7 3 3 3 5 4 4 4 2 4 3 5 Y a385 3 2 5 7 3 3 3 5 4 4 4 2 4 3 4 Y a392 3 3 5 7 3 3 3 5 4 4 4 2 4 3 5 Y a41135533244442435Y a414 3 3 5 7 2 3 3 5 2 4 3 2 4 3 5 Y a427 3 3 5 7 3 3 2 5 4 4 4 2 4 3 5 Y a443 3 3 5 6 3 3 3 4 4 4 4 2 4 3 5 N a462 3 3 5 7 3 3 3 5 4 5 4 2 4 3 4 Y a517 2 3 5 7 3 3 3 5 4 4 4 2 4 3 5 Y a537 3 3 5 6 3 3 3 3 4 4 4 2 4 3 5 N a541 3 3 5 7 3 3 3 4 4 4 4 2 4 3 5 Y a557 4 3 5 7 3 3 3 4 4 4 4 2 4 3 5 Y a567 3 3 5 4 3 3 3 4 4 4 2 2 4 3 5 N a592 3 3 5 5 3 3 3 4 4 5 4 2 3 3 5 Y a609 3 2 6 3 3 2 1 2 2 6 3 3 4 4 3 N a612 3 3 5 7 3 3 3 5 4 4 4 2 4 3 5 Y a611 2 3 5 7 3 3 3 5 4 4 4 2 4 3 6 Y a618 3 3 5 5 3 3 3 4 4 4 4 2 4 3 5 Y a640 3 3 5 7 3 3 3 5 4 6 4 2 4 4 5 Y a649 3 3 5 8 3 3 3 5 4 4 3 2 4 3 5 Y a654 3 2 5 7 3 3 3 5 4 4 4 2 4 3 4 Y ii The International Journal of Tuberculosis and Lung Disease

Table A (continued)

Key MIRU10 MIRU16 MIRU23 MIRU26 MIRU27 MIRU39 MIRU40 Mtub21 Mtub30 Mtub39 ETRA ETRB ETRC ETRD ETRE Beijing a668 335733344542434Y a680 335733354442435Y a689 335633364442435Y a7135533244442435Y a726 335733344642435N a792 335633354442435Y a82335734354442435Y a824 135533244442435Y a834 335733354442435Y a862 335733344442435Y a884 335533234442436N a914 235432532531433N b1011 343533344442436Y b1047 335733354442435Y b1048 235733354332435Y b1075 235532432431433N b1077 135533244442435Y b1084 335633314442435Y b1127 335833354442435Y b1149 335733344442435Y b1151 335633354442435Y b1164 335633364442435Y b1170 335533344442435Y b1173 335733344442435Y b1195 335733254442435Y b1244 335633354442435Y b1244 335733344632435Y b1257 335733354442435Y b1271 235733354442435Y b1280 335733344442435Y b1301 335433234442436N b1312 235233354442435Y b1336 335833354432435Y b1342 335733344542434Y b1381 335933344442435Y b1396 335733354642445Y b1411 325733354442434Y b1424 415733354442435Y b1435 235442132631433N b1444 335733354442435Y b1457 335733554442436Y b1458 335733344442435Y b1491 335733354442435Y b1495 235633364442437Y b15335533234442436N b1530 335633344442435N b1571 335733434542432Y b1579 235522322431433N b1666 335433354442435Y b1668 235733354442435Y b169 325132234452433N b1768 335733154442436Y b1787 235833344442435Y b179 235633354442435Y b183 335733354442435Y b1844 336433344422435Y b189 335832354442435Y b1934 335833334442435Y b1959 335733354442415Y b2026 135633344442434Y b2048 335933344442435Y b206 335733354442435Y b2121 335733364442415Y b2122 335733354442435Y b2124 335733354542415Y b2167 335733354432435Y b2177 335833354442434Y b2190 345533244542435Y b2247 234432212354533N b225 335731354442536Y b2254 435733344442435Y b2277 335733354622435Y Molecular epidemiology of TB in China iii

Table A (continued)

Key MIRU10 MIRU16 MIRU23 MIRU26 MIRU27 MIRU39 MIRU40 Mtub21 Mtub30 Mtub39 ETRA ETRB ETRC ETRD ETRE Beijing b228 335733344442435Y b2482 235733354442435Y b267 135533244442435Y b268 335733354442475Y b289 345733364442433Y b355 335433344442435Y b409 335733344442435Y b412 335733354642435Y b414 335733344442435Y b417 335733454442435Y b432 335434354442435Y b441 335733354432435Y b444 335733354442435Y b469 335833344442435Y b541 335733554442434Y b585 336543342442434Y b612 335733344442435Y b621 335733354442435Y b640 335533234442436N b642 335723354442435Y b672 135633244442435Y b693 335733354442435Y b764 335733354442435Y b779 135533244442435Y b878 135533244442435Y b92335733242442435Y b945 335633334442434Y b945 335733344442435Y b957 235532432431433N b985 335733354642445Y c1014 335733354442435Y c1021 335733354622435Y c1024 335733434542432Y c1027 335633344442435N c1077 335633344442436Y c1118 335633344422435Y c1144 335733354642435Y c1172 335733354442435Y c1185 335633344442435N c1215 645433342442235Y c1222 335733354642445Y c1258 335633524442435N c1265 335733334442435Y c1287 335533344542434Y c1291 335723352432435Y c1297 335733344442435Y c1305 335533234542426Y c1312 345733364442433Y c1389 335733354442435Y c1412 335633364442435Y c142 335433344442435Y c1427 335733344542435Y c1439 335733344442435Y c1447 335733354442435Y c1454 235532332431433N c1544 335733344542432Y c1557 335733454442435Y c1586 335633354442435Y c1587 335733354442435Y c1587 335533344442435Y c1627 235533354442435Y c163 335833354442435Y c1637 335732334442435Y c1644 335533244442435Y c1659 335633344442435N c1683 335733344442435Y c1714 235733334442435Y c1717 336533364442435Y c182 335833354432435Y c1825 335533234442416Y c184 435733344442435Y c1964 335833354432435Y iv The International Journal of Tuberculosis and Lung Disease

Table A (continued)

Key MIRU10 MIRU16 MIRU23 MIRU26 MIRU27 MIRU39 MIRU40 Mtub21 Mtub30 Mtub39 ETRA ETRB ETRC ETRD ETRE Beijing c1992 3 3 5 643344442435N c2007 3 3 5 733414432435Y c201 3 3 5 733434542432Y c2014 3 3 5 733354442436Y c2018 4 3 5 733344442435Y c2024 3 3 5 733344442435Y c2025 3 3 5 733354442435Y c2045 2 3 5 833354442435Y c2137 3 2 5 733354442445Y c214 2 3 5 532432431433N c2164 2 3 5 532332331433N c2182 2 3 5 532432431433N c2184 3 3 5 733354442435Y c2209 3 3 5 733354642435Y c2211 3 3 5 433344442435Y c24335633354332435Y c264 3 3 5 533344442435Y c279 1 3 5 533244442435Y c284 3 3 5 733354442435Y c294 3 3 5 733354442435Y c323 3 3 5 733354442435Y c346 3 3 5 733344442435Y c347 3 3 5 733384442435Y c367 3 3 5 733354442435Y c397 1 3 5 753342442435Y c421 1 3 5 533244442435Y c423 2 3 5 733264442433Y c448 3 3 5 733334442435Y c457 1 3 5 533244442435Y c461 4 3 5 333354442435Y c462 3 3 5 433344442435Y c467 3 3 5 633354442435Y c477 3 3 5 733354642435Y c547 2 3 5 532332431433N c553 3 2 5 533344442434Y c563 1 3 4 733344442434Y c569 3 3 5 733354442435Y c631 3 3 5 733344442435Y c664 3 3 5 933344442435Y c670 3 3 5 733354442435Y c686 3 3 5 733344442435Y c692 3 3 5 733344442435Y c698 2 3 5 432432421433N c719 3 3 5 734354442435Y c753 3 3 5 643344442435N c763 3 3 5 733344442435Y c775 3 3 5 733384442435Y c785 4 2 5 733344442435Y c839 3 3 5 733354442435Y c888 3 3 5 733354442435Y c915 3 3 5 733354442435Y c932 2 3 5 733354442435Y

MIRU ¼ mycobacterial interspersed repetitive unit; VNTR ¼ variable number of tandem repeats; B ¼ Beijing; N ¼ no; Y ¼ yes. Molecular epidemiology of TB in China v

RESUME

CONTEXTE : Une mine d’antimoine dans la province de 35 grappes. L’estimation de la transmission récente de la Jiangxi, Chine. TB aet´ ´ e de 53,2% au sein de la communauté. Les OBJECTIF : Examiner l’incidence de la transmission de patients dont le traitement aechou´ ´ eétaient plus la tuberculose (TB) et des facteurs de risque associés susceptibles de se trouver dans ces regroupements (OR dans une communauté vivant sur le site d’une mine ajusté [aOR] 0,03 ; IC95% 2,12–6,89). Les patients d’antimoine chinoise. ayant des isolats multirésistants avaient plus souvent SCHE´ MA : Etude rétrospective de cohorte. subi unechec ´ du traitement et se trouvaient plus souvent ME´ THODES : L’analyse des unités répétitives dans les regroupements que ceux ayant une souche mycobactériennes intercalées-nombre variable de pharmacosensible (aOR 0,001 ; IC95% 4,89–29,7]). répétitions en tandema ` 15 loci (MIRU-VNTR)15-Chine CONCLUSION : Les individus dont le traitement a aété utilisépourdéterminer le regroupement des echou´´ e sont des sources importantes de transmission patients. Une réaction polymérase multiplex en chaˆıne de la TB, et les isolats multirésistants sont responsables ciblant la délétion région de différence 105 aet´ ´ e adoptée de la majorité de cette transmission. Les plans de lutte pour identifier les souches Beijing. Les facteurs de risque contre la TB doivent se concentrer sur les cas d’échec du de regroupement ontet´ ´ eévalués. traitement dans cette communauté. RE´ SULTATS : Sur 263 patients, 175 ontet´ ´ e regroupés en

RESUMEN

MARCO DE REFERENCIA: Una mina de antimonio de la RESULTADOS: De los 263 pacientes, 175 pertenec´ıan a provincia de Jiangxi en la China. 35 conglomerados. La estimacion ´ de la transmision ´ OBJETIVO: Investigar la tasa de transmisión de la reciente en la comunidad fue de 53,2%. Los pacientes tuberculosis (TB) y los factores de riesgo asociados con con fracaso terapéutico formaban parte de un la misma en una comunidad de explotacion ´ de minas de conglomerado con mayor frecuencia (OR ajustado antimonio en la China. [ORa] 0,03; IC95% 2,12–6,89). La probabilidad de DISENO:˜ Fue este un estudio retrospectivo de cohortes. haber presentado un fracaso terapéutico y de pertenecer ME´ TODOS: Se practico ´ la genotipificacion ´ con a un conglomerado fue mayor en los pacientes con marcadores para locus multiples ´ de las secuencias aislados multirresistentes que en los pacientes con cepas repetitivas en tańdem con 15 loci (MIRU-VNTR15-China), sensibles (ORa 0,001; IC95% 4,8–29,7). a fin de determinar la proporcion ´ de pacientes que CONCLUSIO´ N: Las personas con fracaso terapéutico formaban parte de conglomerados. Las cepas Beijing se representan una fuente importante de transmision ´ de la detectaron mediante una reacción en cadena de la TB y la mayor´ıa alberga cepas multirresistentes. Los polimerasa multiple ´ dirigida a la delecion ´ de la región de planes de control de la TB deben prestar especial diferencia 105. Se evaluaron los factores de riesgo de atencion ´ a los casos de fracaso terapéutico en la aparicion ´ de conglomerados. comunidad.