INT J TUBERC LUNG DIS 16(3):306–311 © 2012 The Union http://dx.doi.org/10.5588/ijtld.11.0304
Recent transmission of W-Beijing family Mycobacterium tuberculosis in rural eastern China W. Wang,* Y. Hu,* B. Mathema,† W. Jiang,* B. Kreiswirth,† B. Xu* * Department of Epidemiology, Key Laboratory of Public Health Safety, Ministry of Education, Fudan University, Shanghai, China; † Tuberculosis Center, Public Health Research Institute, Newark, New Jersey, USA SUMMARY O B J E C T I V E : To understand the degree of recent transmission of tuberculosis (TB) and determine the risk factors associated with recent transmission stratified by W-Beijing genotype in rural China. D E S I G N : A cross-sectional study of bacteriologically confirmed TB patients registered in two rural counties of eastern China over a 1-year period. R E S U LT S : Of 351 patient isolates, spoligotyping identified 243 (69.2%) as W-Beijing family strains, and 53 (15.1%) and 15 (4.3%) as members of T1 Family and Family 33, respectively. Insertion sequence (IS) 6110 based restriction fragment length polymorphism typing revealed that 31 clusters together accounted for 80 of the 351 isolates. Strains with the W-Beijing genotype were more likely to be clustered than non-Beijing strains (42.3% vs.
8.3%, P < 0.001). The proportion of cases due to recent transmission was estimated at 23.1% (32.1% W-Beijing genotype vs. 2.8% non-W-Beijing genotype). Multivariate analysis showed that bacille Calmette-Guérin (BCG) vaccination (aOR 2.97), multidrug resistance (aOR 5.45) and body mass index (aOR 1.13) were independent predictors for clustering among W-Beijing isolates. C O N C L U S I O N S : The low clustering proportions highlight the role of endogenous reactivation of TB as a main concern in rural eastern China. Our findings also suggest that W-Beijing strains were associated with recent transmission in this population, where multidrug resistance and BCG vaccination may play an important role in the mechanism of TB transmission. K E Y W O R D S : IS6110 RFLP; clustering; reactivation
TUBERCULOSIS (TB) and drug-resistant TB are major global public health threats, accounting for the largest number of deaths caused by a single infectious agent. Particularly worrying is the worldwide emergence of multidrug-resistant TB (MDR-TB), Mycobacterium tuberculosis strains that are difficult to treat and are associated with poor therapeutic outcomes.1 A recent survey in 74 geographic countries/ regions worldwide showed proportions of MDR-TB ranging from 0% to 22.3% for new cases.2 China has the second highest burden of TB worldwide, with approximately 1.3 million new TB cases detected annually,3 and has become one of the hot spots of MDR-TB, with overall drug resistance at 27.8% and MDR-TB at 9.3%.4,5 Molecular tools have enhanced our understanding of TB epidemiology by providing insight into the transmission dynamics, source and spread of M. tuberculosis.6,7 Molecular epidemiologic studies have also indicated the existence and worldwide spread of particular strain families in defined geographic regions.8 Molecular epidemiologic methods have refined estimates of recent transmission as an important indicator for surveillance and TB control programs.9
A number of reports have indicated that the W-Beijing genotype is globally widespread.1,8 These strains were first described in China and in neighboring countries and were subsequently observed in several parts of the world, including former Soviet Union and Asian countries.10–12 W-Beijing genotype strains have caused large outbreaks, and have frequently been associated with drug resistance and MDR-TB.8,13,14 The use of insertion sequence (IS) 6110 genotyping and secondary typing methods, particularly spoligotyping, has provided basic tools to differentiate W-Beijing family strains from all other M. tuberculosis complex strains. The distribution of W-Beijing family strains in distinct geographic regions and their ability to predominate and spread in large genotypic clusters suggest that members of this family may have evolutionary refinements facilitating continual spread.1,8,10 Some members of this family bear a unique lipid and phenolic glycolipid that has been associated with a less protective immune response and poorer survival in a murine model.15 In addition, some have suggested that W-Beijing strains have unique properties that render M. bovis bacille Calmette-Guérin (BCG) vaccination
Correspondence to: Biao Xu, School of Public Health, Fudan University, 138 Yi Xue Yuan Road, Shanghai 200032, China. Tel: (+86) 21 5423 7710. Fax: (+86) 21 5423 7811. e-mail:
[email protected] Article submitted 27 April 2011. Final version accepted 6 September 2011.
Recent transmission of W-Beijing strains
less protective, thereby allowing these strains to enrich in certain populations.16,17 In our previous analysis, we found that MIRU (mycobacterial interspersed repetitive units) genotype 223325173533 of the Beijing family is associated with MDR-TB and increased transmissibility.18 In the present study, we further used spoligotyping and IS6110-based restriction fragment length polymorphism (RFLP) typing to understand the transmission of TB in rural China by examining circulating M. tuberculosis strain families and to determine risk factors associated with recent transmission.
MATERIALS AND METHODS Study population A population-based epidemiologic study of active TB was conducted in two rural sites of eastern China, Guanyun County (GY) in Jiangsu Province between 1 April 2004 and 31 March 2005, and Deqing County (DQ) in Zhejiang Province between 1 June 2004 and 31 May 2005. The prevalence of human immunodeficiency virus infection in TB cases for these populations was estimated to be low.19 All sputum specimens from these patients were submitted to the microbiology laboratory at the School of Public Health, Fudan University, for specimen processing, culture, drug susceptibility testing (DST) and genotyping. Inclusion criteria were active TB cases who were bacteriologically confirmed by sputum culture, and who consented to participate in the study. Extrapulmonary TB cases were excluded. A total of 399 bacteriologically positive TB cases were identified and enrolled in the study (182 in DQ and 217 in GY). Ethical approval was granted by the Ethics Committee of the School of Public Health, Fudan University. Informed consent was obtained from all 399 participants. Data collection Subjects were interviewed at the time of TB diagnosis at the county TB dispensaries (CTDs) by physicians who had undergone a 2-day training course for the interview. A semi-structured questionnaire was developed that covered general demographic and socioeconomic characteristics, clinical symptoms and disease history at TB diagnosis. BCG vaccination was determined by self-report and confirmed by the presence of a scar at interview. Body mass index (BMI, kg/m2) was calculated by on-site measuring of weight and height. Individual monthly income was skewed; the variable was therefore included in the logistic model after being separated into four groups by quartiles ranking from low (Quartile 1) to high (Quartile 4). Molecular characterization M. tuberculosis genomic DNA was extracted by standard methods.20 Two genotyping methods were
307
applied to characterize M. tuberculosis isolates: IS6110-based RFLP (IS6110-RFLP) and spoligotyping. Spoligotyping was carried out using commercial kits from Isogen Bioscience BV (Maarssen, The Netherlands).10 Strain family was determined by comparison of spoligotypes with the SpolDB4 database.21 W-Beijing family strain was defined as a M. tuberculosis strain that hybridized to all of the last nine spacer oligonucleotides (spacers 35 to 43).4 IS6110RFLP was conducted following the standard protocol of van Embden et al.,20 with data analyzed using Gelcompar II (Applied Maths, Kortrijk, Belgium). RFLP band-based dendrograms were produced and compared using dice coefficient with an allowed band size deviation of 1%. Estimation of recent transmission To estimate the extent of recent transmission of TB, we calculated the widely used recent transmission index (RTI)n−1= (nc − c)/n, where n is the total number of cases in the sample, c is the number of genotypes represented by at least two cases, and nc is the total number of cases in clusters of two or greater.22 Statistical analysis The χ2 test was used to assess differences in the clustering proportion of M. tuberculosis isolates with different genotypes; when the expected value was
