Testing of tuberculosis infection among Chinese adolescents born after terminating the Bacillus Calmette--Guérin booster vaccination: subgroup analysis of a population-based cross-sectional study

Abstract

The prevalence of tuberculosis infection among adolescents born after terminating the Bacillus Calmette–Guérin (BCG) booster vaccination in China was estimated using tuberculin skin testing (TST) and QuantiFERON-TB Gold assay (QFT) to investigate the influence of neonatal BCG vaccination on the performance of TST. Data analysis was conducted for 2831 eligible participants aged 5–15 years from the baseline survey of a population-based multi-center prospective study. The prevalence rates of TST (induration≥10 mm) and QFT positivity were 9.3% (264/2827) and 2.5% (71/2831), respectively. The rate of QFT indeterminate result was 2.2% (62/2831). The overall agreement between TST and QFT was low (concordance= 88.0%; κ coefficient= 0.125). Only TST was positively associated with BCG vaccination with an adjusted odds ratio of 1.71 [95% confidence interval, 1.26–2.31]. A history of close contact with patients of active TB was significantly associated with positivity for TST and QFT. Our results suggested that BCG neonatal vaccination still affects TST performance, and a two-step approach might be considered for TB infection testing among adolescents in China.

Keywords

tuberculin skin test / interferon-γ release assays / adolescent / agreement

1 Introduction

Tuberculin skin testing (TST) has been a major tool for tuberculosis (TB) infection testing worldwide despite its widely known limitations, such as cross-reactivity with hypersensitivity due to Bacillus Calmette–Guérin (BCG) vaccinations, non-tuberculous mycobacterial (NTM) infections [1], and compromised specificity caused by the booster shots [2]. Interferon-γ release assays (IGRAs) are based on the cellular production of interferon-γ (IFN-γ) in response to Mycobacterium tuberculosis (MTB) specific antigens, i.e., ESAT-6, CFP10, and TB7.7, and have been widely used in various populations. The absence of these protein antigens in all vaccine strains of Mycobacterium bovis-BCG and most NTM guarantees the high specificity levels of IGRAs as compared with TST [3,4]. IGRAs do not induce boosting problems caused by repeated testing. However, IGRAs are expensive compared with TST and require an equipped laboratory with trained personnel. These factors limit the application of IGRAs in less developed areas.

BCG vaccination was included in China’s National Immunization Program as a combination of neonatal and booster vaccinations in 1977–1978; the latter was terminated as recommended by World Health Organization (WHO) in 1997 [5]. LATENTTB-NSTM study is a population-based, multi-centered prospective study that aimed to evaluate the prevalence of latent TB infection (LTBI) in rural China and to estimate the incidence of an active disease among infections. The published baseline results suggested that the prevalence of LTBI in China might be overestimated by TST as compared with IGRA because the performance of the former is influenced by various factors, including BCG vaccination and age [6]. Furthermore, the difference in TST positivity between BCG-scar-present and BCG-scar-absent subgroups of the study population was more evident in participants born between 1978 and 1997 who had theoretically received booster BCG vaccination [7]. QuantiFERON-TB Gold In-Tube assay (QFT), a commercial IGRA, has been recommended for testing LTBI in “all circumstances in which the TST is currently used,” including the testing on children by the Centers for Disease Control and Prevention (CDC) in the US [8]. However, the performance of QFT has not been sufficiently evaluated and compared with that of TST among adolescents in China.

Basing on the baseline survey of the LATENTTB-NSTM study, the present study aims to specifically evaluate the performance of TST and QFT among adolescents born after the booster BCG vaccination was terminated. In the era with BCG neonatal vaccination, our results might provide important references for developing a new TB infection testing approach among adolescents with high risk of acquiring infection and developing an active disease in China.

2 Materials and methods

2.1 Study settings and participants

This cross-sectional study is based on the baseline survey of the LATENTTB-NSTM study. Detailed information of the study was introduced in the reference [6]. A parallel estimation on LTBI prevalence was conducted using TST and QFT on registered residents (≥5 years old) at four study sites in rural China between July 1, 2013 and September 30, 2013. Data analyses were restricted to participants who were born after 1997 when BCG booster vaccination was terminated.

Inclusion criteria of the study population were as follows: births after December 31, 1997 and before June 1, 2008 (5–15 years old at 2013 baseline survey); household residence at the study site for 6 months or longer in the past year; ability to complete the investigations and tests during the study duration; and provision of voluntary written informed consent. Exclusion criteria include the presence or self-reported history of active TB.

2.2 Ethics approval and consent to participate

The study protocol was approved by the Ethics Committee of the Institute of Pathogen Biology, Chinese Academy of Medical Sciences (Beijing, China). Written informed consent from parents or guardians was obtained for each participant.

2.3 Procedure

Sociodemographic data, including ethnic origin, educational level, household income per head in 2012 (i.e., total family income/number of people in the household), and area of household living space, were collected using a standardized questionnaire. Physicians examined the height, weight, pulse, and presence of a BCG scar. Suspected symptoms of pulmonary TB were considered. Self-reported history of active TB, history of close contact, and history of other chronic diseases and immune system disorders (including AIDS, hyperthyroidism, type I diabetes, myasthenia gravis, systemic lupus erythematosus, rheumatoid arthritis, and ankylosing spondylitis) were also collected.

Venous blood was collected for QFT according to the manufacturer’s instructions, and the result was considered positive if the value exceeded the standard cut-off at 0.35 IU/mL. TST was implemented immediately after using Mantoux method by intradermally injecting 0.1 mL of 5 tuberculin units of purified protein derivative (Xiangrui, Beijing, China) into the left forearm as per patient preference [9]. Digital chest radiography was performed for participants with reported suspected symptoms of pulmonary TB or history of close contact. Children with radiographic abnormalities consistent with active pulmonary TB were transferred to the local CDC for diagnosis. Individuals with current active pulmonary TB (pathogenic test positive or clinically suspected) were not included in the analysis of TB infection.

2.4 Statistical analysis

Data from questionnaire and tests were doubly entered into the project-specific data collection and management software. After cleaning, the data were then converted and analyzed using Statistical Analysis System 9.2 (SAS Institute Inc., Cary, NC, USA).

Based on previous reports, here considered potential factors associated with TB infection include history of close contact with patients with pulmonary TB, history of immune system disorders, household income per head, body-mass index (BMI; calculated), and presence of BCG scar. Household income per head was classified based on the national mean level in 2010 (6000 RMB) [10], and BMI was categorized as thinness, normal, and overweight according to the WHO Growth reference for 5–19 years [11].

Distribution of categorical variables in participants was compared by using the study sites for Pearson’s χ² test. Univariate analysis was conducted by Pearson’s χ ² and Fisher exact tests to identify the potential variables related to TST and QFT positivity. All variables with P values less than 0.05 in univariate analysis were entered into the multiple logistic regression analyses, and the strength of the association was estimated by odds ratios (OR) and 95% confidence interval (CI). The agreement between QFT and TST was assessed by calculating the Cohen’s κ coefficient [12].

3 Results

A total of 2831 eligible participants aged 5–15 years old without HIV infection or any immune system disorder were included. Table 1 shows that the participants from different study sites exhibited similar gender distribution with an overall mean age of 9.6 years (SD 2.9). Participants from site A showed the highest household income per head and overweight proportion. BCG scar was present in 64.2% of the participants distributed among the study sites. Meanwhile, 2.2% of the participants (61/2831) reported a history of close contact with active TB patients.

Four of the participants had missing data for TST. In summary, 18.3% of the participants (517/2827) had an induration≥5 mm, 9.3% of the participants (264/2827) had an induration≥10 mm, and 3.9% of the participants (111/2827) had an induration≥15 mm. For QFT, 2.5% (71/2831) of the participants tested positive, and 2.2% (62/2831) reported indeterminate results. As shown in Table 2, the factors significantly associated with TST≥10 mm in the univariate analysis were the presence of BCG scar, history of close contact with active TB patients, BMI, and study site. For QFT, only history of close contact with active TB patients was associated with positive results. In the multivariable analysis, the presence of BCG scar (adjusted OR= 1.71; 95% CI, 1.26–2.31), history of close contact with active TB patients (adjusted OR= 3.24; 95% CI, 1.71–6.14), and study site remained statistically associated with TST positivity. Individuals with history of close contact had an increased risk for QFT positivity with adjusted OR of 5.41 (95% CI, 2.37–12.36).

As shown in Fig. 1, the overall agreement between TST and QFT was 88.0% with a κ coefficient of 0.125 (95% CI, 0.072–0.179) when using 10 mm as cut-off value of TST. When the cut-off value increased to 26 mm, the highest κ coefficient was recorded as 0.354 (95% CI, 0.233–0.476), which was still acceptable. The agreement between TST and QFT was higher in BCG-scar-absent participants than in BCG-scar-present participants.

Table 3 presents the detailed information of the concordance and discordance between QFT and TST results. Among the participants aged 5–9 years, 0.8% (11/1457) were TST⁺/QFT⁺, which had no statistical significant difference with those for aged 10–15 years (1.0%, 15/1370, P = 0.8). In the TST⁺/QFT⁻ results, 7.9% (116/1457) were aged 5–9 years, and 8.3% (113/1370) were aged 10–15 years, which also showed no significant difference (P = 0.07). Among those participants who reported a history of close contact with active TB patients, 11.5% (7/61) were TST⁺/QFT⁺, but none was TST⁻/QFT⁺ (P<0.001).

QFT quantitative values (median of the TB antigen minus Nil control) were calculated with regard to the TST results in Table 4 to further assess the characteristic of QFT positives. When using 5 mm as the TST cut-off value, 38 out of 71 QFT positives were TST negative and showed significantly lower QFT quantitative values than the 33 TST positives. However, when restricting the analysis for those with a history of close contact, all of the seven QFT positives were TST positive. Similar results were obtained when using different cut-off values for TST.

4 Discussion

In this cross-sectional epidemiological study, parallel testing on the prevalence of TB infection was conducted by using TST and QFT in 2831 adolescents (aged 5–15 years) born after 1997 when booster BCG vaccination was terminated in China. The overall rates of TST (≥10 mm) and QFT positivity were 9.3% (264/2827) and 2.5% (71/2831), respectively. Association analysis showed BCG neonatal vaccination influences the performance of TST but not that of QFT.

Two previously published meta-analyses suggested that TST and IGRAs have similar accuracy for the detection of TB infection in children [13,14]. WHO strongly recommended that IGRAs should not replace TST for the detection of TB infection in children from low or middle-income countries because IGRAs are unsuitable for less-developed areas, are expensive, and require professional laboratory support and trained personal for accurate performance [15]. However, in Korea and UK, a two-step approach was suggested to detect TB infection by IGRA in TST-positive subjects [16]. The National Institute for Health and Care Excellence guidelines proposed the use of IGRAs for all children who are aged 5 years or older and have TST-positive results [17]. Several studies confirmed that QFT is more specific than TST for TB infection testing in children [18–20]. As a potential underlying mechanism, BCG vaccination was suggested to cause the false-positive TST results among children; such effect varies with many factors, including the age of first vaccination [21–25]. A meta-analysis on 240 203 subjects who received BCG vaccination during infancy reported that only 1% of the study participants were TST positive 10 years after BCG vaccination [26]. However, our results revealed that the presence of BCG scar was significantly associated with the increased risk of TST positivity (TST⁺/QFT⁺ and TST⁺/QFT⁻) even after 10 years (Table 3). TST specificity is compromised by BCG boosters, and this effect might persist for more than 20 years [7]. Our analysis on adolescents suggests that the terminated BCG neonatal vaccination still affects TST performance.

Inconsistent with some previous studies [27,28], no significant difference in the positivity of TST and QFT was observed among the age groups in our study. A potential explanation might be that the burden of TB infection in our study population is lower than that in studies from high-burden areas. The association of TB infection with aging-related accumulation of exposure is difficult to identify in a setting with low prevalence. In addition, the rate of TST positivity varied by study sites, which might be partly explained by the various rates of BCG vaccination among the study sites. However, this variation on QFTs was consistent with the distribution pattern of TB epidemic among the study sites (please refer to Supplementary Figure). The difference in BCG vaccination implementation status at the study sites might affect the performance of TST and thus render QFT a better approach to determine TB infection prevalence in different areas.

Another concern is that the 71 QFT-positive individuals showed significantly higher QFT quantitative values for TST positives than those with TST negatives regardless of the TST cut-off values. This phenomenon can be explained by the possibility that the individuals with QFT quantitative values around the cut-off value might be misclassified as false positives. This result is consistent with previous report on the reversion of QFT positivity in the serial tests; in this study, the QFT quantitative values of those reversions were significantly lower and closer to the cut-off value than those with persistent positives [29–31]. These findings also suggested that those reversions might have been misclassified as positives. Considering the lack of a fixed standard for LTBI test, we could not exclude the possibility of false negative for QFT. Therefore, further studies are needed to evaluate the performance of QFT with regard to sensitivity and specificity in Chinese adolescents.

A history of close contact was significantly associated with positivity for TST and QFT, and the overall agreement between TST and QFT was low. However, our results indicated that almost all QFT positives with a history of close contact were TST positives as shown in Table 4. This finding suggested that the agreement of these two tests might be better in high-risk populations. Machado et al. found an agreement of 76% (k = 0.53) between TST and QFT results among household contacts, which is consistent with the results of our study [32]. Cost-effectiveness analysis performed in our study population indicated that TST screening followed by QFT confirmation would be more cost saving compared with using QFT only (RMB 9181 versus RMB 30 460 per positive). Therefore, this two-step approach might be considered in children to reliably and cost-efficiently estimate the prevalence of infection. This two-step approach replaces only the use of TST in the high-risk subgroups, such as close contacts, and will largely reduce the number of patients eligible for preventive therapy among those identified with LTBI.

The limitations of the study should be considered when interpreting the results. First, the analysis on risk factor identification was limited because of the small number of QFT positives. Second, QFT indeterminate result (2.2%) was mostly equal to the positive rate (2.5%). Indeterminate result is defined as an abnormally high level of IFN-γ in the nil tube or a low response in the mitogen tube [33]. This outcome might be caused by technical error, such as incubation delay and host health problems, including fever, acute pneumonia, and chronic renal diseases [34,35]. Bias caused by misclassification could not be excluded for the analysis of QFT even if most of the QFT indeterminate results (85.5%, 53/62) were TST negative as shown in Table 3. The proportion of indeterminate result in this study population is consistent to the findings in the other age groups, the possibility of nonstandard laboratory operation can be excluded, and the results of quality control for this test were provided in a previous report [6]. In addition, a sensitivity analysis, including “indeterminate” among the negative results, was performed, and the results of association analysis were not significantly changed (Supplementary Table). Third, the 2831 children included in the analysis were enrolled from the household sample designed for the whole registered resident population. This factor results in the imbalance of the samples among the four study sites, thereby rendering the comparison of TB infection burdens between the study sites difficult. Sex and age were standardized when evaluating the consistency of TST and QFT results with local TB epidemic to partially overcome this limitation. Fourth, stratified analysis was limited by the small number of infections in the study population, which is also one of the constraints of this study. Finally, infection with other types of mycobacteria other than MTB may also lead to false-positive TST. The potential bias caused by such factors in our data analysis is difficult to minimize. Despite these limitations, we believe our data will be helpful in understanding the performance of TST and IGRAs in Chinese children.

5 Conclusions

This study addressed the prevalence of TB infection (using both IGRAs and TST) in Chinese adolescents born after the booster BCG vaccination was terminated and strongly suggested that BCG neonatal vaccination still affects TST performance. In consideration of the limitations of these two tests, the two-step approach might be applied for TB infection test among adolescents in China.

References

[1] Huebner RE, Schein MF, Bass JB Jr. The tuberculin skin test. Clin Infect Dis 1993; 17(6): 968–975

[2] Menzies D. What does tuberculin reactivity after bacille Calmette-Guérin vaccination tell us? Clin Infect Dis 2000; 31(Suppl 3): S71–S74

[3] Pai M, Riley LW, Colford JM Jr. Interferon-γ assays in the immunodiagnosis of tuberculosis: a systematic review. Lancet Infect Dis 2004; 4(12): 761–776

[4] Andersen P, Munk ME, Pollock JM, Doherty TM. Specific immune-based diagnosis of tuberculosis. Lancet 2000; 356(9235): 1099–1104

[5] Wang L, Liu J, Chin DP. Progress in tuberculosis control and the evolving public-health system in China. Lancet 2007; 369(9562): 691–696

[6] Gao L, Lu W, Bai L, Wang X, Xu J, Catanzaro A, Cárdenas V, Li X, Yang Y, Du J, Sui H, Xia Y, Li M, Feng B, Li Z, Xin H, Zhao R, Liu J, Pan S, Shen F, He J, Yang S, Si H, Wang Y, Xu Z, Tan Y, Chen T, Xu W, Peng H, Wang Z, Zhu T, Zhou F, Liu H, Zhao Y, Cheng S, Jin Q; LATENTTB-NSTM study team. Latent tuberculosis infection in rural China: baseline results of a population-based, multicentre, prospective cohort study. Lancet Infect Dis 2015; 15(3): 310–319

[7] Gao L, Jin Q. Differences in BCG vaccination and tuberculin skin-test positivity—Authors’ reply. Lancet Infect Dis 2015; 15(9): 1003–1005

[8] Mazurek GH, Jereb J, Lobue P, Iademarco MF, Metchock B, Vernon A, Division of Tuberculosis Elimination, National Center for HIV, STD, and TB Prevention, Centers for Disease Control and Prevention (CDC). Guidelines for using the QuantiFERON-TB Gold test for detecting Mycobacterium tuberculosis infection, United States. MMWR Recomm Rep 2005; 54(RR-15): 49–55

[9] Cheng S, Wang G, Wang L, Zhou L. Guidance of Tuberculin Skin Test. Bengjing: People’s Medical Publishing House, 2014

[10] Sheng YL. China Statistical Yearbook—2010. Beijing: China Statistics Press, 2010

[11] World Health Organization. Growth reference 5–19 years. 2007.

[12] Kang YA, Lee HW, Yoon HI, Cho B, Han SK, Shim YS, Yim JJ. Discrepancy between the tuberculin skin test and the whole-blood interferon γ assay for the diagnosis of latent tuberculosis infection in an intermediate tuberculosis-burden country. JAMA 2005; 293(22): 2756–2761

[13] Mandalakas AM, Detjen AK, Hesseling AC, Benedetti A, Menzies D. Interferon-γ release assays and childhood tuberculosis: systematic review and meta-analysis. Int J Tuberc Lung Dis 2011; 15(8): 1018–1032

[14] Machingaidze S, Wiysonge CS, Gonzalez-Angulo Y, Hatherill M, Moyo S, Hanekom W, Mahomed H. The utility of an interferon γ release assay for diagnosis of latent tuberculosis infection and disease in children: a systematic review and meta-analysis. Pediatr Infect Dis J 2011; 30(8): 694–700

[15] World Health Organization. Use of tuberculosis interferon-γ release assays (IGRAs) in low-and middle-income countries: policy statement. 2011.

[16] Mazurek GH, Jereb J, Vernon A, LoBue P, Goldberg S, Castro K; IGRA Expert Committee; Centers for Disease Control and Prevention (CDC). Updated guidelines for using interferon γ release assays to detect Mycobacterium tuberculosis infection — United States, 2010. MMWR Recomm Rep 2010; 59(RR-5): 1–25

[17] National Institute for Health and Care Excellence. Clinical diagnosis and management of tuberculosis, and measures for its prevention and control. Manchester: NICE, 2011

[18] Lighter J, Rigaud M, Eduardo R, Peng CH, Pollack H. Latent tuberculosis diagnosis in children by using the QuantiFERON-TB Gold In-Tube test. Pediatrics 2009; 123(1): 30–37

[19] Tsiouris SJ, Austin J, Toro P, Coetzee D, Weyer K, Stein Z, El-Sadr WM. Results of a tuberculosis-specific IFN-γ assay in children at high risk for tuberculosis infection. Int J Tuberc Lung Dis 2006; 10(8): 939–941

[20] Bianchi L, Galli L, Moriondo M, Veneruso G, Becciolini L, Azzari C, Chiappini E, de Martino M. Interferon-γ release assay improves the diagnosis of tuberculosis in children. Pediatr Infect Dis J 2009; 28(6): 510–514

[21] Nelson LJ, Wells CD. Global epidemiology of childhood tuberculosis. Int J Tuberc Lung Dis 2004; 8(5): 636–647

[22] Starke JR. Childhood tuberculosis: ending the neglect. Int J Tuberc Lung Dis 2002; 6(5): 373–374

[23] Piñeiro R, Mellado MJ, Cilleruelo MJ, García-Ascaso M, Medina-Claros A, García-Hortelano M. Tuberculin skin test in bacille Calmette-Guérin-vaccinated children: how should we interpret the results? Eur J Pediatr 2012; 171(11): 1625–1632

[24] Rieder HL. Epidemiological Basis of TB Control. Paris: International Union Against Tuberculosis and Lung Disease, 1999. 26–43

[25] Wang L, Turner MO, Elwood RK, Schulzer M, FitzGerald JM. A meta-analysis of the effect of Bacille Calmette Guérin vaccination on tuberculin skin test measurements. Thorax 2002; 57(9): 804–809

[26] Farhat M, Greenaway C, Pai M, Menzies D. False-positive tuberculin skin tests: what is the absolute effect of BCG and non-tuberculous mycobacteria? Int J Tuberc Lung Dis 2006; 10(11): 1192–1204

[27] Mahomed H, Hawkridge T, Verver S, Geiter L, Hatherill M, Abrahams DA, Ehrlich R, Hanekom WA, Hussey GD; SATVI Adolescent Study Team. Predictive factors for latent tuberculosis infection among adolescents in a high-burden area in South Africa. Int J Tuberc Lung Dis 2011; 15(3): 331–336

[28] Connell TG, Ritz N, Paxton GA, Buttery JP, Curtis N, Ranganathan SC. A three-way comparison of tuberculin skin testing, QuantiFERON-TB gold and T-SPOT.TB in children. PLoS One 2008; 3(7): e2624

[29] Gao L, Bai L, Liu J, Lu W, Wang X, Li X, Du J, Chen X, Zhang H, Xin H, Sui H, Li H, Su H, He J, Pan S, Peng H, Xu Z, Catanzaro A, Evans TG, Zhang Z, Ma Y, Li M, Feng B, Li Z, Guan L, Shen F, Wang Z, Zhu T, Yang S, Si H, Wang Y, Tan Y, Chen T, Chen C, Xia Y, Cheng S, Xu W, Jin Q; LATENTTB-NSTM study team. Annual risk of tuberculosis infection in rural China: a population-based prospective study. Eur Respir J 2016; 48(1): 168–178

[30] Joshi M, Monson TP, Joshi A, Woods GL. IFN-γ release assay conversions and reversions. Challenges with serial testing in U.S. health care workers. Ann Am Thorac Soc 2014; 11(3): 296–302

[31] Slater ML, Welland G, Pai M, Parsonnet J, Banaei N. Challenges with QuantiFERON-TB Gold assay for large-scale, routine screening of U.S. healthcare workers. Am J Respir Crit Care Med 2013; 188(8): 1005–1010

[32] Machado A Jr, Emodi K, Takenami I, Finkmoore BC, Barbosa T, Carvalho J, Cavalcanti L, Santos G, Tavares M, Mota M, Barreto F, Reis MG, Arruda S, Riley LW. Analysis of discordance between the tuberculin skin test and the interferon-γrelease assay. Int J Tuberc Lung Dis 2009; 13(4): 446–453

[33] Kobashi Y, Sugiu T, Mouri K, Obase Y, Miyashita N, Oka M. Indeterminate results of QuantiFERON TB-2G test performed in routine clinical practice. Eur Respir J 2009; 33(4): 812–815

[34] Lange B, Vavra M, Kern WV, Wagner D. Indeterminate results of a tuberculosis-specific interferon-γ release assay in immunocompromised patients. Eur Respir J 2010; 35(5): 1179–1182

[35] Zrinski Topić R, Zoričić-Letoja I, Pavić I, Dodig S. Indeterminate results of QuantiFERON-TB Gold In-Tube assay in nonimmunosuppressed children. Arch Med Res 2011; 42(2): 138–143

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag GmbH Germany