The effect of bacille Calmette-Guérin vaccination at birth on immune response in China

The effect of bacille Calmette-Guérin vaccination at birth on immune response in China

Vaccine 33 (2015) 209–213 Contents lists available at ScienceDirect Vaccine journal homepage: www.elsevier.com/locate/vaccine The effect of bacille...

740KB Sizes 0 Downloads 12 Views

Vaccine 33 (2015) 209–213

Contents lists available at ScienceDirect

Vaccine journal homepage: www.elsevier.com/locate/vaccine

The effect of bacille Calmette-Guérin vaccination at birth on immune response in China Yu Pang a,b,1 , Wanli Kang a,1 , Aihua Zhao c,1 , Guan Liu a , Weixin Du c , Miao Xu c , Guozhi Wang c,2 , Yanlin Zhao b,3 , Suhua Zheng a,∗ a

Beijing Chest Hospital, Beijing Tuberculosis and Thoracic Tumor Research Institute, Capital Medical University, Beijing, China National Center for Tuberculosis Control and Prevention, Chinese Center for Disease Control and Prevention, Beijing, China c National Institute for Food and Drug Control, Beijing, China b

a r t i c l e

i n f o

Article history: Received 19 May 2014 Received in revised form 12 August 2014 Accepted 14 October 2014 Available online 6 November 2014 Keywords: Tuberculosis BCG Vaccination Immune response Birth

a b s t r a c t Bacille Calmette-Guérin (BCG) vaccine is still the most effective approach to prevent tuberculosis in childhood. In order to provide protection against severe forms of childhood tuberculosis, it is customary to give BCG vaccination at birth in China. Tuberculin skin testing after vaccination is usually used to evaluate the immunogenic activity and protective efficacy of the BCG. We report the results of a multi-site prospective cohort study to evaluate the immunological reactivity against BCG in four prefectural cities in China. A total of 59,022 newborn infants were vaccinated between January 2011 and March 2012, and follow-up data on 27,517 vaccinated infants were available for this study. Of these, 679 (2.5%) had PPD readings of 0–5 mm, 17,072 (62.0%) had PPD readings of 5–10 mm of induration, 8864 (32.2%) had readings of 10–15 mm, 815 (3.0%) had readings of 15–20 mm, and 87 (0.3%) had readings of >20 mm of induration. The size of PPD reaction varied significantly with the geographic location, gender, season of vaccination, and grade of hospital administering the BCG vaccine (P < 0.001). 97.8% of the infants with a BCG scar of >1 mm had a positive TST reaction. However, only 56.9% of infants without a BCG scar had a positive PPD reaction. Our results demonstrate that the BCG immunization among newborn infants in China induces satisfactory immune response. In addition, BCG scars provide a useful indicator of vaccination response in Chinese infants. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction Tuberculosis (TB) is one of the leading global causes of death from an infectious disease worldwide [1,2]. Mycobacterium bovis bacilli Calmette-Guérin (BCG), the only vaccine currently available against TB, is generally considered to provide protection against the severe forms of childhood tuberculosis, but provides unsatisfactory protection against tuberculosis in adults [3,4]. Several candidate vaccines, evaluated in clinical trials since 1990s, have induced high-level protection in animal models [1]. Unfortunately,

∗ Corresponding author. Tel.: +86 10 6954 4491; fax: +86 10 6954 4491. E-mail addresses: [email protected] (G. Wang), [email protected] (Y. Zhao), [email protected] (S. Zheng). 1 These authors contributed equally. 2 National Institute for Food and Drug Control, No.2, Tiantan Xili, Dongcheng District, Beijing, China. Tel.: +86 10 5877 6543; fax: +86 10 5877 6543. 3 National Center for Tuberculosis Control and Prevention, Chinese Center for Disease Control and Prevention, No. 155, Changbai Road, Changping District, Beijing, China. Tel.: +86 10 5890 0777; fax: +86 10 5890 0778. http://dx.doi.org/10.1016/j.vaccine.2014.10.030 0264-410X/© 2014 Elsevier Ltd. All rights reserved.

the most promising vaccine, modified Vaccinia Ankara virus expressing antigen 85A (MVA85A), failed to enhance the protective efficacy of BCG in infants [3]. Thus, to date, BCG vaccination is still the most effective approach in preventing childhood tuberculosis. China has the second largest population of tuberculosis patients globally [5], and more than 40% of population is infected with M. tuberculosis. Since the early 1950s, the BCG strain Copenhagen 1331 has been widely used to vaccinate newborn infants [6]. Tuberculin skin tests (TST) are administered within 8–12 weeks after vaccination to evaluate the immunogenic activity and protective efficacy of the BCG [7]. Several previous studies have demonstrated that the subsequent reactivity of BCG vaccination can vary depending on injection dose, manufacturers of the vaccine, the latitude of residence, and human genetic factors [8–15]. Here, we report the results of a multi-site prospective cohort study evaluating the immunological reactivity against BCG in newborns in China. Our aim was to investigate the factors associated with TST positivity and the relationship between BCG scar and PPD reaction among newborn infants.

210

Y. Pang et al. / Vaccine 33 (2015) 209–213

2. Methods and materials

All the staff in this study were trained and approved by the National Clinical Center on Tuberculosis.

2.1. Study design and sample size The Epidemiological Research Office of Beijing Tuberculosis and Thoracic Tumor Research Institute, affiliated with the Chinese Center for Disease Control and Prevention, was responsible for implementing this study. From January 2011 to March 2012, we carried out this study in four districts of China: Haidian District (January 2011–December 2011) and Changping District (January 2011–December 2011) of Beijing Municipality, Jiulongpo District (April 2011–March 2012) of Chongqing Municipality, and Changzhou City (February 2011–January 2012) of Jiangsu Province (Fig. 1). These districts represent eastern, central, and western regions of China, respectively. There were 3, 5, 12, and 6 hospitals enrolled from Haidian District,Changping District, Jiulongpo District, and Changzhou City, respectively. All newborn infants received the BCG vaccine at birth in the hospital where they were born, according to Chinese policy. Vaccinated infants were followed up after 3 months. A tuberculin skin test was used to detect the immune efficiency, and the presence of a BCG scar was also recorded. Reports of adverse reaction to the vaccination was obtained by interviewing the infants’ parents. The sample size calculation assumed a 1% rate of negative response to BCG and a 95% confidence interval (˛ = 0.05). Based on this calculation, our enrollment target was at least 38,032 BCGvaccinated infants.

2.2. BCG vaccine and tuberculin skin test All vaccines were produced and supplied by Chengdu Institute of Biological Products, which is responsible for the supply of BCG vaccine for China. The vaccine produced by this institute has been certified by local authorities and the World Health Organization. Vaccine was prepared using Danish strain BCG823. All vaccine was transported to the local hospitals by a cold chain system, and then stored in the refrigerator. A Mantoux test with purified protein derivative was used to measure the tuberculin skin test reaction at 3 months of age. The test was performed at the local Center for Disease Control and Prevention (CDC) or TB dispensary. Two measurements of induration diameter were taken after 48–72 h [16]. The mean of the two measurements was calculated for subsequent data analysis. 2.3. Definitions Hospitals in China are classified into three levels: third-class hospitals are those with more than 500 beds, second-class hospitals are those with 100–499 beds, and first-class hospitals those with fewer than 100 beds [17]. Infants were considered to have BCG scars if the scar measured >1 mm. PPD positivity was defined as an induration diameter of >5 mm.

Fig. 1. Distribution map of vaccinated infants included in this study. The number following the pilot name represents the number enrolled in each pilot. The outer pie charts show the composition of boys and girls in each pilot.

Y. Pang et al. / Vaccine 33 (2015) 209–213

The four quarters that make up the year are: January, February and March (the first quarter); April, May and June (the second quarter); July, August and September (the third quarter); and October, November and December (the fourth quarter). 2.4. Data analysis All collected data were entered with Epi Data 3.02 software. In order to ensure accuracy, data were entered by two individuals, and 10% of entered data was randomly selected and rechecked. Analysis was performed using SPSS14.0 (SPSS Inc., USA). Pearson’s chi-square test, Fisher’s exact test, and T test were performed to compare categorical variables. The Pearson correlation coefficient was used to quantify the degree of correlation between the PPD reaction size and BCG scar size. Differences with a P value of <0.05 were considered statistically significant. 2.5. Ethics The protocols applied in this study were approved by the Ethics Committee of Beijing Tuberculosis and Thoracic Tumor Research Institute. All parents of newborn infants enrolled in the study provided written informed consent. 3. Results A total of 59,022 newborn infants were vaccinated in the four study sites between January 2011 and March 2012. Follow-up data on 27,517 vaccinated infants were available for this study. Of these, 12,588 (45.7%) were female, and 14,929 (54.3%) were male. Table 1 shows the distribution of infants among the four study sites. 3.1. Tuberculin Skin Test Of the 27,517 vaccinated infants, 679 (2.5%) had PPD readings of 0–5 mm of induration, 17,072 (62.0%) had readings of 5–10 mm, 8864 (32.2%) had readings of 10–15 mm, 815 (3.0%) had readings of 15–20 mm, and 87 (0.3%) had readings of >20 mm. The distribution of induration sizes was significantly different among the four study sites (Table 1). The proportion of infants with 5–10 mm of induration in Changping District was more than 95%, while the 10–15 mm of induration was the predominant

211

type of PPD reaction in Jiulongpo District (60.1%). The mean PPD induration sizes of Changping, Haidian, Changzhou and Jiulongpo were 6.5 ± 1.3, 9.4 ± 2.8, 8.0 ± 2.4 and 11.0 ± 3.3 mm, respectively. The statistic analysis showed that the differences of mean PPD sizes between any two groups were all significant (P < 0.001). In addition, there was a small but statistically significant difference in mean PPD size between two gender group; 8.9 ± 3.1 mm and 8.7 ± 3.1 in boys and girls groups, respectively (P < 0.001). We also analyzed the distribution PPD induration size among different quarters. As shown in Table 1, the mean PPD induration of the infants who had BCG vaccination during the first quarter was 8.6 ± 3.1 mm, which was significantly lower than that of the second (8.9 ± 3.2, P < 0.001), third (9.1 ± 3.0, P < 0.001), and fourth quarter (8.9 ± 2.9, P < 0.001), respectively. The infants vaccinated in the third quarter harbored the largest mean PPD induration size. Statistic analysis results also showed that the difference between the second and fourth quarters was not significant (P = 0.587). Comparison of the mean induration by the hospital-level where the new-born received the BCG vaccination showed that the size of the reaction in the first-class hospital (9.4 ± 3.3) was significantly bigger than that in the second-class hospital (8.7 ± 3.1) (P < 0.001). When the PPD reaction sizes of the two groups, second-class group and third-class group (8.2 ± 2.5), were compared, this showed significant differences (P < 0.001) (Table 2).

3.2. BCG scar and skin test status Ninety-nine per cent (27,293/27,517) of all vaccinated infants had a BCG scar. We analyzed the relationship between BCG scar and PPD induration diameter. In order to exclude possible false positive PPD results caused by tuberculosis infection in the community rather than vaccine inoculation, infants with a PPD induration larger than 15 mm were excluded from analysis. Of the remaining infants, 97.8% of those with a BCG scar had a positive PPD reaction (Table 3). However, out of the group without BCG scar, one hundred and twenty-six of 224 (56.9%) infants had a positive PPD reaction. In an analysis of the relationship between the PPD reaction size and BCG scar size, we found that the size of the tuberculin skin test was not associated with that of the BCG scar.

Table 1 PPD induration size by pilot, gender, and quarter. Characteristics

N

Mean size of PPD (±SDa ) (mm)

No. of infants with different PPD induration size (%) 0–5 mm

Pilotb Changping Haidian Changzhou Jiulongpo Genderc Boys Girls Quarterd First Second Third Fourth Total a

5–10 mm

10–15 mm

>15 mm

5401 5278 9239 7599

6.5 ± 1.3 9.4 ± 2.8 8.0 ± 2.4 11.0 ± 3.3

2 (0.0) 55 (1.0) 383 (4.2) 239 (3.2)

5136 (95.1) 2841 (53.8) 7053 (76.3) 2042 (26.9)

260 (4.8) 2290 (43.4) 1749 (18.9) 4565 (60.1)

3 (0.1) 92 (1.8) 54 (0.5) 753 (9.1)

14929 12588

8.9 ± 3.1 8.7 ± 3.1

358 (2.4) 321 (2.6)

9153 (61.3) 7919 (62.9)

4931 (33.0) 3933 (31.2)

487 (3.2) 415 (3.3)

9412 6272 5713 6120 27517

8.6 ± 3.1 8.9 ± 3.2 9.1 ± 3.0 8.9 ± 2.9 8.8 ± 3.1

317 (3.4) 153 (2.4) 99 (1.7) 110 (1.8) 679 (2.5)

6082 (64.6) 3716 (59.2) 3377 (59.1) 3897 (63.7) 17072 (62.0)

2724 (28.9) 2177 (34.7) 2055 (36.0) 1908 (31.2) 8864 (32.2)

289 (3.0) 226 (3.7) 182 (3.2) 205 (3.3) 902 (3.3)

SD–standard deviation. The statistics values for pilot are as follows: P < 0.001 (Changping vs. Haidian), P < 0.001 (Changping vs. Changzhou), P < 0.001 (Changping vs. Jiulongpo), P < 0.001 (Haidian vs. Changzhou), P < 0.001 (Haidian vs. Jiulongpo) and P < 0.001 (Changzhou vs. Jiulongpo). c The statistics value for gender is as follows: P < 0.001. d The four quarters that make up the year are: January, February and March (the first quarter); April, May and June (the second quarter); July, August and September (the third quarter); and October, November and December (the fourth quarter). The statistics values for quarter are as follows: P < 0.001 (first vs. second), P < 0.001 (first vs. third), P < 0.001 (first vs. fourth), P = 0.006 (Second vs. Third), P = 0.578 (second vs. fourth) and P = 0.566 (third vs. fourth). b

212

Y. Pang et al. / Vaccine 33 (2015) 209–213

Table 2 Distribution of diameter of PPD among the infants injected in different class hospitals. Class of hospitala

Mean size of PPD (±SDb ) (mm)

9.4 ± 3.3 8.7 ± 3.1 8.2 ± 2.5 8.8 ± 3.1

First-classc Second-class Third-class Total a b c

Diameter of PPD (%)

Total

0–5 mm

5–15 mm

>15 mm

217 (3.2) 263 (1.6) 199 (4.2) 679 (2.5)

6247 (92.8) 15388 (96.0) 4543 (95.4) 26178 (95.1)

269 (4.0) 373 (2.3) 18 (4.0) 660 (2.4)

6733 16024 4760 27517

The clinical hospitals was classified into three class according to the government documents in China. The third-class hospital was the top-level hospital. SD—standard deviation. The statistics values for pilot are as follows: P < 0.001 (first-class vs. second-class), P < 0.001 (First-class vs. Third-class), and P < 0.001 (second-class vs. third-class).

Table 3 Performance of BCG scar for evaluating the effect of BCG. Scar

0–1 mm >1 mm Total

Diameter of PPD (%)

Total

0–5 mm

5–15 mm

93 (43.1) 586 (2.2) 679

126 (56.9) 25810 (97.8) 25936

219 26396 26615

3.3. Adverse reactions A total of 19 adverse reactions were reported in this study, corresponding to an incidence of adverse reactions of 0.07%. Lymphadenitis accounted for 11 (57.9%) of these reactions, with a median time to onset of 57 days after vaccination (range, 21–88 days). Dermatitis was reported in five (26.3%) of the infants with adverse reactions, with a median time to onset of 2 days after vaccination (range, 1–24 days). In addition, there was one infant each with abscess, fever, and vomit. 4. Discussion In several countries with tuberculosis epidemics, including China, neonatal BCG vaccination is still the major tool for preventing severe forms of tuberculosis in children [4]. Following BCG vaccination, tuberculin reactivity has been the most common measure of the immune efficacy of BCG vaccine [7]. Our results suggest that more than 95% of BCG-vaccinated infants in China had positive PPD reactivity, which is in contrast to the results of several studies that demonstrated high rates of PPD negativity in children despite neonatal BCG vaccination [4]. One possible explanation for this difference may be that the efficacy of BCG declines with time, resulting in more negative PPD reactions in older children [18]. Although the proportion of vaccinated infants with no immune response to the BCG vaccine is low (2.5%), given the large population of China, an estimated 401,000 vaccinated infants per year in China will receive no protection from the BCG vaccine. Children without a positive tuberculin skin test-reaction have substantially higher mortality during follow-up [16,19,20]. Hence, revaccination may be needed in some children to overcome the lack immune response to the first vaccination. Of infants with positive PPD reactions, 902 (3.3%) had indurations >15 mm, which may be attributable infection with M. tuberculosis rather than BCG vaccination [21,22]. This observation indicates that not all adult tuberculosis patients are currently being identified and treated, resulting in the infection of infants in household. In China, laboratory diagnosis of pulmonary tuberculosis continues to rely on direct sputum smear microscopy, which has poor sensitivity [23]. A major effort to improve laboratory capacity, especially in resource-poor settings, is needed. Fortunately, based on the national plan for tuberculosis laboratory construction in China, 80% of county-level laboratories will obtain the capability of mycobacterial culture by 2015 [24], and GeneXpert may serve as

an important supplement to mycobacterial culture in these laboratories. Thus, with expanded laboratory capacity for tuberculosis diagnosis, better case-finding and early treatment of adult tuberculosis patients should reduce the risk of tuberculosis infection among infants. Several factors were associated with induration size in our analysis. The differences among study sites may be attributable to the differing prevalences and types of non-tuberculous mycobacteria (NTM) in these locations [11]. For example, the higher prevalence of NTM in Changzhou may have been one reason for the larger size of PPD reactions observed [25–28]. The association between male sex and increased PPD reaction size has been reported in a previous study [29]. However, our observation that induration size of infants vaccinated in the first quarter was significantly smaller than those of other quarters has not been reported in previous studies. One possible explanation for lower temperatures in the winter are associated with poor immune status of infants born in the first quarter. In addition, this temporal variation could have been caused by batch-to-batch variation in BCG vaccine. Both BCG scarring and tuberculin skin testing can provide useful measures of an individual’s immune response to BCG vaccination [30]. In the present study, our findings demonstrate that the presence of a BCG scar is a good indicator of successful immunization among vaccinated infants, but lack of a scar is not predictive of a poor immune response. In China, the tuberculin skin test is widely used to evaluate the immune response to BCG newborns. Considering the complexity of administering and reading the test [31], our results suggest that the tuberculin skin test need only used in vaccinated infants without BCG scars. The incidence of adverse reactions (0.07%) in the present study was lower than what has been reported previously [32–35]. The experience of the staff performing the BCG vaccination, each of whom typically vaccinates more than 100 newborns per year, likely contributed to the low incidence of adverse reactions [35]. However, because the occurrence of adverse reactions was only obtained by interviewing infants’ parents, the incidence of adverse reactions may have been underestimated. In conclusion, our data demonstrate that the BCG vaccination among newborn infants in China provides satisfactory immune response, as evaluated by tuberculin skin test. In addition, adverse reactions following vaccination are uncommon. The BCG scar may serve as good indicator of vaccination response, precluding the need for tuberculin skin testing. Further follow up of vaccinated infants will be carried out to investigate the protective effect of BCG against tuberculosis in settings of high tuberculosis prevalence.

Acknowledgements The study was supported by the National Key Project (2008ZX10003-009). We thank the BCG vaccination provides who contributed data for this study. We also would like to thank all the staffs in the pilots.

Y. Pang et al. / Vaccine 33 (2015) 209–213

References [1] Andersen P, Doherty TM. The success and failure of BCG—implications for a novel tuberculosis vaccine. Nat Rev Microbiol 2005;3:656–62. [2] Lienhardt C, Glaziou P, Uplekar M, Lonnroth K, Getahun H, Raviglione M. Global tuberculosis control: lessons learnt and future prospects. Nat Rev Microbiol 2012;10:407–16. [3] Tameris MD, Hatherill M, Landry BS, Scriba TJ, Snowden MA, Lockhart S, et al. Safety and efficacy of MVA85A, a new tuberculosis vaccine, in infants previously vaccinated with BCG: a randomised, placebo-controlled phase 2b trial. Lancet 2013;381:1021–8. [4] Mudido PM, Guwatudde D, Nakakeeto MK, Bukenya GB, Nsamba D, Johnson JL, et al. The effect of bacille Calmette-Guerin vaccination at birth on tuberculin skin test reactivity in Ugandan children. Int J Tuberc Lung Dis 1999;3:891–5. [5] World Health Organization. Global tuberculosis report 2012. Geneva, Switherland: World Health Organization; 2012. [6] Wang GZ, Balasubramanian V, Smith DW. The protective and allergenic potency of four BCG substrains in use in China determined in two animal models. Tubercle 1988;69:283–91. [7] Menzies D. What does tuberculin reactivity after bacille Calmette-Guerin vaccination tell us? Clin Infect Dis 2000;31:S71–4. [8] Karalliedde S, Katugaha LP, Uragoda CG. Tuberculin response of Sri Lankan children after BCG vaccination at birth. Tubercle 1987;68:33–8. [9] Padungchan S, Konjanart S, Kasiratta S, Daramas S, ten Dam HG. The effectiveness of BCG vaccination of the newborn against childhood tuberculosis in Bangkok. Bull WHO 1986;64:247–58. [10] Santosa G, Syamsuri MM, Gusti I, Djelantik G, Projogo E, Nyoman IG, et al. Difference in severity of tuberculosis in children with or without a BCG scar. Paediatr Indones 1985;25:87–92. [11] Swai OB, Aluoch JA, Kamunvi F, Agwanda R, Kwamanga D, Kenya PR. A national BCG scar survey in the estimation of BCG vaccination coverage in Kenya. East Afr Med J 1985;62:842–51. [12] Fine PE. Variation in protection by BCG: implications of and for heterologous immunity. Lancet 1995;346:1339–45. [13] Fine PE, Rodrigues LC. Modern vaccines. Mycobacterial diseases. Lancet 1990;335:1016–20. [14] Colditz GA, Brewer TF, Berkey CS, Wilson ME, Burdick E, Fineberg HV, et al. Efficacy of BCG vaccine in the prevention of tuberculosis. Meta-analysis of the published literature. J Am Med Assoc 1994;271:698–702. [15] Wilson ME, Fineberg HV, Colditz GA. Geographic latitude and the efficacy of bacillus Calmette-Guerin vaccine. Clin Infect Dis 1995;20:982–91. [16] Roth A, Sodemann M, Jensen H, Poulsen A, Gustafson P, Weise C, et al. Tuberculin reaction, BCG scar, and lower female mortality. Epidemiology 2006;17:562–8. [17] Yip WC, Hsiao W, Meng Q, Chen W, Sun X. Realignment of incentives for healthcare providers in China. Lancet 2010;375:1120–30.

213

[18] Sterne JA, Rodrigues LC, Guedes IN. Does the efficacy of BCG decline with time since vaccination. Int J Tuberc Lung Dis 1998;2:200–7. [19] Garly ML, Martins CL, Bale C, Balde MA, Hedegaard KL, Gustafson P, et al. BCG scar and positive tuberculin reaction associated with reduced child mortality in West Africa. A non-specific beneficial effect of BCG? Vaccine 2003;21:2782–90. [20] Roth A, Gustafson P, Nhaga A, Djana Q, Poulsen A, Garly ML, et al. BCG vaccination scar associated with better childhood survival in Guinea-Bissau. Int J Epidemiol 2005;34:540–7. [21] Yan B, Dumu H. Tuberculosis. Beijing: Beijing Press; 2003. p. 45–70. [22] Joncas JH, Robitaille R, Gauthier T. Interpretation of the PPD skin test in BCGvaccinated children. Can Med Assoc J 1975;113:127–8. [23] Xia H, Song YY, Zhao B, Kam KM, O’Brien RJ, Zhang ZY, et al. Multicentre evaluation of Ziehl–Neelsen and light-emitting diode fluorescence microscopy in China. Int J Tuberc Lung Dis 2013;17:107–12. [24] The General Office of the State Council. Five-year plan for tuberculosis control and prevention from 2011 to 2015 in China; 2011. [25] Wang X, Li H, Jiang G, Zhao L, Ma Y, Javid B, et al. Prevalence and drug resistance of nontuberculous mycobacteria, northern China, 2008–2011. Emerg Infect Dis 2014;20:1252–3. [26] Jing H, Wang H, Wang Y, Deng Y, Li X, Liu Z, et al. Prevalence of nontuberculous mycobacteria infection, China, 2004–2009. Emerg Infect Dis 2012;18:527–8. [27] Wang HX, Yue J, Han M, Yang JH, Gao RL, Jing LJ, et al. Nontuberculous mycobacteria: susceptibility pattern and prevalence rate in Shanghai from 2005 to 2008. Chin Med J (Engl) 2010;123:184–7. [28] Li X, Tan S, Huang Y, Cai X, Liu Z. Analysis on the epidemiological characteristics of 812 non-tuberculous mycobacteria strains. Chin Anti-tuberculosis J 2010;32:811–4 (Chinese). [29] Roth A, Sodemann M, Jensen H, Poulsen A, Gustafson P, Gomes J, et al. Vaccination technique, PPD reaction and BCG scarring in a cohort of children born in Guinea-Bissau 2000–2002. Vaccine 2005;23:3991–8. [30] Floyd S, Ponnighaus JM, Bliss L, Warndorff DK, Kasunga A, Mogha P, et al. BCG scars in northern Malawi: sensitivity and repeatability of scar reading, and factors affecting scar size. Int J Tuberc Lung Dis 2000;4:1133–42. [31] Enarson DA. Use of the tuberculin skin test in children. Paediatr Respir Rev 2004;5:S135–7. [32] Bannon MJ. BCG and tuberculosis. Arch Dis Child 1999;80:80–3. [33] Milstien JB, Gibson JJ. Quality control of BCG vaccine by WHO: a review of factors that may influence vaccine effectiveness and safety. Bull World Health Organ 1990;68:93–108. [34] Praveen KN, Smikle MF, Prabhakar P, Pande D, Johnson B, Ashley D. Outbreak of bacillus Calmette-Guerin-associated lymphadenitis and abscesses in Jamaican children. Pediatr Infect Dis J 1990;9:890–3. [35] Turnbull FM, McIntyre PB, Achat HM, Wang H, Stapledon R, Gold M, et al. National study of adverse reactions after vaccination with bacille CalmetteGuerin. Clin Infect Dis 2002;34:447–53.