The Pediatric Infectious Disease Journal • Volume 33, Number 12, December 2014
TABLE 1. Total Number of Enterovirus With Different Clinical Diagnosis, Beijing, 2013 Virus No. Serotype Cases CVA6 EV71 CVA10 CVA16 CVA2 CVA5 CVA12 E9 CVA4 CVB4 E7 Negative Total
56 19 6 5 4 3 2 2 1 1 1 30 130
Percentage (%)
No. Cases With Different Clinical Diagnosis
43.08 14.62 4.62 3.85 3.08 2.31 1.54 1.54 0.77 0.77 0.77 23.08 100
50 mild and 6 severe HFMD 9 mild and 10 severe HFMD Mild HFMD Mild HFMD 3 mild and 1 severe HFMD Mild HFMD Mild HFMD Mild HFMD Mild HFMD Severe HFMD Mild HFMD 22 mild and 8 severe HFMD 104 mild and 26 severe HFMD
less than those in EV71 patients (P = 0.027), due to the normal glucose levels observed in most CVA6 patients (76.36%) and the increased levels identified in 61.11% of EV71 patients. Significant differences were not observed in other demographic characteristics and laboratory findings from the CVA6 and EV71 patients. Meanwhile, 100 HFMD patients (Table 1) were interviewed for the presence of onychomadesis. By definition, 13 patients displayed onychomadesis, including 12 with CVA6 infection (22.64%, 12/53) and 1 with CVA10 infection (16.67%, 1/6). Onychomadesis was significantly associated with CVA6 infection (P = 0.002).
DISCUSSION In this study, the 243 HFMD cases were mainly concentrated in suburban areas and at the edges of urban areas, and cases were rare in the central cities and outer suburban areas. Dense population and poor sanitation might contribute to the increased morbidity in these locations. Moreover, the duration of HFMD was 1 month longer (May to August) than that reported by Qian et al9 (May to July). However, this finding was not unexpected given that the temperature in August 2013 was approximately 1–2°C higher than in previous years. Since 1981, EV71 and CVA16 have been the major causes of HFMD in China.10 However, in this study, we found that the presence of EV71 and CVA16 decreased dramatically and that CVA6 was the main causative agent of HFMD in Beijing in 2013. To our knowledge, this is the first study to report HFMD caused by CVA6 in mainland China. CVA6 is one of the major causes of herpangina.4,11 However, the detection rate of CVA6 in HFMD patients is gradually increasing, and it has now become the main etiologic agent of HFMD in Beijing. Moreover, we found that CVA6 could also cause severe cases of HFMD, and the manifestations of these cases are very difficult to distinguish from EV71 infection. It is likely that CVA6 possesses a pathogenic mechanism similar to EV71. Hence, changes in both the HFMD etiologic agent and clinical manifestations caused by CVA6 are worthy of attention. Onychomadesis is an acute, painless and noninflammatory disease.12 Several onychomadesis outbreaks linked to HFMD have been reported.3–6 In our study, we reported that CVA6 infection was significantly associated with onychomadesis (P = 0.002), which suggested that CVA6 might serve as the main causative agent of onychomadesis after HFMD; however, this hypothesis requires further study. In addition, both the fever duration and glucose levels of CVA6 patients were significantly reduced compared with EV71 patients, suggesting that the symptoms caused by CVA6 were relatively milder. However, the mean serum C-reactive protein levels © 2014 Lippincott Williams & Wilkins
HFMD and Coxasackievirus A6
of CVA6 patients were considerably higher (P < 0.001), which was not consistent with the illness severity. Hence, the severity of CVA6 HFMD could not be predicted by the elevated C-reactive protein level, which was consistent with a report from Taiwan.13 In all, this study is the first to report a significant change in the HFMD etiologic agents in Beijing in 2013, which is worthy of attention and further research.
ACKNOWLEDGMENTS The authors thank Qu Pei, Liu Yanan, Zhang Ling, Zhan Yongjing, Zhang Baomin and Ji Tianjiao for excellent technical assistance. The authors also thank the team in the Department of Infectious Diseases and the Department of Pediatrics of Beijing Ditan Hospital. REFERENCES 1. Zhang Y, Tan XJ, Wang HY, et al. An outbreak of hand, foot, and mouth disease associated with subgenotype C4 of human enterovirus 71 in Shandong, China. J Clin Virol. 2009;44:262–267. 2. Gopalkrishna V, Patil PR, Patil GP, et al. Circulation of multiple enterovirus serotypes causing hand, foot and mouth disease in India. J Med Microbiol. 2012;61(pt 3):420–425. 3. Osterback R, Vuorinen T, Linna M, et al. Coxsackievirus A6 and hand, foot, and mouth disease, Finland. Emerg Infect Dis. 2009;15:1485–1488. 4. Wei SH, Huang YP, Liu MC, et al. An outbreak of coxsackievirus A6 hand, foot, and mouth disease associated with onychomadesis in Taiwan, 2010. BMC Infect Dis. 2011;11:346. 5. Bracho MA, González-Candelas F, Valero A, et al. Enterovirus co-infections and onychomadesis after hand, foot, and mouth disease, Spain, 2008. Emerg Infect Dis. 2011;17:2223–2231. 6. Fujimoto T, Iizuka S, Enomoto M, et al. Hand, foot, and mouth disease caused by coxsackievirus A6, Japan, 2011 [letter]. Emerg Infect Dis. 2012;18:337–339. 7. Feder HM Jr, Bennett N, Modlin JF. Atypical hand, foot, and mouth disease: a vesiculobullous eruption caused by Coxsackie virus A6. Lancet Infect Dis. 2014;14:83–86. 8. Oberste MS, Maher K, Williams AJ, et al. Species-specific RT-PCR amplification of human enteroviruses: a tool for rapid species identification of uncharacterized enteroviruses. J Gen Virol. 2006;87(pt 1):119–128. 9. Qian HQ, Tian H, Li XT, et al. Epidemiologic investigation of hand foot and mouth disease in Beijing from 2007 to 2012 [in Chinese]. Int J Virol, 2013; 20:6–10. 10. Wu J. [Epidemiology of hand, foot and mouth disease and severe enterovirus infection]. Clin Pediatr Emerg Med. 2008;15:100–102. 11. YamashitaT, Ito M, Taniguchi A, et al. Prevalence of coxsackievirus A5, A6, and A10 in patients with herpangina in Aichi Prefecture, 2005. Jpn J Infect Dis. 2005;58:390–391. 12. Clementz GC, Mancini AJ. Nail matrix arrest following hand-foot-mouth disease: a report of five children. Pediatr Dermatol. 2000;17:7–11. 13. Lo SH, Huang YC, Huang CG, et al. Clinical and epidemiologic features of Coxsackievirus A6 infection in children in northern Taiwan between 2004 and 2009. J Microbiol Immunol Infect. 2011;44:252–257.
INCREASED RISK OF MYCOBACTERIUM TUBERCULOSIS INFECTION IN HOUSEHOLD CHILD CONTACTS EXPOSED TO PASSIVE TOBACCO SMOKE Saranya Sridhar, MBBS, MS, DPhil, DFPH,* Nisha Karnani, BA,* David W. Connell, MRCP,* Kerry A. Millington, DPhil,* Davinder Dosanjh, MD,* Mustafa Bakir, MD,† Ahmet Soysal, MD,† Jonathan Deeks, DPhil,‡ and Ajit Lalvani, DM* Abstract: Risk factors associated with Mycobacterium tuberculosis infection were investigated in a prospective cohort of household child tuberculowww.pidj.com | 1303
The Pediatric Infectious Disease Journal • Volume 33, Number 12, December 2014
Sridhar et al
sis contacts. A significantly increased risk of acquiring infection was associated with exposure to passive cigarette smoke, higher number of index cases, younger age and reduced household monthly income. Key Words: childhood tuberculosis, infection, interferon-gamma release assay, smoking Accepted for publication July 29, 2014. From the *Tuberculosis Research Centre, Department of Respiratory Medicine, National Heart and Lung Institute, Imperial College London, Norfolk Place, London; †Department of Paediatrics, Marmara University School of Medicine, Istanbul, Turkey; and ‡Department of Public Health, Epidemiology and Biostatistics, Public Health School of Health and Population Sciences, University of Birmingham, Edgbaston, United Kingdom The authors have no conflicts of interest to disclose. This study was funded by the UNICEF/UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR) and the Wellcome Trust. A.L. is a Wellcome Senior Research Fellow in Clinical Science. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Ethical approval was granted by the IRB of Marmara University School of Medicine, Istanbul; The Turkish Ministry of Health, Ankara and the WHO Steering Committee on research involving human subjects, Geneva. Address for correspondence: Ajit Lalvani, MA, DM, FRCP, Tuberculosis Research Centre, Department of Respiratory Medicine, National Heart and Lung Institute, Imperial College London, 2 Norfolk Place, London W2 1PG, United Kingdom. E-mail: [email protected]. Copyright © 2014 by Lippincott Williams & Wilkins DOI: 10.1097/INF.0000000000000506
A
n estimated, one third of the global population is infected with Mycobacterium tuberculosis (Mtb)1 with 8.6 million incident cases in 2012 alone.2 Individuals with latent tuberculosis infection, at risk of progression to active tuberculosis (TB), act as a significant reservoir of incident cases and are therefore a major target for TB control programs.3 Control of latent tuberculosis infection necessitates knowledge of biological and environmental risk factors that determine acquisition of infection to allow design and implementation of cost-effective interventions. While host characteristics such as age, exposure to an infectious case, bacillus Calmette–Guérin (BCG) vaccination and ethnicity have been well studied,4 data on environmental and socioeconomic determinants more amenable to intervention are more limited. Investigating these risk factors in children, who are most susceptible to developing severe disseminated TB is therefore of public health importance. Accurate diagnosis of Mtb infection is a barrier to investigating risk factors for Mtb infection. The challenges in identifying risk factors lie in distinguishing infection from progression to disease, use of the less specific tuberculin skin test (TST) rather than the interferon-gamma release assay (IGRA) and testing for Mtb infection at a single time point when using IGRAs with potential of misclassifying individuals who subsequently revert or convert their IGRA results at a later time point.5 We report a new analysis of a previously described child cohort of household TB contacts6,7 incorporating hitherto unavailable follow-up data. Use of a 2-time point IGRA screen offering a more robust diagnosis of Mtb infection, by excluding IGRA reverters and correctly classifying IGRA converters, is used to identify risk factors associated with acquisition of infection in children exposed to TB in the household.
METHODS Study Population We used data from a Turkish cohort described previously7 in which childhood contacts of sputum smear-positive TB cases were recruited. All participants were tested with TST and an ex-vivo
1304 | www.pidj.com
IFN-g ELISpot assay at the time of recruitment and a second ELISpot assay was repeated 6 months later. At enrollment, BCG vaccination status was assessed by the presence of a scar and demographic and socioeconomic information recorded. Details of the follow-up and incident cases have been previously published.6
Ex vivo IFN-g ELISPOT Assay ELISpot was performed as described previously using ESAT-6 and CFP-10 peptides7 which is comparable to the commercially available T-Spot.TB IGRA. Responses were scored as positive if the test wells contained a mean of at least 5 spot-forming cells more than the mean of the negative control wells, and, additionally, this number was at least twice the mean of the negative control wells.
Statistical Analysis The primary outcome for this analysis is TB infection assessed by IGRA results at baseline and 6-month follow-up time point. Individuals with TB infection were defined as those with either persistent positive IGRA result (at baseline and 6-months) or IGRA converters (IGRA negative at baseline and IGRA positive at 6 months). Those without TB infection were defined by persistent negative IGRA (at baseline and 6-months). Individuals with an IGRA reversion (IGRA positive at baseline and IGRA negative at 6 months) were excluded as the clinical and biological significance of an IGRA reversion is uncertain. Descriptive statistics were used to describe characteristics of the 2 groups in each analysis, TB infection versus no TB infection, including frequencies (percentages) and means with standard deviations. Where relevant, the Fisher exact test or the χ2 test were used to compare proportions and the independent groups’ t test was used to compare means between the 2 groups. A multivariable logistic regression model was built to investigate the effect of different risk factors on TB infection and progression to active disease. Before fitting the main model, univariable logistic regression models were used to investigate the association between variables and IGRA reversion. A model was then fitted containing variables associated with TB infection on univariate regression (P < 0.2) along with age and sex (irrespective of their P value) and stepwise backward selection was conducted to remove variables 1 at a time based on Wald statistics. An overall P value of 0.05 was considered to be statistically significant. Statistical analyses were undertaken using Stata version 11 (StataCorp, College Station, TX).
RESULTS There were 714 children with available IGRA results at baseline and the 6-month follow-up time point. As defined above, 341 (47.76%) children were classified as infected with Mtb, which comprised 273 children with a persistent positive IGRA and 68 IGRA converters. Three hundred and seventy-three (52.24%) individuals had a persistent negative IGRA and were defined as not being infected with Mtb. The median age of the cohort was 7.5 years (range 1 month to 16 years). Univariate analysis of host, environmental and socioeconomic risk factors for Mtb infection (Table 1) showed a significant association (P < 0.05) with age, number of index patients in the household, having a smoker in the household, monthly household income and mother’s education. Independent risk factors for TB infection, after fitting a multivariate model, were age, BCG status, number of index patients in the household, having a smoker in the household and monthly household income (Table 1). A smoker in the household [odds ratio (OR): 1.52, 95% confidence interval (CI): 1.09–2.12) increased risk of acquiring TB infection in child contacts while a household monthly income ≥US$75 decreased risk of infection (OR: 0.55, 95% CI: 0.38–0.79). Children were more likely to be infected with © 2014 Lippincott Williams & Wilkins
The Pediatric Infectious Disease Journal • Volume 33, Number 12, December 2014
HFMD and Coxasackievirus A6
TABLE 1. Risk Factors for Mtb Infection in Child Contacts Uninfected* Host risk factors Sex‡ Female 175 (50.87%) Male 198 (53.66%) Median age in months (inter84(38–126) quartile range) BCG vaccination‡ 0 63 (45.32%) ≥1 310 (54.01%) Environmental risk factors Number of index patients in household‡ 1 369 (53.48%) >1 4 (17.39%) Smoker in the household No 138 (58.97%) Yes 235 (48.96%) Socioeconomic risk factors Monthly income‡
TABLE 1. Total Number of Enterovirus With Different Clinical Diagnosis, Beijing, 2013 Virus No. Serotype Cases CVA6 EV71 CVA10 CVA16 CVA2 CVA5 CVA12 E9 CVA4 CVB4 E7 Negative Total
56 19 6 5 4 3 2 2 1 1 1 30 130
Percentage (%)
No. Cases With Different Clinical Diagnosis
43.08 14.62 4.62 3.85 3.08 2.31 1.54 1.54 0.77 0.77 0.77 23.08 100
50 mild and 6 severe HFMD 9 mild and 10 severe HFMD Mild HFMD Mild HFMD 3 mild and 1 severe HFMD Mild HFMD Mild HFMD Mild HFMD Mild HFMD Severe HFMD Mild HFMD 22 mild and 8 severe HFMD 104 mild and 26 severe HFMD
less than those in EV71 patients (P = 0.027), due to the normal glucose levels observed in most CVA6 patients (76.36%) and the increased levels identified in 61.11% of EV71 patients. Significant differences were not observed in other demographic characteristics and laboratory findings from the CVA6 and EV71 patients. Meanwhile, 100 HFMD patients (Table 1) were interviewed for the presence of onychomadesis. By definition, 13 patients displayed onychomadesis, including 12 with CVA6 infection (22.64%, 12/53) and 1 with CVA10 infection (16.67%, 1/6). Onychomadesis was significantly associated with CVA6 infection (P = 0.002).
DISCUSSION In this study, the 243 HFMD cases were mainly concentrated in suburban areas and at the edges of urban areas, and cases were rare in the central cities and outer suburban areas. Dense population and poor sanitation might contribute to the increased morbidity in these locations. Moreover, the duration of HFMD was 1 month longer (May to August) than that reported by Qian et al9 (May to July). However, this finding was not unexpected given that the temperature in August 2013 was approximately 1–2°C higher than in previous years. Since 1981, EV71 and CVA16 have been the major causes of HFMD in China.10 However, in this study, we found that the presence of EV71 and CVA16 decreased dramatically and that CVA6 was the main causative agent of HFMD in Beijing in 2013. To our knowledge, this is the first study to report HFMD caused by CVA6 in mainland China. CVA6 is one of the major causes of herpangina.4,11 However, the detection rate of CVA6 in HFMD patients is gradually increasing, and it has now become the main etiologic agent of HFMD in Beijing. Moreover, we found that CVA6 could also cause severe cases of HFMD, and the manifestations of these cases are very difficult to distinguish from EV71 infection. It is likely that CVA6 possesses a pathogenic mechanism similar to EV71. Hence, changes in both the HFMD etiologic agent and clinical manifestations caused by CVA6 are worthy of attention. Onychomadesis is an acute, painless and noninflammatory disease.12 Several onychomadesis outbreaks linked to HFMD have been reported.3–6 In our study, we reported that CVA6 infection was significantly associated with onychomadesis (P = 0.002), which suggested that CVA6 might serve as the main causative agent of onychomadesis after HFMD; however, this hypothesis requires further study. In addition, both the fever duration and glucose levels of CVA6 patients were significantly reduced compared with EV71 patients, suggesting that the symptoms caused by CVA6 were relatively milder. However, the mean serum C-reactive protein levels © 2014 Lippincott Williams & Wilkins
HFMD and Coxasackievirus A6
of CVA6 patients were considerably higher (P < 0.001), which was not consistent with the illness severity. Hence, the severity of CVA6 HFMD could not be predicted by the elevated C-reactive protein level, which was consistent with a report from Taiwan.13 In all, this study is the first to report a significant change in the HFMD etiologic agents in Beijing in 2013, which is worthy of attention and further research.
ACKNOWLEDGMENTS The authors thank Qu Pei, Liu Yanan, Zhang Ling, Zhan Yongjing, Zhang Baomin and Ji Tianjiao for excellent technical assistance. The authors also thank the team in the Department of Infectious Diseases and the Department of Pediatrics of Beijing Ditan Hospital. REFERENCES 1. Zhang Y, Tan XJ, Wang HY, et al. An outbreak of hand, foot, and mouth disease associated with subgenotype C4 of human enterovirus 71 in Shandong, China. J Clin Virol. 2009;44:262–267. 2. Gopalkrishna V, Patil PR, Patil GP, et al. Circulation of multiple enterovirus serotypes causing hand, foot and mouth disease in India. J Med Microbiol. 2012;61(pt 3):420–425. 3. Osterback R, Vuorinen T, Linna M, et al. Coxsackievirus A6 and hand, foot, and mouth disease, Finland. Emerg Infect Dis. 2009;15:1485–1488. 4. Wei SH, Huang YP, Liu MC, et al. An outbreak of coxsackievirus A6 hand, foot, and mouth disease associated with onychomadesis in Taiwan, 2010. BMC Infect Dis. 2011;11:346. 5. Bracho MA, González-Candelas F, Valero A, et al. Enterovirus co-infections and onychomadesis after hand, foot, and mouth disease, Spain, 2008. Emerg Infect Dis. 2011;17:2223–2231. 6. Fujimoto T, Iizuka S, Enomoto M, et al. Hand, foot, and mouth disease caused by coxsackievirus A6, Japan, 2011 [letter]. Emerg Infect Dis. 2012;18:337–339. 7. Feder HM Jr, Bennett N, Modlin JF. Atypical hand, foot, and mouth disease: a vesiculobullous eruption caused by Coxsackie virus A6. Lancet Infect Dis. 2014;14:83–86. 8. Oberste MS, Maher K, Williams AJ, et al. Species-specific RT-PCR amplification of human enteroviruses: a tool for rapid species identification of uncharacterized enteroviruses. J Gen Virol. 2006;87(pt 1):119–128. 9. Qian HQ, Tian H, Li XT, et al. Epidemiologic investigation of hand foot and mouth disease in Beijing from 2007 to 2012 [in Chinese]. Int J Virol, 2013; 20:6–10. 10. Wu J. [Epidemiology of hand, foot and mouth disease and severe enterovirus infection]. Clin Pediatr Emerg Med. 2008;15:100–102. 11. YamashitaT, Ito M, Taniguchi A, et al. Prevalence of coxsackievirus A5, A6, and A10 in patients with herpangina in Aichi Prefecture, 2005. Jpn J Infect Dis. 2005;58:390–391. 12. Clementz GC, Mancini AJ. Nail matrix arrest following hand-foot-mouth disease: a report of five children. Pediatr Dermatol. 2000;17:7–11. 13. Lo SH, Huang YC, Huang CG, et al. Clinical and epidemiologic features of Coxsackievirus A6 infection in children in northern Taiwan between 2004 and 2009. J Microbiol Immunol Infect. 2011;44:252–257.
INCREASED RISK OF MYCOBACTERIUM TUBERCULOSIS INFECTION IN HOUSEHOLD CHILD CONTACTS EXPOSED TO PASSIVE TOBACCO SMOKE Saranya Sridhar, MBBS, MS, DPhil, DFPH,* Nisha Karnani, BA,* David W. Connell, MRCP,* Kerry A. Millington, DPhil,* Davinder Dosanjh, MD,* Mustafa Bakir, MD,† Ahmet Soysal, MD,† Jonathan Deeks, DPhil,‡ and Ajit Lalvani, DM* Abstract: Risk factors associated with Mycobacterium tuberculosis infection were investigated in a prospective cohort of household child tuberculowww.pidj.com | 1303
The Pediatric Infectious Disease Journal • Volume 33, Number 12, December 2014
Sridhar et al
sis contacts. A significantly increased risk of acquiring infection was associated with exposure to passive cigarette smoke, higher number of index cases, younger age and reduced household monthly income. Key Words: childhood tuberculosis, infection, interferon-gamma release assay, smoking Accepted for publication July 29, 2014. From the *Tuberculosis Research Centre, Department of Respiratory Medicine, National Heart and Lung Institute, Imperial College London, Norfolk Place, London; †Department of Paediatrics, Marmara University School of Medicine, Istanbul, Turkey; and ‡Department of Public Health, Epidemiology and Biostatistics, Public Health School of Health and Population Sciences, University of Birmingham, Edgbaston, United Kingdom The authors have no conflicts of interest to disclose. This study was funded by the UNICEF/UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR) and the Wellcome Trust. A.L. is a Wellcome Senior Research Fellow in Clinical Science. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Ethical approval was granted by the IRB of Marmara University School of Medicine, Istanbul; The Turkish Ministry of Health, Ankara and the WHO Steering Committee on research involving human subjects, Geneva. Address for correspondence: Ajit Lalvani, MA, DM, FRCP, Tuberculosis Research Centre, Department of Respiratory Medicine, National Heart and Lung Institute, Imperial College London, 2 Norfolk Place, London W2 1PG, United Kingdom. E-mail: [email protected]. Copyright © 2014 by Lippincott Williams & Wilkins DOI: 10.1097/INF.0000000000000506
A
n estimated, one third of the global population is infected with Mycobacterium tuberculosis (Mtb)1 with 8.6 million incident cases in 2012 alone.2 Individuals with latent tuberculosis infection, at risk of progression to active tuberculosis (TB), act as a significant reservoir of incident cases and are therefore a major target for TB control programs.3 Control of latent tuberculosis infection necessitates knowledge of biological and environmental risk factors that determine acquisition of infection to allow design and implementation of cost-effective interventions. While host characteristics such as age, exposure to an infectious case, bacillus Calmette–Guérin (BCG) vaccination and ethnicity have been well studied,4 data on environmental and socioeconomic determinants more amenable to intervention are more limited. Investigating these risk factors in children, who are most susceptible to developing severe disseminated TB is therefore of public health importance. Accurate diagnosis of Mtb infection is a barrier to investigating risk factors for Mtb infection. The challenges in identifying risk factors lie in distinguishing infection from progression to disease, use of the less specific tuberculin skin test (TST) rather than the interferon-gamma release assay (IGRA) and testing for Mtb infection at a single time point when using IGRAs with potential of misclassifying individuals who subsequently revert or convert their IGRA results at a later time point.5 We report a new analysis of a previously described child cohort of household TB contacts6,7 incorporating hitherto unavailable follow-up data. Use of a 2-time point IGRA screen offering a more robust diagnosis of Mtb infection, by excluding IGRA reverters and correctly classifying IGRA converters, is used to identify risk factors associated with acquisition of infection in children exposed to TB in the household.
METHODS Study Population We used data from a Turkish cohort described previously7 in which childhood contacts of sputum smear-positive TB cases were recruited. All participants were tested with TST and an ex-vivo
1304 | www.pidj.com
IFN-g ELISpot assay at the time of recruitment and a second ELISpot assay was repeated 6 months later. At enrollment, BCG vaccination status was assessed by the presence of a scar and demographic and socioeconomic information recorded. Details of the follow-up and incident cases have been previously published.6
Ex vivo IFN-g ELISPOT Assay ELISpot was performed as described previously using ESAT-6 and CFP-10 peptides7 which is comparable to the commercially available T-Spot.TB IGRA. Responses were scored as positive if the test wells contained a mean of at least 5 spot-forming cells more than the mean of the negative control wells, and, additionally, this number was at least twice the mean of the negative control wells.
Statistical Analysis The primary outcome for this analysis is TB infection assessed by IGRA results at baseline and 6-month follow-up time point. Individuals with TB infection were defined as those with either persistent positive IGRA result (at baseline and 6-months) or IGRA converters (IGRA negative at baseline and IGRA positive at 6 months). Those without TB infection were defined by persistent negative IGRA (at baseline and 6-months). Individuals with an IGRA reversion (IGRA positive at baseline and IGRA negative at 6 months) were excluded as the clinical and biological significance of an IGRA reversion is uncertain. Descriptive statistics were used to describe characteristics of the 2 groups in each analysis, TB infection versus no TB infection, including frequencies (percentages) and means with standard deviations. Where relevant, the Fisher exact test or the χ2 test were used to compare proportions and the independent groups’ t test was used to compare means between the 2 groups. A multivariable logistic regression model was built to investigate the effect of different risk factors on TB infection and progression to active disease. Before fitting the main model, univariable logistic regression models were used to investigate the association between variables and IGRA reversion. A model was then fitted containing variables associated with TB infection on univariate regression (P < 0.2) along with age and sex (irrespective of their P value) and stepwise backward selection was conducted to remove variables 1 at a time based on Wald statistics. An overall P value of 0.05 was considered to be statistically significant. Statistical analyses were undertaken using Stata version 11 (StataCorp, College Station, TX).
RESULTS There were 714 children with available IGRA results at baseline and the 6-month follow-up time point. As defined above, 341 (47.76%) children were classified as infected with Mtb, which comprised 273 children with a persistent positive IGRA and 68 IGRA converters. Three hundred and seventy-three (52.24%) individuals had a persistent negative IGRA and were defined as not being infected with Mtb. The median age of the cohort was 7.5 years (range 1 month to 16 years). Univariate analysis of host, environmental and socioeconomic risk factors for Mtb infection (Table 1) showed a significant association (P < 0.05) with age, number of index patients in the household, having a smoker in the household, monthly household income and mother’s education. Independent risk factors for TB infection, after fitting a multivariate model, were age, BCG status, number of index patients in the household, having a smoker in the household and monthly household income (Table 1). A smoker in the household [odds ratio (OR): 1.52, 95% confidence interval (CI): 1.09–2.12) increased risk of acquiring TB infection in child contacts while a household monthly income ≥US$75 decreased risk of infection (OR: 0.55, 95% CI: 0.38–0.79). Children were more likely to be infected with © 2014 Lippincott Williams & Wilkins
The Pediatric Infectious Disease Journal • Volume 33, Number 12, December 2014
HFMD and Coxasackievirus A6
TABLE 1. Risk Factors for Mtb Infection in Child Contacts Uninfected* Host risk factors Sex‡ Female 175 (50.87%) Male 198 (53.66%) Median age in months (inter84(38–126) quartile range) BCG vaccination‡ 0 63 (45.32%) ≥1 310 (54.01%) Environmental risk factors Number of index patients in household‡ 1 369 (53.48%) >1 4 (17.39%) Smoker in the household No 138 (58.97%) Yes 235 (48.96%) Socioeconomic risk factors Monthly income‡
