Secondary Logo

Journal Logo

Original Studies

Culture-positive Pediatric Tuberculosis in Toronto, Ontario

Sources of Infection and Relationship of Birthplace and Mycobacterial Lineage to Phenotype

Rayment, Jonathan H. MDCM, MSc*; Guthrie, Jennifer L. MSc; Lam, Karen BSc; Whelan, Michael MSc; Lee, Brenda MHSc; Jamieson, Frances B. MD, MHSc, FRCP(C)‡§; Kitai, Ian MB, BCh, FRCP(C)†¶

Author Information
The Pediatric Infectious Disease Journal: January 2016 - Volume 35 - Issue 1 - p 13-18
doi: 10.1097/INF.0000000000000915


The clinical manifestations of pediatric tuberculosis (TB) disease are wide ranging and distinct from adult disease.1 Host factors that contribute to TB disease phenotype include age, socioeconomic status, nutritional status and comorbid medical conditions, especially HIV infection.1 Specific polymorphisms of genes involved in the host immune response have been associated with disease phenotypes (reviewed by Di Pietrantonio and Schurr2).

Different strain types or lineages of Mycobacterium tuberculosis (Mtb) vary in virulence.3 Advances in genetic and bioinformatic techniques have allowed for standardized classification of Mtb isolates into phylogeographic lineages,4,5 and the intrinsic biology of these lineages may play a role in the clinical manifestations of TB disease.6–9 There are few data linking Mtb lineage and clinical disease phenotype in the pediatric population.10

History of host travel to a TB-endemic country is an important risk factor for TB infection and disease.11,12 However, a detailed travel history is not systematically collected on patients with TB in the US and Canada.

Our institution is a major pediatric TB referral centre in Ontario, treating approximately 15–20 new cases of TB disease per year (30%–40% of all pediatric cases in Ontario). Demographic and other data including travel history are prospectively collected at clinic entry for all TB patients. In addition, Public Health Ontario Laboratories (PHOL) have maintained a database [the Ontario Universal Typing-Tuberculosis (OUT-TB) program13] and an archive of all isolates from culture-positive TB cases in Ontario since 1997. Genotype data are available since 1997 and for 92% of culture-confirmed cases since 2007.

This study is a retrospective analysis of all culture-positive pediatric TB cases managed at our institution over an 11-year period, between 2002 and 2012. The objectives of this study were to describe the clinical and patient demographic features of these cases, to determine the Mtb lineage distribution of the case isolates and to identify potential associations of these features with disease phenotype. We also compared the Mtb lineage distribution of the case isolates in our population with the adult data collected in OUT-TB. Finally, we sought to explore the differences between foreign-born and Canadian-born patients in their disease characteristics and potential sources of TB acquisition.


Study Setting

This study was undertaken at The Hospital for Sick Children (SickKids) in Toronto, Ontario, Canada. Public health units in the Greater Toronto Area make referrals to the SickKids TB program for evaluation of childhood TB contacts and for management of TB disease.

Study Design

This was a retrospective study that included consecutive patients with a clinical diagnosis of TB and a positive culture for Mtb between January 1, 2002 and December 31, 2012. Inclusion criteria were a clinical diagnosis of TB, age younger than 18 years at the time of diagnosis, at least 1 TB clinic visit at SickKids or 1 inpatient infectious disease service consultation and at least 1 positive Mtb culture. Patients without an available M. tuberculosis isolate were excluded. Patients were identified by cross-referencing the SickKids TB Clinic database, the SickKids microbiology database and the PHOL OUT-TB database.

Ethics Approval

Ethics approval for this study was obtained from the Research Ethics Board at SickKids.

Clinical Data

Clinical data were collected retrospectively from the SickKids TB Clinic records and through the Electronic Patient Chart at SickKids. Since 2002, all patients followed by the program have data recorded at first visit on a standardized clinic intake form. Cases of extrathoracic disease were identified based on extrathoracic culture positivity or clinical evidence of extrathoracic disease and a positive culture from an intrathoracic site.

Countries of birth were categorized into geographic regions as previously described.4 The combination of the foreign-born patients’ region of birth and the Canadian-born patients’ parents’ region of birth is referred to as “ethnic community” in the analysis. TB-endemic countries were defined as those with an annual TB incidence of at least 15 per 100,000 population.14

Ontario Data Collection

Incidence of all adult and pediatric cases in Ontario, 2002–2012, was determined using public health reporting data from Ontario’s Integrated Public Health Information System. To determine culture positivity, these records were cross-referenced with the OUT-TB database, which contains records of all culture-positive cases in the province of Ontario. Mtb lineage was recorded for culture-positive adult cases in the OUT-TB database with complete spoligotype patterns.

Microbiology and Bioinformatics

All Mtb isolates were grown in liquid media (MGIT 960, Becton Dickinson, Franklin Lakes, NJ) and/or Lowenstein–Jensen solid media at PHOL. Drug susceptibility testing and DNA extraction were performed on the first culture-positive isolate for each individual patient. DNA was extracted following standard protocols.15Mtb strains were genotyped using polymerase chain reaction-based rapid typing methods, 24-locus Mycobacterial Interspersed Repetitive Units–Variable Number Tandem Repeat (MIRU-VNTR) and spoligotyping, following standard protocols.16,17Mtb lineage was estimated using the “Rules” method with the TB-insight tool (


Statistics were calculated using R (v3.1.2, R Foundation for Statistical Computing, Vienna, Austria). Significance of unadjusted odds ratios (OR) was calculated using Fisher exact test or χ2 test, where applicable. Where applicable, OR were adjusted using a multiple logistic regression model. Median values were compared using the Mann–Whitney U test.


Study Population

In the 11-year study period, 141 clinical cases of TB were treated at SickKids. Of those, 79 had positive cultures for Mtb complex (excluding Mycobacterium bovis). Of those 79 cases, 2 cases were excluded as the isolates were not available for further genotypic analysis. The remaining 77 case isolates were included in the analysis.

The TB cases diagnosed and treated at SickKids represented 29.7% (62 of 209) of the clinically diagnosed and 31.2% (79 of 253) of the culture-positive cases of pediatric TB in Ontario in the study time frame (see Table, Supplemental Digital Content 1, The rate of culture positivity in clinically diagnosed TB at SickKids was 56.0% (79 of 141), which was similar to 54.8% (253 of 462) in all other Ontario pediatric cases. By comparison, the adult culture positivity rate in Ontario of 78.4% (5409 of 6902) was significantly higher (P < 0.001) than in the overall pediatric population (see Table, Supplemental Digital Content 1,


Demographic characteristics of the study population are summarized in Table 1. The age distribution of our study population was bimodal. Age peaks were seen in early childhood, younger than 5 years (14 of 77, 18.2%), and in later childhood/adolescence, 10 years or older (58 of 77, 75.3%) with a nadir in the early school-age years, between 5 and 9 years of age (5 of 77, 6.5%). There was a slight male predominance (42 of 77, 54.5%). The majority of our patients were born outside of Canada (46 of 77, 59.7%). Most foreign-born patients were from the Indian subcontinent (24 of 46, 52.2%), East/Southeast Asia (11 of 46, 23.9%) and Africa (3 of 46, 6.5%). Of the Canadian-born patients, all but 1 (30 of 31, 96.8%) had at least 1 foreign-born parent. Overall, 76 of 77 (98.7%) of our patients were either immigrants or first-generation Canadians. Most cases in foreign-born patients were diagnosed within 5 years of immigration (28 of 46, 60.9%), with many being diagnosed within the first 2 years of immigration (17 of 46, 37.0%). Twenty-five (25 of 77, 32.5%) of the patients in the study had a history of travel to a TB-endemic country within 5 years of diagnosis, 28 (28 of 77, 36.4%) had a known clinical contact, and 32 (32 of 77, 41.6%) had a genotype match in the OUT-TB database.

Demographic and Clinical Features of the 77 Pediatric Patients with Culture-positive Tuberculosis Diagnosed at SickKids Between 2002 and 2012

A comparison of Canadian-born with foreign-born patients is shown in Table 2. Canadian-born patients were younger than foreign-born patients (median ages, 6.4 ± 6.2 and 15.5 ± 2.3 years, respectively; P < 0.001) and were more likely to have a positive genotype match in the OUT-TB database {OR = 3.2 [95% confidence interval (CI): 1.1–9.2]; P = 0.02}. There was a trend toward a higher rate of travel to TB-endemic countries in Canadian-born (14 of 31, 45.2%) when compared with foreign-born patients [11 of 46, 23.9%; OR = 2.6 (95% CI: 0.90–7.9); P = 0.05]. Of those with no reported TB contact, Canadian-born patients were more likely (12 of 15, 80.0%) than foreign-born patients (8 of 34, 23.5%) to have travelled to a TB-endemic country [OR = 13.0 (95% CI: 2.5–78.5); P < 0.001]. Similarly, in Canadian-born patients without a genotype match in OUT-TB, there was a trend toward being more likely (8 of 13, 61.5%) than foreign-born patients (8 of 34, 23.5%) to have travelled to a TB-endemic country [OR = 3.5 (95% CI: 0.77–16.9); P = 0.09].

Comparison of Demographics of Canadian-born and Foreign-born Pediatric Patients with Culture-positive TB Diagnosed at SickKids Between 2002 and 2012

Clinical Characteristics

Clinical characteristics of the study population are summarized in Table 1. Forty of 77 (51.9%) cases were exclusively intrathoracic, with no evidence of extrathoracic or pleural involvement. Cases of extrathoracic disease were identified either by a positive extrathoracic Mtb culture (36 of 37) or by clinical evidence of extrathoracic disease with a positive-intrathoracic culture (1 of 37). Sixteen of 77 (20.8%) were exclusively extrathoracic, and 21 of 77 (27.3%) were mixed intrathoracic and extrathoracic. In the 77 cases, there were 95 unique sites of positive culture (1.2 sites per case). The most common thoracic site of culture positivity was sputum (29 of 77 patients, 37.7%). The most common extrathoracic site of culture positivity was open lymph node biopsy (13 of 77, 16.8%). Case isolates were predominantly pan-sensitive (68 of 77, 88.3%). Drug resistance to 1 or more agents was identified in 9 of 77 cases (11.7%).

Microbiology and Genetics

Mtb phylogeographic lineage was successfully estimated using the Rules method of the TB-Insight tool5 for all case isolates with 24-locus MIRU-VNTR data (Table 1). The case isolates were found to be belong to 4 lineages4: Euro-American/lineage 4 (36.4%), East-African Indian/lineage 3 (29.9%), Indo-Oceanic/lineage 1 (22.1%) and East Asian/Lineage 2 (11.7%). This distribution of lineages was very similar to the overall lineage distribution of the adult TB population in Ontario in the study period (see Table, Supplemental Digital Content 2,

Clinical Associations

The association of all demographic and clinical features with disease site was assessed using univariate analysis, when compared with a dichotomized population. Using this approach, associations were identified between Mtb lineage, place of birth and ethnic community (Table 3). Because of small sample size, only the Mtb lineage association was adjusted for age and sex. This adjustment did not have a significant effect on the results, and only the unadjusted OR are reported. Complete univariate and multivariate analyses are shown in Tables (Supplemental Digital Content 3, and Supplemental Digital Content 4,

Distribution of Extrathoracic Disease (Isolated or Mixed) Based on Mtb Lineage, Patient Ethnic Community or Place of Birth

On univariate regression analysis, when compared with a dichotomous population, case isolates of the East Asian lineage were less likely [OR = 0.11 (95% CI: 0.01–0.94); P < 0.05] to be associated with extrathoracic disease. Case isolates of the Indo-Oceanic lineage were more likely [OR = 4.9 (95% CI: 1.4–16.7); P < 0.05] to be associated with extrathoracic disease, when compared with a dichotomized population. On univariate regression analysis, foreign (non-Canadian) birth was associated with extrathoracic disease [OR = 3.0 (95% CI: 1.04–8.71); P = 0.025]. In addition, inclusion in the Indian subcontinent ethnic community is negatively associated [OR = 0.30 (95% CI: 0.10–0.90); P = 0.029] with extrathoracic disease.


The 2013, the World Health Organization document “Roadmap for Childhood Tuberculosis” recommended better characterization of pediatric TB disease using modern biological and epidemiological methods and advancement of understanding of host–pathogen interaction.18 The current retrospective study included all consecutive culture-positive pediatric TB patients diagnosed at SickKids between 2002 and 2012. We identified associations of disease site with mycobacterial lineage and patient ethnic community and identified differences between Canadian-born and foreign-born patients in age of disease onset and potential sources of TB acquisition.

The proportion of foreign-born patients in our study (60%) approximates that in a previous study of TB in adolescents in Ontario19 but is significantly lower than the overall Ontario TB population.20 However, when we included the birth place of the patients’ parents in the analysis, 99% of patients were from immigrant families. In the single exceptional case, the disease was nosocomially acquired in the neonatal intensive care unit.21 By comparison, 75%–80% of children with TB in the US are foreign-born or have foreign-born parents.22,23

Clustering of TB cases within immigrant communities in low-burden countries is well described,24–26 and our data identify mechanisms by which this may occur. When compared with foreign-born patients, Canadian-born patients were more likely to have an Mtb genotype match in the OUT-TB database. Canadian-born patients without a recorded contact history or a genotype match were more likely to have travelled to a TB-endemic country. This suggests that Canadian-born patients are likely contracting their disease either from contacts in Canada or upon travel to their parents’ home country. Conversely, foreign-born patients were less likely to have a history of travel or a known genotype match or contact, suggesting that most foreign-born patients activate latent infection that was present on immigration.

These data may advise strategies to prevent pediatric TB in low-burden settings. Screening those who emigrate from high-burden countries for latent TB infection may be helpful27 but would not prevent travel-acquired infection in both Canadian-born and some foreign-born individuals. Pretravel education and pretravel and posttravel TB skin tests are potential methods to prevent travel-acquired TB in people who are planning for long-duration travel to high-burden countries.12 In addition, we propose that detailed travel history should be included in TB notification data in North America.

The rates of culture positivity in our study population and in the overall pediatric population in Ontario are lower than reported rates in adult disease,28 but similar to rates previously reported in pediatric TB.22,23 The rate of mixed and exclusive extrathoracic disease in our study was higher than the rates reported in the adult literature28 and in other pediatric studies.22,23 These higher rates of extrathoracic disease could represent a selection bias within our study because of the fact that we included only culture-positive patients, and the yield of respiratory specimen culture in pediatric TB is low.1,29,30

Although numbers are small, we found a trend toward less extrathoracic disease in children younger than 5 years (see Table, Supplemental Digital Content 3, similar to findings in another Canadian studies19,31 but different from classical descriptions of childhood TB.1,10 Our findings could partly be an effect of the immigration medical examination for Canada, which requires chest radiogram for only those older than 11 years and may identify those with untreated pulmonary disease before arrival.

When the lineage of the infecting Mtb strain and site of disease were compared (Table 3), infection with an Indo-Oceanic lineage strain was positively associated with extrathoracic disease, whereas infection with an East Asian lineage strain was negatively associated with extrathoracic disease. This association reached statistical significance when comparison was made with a dichotomized population. This finding is novel in the pediatric TB population and deserves further study.

Recently, 2 large cohort studies in the US and the United Kingdom reported associations between Mtb lineage and site of TB disease in adults.7,8 Click et al7 described results similar to ours, with Indo-Oceanic lineage having the highest rate of extrathoracic disease and East-Asian lineage having the lowest. However, Pareek et al8 described contradicting findings, with both Indo-Oceanic and East Asian lineage being associated with higher rates of extrathoracic disease. The reason for these discrepancies is unclear; however, it is possible that certain Mtb sublineages may exhibit intrinsic differences in virulence and transmissibility.32,33

On univariate analysis, foreign birth was associated with extrathoracic disease. This has been observed in some7,8,28 but not all22,23 studies of TB in North America. The reasons for this are unclear, but Canadian immigration health screening processes,27 which may identify pulmonary disease before landing, could partly explain this finding. Finally, we found that inclusion within the Indian Subcontinent ethnic community was negatively associated with extrathoracic disease in our study population. Although this association seems quite strong, it was identified on univariate analysis, and there is no clear explanation. This observation warrants further investigation.

Strengths of this study include the long time frame reviewed, the prospectively collected information on all patients upon clinic entry, the existence of all genotyping data for the province in a single database and the integration of detailed clinical and genotyping data. However, this study has several limitations. Despite the 11-year time frame, we were limited in our analyses because of a small sample size (77 cases) from a single tertiary care centre. The OUT-TB database contains complete data from 2007 onward, and thus, the reported rate of genotype match is likely underestimated. Additionally, our centre services an urban region with a high immigrant population and cares for very few patients from First Nations communities. Finally, our study included only culture-positive cases of TB, which excludes a significant number of patients with clinically diagnosed TB30 from our demographic analysis.

Despite the limitations outlined above, this study adds to the body of knowledge of pediatric TB. We document that distribution of Mtb lineages in children mirrors those for all TB patients in Ontario and is among the first to explore associations between lineage and pediatric TB phenotype. Our clinical and genotyping data highlight travel as an important mode of TB acquisition, especially in Canadian-born children. We also report associations between Mtb lineage, place of birth, ethnic community and extrathoracic disease. Although the strength of these associations requires confirmation, these findings will be useful in generating hypotheses and directing future study.

Future studies of larger pediatric populations are required to confirm these results and to better establish the association between lineage and other possible covariates such as age and ethnic community and to further explore travel-related TB in children in North America. In addition, inclusion of Canadian First Nations communities within the study population would provide valuable insight into the disease pathophysiology within this at-risk population.


1. Newton SM, Brent AJ, Anderson S, et al. Paediatric tuberculosis. Lancet Infect Dis. 2008;8:498–510
2. Di Pietrantonio T, Schurr E.. Host-pathogen specificity in tuberculosis. Adv Exp Med Biol. 2013;783:33–44
3. Valway SE, Sanchez MP, Shinnick TF, et al. An outbreak involving extensive transmission of a virulent strain of Mycobacterium tuberculosis. N Engl J Med. 1998;338:633–639
4. Gagneux S, DeRiemer K, Van T, et al. Variable host-pathogen compatibility in Mycobacterium tuberculosis. Proc Natl Acad Sci U S A. 2006;103:2869–2873
5. Shabbeer A, Cowan LS, Ozcaglar C, et al. TB-Lineage: an online tool for classification and analysis of strains of Mycobacterium tuberculosis complex. Infect Genet Evol. 2012;12:789–797
6. Thwaites GE, Chau TT, Caws M, et al. Isoniazid resistance, mycobacterial genotype and outcome in Vietnamese adults with tuberculous meningitis. Int J Tuberc Lung Dis. 2002;6:865–871
7. Click ES, Moonan PK, Winston CA, et al. Relationship between Mycobacterium tuberculosis phylogenetic lineage and clinical site of tuberculosis. Clin Infect Dis. 2012;54:211–219
8. Pareek M, Evans J, Innes J, et al. Ethnicity and mycobacterial lineage as determinants of tuberculosis disease phenotype. Thorax. 2013;68:221–229
9. Coscolla M, Gagneux S.. Does M. tuberculosis genomic diversity explain disease diversity? Drug Discov Today Dis Mech. 2010;7:e43–e59
10. Hesseling AC, Marais BJ, Kirchner HL, et al. Mycobacterial genotype is associated with disease phenotype in children. Int J Tuberc Lung Dis. 2010;14:1252–1258
11. Rieder HL.. Risk of travel-associated tuberculosis. Clin Infect Dis. 2001;33:1393–1396
12. Cobelens FG, van Deutekom H, Draayer-Jansen IW, et al. Risk of infection with Mycobacterium tuberculosis in travellers to areas of high tuberculosis endemicity. Lancet. 2000;356:461–465
13. Bolotin S, Alexander DC, Guthrie JL, et al. The Ontario universal typing of tuberculosis (OUT-TB) surveillance program–what it means to you. Can Respir J. 2010;17:e51–e54
14. Menzies D Canadian Tuberculosis Standards. 20147th ed Ottawa Canadian Thoracic Society and The Public Health Agency of Canada
15. van Soolingen D, Hermans PW, de Haas PE, et al. Occurrence and stability of insertion sequences in Mycobacterium tuberculosis complex strains: evaluation of an insertion sequence-dependent DNA polymorphism as a tool in the epidemiology of tuberculosis. J Clin Microbiol. 1991;29:2578–2586
16. Supply P, Allix C, Lesjean S, et al. Proposal for standardization of optimized mycobacterial interspersed repetitive unit-variable-number tandem repeat typing of Mycobacterium tuberculosis. J Clin Microbiol. 2006;44:4498–4510
17. Cowan LS, Diem L, Brake MC, et al. Transfer of a Mycobacterium tuberculosis genotyping method, spoligotyping, from a reverse line-blot hybridization, membrane-based assay to the Luminex multianalyte profiling system. J Clin Microbiol. 2004;42:474–477
18. World Health Organization. Roadmap for Childhood Tuberculosis: Towards Zero Deaths. 2013 Geneva, Switzerland WHO Document Production Services
19. Phongsamart W, Kitai I, Gardam M, et al. A population-based study of tuberculosis in children and adolescents in Ontario. Pediatr Infect Dis J. 2009;28:416–419
20. Tuite AR, Guthrie JL, Alexander DC, et al. Epidemiological evaluation of spatiotemporal and genotypic clustering of Mycobacterium tuberculosis in Ontario, Canada. Int J Tuberc Lung Dis. 2013;17:1322–1327
21. Crockett M, King SM, Kitai I, et al.Outbreak Investigation Team. Nosocomial transmission of congenital tuberculosis in a neonatal intensive care unit. Clin Infect Dis. 2004;39:1719–1723
22. Pang J, Teeter LD, Katz DJ, et al.Tuberculosis Epidemiologic Studies Consortium. Epidemiology of tuberculosis in young children in the United States. Pediatrics. 2014;133:e494–e504
23. Winston CA, Menzies HJ.. Pediatric and adolescent tuberculosis in the United States, 2008-2010. Pediatrics. 2012;130:e1425–e1432
24. Hirsh AE, Tsolaki AG, DeRiemer K, et al. Stable association between strains of Mycobacterium tuberculosis and their human host populations. Proc Natl Acad Sci U S A. 2004;101:4871–4876
25. Lillebaek T, Andersen AB, Bauer J, et al. Risk of Mycobacterium tuberculosis transmission in a low-incidence country due to immigration from high-incidence areas. J Clin Microbiol. 2001;39:855–861
26. Fenner L, Gagneux S, Helbling P, et al.Swiss HIV Cohort Study Group; Molecular Epidemiology of Tuberculosis Study Group. Mycobacterium tuberculosis transmission in a country with low tuberculosis incidence: role of immigration and HIV infection. J Clin Microbiol. 2012;50:388–395
27. Greenaway C, Sandoe A, Vissandjee B, et al.Canadian Collaboration for Immigrant and Refugee Health. Tuberculosis: evidence review for newly arriving immigrants and refugees. CMAJ. 2011;183:E939–E951
28. Peto HM, Pratt RH, Harrington TA, et al. Epidemiology of extrapulmonary tuberculosis in the United States, 1993-2006. Clin Infect Dis. 2009;49:1350–1357
29. Zar HJ, Hanslo D, Apolles P, et al. Induced sputum versus gastric lavage for microbiological confirmation of pulmonary tuberculosis in infants and young children: a prospective study. Lancet. 2005;365:130–134
30. Anderson ST, Kaforou M, Brent AJ, et al.ILULU Consortium; KIDS TB Study Group. Diagnosis of childhood tuberculosis and host RNA expression in Africa. N Engl J Med. 2014;370:1712–1723
31. de Pontual L, Balu L, Ovetchkine P, et al. Tuberculosis in adolescents: A French retrospective study of 52 cases. Pediatr Infect Dis J. 2006;25:930–932
32. Kato-Maeda M, Kim EY, Flores L, et al. Differences among sublineages of the East-Asian lineage of Mycobacterium tuberculosis in genotypic clustering. Int J Tuberc Lung Dis. 2010;14:538–544
33. Kato-Maeda M, Shanley CA, Ackart D, et al. Beijing sublineages of Mycobacterium tuberculosis differ in pathogenicity in the guinea pig. Clin Vaccine Immunol. 2012;19:1227–1237

tuberculosis; pediatric; lineage; travel

Supplemental Digital Content

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.