Infection with Mycobacterium tuberculosis (Mtb ) is a global public health issue. It is estimated that 1 of 3 of the world’s population is infected with Mtb , and tuberculosis (TB) is the second leading cause of mortality because of an individual infectious organism.1 2 Mtb infection (LTBI) in children is complicated by the lack of a diagnostic standard, potential impact of previous Bacille Calmette–Guérin (BCG) vaccination and exposure to environmental mycobacteria. One avenue for control of TB is through identification and treatment of LTBI before progression to active disease.3 4 5 , 6 Mtb infection remains a concern. Interferon-gamma release assays (IGRAs) used for detection of Mtb infection rely upon the production of interferon-gamma by T-cells that have had previous exposure to antigens relatively specific for Mtb .7–10 Mtb infection, with an apparent increase in specificity.8 , 11–15 Mtb and environmental mycobacteria. The goals of the current study were to (1) evaluate correspondence between TST and a commercial IGRA (T-SPOT.TB , Oxford Immunotec Ltd., Abingdon, United Kingdom) in international adoptees; (2) determine factors associated with positive results on both TST and the T-SPOT.TB and (3) specifically address the issue of age in relation to performance of these diagnostic tests.
METHODS
Study Participants
Participants were 109 children 4 months to 16 years of age (median 22 months) seen in the International Adoption Clinic of Nationwide Children’s Hospital, Columbus, OH between November 2008 and January 2011. All children seen were eligible for participation and were typically evaluated within 1 month of arrival in the United States. Consent was obtained from the parent or legal guardian.
The study was approved by the Nationwide Children’s Hospital IRB (IRB08-00281).
History and Clinical Data
Medical information was collected from a standard adoption clinic form completed for each adoptee. Information included past medical, surgical and vaccination history, current medical issues, country of birth, history of Mtb exposure, documentation of BCG vaccination and the presence of one or more BCG scars.
Tuberculin Skin Test
TST was performed at the initial visit using the Mantoux method with 5 tuberculin units (0.1 mL) of PPD administered by intradermal injection on the volar surface of the forearm by 1 of 2 physicians or a nurse practitioner trained and experienced in TST placement. Reading was made at 48–72 hours by the nurse practitioner, with results recorded to the nearest millimeter of induration in the largest diameter. For recent arrivals from TB endemic countries, ≥10 mm was considered positive; in immunocompromised children or those with known Mtb exposure, TST ≥ 5 mm was considered positive. Chest radiograph (CXR) was obtained if the TST result was positive. No child had history or CXR findings suggestive of active tuberculosis.
Interferon-Gamma Release Assay
Whole blood samples (4 mL if ≥2 years of age, 2 mL if <2 years of age) were obtained using Vacutainer CPT tubes (BD, Franklin Lakes, NJ) during collection of routine clinic labs. This was generally done 1–2 hours after TST. The commercial T-SPOT.TB IGRA includes early secretory antigenic target 6 and culture filtrate protein 10 antigens. These antigens are relatively specific for Mtb and will not stimulate memory T-cell responses induced by prior exposure to BCG vaccination or most nontuberculous mycobacteria. Peripheral blood mononuclear cells were obtained after centrifugation, and assays were performed according to manufacturer’s instructions. Four wells of an 8-well plate were used for each study sample: (1) nil control to identify nonspecific cell activation, (2) panel A (early secretory antigenic target 6), (3) panel B (culture filtrate protein 10) and (4) positive control containing phytohemagglutinin (a known polyclonal activator) as a test for peripheral blood mononuclear cell functionality. Spots were counted manually, and results were interpreted as positive, negative, borderline or invalid as per manufacturer’s instructions. Borderline/invalid tests were not repeated as blood samples were collected at the time of routine clinical labs, and results of the T-SPOT.TB assay were not used clinically to determine the need for CXR or for initiating or completing therapy for LTBI. Clinical decisions were made on the basis of the results of the TST. The T-SPOT.TB assay was utilized in preference to the QuantiFERON test based on data indicating higher rates of indeterminate/invalid results in younger children with the latter.16–18
Laboratory Data
Specimens for standard laboratory tests were collected per adoption clinic protocol and included, but were not limited to, complete blood count, protein, albumin, RPR and serologic screening for hepatitis viruses and HIV.
Statistical Analysis
Concordance was calculated between TST and standard T-SPOT.TB assay. Primary analysis excluded children with borderline or invalid IGRA results. The χ2 statistic was used to evaluate relationships between categorical variables, whereas the t -test or Mann–Whitney U test was used for comparison of continuous variables. Logistic regression was used to determine factors contributing to positive TST and T-SPOT.TB results. The level of statistical significance was set at P = 0.05. Data were analyzed using Stata 10 (Stata, Inc, College Station, TX).
RESULTS
Demographics
One hundred nine children had a median age of 22 months (range 4–193 months; mean age 39.6 months); 62 children (56.9%) were female. The majority of children (87 [79.8%]) were adopted from an orphanage, and the remainder were from foster care or other settings. Children were adopted from 14 countries, with the majority from China (40 of 109 [36.7%]), Russia (27 of 109 [24.8%]) and Ethiopia (12 of 109 [11.0%]). BCG vaccination was documented in 46.8% (51 of 109), and 68.8% (75 of 109) had BCG scar; consequently, 72.5% (79 of 109) had evidence of previous BCG vaccination. In contrast, over 90% of the children had documentation of vaccination for polio and diphtheria/tetanus/pertussis. Although guidelines from 4 of the birth countries recommend BCG booster vaccination, only 22% (9 of 41) of the children from these countries were old enough to have received a booster. There was no documentation of multiple BCG vaccinations, and only 2 children had more than 1 BCG scar. History of specific TB contact in the past, as documented in information provided by the adoption agency, was noted for 12 of 109 children (11.0%). Seven of these children had household contact with a caregiver treated for active TB. Negative specific TB contact history was not typically recorded. No child was HIV-positive or otherwise known to be immunocompromised.
Tuberculin Skin Test
TST was administered to 107 children (98.2%), and 25 skin reactions (23.4%) were positive. Two children had documentation of TST performed before departure from the country of origin. No child was diagnosed with active TB based on clinical history, examination and/or radiographic findings. In univariate analyses, those with positive TST differed from negative TST in age, gender, history of TB exposure and world region of birth (Table 1 ). Patients testing positive were older and were more likely to be male and have a history of known TB exposure. Children from Africa (5 of 14, 35.7%) and Eastern Europe (16 of 40, 40.0%) were more likely to test positive than those from Asia (3 of 49, 6.1%; P = 0.001). This may partially reflect differences in BCG strain used in the varying regions.19 20 , 21
TABLE 1: Results on Selected Variables by Latent Tuberculosis Infection Test Results (Positive/Negative)
Logistic regression utilizing the variables significant from univariate testing was performed. Only history of TB exposure remained a significant independent predictor. Children with TB exposure were about 9 times more likely to have positive TST (odds ratio: 9.24, 95% confidence interval: 1.54–55.54).
T-SPOT.TB
T-SPOT.TB was performed for 109 children. Results were positive for 5 (4.6%) and negative for 91 (83.5%); number of spots in the positive samples ranged from 12 to 32. Eleven assays were invalid (10.1%), and 2 were borderline (1.8%). Invalid results were largely related to failure of the mitogen control. Only 2 of the 13 invalid/borderline results were from children with positive TST. Overall agreement between TST and T-SPOT.TB was 71.0% (76 of 107; Table 2 ). Among those children with TST, non-borderline non-invalid T-SPOT.TB , overall agreement was 80.8% (76 of 94). Within this cohort, Cohen kappa was 0.30, and prevalence-adjusted, bias-adjusted Kappa was 0.68.22
TABLE 2: TST and T-SPOT.TB Results
Univariate analyses were performed to identify factors differentiating those with positive/negative versus borderline/invalid T-SPOT.TB results. These groups did not differ in demographic characteristics or measured variables except for WBC count, with children with valid T-SPOT.TB results having a higher mean WBC count (10.1 vs. 7.8, P = 0.02). The groups did not differ in lymphocyte percentage or absolute lymphocyte count. Subsequent univariate and multivariate analyses did not include the 13 children with invalid and borderline T-SPOT.TB results.
Children with positive T-SPOT.TB differed from those with negative results in age, history of TB exposure and size of TST induration (Table 1 ). Children with positive T-SPOT.TB were older, and all had history of TB exposure. Those with a positive T-SPOT.TB assay had larger diameter of TST induration (ranging from 15 to 25 mm). Although not statistically significant because of study numbers, 4 of the 5 patients with positive T-SPOT.TB were from Eastern Europe.
Age
None of the 17 patients less than 1 year of age had positive TST or T-SPOT.TB . None of these children had specific history of TB exposure. Figure 1 shows TST results by age group. Of the 17 children <1 year of age, 8 (47.1%) had documentation of BCG vaccination, 9 (52.9%) had BCG scar, and overall 11 (64.7%) had evidence of BCG vaccination. Two children (ages 30 and 59 months) had 2 BCG scars; neither had a positive T-SPOT.TB . One of these children was TST-positive (18 mm), and one was TST-negative (0 mm). Ten of 20 (50%) of those 5 years or older had positive TST. Three of 5 patients with positive T-SPOT.TB were above age 5 years.
FIGURE 1: TST results by age group. All results in the 0–4 mm category were 0 mm except for 1 child with a result of 4 mm in the 12–23 months age group.
DISCUSSION
Specificity of the TST has been questioned given the relatively common finding of a positive TST and a negative IGRA in children.11–14 TB .
Positive results on TST were most related to history of TB exposure. Age (increased likelihood of positive test with increasing age) and region of birth (increased likelihood of positive test if born in Africa or Eastern Europe when compared with Asia) were associated with positive TST on univariate testing. TST responses have been shown to vary with the specific BCG vaccine used, and this may help to explain the impact of region of birth on TST reactivity.19 23 24 , 25 6 , 8
Positive results on a commercial IGRA were uncommon, occurring in only 5 of 109 children (4.6%). Invalid results (11 of 109, 10.1%) were largely related to failure of mitogen control and are consistent with findings from other studies, although invalid or failed test rates vary widely across studies and are more common with QuantiFERON Gold than T-SPOT.TB .8 , 16 , 17 , 24 , 26 , 27 TB results had negative TST, making it unlikely that T-SPOT.TB was failing to detect LTBI. No child with positive T-SPOT.TB had negative TST.
Positive T-SPOT.TB was associated with TB exposure history and size of TST induration. All patients with positive T-SPOT.TB IGRA had a history of TB contact. In the current study, 5 of 25 (20%) TST positive children were T-SPOT.TB positive. This is consistent with the findings of Soysal et al26 TB assay. These authors also identified an association between positive T-SPOT.TB and size of TST reaction. However, 15 of 91 (16.5%) children in our study with negative IGRA also had TST ≥15 mm. In our context, decisions regarding treatment were based on TST results, not results of the T-SPOT.TB assay.
Age was consistently associated with results of TB testing, with positive findings associated with older age. Similar association of age with positive TST and IGRA testing has been reported in several studies.6 , 24 , 28–30 TB . In contrast, 10 of 20 (50.0%) children 5 years of age or older had positive TST. Additionally, 3 of the 5 children with positive T-SPOT.TB were above age 5 years; patients testing positive on T-SPOT.TB assay were 16, 45, 71, 105 and 127 months of age. Similarly, in the study of Cruz et al,13 n = 5) had positive T-SPOT.TB , and only 1 had positive TST; 93 of 133 (69.9%) of the children ≥5 years of age had positive TST. In the study of Lucas et al,24 TB result was 37 months of age. These findings may reflect increased likelihood of Mtb or environmental mycobacterial exposure with increasing age or a less robust immune response with exposure to Mtb antigens in younger children even within the context of previous natural exposure to mycobacteria (tubercular or nontubercular).
Limitations of the study include the relatively small number of children evaluated, especially in the younger (<1 year) age group. Additionally, although clinical and laboratory data were collected prospectively, historical information was by necessity obtained from review of orphanage/adoption agency summaries and documentation of varying quality. This included data on BCG vaccination, which was documented in only 46.8% of study participants. An additional 28 (25.7%) had physical evidence of BCG vaccination, giving BCG coverage of 72.5% in our study population. This seems relatively low given that BCG vaccination is generally recommended in the involved countries of birth. Inappropriate assignment of BCG status could have impacted upon findings with regard to BCG associations with TST and T-SPOT.TB .
In summary, TST was positive (≥10 mm) in a significant percentage (25 of 107, 23.4%) of international adoptees who had placement of PPD after arrival in the US. Results on T-SPOT.TB were rarely positive (5 of 109, 4.6%), and discordant results on TST and T-SPOT.TB reflected negative IGRA in the context of positive TST. Results suggest that in this population, LTBI may be overestimated by TST because of poor specificity. Longitudinal follow-up and repeat T-SPOT.TB IGRA on nontreated children may help to clarify this issue. We found no clear evidence that the positive findings on TST were related to previous BCG vaccination. Age was consistently associated with findings on the T-SPOT-TB assay and TST with no positive results noted for any child <1 year of age.
ACKNOWLEDGMENTS
Oxford Immunotec Ltd. provided the T-SPOT.TB assay kits and Vacutainer CPT tubes. Oxford Immunotec Ltd. was not involved in design and conduct of the study; collection, management, analysis and interpretation of the data or in preparation, review, or approval of the manuscript.
REFERENCES
1. World Health Organization. Tuberculosis Fact Sheet #104. Available at:
http://www.who.int/mediacentre/factsheets/fs104/en/index.html . Accessed on December 1, 2014
2. Cuevas LE, Browning R, Bossuyt P, et al. Evaluation of tuberculosis diagnostics in children: 2. Methodological issues for conducting and reporting research evaluations of tuberculosis diagnostics for intrathoracic tuberculosis in children. Consensus from an expert panel. J Infect Dis. 2012;205(Suppl 2):S209–S215
3. Horsburgh CR Jr. Priorities for the treatment of latent tuberculosis infection in the United States. N Engl J Med. 2004;350:2060–2067
4. George SA, Ko CA, Kirchner HL, et al. The role of chest radiographs and tuberculin skin tests in tuberculosis screening of internationally adopted children. Pediatr Infect Dis J. 2011;30:387–391
5. Wang L, Turner MO, Elwood RK, et al. A meta-analysis of the effect of Bacille Calmette Guérin vaccination on tuberculin skin test measurements. Thorax. 2002;57:804–809
6. Piñeiro R, Mellado MJ, Cilleruelo MJ, et al. Tuberculin skin test in bacille Calmette–Guérin-vaccinated children: how should we interpret the results? Eur J Pediatr. 2012;171:1625–1632
7. Meier T, Eulenbruch HP, Wrighton-Smith P, et al. Sensitivity of a new commercial enzyme-linked immunospot assay (T SPOT-TB) for diagnosis of tuberculosis in clinical practice. Eur J Clin Microbiol Infect Dis. 2005;24:529–536
8. Ferrara G, Losi M, D’Amico R, et al. Use in routine clinical practice of two commercial blood tests for diagnosis of infection with
Mycobacterium tuberculosis : a prospective study. Lancet. 2006;367:1328–1334
9. Mahomed H, Hughes EJ, Hawkridge T, et al. Comparison of Mantoux skin test with three generations of a whole blood IFN-gamma assay for tuberculosis infection. Int J Tuberc Lung Dis. 2006;10:310–316
10. Wang SH, Powell DA, Nagaraja HN, et al. Evaluation of a modified interferon-gamma release assay for the diagnosis of latent tuberculosis infection in adult and paediatric populations that enables delayed processing. Scand J Infect Dis. 2010;42:845–850
11. Menzies D, Pai M, Comstock G.. Meta-analysis: new tests for the diagnosis of latent tuberculosis infection: areas of uncertainty and recommendations for research. Ann Intern Med. 2007;146:340–354
12. Connell TG, Ritz N, Paxton GA, et al. A three-way comparison of tuberculin skin testing, QuantiFERON-TB gold and T-SPOT.TB in children. PLoS One. 2008;3:e2624
13. Cruz AT, Geltemeyer AM, Starke JR, et al. Comparing the tuberculin skin test and T-SPOT.TB blood test in children. Pediatrics. 2011;127:e31–e38
14. Zhang S, Shao L, Mo L, et al. Evaluation of gamma interferon release assays using
Mycobacterium tuberculosis antigens for diagnosis of latent and active tuberculosis in Mycobacterium bovis BCG-vaccinated populations. Clin Vaccine Immunol. 2010;17:1985–1990
15. Sollai S, Galli L, de Martino M, Chiappini E.. Systematic review and meta-analysis on the utility of Interferon-gamma release assays for the diagnosis of
Mycobacterium tuberculosis infection in children: a 2013 update. BMC Infect Dis. 2014;14(Suppl 1):S6
16. Bergamini BM, Losi M, Vaienti F, et al. Performance of commercial blood tests for the diagnosis of latent tuberculosis infection in children and adolescents. Pediatrics. 2009;123:e419–e424
17. Haustein T, Ridout DA, Hartley JC, et al. The likelihood of an indeterminate test result from a whole-blood interferon-gamma release assay for the diagnosis of
Mycobacterium tuberculosis infection in children correlates with age and immune status. Pediatr Infect Dis J. 2009;28:669–673
18. Lighter J, Rigaud M, Eduardo R, et al. Latent tuberculosis diagnosis in children by using the QuantiFERON-TB Gold In-Tube test. Pediatrics. 2009;123:30–37
19. Castro-Rodriguez JA, Mallol J, Andrade R, et al. Comparison of tuberculin skin test response after three modalities of neonatal BCG vaccination. Trans R Soc Trop Med Hyg. 2007;101:493–496
20. World Health Organization. Growth charts. Available at:
http://www.cdc.gov/growthcharts/who_charts.htm . Last accessed December 1, 2014
21. Centers for Disease Control and Prevention. Growth charts. Available at:
http://www.cdc.gov/growthcharts/cdc_charts.htm . Last accessed December 1, 2014
22. Sim J, Wright CC.. The kappa statistic in reliability studies: use, interpretation, and sample size requirements. Phys Ther. 2005;85:257–268
23. Centers for Disease Control and Prevention. Latent tuberculosis infection: a guide for primary health care providers. Available at:
http://www.cdc.gov/tb/publications/LTBI/default.htm . Last accessed December 1, 2014
24. Lucas M, Nicol P, McKinnon E, et al. A prospective large-scale study of methods for the detection of latent
Mycobacterium tuberculosis infection in refugee children. Thorax. 2010;65:442–448
25. Yassin MA, Petrucci R, Garie KT, et al. Use of tuberculin skin test, IFN-γ release assays and IFN-γ-induced protein-10 to identify children with TB infection. Eur Respir J. 2013;41:644–648
26. Soysal A, Türel O, Toprak D, et al. Comparison of positive tuberculin skin test with an interferon-gamma-based assay in unexposed children. Jpn J Infect Dis. 2008;61:192–195
27. Bamford AR, Crook AM, Clark JE, et al. Comparison of interferon-gamma release assays and tuberculin skin test in predicting active tuberculosis (TB) in children in the UK: a paediatric TB network study. Arch Dis Child. 2010;95:180–186
28. Nakaoka H, Lawson L, Squire SB, et al. Risk for tuberculosis among children. Emerg Infect Dis. 2006;12:1383–1388
29. Dogra S, Narang P, Mendiratta DK, et al. Comparison of a whole blood interferon-γ release assay with tuberculin skin testing for the detection of tuberculosis infection in hospitalized children in rural India. J Infect. 2007;54:267–76
30. Banfield S, Pascoe E, Thambiran A, et al. Factors associated with the performance of a blood-based interferon-γ release assay in diagnosing tuberculosis. PLoS One. 2012;7:e38556
