The Spanish Society for Pediatric Infectious Diseases (SEIP) recommends performance of a tuberculin skin test (TST) as part of the immigrant child's first health check in our country.1 Most of these children have been vaccinated with Bacillus Calmette-Guérin (BCG), and the interpretation of TST for them is a controversial question.2 The degree of induration in vaccinated children must be taken into account at the time of the diagnosis of latent tuberculosis infection (LTBI). Some countries adjust TST cut-off points for vaccinated children, even though there is no consensus as to whether, or how much, the cutoff point should be increased.
The new interferon-gamma release assays (IGRAS) have greater specificity than the TST in detection of LTBI.3 Studies in children show excellent sensitivity of IGRAS in the diagnosis of tuberculosis disease (TBD).4 Thus, we consider that the QuantiFERON-TB Gold Test In Tube (QTF) could be used as a diagnostic referent to analyze the sensitivity and specificity of the TST in BCG vaccinated and nonvaccinated children.
PATIENTS AND METHODS
A transversal multicenter study was performed in 9 hospitals in Madrid, Spain. The inclusion of patients began in March 2007, and the study finished in February 2009. Immunocompetent children less than 15 years of age who immigrated from TB endemic areas and who arrived in Spain during the prior 2 years,4 children who were in contact with a person with active tuberculosis and children admitted with symptoms suggestive of TBD were included.
Human immunodeficiency virus-infected cases, children with congenital immunodeficiencies, being treated with immunosuppressants, with autoimmune diseases, malignant proliferative processes, and solid organ or bone marrow transplants were excluded. Likewise, patients were excluded if they had previously received antituberculosis treatment for LTBI or TBD, or if they had received treatment for more than 3 days at the time of inclusion.
Both tests (TST and QTF) were performed on all children. Considered to be BCG vaccinated, were those children with postvaccinal scarring on the deltoid area and who come from countries in which BCG is included in the vaccination schedule.2
LTBI was diagnosed in asymptomatic children with normal chest x-rays and/or chest computed tomography scan and with a positive TST under the reading criteria of the SEIP (≥5 mm for children with close contact to TBD and ≥10 mm in screening of immigrant children).1
TBD was diagnosed thus (1) Children with microbiological confirmation by growth of Mycobacterium tuberculosis in solid or liquid medium or with positive molecular polymerase chain reaction studies for M. tuberculosis. (2) Children with symptoms compatible with TBD and without microbiological confirmation, but that presented a TST ≥5 mm and radiologic studies compatible with TBD or the presence of caseous granulomas in histhological specimens (excluding other causes of granulomatous disease).
The TST was carried out according to the recommendations of the SEIP.1 The TST reading was performed after 72 hours by a single professional in each center. All the QTF assays were performed at the Immunology Department of Carlos III Hospital (Madrid) by a single immunologist following Centers for Disease Control and Prevention recommendations.5 All the samples were processed within 3 hours of its extraction. The signed informed consent was obtained from parents or guardians beforehand. Reading of results from the 2 tests was blind, the laboratory being without access to the clinical data and the doctors of each hospital not knowing the QTF results prior to TST interpretation.
The data were analyzed at the Biostatistics Department of La Paz Hospital (Madrid) using the SPSS 9.0 program (SPSS Inc.). The data description was expressed in absolute frequencies, using mean and standard deviation, or median and range. The Spearman correlation coefficient was used to analyze the association between quantitative variables. To evaluate the TST sensitivity and specificity of the different reading points recommended in Spain by the SEIP, the area under the receiver operating characteristic curve (ROC curve) was calculated. The study was approved by the Clinical Testing and Ethics Committees of the participating hospitals.
A total of 459 children were included, 264 girls (57.5%) and 195 boys (42.4%), with an average age of 4.73 ± 3.68 years (1 month–15 years). Of total, –318 children were included by screening of immigrants proceeding from TB endemic areas (69.2%), 83 children for contact with TBD (18%), and 58 for suspicion of active TBD (12.6%). In all, 86% of the patients (396/459) came from TB endemic areas (131 Latin America [28.5%], 115 Sub-Saharan Africa [25%]; 67 China [14.5%], 65 Eastern European countries [14.1%], 52 India [11.3%], and 29 Morocco [6.3%]). Scar of the deltoid area compatible with prior BCG vaccination was observed in 46.4% (213/459) of the children. There were no statistically significant differences between the ages of the vaccinated population (4.72 ± 3.89 years) and that of the nonvaccinated population (4.73 ± 3.44 years) (P = 0.5). In 318 children (69.2%) LTBI and TBD were excluded, 73 cases (15.9%) were diagnosed with LTBI and 68 (14.8%) with TBD.
Performance of QTF obtained 343 negative, 96 positive, and 20 indeterminate results due to lack of lymphocyte activation after stimulation with mitogen (4.3%). QTF sensitivity was calculated by extrapolating the results obtained in TBD patients.5 The QTF was indeterminate in 6 of the 68 children with TBD (8.8%) and was positive in 61 of the 62 TBD children from whom we obtained valid results (sensitivity, 98.3%; 95% confidence interval [CI], 94%–100%). A patient with microbiologically confirmed tuberculosis meningitis presented negative QTF and TST. Specificity was calculated by observing the percentage of negative results among asymptomatic children with negative TST.3,5 QTF result was indeterminate in 11 of the 318 healthy children (3.4%) and was negative in 304 of the 307 healthy children from whom we obtained valid results (specificity, 99%; 95% CI, 97%–100%).
QTF was positive in 0.9% of children with TST indurations <5 mm (3/313), 22.2% of those with indurations between 5 and 9 mm (4/18), in 50.9% of those with indurations between 10 and 14 mm (26/51), and in 81.8% of cases with indurations ≥15 mm (63/77). Analyzing the concordant results from both tests, a light-moderate correlation was observed between millimetres of TST indurations and amount of interferon-γ produced after stimulation with ESAT-6 and CFP-10 (r = 0.743) (Fig., Supplemental Digital Content 1, http://links.lww.com/INF/A649).
As the QTF in our sample demonstrated excellent sensibility and specificity, we consider the test as a diagnostic reference to analyze TST sensitivity and specificity in BCG vaccinated and nonvaccinated populations.1
Two hundred forty-six children without BCG scar were included (53.5%). The ROC curve obtained allowed adequate discrimination of points of sensitivity and specificity (Area under curve, 0.977; 95% CI, 0.95–0.99). In our series, a TST induration equal to or greater than 10 mm presented a sensitivity of 95.9% (95% CI, 88.6%–98.6%) and a specificity of 95.5% (95% CI, 91.1%–97.8%). On reducing the reading point to ≥5 mm, the TST presented excellent sensitivity (100%; 95% CI, 95%–100%) and specificity (93.6%; 95% CI, 88.7%–96.5%).
Two hundred thirteen children were included with scarring on the deltoid area compatible with prior BCG vaccination. The ROC curve obtained allowed acceptable discrimination of appropriate points of sensitivity and specificity (0.865; 95% CI, 0.76–0.96). In this population, considering as positive TST indurations equal to or greater than 10 mm, TST specificity was only 86% (95% CI, 80%–90%) and sensitivity 82.6% (95% CI, 63%–93%). If we take those indurations of ≥15 mm as positive, TST sensitivity reduces to 60.9% (95% CI, 40.8%–77.8%) although specificity improves to 97% (95% CI, 92.4%–98.2%).
Interpretation of the TST in vaccinated children continues to be a controversial question. Some meta-analyses have shown that the effect of BCG can endure for up to 15 years.6,7 However, some authors consider that one should not take previous BCG vaccination history into account when interpreting the TST due to the risk of not treating real LTBI.8
In Spain, the first consensus document from the SEIP in 2003 established positively those indurations equal to or greater than 15 mm in BCG-vaccinated children during the last 3 years, and negatively those of less than 10 mm.1 Individualized reading was recommended in the evaluation of indurations between 11 and 14 mm. However, when there is a TST reaction of ≥10 mm in a vaccinated child, given the high risk of developing TBD after the infection, the effect of BCG is usually ignored.
The development of more specific tests such as QTF can help us to differentiate false positive TST results due to prior BCG vaccination. In BCG vaccinated children diagnosed with LTBI, the appearance of discordant positive TST results with negative IGRAS is frequent,4,9 but this phenomenon does not seem to be observed in children with TBD.4 Also, it is well known that TST indurations produced after BCG vaccination or by infection by other mycobacterias normally do not exceed 15 mm. However, it is difficult to differentiate whether an indurations between 11 and 14 mm is or is not a false positive.
In our study, excluding indeterminate results, QTF had sensitivity of 98% and specificity of 99%, thus we consider it could be used as a diagnostic reference to analyze TST sensitivity and specificity. We observed how the increase in TST reading point to 15 mm in BCG-vaccinated children could give rise to an important reduction in test sensitivity. As a result, the new consensus document published in 2010 in Spain definitively dismissed the 15 mm cut-off point for those children already vaccinated.10 The American Academy of Pediatrics and the American Thoracic Society also recognize a cut-off point of 10 mm for these same cases.11,12
On the other hand, according to our data, in children without BCG scar and with TB risk factors or suspicion of TBD, a TST ≥5 mm presents excellent specificity and sensitivity. In an analysis similar to our own, Bakir et al studied the TST sensitivity and specificity in 979 children in contact with adults diagnosed of pulmonary tuberculosis using an IGRA as a “reference standard.” The author found that the sensitivity of TST in infants under 2 years of age using a cut-off point equal to or greater than 10 mm was only 66%. Reducing the cut-off point to 5 mm increased sensitivity to 78%.13 This fact supports the hypothesis of considering positive indurations above 5 mm in nonvaccinated infants belonging to risk populations.
The current epidemiological situation in Spain, with increase in immigration from developing countries, has led to high rates of TB in our community. TST sensitivity observed using QTF as a reference in immigrant children without BCG scar raises the question of whether in this population a cut-off point of 5 mm is correct. The performance of an IGRA in these children with indurations between 5 and 10 mm would confirmed tuberculosis infection and would rule out the presence of the majority of environmental mycobacterias.
The Spanish Collaborative Group for the study of QuantiFERON-TB GOLD Test in children Collaborators are Daniel Blázquez, MD; María Penín, MD; Maria Luisa Navarro, MD, PhD; Jesús Saavedra-Lozano, MD, PhD; Maria Isabel González-Tomé, MD, PhD; Cristina Calvo, MD, PhD; Marta Ruiz, MD; Sara Guillén, MD, PhD; Jose Tomás Ramos, MD, PhD; Teresa Hernandez-Sampelayo, MD, PhD; Ramón Velázquez, MD; Beatriz Pérez-Gorricho, MD, PhD; and Jorge Martinez, MD.
1. Grupo de trabajo de Tuberculosis de la Sociedad Espanola de Infectologia Pediatrica. Interpretación de la prueba de tuberculina en niños. An Pediatr (Barc)
2. Araujo Z, de Waard JH, de Larrea CF, et al. The effect of Bacille Calmette-Guerin vaccine on tuberculin reactivity in indigenous children from communities with high prevalence of tuberculosis. Vaccine
3. 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
4. Bianchi L, Galli L, Moriondo M, et al. Interferon-gamma release assay improves the diagnosis of tuberculosis in children. Pediatr Infect Dis J
5. Mazurek GH, Jereb J, Lobue P, et al. Guidelines for using the QuantiFERON-TB Gold test for detecting Mycobacterium tuberculosis
infection, United States. MMWR Recomm Rep
6. Pai M, Zwerling A, Menzies D. Systematic review: T-cell-based assays for the diagnosis of latent tuberculosis infection: an update. Ann Intern Med
7. Wang L, Turner MO, Elwood RK, et al. A meta-analysis of the effect of Bacille Calmette Guerin vaccination on tuberculin skin test measurements. Thorax
8. Sleiman R, Al-Tannir M, Dakdouki G, et al. Interpretation of the tuberculin skin test in Bacille Calmette-Guerin vaccinated and nonvaccinated school children. Pediatr Infect Dis J
9. Lighter J, Rigaud M, Eduardo R, et al. Latent tuberculosis diagnosis in children by using the QuantiFERON-TB Gold In-Tube test. Pediatrics
10. Moreno-Pérez D, Andrés-Martín A, Altet Gómez MN, et al. Diagnosis of tuberculosis in paediatrics. An Pediatr (Barc)
. 2010;72:283.e1–283.e14. doi: 10.1016/j.anpedi.2010.01.002.
11. Targeted tuberculin testing and treatment of latent tuberculosis infection. This official statement of the American Thoracic Society was adopted by the ATS Board of Directors, July 1999. This is a Joint Statement of the American Thoracic Society (ATS) and the Centers for Disease Control and Prevention (CDC). This statement was endorsed by the Council of the Infectious Diseases Society of America. (IDSA), September 1999, and the sections of this statement. Am J Respir Crit Care Med
. 2000;161(4 pt 2):S221–S247.
12. American Academy of Paediatrics. Tuberculosis. In: Pickering L, Baker C, Long SS, eds. Red Book: 2006 Report of the Committee on Infectious Diseases. 25th ed
. Elk Grove Village, IL: American Academy of Paediatrics; 2006:678–698.
13. Bakir M, Dosanjh DP, Deeks JJ, et al. Use of T cell-based diagnosis of tuberculosis infection to optimize interpretation of tuberculin skin testing for child tuberculosis contacts. Clin Infect Dis