Tuberculosis (TB) is the leading cause of death from an infectious disease worldwide.1 It is estimated that around 1.1 million children develop TB annually.1 In children, microbiologic confirmation is difficult to achieve due to the paucibacillary nature of TB in this age group and the difficulties in obtaining sputum.2 Therefore, TB diagnosis in children is often based on a combination of clinical and radiologic findings, epidemiologic risk factors and a positive tuberculin skin test (TST) or interferon-gamma release assay (IGRA) result.2
IGRAs are immunoassays that rely on the detection of interferon-gamma produced by T cells following stimulation with relatively TB-specific peptides.2 In Europe QuantiFERON-TB assays (Cellestis/Qiagen, Hilden, Germany) are more widely used than T-SPOT.TB assays (Oxford Immunotec, Abingdon, United Kingdom).3 In 2017, the QuantiFERON-TB Gold Plus assay (QFT-Plus) was launched, replacing the previous-generation QuantiFERON-TB Gold in-Tube (QFT-GIT) assay. QFT-Plus uses 2 antigen tubes: TB1—intended to primarily induce CD4+ T-cell response—contains ESAT-6 and CFP-10 antigens that were also part of the QFT-GIT, but in contrast to the previous-generation assay lacks TB7.7; TB2 is a new tube that contains additional short peptides—aimed at eliciting CD8+ T-cell responses—which are claimed to be more pronounced in recent infection (compared with remote infection) and to potentially reflect TB exposure intensity.4
We recently demonstrated that QFT-Plus assays do not have greater sensitivity than previous-generation QFT assays in the diagnosis of TB disease in children and adolescents.5 However, data on the performance of QFT-Plus in the pediatric age group remain limited, with most studies primarily having focused on test sensitivity.5–8 This study aimed to determine the specificity of QFT-Plus assays in children and adolescents with suspected TB in a low TB-burden setting.
MATERIALS AND METHODS
We performed a cross-sectional study within the Spanish Pediatric TB Research Network (pTBred), which includes 83 participating centers.5 Data are collected prospectively using RedCap tools, hosted at Instituto de Investigación Sanitaria Gregorio Marañón.9 Approvals for pTBred and for the study reported here were obtained from the Hospital Carlos III Ethics Committee (Madrid, Spain; ref. P13/12) and from the Hospital Sant Joan de Déu Ethics Committee (Esplugues, Spain; ref. PIC-168-16). Informed consent was obtained from parents/guardians at inclusion.
For this study, we enrolled patients <18 years of age in whom QFT-Plus was performed due to clinically or radiologically suspected TB from September 2016 to March 2020. Patients with a previous history of TB disease or latent TB infection (LTBI) were excluded. All QFT-Plus assays were performed in fully accredited diagnostic laboratories at each participating institution, and their results were interpreted according to manufacturer’s instructions. In brief, QFT-Plus results were classified as positive (i.e., TB1-nil and TB2-nil ≥0.35 IU/mL), negative or indeterminate. TSTs were performed by intradermal injection of 2 tuberculin-units of purified protein derivative (PPD RT23; Statens Serum Institut, Copenhagen, Denmark), with results read after 48–72 hours by specifically trained nurses. According to national guidelines, a 5-mm induration cutoff was used to define a positive test result, irrespective of BCG vaccination status.
Patients were categorized according to their final diagnosis as “TB” or “not TB.” In pTBred, the diagnosis of TB disease is based on epidemiologic, clinical, radiologic, and microbiologic features according to criteria published elsewhere.10 TB disease was further categorized as microbiologically confirmed or unconfirmed and as intrathoracic or extrathoracic according to established consensus definitions.11
Data are reported as proportions with 95% confidence intervals (CI; using binomial exact calculation) or medians with interquartile ranges (IQR). Test specificity was calculated based on the proportion of negative QFT-Plus results in patients classified as “not TB.” Total percentage agreement and Cohen kappa coefficient (κ) were used to quantify concordance between TST and QFT-Plus results. Indeterminate QFT-Plus results were excluded from these analyses. Statistical analyses were performed using SPSS (V24; IBM, Armond, NY), with statistical significance defined as a 2-tailed P value <0.05.
During the study period, QFT-Plus was performed in 286 children with clinical or radiologic suspicion of TB. Two patients were excluded due to a previous history of TB. Of the 284 remaining patients, 102 (35.9%) were diagnosed with TB, while 182 (64.1%) had an alternative diagnosis (i.e., “not TB”; Table 1).
TABLE 1. -
QFT-Plus Assay Results in Relation to Clinical Characteristics and Tuberculin Skin Test Results in the 182 Children With Final Classification of “not TB”
|Characteristics, n (%)
||Specificity, % (95% CI)
|n = 182
||n = 151
||n = 14
||n = 17
|BCG vaccination status
|Tuberculin skin test result
Indeterminate assay results were excluded in the calculation of specificity estimates.
*Excluded from calculation of corresponding P value.
BCG inidcates bacillus Calmette-Guérin vaccine; n, number.
Of the 102 TB cases (50 [49.0%] females, median age: 8.5 [IQR: 1.9–12.5] years), 89 (87.3%) had positive QFT-Plus results; the remaining 13 (12.7%) had negative results (no indeterminate results), corresponding to a test sensitivity of 87.3% (95% CI: 79.2–93.0). Detailed data on disease phenotypes and test performance are shown in Table (Supplemental Digital Content 1, http://links.lww.com/INF/E387).
Of the 182 “not TB” cases (93 [51.1%] females, median age: 7.0 [IQR: 3.2–12.8] years), 14 (7.7%) had positive QFT-Plus results (see Table, Supplemental Digital Content 2, http://links.lww.com/INF/E388), 17 (9.3%) had indeterminate results, and the remaining 151 cases (83.0%) had negative results. Seventy-three (40.1%) had a thoracic disease focus, while 109 (59.9%) had extrathoracic disease (see Table, Supplemental Digital Content 3, http://links.lww.com/INF/E388). The most common final diagnoses were community-acquired pneumonia (n = 46), selflimiting cervical lymphadenitis (n = 30), nontuberculous mycobacteria (NTM) lymphadenitis (n = 17), and selflimiting fever (n = 15).
The overall specificity of QFT-Plus for TB disease was 91.5% (95% CI: 86.2–95.3%), having excluded indeterminate results. The assay specificity was ≥85% in all subgroups analyzed (Table 1), except in TST-positive patients (79.3% [95% CI: 60.3–92.0%]), in whom specificity was significantly lower than in TST-negative patients (95.6% [95% CI: 88.9–98.8%]; P = 0.025). The positive and negative predictive values of the assay were 86.4% (95% CI: 78.3–92.4%) and 92.1% (95% CI: 86.8–95.7%), respectively.
Of the 14 “not TB” patients with a positive QFT-Plus result (see Table, Supplemental Digital Content 2, http://links.lww.com/INF/E388), 8 had risk factors for TB infection (born in or travel to high TB-burden country, n = 6; contact with TB patient, n = 2), and were therefore deemed to have LTBI. In 4 patients with initial borderline-positive QFT-Plus results, repeat testing yielded negative results, indicating the initial result had been false positive.
TSTs were performed in 128 “not TB” patients; 30 (23.4%) had positive and 98 (76.6%) negative results. Therefore, TST had significantly lower specificity (76.6% [68.3%–83.6%]) than QFT-Plus (P < 0.001). Concordance between both tests was weak (77.1%, κ = 0.208). Of the 23 (19.5%) patients with a TST-positive/QFT-Plus-negative result constellation, 8 were BCG-vaccinated; 8 had clinical features consistent with NTM lymphadenitis (3 subsequently microbiologically confirmed; all 8 recovered without TB treatment); 5 had epidemiologic risk factors for LTBI; in the remaining 2 cases, no potential explanation for the positive TST result was found. Four patients (3.4%) had a TST-negative/QFT-Plus-positive result constellation (patients 10–13; see Table, Supplemental Digital Content 2, http://links.lww.com/INF/E388). Only 1 of those (patient 10) had risk factors for LTBI; a repeat QFT-Plus assay in this patient produced a negative result. QFT-Plus result conversion was also observed in 2 of the remaining 3 patients.
To our knowledge, this is one of the largest studies evaluating the specificity of QFT-Plus assays in children and adolescents with suspected TB disease. The key strength of this study lies in the use of a control group with a broad spectrum of alternative diagnoses, thereby reflecting clinical practice. Most previous studies investigating the specificity of previous-generation IGRAs (i.e., QFT-GIT and T-SPOT.TB assays) in children had significant limitations. Studies focusing on LTBI were hampered by the lack of a gold standard for LTBI, and often made broad assumptions concerning TST/IGRA result discordance. Fewer studies focused on children with suspected TB disease, based on suggestive clinical features and radiologic findings, but many of those were constrained by the small number of pediatric TB cases included.12
The specificity of the QFT-Plus assay has been broadly evaluated in adults. A large meta-analysis including more than 2700 patients estimated the assay specificity to be as high as 96%.13 However, adult data cannot directly be extrapolated to children, given that previous studies, including our own, have shown that IGRAs perform less well in young children than in adults.14,15
Despite QFT-Plus assays having been introduced in 2016, pediatric data on assay specificity remain scarce. One recently published study from Italy, which included 43 children with suspected TB disease, estimated the assay specificity to be 90.1%, which aligns well with the estimate of 91.5% in our study.8 However, the aforementioned study only included 12 children with TB disease, limiting the precision of its estimates. The negative predictive value of the assay was 92.1% in our study. In low-burden settings, such as Spain, these figures translate into about 1 in 10 patients with an alternative diagnosis having a positive QFT-Plus result. The latter may be due to either coincidental LTBI or a false-positive test result. We found that most patients in the “not TB” group with borderline interferon-gamma levels in TB1 or TB2 antigen tubes converted to negative results on repeat testing. In patients with an alternative diagnosis and repeatedly positive QFT-Plus results, potential risk factors for LTBI should be sought and chemoprophylaxis be considered.
Our data suggest that the QFT-Plus assay has significantly greater specificity for TB disease than the TST.12 This likely results from the former using 2 well-defined mycobacterial antigens, which are absent from all known BCG vaccine strains and most NTM species, to elicit T-cell responses.2 In contrast, the TST test substance, PPD, contains a broad array of mycobacterial antigens that are also expressed by BCG and NTM, resulting in potential confounding in BCG-vaccinated individuals and patients with NTM infection.2 This hypothesis is further supported by the observation that most children in our study with a TST-positive/QFT-Plus-negative result constellation either had previously been vaccinated with BCG or had probable or confirmed NTM lymphadenitis.
Our study has a number of limitations. First, several patients in the “not TB” control group did not have a histologically or microbiologically confirmed alternative diagnosis. Additionally, a small number of patients in the “not TB” group had positive QFT-Plus results, and—due to the presence of significant risk factors—were deemed to have incidental LTBI, which cannot be proven in the absence of a gold standard for LTBI. Finally, we did not collect data on the precise timing between QFT-Plus and TST. Potential boosting of IGRA responses resulting from prior TST administration has previously been hypothesized16; however, we have subsequently demonstrated that this does not occur.17,18
In summary, our data show that in children with suspected TB the specificity of the QFT-Plus assay is greater than 90%, similar to previous-generation IGRAs. In our study, the QFT-Plus assay had significantly higher specificity than the TST. However, a small proportion of patients without TB disease had false-positive QFT-Plus assay results, which converted to negative results on repeat testing. Therefore, while a positive QFT-Plus result in children presenting with clinical or radiologic features of TB lends substantial support to a putative diagnosis of TB, it does not rule out an alternative diagnosis.
1. WHO. Global tuberculosis
report 2018. WHO 2019. Available at: https://www.who.int/tb/publications/global_report/en/
. Accessed January 10, 2021.
2. Tebruegge M, Ritz N, Curtis N, et al. Diagnostic Tests for childhood tuberculosis
: past imperfect, present tense and future perfect? Pediatr Infect Dis J. 2015;34:1014–1019.
3. Tebruegge M, Ritz N, Koetz K, et al. Availability and use of molecular microbiological and immunological tests for the diagnosis of tuberculosis
in Europe. PLoS One. 2014;9:e99129.
4. Barcellini L, Borroni E, Brown J, et al. First evaluation of quantiFERON-TB gold plus performance in contact screening. Eur Respir J. 2016;48:1411–1419.
5. Soler-Garcia A, Gamell A, Santiago B, et al.; QFT-Plus Study Group of the Spanish Pediatric TB Research Network (pTBred). Diagnostic accuracy of QuantiFERON-TB gold plus assays in children and adolescents with tuberculosis
disease. J Pediatr. 2020;223:212–215.e1.
6. Kay AW, DiNardo AR, Dlamini Q, et al. Evaluation of the QuantiFERON-tuberculosis
gold plus assay in children with tuberculosis
disease or following household exposure to tuberculosis
. Am J Trop Med Hyg. 2019;100:540–543.
7. Nguyen DT, Phan H, Trinh T, et al. Sensitivity and characteristics associated with positive QuantiFERON-TB Gold-Plus assay in children with confirmed tuberculosis
. PLoS One. 2019;14:e0213304.
8. Buonsenso D, Delogu G, Perricone C, et al. Accuracy of QuantiFERON-TB gold plus test for diagnosis of mycobacterium tuberculosis
infection in children. J Clin Microbiol. 2020;58:e00272–e00220.
9. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)–a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–381.
10. Graham SM, Ahmed T, Amanullah F, et al. Evaluation of tuberculosis
diagnostics in children: 1. Proposed clinical case definitions for classification of intrathoracic tuberculosis
disease. Consensus from an expert panel. J Infect Dis. 2012;205(suppl 2):S199–S208.
11. Wiseman CA, Gie RP, Starke JR, et al. A proposed comprehensive classification of tuberculosis
disease severity in children. Pediatr Infect Dis J. 2012;31:347–352.
12. Laurenti P, Raponi M, de Waure C, et al. Performance of interferon-γ release assays in the diagnosis of confirmed active tuberculosis
in immunocompetent children: a new systematic review and meta-analysis. BMC Infect Dis. 2016;16:131.
13. Pourakbari B, Mamishi S, Benvari S, et al. Comparison of the QuantiFERON-TB gold plus and QuantiFERON-TB gold in-tube interferon-γ release assays: a systematic review and meta-analysis. Adv Med Sci. 2019;64:437–443.
14. Velasco-Arnaiz E, Soriano-Arandes A, Latorre I, et al. Performance of tuberculin skin tests and interferon-γ release assays in children younger than 5 years. Pediatr Infect Dis J. 2018;37:1235–1241.
15. Tebruegge M, de Graaf H, Sukhtankar P, et al. Extremes of age are associated with indeterminate QuantiFERON-TB gold assay results. J Clin Microbiol. 2014;52:2694–2697.
16. Leyten EM, Prins C, Bossink AW, et al. Effect of tuberculin skin testing on a Mycobacterium tuberculosis
-specific interferon-gamma assay. Eur Respir J. 2007;29:1212–1216.
17. Velasco-Arnaiz E, Soriano-Arandes A, Espiau M, et al. Impact of baseline tuberculin skin test and isoniazid chemoprophylaxis on subsequent quantiferon-TB gold in-tube performance in young children assessed after tuberculosis
contact in Catalonia. Pediatr Infect Dis J. 2020;39:e22–e25.
18. Ritz N, Yau C, Connell TG, et al. Absence of interferon-gamma release assay conversion following tuberculin skin testing. Int J Tuberc Lung Dis. 2011;15:767–769.
APPENDIX: QFT-Plus Study Group of the Spanish Pediatric TB Research Network (pTBred)
Olga Calavia, Rebeca Lahoz (Servei de Pediatria, Hospital Universitari Joan XXIII, Tarragona, Spain); Teresa Vallmanya (Servei de Pediatria, Hospital Universitari Arnau de Vilanova, Lleida, Spain); María-Isabel Garrote-Llanos (Sección de Infectología Pediátrica, Hospital Universitario Basurto, Bilbao, Vizcaya, España; Departamento de Pediatría, Universidad del País Vasco, UPV/EHU, Bilbao, Vizcaya, España); María Montero, Estrella Peromingo-Matute (Unidad de Pediatría y sus áreas específicas, Hospital Universitario Puerta del Mar, Cádiz, España); Queralt Soler-Campins, Marta Velázquez, Marina Fenoy (Servicio de Pediatría, Consorci Sanitari de Terrassa, Terrassa, Barcelona, España); Beatriz Soto-Sánchez, Marta Ruiz, Sara Guillén (Pediatrics Department, Hospital de Getafe, Madrid, Spain); Elena Colino, Saro Quintana (Pediatrics Department, Complejo Hospitalario Universitario Insular-Materno Infantil de Las Palmas de Gran Canaria, Las Palmas de Gran Canaria, Spain); Cristina Calvo, María Luz García-García (Pediatrics Department, Hospital Severo Ochoa, Madrid, Spain; Pediatric Infectious and Tropical Diseases Department, Hospital La Paz, Madrid, Spain; Health Research Institute IdiPAZ, Hospital La Paz, Madrid, Spain; Red de Investigación Translacional en Infectología Pediátrica, RITIP, Madrid, Spain); Nerea Azurmendi Gundín (Servicio de Pediatría, Hospital Universitario Donostia-Instituto BioDonostia, San Sebastián, Guipúzcoa, España; Departamento de Pediatría, Facultad de Medicina, EHU-UPV, San Sebastián, Guipúzcoa, España); Pere Soler-Palacín, Andrea Martín-Nalda (Pediatric Infectious Diseases and Immunodeficiencies Unit, Hospital Universitari Vall d’Hebron, Vall d’Hebron Research Institute, Barcelona, Spain); María Teresa Tórtola (Microbiology Department, Vall d’Hebron University Hospital, Barcelona, Spain); Antoni Soriano-Arandes (Paediatric Infectious Diseases and Immunodeficiencies Unit, Hospital Universitari Vall d’Hebron, Vall d’Hebron Research Institute, Barcelona, Spain); Eva María López-Medina (Pediatric Infectious Diseases Unit, Hospital Universitario y Politécnico La Fe, Valencia, Spain); María del Mar Santos, Ángel Hernández, Teresa Hernández-Sampelayo (Pediatric Infectious Diseases Unit, Gregorio Marañón Hospital, Madrid, Spain); Ana Méndez-Echevarría, María José Mellado, Fernando Baquero-Artigao, Talía Sainz (Pediatric Infectious and Tropical Diseases Department, Hospital La Paz, Madrid, Spain); Milagros García Hortelano (Department of Pediatrics, Hospital Carlos III, Madrid, Spain, Javier Álvarez, Enrique Villalobos (Pediatric Infectious Diseases Unit, Department of Pediatrics, Hospital Infantil Universitario Niño Jesús, Madrid, Spain, Inés Gale (Unidad de Enfermedades Infecciosas, Servicio de Pediatría, Hospital Universitario Miguel Servet, Zaragoza, Spain); Miguel Lillo (Pediatric Oncology, Hospital General Universitario de Albacete, Albacete, Spain); Marta Dapena (Department of Pediatrics, Hospital General de Castelló, Infectious Diseases Unit, Castelló); David Gómez-Pastrana (Neumología Pediátrica, Servicio de Pediatría, Hospital Universitario Jerez de la Frontera, Cádiz, Spain); Elisenda Moliner (Neonatology Unit, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain); Lourdes García (Pediatrics, Consorci Sanitari del Maresme, Mataró, Spain); Neus Rius, María Teresa Pascual (Pediatrics, Hospital Universitari Sant Joan de Reus, Reus, Spain); Laia Sánchez-Torrent (Pediatrics, Parc Sanitari Sant Joan de Déu, Sant Boi de Llobregat, Spain); and Eneritz Velasco-Arnaiz, Clàudia Fortuny, Miguel Lanaspa (Malalties Infeccioses i Resposta Inflamatòria Sistèmica en Pediatria, Unitat d´Infeccions, Servei de Pediatria, Institut de Recerca Sant Joan de Déu, Barcelona, Spain).