e226 | www.pidj.com The Pediatric Infectious Disease Journal •  Volume 33, Number 9, September 2014 ORIGINAL STUDIES Background: The diagnosis of latent or active tuberculosis in children is often challenging. Recently, interferon-γ release assays have been licensed, but their diagnostic accuracy in young children remains questionable as fre- quent false-negative or indeterminate results have been reported. Methods: We performed a multicenter, retrospective study in children 0–24 months of age who were tested at least once with QuantiFERON-TB Gold- in-tube (QTF-IT) ± tuberculin skin test (TST), to analyze its use and perfor- mance in clinical practice. Results: Eight-hundred and twenty-three children (449 males, median age 13.5 months) were enrolled. QTF-IT sensitivity and specificity for active tuberculosis were 92.4% and 98.6%, respectively. Indeterminate tests (4.2 %) were not related to age (P = 0.838) or gender (P = 0.223); 32 children (91.4 %) with an indeterminate QTF-IT ultimately resulted uninfected. In the 616 subjects with valid paired results of QTF-IT and TST, sensitivity and specificity were comparable (91.1% vs. 85.1% and 98.1% vs. 97.9%, respectively). Diagnostic concordance between tests was higher in Bacillus Calmétte-Guerin nonvaccinated children (κ = 0.802). A high rate of dis- cordant tests was observed in latent infections. Conclusions: QTF-IT showed good sensitivity and specificity, and a low rate of indeterminate results in the first 2 years of life, supporting its use at this age. However, considering costs and the similar performance between QTF- IT and TST, it is reasonable to suggest the latter as first-line testing in young children. The complementary use of TST and interferon-γ release assays may be considered in selected cases to improve the accuracy of testing. Key Words: tuberculosis, children, interferon-γ release assay, quantiF- ERON, tuberculin skin testing, infants (Pediatr Infect Dis J 2014;33:e226–e231) I nfants and young children are at high risk of progression of tubercu- losis (TB) infection to severe pulmonary or disseminated disease as their immature immune system, with impaired T-cell response, fails to control intracellular pathogens such as mycobacteria. 1 The prompt identification and treatment of the infection in young children are thus a crucial step in worldwide TB control programs. 2 However, the diagnosis of latent or active TB infection is often challenging in the first years of life. 1,2 The initial phase of infection may be asympto- matic or with nonspecific symptoms and signs. Microscopic exami- nation of sputum samples is often useless because young children are unable to expectorate and generally have a paucibacillary disease; also with induced sputum or gastric aspirate, the culture is positive in only one quarter of cases, rising up to 50% with intensive sam- pling techniques. 3,4 Therefore, the diagnosis of TB or latent tuber- culosis infection (LTBI) often relies upon tuberculin skin test (TST), which has however several limitations, such as reduced sensitivity in patients with an impaired cellular immune response and poor speci- ficity in Bacillus Calmétte-Guerin (BCG)-vaccinated individuals or in nontuberculous mycobacteria infections. 5–7 Furthermore, TST is reliant on correct administration and interpretation and it requires the patient’s recall after 48–72 hours, with possible loss to follow up. 8 Recently, new diagnostic tests, based on the measurement of interferon-γ (IFN-γ) released by T cells in response to specific anti- gens of Mycobacterium tuberculosis (interferon-γ release assays, IGRAs), have become available. 9–11 IGRAs have higher specificity than TST and they are not influenced by BCG vaccination or by most nontuberculous mycobacteria infections; furthermore, IGRAs have higher positive predictive values for the development of active TB than TST. 12,13 Although IGRAs are routinely used for the diagnosis of TB in adults, most guidelines recommend their use in children only after 4 years of age. 14–17 Indeed, the likelihood of an indeterminate test inversely relates to age and frequent false-negative results have been reported in infants. 19,20 Despite the increasing number of tar- geted studies, the diagnostic accuracy of IGRAs in the first years of life remains questionable. 21 QuantiFERON-TB Gold-in-tube (QTF-IT; Cellestis Inc, Valencia, CA) is one of the available IGRAs: it is based on a whole blood enzyme-linked immunosorbent assay and it measures IFN-γ response to antigens CFP-10, ESAT-6 and TB 7.7, encoded by the RD1 region of M. tuberculosis. Aim of our study was to evaluate the extent of use and performance of QTF-IT in routine clinical practice among infants up to 24 months of age. MATERIALS AND METHODS We performed a retrospective, multicenter study that involved 18 Italian pediatric centers. Medical records of children Copyright © 2014 by Lippincott Williams & Wilkins ISSN: 0891-3668/14/3309-e226 DOI: 10.1097/INF.0000000000000353 Performance of Interferon-γ Release Assay for  the Diagnosis of Active or Latent Tuberculosis in  Children in the First 2 Years of Age A Multicenter Study of the Italian Society of Pediatric Infectious Diseases Silvia Garazzino, MD,* Luisa Galli, MD,† Elena Chiappini, MD,† Michele Pinon, MD,* Barbara Maria Bergamini, MD,‡ Salvatore Cazzato, MD, PhD,§ Paola Dal Monte, BScD, PhD,¶ Icilio Dodi, MD,‖ Laura Lancella, MD,** Susanna Esposito, MD,†† Lorenzo Iughetti, MD,‡ Carlotta Montagnani, MD,† Maurizio De Martino, MD,† and Pier-Angelo Tovo, MD,* for The SITIP IGRA Study Group Accepted for publication March 10, 2014. From the *Department of Pediatrics, University of Turin, AOU Città della Salute e della Scienza, Regina Margherita Children’s Hospital, Turin; †Department of Health Sciences, University of Florence, Anna Meyer Children’s Univer- sity Hospital, Florence; ‡Department of Pediatrics, University of Modena and Reggio Emilia, Modena; §Department of Pediatrics, University of Bolo- gna, S. Orsola-Malpighi General Hospital, Bologna; ¶Unit of Microbiology, Department DIMES, University of Bologna, S. Orsola-Malpighi General Hospital, Bologna; ‖Pietro Barilla Children’s Hospital, Parma; **Pediat- rics and Infectious Diseases, Ospedale Pediatrico Bambino Gesù, IRCCS, Rome; and ††Pediatric Clinic 1, Department of Pathophysiology and Trans- plantation, Università degli Studi di Milano, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy. The authors have no funding or conflicts of interest to disclose. Participants of The SITIP IGRA Study Group are listed in Appendix. Address for correspondence: Silvia Garazzino, MD, Regina Margherita Children’s Hospital, Piazza Polonia 94, 10126, Turin, Italy. E-mail: silvia.garazzino@unito.it.