QuantiFERON-TB Gold In-Tube test performance in a large pediatric population investigated for suspected tuberculosis infection.

Background: The performance of QuantiFERON-TB Gold In-Tube (QTF-IT) in children is under debate, especially in those under 5 years of age. Moreover, interpretation of discordant QFT-IT/Tuberculin-Skin-Test (TST) results remains controversial. This study aims at studying the sensitivity of QFT-IT and TST in children with active TB cases and exploring risk factors associated with discordant TST+/QFT-IT-.

Methods: Children consecutively referred to one single pediatric center between 2010 and 2017 for suspected tuberculosis infection (TB) were enrolled. All children underwent clinical evaluation, TST and QFT-IT. Sensitivity of QFT-IT and TST in active TB cases and risk factors associated with discordant TST+/QFT-IT- results were assessed. Uni- and multi-variate logistic regression analyses were performed.

Results: Overall, 4631 children (median age 5.67 years) were enrolled, and 205 active TB cases were diagnosed (83 microbiologically confirmed). A high QFT-IT sensitivity was observed in children between 2 and 4 years of age (95.0%; 95%CI: 85.4-100) and in those between 5 and 18 years (89.1%; 95%CI:79.2-99.2) with microbiologically confirmed active TB. However, sensitivity was suboptimal in children younger than 2 years (84.6%; 95%CI: 65.0-100). Independent risk factors associated with discordant TST+/QFT-IT- results in children with latent tuberculosis infection (LTBI) were previous BCG vaccination (aOR:2.18; 95%CI:1.33-3.58; p = 0.002), age <2 years vs. 5-18 years (aOR:7.54; 95%CI:2.52-22.59; p < 0.0001), age 2-4 years vs. 5-18 years (aOR:4.63; 95%CI:2.66-8.06; p < 0.0001), and investigation for screening rather than for contact with a suspected or confirmed case (aOR:3.58; 95%CI:2.30-5.59; p < 0.0001).

Conclusion: Our data suggest that QFT-IT might be used as unique assay in children over 2 years of age investigated for recent immigration/adoption screening and in cases of recent low risk TB contact. This approach could considerably reduce the number of children undergoing pharmacological treatment. Conversely, both tests are recommended in cases of strong clinical suspicion or high risk TB contact in children less than 5 years of age, in order to avoid misdiagnosis.