<![CDATA[BACKGROUND: The diagnosis of tuberculosis (TB) in children is challenging by the absence of a practical gold standard. Interferon (IFN)-ginducible protein 10 (IP-10) is a chemokine that may serve as the leading candidate marker in child TB diagnosis. The aim of this study is to assess the diagnostic value of IP-10 in the diagnosis of TB in children.METHODS: We recruited eligible symptomatic and asymptomatic children aged <15 years actively by contact investigation and passively from inpatient and outpatient clinics in two hospitals, in Yogyakarta, Indonesia. We conducted clinical examination and chest X-ray in all eligible children. Sputum smear and the rapid molecular TB test were performed in children with TB symptoms. All participants underwent blood sampling for IFN-g Release Assay and IP-10 test.RESULTS: A total of 79 children were recruited to this study. Twelve children were with TB disease, 16 with latent TB infection (LTBI), 40 were TB-exposed only and 11 were non-TB. Children with evidence of TB infection either with TB disease or LTBI had higher levels of antigen-stimulated IP-10 compared to non-infected children, both TB exposed only and non-TB (p=0.000). A cut-off 408.74 pg/mL for antigen-stimulated IP-10 showed high diagnostic accuracy for diagnosis of TB infection (AUC: 0.97, 95% CI: 0.92-1.00, sensitivity: 92.3%, and specificity: 91.9%). However, the stimulated levels of IP-10 between children with TB disease and LTBI were not significantly different (p=0.268).CONCLUSION: IP-10 performed well to diagnose TB infection in children. However, it cannot be used to differentiate TB infection from TB disease.KEYWORDS: IFN-g, IP-10, latent TB, active TB, children]]>
