The evaluation of whole blood cytokine assay for diagnosis of M.tuberculosis infection in South African children with household tuberculosis contact.

Abstract

Background: There are critical unmet needs for improved strategies in the detection and diagnosis of M.tuberculosis infection in children, and for prevention of tuberculosis disease in children. Bacillus Calmette-Guérin (BCG) vaccination has limited the utility of tuberculin skin testing (TST) in areas with high vaccine coverage. Objectives: The aim of this study was to estimate the prevalence of M.tuberculosis infection in children with household tuberculosis contacts, using QFT-GIT testing in comparison with TST. Methods: This study was a cross-sectional design to assess the performance of a new T-cell based blood test, namely QuantiFERON-TB Gold In Tube (QFT-GIT), for diagnosis of tuberculosis infection in the children (n=182) of adults (n=124) with pulmonary tuberculosis, additionally to determine the prevalence of M.tuberculosis infection in children with household tuberculosis contacts, using QFT-GIT testing in comparison with TST. The study was carried out at Chris Hani Hospital. For children involved in the study, tuberculosis exposure information was obtained, together with TST, QFT-GIT, and HIV testing. Data obtained from both experiments was statistically analysed using SPSS version 24 to determine whether there was a significant agreement between QFT-GIT and TST on the detection of M.tuberculosis prevalence in children with house hold contacts with confirmed M.tuberculosis infection. Results: This study examined the sensitivity and specificity of the QFT-GIT tests compared with the standard TST for diagnosing latent tuberculosis disease in paediatric contacts. Because of the lack of a latent tuberculosis “gold standard”, the specificity and sensitivity of QFT-GIT was calculated with a two-by-two table method. The specificity of the QFT-GIT was 84% and the sensitivity was 85%. There was a good correlation between QFT-GIT and TST (Cohen’s kappa of 0.705). Seventeen percent (17%) of the 182 children tested by QFT-GIT yielded indeterminate results. Age was associated with indeterminate QFT-GIT results in paediatric tuberculosis contacts. Point prevalence for QFT-GIT was recorded as 31% at baseline and 39.5% after six months indicating variability between QFT-GIT results at baseline and after six months. Conclusion: It was concluded that the prevalence of tuberculosis infection was common among South African children who live with an adult with active tuberculosis. The agreement between QFT-GIT assay and TST for the diagnosis of latent tuberculosis in children was high. Although TST and QFT-GIT assays appeared comparable, QFT-GIT showed higher positivity rate amongst those contacts with reported household tuberculosis exposure compared to TST. The QFTGIT assay was a better indicator of the risk of M.tuberculosis infection than TST in a BCG-vaccinated population.