Setting: Improved strategies are needed for detecting Mycobacterium tuberculosis infection in children in TB-endemic settings.
Objective: To determine the prevalence of M. tuberculosis infection by tuberculin skin testing (TST) and by the QuantiFERON-TB Gold In-Tube (QFT-GIT) test in children with an adult household contact with pulmonary TB in South Africa.
Design: Cross-sectional study.
Results: A total of 167 adult pulmonary TB cases (153/167, 92% human immunodeficiency virus [HIV] infected) and 270 pediatric contacts (median age 6 years, 14/270, 5% HIV-infected) were enrolled. All children completed QFT-GIT testing and 254 (94.1%) completed TST testing. Prevalence of M. tuberculosis infection was 28% (71/254, 95%CI 23-34) using TST (5 mm cut-off) and 29% (79/270, 95%CI 24-35) using QFT-GIT (P = 0.49). Agreement between TST and QFT-GIT was 81% (kappa 0.58). Nineteen (7%) QFT-GIT results were indeterminate. Children aged <2 years were more likely than older children to have indeterminate QFT-GIT results (aOR 5.7, 95%CI 1.5-22, P = 0.01) and discordant QFT-GIT and TST results (aOR 3.5, 95%CI 1.7-7.6, P = 0.001).
Conclusion: Prevalence of M. tuberculosis infection in pediatric contacts was high regardless of the diagnostic method used. TST should not be excluded for the detection of pediatric M. tuberculosis infection in this setting, but QFT-GIT may be a feasible alternative in children aged ≥ 2 years.