Abstract
BACKGROUND: The diagnosis of childhood tuberculosis remains a challenge worldwide. The Xpert MTB/RIF test, a rapid mycobacteria tuberculosis diagnostic tool, was recommended for use in children based on data from adult studies. We evaluated the performance of the Xpert MTB/RIF test for the diagnosis of childhood pulmonary tuberculosis using one induced sputum sample and described clinical characteristics associated with a positive Xpert MTB/RIF test. The sputum culture on both Lowenstein-Jensen (LJ) and Mycobacteria Growth Indicator Tube (MGIT) was the gold standard. METHODS: We consecutively enrolled 250 Ugandan children aged 2 months to 12 years with suspected pulmonary tuberculosis between January 2011 and January 2012 into a cross-sectional diagnostic study at a tertiary care facility in Uganda. RESULTS: We excluded data from 15 children (13 contaminated culture and 2 indeterminate MTB/RIF test results) and analysed 235 records. The Xpert MTB/RIF test had a sensitivity of 79.4% (95% CI 63.2 - 89.7) and a specificity of 96.5% (95% CI 93 - 98.3). The Xpert MTB/RIF test identified 13 of the 14 (92.9%) smear positive-culture positive and 14 of the 20 (70%) smear negative -culture positive cases. The Xpert MTB/RIF identified twice as many cases as the smear microscopy (79.4% Vs 41.2%). Age > 5 years (OR 3.3, 95% CI 1.4 - 7.4, p value 0.005), a history of Tuberculosis (TB) contact (OR 2.4, 95% CI 1.1 - 5.2, p value 0.03), and a positive tuberculin skin test (OR 4.1, 95% CI 1.7 - 10, p value 0.02) was associated with a positive Xpert MTB/RIF test. The median time to TB detection was 49.5 days (IQR 38.4-61.2) for LJ, and 6 days (IQR 5 - 11.5) for MGIT culture and 2 hours for the Xpert MTB/RIF test. CONCLUSION: The Xpert MTB/RIF test on one sputum sample rapidly and correctly identified the majority of children with culture confirmed pulmonary tuberculosis with high specificity.