Performance of a Novel Stool Quantitative Polymerase Chain Reaction Assay for Pediatric Tuberculosis Detection in Sub-Saharan Africa.

Background: Children have paucibacillary tuberculosis and cannot provide expectorated sputum. Invasive specimen collection, by gastric aspiration or sputum induction, has a low diagnostic yield. In this study, we aimed to evaluate the diagnostic performance and additive yield of a novel stool-based assay in children diagnosed with tuberculosis in sub-Saharan Africa.

Methods: We conducted a prospective case-control study from October 2020 to June 2023 in Eswatini, Mozambique, and Tanzania. Children under 15 years newly diagnosed with tuberculosis completed clinical examination, chest radiography, culture, sputum Xpert Ultra, stool Xpert Ultra, and stool-based quantitative polymerase chain reaction (stool qPCR) assessment. Stool qPCR sensitivity was calculated against culture, a composite microbiological reference standard, and a clinical reference standard. Specificity was calculated in a control population of healthy, TB disease-free, child household contacts.

Results: Among 456 children, 232 were diagnosed with TB and 224 controls. Stool sample collection was achieved in 95.6% of children. The qPCR was positive in 17.2% (40/232) of clinically diagnosed participants. In the same population, test positivity was 8% (13/162) for culture, 13.4% (27/202) sputum Xpert Ultra, and 14.8% (33/223) stool Xpert Ultra. When compared to a microbiological reference standard (any positive test), the sensitivity of stool qPCR was 35.6% (21/59). Specificity in the control population was 96.1% (196/204), and the additive yield of qPCR with all tests performed was of 8.7%.

Conclusion: This stool qPCR assay can increase the microbiologic confirmation of tuberculosis in pediatric populations from TB high-burden settings. It may be particularly useful where resource limitations or clinical capacity impedes diagnostic specimen collection via sputum induction or gastric aspiration.