Diagnostic accuracy of stool based Xpert MTB/RIF ultra assay for diagnosis of pulmonary tuberculosis in children, Northwest Ethiopia

Diagnosing childhood pulmonary tuberculosis (PTB) presents unique challenges, including paucibacillary disease, difficulties in specimen collection, and limited diagnostics. Hence, millions of children with tuberculosis (TB) infection remain underdiagnosed. Therefore, our research team evaluated diagnostic accuracy of stool-based Xpert ultra for detecting PTB in children. Five ml respiratory and 3 g stool specimens were analyzed. Strict adherence to laboratory standard operating procedures were warranted. Diagnostic performance of stool Xpert Ultra was compared against Xpert ultra and culture on respiratory samples, and composite reference standards. Of 557 children recruited; 510 had complete microbiological results. PTB was diagnosed in 52/510 (10.2%) children with presumptive TB. Four had microbiologically unconfirmed PTB, who were clinically diagnosed with positive response to anti-TB, remaining 48 were confirmed microbiologically. Stool specimen Xpert ultra had sensitivity 95.8%(95%CI: 82.8-98.6) and specificity of 99.8% (95%CI: 98.7-100) compared to culture; yet, sensitivity of stool was 88.5% (95%CI: 72-95.6) and specificity 100% (95%CI: 99.2-100) when compared to CRS. Xpert ultra on respiratory specimen had sensitivity and specificity 97.9% (95%CI: 85.8- 99.5) and 99.8% (95%CI: 98.7-100) compared to culture and 92.3%(95%CI: 81.4-97.9) and 100% (95%CI: 99.2-100) compared to CRS. Thus, stool samples present promising alternative for diagnosing PTB in children when respiratory specimens cannot be obtained.