Evaluation of stool-based testing to diagnose tuberculosis in children using the Truenat platform in routine settings in Nigeria

Abstract

The World Health Organization (WHO) recommends testing stool with Xpert MTB/RIF Ultra (Xpert-Ultra) as an initial diagnostic test for detection of tuberculosis (TB) and rifampicin resistance in children. The option of testing stool on the Truenat platform could benefit children and their caregivers in remote areas where GeneXpert is not available. We report the results of research to validate a protocol for processing and testing stool on Truenat using an adapted Simple One Step (SOS) stool method in routine settings in Nigeria. Stool specimens from children with presumptive TB were tested with Truenat MTB Plus (Truenat) and Xpert-Ultra using a comparative cross-sectional study design. A total of 510 children were enrolled and submitted a stool specimen. Of these, 482 (94.5%) had valid results on both platforms, 28 (5.8%) with MTB detected and 454 MTB not detected. Of the 28 with MTB detected, eight were detected on both platforms, seven on Truenat only, and 13 on Xpert-Ultra only. The concordance rate between Truenat and Xpert-Ultra was high at 95.8%. Significantly more non-determinate (error/invalid) results were observed with Truenat compared to Xpert-Ultra (4.7% versus 1.2% respectively; P < 0.001). Truenat can accurately detect MTB using 100 mg of stool with an adapted SOS stool protocol. Further optimizing extraction methods could reduce the frequency of non-determinate results. These results hold promise for expansion of childhood TB diagnosis services to hard-to-reach areas with limited infrastructure that are suitable for Truenat placement. Global scale-up of these alternative testing approaches will be necessary to close the gap in childhood TB case finding.