Ann Clin Microbiol Antimicrob. 2026 Mar 8. doi: 10.1186/s12941-026-00854-7. Online ahead of print.
ABSTRACT
BACKGROUND: Paediatric tuberculosis (TB) diagnosis remains a challenge due to the paucibacillary nature of the disease, resulting in 51.00% of TB cases remaining undiagnosed, rising to 58% in children under five years of age.
METHODS: This study evaluates the effectiveness of different microbiological methods (Xpert® MTB/RIF Ultra, Anyplex™ MTB/NTM, culture and microscopy) for TB detection in gastric aspirate (GA) samples collected from 483 paediatric patients in Slovakia and the Czech Republic. The sensitivity of each diagnostic method was further analyzed according to patient age.
RESULTS: The highest sensitivity was observed with the Anyplex MTB/NTM assay (38.94%). Interestingly, Xpert MTB/RIF Ultra exhibited lower sensitivity than culture (16.67% versus 21.27%). Considering all methods together, the positivity was not significantly associated with age. The overall positivity rate of all diagnostic methods combined was higher in children under five years (46.7%) and adolescents aged 16-18 years (54.2%) compared with those aged 6-15 years (31.2%). Similarly, the sensitivity of individual diagnostic methods (except Xpert MTB/RIF Ultra) followed the same trend.
CONCLUSION: These data indicate that gastric aspirate microbiology provides only moderate confirmation of paediatric TB in routine practice, and that nucleic acid amplification-based technologies (NAAT) should be used as part of a complementary diagnostic strategy alongside culture and smear microscopy, with culture remaining essential for drug susceptibility testing. Because NAAT platforms were applied in different centre-specific cohorts, the findings should be interpreted as real-world yields rather than a head-to-head comparison, and future work should prioritise standardised sampling and evaluation of non-invasive alternatives (e.g., stool) across age groups.
PMID:41796328 | DOI:10.1186/s12941-026-00854-7