Microbiological detection of childhood tuberculosis from gastric aspirate samples delivers only moderate diagnostic confirmation under real-world clinical conditions, with overall sensitivity varying substantially across different testing methodologies and age cohorts. According to findings from an extensive retrospective evaluation conducted across pediatric centers in Slovakia and the Czech Republic, molecular testing using the Anyplex MTB/NTM assay identified Mycobacterium tuberculosis complex in 38.94 percent of clinically diagnosed cases within its cohort, while Xpert MTB/RIF Ultra achieved a sensitivity of 16.67 percent in its respective patient group. The research examined routine clinical practice records to establish how modern nucleic acid amplification tests perform alongside traditional culture and microscopy in young patients who cannot naturally expectorate sputum, providing detailed empirical data on routine diagnostic yields from gastric lavage procedures.
Diagnosing tuberculosis in children has remained an enduring challenge in clinical microbiology due to the non-specific clinical presentations, difficulties in securing adequate specimens, and the inherently paucibacillary nature of childhood mycobacterial disease. Estimates from the World Health Organization cited in the investigation indicate that worldwide tuberculosis resurgences reached 8.2 million new cases in 2023, representing the highest recorded incidence since 1995, with an estimated 1.25 million of those cases occurring among children and adolescents. Because younger pediatric patients typically cannot generate sputum and frequently swallow respiratory secretions, conventional diagnostic pathways often fail, leaving roughly 51.00 percent of childhood tuberculosis cases undiagnosed worldwide and rising to 58 percent in children under five years of age.
To evaluate microbiological detection under clinical conditions, investigators reviewed electronic medical records from January 2020 through June 2024 for 483 pediatric patients under eighteen years of age with suspected tuberculosis who were hospitalized at the National Institute of Children Tuberculosis and Respiratory Diseases in Slovakia and Thomayer University Hospital in Prague. The overall cohort comprised 249 boys and 234 girls with a mean patient age of 7.2 years, all of whom successfully underwent gastric aspirate collection. Diagnostic evaluation was typically prompted by routine contact tracing, with clinical diagnosis established using a composite reference standard that required both immunological evidence of infection—such as a positive tuberculin skin test or an interferon-gamma release assay—and supportive radiological findings on chest radiographs or computed tomography.
The procedural protocol for gastric aspirate collection involved early morning transesophageal placement of an appropriately sized nasogastric tube into the stomach of a fasting patient. Clinicians performed gastric lavage using 50 milliliters of normal saline administered through the tube, followed by aspiration of 35 milliliters of gastric fluid into sterile collection tubes. In these clinical settings, gastric aspiration served as the primary diagnostic alternative not only for infants and toddlers but also for older children and adolescents above five years of age who were unable to cooperate with sputum collection protocols or lacked access to specialized respiratory physiotherapy staff needed for sputum induction.
Laboratory processing standardized the handling of all collected aspirate specimens prior to diagnostic inoculation. Technicians concentrated the raw gastric fluid through centrifugation at 3,000 revolutions per minute for 15 minutes, after which the resulting sediment was resuspended and decontaminated using N-acetyl-L-cysteine-sodium hydroxide. Following subsequent neutralization with sodium hydroxide and washing in phosphate-buffered saline, the final processed sediment volume measured approximately 3 milliliters. From this single prepared sediment, technicians allocated 200 microliters for direct smear microscopy, 500 microliters across each of three solid Löwenstein-Jensen culture media slants, and 1,000 microliters for nucleic acid extraction in the Slovak cohort or direct loading into the automated cartridge system in the Czech cohort.
Among the 483 patients evaluated, 265 received a final clinical diagnosis of active tuberculosis disease, encompassing 246 cases of pulmonary involvement, 11 cases of extrapulmonary tuberculosis, and 8 cases presenting with acute miliary disease. Overall, 123 children were bacteriologically confirmed as tuberculosis-positive by at least one microbiological method, comprising 98 out of 421 patients evaluated in the Slovakian workflow and 25 out of 62 patients tested in the Czech Republic workflow. Only 32 of these 123 confirmed patients tested positive by more than one microbiological method, and just 10 patients yielded positive results across all three diagnostic modalities encompassing nucleic acid amplification, solid culture, and smear microscopy.
Comparing assay sensitivities against the composite clinical reference standard highlighted notable performance differences between individual laboratory technologies. The Anyplex MTB/NTM real-time polymerase chain reaction assay demonstrated the highest individual sensitivity at 38.94 percent, showing significantly higher detection rates than traditional microbiological techniques and detecting 41 positive gastric aspirate samples that were entirely negative on both culture and smear microscopy. Furthermore, Anyplex MTB/NTM successfully detected mycobacterial DNA in all samples that were positive by culture, demonstrating that automated real-time molecular assays can capture paucibacillary specimens that conventional mycobacterial growth systems fail to isolate.
Conversely, the observed sensitivity of Xpert MTB/RIF Ultra reached 16.67 percent when evaluated against the clinical composite reference standard, proving lower than solid culture, which demonstrated an overall sensitivity of 21.27 percent across the cohort. When evaluated specifically against culture-positive samples as a reference standard, Xpert MTB/RIF Ultra attained a sensitivity of 40.00 percent and a specificity of 100.00 percent, remaining negative in 16 culture-positive gastric aspirates from patients who met the clinical active tuberculosis case definition. Direct acid-fast smear microscopy exhibited the lowest diagnostic yield across all evaluated modalities, identifying only 5.60 percent of active clinical cases.
Analysis of diagnostic positivity rates across age brackets revealed substantial demographic variation across the study population. Individual diagnostic methods, with the sole exception of Xpert MTB/RIF Ultra, mirrored this demographic distribution, reflecting established biological transitions in host immunity, disease manifestation, and bacillary burden across developmental stages.
The authors contextualized these age-associated disparities by noting that school-aged children generally experience a relatively effective immune containment phase characterized by localized intrathoracic granulomatous lesions with minimal bacillary dissemination, which substantially depresses the concentration of mycobacteria in swallowed secretions. Younger children under five years frequently swallow higher volumes of respiratory secretions and may present with higher bacillary concentrations, while older adolescents develop adult-like pulmonary manifestations such as cavitary lesions that naturally generate higher bacterial loads. Disease severity also dictated detection success across all age groups, with gastric aspirates yielding positive results in 75.0 percent of acute miliary tuberculosis cases compared to 27.27 percent in extrapulmonary presentations.
In addition to detecting tuberculosis complex, the Anyplex assay identified nontuberculous mycobacterial DNA in seven gastric aspirate specimens, all of which yielded negative smear microscopy and negative cultures. Because none of these seven pediatric patients met established clinical criteria for nontuberculous mycobacterial disease, investigators interpreted these findings as transient mucosal colonization or environmental contamination from swallowed non-sterile matter rather than definitive invasive disease. Four of these seven children had clinically confirmed active tuberculosis, underscoring the clinical necessity of interpreting positive nontuberculous molecular signals cautiously within the complete clinical and bacteriological context of the patient.
The study authors emphasized several critical methodological limitations that prevent interpreting these retrospective yields as direct head-to-head proof of technical superiority. The two molecular platforms were deployed across separate center-specific patient populations with different baseline cohort sizes, precluding randomized sample-matched comparisons. Furthermore, the volume and bacterial concentration of gastric aspirates depend heavily on patient fasting compliance and aspiration technique, which may particularly restrict fixed-volume automated cartridge systems. Because solid culture remains indispensable for comprehensive drug susceptibility testing, the researchers concluded that nucleic acid amplification tests must serve as complementary tools alongside conventional microbiological workflows, while future pediatric research prioritizes standardized sampling protocols and non-invasive diagnostic alternatives such as stool testing.
Subject of Research: Diagnostic accuracy of microbiological and molecular tests for pediatric tuberculosis using gastric aspirate specimens
Article Title: Comparative evaluation of Anyplex™ MTB/NTM, Xpert MTB/RIF ultra, culture, and microscopy for the diagnosis of pediatric tuberculosis using gastric aspirates
Article References: Dohál, M., Dvořáková, V., Hromádková, M., Kunč, P., Fábry, J., Porvazník, I., Solovič, I., Doležalová, K., Krivošová, M., & Mokrý, J. (2026). Comparative evaluation of Anyplex™ MTB/NTM, Xpert MTB/RIF ultra, culture, and microscopy for the diagnosis of pediatric tuberculosis using gastric aspirates. Annals of Clinical Microbiology and Antimicrobials, 25(1), Article 18. https://doi.org/10.1186/s12941-026-00854-7
Image Credits: AI Generated
DOI: 10.1186/s12941-026-00854-7
Keywords: pediatric tuberculosis, gastric aspirates, Anyplex MTB/NTM, Xpert MTB/RIF Ultra, mycobacterial culture, smear microscopy, Mycobacterium tuberculosis, nontuberculous mycobacteria, childhood diagnostics, paucibacillary tuberculosis, nucleic acid amplification, molecular diagnostics, clinical microbiology, pulmonary tuberculosis, retrospective cohort, diagnostic sensitivity
News Source: Harold Sullivan. (November 12, 2026). Gastric Aspirate Testing Shows Variable Sensitivity in Diagnosing Pediatric Tuberculosis. Scienmag.
