Sensitive and Feasible Specimen Collection and Testing Strategies for Diagnosing Tuberculosis in Young Children

Area of Science:

• Pediatric infectious diseases
• Diagnostic microbiology
• Global health

Background:

• Tuberculosis diagnosis in children under 5 often relies on invasive gastric aspirate (GA) or induced sputum, limiting use in resource-limited settings.
• Developing less invasive, sensitive diagnostic methods is crucial for effective childhood tuberculosis management.

Purpose of the Study:

• To evaluate the sensitivity of minimally invasive specimen and assay combinations for diagnosing tuberculosis in young children.
• To compare the diagnostic yield of various specimen types and testing strategies against established reference standards.

Main Methods:

• A prospective cross-sectional study in Kenya involving children under 5 with suspected tuberculosis.
• Multiple specimen types (including gastric aspirate, induced sputum, nasopharyngeal aspirate, stool, urine) were tested using Xpert MTB/RIF assay and culture.
• Combinations of specimens and tests were analyzed to determine their diagnostic yield relative to the maximum observed yield.

Main Results:

• Gastric aspirate (GA) had a sensitivity of 77%, and induced sputum had 64% for confirmed tuberculosis cases.
• Nasopharyngeal aspirate (NPA) alone or combined with stool or urine showed comparable yields to reference standards (74%, 71%, 69% sensitivity, respectively).
• Combining GA and NPA samples achieved a 90% yield, exceeding current reference standards.

Conclusions:

• Nasopharyngeal aspirate (NPA), especially in duplicate or combined with stool/urine, is a viable, less invasive option for pediatric tuberculosis diagnosis.
• The combination of gastric aspirate (GA) and nasopharyngeal aspirate (NPA) demonstrates superior diagnostic yield and potential for improving tuberculosis detection in resource-limited settings.