Proportions of Xpert MTB/RIF Ultra 'trace' results vary widely among different populations with presumptive TB

BACKGROUND Microbiological confirmation of Mycobacterium tuberculosis (MTB) with Xpert ® MTB/RIF Ultra is recommended, but the interpretation of a 'trace' result is uncertain. We aimed to determine the proportion of Xpert Ultra-positive results reported as 'trace' and how the 'trace' result informed patient care. METHODS We searched MEDLINE OVID, EMBASE, PubMed, Scopus, and Web of Science. All studies until September 2023 that reported Xpert Ultra 'trace' and non-trace MTB-positive results were included for analysis. The Joanna Briggs Institute tool evaluated methodological quality. RESULTS The review identified 62 studies conducted across 23 countries with 124,749 participants. The proportions of 'trace' varied widely from 3.1% to 80%; this proportion was 30% or higher in 26 (42%) studies. Higher 'trace' proportions were reported among children and in extrapulmonary samples. MTB culture, when done, was positive in 23.5% of 'trace' compared to 60.5% of non-trace positive results. Data on treatment decisions for those with 'trace' results were usually not reported, although some who were not treated developed TB. CONCLUSION Xpert Ultra 'trace' results are common, with wide variation between populations and inconsistencies in interpretation. The implications of a 'trace' result for clinical management and reporting categorisation require greater clarity. .