Interpreting trace results from the Xpert® MTB/RIF Ultra assay, a frontline diagnostic tool for tuberculosis, remains challenging, particularly in low-tuberculosis burden settings. This study aimed to provide practical recommendations for managing trace-positive patients in Lebanon based on clinical, radiological, and bacteriological data. A retrospective observational study was conducted at the National Tuberculosis Reference Laboratory in Lebanon, Laboratoire Rodolphe Mérieux, from January 2020 to December 2024. Samples referred by the Lebanese National Tuberculosis Program for both Xpert® MTB/RIF Ultra and mycobacterial culture were included. Non-conforming samples or those with invalid/error Xpert® MTB/RIF Ultra results were excluded. For patients with trace-positive results, correlations were assessed with microbiological parameters (microscopy and culture), clinical characteristics (symptoms, prior TB history and treatment initiation) and radiological findings. Diagnostic accuracy and case classification were assessed using a Composite Clinical Reference Standard (CRS) that excluded the index test to avoid incorporation bias. Among 1,061 Ultra tests, 42 (4.0%) yielded trace-positive results. Patients with trace results had a mean age of 42 years (range: 0–75), and 59.5% were female. Two patients (4.8%) had a history of prior tuberculosis treatment, and one (2.4%) was HIV-positive. Most trace-positive results (33 cases; 78.6%) were derived from respiratory specimens. Culture confirmed Mycobacterium tuberculosis complex in 9 cases (21.4%). Using the CRS, 20 patients (47.6%) were classified as Definite TB (9 patients) or Probable TB (11 patients) and initiated first‑line therapy; 17 of them (85.0%) achieved favorable outcomes. In low‑burden settings like Lebanon, the interpretation of Xpert® MTB/RIF Ultra trace results is strengthened by applying a Composite Clinical Reference Standard that integrates clinical, radiological, and microbiological criteria. This approach supports more informed clinical classification, guides appropriate treatment decisions, and was associated with favorable outcomes in our cohort. These exploratory findings support the potential value of CRS‑based frameworks and warrant confirmation in prospective studies.
Soutou et al. (Thu,) studied this question.