Tuberculosis (TB) is a multisystemic contagious disease produced by Mycobacterium tuberculosis complex bacteria (MTBC), with a prevalence of 65:100,000 inhabitants in Romania (six times higher than the European average). The diagnosis usually relies on the detection of MTBC in culture. Although this is a sensitive method of detection and remains the “gold standard”, the results are obtained after several weeks. Nucleic acid amplification tests (NAATs), being a quick and sensitive method, represent progress in the diagnosis of TB. The aim of this study is to assess the assumption that NAAT using Xpert MTB/RIF is an efficient method of TB diagnosis and has the capacity to reduce false-positive results. Pathological samples from 862 patients with TB suspicion were tested using microscopic examination, molecular testing and bacterial culture. The results show that the Xpert MTB/RIF Ultra test has a sensitivity of 95% and a specificity of 96.4% compared with 54.8% sensitivity and 99.5% specificity for Ziehl–Neelsen stain microscopy, and an average of 30 days gained in the diagnosis of TB compared with bacterial culture. The implementation of molecular testing in TB laboratories leads to an important increase in early diagnostics of the disease and the prompter isolation and treatment of infected patients.

Background: The introduction of the “trace” category in the Xpert MTB/RIF Ultra assay has significantly improved the sensitivity of molecular tuberculosis diagnostics. While it enhances sensitivity, especially in paucibacillary and extrapulmonary cases, its specificity remains debatable, making its interpretation outside select populations a topic of clinical uncertainty. Objectives: This study evaluates the diagnostic and clinical significance of trace-positive results obtained with the Xpert MTB/RIF Ultra assay in the context of a high-incidence TB setting, examining their association with clinical, imaging, and microbiological findings. Methods: A retrospective analysis was conducted on 65 samples with trace-positive Xpert Ultra results, collected over a six-year period from 59 distinct patients in a general hospital in Romania. Correlations were assessed with microscopy, culture, clinical features, imaging, treatment initiation, and prior TB history. A composite reference standard was used for diagnostic accuracy evaluation. Results: Of the 65 trace-positive samples, 29 (44.6%) were culture-positive and 5 (7.7%) were smear-positive. A high proportion of patients, 56 (94.9%), presented with TB-compatible symptoms, and 47 (79.6% of those with imaging) had highly suggestive radiological findings. Based on the composite reference standard, 47 patients (79.7%) were ultimately diagnosed with active TB. Anti-TB treatment was initiated in 44 patients (74.5%). Trace positivity was observed across various specimen types, including sputum, pleural fluid, and cerebrospinal fluid. Conclusions: In high TB burden environments, trace-positive Xpert Ultra results frequently reflect true disease when interpreted within the appropriate clinical and imaging framework. Our findings indicate that, in regions with high tuberculosis incidence such as Romania, trace-positive Xpert Ultra results may contribute meaningfully to clinical decision-making when interpreted alongside clinical and radiological findings, in alignment with current WHO guidance.