RESUMEN
INTRODUCTION: Since 2018, World Health Organization (WHO) recommended the Xpert MTB/RIF Ultra use for pulmonary and extrapulmonary TB diagnosis, and suggested that Xpert Ultra should be tested in various populations, with different geographical and epidemiological settings. METHODS: Cross-sectional study with prospective data collection. Outpatients aged >18 years with respiratory symptoms suggestive of pulmonary TB were invited to participate. Sensitivity, specificity, positive and negative predictive values of the test were calculated and compared with the traditional Xpert MTB/RIF. RESULTS: During the study period, 180 patients met the inclusion and were included in the analysis. Xpert MTB/RIF Ultra test was positive in 33 patients (18.3%), and RIF resistance was detected in 1 (3.1%) patient. Considering culture as the gold standard, the sensitivity, specificity, positive predictive value, and negative predictive value of Xpert MTB/RIF Ultra were 100.0% (95% CI 85.2-100.0), 93.6% (95% CI 88.6-96.9), 69.7% (95% CI 55.8-80.7), and 100.0% (95% CI 87.2-100.0), respectively. The area under the ROC curve was 0.97 for the Xpert MTB/RIF Ultra test (95% CI 0.93 to 0.99; p < 0.0001). There was no difference statistically significant between sensitivities and specificities of Xpert MTB/RIF and Xpert MTB/RIF Ultra (p > 0.05). CONCLUSIONS: This is the first study in Brazil to evaluate the accuracy of Xpert MTB/RIF Ultra in individuals with presumptive pulmonary TB. The test showed an excellent sensitivity and a high specificity, demonstrating that it is a useful tool for pulmonary TB diagnosis.