RESUMO
INTRODUCTION: In July 2012, the United States Agency for International Development (USAID) Quality Health Care Project introduced the Xpert® MTB/RIF assay at the facility level of the primary health care system in Kyrgyzstan. This study analysed the results of the implementation of Xpert. MATERIALS AND METHODS: Test results from 2734 patients from July 2012 to December 2014 were analysed. The sensitivity and specificity of Xpert in routine programme conditions were evaluated using culture and phenotypic drug susceptibility testing (DST) as gold standard. Contribution to early start of treatment for multidrug-resistant tuberculosis (MDR-TB) was expressed as the median time between availability of the test result and start of treatment. RESULTS: Compared to culture, the sensitivity and specificity of Xpert were respectively 92.7% and 90.4%. For the detection of rifampicin (RMP) resistance, Xpert sensitivity and specificity were respectively 90.1% and 90.7%. The median time to initiation of MDR-TB treatment decreased to 10 days (interquartile range [IQR] 6-16) in 2014 from 20 days (IQR 12-40, P < 0.001) in 2013. CONCLUSION: The Xpert assay demonstrated good agreement in the detection of both Mycobacterium tuberculosis and RMP-resistant pulmonary TB in routine clinical practice. Although Xpert improved the time to treatment initiation from 2013 to 2014, more efforts are needed to further reduce this delay.