INT J TUBERC LUNG DIS 20(1):115–120 Q 2016 The Union http://dx.doi.org/10.5588/ijtld.15.0227 Accuracy of Xpert W MTB/RIF assay compared with AdvanSure E TB/NTM real-time PCR using bronchoscopy specimens Y. Ko,* † H-K. Lee, † Y. S. Lee, †‡ M-Y. Kim, † J. H. Shin, § E-J. Shim, ¶ S. Y. Park,* E. K. Mo,* Y. B. Park* *Department of Pulmonary and Critical Care Medicine, Kangdong Sacred Heart Hospital, Hallym University College of Medicine, Seoul, † Division of Pulmonary, Allergy and Critical Care Medicine, Department of Internal Medicine, Busan Paik Hospital, Inje University College of Medicine, Busan, ‡ Division of Pulmonology, Department of Internal Medicine, Institute of Chest Disease, Severance Hospital, Yonsei University College of Medicine, Seoul, § Department of Laboratory Medicine, Busan Paik Hospital, Inje University College of Medicine, Busan, ¶ Department of Pharmacology and PharmacoGenomics Research Center, Inje University College of Medicine, Inje, Republic of Korea SUMMARY BACKGROUND: The performance of Xpert w MTB/RIF assay, an automated nucleic acid amplification test (NAAT) that was developed for the detection of tuberculosis (TB), has been evaluated in various clinical settings. However, few studies have compared Xpert with other NAATs, especially its performance using lower respiratory tract specimens (LRTS). OBJECTIVE: To compare the practical diagnostic per- formance of the Xpert assay with that of the Ad- vanSure e TB/NTM RT-PCR kit in the detection of pulmonary TB (PTB), using LRTS obtained through bronchoscopy. RESULTS: Of 249 patients included, 105 had culture- confirmed PTB. Using culture as reference, the overall sensitivity of Xpert and AdvanSure was respectively 92.4% and 83.8%. When acid-fast bacilli smear results were taken into consideration, the sensitivity of Xpert for smear-positive and smear-negative LRTS was respec- tively 100% and 88.9%, while that of the AdvanSure was 100% and 76.4%. Xpert showed better results than AdvanSure in terms of sensitivity in smear-negative LRTS (P ¼ 0.012), but no difference in smear-positive LRTS. CONCLUSIONS: Xpert may be advantageous in the detection of PTB using LRTS, particularly in low microbiological burden settings. KEY WORDS: Xpert w MTB/RIF assay; Mycobacterium tuberculosis; bronchoscopy; lower respiratory tract specimen ALTHOUGH GLOBAL EFFORTS have reduced the overall incidence of tuberculosis (TB), the disease remains a major worldwide health problem. 1 In 2012, 8.6 million new cases and 1.3 million deaths due to TB were reported. 2 To reduce TB incidence and prevent its transmission, rapid and early diagno- sis of TB and timely initiation of appropriate treatment are essential. Although identification of Mycobacterium tuber- culosis using microbiological culture is the standard method, it is relatively time-consuming and usually requires 2–6 weeks to confirm the diagnosis of TB. 3 Since the introduction of molecular methods for the rapid detection of TB, several commercial systems using nucleic acid amplification tests (NAATs) have been developed. 4 The Xpert w MTB/RIF assay (Ce- pheid Inc, Sunnyvale, California, USA), recently endorsed by the World Health Organization (WHO) for the rapid diagnosis of TB, allows fully automated sample preparation, amplification, and simultaneous detection of M. tuberculosis and rifampicin resistance by real-time (RT) polymerase chain reaction (PCR) using a single-use disposable cartridge. 5 The WHO now recommends the Xpert assay over conventional tests for the diagnosis of extra-pulmonary TB, such as TB of the lymph nodes and other tissues, and as the preferred initial test for the diagnosis of TB menin- gitis. In presumed PTB cases, Xpert may be used rather than conventional microscopy and culture as the initial diagnostic test. 6,7 Since the introduction of the Xpert assay for rapid molecular detection of TB, several studies have evaluated the diagnostic performance of the assay in comparison with that of other NAATs using respira- tory specimens such as sputum and lower respiratory tract specimens (LRTS). 8–13 However, few studies YSK and HKL contributed equally to this work. Correspondence to: Yousang Ko, Department of Pulmonary and Critical Care Medicine, Hallym University Kangdong Sacred Heart Hospital, 150 Seongan-ro Gangdong-gu, Seoul 134-701, Republic of Korea. Tel: ( þ 82) 2 2224 2754. Fax: ( þ 82) 2 2224 6925. e-mail: koyus@naver.com Article submitted 12 March 2015. Final version accepted 20 July 2015.