Three year evaluation of Xpert MTB/RIF in a low prevalence tuberculosis setting: A Scottish perspective. Issue 5 (May 2017)

• Record Type: Journal Article
• Authors: Parcell, Benjamin J.; Jarchow-MacDonald, Anna A.; Seagar, Amie-Louise; Laurenson, Ian F.; Prescott, Gordon J.; Lockhart, Michael
• Is Part Of: Journal of infection. Volume 74:Issue 5(2017)
• Page Start: 466
• Page End: 472
• Publication Date: 2017-05
• Subjects: Xpert MTB/RIF -- Mycobacterium tuberculosis -- Tuberculosis -- Molecular testing -- Early diagnosis -- Nucleic acid amplification; Infection -- Periodicals; Bacterial Infections -- Periodicals; Communicable Diseases -- Periodicals; Electronic journals; 616.905
• DOI: 10.1016/j.jinf.2017.02.005
• Languages: English
• ISSNs: 0163-4453

Abstract

Summary: Objectives: Xpert MTB/RIF (Cepheid) is a rapid molecular assay shown to be sensitive and specific for pulmonary tuberculosis (TB) diagnosis in highly endemic countries. We evaluated its diagnostic performance in a low TB prevalence setting, examined rifampicin resistance detection and quantitative capabilities predicting graded auramine microscopy and time to positivity (TTP) of culture. Methods: Xpert MTB/RIF was used to test respiratory samples over a 3 year period. Samples underwent graded auramine microscopy, solid/liquid culture, in-house IS 6110 real-time PCR, and GenoType MTBDRplus (HAIN Lifescience) to determine rifampicin and/or isoniazid resistance. Results: A total of 2103 Xpert MTB/RIF tests were performed. Compared to culture sensitivity was 95.8%, specificity 99.5%, positive predictive value (PPV) 82.1%, and negative predictive value (NPV) 99.9%. A positive correlation was found between auramine microscopy grade and Xpert MTB/RIF assay load. We found a clear reduction in the median TTP as Xpert MTB/RIF assay load increased. Rifampicin resistance was detected. Conclusions: Xpert MTB/RIF was rapid and accurate in diagnosing pulmonary TB in a low prevalence area. Rapid results will influence infection prevention and control and treatment measures. The excellent NPV obtained suggests further work should be carried out to assess its role in replacing microscopy. Highlights: Xpert MTB/RIF gave accurate results in a low TB and HIV prevalence setting. It had a very good NPV useful for ruling out pulmonary TB. Auramine microscopy grade and Xpert MTB/RIF assay load correlated. Median TTP reduced as Xpert MTB/RIF assay load increased. Molecular testing respiratory samples may be appropriate for infectivity assessment.