作者: Rocio Sanjuan-Jimenez , Inmaculada Toro-Peinado , Pilar Bermudez , Juan D. Colmenero , Pilar Morata
DOI: 10.1371/JOURNAL.PONE.0143025
关键词:
摘要: Background Nucleic acid amplification tests are increasingly used for the rapid diagnosis of tuberculosis. We undertook a comparative study efficiency and diagnostic yield real-time PCR senX3-regX3 based assay versus classical IS6110 target new commercial methods. Methods This single-blind prospective included 145 consecutive samples: 76 from patients with culture-confirmed tuberculosis (86.8% pulmonary 13.2% extrapulmonary tuberculosis: 48.7% smear-positive 51.3% smear-negative) 69 control samples (24 diagnosed non-tuberculous mycobacteria infections 45 suspected which was eventually ruled out). All were tested by two CE-marked assays (Xpert®MTB/RIF AnyplexTM plus MTB/NTM) in-house targeting gene. Results The detection limit ranged 1.00E+01 fg Anyplex, to 1.00E+04 Xpert. three Xpert, detected all 37 cases. Conversely, Anyplex positive in 34 (91.9%) In smear-negative tuberculosis, differences observed between assays; Xpert 22 (56.41%) 39 samples, 24 (61.53%), 28 (71.79%) 35 (89.74%). negative samples; however, false rate 8.7% 13% IS6110, respectively. The overall sensitivity 77.6%, 85.7%, 77.3% 94.7% specificity 100%, 90.8% 87.0% senX3-regX3, assays, respectively. Conclusion Real-time lack desired specificity. MTB/RIF both sensitive specific MTBC samples. Therefore, real time could be useful complementary tool