C-206. Clinical Evaluation of Commercial Strand Displacement Amplification (SDA) System in Detecting Mycobacterium tuberculosis complex (MTBC) in Smear-Negative Respiratory and Non-Respiratory Specimens

C-206. Clinical Evaluation of Commercial Strand Displacement Amplification (SDA) System in Detecting Mycobacterium tuberculosis complex (MTBC) in Smear-Negative Respiratory and Non-Respiratory Specimens

L-H. Sng, P. Y. F. Cheah, M. Y. Choy, M. D. G. Quieng, A. J. V. Ramirez, S. Wang;
Singapore Gen. Hosp., Singapore, SINGAPORE.

The ProbeTec ET^® Direct Detection Assay (DTB) (Becton Dickinson Diagnostic Instruments, Sparks, MD) is able to provide rapid evidence for the presence of MTBC in clinical specimens. However, information on its performance compared against culture techniques is sparse, particularly among smear-negative specimens. We aimed to compare the performance of this technique against culture methods, primarily the automated BACTEC 960 MGIT system (Becton Dickson Diagnostic Systems, Sparks, MD), among smear-negative specimens. Of 505 respiratory and non-respiratory specimens processed using standard methods, including digestion-decontamination using BBL™ Myco-Prep™, sediments were aliquoted out for SDA and for inoculation into MGIT and Lowenstein-Jensen media. Processed tissue specimens were also inoculated into Bactec 12B vials and incubated at room temperature. When SDA and culture results were discrepant, they were adjudicated independently by an observer using laboratory and radiological findings, clinical history, clinical decision to initiate anti-tuberculous therapy and response to treatment. The overall sensitivity and specificity of the SDA method was 97.8% and 99.0% respectively, with positive and negative predictive values of 91.7% and 99.8% respectively. Comparing between respiratory (n=83) and non-respiratory (n=422) specimens, there was no difference in the sensitivity (92.3 vs 100%; p=0.30), specificity (97.1 vs 99.2%; p=0.31), positive (85.7 vs 91.2%; p=0.60) and negative (98.6 vs 100%; p=0.32) predictive values. According to the test and corresponding internal MOTA scores, there was only one specimen affected by inhibitory reactions (0.2%). Our findings showed that the DTB has a high degree of accuracy for the detection of MTBC in smear-negative respiratory and non-respiratory specimens in a clinical laboratory.