Bacteriophage-based assays for the rapid detection of rifampicin resistance in Mycobacterium tuberculosis: a meta-analysis

Abstract

Objective: To summarize, using meta-analysis, the accuracy of bacteriophage-based assays for the detection of rifampicin resistance in Mycobacterium tuberculosis.

Methods: By searching multiple databases and sources we identified a total of 21 studies eligible for meta-analysis. Of these, 14 studies used phage amplification assays (including eight studies on the commercial FASTPlaque-TB kits), and seven used luciferase reporter phage (LRP) assays. Sensitivity, specificity, and agreement between phage assay and reference standard (e.g. agar proportion method or BACTEC 460) results were the main outcomes of interest.

Results: When performed on culture isolates (N=19 studies), phage assays appear to have relatively high sensitivity and specificity. Eleven of 19 (58%) studies reported sensitivity and specificity estimates > or =95%, and 13 of 19 (68%) studies reported > or =95% agreement with reference standard results. Specificity estimates were slightly lower and more variable than sensitivity; 5 of 19 (26%) studies reported specificity <90%. Only two studies performed phage assays directly on sputum specimens; although one study reported sensitivity and specificity of 100 and 99%, respectively, another reported sensitivity of 86% and specificity of 73%.

Conclusions: Current evidence is largely restricted to the use of phage assays for the detection of rifampicin resistance in culture isolates. When used on culture isolates, these assays appear to have high sensitivity, but variable and slightly lower specificity. In contrast, evidence is lacking on the accuracy of these assays when they are directly applied to sputum specimens. If phage-based assays can be directly used on clinical specimens and if they are shown to have high accuracy, they have the potential to improve the diagnosis of MDR-TB. However, before phage assays can be successfully used in routine practice, several concerns have to be addressed, including unexplained false positives in some studies, potential for contamination and indeterminate results.