Simplified detection of Mycobacterium tuberculosis in sputum using smear microscopy and PCR with molecular beacons

Abstract

The prompt diagnosis of smear-negative cases is a prerequisite to controlling tuberculosis (TB). Several new laboratory approaches, including nucleic acid amplification (NAA), are being evaluated in various disease settings to meet this challenge. However, NAA needs simplification before it is widely accepted. Furthermore, a supporting smear result improves confidence in and reliability of PCR. In this context, an asymmetric devR PCR assay using two molecular beacon probes for visual or fluorimetric end-point detection of Mycobacterium tuberculosis was developed. The assays reproducibly detected 25 fg M. tuberculosis DNA versus 100 fg by conventional gel electrophoresis (henceforth referred to as gel assay). The devR and IS6110 PCR assays were blindly evaluated on sputum specimens obtained from a directly observed-treatment short-course centre. Universal sample processing (USP) smear microscopy and culture were used as a supportive test and the 'gold' standard, respectively. Among the 148 specimens analysed, 120 were M. tuberculosis culture-positive. Amongst the 122 direct smear-negative samples, 96 were culture-positive, of which 61 were detected by USP smear microscopy. All 35 USP smear-negative samples were positive by three or more PCR methods. devR PCR had a sensitivity of 92.5 % in the fluorimetric assay versus 86.7 % by visual inspection and 90.8 % by the gel method. IS6110 PCR performed at almost equivalent levels. devR visual and fluorimetric assays considered together yielded an increased sensitivity of 95 % without compromising on a specificity of 92.9 %. The results suggest that the USP smear test is useful for diagnosing direct smear-negative TB and judiciously restricting PCR testing to only smear-negative samples. When used together, these tests can provide rapid diagnosis of smear-negative TB in a cost-effective manner.