Comparison of Loop-Mediated Isothermal Amplification, Microscopy, Culture, and PCR for Diagnosis of Pulmonary Tuberculosis

Abstract

The diagnosis of tuberculosis (TB) in endemic countries is challenging due to high caseloads and limited resources. A simple and cost-effective diagnostic test for the rapid detection of Mycobacterium tuberculosis (M. tuberculosis) in clinical specimens is crucially needed. We evaluated the performance of an in-house assay based on loop-mediated isothermal amplification (LAMP) targeting the M. tuberculosis 16S ribosomal RNA (rRNA) gene for the diagnosis of TB in Thailand. A total of 252 sputum samples from suspected cases of pulmonary TB were analyzed. The sensitivity of LAMP was 99.04% (103/104; 95% confidence interval [CI]: 94.76–9.98%) and 72.73% (16/22; 95% CI: 49.78–89.27%) for smear-positive and smear-negative samples with TB-culture positivity, respectively. LAMP detected 20.69% (24/116) of TB culture negative samples but all those were positive by conventional polymerase chain reaction (PCR). The sensitivity of LAMP was higher than that of sputum microscopy while the performance of LAMP was similar to PCR. None of the samples positive for non-tuberculous mycobacteria by culture and PCR were positive by LAMP. Compared to TB culture, the positive predictive value (PPV), negative predictive value (NPV), and kappa coefficient of LAMP were 83.22%, 88.33%, and 0.75 respectively. Based on the diagnostic performance, we propose that LAMP would be suitable as a potential diagnostic test for rapid TB diagnosis in resource-limited laboratory settings.