Detection of Multi-Drug Resistant Tuberculosis Using the Genotype MTBDR DNA Strip Assay and Lowenstein Jensen Proportion Method

Abstract

Pulmonary tuberculosis (TB) remains a significant public health issue in low-middle income countries. Accurate and timely diagnosis is key to effective management. Diagnosing Multi Drug Resistance Tuberculosis (MDR-TB) is mostly done with phenotypic Lowenstein Jenson (LJ) proportion method with long turn around which delays treatment. The genotypic MTBDR plus was introduced by World Health Organisation (W.H.O) for the same purpose hence, this study aims to detect MDR-TB using both two methods. Sputum samples were collected from cases of pulmonary TB diagnosed with Genexpert and Ziehl Neelsen stain. Positive samples were subjected to MTBDR plus and the LJ proportion method with the LJ method considered gold standard. Chi square analysis was used to evaluate the Sensitivity, Specificity, Positive Predictive value (PPV), Negative Predictive value (NPV), of the MTBDR plus method compared to the LJ Proportion method. Kappa values were also estimated as a measure of agreement between the two methods. In evaluating the performance of MTBDR plus compared to the LJ proportion method, the sensitivity, specificity, PPV and NPV for the detection of rifampicin and isoniazid  was  93.7%, 90.2%, 92.5% ,91.7% and 88.5%, 95.5%,92.0% and 93.3% respectively, while evaluation MDR-TB was 74.5%, 94.4%,  88.4% and 86.6% with a Kappa value of 0.85,0.84 and 0.71 for Isoniazid, Rifampicin and MDR-TB which indicate almost perfect agreement for both rifampicin and isoniazid and substantial agreement for Multi Drug Resistance (MDR). Compared to LJ proportion method, MTBDRplus performed well in the detection of drug resistance to rifampicin, isoniazid and MDR-TB, hence, is a rapid and efficient tool for the diagnosis and initiation of treatment for MDR-TB.