OBJECTIVE Accurate diagnosis of tuberculosis (TB) is crucial to facilitate early treatment of the patients, and to reduce its spread. Clinical presentation of Mycobacterium tuberculosis complex (MTBC) and non tuberculous mycobacteria (NTM) may or may not be the same, but the treatment regimen is always different for both the infections. Differentiation between MTBC and NTM by routine laboratory methods is time consuming and cumbersome. This study was aimed to evaluate an immunochromatographic test (ICT), based on mouse monoclonal anti-MPT64, for simple and rapid discrimination between MTBC and NTM in clinical isolates from extra-pulmonary tuberculosis cases. METHODS A total of 800 clinical samples were collected from patients suspected to have extra-pulmonary tuberculosis. Preliminary diagnosis has been done by direct Ziehl-Neelsen (ZN) staining followed by culture in BACTEC system. A total of 150 clinical isolates, which were found positive in BD 460 TB system during September 2009 to September 2010 were selected for the screening by ICT test. p-nitro-α-acetylamino- β-hydroxy propiophenone (NAP) test was performed for differentiation of MTBC and NTM. M. tuberculosis complex was further confirmed by IS6110 PCR of BACTEC culture positive isolates, this served as the reference method for MTBC identification and comparative evaluation of the ICT kit. RESULTS Of the 150 BACTEC culture positive isolates tested by ICT kit, 101 (67.3%) were found positive for MTBC and remaining 49 (32.7%) were considered as NTM. These results were further confirmed by IS6110 PCR that served as the reference method for detection of MTBC. H 37 Rv reference strain was taken as a control for ICT test and IS6110 PCR. The reference strain showed the presence of MPT64 antigen band in the ICT test. Similar bands were formed in 101 of 102 MTBC isolates tested, proving 99.1 per cent sensitivity and no bands were detected in 48 (100%) NTM isolates tested, proving 100 per cent specificity of the ICT kit. CONCLUSIONS Our findings show that ICT test can be used on direct culture positive specimens. It does not require any special equipment, is simple and less time consuming. It can easily discriminate between MTBC and NTM and thus can help in appropriate management of tuberculosis.