AbstractRapid TB diagnosis and drug susceptibility testing is limited by the slow growth of the causative organism, Mycobacterium tuberculosis (MTB). Effective and rapid methods for the detection and drug susceptibility testing of TB, that are suitable for implementation in low income countries, which bear the highest TB burden, are urgently needed for effective identification, treatment and control of the disease. Mycobacteriophages have the potential to become useful tools in the diagnosis of TB, as they are specific for mycobacteria and only replicate in, and hence detect, viable cells. Phage‐based techniques involve simple manual manipulations that typically require little capital expense and yield results rapidly. Two techniques have been developed recently which use mycobacteriophage for the antibiotic susceptibility testing and direct diagnosis of TB from sputum. Luciferase Reporter Phages (LRP) are recombinant phages that contain the gene for the fire‐fly luciferase enzyme. Upon infection of mycobacteria by these phages, this enzyme is produced within a cell. In the presence of this enzyme, light is produced by viable cells that also contain the substrates for the enzyme: ATP and luciferin (added exogenously). The ‘Bronx Box’ has been developed as a low cost means of detecting light signal from mycobacteria infected with LRP using photographic film. A MTB strain's susceptibility to an antibiotic can be assessed by comparing light production, as an indication of the presence of viable cells, in samples pre‐incubated in the presence and absence of the test antibiotic. Susceptibility testing of some of the front line drugs for TB chemotherapy has given excellent results (detection of all resistant isolates), using LRP and the Bronx Box. Phage amplification uses the production of progeny phage, instead of a reporter gene product, to reflect the presence of target bacteria. The technique has been successfully applied to drug susceptibility testing and for diagnosis to the TB from sputum specimens. Viable MTB cells within a specimen are uniquely infected by a bacteriophage reagent. A potent selective virucide destroys all the phage particles that remain outside the cells while remaining inactive against the MTB bacilli (and the phages that have infected them). After the virucide has been neutralised, and phages that are released from infected cells are indicative of TB cells being present in the original specimen. These phages can be easily detected by propagation on lawns of non‐pathogenic rapidly growing mycobacteria. A commercially available test, termed FASTPlaqueTB‐RIF, determines the rifampicin susceptibility of isolates by comparing the viability of cells pre‐treated with and without rifampicin. Results give excellent agreement (97–98% overall) with conventional ‘gold standard’ methods, but in 2 days rather than 3 weeks. Preliminary work has shown that it should be possible to reduce the assay time to 24 h. A variation to this method, FASTPlaqueTB, detects TB directly from sputum specimens and is capable of rapidly reporting the presence of viable MTB cells within 48 h. This technique has achieved excellent specificity (close to 100%) when evaluated in the clinical setting.© 2001 Society of Chemical Industry