Five hundred twenty processed respiratory specimens from 326 patients received for the diagnosis of tuberculosis or other mycobacterial infections were tested by means of the LCx Mycobacterium tuberculosis Assay from Abbott Laboratories, which uses ligase chain reaction technology for the direct detection of M. tuberculosis complex in respiratory specimens. The results of the LCx M. tuberculosis Assay were compared with the results of culture and staining techniques. After a combination of culture results and the patient's clinical data, a total of 195 specimens were collected from 110 patients who were positively diagnosed as having pulmonary tuberculosis. Twenty-three of these 195 specimens which corresponded to 10 patients with a history of pulmonary tuberculosis (TB) and anti-TB treatment ranging from 1 to 6 months were culture negative. The other 172 specimens were culture positive for M. tuberculosis. With an overall positivity rate of 37.5% (195 of 520 specimens), the sensitivity, specificity, and positive and negative predictive values were 90.8, 100, 100, and 94.7%, respectively, for the LCx M. tuberculosis Assay; 88.2, 100, 100, and 93.4%, respectively, for culture; and 82.6, 92, 72.9, and 97.6%, respectively, for acid-fast staining. For 161 specimens (82.6%) from patients smear positive for the disease and 34 specimens (17.4%) from patients smear negative for the disease, the sensitivity values for the LCx M. tuberculosis Assay were 98.8 and 53%, respectively. There were no statistically significant differences in the sensitivities and specificities between the LCx M. tuberculosis Assay and culture (P > 0.05). Conclusively, the LCx M. tuberculosis Assay has proved to have an acceptable sensitivity and a high specificity in detecting M. tuberculosis and has the potential of reducing the diagnosis time to an 8-h working day.