Background: The inhibition rates for nucleic acid tests of Mycobacterium tuberculosis have been reported to range from less than 1% to more than 10%. Specimen dilution, boiling, addition of bovine serum albumin (BSA), and a silica membrane can be used to override amplification inhibitors in nucleic acid tests of M. tuberculosis. The inhibition rate for real-time PCR of M. tuberculosis (COBAS TaqMan MTB test; Roche Diagnostics, Manheim, Germany) and effective strategies to override PCR inhibitors were investigated in this study. 

Methods: The inhibition rate for COBAS TaqMan MTB test was investigated in 980 clinical specimens. The effectiveness of PCR inhibitor removal by repeated run, dilution, boiling, addition of BSA, and use of silica membrane were evaluated in the inhibited specimens. 

Results: Inhibitory substances were present in 4.1% of specimens (40/980). Among 40 inhibited specimens, inhibitory substances were removed in 12 (30%), 30 (75%), 27 (67.5%), 25 (62.5%) and 12 (30%) specimens with repeated run, dilution, addition of RBS, boiling and use of silica membrane, respectively. 

Conclusion: The overall inhibition rate for the COBAS TaqMan MTB test was 4.1%. Dilution, boiling and addition of BSA were shown to be more effective than repeated run and use of silica membrane for removal of PCR inhibitors. A combination of two methods might be useful and should be studied in the future. (Korean J Clin Microbiol 2011;14:97-102)