Evaluation of a PCR assay for detection of Mycobacterium tuberculosis in clinical specimens.

A polymerase chain reaction (PCR) assay for detection of M. tuberculosis was optimized for application to clinical specimens, which were prepared for amplification by boiling in buffer. The buffer contained a synthetic DNA fragment to determine if DNA amplification from the individual prepared specimens was subject to inhibition because of substances present in the specimen, or by the process of specimen preparation or storage. The PCR test was less sensitive than direct microscopy (75% as against 87.5%) and had a specificity of 97%. Invalid results due to inhibition of amplification occurred in 12% of specimens. Incorporation of the internal standard into the specimen preparation buffer ensures that any step in the process which inhibits DNA amplification is detected in the failure of amplification of the internal standard. The use of internal standard in this way should be considered in developing diagnostic protocols.