Laboratory experience and guidelines for avoiding false positive polymerase chain reaction results.

Despite the widespread use of polymerase chain reaction (PCR) for diagnosis of infectious diseases, the technology has not been generally introduced into routine diagnostic laboratories. One of the most serious problems which has influenced the acceptance of this technology is the occurrence of false positive PCR results. This study describes the experience, in a hospital laboratory setting, of using PCR for the diagnosis of heat-labile enterotoxin-producing E. coli, M. tuberculosis, M. paratuberculosis and human papillomavirus. Results indicate that a build-up of amplicons, generated during the amplification process in the laboratory, is the main source of PCR-contamination. Protocols are described that include both physical and chemical procedures to prevent contamination. The use of photo-induced psoralen is recommended for those laboratories already involved in PCR work where amplicons are likely to be present. An enzymatic system (uracil-N-glycosylase) was evaluated and is recommended for workers intending to start diagnostic PCR. Attention was given to simple control measures which are easily implemented in a routine diagnostic laboratory. Protocols such as these are likely to have a major impact on the introduction of PCR-based methods into routine laboratories.