Abstract

BACKGROUND: The incidence of mycobacterial infections has increased during the past five years. A prompt diagnosis is indispensable for initiating appropriate treatment. Because culturing of mycobacteria takes three to six weeks and sensitivity of microscopic detection of acid fast bacilli is low, amplification methods provide promising possibilities. Recently, the polymerase chain reaction (PCR) has been shown to be useful for confirming a mycobacterial infection, especially in cases with unexpected histological findings or lack of suitable material for culturing. AIMS: To evaluate the impact of PCR based techniques in the detection of mycobacterial infections in uncultured routine histological specimens as an alternative to surgical pathology. METHODS: Two hundred and twenty nine formalin fixed and paraffin wax embedded samples from 141 patients with clinical or histological suspicion of a mycobacterial infection were investigated using three different PCR assays and Southern blotting. PCR results were compared with histology and culture and the patients' clinical findings. RESULTS: When using culture as the reference method, the sensitivity for the detection of mycobacteria of the tuberculosis complex was 90%, specificity was 92%, the positive predictive value was 81%, and the negative predictive value was 96%. The sensitivity for the detection of nontuberculous mycobacteria was 100% and specificity was 78%, the positive predictive value was 26%, and the negative predictive value was 100%. The patients' clinical findings supported the PCR positive results, indicating a mycobacterial infection in 11 of 18 initially culture negative cases and in 21 of 35 PCR positive cases without culture results. CONCLUSIONS: These results indicate that PCR based techniques are sensitive, specific, and rapid methods for the detection of mycobacteria in routinely processed paraffin wax embedded and formalin fixed histological samples.