ABSTRACT
There is a need for rapid and sensitive detection of Mycobacterium tuberculosis in tissue specimens. A polymerase chain reaction (PCR)-based assay for the diagnosis of tuberculosis was evaluated in 60 formalin-fixed tissue specimens, the target for the amplification being a segment of IS6110 in the M. tuberculosis chromosome. Of the 60 formalin-fixed, paraffin-embedded tissue specimens studied, 57 showed granulomatous inflammation and 53 had been cultured for mycobacteria; 10 were positive for M. tuberculosis and three were positive for other mycobacteria. Of 60 samples, 15 showed acid-fast bacilli on special staining. When done comparatively on a positive culture for M. tuberculosis, PCR for M. tuberculosis DNA in 60 tissue samples was 100% sensitive and 93% specific, having a positive predictive value of 76.9% and negative predictive value of 100%. PCR for M. tuberculosis DNA done on tissue samples was positive for 14 of 19 patients who had a clinical diagnosis of tuberculosis, negative for all six patients with nontuberculous mycobacterial infections, and negative for all 33 patients who had a diagnosis of a disease other than mycobacterial infection. When compared with the clinical diagnosis of tuberculosis, PCR for M. tuberculosis DNA in these patients' tissues was 73.6% sensitive and 100% specific, having a positive predictive value of 100% and negative predictive value of 88.6%. These data indicate that PCR amplification is useful for detecting M. tuberculosis DNA in formalin-fixed tissue specimens, and that it can be used to increase diagnostic accuracy in patients who have perplexing diagnostic problems associated with a granulomatous tissue response.
Subject(s)
DNA, Bacterial/analysis , Histocytological Preparation Techniques , Mycobacterium tuberculosis/isolation & purification , Polymerase Chain Reaction , Bacteriological Techniques , Chromosomes, Bacterial/genetics , Coloring Agents , DNA, Bacterial/genetics , Fixatives , Formaldehyde , Gene Amplification , Granuloma/microbiology , Granuloma/pathology , Humans , Mycobacterium/classification , Mycobacterium Infections/diagnosis , Mycobacterium Infections/microbiology , Mycobacterium avium/isolation & purification , Mycobacterium tuberculosis/genetics , Paraffin Embedding , Predictive Value of Tests , Sensitivity and Specificity , Tuberculosis, Pulmonary/diagnosis , Tuberculosis, Pulmonary/microbiology , Tuberculosis, Pulmonary/pathologyABSTRACT
Use of the polymerase chain reaction (PCR) to detect Mycobacterium tuberculosis in formalin-fixed, paraffin-embedded tissue would be of great diagnostic value. However, formaldehyde has been reported to decrease the efficiency of amplification by structurally altering the polynucleotide chain. We sought to determine the ability of the PCR assay to detect M. tuberculosis in formalin-fixed, paraffin-embedded tissue in a mouse experimentally infected with the H37Rv strain of M. tuberculosis. Lung tissue from the infected mouse was cultured to determine the number of organisms per gram of tissue. The remaining lung tissue was divided into eight portions, seven of which were fixed in 10% neutral buffered formalin for 24-h intervals over periods lasting from 1 to 7 d, and a control portion was placed in isotonic saline. The tissue samples were then paraffin-embedded, and sections were obtained from each tissue block for PCR analysis. We show that the PCR assay can detect as few as nine organisms in a 5-micron section of tissue, and that up to 7 d of fixation in 10% neutral buffered formalin has a negligible effect on the assay. The PCR assay can detect low numbers of M. tuberculosis organisms in formalin-fixed, paraffin-embedded tissue.
