Assay of Mycobacterium Tuberculosis in Biopsy Specimens Taken From Hepatic Granuloma Patients Using Polymerase Chain Reaction Method

Abstract

Background/aims: In the diagnosis of hepatic tuberculosis, classical laboratory methods (including cultures and direct staining of the acid resistant bacilli) are insufficient. Diagnosis is commonly made by clinical and histological findings and then confirmed by response to antituberculous treatment. In the diagnosis of pulmonary and extra-pulmonary tuberculosis, the use of PCR to detect tuberculous bacilli is being evaluated and some authors consider it a very reliable and sensitive method. However, the use of PCR in the diagnosis of hepatic tuberculosis has not been sufficiently supported by clinical studies. The aim of this study was to assess the sensitivity of this method in demonstrating M. tuberculosis bacilli in liver tissue specimens of hepatic granuloma patients, which were fixed with formalin and imbedded in paraffin blocks. Methods: Thirty-two liver biopsy specimens taken from cases diagnosed with granulomatous hepatitis and fixed with formalin in paraffin blocks were included in the study. The specimens were divided into three groups according to clinical and laboratory findings, histopathological diagnosis of hepatic granuloma and the response to appropriate treatment: Group A (n=12): hepatic granuloma with caseification necrosis (liver tuberculosis), Group B (n=10): noncaseous hepatic granuloma (liver tuberculosis), Group C (n=10) nontuberculous hepatic granuloma patients. All biopsy materials were stained by the Ziehl-Neelsen method and with allocrom stain in a direct search for bacilli and the PCR study was then performed. DNA amplification of M. tuberculosis IS6110 gene was also done by PCR study Results: Direct stains were negative in all specimens. M. tuberculosis gene amplification was found to be positive with the first step PCR method in three (25%) of 12 patients in group A, while with nested PCR, four (33%) of 12 patients in group A, three (30%) of 10 patients in group B and one (10%) of 10 patients in group C were positive. Only seven (32%) of all 22 patients of groups A and B who were diagnosed with tuberculous granulomatous hepatitis had positive PCR tests. In the control group however, only one (10%) of the patients had a positive PCR test. Conclusion: The findings of this study show that in our patient groups, PCR is insufficiently sensitive in diagnosing TB. Possible explanations for this low sensitivity could be that DNA concentrations were low in our specimens that paraffin blocks used contained inhibitors or that the primers used were inappropriate.