A research report submitted to the Faculty of Health Sciences, University of the Witwatersrand, in partial fulfilment for the Degree of Master of Medicine In the branch of Anatomical Pathology Johannesburg 2015 / Background: Mycobacterium tuberculosis (Mtb) poses a major global health problem. According to the World Health Organization, South Africa is a country with one of the highest reported incidence rates of this disease. Key to overcoming this preventable and treatable disease lies in establishing a reliable and rapid diagnostic approach. Aims and Objectives: This study aims to investigate the validity of the mycobacterium polymerase chain reaction (PCR) assay applied to formalin-fixed, paraffin-embedded tissue in which the histology showed granulomatous inflammation with no demonstrable acid-fast bacilli. Methods: A retrospective, cross sectional and non-interventional study was conducted on 121 histopathology cases showing granulomatous inflammation with a negative Ziehl-Neelsen (ZN) stain. The mycobacterium PCR results obtained in these cases were compared against the results of mycobacterium culture obtained from a specimen derived from the same or related site as the biopsy. Results: The mean age of the study population was 35.3 years and the study cohort included 63 males and 58 females. The sensitivity of nested mycobacterium PCR (detecting the 133 base pair product of the heat shock protein 65 kilo Dalton gene), was 64.1% and the specificity was 68.2%. The positive and negative predictive values were 49% and 80% respectively. Twenty six of the 121 cases studied had a false positive result (21.5%). CONCLUSION: There are many factors that may influence the result of a PCR assay and the interpretation thereof. Some of these factors include the inability of the test to distinguish between live and dead bacilli, the high risk of carry over contamination, and the paucibacillary nature of certain samples with an unequal distribution of the few bacilli that may be present. Although the sensitivity and specificity of mycobacterium PCR on paucibacillary, formalin-fixed, paraffin embedded tissue is suboptimal, the interpretation of these results must be performed in conjunction with the overall clinical presentation of the patient. / MT2016