Abstract
INTRODUCTION: the signs and symptoms of tuberculous (TB) colitis were similar with other diseases, such as inflammatory bowel disease. Therefore, finding the diagnostic modality to help differentiate TB colitis with other diseases was a challenge. In this study we aimed to find the proportion of positive stool TB-PCR in suspected TB colitis subjects and also the diagnostic value of the stool TB-PCR if compared to colonoscopy, histopathology and clinical evaluation. METHODS: a cross-sectional study was done on subjects suspected to have TB colitis who undergone colonoscopy and histopathology examination between February-April 2019. Stool samples from those subjects were collected and extracted with the QIAamp® Fast Stool DNA Mini Kit. The TB-PCR was done using artus® M. tuberculosis RG PCR kit, which targeted on 16s rRNA gene. The results of stool TB-PCR then were compared with the combination of colonoscopy, histopathology and clinical evaluation as the gold standard. RESULTS: from sixty subjects who were recruited, there were 26/60 (43.3%) subjects with positive stool TB-PC. It was consisted of 7/8 TB colitis subjects and 19/52 non-TB colitis subjects. The diagnostic value of the stool TB-PCR was: sensitivity 87.5%, specificity 63.5%, positive predictive value 26.9% and negative predictive value 97.1%. CONCLUSION: stool TB-PCR has good sensitivity but low specificity for diagnosing TB colitis. Therefore, stool TB-PCR could be utilized as a screening test for TB colitis.