Abstract
OBJECTIVE:the aim of this study is to evaluate the performance of stool microscopy and culture in the diagnosis of pulmonary tuberculosis (PTB).MATERIAL AND METHOD:In resource-limited settings, PTB is mainly diagnosed in sputum by microscopy methods. In patients unable to provide sputum, invasive procedures are necessary to obtain alternative respiratory tract specimens. Because Mycobacterium tuberculosis complex (MTC) organisms are known to survive in gastric fluid, we hypothesized that swallowed MTC organisms would be detectable in stool samples. We compared the presence of MTC organisms in sputum and stool specimens collected at the same time from the same patients.RESULTS:We included samples routinely submitted to our laboratory of microbiological diagnosis from patients with suspected PTB. In addition, a stool specimen was collected within 24 h of the sputum collection or gastric aspirate. In the 57 patients included, sputum microcopy and culture confirmed respectively 9 (21%) and 15 (26%) cases. Stool samples made it possible to confirm 9 cases by microscopy and 5 cases by culture. The sensitivity of microscopy was 60% and that of culture 33%. Sputum analysis was negative for one HIV-infected patient, but the stool sample was positive and permitted the diagnosis of PTB.CONCLUSION:This study proves that despite the low sensitivity of stool cultures it can be an alternative or additional interesting sample for the diagnosis of PTB in patients who have difficulty expectorating.