Effective testing for pulmonary tuberculosis using Xpert MTB/RIF assay for stool specimens in immunocompetent Pakistani children.

OBJECTIVE/BACKGROUND: Childhood tuberculosis (TB) is largely a paucibacillary disease and difficult to diagnose. It is difficult to obtain a sputum or gastric aspirate (GA) sample, and patients are often undiagnosed and treated empirically. Stool is a noninvasive specimen not usually used for TB testing in Pakistan. We investigated the value of Xpert MTB/RIF to diagnose Mycobacterium tuberculosis (MTB) in children with pulmonary TB cases, by performing comparative testing of GA and stool samples.

METHOD

We recruited 60 children aged 1-15years, suspected of TB, from the Department of Pediatrics, Civil Hospital, Karachi, Pakistan and The Aga Khan University Hospital, Karachi, Pakistan. All were immunocompetent. Patients had a Kenneth Jones TB score of ⩾5. Paired GA/sputum and stool samples were collected for testing. All GA samples were tested by Xpert MTB/RIF assay and MTB culture, while stool was tested by Xpert MTB/RIF.

RESULTS

The study participants included 27 males and 23 females with a mean age of 6years and a mean TB (Kenneth Jones) score of 7. Stool was received in the laboratory within 1-2days of the GA sample for all but one participant, who expired. The rates of MTB detection were as follows: 22% (11 cases) based on Xpert MTB testing of GA, 21% (10 cases) based on MTB culture of GA, and 21% (10 cases) based on Xpert MTB testing of stool. No rifampicin resistance was detected. Overall, there was concordance between testing of GA and stool. One case had GA with low positive Xpert and positive MTB culture, but negative stool Xpert result. In another case, there was low positive GA Xpert, positive GA MTB culture, and positive stool Xpert. A positive Xpert MTB stool test was associated with a higher TB score (>5) and a greater bacillary load. All 11 cases of TB diagnosed were put on antituberculous therapy and responded well to treatment.

CONCLUSION

Use of Xpert MTB/RIF assay for stool-based diagnosis of pulmonary TB in immunocompetent children is useful in a resource poor setting. This is a valuable and noninvasive diagnostic alternative for the diagnosis of childhood TB and can be adapted by pediatric arms of national TB programs.