A new tuberculosis (TB) test (Xpert MTB/RIF*), using two sputum specimens from each child, is substantially more sensitive and faster than conventional smear microscopy, detecting twice as many cases among children aged 15 and younger. The findings of the first study to report the accuracy of this new test in children, published Online First in The Lancet Infectious Diseases, suggest that Xpert MTB/RIF should become the first-choice diagnostic test for childhood TB.
Currently, diagnosing TB in resource-poor settings relies mainly on the use of smear microscopy of sputum samples. But, this 125 year-old method is unable to determine drug susceptibility, and has low sensitivity (particularly in people with HIV and children) which can lead to delayed diagnosis and treatment and increases the likelihood of transmitting the infection to others.
Previous studies have shown that the MTB/RIF test can rapidly detect TB and multi-drug resistant TB with very high sensitivity and specificity in adults. In 2010, WHO recommended that MTB/RIF be the initial diagnostic test in people with suspected drug-resistant TB and TB complicated by HIV infection, which are more difficult to diagnose. But until now, its accuracy has not been tested in children.
In this study, Mark Nicol, Heather Zar and colleagues from the University of Cape Town, Cape Town, South Africa enrolled 452 children with suspected TB from two hospitals in Cape Town, South Africa, an area with a high TB and HIV burden. Two induced sputum specimens were taken from each child and examined for Mycobacterium tuberculosis using standard liquid culture and the MTB/RIF test.
MTB/RIF tests done on both samples doubled the sensitivity of detecting TB compared with conventional smear microscopy, identifying 75.9% of confirmed cases compared with 37.9%—including all smear-positive cases but only 61.1% of smear-negative cases—with a specificity of 98.8%, and in a substantially faster time than culture (1 day vs 12 days).
Importantly, testing a second specimen increased the sensitivity of MTB/RIF for smear-negative TB by 27.8%, with the first MTB/RIF test detecting only about a third of smear-negative cases. By comparison, a second culture test increased the sensitivity by just 13.8%.
MTB/RIF detected all cases of confirmed TB in children with HIV, although future studies are needed to corroborate whether sensitivity is increased in these patients.
The authors say: "We showed that the sensitivity of MTB/RIF for smear-negative tuberculosis is lower in children compared to adults, but is twice as sensitive compared with smear microscopy. We also showed that the incremental benefit in testing a second induced sputum specimen is substantial, suggesting that, in children, a second specimen should be tested to optimise sensitivity."
They conclude: "To maximise the effect of this technology and benefit child health, increased capacity for sputum induction in children at health-care facilities is needed."
In a Comment, Eduardo Gotuzzo from Universidad Peruana Cayetano Heredia, Lima, Peru says: "Overall sensitivity was higher with a second test…but this approach would likely increase diagnosis costs…Further studies should be done in settings with different HIV and tuberculosis prevalence, and implementation studies and cost-effectiveness studies are needed to support the use of a second test in paediatric populations."
Professor Mark Nicol, University of Cape Town, Cape Town, South Africa. T) +27 21 406 6083/6727 E) mark.nicol@uct.ac.za
Dr Eduardo Gotuzzo, Universidad Peruana Cayetano Heredia, Lima, Peru. T) +51 1 382 1021 E) eduardo.gotuzzo@upch.pe
Notes to Editors: *Xpert MTB/RIF is a new molecular test designed to identify DNA of Mycobacterium tuberculosis (MTB) and resistance to rifampicin (RIF). Xpert MTB/RIF is simple to perform with little training, and is fully automated so needs minimum biosafety facilities.
Journal
The Lancet Infectious Diseases