Rapid urine-based screening for TB to reduce AIDS-related mortality in hospitalised patients in Africa
Post-mortem studies show that TB is the cause of between one-third and two-thirds of adult HIV/AIDS-related deaths recorded in health facilities in sub-Saharan Africa. However, around one half of these TB cases are undiagnosed at the time of death, highlighting the urgent need for new diagnostic approaches. Background studies conducted in Cape Town by Stephen Lawn show that a large majority of TB cases can be rapidly diagnosed from a urine sample within the first 24 hours of acute hospital admission, using a combination of two techniques: Determine TBLAM lateral-flow urine test and Xpert MTB/RIF testing of concentrated urine.
The three-year Screening for Tuberculosis to Reduce AIDS-Related Mortality in Hospitalized Patients in Africa (STAMP) study, started in 2015, is an individually randomised controlled trial that will assess the clinical outcomes of standard sputum-based testing with Xpert MTB/RIF plus additional urine-based screening, compared with the standard screening alone.
Funded by a UK Medical Research Council, Department for International Development and Wellcome Trust Global Clinical Trials Scheme grant award of £2.1 million over three years, the trial is underway in both Malawi and KwaZulu Natal, South Africa.
The XTEND study: evaluating Xpert MTB/RIF as a first line TB test in South Africa
XTEND was a trial evaluating the impact of Xpert MTB/RIF in the context of national roll-out of this new diagnostic test in South Africa. It was a cluster-randomised trial, where the 20 clusters each comprised a laboratory and two associated (but not co-located) clinics. Intervention clusters had already implemented Xpert MTB/RIF as their first line TB diagnostic test; control clinics were still using smear microscopy. We enrolled people giving a sputum sample for TB investigation in the 40 study clinics; the primary outcome was mortality at 6 months after enrolment.
The XTEND study found that mortality at 6 months was not reduced in the Xpert MTB/RIF arm compared to the smear microscopy arm (3.9% vs. 5.0%; adjusted risk ratio 1.1). The number of people starting TB treatment by 6 months (11.6%) was also not changed. However the proportion of people with a confirmed TB diagnosis was higher in the Xpert arm (9.2% vs. 7.8%; adjusted prevalence ratio 1.49). There was no difference in the proportion of study participants with a positive TB test result who did not start TB treatment (17.0% vs. 14.9%). The risk of death was higher among those who were HIV-positive and not on ART at enrolment, or did not know their ART status.
These results imply that implementation of a new diagnostic test with higher sensitivity may not, in isolation, improve patient outcomes for drug-sensitive TB. Improved outcomes may need better tests, but also better linkage to TB and HIV care.
Xpert MTB/RIF versus sputum microscopy as the initial diagnostic test for tuberculosis: a cluster-randomised trial embedded in South African roll-out of Xpert MTB/RIF. Churchyard GJ, Stevens WS, Mametja LD, McCarthy KM, Chihota V, Nicol MP,Erasmus LK, Ndjeka NO, Mvusi L, Vassall A, Sinanovic E, Cox HS, Dye C, Grant AD, Fielding KL. Lancet Glob Health. 2015 Aug;3(8):e450-7