Performing a molecular test for tuberculosis during a patient visit to a rural clinic in South Africa, rather than performing the test at a centralised laboratory, greatly reduced the time to treatment for patients who did not have a drug-resistant form of the disease.  This is according to the findings of a randomised, controlled trial published online in the American Thoracic Society’s American Journal of Respiratory and Critical Care Medicine.

In ‘Impact of Point-of-Care Xpert MTB/RIF on Tuberculosis Treatment Initiation: A Cluster Randomized Trial’, researchers from the London School of Hygiene & Tropical Medicine and Africa Health Research Institute (AHRI) report on a trial conducted at a rural clinic in northern KwaZulu-Natal, which has among the highest rates of HIV and HIV-associated TB in the world. Unlike most studies of diagnostics, which investigate how accurate a test is, this trial explored whether a different way of delivering the test would lead to better patient outcomes.

The researchers found that by providing a TB diagnosis within hours of the patient visit (typically referred to as point-of-care), three out of four patients who did not have drug-resistant TB began treatment the same day. This compared to seven days, on average, to treatment for those patients whose diagnostic test was performed at a centralised laboratory. Initiation of appropriate treatment within 30 days of diagnosis was better with point-of-care than laboratory diagnosis, but the difference was not statistically significant.

Point-of-care diagnosis did not reduce time to treatment for drug-resistant TB.  Researchers said this fact likely reflects the absence of a local treatment programme for drug-resistant TB during the time the study was conducted. They noted that since the study ended, a local treatment programme has been established and time to treatment has been shortened.

“Research has shown that in South Africa and other countries with a high TB burden, up to one in four people with a laboratory diagnosis of TB do not start TB treatment,” said Dr Richard Lessells, a former clinical research fellow at AHRI, who led the study.

“Theoretically, we now have the technology to diagnose TB and initiate treatment during a single visit – something that happens routinely with HIV and malaria – but we wanted to test this technology in the real world to see if it could actually deliver this.”

Dr Lessells and colleagues compared a testing system where Xpert MTB/RIF was performed by a nurse at the clinic to transporting sputum samples to a central laboratory, where Xpert MTB/RIF was performed by a laboratory technician. The test, which has been in use for less than a decade, is highly accurate in detecting the presence of TB bacteria in about two hours’ time. It also can determine resistance to rifampicin, a key first-line TB drug.

(Above: A nurse tests for TB using Xpert MTB/RIF at a clinic in uMkhanyakude district, northern KwaZulu-Natal)

Adults reporting cough who were HIV positive and/or at high risk of drug-resistant TB were randomized to either point-of-care (651 patients) or laboratory testing (646 patients) based on which two-week period between August 2011 and March 2013 they were seen in the clinic.

“To fight TB in South Africa and elsewhere, we need to decentralize both diagnosis and treatments in a coordinated manner,” Lessells added. “Nurses are at the frontline of TB and HIV care in South Africa, and we need to give them the tools to make the correct diagnosis and initiate the correct TB treatment in a single clinical encounter.”

Although the Xpert MTB/RIF trial demonstrated important benefits of this relatively new diagnostic technology, Dr. Lessells said that it still has shortcomings.

“The test fails to detect all cases of TB, takes too long to get a result and is too expensive for it to be deployed in many countries,” he said. “Diagnostics for TB are evolving but we really need transformative tools if we are serious about ending TB.

“The holy grail of TB remains to be developed:  a simple, rapid test that does not rely on sputum, that can be performed in a few minutes by a nurse or community health worker and that detects all forms of TB.”

(Top photograph: A nurse loads a Xpert MTB/RIF cartridge with a sputum sample) 

The Wellcome Trust funded this study.