Xpert MTB/RIF assay is a novel automated molecular tool for the diagnosis of TB. Xpert can detect TB genetic material in sputum samples as well as test for genetic resistance to rifampicin providing results within 2 hours. Xpert received WHO endorsement in December 2010. There is limited data on the impact of Xpert on time-to-treatment and TB-related patient morbidity in primary care clinics. No studies have yet evaluated Xpert performed at the point-of-treatment (POT) i.e. in primary care clinic location. The investigators hypothesize that one sputum GeneXpert MTB/RIF assay performed at the POT will improve time-to-diagnosis, time-to-treatment and TB related patient morbidity for patients with suspected TB presenting to primary level TB clinics in high HIV prevalent settings.

Tuberculosis (TB) is one of the world's most important infectious causes of mortality and continues to kill 1.8 million people annually. Despite intensified standard measure of TB control, TB case detection rates are low, posing major hurdles for TB control. It is estimated that approximately 50% of patients with TB are still not diagnosed and treated appropriately. The problem is compounded by the increasing prevalence of multi drug resistant (MDR) and extensively drug resistant (XDR) TB and the close association between TB and HIV infection.

Diagnostic tools introduced 100 years ago are still in routine use and increasingly inaccurate if the face of the HIV and TB syndemics. Consequently, many patients with active TB remain undiagnosed and continue to spread the disease within the community. Thus, missed or delayed diagnosis results in ongoing transmission, patient morbidity and mortality, and social and economic consequences. Currently, there is no available point-of-care, or even point-of-treatment test that allows early detection of active tuberculosis at the peripheral health clinic level. Lack of rapid, simple and accurate diagnostic tests at this level is a major hurdle in controlling the global burden of TB. A number of promising new TB diagnostics have shown initial promise but there remains an urgent need to assess their impact when used at the point-of-treatment in primary care level.

In 2009, Cepheid released the Xpert® MTB/RIF Assay, which is the only system able to deliver answers directly from unprocessed samples by combining on-board preparation of the sample with real-time polymerase chain reaction (PCR) in less than 2 hours. Additionally, the Xpert® MTB/RIF Assay allows for simultaneous on-demand molecular testing for the detection of mycobacterium tuberculosis (M.tb) and rifampicin (frontline anti-TB drug) resistance. . The GeneXpert™ system consists of a GeneXpert instrument, personal computer and disposable fluidic cartridges. The system combines cartridge-based sample preparation with amplification and detection in a fully integrated and automated nucleic acid analysis instrument. Xpert has now been shown to be an accurate tool for the rapid diagnosis of tuberculosis in both smear-positive and smear-negative samples in both a multicentre evaluation and demonstration study with a sensitivity of approximately 70% in smear negative culture positive TB. Xpert testing in both these studies was performed at microscopy laboratories. In December 2010, on the basis of these results, Xpert was endorsed for TB diagnosis by the World Health Organisation (WHO) but is yet to be integrated into national tuberculosis control programmes.

Limited data is available on the impacts of Xpert on patient important outcomes such as TB-related morbidity. No data is available about the feasibility and robustness of performing Xpert in primary care clinics at the POT using minimally trained nursing staff. The objective of this study will be to examine the feasibility and impact of a single point-of-treatment Gene Xpert MTB/RIF Assay performed by clinic staff compared to standard microscopy-centre based diagnostics. Special focus will be on the patient-related outcomes of time-to-treatment initiation, drop out rates and the mean difference in TB-morbidity scores in patients diagnosed with Xpert.

The Xpert POT study will be a multicentre patient-level randomised controlled trial comparing a single sputum GeneXpert MTB/RIF Assay performed at point-of-treatment with same-day standard fluorescent smear microscopy for TB diagnosis at the primary level of care. A single liquid MGIT culture performed a regional laboratory will be used as the reference standard.

Inclusion Criteria:

1. Able and willing to give informed consent

2. Ambulant patient presenting to TB clinic

   IF HIV negative requires 2 or more of the following:

   • Cough ≥ 2 weeks
   • loss of weight
   • persistent fever ≥ 2 weeks and/or
   • a single recorded temp > 38°C
   • night sweats
   • generalized fatigue
   • hemoptysis or
   • chest pain

   OR if HIV positive - any one of the following:

   • current cough
   • night sweats
   • fever
   • loss of weight

3. Patient 18 years or above

Exclusion Criteria:

1. Inability to provide informed consent (e.g. mentally impaired)
2. Unable to produce 2 sputa of ≥ 1ml
3. TB treatment within the last 60 days
4. Unable to potentially return for study follow-up at 2 and 6 months (i.e. leaving community)
5. Patient not meeting inclusion criteria