A Randomised Control Trial of Sputum Induction, and New and Emerging Technologies in a High HIV Prevalence Primary Care Setting

Tuberculosis is on the increase in Africa. A key area of weakness in TB control efforts is the inability to make a rapid diagnosis. This is, in part, due to the inability to obtain representative biological samples and the non-availability of cheap, effective, rapid and field-friendly diagnostic tools. Indeed, in HIV positive patients, where the sensitivity of sputum smear is as low as 20%, culture results take several weeks and a significant proportion of patients do not expectorate sputum. Sputum induction for the diagnosis of TB has been evaluated in several studies, has been shown to have a good yield, feasibility and safety when performed correctly, and with diagnostic comparability to bronchoscopy. Few studies have evaluated the performance of induced sputum outside of the hospital environment and the tolerability, yield and performance outcome of sputum induction in a primary care facility has yet to be evaluated. The objective of this study is two-fold. We seek to evaluate, through a randomized controlled trial, the feasibility, performance outcomes and impact on time-to-diagnosis and -treatment of sputum induction, in a high HIV prevalence primary care setting, for the diagnosis of smear negative/ sputum scarce TB. We hypothesize that acquiring an induced sputum sample for smear microscopy and liquid TB culture will decrease time-to-diagnosis and time-to-treatment initiation in smear negative/sputum scarce TB patients in a primary care clinic in a resource-limited high TB HIV prevalent setting. We will also evaluate the potential incremental benefit of novel technologies to improve the rapidity and diagnostic yield using induced sputum samples (Xpert MTB/RIF assay, microscopic observation drug susceptibility testing (MODS) and the Genotype MTB DRplus line probe assay).