Pooling sputum samples to improve the feasibility of Xpert® MTB/RIF in systematic screening for tuberculosis

Abstract

Setting: Systematic screening for tuberculosis (TB) using Xpert® MTB/RIF.

Objective: To determine whether pooling sputum samples for Xpert testing may improve the feasibility and cost-effectiveness of Xpert by reducing the number of Xpert tests required.

Design: Mycobacterium tuberculosis-spiked sputum samples at low organism concentrations were used to mimic samples that are more likely to be found in the screening, compared to the diagnostic, setting. Using Xpert, pooled sputum samples were tested from a pooling ratio of 1 in 2 to 1 in 12.

Results: A linear relationship between the pooling ratio and the Xpert MTB cycle threshold (Ct) value was found. As the sputum pooling ratio increased, the Ct value also increased. However, the slope of this increase was relatively small. In the majority of the samples pooled (75/96, 78.1%), Xpert was able to detect M. tuberculosis.

Conclusion: These findings suggest that sputum pooling may be a viable method of improving the feasibility and cost-effectiveness of large-scale sputum testing using Xpert in the TB screening setting.