Xpert® MTB/RIF assay for extrapulmonary tuberculosis and rifampicin resistance

This is a protocol for a Cochrane Review (Diagnostic test accuracy). The objectives are as follows:

To determine the diagnostic accuracy of Xpert® MTB/RIF for the detection of extrapulmonary TB by site of disease in people suspected of having extrapulmonary TB. The role of Xpert® MTB/RIF would be a replacement for standard practice.

To determine the diagnostic accuracy of Xpert® MTB/RIF for detection of rifampicin resistance in people suspected of having extrapulmonary TB. The role of Xpert® MTB/RIF would be an initial test for replacement for culture-based DST.

We plan to investigate the impact of covariates on heterogeneity in estimates of test accuracy across the included studies.

Detection of extrapulmonary TB

The covariates of interest are: HIV prevalence; condition of the specimen (fresh versus frozen); sample input volume (for liquid specimens); homogenization step (for tissue specimens); smear status of specimen; past history of TB; and prevalence of extrapulmonary TB. In addition, we plan to explore whether the WHO standard operating procedures for a given type of specimen were followed and can explain the observed heterogeneity in accuracy estimates. For TB meningitis we plan to investigate whether the use of a concentration step prior to the use of Xpert® MTB/RIF has an impact on accuracy estimates. In addition, for detection of lymph node TB, pleural TB, and TB meningitis, we plan to adjust accuracy estimates by applying a latent class meta-analysis model to account for the imperfect nature of culture as the reference standard.

Detection of rifampicin resistance detection

The covariate of interest is the prevalence of rifampicin resistance.