Setting: The first national anti-tuberculosis drug resistance survey in Pakistan, a high tuberculosis (TB) and low human immunodeficiency virus (HIV) burden country.
Objective: To determine the proportion of patients with multidrug-resistant TB (MDR-TB) and to compare the performance of Xpert(®) MTB/RIF with conventional phenotypic drug susceptibility testing (DST).
Methods: Sputum samples were collected from 1972 consecutively enrolled pulmonary TB patients from 40 clusters. Phenotypic DST was performed in parallel with Xpert.
Results: The proportion of MDR-TB patients was 3.7% (95%CI 2.5-5.0) among new and 18.1% (95%CI 13.0-23.4) among previously treated cases. A valid rifampicin (RMP) testing result was available from substantially more cases with Xpert (n = 1809) than with phenotypic DST (n = 1592). Among strains with discordant results, rpoB sequencing revealed only one false-resistant result (new TB case) with Xpert and 7.7% (8/104) of RMP-resistant cases missed with Xpert against 3.8% (4/14) by phenotypic DST. This difference was not significant.
Conclusions: This survey provides the first representative data for Pakistan on its MDR-TB burden. The Xpert assay had nearly 100% specificity, even in a low MDR-TB prevalence setting. The use of this assay greatly simplifies survey logistics, making it a feasible option for survey implementation, especially in resource-constrained settings.