Page 201 - SAHCS HIVMed Journal Vol 20 No 1 2019

Page 5 of 10 Original Research

definitive test for drug-resistant TB identification. This, Different methods of drug susceptibility testing could explain
however, needs to be evaluated in additional settings. discrepant results in some cases. 33,34 The majority of drug
susceptibility testing on cultured isolates in both studies was
Xpert Ultra is an updated, next-generation sample cartridge PCR-based; however, we also captured results of all TB tests
for the Xpert platform that is now recommended by the performed in-service and cannot reliably differentiate
WHO as a replacement for the current Xpert MTB/RIF between drug susceptibility testing performed with other
cartridge 27,28 and has been implemented in South Africa. It methods such as liquid or solid media for all samples for the
provides increased sensitivity for the detection of MTB in duration of the study. Because of the early implementation of
sputum (especially smear-negative and pauci-baciliary disposable bedpans, we do not suspect undetected
disease). Xpert Ultra utilises a new melt curve analysis to contamination beyond that described above. Furthermore,
detect RIF-resistance; however, its diagnostic accuracy because most patients with positive urine rifampicin results
(including specificity) for the detection of rifampicin did not have paired urine culture isolates available for further
resistance is similar to that of Xpert. The results of this study genotypic or phenotypic drug-susceptibility testing, patients
29
suggest that urine Xpert Ultra rifampicin resistance results classified as having false positive rifampicin results may
should be interpreted cautiously and confirmed by alternative have had heteroresistance with compartmentalised true
drug susceptibility testing (either phenotypic or alternative rifampicin-resistant urinary TB and rifampicin-susceptible
genotypic assays) until the specificity of Xpert Ultra for TB at other anatomic sites. However, the favourable clinical
rifampicin resistance detection has been confirmed to be course of most of these patients on first-line drug-sensitive
adequately high to warrant stand-alone testing on urine TB treatment counts against this possibility. Notably, a large
samples. proportion of false positive rifampicin results were among
those already receiving anti-TB therapy, where 50% of urine
Of interest, in this cohort we describe three patients with a Xpert rifampicin resistance results were classified as false
confirmed urine Xpert rifampicin resistance result who resistance; this suggests that further caution should be
also had drug-sensitive strains from independent samples applied when interpreting urine Xpert rifampicin resistance
during the same admission suggesting likely results in treatment-experienced patients.
heteroresistance (either polyclonal infection or acquired
heteroresistance). The prevalence of heteroresistance in An additional limitation of the study is that sequencing of
MTB infections has previously been described. 30,31,32 isolates was not performed as part of either study. Sequencing
Although not well-studied, these are likely associated of the rpoB gene would have been particularly useful in the
31
with increased rates of treatment failure for the individual cases that we could not classify as true or false resistance and
and could complicate TB control efforts at a population the heteroresistant cases. Furthermore, urine TB cultures
level. Xpert may miss heteroresistance if used as a stand- were not routinely performed in either study and it may have
alone test for the detection of rifampicin resistance, been useful to compare drug susceptibility results on isolates
however, early studies show that Xpert Ultra may detect cultured from urine samples collected at the same time as the
heteroresistance when the resistant DNA comprises 5% or urine Xpert samples.
more of the sample. 28
In conclusion, urine testing using Xpert provides important
Strengths of this study include a large number of urine Xpert diagnostic yield for hospitalised HIV-infected patients being
rifampicin results from two geographically and clinically investigated for HIV-associated TB, especially in those unable
comparable cohorts where patients were prospectively to produce sputum samples. Although the overall proportion
recruited and underwent systematic testing for TB. of patients with urine Xpert rifampicin resistance in this
Additionally, all TB assays including urine Xpert testing were cohort was relatively low, the proportion of those classified
performed at the same laboratory according to standard as false rifampicin resistance was substantially higher than
protocols. After an error yielded two likely false Xpert has previously been reported on sputum. Urine Xpert
rifampicin resistant urine cases due to contamination soon rifampicin resistant results should therefore be interpreted
after recruitment initiation, disposable bedpans (single-use) with caution, repeated on a second sample in patients at low-
were implemented for the duration of both studies. We risk for drug resistant TB (as currently recommended by the
therefore recommend that clinicians use single-use specimen WHO for sputum samples) and confirmed using additional
collection bedpans and containers when utilising Xpert culture-based or molecular assays when possible. Whether
or Xpert Ultra testing on urine to prevent DNA-cross- these findings apply to Xpert Ultra is an issue that requires
contamination between samples.
further study.
The reason(s) for the high proportion of false positive urine Acknowledgements
rifampicin resistance is not entirely clear, but the proportion
was higher among those already receiving TB therapy. A The late Stephen D. Lawn was PI on the Jooste Hospital
limitation of this study is that we did not have data available study. The investigators are grateful to the clinical and
to systematically evaluate the Xpert probe features associated administrative staff of the Western Cape Department of
with our classification of false rifampicin resistance. Health.

http://www.sajhivmed.org.za 194 Open Access