Page 198 - SAHCS HIVMed Journal Vol 20 No 1 2019

Page 2 of 10 Original Research

Xpert has not been associated with a mortality reduction in mycobacteriology (TB culture and/or Xpert). Information
most trials to date, 8,9,10 its implementation has been associated about any additional specimens, that were clinically indicated
with overall shorter times to starting anti-TB therapy, and collected by the medical teams were also recorded – for
including DR TB. 8,11,12,13 It has also increased the diagnostic example, lymph node aspirates, cerebrospinal fluid TB
yield by 1.4% – 15% (compared to sputum microscopy) in cultures, pleural TB cultures and urine TB cultures.
clinical trials in Sub-Saharan Africa, Brazil and Indonesia.
In the first study, two cases of false urine rifampicin
8,9,10,11,13,14,15,16
resistance occurred 3 months after study initiation
Against the backdrop of improved case detection, previous (Appendix Table 1-A1 – patients JTBS097 & JTBS099). Both
studies have reported on false rifampicin resistance results patients’ urine Xpert samples were collected after a sample
associated with the Xpert MTB/RIF assay, and meta-analyses was taken from an MDR patient earlier on the same day.
found the overall specificity of the Xpert for rifampicin- It was determined that both samples were likely
resistance in sputum samples to be 98% (i.e. 2% showed false contaminated due to inadequate cleaning of the reusable
rifampicin resistant results) and 99% in extra-pulmonary bedpan, although laboratory cross-contamination could
samples. 17,18 This however appeared to be associated in part not be ruled out. We subsequently introduced single-use
19
with earlier Xpert cartridge generations. An implementation disposable bed pans (Litha Healthcare Group,
study from South Africa found the Xpert G4 cartridge to have Johannesburg, South Africa) and these were used for the
excellent positive predictive value for rifampicin resistance remainder of the Jooste Hospital TB study and the duration
of 99.5% (95% CI 98.5–100) in sputum samples. 20 of the Khayelitsha Hospital TB study. There were no repeat
episodes of suspected cross-contamination. Urine was
We have previously found that among HIV-patients transferred to a polypropylene tube using a sterile syringe.
requiring acute medical hospitalisation, testing of a single Patients from both cohort studies had urine Xpert testing
concentrated urine sample detected 2.2 times more TB performed. Demographic details and clinical symptoms
cases than sputum Xpert testing, largely because of the were recorded for all patients at study entry. Patients were
inability of sick inpatients to produce a sputum sample. managed by the hospital and clinic staff, and all TB
21
Additionally, a recent randomised, multi-country trial diagnostic test results were made available by study staff
found that the addition of rapid urine-based assays and could be utilised to inform patient care.
(including urine Xpert) to sputum Xpert testing was
associated with reduced mortality among hospitalised HIV-
infected patients in sub-group analyses. This suggests that Laboratory methods
22
urine-based testing using Xpert may have an important role Urine Xpert testing for both studies was performed at the
in the TB diagnostic algorithm among hospitalised patients Groote Schuur Hospital National Health Laboratory Service
with advanced HIV, especially those too ill to produce a laboratory using Xpert MTB/RIF Assay G4 version 5. All
sputum sample. However, the proportion of false-positive specimens were processed using standardised protocols and
rifampicin resistance results using Xpert on urine samples quality assurance procedures as previously described. In
24
has not been reported. We sought to determine the brief, for the GF Jooste Hospital study, Xpert testing of urine
proportion of urine Xpert false rifampicin resistance results samples was conducted in two ways on each sample. The
among hospitalised HIV-infected patients being investigated first method (unconcentrated) utilised 2.0 mL of fresh urine
for HIV-associated TB in Cape Town, South Africa. that was centrifuged, resuspended in 0.75 mL phosphate
buffer and then tested using Xpert. The second method
25
Methods (concentrated) used a 30 mL – 40 mL urine sample that was
Patients and setting centrifuged at 3000 g for 15 min. The resultant supernatant
was removed and the pellet was resuspended in the residual
Patients from two parent cohort studies were included. In the urine volume (without the addition a phosphate buffer);
first, patients were recruited at GF Jooste Hospital, South 0.75 mL was then tested using Xpert. For both methods,
21
Africa from June 2012 to October 2013. Unselected HIV- Xpert sample reagent (1.5 mL) was added to the samples as
infected patients admitted to the medical wards were per manufacturer’s instructions. The Khayelitsha Hospital
recruited within 24 h of admission, regardless of TB treatment study only used Xpert testing on concentrated urine samples
at the time of admission. 21,23 GF Jooste Hospital was closed at and was undertaken using the same methods as described
the end of 2013 and two new hospitals (including Khayelitsha above. The reference standard for drug resistance, including
Hospital) were opened serving the same communities at the rifampicin resistance for both studies, was a molecular line
time the second study was conducted. The second study was probe assay (MTBDR plus; Hain Lifescience Nehren,
undertaken at Khayelitsha Hospital from January 2014 until Germany) undertaken on culture isolates from any clinical
October 2016 and recruited HIV-infected patients with a low specimen (not necessarily urine).
CD4 T-cell count (< 350 cells/µL) admitted to hospital with
a suspected new diagnosis of TB. Patients already on TB Analysis
treatment were excluded from this study. Sputum, blood
and urine samples were systematically obtained (when Patient populations were from overlapping referral areas
possible) as part of both study protocols and submitted for in the Cape Town townships and both cohorts included

http://www.sajhivmed.org.za 191 Open Access