Xpert Carba-R to identify people carrying carbapenemase-producing organisms

Study component

Description

Objectives/hypotheses

To compare the performance of 3 commercial assays in detecting carbapenemases:

Study design

Comparative diagnostic accuracy study.

The carbapenemase genes had been previously detected using in‑house PCR assays or a commercial microarray (Check‑MDR CT102). These data were regarded as the gold standard against which the commercial assays were compared. A modified Xpert Carba‑R version 2 kit was subsequently used to test any isolates with an OXA‑48 variant carbapenemase gene that were not detected by the original Xpert Carba‑R assay.

Setting

Non‑consecutive isolates submitted to Public Health England's AMRHAI Reference Unit from laboratories across the UK between July 2009 and April 2014.

Inclusion/exclusion criteria

No inclusion/exclusion criteria were reported.

Isolates were not consecutive referrals, but were selected for geographical, temporal (within the above timeframes) and carbapenemase diversity.

Primary outcomes

Performance characteristics – sensitivity.

Statistical methods

Not reported.

Results

The overall test sensitivity was 94.3% (402/426).

The Xpert Carba‑R test successfully detected the correct carbapenemase gene in all 302 isolates with a KPC, NDM or VIM variant gene.

A modified Xpert Carba‑R version 2 assay correctly identified the OXA‑181 variants not detected by Xpert Carba‑R.

Conclusions

The authors concluded that the commercial assays offer a reliable means of detecting bacteria with clinically significant carbapenemases.

Abbreviations: AMRHAI, Antimicrobial Resistance and Healthcare Associated Infections; PCR, polymerase chain reaction.

Isolates included

450 isolates comprising of 100 samples each with KPC, NDM, VIM and OXA‑48 variant genes, 2 isolates containing both NDM and OXA‑48 genes, 24 isolates containing IMP genes and 24 isolates that were carbapenem resistant, but did not contain a known carbapenemase gene.

Primary outcome results:

Assay performance (sensitivity)

Xpert Carba‑R

Check‑Direct CPE with ABI 7500 platform

Check‑Direct CPE with BD MAX platform

eazyplex SuperBug complete A

KPC

100%

100%

100%

100%

OXA‑48 variant

83%^a,b

100%

100%

83%^a,b

NDM

100%

100%^c

100%^c

100%

VIM

100%

100%^c

100%^c

100%

IMP

71%^d

N/A

N/A

N/A

NDM + OXA‑48 variant

2 × NDM; 1 × OXA‑48

2 × NDM; 2 × OXA‑48

2 × NDM; 2 × OXA‑48

2 × NDM; 1 × OXA‑48

Non‑carbapenemase

0%

0%

0%

0%

The overall test sensitivity was 94.3% (402/426).

The Xpert Carba‑R test successfully detected the correct carbapenemase gene in all 302 isolates with a KPC, NDM or VIM enzyme.

17.6%^e (18/102) of the isolates with an OXA‑48 variant carbapenemase gene were not detected by Xpert Carba‑R. A modified Xpert Carba‑R version 2 kit was subsequently provided and correctly identified these as OXA‑181 variants.

Abbreviations: IMP, imipenemase metallo‑beta‑lactamase; KPC, Klebsiella pneumoniae carbapenemases; NDM, New Delhi metallo‑beta‑lactamase; OXA, oxacillinases; VIM, Verona integron‑encoded metallo‑beta‑lactamase.

^aSequencing identified OXA‑181 in 18 isolates where the OXA‑48 variant gene was not detected.

^bCalculated to be 82.4% by the authors of this briefing.

^cThe Check‑Direct CPE kit does not distinguish between NDM and VIM producers on the ABI 7500 platform, but does on the BD Max platform.

^dSequencing identified close matches with IMP‑4, IMP‑7, IMP‑8, IMP‑13 and IMP‑14 in 7 isolates where an IMP‑1 variant gene was not detected.

^eCalculated by the authors of this briefing, not explicitly reported in the study.

Study component

Description

Objectives/hypotheses

To determine the sensitivity and specificity of the Xpert MDRO for detecting carbapenem resistance genes (KPC, NDM, VIM) when compared to the results of culture with and without a broth enrichment step followed by Check‑Points microarray detection.

Study design

Diagnostic accuracy study.

Setting

Swab samples were obtained from 2 hospitals in the United States and 1 hospital in Spain. Stool samples were obtained from another US hospital.

Inclusion/exclusion criteria

None reported.

Primary outcomes

Performance characteristics: sensitivity, specificity, PPV, NPV.

Statistical methods

95% confidence intervals were calculated using the Clopper‑Pearson/Fisher exact.

Conclusions

The authors concluded that the Xpert MDRO assay can detect the KPC, NDM and VIM carbapenemase genes directly from rectal swab samples.

Abbreviations: CI, confidence intervals; IMP, imipenemase metallo‑beta‑lactamase; KPC, Klebsiella pneumoniae carbapenemases; MDRO, multi‑drug resistant organisms; NDM, New Delhi metallo‑beta‑lactamase; NPV, negative predictive value; PPV, positive predictive value; US, United States; VIM, Verona integron‑encoded metallo‑beta‑lactamase.

Samples included

328 swab samples were included; 121 single rectal swabs in Amies medium, 74 single rectal swabs in Stuart's medium, 35 single peri‑rectal swabs in Stuart's medium, and 98 double swabs that were dipped in discarded stool specimens and placed in Stuart's medium.

66 contrived stool samples were prepared and confirmed to have various dilutions using 3 Klebsiella pneumoniae isolates containing the NDM gene.

Primary outcome results:

Reported Xpert Carba‑R performance characteristics from swab samples (95% CI):

VIM

KPC

Sensitivity

100% (71.7–100%)

100% (92.8–100%)

Specificity

99.4% (97.7–99.9%)

99.0% (97.0–99.8%)

PPV

81.8% (48.2–97.7%)

93% (80.9–98.5%)

NPV

100% (99.1–100%)

100% (98.9–100%)

Positivity of the Xpert Carba‑R on contrived stool samples with NDM‑containing organisms (%) (95% CI):

Dilution of Klebsiella pneumoniae isolates

Positivity

150 CFU/ml (n=15)

93.3% (68.1–99.8%)

300 CFU/ml (n=15)

100% (81.9–100%)

600 CFU/ml (n=12)

100% (77.9–100%)

1200 CFU/ml (n=12)

100% (77.9–100%)

1800 CFU/ml (n=12)

100% (77.9–100%)

Abbreviations: CI, confidence intervals; CFU, colony‑forming unit; IMP, imipenemase metallo‑beta‑lactamase; KPC, Klebsiella pneumoniae carbapenemases; MDRO, multi‑drug resistant organisms; NDM, New Delhi metallo‑beta‑lactamase; NPV, negative predictive value; PPV, positive predictive value; VIM, Verona integron‑encoded metallo‑beta‑lactamase.

Study component

Description

Objectives/hypotheses

To evaluate the performance of the Xpert Carba‑R assay using clinical isolates and faecal specimens directly when compared to conventional multiplex PCR.

Study design

Diagnostic accuracy.

Setting

Clinical isolates of carbapenem resistant E. coli and K. pneumoniae were collected from bloodstream infections between January and December 2013.

Faecal specimens were also directly tested for the presence of carbapenemase genes using the Xpert Carba‑R test.

Inclusion/exclusion criteria

Not reported.

Primary outcomes

Performance characteristics: sensitivity, specificity, PPV and NPV.

Statistical methods

Not reported.

Conclusions

The authors concluded that the Xpert Carba‑R would be useful for the prompt detection and isolation of patients infected or colonised with strains that may harbour carbapenemase genes. However, they highlight that the incorporation of OXA‑48 variant specific sequences in the panel may help improve its sensitivity and maximise the coverage of the assay.

Abbreviations: E. coli, Escherichia coli; K. pneumoniae, Klebsiella pneumoniae; NPV, negative predictive value; OXA, oxacillinases; PPV, positive predictive value.

Samples included

120 clinical isolates: E. coli (n=32), K. pneumoniae (n=88)

26 faecal specimens.

Primary outcome results:

Prevalence of each carbapenemase gene

Conventional multiplex PCR

Xpert Carba‑R

NDM

40% (48/120)

55% (66/120)

OXA‑48 variant

39.2% (47/120)

0% (0/120)

Co‑producers of NDM and OXA‑48 variant

12.5% (15/120)

0% (0/120)

Reported performance characteristics of Xpert Carba‑R test (as reported, excluding the OXA‑48 variant results):

Sensitivity: 100%,

Specificity: 77%,

PPV: 96%

NPV: 100%

Of the tested faecal samples, 46% (12/26) were identified as carbapenemase producers, 9 with NDM, 2 with the co‑production of NDM and VIM and one with the co‑production of NDM and KPC.

Abbreviations: E. coli, Escherichia coli; IMP, imipenemase metallo‑beta‑lactamase; K. pneumoniae, Klebsiella pneumoniae; KPC, Klebsiella pneumoniae carbapenemases; NDM, New Delhi metallo‑beta‑lactamase; NPV, negative predictive value; OXA, oxacillinases; PPV, positive predictive value; VIM, Verona integron‑encoded metallo‑beta‑lactamase.