The Journal of Hospital Infection - Supplement 1

advantages and disadvantages. The Royal Free London NHS

Foundation Trust screens all patients considered

‘

high-risk

’

for

colonisation with CPO.

Aim(s)/Objective(s):

To review current CPO screening methodology

and results from samples received during Jan-Apr 2016, with

particular focus on carbapenemase types identified and turn-around

times.

Method(s):

Screening samples are cultured on chromogenic media;

the penicillin/cephalosporin antibiogram, together with MICs to

relevant carbapenems, are then used to identify potential CPOs

(pCPOs). These are tested by KPC/MBL Confirm Kit (Enterobacteria-

ceae) or KPC/MBL Confirm in

P. aeruginosa/Acinetobacter

Kit [Rosco

Diagnostica]. Possible OXA-48 positive Enterobacteriaceae are high-

lighted by resistance to Temocillin and require Reference Laboratory

confirmation.

Results:

During Jan

–

Apr 2016 inclusive, 3775 screening samples (1735

patients) were tested for CPO. 3664/3775 (97.06%) samples were

shown to be negative for CPO using chromogenic media and/or

antibiogram analysis.

pCPOs were isolated from 111/3775

(2.94%) samples from 77/1735 (4.43%) patients, of which 54 [from 52

patients] represented non-duplicate isolates that were tested for

carbapenemase production. 22/54 were confirmed CPOs (41% of

pCPO samples, 42% of pCPO patients, 0.6% of screening samples,

1.3% of patients screened): 8 OXA-48, 5 VIM, 4 NDM, 4 OXA-23 and

1 KPC.

Discussion and/or Conclusion(s):

OXA-48 represented 36.4% of the

CPOs detected, highlighting a need to adapt our protocol to improve

detection of this genotype, particularly with regard to turn-around

time which currently relies on Reference Laboratory confirmation.

Consequently, we are evaluating RapidEC CarbaNP [Biomerieux] and

Xpert Carba-R PCR [Cepheid] to improve detection of CPOs.

ID: 4870

Review of

Mycobacterium tuberculosis

sensitivity testing protocol

for resistant strains at the Scottish Mycobacteria Reference

Laboratory

Alex Kennett

, Pauline Claxton

, Ian Laurenson

NHS Lothian,

Scottish Mycobacteria Reference Laboratory

Background:

The World Health Organisation (WHO) estimated that

multi-drug resistance (MDR) was found in 3.3% of new cases of

Mycobacterium tuberculosis

(TB) in 2014 and 20% of previously treated

cases.

Aim(s)/Objective(s):

To assess the utility of extended Drug

Susceptibility Testing (DST) carried out on resistant TB isolates at the

Scottish Mycobacterial Reference Laboratory (SMRL).

Method(s):

The use of the BACTEC MGIT 960 system, second line TB

eXiST testing, TREK sensititre MYCOTB MIC plates, Hain GenoType

MTBDRplus and MTBDRsl genotypic testing on 44 resistant clinical

isolates was examined between the end of January 2014 and 11th June

2015.

Results:

18 isolates showed mono-drug resistance; isoniazid (10),

pyrazinamide (7) and streptomycin (1). 26 isolates were resistant to

more than one drug. 6 were multi-drug resistant and 2 were extremely

drug resistant. Complete phenotypic and genotypic testing was carried

on the majority of samples with 100% isolates tested using MGIT 960

system, TB eXist software and MYCOTB MIC plates, 100% were tested

with MTBDRplus and 93.2% (n = 41) with MTBDRsl testing. In total,

only 3 (6.8%) of isolates showed inconsistent results between the

different modalities of testing and 1 demonstrated ethambutol

resistance after extended incubation.

Discussion and/or Conclusion(s):

On the basis of this review, resis-

tance testing was streamlined to carry out MTBDRplus and second line

TB eXist testing on mono-resistant isolates, reserving MYCOTB plates

and MTBDRsl testing for multiple drug resistant isolates.

ID: 4922

Improving detection of CPOs in the diagnostic microbiology

laboratory

Hema Sharma

, Gemma Vanstone

, Indran Balakrishnan

Royal Free

London NHS Foundation Trust,

HSL Analytics

Background:

To aid identification of carbapenemase producing

organisms (CPOs), The Royal Free London NHS Foundation Trust uses

susceptibility profiles to penicillins and cephalosporins, alongside

MICs to relevant carbapenems. Possible CPOs are then tested by KPC/

L Confirm (Enterobacteriaceae) or KPC/M

L Confirm (

P. aeruginosa

Acinetobacter

) Kits [Rosco Diagnostica]

–

results are available 24 h

later. Due to the inability of Rosco assays to detect OXA-48, resistance

to temocillin is used as a surrogate marker for this genotype; possible

OXA-48s require Reference Laboratory (AMRHAI, PHE) confirmation,

which incurs a delay of 3

–

5 d.

Patients with possible CPOs are isolated in side-rooms and removed

if/when isolates are confirmed negative. Reducing turnaround-time

(TAT) to CPO detection will optimise both infection control and

antibiotic stewardship.

Aim(s)/Objective(s):

To improve CPO detection TAT with use of

RapidEC CarbaNP [Biomerieux] and Xpert Carba-R [Cepheid] assays.

Method(s):

42 unique isolates from screening specimens received

between Jan-Apr 2016 meeting susceptibility criteria for carbapene-

mase testing were additionally tested by RapidEC CarbaNP and Xpert

Carba-R. All isolates were sent to AMRHAI for confirmation.

Results:

AMRHAI identified 19/42 CPOs [8 OXA-48, 5 NDM, 4 VIM, 1

KPC, 1 OXA-23]. Our in-house method of Rosco plus temocillin

susceptibility (100% sensitivity, 43.5% specificity) lacked specificity

due to its inability to detect OXA-48s. The RapidEC CarbaNP (100.0%

sensitivity, 91.3% specificity) and Xpert CarbaR (94.7% sensitivity, 100%

specificity) performed well and detected all OXA-48 positive CPOs

with results available within 1.5

–

3 h.

Discussion and/or Conclusion(s):

Both RapidEC CarbaNP and Xpert

CarbaR assays improved TAT to CPO detection (particularly OXA-48s).

Xpert Carba-R had the additional advantage of determining carbape-

nemase type.

ID: 4929

Diagnostic accuracy evaluation of the RenDx Fungiplex assay for

detection of candidaemia

Cara McKeating

, Raquel Posso

, Lewis White

, Ronan McMullan

1,3

Microbiology, Belfast Health and Social Care Trust,

Regional Mycology

Unit, Public Health Wales Microbiology Cardiff, University Hospital of

Wales,

Centre for Experimental Medicine, Queens University Belfast

Background:

The Renishaw RenDx Fungiplex is a new PCR-based

test that detects

Candida

Aspergillus

DNA in blood/serum/plasma

specimens.

Aim(s)/Objective(s):

The aimwas to provide preliminary evaluation of

the diagnostic accuracy of this test, for detection of candidaemia in the

NHS service context.

Method(s):

Twelve consecutive adults with candidaemia were identi-

fied prospectively. For 11 of these patients serum taken contempor-

aneously with the blood culture was available and was stored at

−

70°C

for batch analysis. Where possible, serum obtained the preceding

and/or following day was also collected. Serum from thirty-nine adults

with a paired negative blood culture was also collected in a similar

manner for negative controls. Staff performing PCR testing were

blinded to the culture result. Sensitivity and specificity were

calculated, expressed with 95% confidence intervals.

Results:

In the primary analysis, a comparison of paired blood culture

and contemporaneous serum was undertaken. Of 11 candidaemic

patients, the paired serumwas PCR-positive in 4, generating sensitivity

of 36.4% (8.0

–

64.8%). Of the 39 control sera, 5 were PCR-positive,

generating specificity of 87.2% (76.7

–

97.7%). The sensitivity increased

to 63.3% (34.8

–

91.8%) in a secondary analysis when the serum samples

from the day before/after the positive blood culturewere included (7 of

11 detected).

Abstracts of FIS/HIS 2016

–

Poster Presentations / Journal of Hospital Infection 94S1 (2016) S24

–

S134

S76