The Journal of Hospital Infection - Supplement 1

Among the infected patients within last 90 days, 85% exposure to

antibiotics, and 67% have history of MDR. 97% were hospital acquired.

Among the patients of MDR-PA, 56% were colonization and 44% were

infection.

The overall, 96.6% MDR-PA isolates shown resistance to cefepime,

91.2% ciprofloxacin, 90.7% piperacillin/tazobactam, 90.2% meropenem,

73.2% gentamicin, 58% amikacin, 54.6% tobramycin, and only 3.4% to

colistin.

Clinical outcome: Overall 60% were cured, 36% died and 4% relapsed.

Discussion and/or Conclusion(s):

Our study showed a relatively low

prevalence (8%), but there were five isolates resistant to all antibiotics

tested in Qatar (Pandrug-resistance). Colistin shows high sensitivity

(96.6%) and can be used for managing severe patients with suspected

infections with MDR-PA.

ID: 5082

Using a simple Point-Prevalence Survey to define appropriate

antibiotic prescribing in hospitalised children across the United

Kingdom

Myriam Gharbi

, Katja Doerholt

, Stefania Vergnano

Julia Anna Bielicki

, Stéphane Paulus

, Esse Menson

Andrew Riordan

, Hermione Lyall

, Sanjay Valabh Patel

Jolanta Bernatoniene

, Ann Versporten

, Maggie Heginbothom

Herman Goossens

, Mike Sharland

Imperial College London,

George

’

s Hospital NHS Trust,

St. Georges University of London,

Alder Hey

Children

’

s NHS Foundation Trust,

Evelina London Children

’

s Hospital,

St Mary

’

s Hospital Imperial College Healthcare NHS Trust,

Southampton

Children

’

s Hospital,

University Hospitals Bristol NHS Foundation Trust,

University of Antwerp,

Public Health Wales

Background:

The National Health Service England, Commissioning

for Quality and Innovation for antimicrobial resistance aims to reduce

the total antibiotic consumption and the use of broad-spectrum

antibiotics in secondary care. However, robust baseline antibiotic use

data are lacking for hospitalised children.

Aim(s)/Objective(s):

To describe and explain the prescription patterns

of antibiotics within and between paediatric units in the UK and

provide a baseline for antibiotic prescribing for future improvement

using CQUIN AMR guidance.

Method(s):

Point prevalence survey (PPS) in 61 paediatric units across

the UK. The standardised study protocol from the Antimicrobial

Resistance and Prescribing in European Children (ARPEC) project was

used. All inpatients under 18 years-old were included except neonates.

Results:

A total of 1247 (40.9%) of 3047 children hospitalised on

the day of the PPS were on antibiotics. The proportion of children

receiving antibiotics showed a wide variation between district general

and tertiary hospitals, with 36.4% (33.4

–

39.4) and 43.0% (40.9

–

45.1)

of children prescribed antibiotics respectively. About a quarter of

children on antibiotic therapy received either a medical or surgical

prophylaxis with parenteral administration being the main prescri-

bed route for antibiotics (>60% of the prescriptions for both type

of hospitals). General paediatrics units were high prescribers of

critical broad-spectrum antibiotics, i.e. carbapenems and piperacillin-

tazobactam.

Discussion and/or Conclusion(s):

We provide a robust baseline for

antibiotic prescribing in hospitalised children in relation to current

national stewardship efforts in the UK. Repeated PPS with further

linkage to resistance data need to be part of the antibiotic stewardship

strategy to tackle the issue of suboptimal antibiotic use in hospitalised

children.

ID: 5117

Gram-negative bacteraemia

–

a retrospective audit

Arpit Vyas

, Dina El-Zimaity

, Essam Rizkalla

University Hospitals of

Leicester,

Kettering General Hospital

Background:

Gram-negative bacteraemia (GNB) is associated with

high mortality rates. Administering appropriate antibiotic treatment

promptly affects overall prognosis and outcome.

Aim(s)/Objective(s):

This retrospective audit aimed to review the iso-

lates, investigations, management and mortality of hospital patients

with GNB at Kettering General Hospital, England.

Method(s):

This retrospective audit covered a two month period

01.01.2016 to 29.02.2016. All patients with GNB from an existing

database were included (n = 60). We defined GNB as a confirmed

gram-negative organism from a blood culture. We defined a mortality

associated with GNB as any patient deceased by 31.03.2016.

Data was collected using the hospital t-path software, discharge

letters, the hospital haematology and biochemical database and

patient notes where possible. Data was analysed using a Spreadsheet

(Microsoft Excel 2010).

Results:

The results show there were 20 (33%) cases of GNB associated

mortality. The most common empirical antibiotic of choice was

Tazocin monotherapy (54% of cases). On average, Microbiology called

within 1.1 days with the ID of organism and 1.9 days with sensiti-

vity for the organism. This lead to antibiotic changes in 25% of cases.

The most common organism was

E. coli

and urinary tract infection

was the most common presumed source of infection. GNB showed

high resistance rates for amoxicillin (55%), co-amoxiclav (37%) and

trimethoprim (34%). In 25% of cases, a urine sample was not sent for

culture.

Discussion and/or Conclusion(s):

GNB is associated with high

mortality in hospital patients. Urine samples for culture and sensitivity

must be sent for all cases, especially as

E. coli

is the most common

causative organism. To re-audit in 6 months.

ID: 5121

Closed-loop controller systems for the precision delivery of

vancomycin: An in silico proof-of-concept

Akash Philip, Timothy Rawson, Luke Moore, Alison Holmes,

Pantelis Georgiou, Pau Herrero.

Imperial College London

Background:

Wide inter- and intra-individual pharmacokinetic

(PK) variability has been observed in vancomycin therapy. We

developed and investigated,

in-silico

, two closed loop-systems for

precision antimicrobial delivery of vancomycin against defined PK-PD

targets.

Method(s):

A Proportional-Integral-Derivative (PID) controller and an

Iterative Learning Controller (ILC) were designed to control conti-

nuous and intermittent vancomycin infusions, respectively using

therapeutic drug monitoring levels. One- and two-compartment PK

models obtained from 24 patients receiving vancomycin were used

to evaluate the controllers. Variables such as weight, clearance, ethni-

city, gender, and agewere considered. Intra-day variability of clearance

and sensor error were simulated to produce more realistic

in-silico

conditions. For the PID controller, a continuous vancomycin sensor,

was assumed. A 24-hour Area-Under-the-Curve(AUC):Minimum-

Inhibitory-Concentration (MIC) (AUC:MIC) > 400 was defined as

success target for both controllers. The PID controller was evaluated

using the two-compartment model and the set-point (i.e. plasma

concentration) was periodically adjusted (e.g. 6 hourly) in order to

achieve the overall 24-hour target.

The ILC controller was tested using the one-compartment model, a

12-hour sampling time, a 1000 mg initial dose, and a trough level

set-point of 10 mg/L.

Results:

The PID controller achieved the set-point (AUC:MIC > 400) for

all 24 individuals in less than 5 hours. For the ICL controller, 22/24

simulated individuals achieved AUC:MIC > 400. 2/22 individuals had

AUC:MIC > 700 that may be associated with toxicity.

Discussion and/or Conclusion(s):

Both the PID and ILC controller

simulations attained required PK-PD targets. The PK profiles obtained

were within acceptable ranges observed in clinical practice. This

provides evidence to support the potential feasibility of using closed-

loop systems for antimicrobial delivery.

Abstracts of FIS/HIS 2016

–

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

–

S134

S33