FEDERATION OF
infection societies
CONFERENCE 2018

13th – 15th November 2018 | Sage Gateshead – Newcastle

FEDERATION OF
infection societies
CONFERENCE 2018

13th – 15th November 2018 | Sage Gateshead – Newcastle

FEDERATION OF
infection societies
CONFERENCE 2018

13th – 15th November 2018 | Sage Gateshead – Newcastle

Posters

GENERAL BACTERIOLOGY – Poster Nos. 112-134

112
Use of MALDI-Tof for the identification of non fermenting gram negative rods in cystic fibrosis

Abstract - 112

Poster 112

Use of MALDI-Tof for the identification of non fermenting gram negative rods in cystic fibrosis

Rachel Austin-Hutchison, James Greig
Universities Hospital Plymouth

Introduction: Laboratory guidelines for the identification of non-fermenting gram negative rods (NF-GNR) in Cystic Fibrosis (CF) caution against using phenotypic techniques due their poor performance. With the exception of morphologically typical Pseudomonas aeruginosa, achromobacter and stenotrophomonas, for which an excellent phenotypic identification is acceptable, all other NF-GNR should be identified using molecular techniques. Since adoption of these guidelines, the UHPNT Microbiology Laboratory has introduced routine use MALDI-Tof. To assess the accuracy of this as a stand-alone test for the identification of NF-GNR in CF samples, both MALDI-Tof and molecular techniques were performed in parallel and a retrospective review undertaken.

Methods: The laboratory sample data base was interrogated and all NF-GNR isolates submitted to a reference laboratory for identification between 2013 and March 2018 were identified. The correlation between isolates speciated using local MALDI-Tof (identification scores of >2) and molecular reference laboratories (any Public Health England reference laboratory or a locally favoured commercial laboratory) was assessed to genus and species level. If a reference laboratory report gave a selection of possible results then the best that could be achieved was a genus level correlation even if one of the options matched the MALDI-Tof result.

Results: Including duplicates (same NF-GNR isolated on >1 occasion from same patient) a total of 67 isolates of interest were identified. Local MALDI-Tof identification correlated with the Public Health England reference speciation on virtually all occasions with only one reference laboratory report noting an achromobacter that was likely a novel species (locally Achromobacter xylosoxidans MALDI 2.3) Commercial 16s rRNA PCR correlated with MALDI-Tof to genus level on all occasions but failed to speciate either at all ie reported to genus level only or gave a species that did not match exactly the local MALDI result on 17/36 (47%) occasions. On eight of these 17 occasions the local MALDI result matched one of two possible species suggested. Members of the Burkholderia cepacia complex (BCC) were accurately identified to genomovar level using MALDI-Tof but the tested sample consisted of only four patients, three of whom were infected with a Burkholderia multivorans and we have no data to identify if these were different or identical strains

Discussion: The correlation between MALDI and the available reference laboratory testing indicate that MALDI is a reliable stand-alone test for the range of NF-GNR isolated in individuals with CF and appears to out perform some molecular tests. This confirms locally what has already been reported and further strengthens the case for inclusion of MALDI-Tof in routine CF laboratory methods. Based on the finding of this and previous studies we have locally dispensed with molecular reference identification (except on initial isolation) and rely on stand alone local MALDI-Tof for the identification of NF-GNR in CF with the exception of BCC. We encourage the CF Trust to review the current laboratory guidelines regarding the need to use molecular identification techniques and embrace MALDI-Tof.

113
E. coli blood stream infection – a 6 month prospective observational series

Abstract - 113

Poster 113

E. coli blood stream infection – a 6 month prospective observational series

Patrick Lillie, Greta Johnson, Monica Ivan, Peter Moss
Department of Infection, Hull and East Yorkshire Hospitals NHS Trust

Introduction: E. coli blood stream infection is a common infection in secondary care and with increasing antibiotic resistance can be difficult to treat. Enhanced reporting of E. coli bacteraemia has occurred since 2011 and a government initiative to reduce Gram negative infections by 50% by 2021 was announced in 2016. To better clarify the local epidemiology, outcomes and associated features we undertook a prospective 6 month review of adult E. coli bacteraemia in our trust.

Methods: From 1st November 2017 to 30th April 2018 all patients with E. coli blood stream infections were reviewed by an infectious diseases consultant, either at the bedside or a notes review. Data were recorded on a standardised pro forma and spread sheet. Demographic, microbiological, clinical, laboratory, treatment and outcome data were recorded. 30 day mortality, length of stay and 90 relapse were ascertained as well as descriptive statistics of community and trust atributed cases. Reasons for preventable cases were recorded for both community and acute trust attributed cases. Predictors of mortality were analysed using Fishers exact and Mann-Whitney tests respectively.

Results: Of 207 total E. coli episodes, 195 cases in 188 patients were reviewed (111 at the bedside and 84 notes reviews). 45 cases were acute trust attributed, acquired, 1 from and 149 were community attributed, and one was in a private hospital acquired cases. 51% of cases were secondary to a urinary tract source, with 25% having a hepatobiliary source. 50% of acute trust and 17% of community acquired cases were felt to have been preventable, with nearly half of all preventable cases being related to urinary catheters. Antibiotic resistance to commonly used 1st line agents was high with resistance rates of 54% to co-amoxiclav, 46% to co-trimoxazole, 23% to piperacillin-tazobactam and 17% to ciprofloxacin. Co-morbidity was high, with 26% of patients having a diagnosis of cancer, 24% diabetes and 12% dementia. 7 day mortality was 13.3%, with a 30 day mortality of 23.6%. 7 (5.7%) of patients relapsed before 90 days, and the median length of stay post bacteraemia was 7 days. Factors associated with 30 day mortality in a univariate analysis were – increasing age (p<0.0001), coming from residential care (p<0.0001) higher Charlson co-morbidity score (p<0.0001) higher early warning score (p<0.0001), higher serum sodium, creatinine and urea (all p<0.0001) lower serum albumin (p<0.0001), shorter time to blood culture becoming positive (p=0.0466) and the empirical antibiotic regime not including gentamicin (p<0.0001). Source of infection, trust or community attribution acquired and activity of the empiric regime were not associated with mortality.

Discussion: E. coli blood stream infections were associated with a marked early mortality, with a sizable number being potentially preventable. Targeting interventions to improve urinary catheter care and reduce the duration of catheter use would be prudent. Outcomes are poor with an appreciable mortality that appears to be driven by severity of infection and co-morbidity, although the use of gentamicin empirically may offer a potential way to improve outcomes and this may benefit from further clarification in larger studies.

114
Necrotising otitis externa: lessons learnt from a large retrospective cohort

Abstract - 114

Poster 114

Necrotising otitis externa: lessons learnt from a large retrospective cohort

Susanne Hodgson1,2, Victoria Sinclair2, Monique Andersson2
1University of Oxford. 2Oxford University Hospitals NHS Foundation Trust

Introduction: Necrotising otitis externa (NOE) is a rare, invasive infection of the external auditory canal extending into surrounding structures including the skull base. It is associated with considerable morbidity. No widely accepted diagnostic criteria or treatment guidelines exist.

Methods: We performed a retrospective, descriptive analysis of the demographics, co-morbidities, clinical management and outcomes for cases of NOE diagnosed at the Oxford Radcliffe Hospitals NHS Foundation Trust between August 2013 and March 2018. 

Results: Eighty-eight patients were identified. Cases were classified as definite, probable or possible on the basis of radiological and clinical information; definite cases (n=63) were defined as those with fitting radiological findings or cranial nerve palsy and an appropriate clinical syndrome (pain and granulation tissue); probable cases (n=22) were those with a fitting clinical syndrome and risk factors for NOE; possible cases were those with a fitting clinical syndrome alone (n=3). Further analysis of definite cases was undertaken.

As previously described, more male than female cases were observed (65% v 35%). The median age at presentation was 84 years (range 20-104). 62% of cases were diabetic and 8% had no identifiable risk factors for infection.

A pathogen was identified on sampling in 38 cases (60%). Of these cases, Pseudomonas aeruginosa was isolated in 79%. 2 out of 28 (7%) isolates of P. aeruginosa were resistant to ciprofloxacin. Other pathogens included E. faecalis, S. pyogenes, S. aureus, Candida spp. and mixed anaerobes.

Specialist infection team input occurred in 89% of cases. The median duration of intravenous antibiotic therapy was 2 weeks (range 0-9.7) and the median total duration of parenteral antibiotic therapy was 6.3 weeks (range 0.9-44.9). For cases receiving intravenous antibiotics (98%), a single anti-pseudomonal agent was used. Follow on oral antibiotics were prescribed for 89% of cases. Of these cases, ciprofloxacin was prescribed alone (80%) or in combination with rifampicin, co-trimoxazole, clindamycin or co-amoxiclav (20%). Only one individual received anti-fungal therapy, which was prescribed empirically. 93% of patients received topical as well as parenteral antibiotics.

Ceftazidime and Pipercillin-Tazobactam were inappropriately dosed for age and renal function for 28% and 3% of cases respectively. Oral ciprofloxacin was inappropriately dosed for 24% of cases. Of the 14 patients on QT prolonging medications for which ciprofloxacin was indicated, only 2 (18%) had this medication reviewed when ciprofloxacin was started. The Infection Team documented advice regarding QT monitoring, INR monitoring and potential side effects of tendonitis and diarrhea with ciprofloxacin in 14%, 66%, 18% and 16% of appropriate cases respectively.

10% of patients underwent surgery in theatre for the following indications: facial nerve decompression (n=2), sampling (n=4) and tarsorraphy (n=1).

10% of cases required readmission for worsening symptoms. All cases presented after completion of initial course of antibiotics. There was no significance difference in duration of parenteral antibiotics for those patients re-admitted and those not (6.3 v 6.4 weeks median. p=0.428, Mann-Whitney test). Mortality at 1 year was 31% for 48 patients with at least 1-year follow-up post diagnosis.

Discussion: This dataset represents one of the largest retrospective case series of NOE to date. Whilst P. aeruginosa was the predominant pathogen isolated, a significant proportion of cases had no pathogen identified and relied on empirical therapy. The role of superficial swab results to determine antimicrobial therapy is unclear. Dosing errors were common and communication by the Infection team of monitoring requirements, drug interactions and potential side effects associated with ciprofloxacin was variable. Further prospective studies are required to stratify patients’ risk of NOE and inform appropriate management regimens.

115
Endocarditis in injecting drug users in the North East of England

Abstract - 115

Poster 115

Endocarditis in injecting drug users in the North East of England

Chris Lawrence, Brendan MacCarron
South Tees Hospitals, Middlesbrough

Introduction: Injecting drug use is a significant cause of morbidity and mortality in the United Kingdom and globally. Infective endocarditis in UK drug users is poorly characterised and inadequately researched.

Methods: We performed a retrospective review of admissions to a large Teaching Hospital in the North of England with a diagnosis of endocarditis in known intravenous drug users(IDUs) over a 5 years period from May 2013 to April 2018. Notes were screened for a diagnosis of endocarditis from coding and then individual cases were reviewed to confirm the diagnosis and establish demographic, microbiological and pathological data.

Data for endocarditis admission in non-intravenous drug users (non-IDUs) was collected over the same time period and compared to the heroin users. Patient demographics and outcomes were compared between the two groups and analysed using Graphpad software and the Fishers exact test and t-test as appropriate. 

Results: 32 separate cases of endocarditis were identified in 26 IDUs compared to 129 cases with endocarditis in the non-IDU population. IDU endocarditis cases were significantly younger,( mean age 38 vs 68 in non-IDU, mean age difference 27.41 years (95% ci 20.93-33.9 p=0.0001)]. Mean length of hospital stay was similar 40.6 vs 37.5 days p=0.65.

Survival to discharge was of borderline significance lower in IDU vs non-IDU (n=28 [87.5%] versus n=125 [96.9%] p=0.051). Survival in those IDU trackable to 2 years was significantly lower: 59% (13 of 22 patients) versus 75% (97 of 129) in non IDU p=0.023.

In the IDU endocarditis case series 78% of cases were in males and 22% female. No IDU were infected with HIV. 9 (28%) were Hep C positive. Two out of 32 IDU cases were culture negative; Blood culture was positive in 28 of 32 patients and valve culture and 16S PCR positive in 5 of 13 who underwent surgery. 19 patients were positive for MSSA, no patients were positive for MRSA. This cohort occupied a total or 1277 bed days over the study period.

Discussion: Endocarditis in IDUs is a significant cause of morbidity and mortality. These patients require interventions that take into account their age, routes of infection and underlying social issues. The observed microbiology can allow for a tailored approach taken into account their likely community acquired pathogens. Innovative and combined approaches to antibiotic and addictions therapy are required to minimise morbidity, mortality and healthcare burden. Follow-up studies should look into the raised mortality rates in those successfully treated and discharged.

116
A service evaluation of prevention and management of ventricular assist device (VAD) infections in a UK specialist cardiothoracic centre

Abstract - 116

Poster 116

A service evaluation of prevention and management of ventricular assist device (VAD) infections in a UK specialist cardiothoracic centre

Vanessa Wong1, Sumita Pai2, Ruth Kappeler2, Olly Allen2, Huina Yang2
1Addenbrooke’s Hospital, Cambridge. 2Royal Papworth Hospital, Cambridge

Introduction: Mechanical circulatory support with ventricular assist devices (VADs) provides long-term haemodynamic support as a bridge to myocardial recovery or cardiac transplantation. However, VAD infections are linked to significant morbidity and mortality. The Royal Papworth Hospital guidelines for the insertion of VADs were updated on 1st March 2016 and recommended perioperative measures, including screening for multidrug-resistant (MDR) pathogens and universal decolonisation of Staphylococcus aureus to prevent infection, as well as providing advice on the diagnosis and management of infection. We audited the compliance to our guidelines, assessed rates of infection and outcomes, particularly recurrence and mortality, with the aim to improve clinical practice.

Methods: We collected data for patients implanted with VADs from 1st March 2016 until 30th September 2017 to assess compliance with the guidelines and from 1st January 2015 to 30th September 2017 for overall rates of infection, causative agents, recurrence of infection and mortality.Patient demographics, device details, microbiology, radiology, clinical management and outcome were obtained from electronic and paper notes.

Standards

  1. All patients should be screened for MRSA, VRE, ESBLs (Acinetobacter, CPE if patient was from a high-risk centre)
  2. All patients should be screened for infection: sputum, urine, swabs, blood cultures
  3. All patients should be offered decolonisation treatment for S. aureus
  4. Perioperative antibiotic prophylaxis should be commenced as per guidelines
  5. All patients should receive driveline care
  6. In superficial infection, swabs and blood cultures should be sent and imaging should be performed to exclude deep infection
  7. In a superficial/deep infection, the patient should be prescribed an appropriate antibiotic for the correct duration in accordance to local policy.

Results: 10 patients had VAD implants from 1st March 2016 until 30th September 2017. Routine screening was performed in 100% (10/10) patients for MRSA, 40% for VRE and 20% for ESBLs. High-risk patients were swabbed for Acinetobacter (33%; 1/3) and CPE (67%; 2/3). No patients received S. aureus decolonisation prior to procedure. Patients were screened for infection with sputum in 100% (1/1), urine in 90% (9/10), and two blood cultures in 40% (4/10) of cases. Routine driveline care was documented in 40% (4/10) of cases. Out of the seven superficial infections, swabs and blood cultures were sent in 100% (7/7) and 71% (5/7), respectively. Appropriate imaging was performed in 71% (5/7) of cases. Appropriate antibiotics for the correct duration were prescribed in 86% (6/7) of superficial and in 100% (3/3) of deep infections.

Analysis of the infection rates from 1st January 2015 to 30th September 2017 showed VAD infections developed in 17 out of 25 patients (68%) over a follow-up period of six months to three years. Superficial driveline infections were reported in nine patients and eight patients presented with deep driveline infections with or without bacteraemia. The average time from VAD insertion to presentation of infection was four months. The commonest pathogen was Staphylococcus aureus (45%), followed by Streptococci/ Enterococci (23%), coagulase-negative Staphylococci (16%), Enterobacteriaceae (9%) and Pseudomonas aeruginosa 7%. Despite appropriate antimicrobial therapy, 47% (8/17) patients had a recurrence of infection. The mortality rate was 40% (10/25) with 30% of deaths attributable to sepsis.

Discussion: A significant proportion of VADs become infected with the majority occurring within six months of insertion. Our analyses revealed that improvement in compliance to the guidelines is needed in areas such as decolonisation and pre-operative screening for MDR organisms and infection. To facilitate this, we have disseminated the results to the transplant team and set up an electronic topical decolonisation bundle to allow ease of prescribing.

117
Multi-centre study of the in vitro activity of ceftolozane/tazobactam and other commonly used antibiotics against Pseudomonas aeruginosa isolates from patients in the United Kingdom

Abstract - 117

Poster 117

Multi-centre study of the in vitro activity of ceftolozane/tazobactam and other commonly used antibiotics against Pseudomonas aeruginosa isolates from patients in the United Kingdom

Adela Alvarez-Buylla1, Mike Allen1, Dan Betts1, Irene Monahan2, Tim Plance2
1MSD Ltd., Hoddesdon. 2St George’s University Hospital, London

Introduction: Ceftolozane/tazobactam (C/T) is a cephalosporin/β -lactamase inhibitor combination with activity against Gram-Negative bacilli including ESBL-producing Enterobacteriaceae and Pseudomonas aeruginosa. In vitro studies conducted in Europe (e.g. Program to Assess Ceftolozane/Tazobactam Susceptibility – PACTS) and elsewhere have demonstrated C/T to be the most active β-lactam antibiotic against P. aeruginosa, second only to colistin; however, there are little UK data to guide appropriate treatment with C/T. Antimicrobial susceptibility testing is routinely performed by diagnostic microbiology laboratories to guide antimicrobial treatment and to detect resistance in clinical bacterial isolates. Until recently, only MIC gradient strips were available for in vitro testing of C/T and P. aeruginosa; however, disc susceptibility testing is a methodology that is easier to perform, inexpensive, and is being increasingly adopted across UK laboratories as a first-line test. The aim of this study was to evaluate the in vitro activity of C/T and other commonly used antipseudomonal antibiotics against P. aeruginosa isolates in the UK by disc susceptibility.

Methods: Clinically significant consecutive isolates from fourteen sites across the UK, geographically spread to include England, Scotland and Wales, were tested, including isolates from cystic fibrosis patients. Isolates were collected from patients between January 2016 and April 2018. MALDI-ToF MS methodology was used for species identification. In vitro susceptibility testing following the EUCAST disc diffusion methodology was used to assess the susceptibility of P. aeruginosa isolates to C/T and nine other antipseudomonal antibiotics (piperacillin/tazobactam (TZP), ceftazidime (CAZ), imipenem (IPM), meropenem (MEM), aztreonam (ATM), amikacin (AMK), gentamicin (GEN), tobramycin (TOB) and ciprofloxacin (CIP)). Susceptibility results were interpreted using EUCAST 2018 criteria.

Results: A total of 1,326 isolates were included in the study. Twelve of the sites contributed 100 isolates each. The remaining two sites tested 69 and 57 isolates. Isolates came from a variety of sample sources including sputum (n=420), blood (n=251), wound swab (n=238), urine (n=204), pus (n=31), bronchoalveolar lavage (n=20), cerebrospinal fluid (n=5), and other (n=157). Isolates were also collected from patients in various wards including respiratory (n=220), ICU (n=153), acute medicine (n=146), general surgery (n=118), emergency department (n=108), nephrology (n=33), gastroenterology (n=17), and other (n=531). Of the isolates tested, 89.4% (n=1,185) were susceptible to C/T, which included 128 CF and 1,057 non-CF isolates. The majority of other antibiotics had lower susceptibility rates, excluding piperacillin/tazobactam (TZP) and tobramycin (TOB), which had susceptibility rates of 90.7% (n=1,203) and 91.6% (n=1,214), respectively. This trend extended across both CF and non-CF isolates. The overall susceptibility for the other β-lactams was as follows: CAZ (80.5%, n=1,067), IPM (80.6%, n=1,069), MEM (78.4%, n=1,039) and ATM (0%, n=0). Susceptibility rates for the aminoglycosides and fluoroquinolone were as follows: AMK (86.0%, n=1,140), GEN (85.4%, n=1,132), and CIP (67.8%, n=899). Susceptibility rates were higher for non-CF isolates across all antibiotics (C/T 93.8%, n=1,057; TZP 94.8%, n=1,068; CAZ 86.3%, n=972; IPM 85.5%, n=964; MEM 82.9%, n=934; ATM 0%, n=0; AMK 92.6%, n=1,043; GEN 91.3%, n= 1,029; TOB 96.0%, n=1,082; CIP 73.7%, n=830).

Discussion: In vitro susceptibility to C/T was high and consistent with data from the BSAC Resistance Surveillance Programme. Unexpectedly, TZP had higher susceptibility rates than seen in previous studies. Lower susceptibility rates were observed for CF isolates in all antibiotics tested.

118
For the record: annual audit as a tool to improve patient handover and the quality of microbiology reports

Abstract - 118

Poster 118

For the record: annual audit as a tool to improve patient handover and the quality of microbiology reports

Katherine Walton
Microbiology, The Newcastle upon Tyne Hospitals NHS Foundation Trust

Introduction: Microbiology at the Newcastle Hospitals (NUTH) provides 24/7 clinical advice for in-patients and outpatients at an 1800-bedded hospital on two main sites, and supports general practice in the area. NUTH is a tertiary referral centre providing a wide portfolio of clinical specialties, the patients are often very complex and different microbiologists including trainees may be involved in a patient’s management at different times. Good handover is therefore essential. Written handovers are used to support out of hours working, supplemented by laboratory bench-rounds and clinical meetings. However, medical microbiology notes added to the patient’s laboratory results provide a permanent record, which can be accessed remotely. These have been encouraged until new integrated electronic patient record and laboratory computer systems are procured. Blood cultures are among the most important microbiology results. We carried out annual snap-shot audits of the medical microbiology (MM) notes attached to positive blood culture records over the past three years. During this time, the laboratory changed from a single to a dual bottle blood culture system.

Methods: Snap-shot retrospective audits of the use of MM notes on positive blood cultures were carried out annually over the same two week period in June from 2016-18, using the laboratory work book and computer system. The number of positive blood cultures, presence of MM notes, and percentage of patient episodes with significant blood cultures where a relevant MM note was available were assessed. An arbitrary standard for this was set at 90%.

Following each audit, the results were fed back to the microbiologists and interventions made to improve compliance, including additional induction sessions, training and revision of the authorisation manual.
The scope of the audit was extended in 2018 to assess the quality of microbiology reports including interpretative comments for probable contaminants and the reporting of appropriate antibiotic sensitivities.

The Trust changed from a single to a dual bottle (aerobic and anaerobic) blood culture system in December 2017, except paediatric patients where a single aerobic bottle was retained.

Results: The use of MM notes in patient episodes with clinically significant positive blood cultures increased from 2016-18: 77% in 2016, 90% in 2017 and 96% in 2018.

From 1/6/18 – 14/6/18, 259 blood culture bottles flagged positive. Eleven were culture negative leaving 248 culture positive blood cultures from 107 patients. As some patients had both probable contaminants and significant pathogens in different blood cultures during the same two week period, 74/107 patient episodes were regarded as clinically significant positive cultures, 28/107 were probable contaminants and it was not possible to determine the significance in 10/107.

Appropriate antimicrobial sensitivities were reported in only 76% of patient episodes with significant positive cultures. For patient episodes with blood cultures containing probable contaminants, interpretative comments highlighting the doubtful clinical relevance were only reported in 57%; conversely, antimicrobial sensitivities were reported in 14%. 

Introduction of the dual bottle blood culture system resulted in an increase in positive blood cultures (2017: 110, 2018: 248) and in the number of patients with clinically significant positive blood cultures (2017: 59, 2018: 74).

Discussion: The audit cycle has helped increase compliance with the use of MM notes in patient’s laboratory records over the past three years, improving patient safety by strengthening microbiology handover of often very complicated cases.

It is now being used to help to improve the quality of the microbiology reports issued to clinicians, including the appropriate use of interpretative comments and antimicrobial susceptibility reporting.

We plan to develop additional audit tools for other aspects of medical microbiology report authorisation and this work is being used to help inform the service requirements during the procurement process for a new laboratory computer system.

119
BactiVac, a network to support the study, development and implementation of bacterial vaccines

Abstract - 119

Poster 119

BactiVac, a network to support the study, development and implementation of bacterial vaccines

Alex G Richter1, Johanna E Dean1, Susan Pope1, Evelina Balandyte1, Adam F Cunningham1, Calman A MacLennan1,2,3
1University of Birmingham. 2University of Oxford. 3Bill and Melinda Gates Foundation, Seattle, USA

Introduction: Infections account for >20% of all deaths worldwide, and are particularly problematic in low-middle income countries (LMICs). Bacterial infections contribute significantly to this burden, killing approximately 5 million people annually. The crisis of antimicrobial resistance means our options for controlling infections is narrowing. Vaccines save millions of lives yearly and are a cost-effective approach to prevent infectious disease and their devastating sequelae. There are many bacterial infections against which we lack any licensed vaccine. To address these issues, the MRC and BBSRC, through the Global Challenges Research Fund, have funded 5 vaccine-related networks. Our network, BactiVac (www.birmingham.ac.uk/bactivac), was launched in August 2017 following the award of £2.2m and is led by Profs Cal MacLennan and Adam Cunningham.

Methods: BactiVac is a global bacterial vaccinology network established to accelerate the development of vaccines against bacterial infections, particularly those relevant to LMICs. The BactiVac Network brings together academic, industrial and other partners involved in vaccine research against human and animal bacterial infections from the UK and LMICs. 

BactiVac fosters partnerships and provides catalyst pump-priming project and training funding to encourage cross-collaboration between academic and industrial partners. Funding opportunities are available to members from LMIC countries to attend the Annual Network Meetings.

Results: The Network funded 12 pump-priming projects and 5 training awards in first round of catalyst funding. Applications submitted to our second round of funding are currently being reviewed and our third funding call is due to open in March 2019. Full details about our funding calls and proposals that we have funded are available on our website.

BactiVac has over 450 members across 53 countries with 37% based in LMIC countries and 12% in industry. Membership is free – join us at http://bit.ly/applyBactiVac.  

Discussion: Benefits of membership include:

  • Access to our catalyst funding schemes – third round opens in March 2019
  • Invitation to our subsidised Annual Network Meetings – registration is currently open for our next meeting on 20-21 March 2019 in Birmingham, UK
  • Access to our Members’ Directory to develop collaborations and access distinct expertise.

We encourage you to become a member and be involved in this initiative, and to be part of its growing success now and in the future.

120
From laboratory to clinical practice- invasive pneumococcal infection at a London Teaching Hospital

Abstract - 120

Poster 120

From laboratory to clinical practice- invasive pneumococcal infection at a London Teaching Hospital

Luciana Sowole, Clare Logan, Kirstin Khonyongwa, Amanda Fife
King’s College Hospital, London

Introduction: Invasive pneumococcal disease (IPD) is a major cause of morbidity and mortality in adults and children worldwide. This analysis aims to review the clinical management of IPD cases at King’s College Hospital (KCH) and identify areas for improvement.

Methods: All S. Pneumoniae isolates cultured from normally sterile sites e.g. CSF, joint fluid and blood culture specimens between May 2015 and April 2018 were identified from the KCH Microbiology database. An in-depth retrospective review of the laboratory data and electronic patient records was then performed.

Results: A total of 96 S. pneumoniae isolates were identified from 90 patients (CSF = 8, blood culture = 85, synovial fluid=3). Two isolates were found to be penicillin-resistant, with a further 18 classified as penicillin-intermediate.

The IPD cases ranged in age from 5 months – 96 years (male 58%, female 42%). In total, 48% (43/90) of patients were immunocompromised due to diabetes, malignancy, sickle cell disease, liver disease, renal disease and HIV. 12% (11/90) of the cohort were HIV positive, with 6 patients newly diagnosed during their admission with IPD. HIV testing was performed in 70% (63/90) of cases.

With regards to vaccine effectiveness, of the 11 serotypes isolated from children aged 0-9 years, 3 serotypes were present in both the 13 and 23 valent vaccine, 7 were present in the 23 valent only and 1 serotype was not present in either vaccine. In those aged over 65, of the 29 serotypes isolated, 15 were present in the 23 valent vaccine only, 8 serotypes were present in both the 13 and 23 valent and 6 serotypes were not present in either vaccine.

Community acquired pneumonia (CAP) was the commonest suspected source of infection (56%, 50/90), followed by meningitis (24%, 22/90) then bone and joint (9%, 8/90). Of the suspected meningitis cases, 8 had S. pneumoniae isolated from CSF, and the remaining 14 had positive blood cultures associated with inflammatory CSF, or radiological & clinical features that were suggestive.

Following isolation of S. pneumoniae, the majority of patients received either amoxicillin (34%, 31/90) or ceftriaxone (32%, 29/90). Two-thirds of patients with pneumonia (62%, 31/50) received 10-14 days of therapy. Due to complications of IPD, 12% (11/90) of the overall cohort received >4 weeks of therapy.

59% (13/22) of suspected meningitis cases received steroids. Of these 13, 7 patients received them within the recommended 12-hour window post antibiotics, 4 patients received them after 12 hours and in 2 cases the timing was unclear.

Following review of clinical notes and discharge letters, there was no clear documentation of vaccine recommendation in 79% (71/90) of IPD cases.

Discussion: A number of important learning points were highlighted in this review. Adjunctive steroids in S. pneumoniae meningitis have been demonstrated to reduce mortality, hearing loss and improve short-term neurological sequelae, yet only a third of patients with suspected S. pneumoniae meningitis received them within the recommended 12 hour window. We suspect that this is because there is a reluctance to prescribe steroids until a diagnosis has been confirmed. In the revised hospital antimicrobial policy, we plan to include a prompt to consider the administration of steroids alongside the meningitis antimicrobial guidance. 

Approximately 30% of patients with IPD were not tested for HIV. Given the recognised association between HIV and IPD, this represents a potential missed opportunity for HIV diagnosis. Additionally, only a fifth of patients had a documented recommendation for pneumococcal vaccination post-discharge. This is a key secondary prevention strategy for managing pneumococcal infection which infection specialists need to disseminate to clinical teams as part of the routine management of patients with IPD.

121
Multi-disciplinary infective endocarditis ward round findings from University Hospitals Coventry and Warwickshire NHS Trust 2016-2018

Abstract - 121

Poster 121

Multi-disciplinary infective endocarditis ward round findings from University Hospitals Coventry and Warwickshire NHS Trust 2016-2018

Samantha Manyanga1, Harry Thynne1, Evangelos Vryonis2, Vjeran Cajic2, Fionnula Stalker3, Peter. E. Glennon4, Peter Munthali1,5
1Microbiology, University Hospitals Coventry and Warwickshire NHS Trust. 2Infectious Diseases, University Hospitals Coventry and Warwickshire NHS Trust. 3Pharmacy, University Hospitals Coventry and Warwickshire NHS Trust. 4Cardiology, University Hospitals Coventry and Warwickshire NHS Trust. 5University of Warwick Medical School

Introduction: Infective Endocarditis (IE) multidisciplinary team (MDT) ward round was established at University Hospitals Coventry and Warwickshire NHS Trust (UHCW) in 2016 and consists of a cardiology consultant, infectious disease consultants, a microbiologist, a pharmacist and ward specific doctors in charge of the patient’s care. It occurs weekly and ensures regular and consistent input, based on the ESC guidelines and clinical expertise, into individual patient care, particularly as this infection requires many weeks of treatment.

Methods: Cases who were reviewed by the IE MDT ward round between June 2016 and June 2018 were recorded on the microbiology IE database. They have been examined in relation to patient demographics (sex and age), the infection itself (valve affected, causative organism, treatment), trends such as seasonality of infections, whether patient is current or previous intravenous drug-user (IVDU); a known risk factor for IE and outcomes (mortality rates). Surgical intervention was recorded for all patients. Trends across all patients were noted and comparisons made between the IVDU patients and non-IVDU patients.

Results: Sixty-seven patients were recorded on the database, of which eighteen were IVDUs. A male predominance was found across all IE cases and lower patient age for IVDUs vs. non-IVDUs (mean <35 years old vs. >60years old), supporting similar endocarditis research findings. Our results showed the majority of our patients to have left-sided IE (mitral or aortic valve) whether IVDU or not. Right-sided IE (tricuspid) is less common in general (~10%) but the majority of these cases occur in IVDUs. Our results found 5 (28%) IVDU and 3 (7%) non-IVDU to have tricuspid valve IE: two of the IVDU cases were bivalvular infections with the mitral valve.

Of note there was an apparent seasonality among this particular population, with a substantial proportion occurring during winter.

IVDU mortality rates were 28% compared to 17% in non-IVDUs; it is unclear if this is linked to lower likelihood of surgery being performed on these patients (22% vs. 62% non-IVDU patients) with surgery known to improve mortality rates.

S.aureus was the most common organism isolated from blood cultures, with a noteworthy difference between IVDUs and non-IVDUs (72% vs 27%), followed by Streptococci and Enterococci. One-fifth of all cases were culture negative. We have a wide range of organisms isolated over the two year period including the less commonly found Aggregatibacter approphilus, Bartonella henslae and Granulicatella adiacens.

Discussion: In conclusion, our findings from our many IE cases at UHCW NHS Trust and large proportion of IVDU cases have shown certain characteristics and allowed us to compare the groups with regards to valve affected and seasonality. It raises interesting questions about the role of surgery on mortality rates for this group of patients. The ward round has ensured consistency in patient care with specialist contribution regarding important decisions such as treatment regime (particularly for some of the rarer bacteria we have isolated), duration, surgical intervention and appropriate follow-up.

122
Mitral and aortic valve infective endocarditis with cerebral complications in endocarditis patients at University Hospitals Coventry and Warwickshire NHS Trust

Abstract - 122

Poster 122

Mitral and aortic valve infective endocarditis with cerebral complications in endocarditis patients at University Hospitals Coventry and Warwickshire NHS Trust

Samantha Manyanga1, Harry Thynne1, Evangelos Vryonis2, Vjeran Cajic2, Peter E Glennon3, Peter Munthali4,1
1Microbiology, University Hospitals Coventry and Warwickshire NHS Trust. 2Infectious Diseases, University Hospitals Coventry and Warwickshire NHS Trust. 3Cardiology, University Hospitals Coventry and Warwickshire NHS Trust,. 4University of Warwick Medical School

Introduction: Infective endocarditis (IE) is a disease affecting the endocardial layer of the heart including the valves. Incidence is approximately 2-6 cases per 100,000 patient years, with intravenous drug users (IVDUs) being at an increased risk. At University Hospitals Coventry and Warwickshire NHS Trust, IVDUs account for a quarter of our endocarditis cases. There are distinct differences between the IVDU and non-IVDU population, as well as dissimilarities between common findings from IE patients and those from the patients seen on this IE multidisciplinary ward round at UHCW NHS Trust.

Methods: Cases reviewed by the weekly IE MDT ward round between June 2016 and June 2018 were recorded on the microbiology IE database. Data from these IE cases were examined with a specific focus on valve affected, causative organism and occurrence of cerebral complications. Mortality rates were noted and compared, along with number of patients who had valve repair or replacement surgery either during their admission or post IE treatment.

Results: Examination of cases from 2016-2018 demonstrated 67 IE cases (18 IVDUs) of which over half were mitral valve (two bivalvular infections of which the mitral valve was one). A further 28% were aortic in origin. There were only a handful of right-sided IE cases, affecting the tricuspid valve, 3 from non-IVDUs and 5 from IVDUs, with IVDU being an known risk factor for tricuspid valve IE.

S.aureus was the predominant organism accounting for 72% of cases, notably higher than for our non-IVDU patients (26%). Cerebral complications were found in 44% of IVDUs and 6.6% of non-IVDUs. In relation to the former, on occasion this was the reason for admission and in two instances related to the patient’s death. This may be due to the late presentation of these patients, with many having documented stigmata of IE on admission.

Discussion: In conclusion, our IVDU patients showed high rates of mitral and aortic valve IE. Many of them had significant cerebral complications, which had an effect on patient outcome and mortality. This was rarely seen in our non-IVDU population despite similar valves being infected. It highlighted the challenge of whether surgery should be performed to improve outcomes in an already high- risk patient group.

123
Clinical and laboratory characteristics of invasive Aeromonasinfection in a UK Teaching Hospital: a 20-year retrospective review

Abstract - 123

Poster 123

Clinical and laboratory characteristics of invasive Aeromonasinfection in a UK Teaching Hospital: a 20-year retrospective review

Swapna Motukupally1, Nelun Perera2
1Leicester Royal Infirmary, United Hospitals of Leicester. 2Leicester Royal Infirmary, University Hospitals of Leicester

Introduction: Aeromonas species are facultative anaerobic gram-negative bacilli that inhabit aquatic environments. It is a recognised cause of self-limiting diarrhoeal disease in humans. Invasive infections are increasingly recognised among patients with underlying diseases such as hepatobiliary disease, malignancies and other immunocompromised conditions. A. hydrophilais the most common species causing infections in humans. Increasing resistance to βlactam antibiotics is being recognised, posing therapeutic challenges.

Aim: This study was undertaken to define the epidemiology, clinical and laboratory features, treatment and clinical outcomes of laboratory confirmed invasive Aeromonas infections over 20 years in a large teaching hospital.

Methods: All patients who had Aeromonas species isolated in a blood culture or a sterile site fluid between February 1998 and January 2018 were retrieved from the pathology database. For each patient a comprehensive case note review was carried out and all relevant clinical, laboratory and antimicrobial prescribing information were collated. 

All blood cultures were processed using the BacT/ALERT® automated blood culture system, isolates were identified to the species level using either the API-20NE or by MALDI-TOF. The isolates that could not be identified were reported as Aeromonas species. Antimicrobial susceptibility testing was performed on all the isolates using BSAC standardised disc diffusion method and interpreted according to BSAC interpretative criteria. 

Results: A total of 44 patients had a laboratory confirmed invasive Aeromonas infection over the study period. 41 patients had Aeromonas isolated in blood cultures, remaining were from ascitic fluid (2) and peritoneal dialysis fluid (1).

The ages ranged from 3 to 89 years (average 68.5 years), 57% were over the age of 65 years, there were 24 male patients. Access to clinical records was available in 40 patients. Thirty patients had a recognised source of infection. The commonest underlying disease was hepatobiliary (19/40) and 18/19 had evidence of biliary obstruction. Two patients had Aeromonas isolated post-mortem following drowning.
A. hydrophila accounted for majority of the infections (27/44, 61.3%). Twenty-one (48%) patients had polymicrobial infections. Almost all isolates (97.5%) were resistant to amoxicillin, and two thirds (63%) were resistant to amoxicillin-clavulanic acid. Susceptibility to 2ndand 3rdgeneration cephalosporins, piperacillin-tazobactam, ciprofloxacin, gentamicin and carbapenems were over 90%.

Except for 7 patients who died before commencing treatment, the remaining were treated with an antibiotic to which the isolate was deemed susceptible. Eleven (27.5%) patients needed additional surgical intervention. A further 7 patients died after commencing treatment, that were attributed to their underlying disease and not due to Aeromonas infection. None had a recurrence of Aeromonas bacteraemia.

Discussion: This retrospective descriptive study conducted in a large teaching hospital in the United Kingdom demonstrated an average of 2 invasive Aeromonas infections / year (range 1-11). Despite the low numbers, invasive Aeromonas infections have been steadily increasing with 75% of infections occurring in the last decade. 
Most of the findings were similar to other published studies. However there were some notable differences. The very low numbers may reflect low exposure to the organism compared to developing countries. Most common source of infection was the hepatobiliary tract with underlying calculi, malignancy or cirrhosis. This may be due to referral of patients from the region for specialist care in a teaching hospital.

The current empiric recommendation for hepatobiliary infections in our hospital is amoxicillin-clavulanic acid. Since two thirds of our isolates were resistant to amoxicillin-clavulanic acid, physicians should be made aware to consider switching to a more appropriate antibiotic if a laboratory identification of invasive Aeromonas infection is available and avoid treatment failure with increased mortality and morbidity. 

124
9 years experience of Staphylococcus aureus bacteraemia in haemodialysis patients at a London Teaching Hospital

Abstract - 124

Poster 124

9 years experience of Staphylococcus aureus bacteraemia in haemodialysis patients at a London Teaching Hospital

Ann Sturdy1, Stephen Mepham1, Shara Palanivel1, Jennifer Cross2, Sophie Collier1
1Microbiology, Royal Free Hospital, London. 2UCL Centre for Nephrology, Royal Free, London

Introduction: Infection causes significant morbidity and mortality in haemodialysis patients, accounting for 20% of deaths in this cohort in the UK1. In most studies, Staphylococcus aureus is the single most common causative pathogen isolated2. We present 10 years of demographic, microbiological and outcome data for Staphylococcus aureus bacteraemias occurring in the adult haemodialysis population at a London teaching hospital and satellite dialysis units.

Methods: Data was collected prospectively using electronic patient and microbiology records. All patients receiving haemodialysis for end stage renal failure who were identified as having one or more blood cultures positive for Staphylococcus aureus between May 2009 and May 2018 were included.

Results: From May 2009 to May 2018 there were 328 sets of blood cultures positive for Staphylococcus aureus, in 175 separate episodes, from 149 haemodialysis patients. The number of episodes annually fell over the study period in both raw numbers (26 to 9) and rate per 100 patient years (4.20 to 1.26). 21 episodes (12%) were caused by methicillin resistant Staphylococcus aureus (MRSA), with a broadly decreasing incidence (from 0.58 to 0.14/100 patient years). 96 (64%) patients were male, with a mean age of 59.2 years (range 23 – 91). In 106 episodes (60%) dialysis was via a dialysis line and 64 (37%) via arteriovenous (AV) fistula (in 5 cases (3%) both methods were being used). Of the fistula access group, 22 (32%) were documented as using the button-hole access method. 59% of patients who had had nasal swabs taken for methicillin sensitive Staphylococcus aureus (MSSA) in the year before each episode had been positive at some point over the preceding year (97/164). A line was judged to be the source in 92 episodes (53%), a fistula in 23 (13%), a non-dialysis related source in 43 (25%) – the majority of which were soft tissue or bone and joint – and the source was unknown in 17 cases (9%). 30-day mortality was 10% (17/175), with a complication rate of 13% (22/175) – 50% being endocarditis – and 26 patients (17%) having at least 1 recurrent episode during the study period. Mortality was significantly higher in episodes of MRSA (5/21, 24%) than MSSA (12/154, 7%, p 0.02) bacteraemia. 

Discussion: Although incidence of both MSSA and MRSA bacteraemia decreased over the 10 year period, reaching the lowest incidence in the last complete year surveyed, mortality and other complication rates still remain significant, in line with other similar studies in the literature, with a significantly higher mortality in cases caused by MRSA3,4. The majority of infections were felt to be due to a haemodialysis catheter, however a significant proportion were attributed to AV fistulas (where the button-hole access technique may have significantly contributed) and to other sources, particularly soft tissue and bone – possibly partly reflecting the high proportion of diabetic patients within this group. The decreased incidence over the study period is likely multi-factorial, however a reduction in button-hole access of AV fistulas, nasal swabbing for MSSA and subsequent decolonisation, early review by the infection team and adherence to line insertion and care policies may all have contributed.

References

  1. Byrne C et al (2018) UK Renal Registry 20th Annual Report of the Renal Association NEPHRON. 139 (suppl1)
  2. Suzuki M et al (2016) Bacteremia in hemodialysis patients World J Nephrol. Nov 6; 5(6): 489-496
  3. Nielsen LH et al (2015) Risk and prognosis of Staphylococcus aureus bacteremia amoung individuals with and without end-stage renal disease: a Danish population-based cohort study BMC Infect Dis. 15:6
  4. Fitzgerald SF et al (2011) A 12-year review of Staphylococcus aureus bloodstream infections in haemodialysis patients: more work to be done J Hosp Infect. Nov;79(3):218-21

125
MSSA reduction strategies in Paediatric Cardiothoracic Transplant Unit, Newcastle upon Tyne Hospitals NHS Foundation Trust

Abstract - 125

Poster 125

MSSA reduction strategies in Paediatric Cardiothoracic Transplant Unit, Newcastle upon Tyne Hospitals NHS Foundation Trust

Simon Dewar, Hloniphani Mpofu, Julie Samuel, Joanna Lumb
The Newcastle Upon Tyne Hospitals NHS Trust

Introduction: Surgical site infection (SSI) is a common postoperative complication in children undergoing cardiothoracic surgery with SSI rates varying between 0.5-6% according to the National Nosocomial Infection surveillance System (US Figures – there are currently no UK wide surveillance for these types of infections). In the paediatric cardiothoracic department various infection prevention initiatives are in place including; MSSA screening programme for ventricular assisted devices followed by eradication, the introduction of Octenisan washes throughout inpatient stay, theatre surveillance audits and device management audits. However, concerns were raised of a perceived increase in sternal wound infections in the unit. Prospective microbiological surveillance was undertaken, along with a review of infection control practices, to help evaluate the problem and determine if any interventions were warranted.

Methods: During March 2018, using the microbiology laboratory information system, data was collected prospectively for all patients with a sample collected from sternotomy or thoracotomy wound. The Infection Prevention and Control team also conducted a review of clinical practice, theatre environmental audit, echo probe cleaning and introduced the SSI bundle. After the implementation of changes, repeat surveillance was carried out during July 2018.

Results: During March 2018, MSSA was isolated from 4 of 24 samples, which corresponded to 4 of 13 patients (31%). From the infection control review, key issues for improved practice included the appropriate decontamination of echo probes and the need to introductive local guidelines to manage post-operative wounds. A cardiothoracic surgical site infection care bundle was adopted which included guidance on good practice pre-operatively (including Octenisan washes), intra-operatively (including antibiotic prophylaxis, skin prep with 2% Chlorhexidine in 70% alcohol and glycaemia control) and post-operative wound care.

Follow up surveillance was undertaken after instigation of the care bundle. During July 2018, 31 relevant sternal samples were received from 11 patients. MSSA was isolated from 3 of 31 samples which corresponded to 2 of 11 patients (18%).

Discussion: Although small numbers in the study, the introduction of a surgical care bundle may have had a positive impact in reducing post-operative MSSA site infection and was felt to complement our existing strategies. Key areas that continue to be monitored and developed include; IV device management, review of Matching Michigan data, use of simulation in improving IPC practice and addressing competency for carers involved in direct wound care.

126
Audit of maternity and microbiology flowchart guidance for management of UTI in pregnancy

Abstract - 126

Poster 126

Audit of maternity and microbiology flowchart guidance for management of UTI in pregnancy

Simon Dewar, Munjusha Narayanan, Elizabeth Nevins, Racheal Roe, Philipa Marsden
The Newcastle Upon Tyne Hospitals NHS Trust

Introduction: The maternity department of Newcastle Hospitals have established local antenatal guidance concerning the diagnosis and treatment of urinary tract infection (UTI) in pregnancy and a local microbiology laboratory flowchart for authorisation of urine culture results in pregnancy was also established. The aim of the study was to evaluate adherence with local guidance and to determine any areas for improved practice if indicated from findings.

Methods: Using the microbiology laboratory information system, data was collected retrospectively for all patients from whom midstream urine (MSU) sample was sent from inpatient or outpatient antenatal services from February to May 2017. Information concerning symptoms, urine dipstick results, antibiotic treatment and follow up of MSU results was gathered from the medical notes, and microbiology reporting of samples was also examined. 

Results: 121 patients had a MSU sent during the study period. These 121 women had a total of 253 MSUs sent throughout their pregnancies. Out of 253 MSUs 49 (19.4 %) were indicated in accordance to local guidance, 163 MSUs (64.4 %) were not indicated and 41 (16.2 %) had incomplete documentation to evaluate if they were indicated or not. 64 (25.3 %) of the MSU samples sent were positive and 189 (74.7%) were negative for bacterial culture. 14 out of 121 (11.6%) patients received antibiotics with a total of 33 antibiotic episodes. Of the 33 antibiotic episodes the microbiological reporting adhered to local guidance in 25 cases (75.8%), in the 8 cases of non-adherence this was either due to the lack of interpretive comment (5 cases) or inappropriate release of antibiotic sensitives (3 cases).

Discussion: The study showed a large numbers of inappropriate MSUs being sent in pregnancy, this has a negative impact on antenatal and microbiology services. Increased education, familiarity and compliance with guidelines are required to improve this.

127
Brucella post-exposure management in laboratory settings: a systematic review

Abstract - 127

Poster 127

Brucella post-exposure management in laboratory settings: a systematic review

Astrid Tulsson1, Tom Wingfield1,2, Nick Beeching1
1Liverpool School of Tropical Medicine. 2University of Liverpool

Introduction: Brucellosis is an important global zoonosis with an estimated 500,000 cases annually. Laboratory-acquired brucellosis (LAB) is one of the most frequent laboratory-acquired diseases with an incidence of 2% in laboratory workers worldwide. Despite the attack rate following laboratory exposure being reported as 30-100%, there are no standardized international guidelines on post-exposure management. The purpose of this study was to provide an updated evaluation of Brucella post-exposure management in laboratory settings by characterizing LAB, evaluating the reporting of Brucella exposures, and identifying challenges to post-exposure management. 

Methods: A systematic literature review on Brucella laboratory exposure was carried out using databases including PubMed, MEDLINE, Web of Science, Google Scholar and the Belgian Biosafety Server. To allow comparison of findings with those of the last major review of LAB published in 2008 (Traxler et al), case-reports published in English or Scandinavian from 2008 onwards describing laboratory Brucella exposures with sufficient data to determine if post-exposure prophylaxis (PEP) was received, risk classification (either stated in paper or deduced using the CDC guidelines for Brucella exposure to divide into high, low or no risk), outcome (LAB or no LAB) and total numbers exposed were included. Records on health care personnel exposures in a non-laboratory setting or case-reports of individual LAB cases were excluded. Variables describing exposure, characteristics of those exposed, setting, intervention and outcome were extracted. Data was analysed descriptively and attack rates calculated. 

Results: Of a total of 29 papers reviewed, seven papers met eligibility criteria and were included in the study. Five papers were from Brucella non-endemic countries and two from endemic countries. A total of 18 exposure events (defined as a laboratory exposure incident where one or several people are exposed to Brucella) and 275 exposures (defined as the number of people exposed to Brucella) were described. Of these, five of the exposed developed LAB. The definition of LAB varied among the reports. Of 77/151 (51%) people with high-risk exposure received PEP and 74/151 (49%) did not. Attack rates were similar regardless of PEP and are lower than those previously reported by Traxler et al, which were 0% and 73% for those who received versus did not receive PEP, respectively. Incompletion of recommended serological follow-up interval was common and so was insufficient data in many of the records. The major sources of Brucella of the index cases were local wild animals, laboratory exposures, and a combination of travel and consumption of unpasteurized dairy products. 

Discussion: In this post-2008 review, LAB attack rates were lower than those previously reported by Traxler et al. Varying definitions of LAB were frequent and presented challenges to comparison of results. Wild local animals should be considered a major source of Brucella in non-endemic countries. The results support processing and handling of unknown gram-negative coccobacilli in class II biosafety cabinets until high-risk pathogens have been excluded. Symptom surveillance after laboratory exposure can capture brucellosis in early stages and is thus recommended. Thorough information about brucellosis in exposed personnel is important to prevent delays in seeking of medical attention. Finally, there is a need for standardized reporting on Brucella exposures to enable more streamlined retrospective studies on LAB.

128
Prevalence, microbiological and epidemiological characterisation of clinical meningitis in children. Experience from Infectious Diseases tertiary referral centre, Egypt

Abstract - 128

Poster 128

Prevalence, microbiological and epidemiological characterisation of clinical meningitis in children. Experience from Infectious Diseases tertiary referral centre, Egypt

Asmaa Farrag1, Noha Elsakka2, Bayoumi Gharib3, Elham Elsakka3
1Alexandria University Children’s Hospital, Alexandria, Egypt. 2NHS Grampian, Aberdeen. 3Alexandria Faculty of Medicine, Alexandria, Egypt

Introduction: Meningitis is a major cause of public health concern in developing countries, with high rates of morbidity and mortality. Early diagnosis and management helps improve outcome of disease. However, it also represents a resource and financial challenge in countries with limited resources. In this study we have undertaken a retrospective analysis of the microbiological and epidemiological characterisation of paediatrics patients admitted with meningitis to Alexandria Fever hospital (AFH), Egypt. The aim is to identify the epidemiology of meningitis in local population, identify current clinical practice and highlight areas for improvement. In addition, to indicate best place to direct resources in order to improve disease outcome in the local population.

Methods: The study included retrospective analysis of data from 100 consecutive cases admitted to AFH during the period of January-December 2017 with a diagnosis of meningitis. Clinical data and lab results were collected from medical records after obtaining ethical approval. Patients aged 1 month -15 years were included. Exclusion criteria included patients with recurrent meningitis due to structural CNS defects, immunodeficient and asymptomatic patients. 

Results: During the study period, meningitis was more prevalent in males (75%) and was more in the younger age group of 1 month – 1 year (53%), and 31% were among 1 -5 years old children. The disease was more common in rural areas (62%), which is associated with lower socioeconomic status.

The main route of patient admission was referral from Alexandria University teaching hospital (AUTH) representing 80% of the admissions and 17% were admitted directly through AFH A&E department. The main presentation on admission was fever (96%), toxic look (59%) vomiting (49%), convulsions (36%) and DIC (30%). Only 3 patients had rash on presentation.

CSF biochemical analysis was performed to all patients, while culture was performed only on patients admitted directly to AFH A&E (17%). Of the 17 CSF cultures, 15 samples were positive with a growth of Neisseria Meningitidis (6), streptococcus pneumonia (5), haemophilus influenzae (4) and staph aureus (1). No virology PCR was performed on any of the CSF samples, however 6 of the patients were diagnosed as viral meningitis on clinical grounds. Isolation and infection control precautions were observed for the 6 patients with positive N. Meningitidis culture. 

All patients received an initial empiric treatment of IV Benzyl Penicillin and ceftriaxone. Acyclovir was added if viral meningitis was clinically suspected. Hospital length of stay averaged between 14-15 days for 68% of the patients, and 5% of the patients stayed in the hospital for more than 21 days. Complications included convulsions (10%), hydrocephalus (1%) and squint (1%). There was favourable outcome for 94% of cases with no mortalities, however 6% of the cases did not attend to their follow up appointments.

Discussion: Robust process for diagnosis and evaluation of cases of meningitis is required. More focus is needed in rural areas with limited resources, this includes improving healthcare facilities in addition to health education and raising public awareness of the disease. 

Implementation of CSF microbiology culture for all cases of suspected meningitis as part of the initial assessment process is essential. A clear documentation of the culture and sensitivity results in patient’s notes is important to help guide the treatment process and rationalise the antibiotic use. 

There is minimal work don’t to identify the prevalence of viral meningitis in the local community. It is recommended that focus and resources would be directed to introducing a routine viral PCR on cases of suspected meningitis. 

129
The clinical management of malignant otitis externa – a review of the last 20 cases presenting to NHS Lothian and development of a joint management protocol with ENT

Abstract - 129

Poster 129

The clinical management of malignant otitis externa – a review of the last 20 cases presenting to NHS Lothian and development of a joint management protocol with ENT

Naomi Henderson1, Rebecca Sutherland2
1Lothian Infection Service, Royal Infirmary Edinburgh. 2Lothian Infection Service, Western General Hospital, Edinburgh

Introduction: Malignant Otitis Externa (MOE) is a progressive infection of the auditory canal which extends to the temporal bone and adjacent structures. If left untreated this process has a high mortality rate. The incidence of this condition is increasing.

We performed a review of the last 5 years of cases presenting to NHS Lothian to ascertain whether there were any trends that would enable us to identify at risk patient groups, optimise referral pathways and advise on antibiotic treatment.

Methods: The NHS Lothian electronic patient data base (TRAK) was searched for all patients with a diagnosis of MOE between 2013 and 2018. Recurrent cases were excluded from the analysis.

Results: 18 confirmed cases and 2 probable cases were captured in the data search. All cases were admitted through ENT. The number of cases per year have increased from 1 to an average of 5 per year.

The median age was 78 (range 28-94). There was a male preponderance (14:6 M:F). 13 out of 20 patients had diabetes mellitus, 17 out of 20 patients had a co morbidity (immunosuppression including diabetes mellitus, malignancy and recent high dose steroid use). 4 out of 20 patients had undergone prior ENT surgery (3 mastoidectomy and 1 canaloplasty).

The causative organism isolated from a superficial swab of the ear canal in this cohort was predominantly Pseudomonas aeruginosa in 12/20 cases (60%). Reassuringly only 1 Pseudomonal isolate was ciprofloxacin resistant. The second most common pathogen was Staphylococcus aureus 5/20 (25%) with one methicillin resistant isolate. Fungal/yeast causes of MOE in this cohort were rare. In the 15 patients who had a deep ENT sample only 2 of these were sent to microbiology (all were sent to pathology).

Presenting complaints common to this patient cohort were severe otalgia (20/20), otorrheoa (19/20) and presence of granulation tissue in the ear canal (15/20). 5 out of 20 patients presented with, or subsequently developed, a facial nerve palsy. All patients had a CT scan during their initial diagnosis, all of which were consistent with MOE, with 13/20 showing extensive involvement of bones out with the external auditory canal. CRP was not discriminative in identifying patients with severe MOE.

The length of time from ENT initial review to time of diagnosis ranged from 1 day to greater than 100 days (median 6 days). Delay to diagnosis in this cohort was due to atypical presentation, and delay to initial cross sectional imaging.

Factors associated with disease progression despite initial treatment course were: extensive bony changes on initial imaging, initial oral antibiotic choice without antipseudomonal activity and a short antibiotic treatment course (2 weeks or less). The length of course of iv therapy ranged from 1 to 22 weeks (median 6 weeks). Of the 16 patients whose symptoms resolved, 9 were given early empirical systemic antipseudomonal antibiotic therapy and 14 were given appropriate antimicrobial therapy once the diagnosis was established.

Discussion: This audit review shows a range of treatment strategies for treating MOE. This is in part due to heterogeneity of the presentation of the infection. There are difficulties in predicting which cases will develop significant complications as a result of invasive disease, and which will resolve with early directed antibiotic therapy and local debridement. The results of the review have enabled discussions between the Departments of ENT and Departments of Infection to form a joint protocol in the management of this infection. The protocol focuses on early sampling (with samples sent to both microbiology and pathology), early referral from primary care, early imaging, clinical infection review on the ENT wards and subsequent joint care on the wards and in the OPD.

130
Infective endocarditis caused by Panton-Valentine leukocidin producing methicillin-susceptible Staphylococcus aureus: first reported case in United Kingdom

Abstract - 130

Poster 130

Infective endocarditis caused by Panton-Valentine leukocidin producing methicillin-susceptible Staphylococcus aureus: first reported case in United Kingdom

Chathuri Gunasekera1,2, Ian Gould1,3, Marjory Greig1
1Aberdeen Royal Infirmary. 2University of Colombo, Sri Lanka. 3University of Aberdeen

Introduction: Staphylococcus aureus (SA) is a well known cause of soft tissue infections, bacteraemia and infective endocarditis (IE). In industrialised countries SA is a leading cause of IE. Panton-Valentine leukocidin (PVL) is a bacteriophage-associated, bi-component cytotoxin produced by community acquired SA strains, contributing to its virulence. PVL production of SA is often linked to acute and recurrent soft tissue infections. IE caused by PVL producing MSSA has never been reported in the United Kingdom. Here we present a case of infective endocarditis caused by PVL producing MSSA in a diabetic patient who had heart failure due to aortic stenosis.

Methods

Case report: A 63 year old female diagnosed with diabetes mellitus, decompensated heart failure due to severe aortic stenosis and atrial fibrillation with left ventricular systolic dysfunction was admitted to the Cardiothoracic Surgical ward of Aberdeen Royal Infirmary for a routine aortic valve replacement surgery. In a recent Cardiology admission, she was managed with diuretics and has developed cellulitis around a cannula site for which flucloxacillin was given. Also, her oral hypoglycameic was revised due to elevated sugar levels. A preoperative transthoracic echocardiography showed calcified aortic valve leaflets but no obvious vegetations. Coronary angiography revealed no significant coronary artery disease.

On admission to the cardiothoracic ward patient was afebrile and not very unwell. Her inflammatory markers were normal. During surgery the native aortic valve with calcified leaflets were shown to bear numerous vegetations on the ventricular side of each leaflet. All leaflets were excised and sent to Microbiology for culture. Aortic valve replacement was done with a bioprosthesis.

The aortic valve culture yielded SA upon enrichment. Direct microscopy and primary culture were negative. This SA isolate was methicillin-sensitive and penicillin-resistant. Isolate was sent to reference laboratory to check for genes responsible for production of toxins and only pvl PCR was positive. The spa-type of the isolate was t1149 and inferred clonal complex was CC45. Two sets of blood cultures taken post operatively yielded no growth.

Patient was subsequently treated for infective endocarditis. Intravenous flucloxacillin and oral rifampicin were given for 6 weeks. Two postoperative trans-thoracic echocardiographies showed no obvious vegetations and the tissue atrial valve was functioning well. On discharge she was well, apyrexial and had a sinus rhythm; chest wound was well healed and her blood sugar levels and inflammatory markers were within normal.

Discussion: SA is a leading cause of bacterial IE, its resulting mortality remaining high despite the improved diagnostics and therapeutics. This may be attributed to the numerous virulence factors of this organism, which includes an array of toxins. In IE, due to various epidemiological factors such as increasing elderly population, intravenous drug usage, prosthetic valves etc. a microbiological shift from streptococci to staphylococci can be appreciated.

The diagnosis of IE was not straightforward in this case. In fact it was a diagnosis arrived at surgery. A cannula site infection leading to bacteraemia (undetected) which resulted in IE may be a work-up in this case. Here, the Duke criteria which are specific but not very sensitive could not be fulfilled. A transoesophageal echocardiography would have helped more.

PVL, a bi-component cytotoxin, is associated with community acquisition and is involved in evading/ destroying host defenses, hence a major virulence factor of SA. It usually gives rise to acute skin/soft tissue infections which may be recurrent. The pathological role of PVL -producing MSSA in IE is not yet fully understood. To the best of our knowledge, this is the first report of IE caused by PVL producing MSSA, in the western world.

131
Prosthetic vascular graft infections at a UK Tertiary Vascular Centre

Abstract - 131

Poster 131

Prosthetic vascular graft infections at a UK Tertiary Vascular Centre

Stephanie Harris1, Ishapreet Kaur2, Janice Tsui3, Sophie Collier3
1King’s College Hospital, London. 2Royal College of Surgeons in Ireland, Dublin, Ireland. 3Royal Free Hospital, London

Introduction: Prosthetic vascular graft infections (PVGIs) are rare; thought to affect 0.5 – 1% of grafts, but these infections can result in significant morbidity and have a high risk mortality. Unlike many chronic infections there are not as yet clear guidelines about the best management of PVGIs and so there can be considerable heterogenisity in the management of them. This study looked at collecting data on the characteristics of presentation, microbiological findings, medical and surgical management and the outcomes of patients with PVGIs managed at a London tertiary care vascular surgery centre.

Methods: Patients with vascular graft infections were identified from a search of our outpatient parenteral antibiotic treatment (OPAT) database and data was retrospectively collected on demographics and risk factors, primary surgical procedure and antimicrobial prophylaxis used at that time, the index admission where the PVGI was identified, microbiological findings, antimicrobial therapy and surgical management and outcomes at completion of treatment and mortality at 30 days and 90 days from presentation with infection. Patients who had presented with their infection episode from 1st May 2013 until 1st May 2018 were included. Patients with haemodialysis graft infections, inflammatory aortitis and those where the majority of treatment was at another hospital were excluded. Results were analysed using chi-squared and t-tests where appropriate.

Results: 83 patients were identified of which 24 were excluded and 59 were included in the study. Average age of the patients was 69 years (median 70, SD 11) and 72.9% were male. Co-morbidity burden as measured by the Charlson Co-morbidity Index was 6 (median 6, SD 2.43) and 69.5% were smokers or ex- smokers. The most frequent graft sites involved were femoral (42.4%), aortic (30.5%) and axillo-femoral grafts (15.3%). 8.5% were primary mycotic aneurysms with prosthetic material placed in an infected field and managed subsequently as a vascular graft infection. 32.2 % presented within 3 months of surgery. The majority of patients (61%) presented with suspicion of sepsis for the index admission where the graft infection was identified. In this cohort 61% of patients were treated for a least 12 weeks and 10.2% of patients continued lifelong antimicrobial suppression treatment. There was no difference in success of treatment (80.6% v 69.6%, p = 0.94). or survival (p) between those treated for at least 12 weeks and those treated for less then 12 weeks. A likely organism was identified in 84.6 % of cases. Gram positive organisms were identified in 42.4%, gram negative in 40.7%, anaerobes in 18.6% % and yeasts in 3.4%. Overall 90 day survival from presentation with infection was 89.8% in this cohort. At completion of the initial treatment course 76.3% had an outcome of cured or improved.

Discussion: This study shows that patients at this centre often present at some time from the initial procedure and may not present with obvious infection, which can cause difficultly in recognition and subsequent management and that PVGIs should be borne in mind when patients present with surgical problems related to their grafts. Microbiologically there was a variety of organisms isolated and many patients had a polymicrobial infection which highlights the importance of getting adequate samples to guide therapy. There was no significant difference with success of treatment with those treated with longer or shorter lengths of treatment, although it is difficult to draw conclusions given the small numbers and heterogenicity of the population. In this study the 90 day survival was very high, and this may represent that patients who were most unwell were never referred to the OPAT service and so not included. PVGIs are infrequent complications and monitoring the findings and outcomes can help inform future practice. 

132
An unusual cause of osteomyelitis in a 17-year-old

Abstract - 132

Poster 132

An unusual cause of osteomyelitis in a 17-year-old

Victoria Shivji1, Shiu-Shing Soo2, Susan Snape3
1Infectious Diseases, Nottingham City Hospital. 2Clinical Microbiology, QMC, Nottingham. 3OPAT and Clinical Microbiology, Nottingham

Introduction: A 17-year-old male with a past medical history comprising only of childhood asthma, who took no regular medications, presented to the Trauma & Orthopaedic outpatient clinic in August 2015 after GP referral. He had a one year history of intermittent swelling and pain over the proximal aspect of his left lower leg. Imaging of the area revealed a proximal tibial lesion with an unusual appearance, and later a PET-CT demonstrated foci of increased metabolic activity over the left proximal tibia and distal femur. He was believed to have multifocal osteomyelitis. He had had no known overt trauma to the area, although was a keen footballer and had sustained many minor injuries in the past. He had never had any surgery to the area. He hadn’t been unwell with any other illness prior to his initial symptoms developing. He had never travelled abroad, and he and his family were from the UK. He denied any sexual contact. His only animal contact was that of a pet lizard that had died several months previously. He was not knowingly immunosuppressed; his HIV test was negative. He lived with his mother and stepfather, who were both well.

Methods: Initial bone biopsies of the site were taken (x 5) intraoperatively, and sent for Microbiological culture and Histological examination. A copious amount of pus superficial to the lesion was drained. Histology revealed a diffuse inflammatory infiltrate, with no features of malignancy. Somewhat surprisingly, microbiological culture was negative after 48 hours. Following surgery, he developed a sinus tract at the point of the incision. Four weeks later he underwent extensive debridement of the proximal tibia, distal femur and sinus tract. Extensive amounts of reaccumulated pus were again drained. Pus was cleared out from the bony cortex and dead tissue debrided. Multiple samples were sent to microbiology. Bone cavities were packed with antibiotic-impregnated cement beads, and wound coverage obtained with a gastrocnemius flap and a split skin graft.

Deep solid specimens were homogenised using Ballotini beads. All samples were inoculated onto blood, CLED, and fastidious anaerobic agar and incubated appropriately. At the 48-hour plate read, growth was seen on all plates from the majority of samples. 2-3mm moist grey colonies were seen on blood agar (with smaller variants on FAA), and colourless non-lactose fermenting colonies described on CLED. Gram stain revealed a Coliform-like gram-negative rod.

Discussion: This is the first known incidence to the literature of a human infection with this organism. We believe the likely source to be the patient’s previously owned lizard. The organism tested broadly sensitive, and the patient was treated successfully via OPAT with 3 weeks of IV ceftriaxone followed by 3 weeks of oral ciprofloxacin. It reminds us to keep an open mind when considering microbiological causative agents of osteomyelitis.

133
Diabetic foot osteomyelitis treatment: a service review of success rates involving patients requiring intravenous antibiotics for resolution of their osteomyelitis and wound healing. 4 factors were specifically studied: location of osteomyelitis, presence/absence of peripheral vascular disease, failure of prior oral antibiotics and number and type of organisms cultured

Abstract - 133

Poster 133

Diabetic foot osteomyelitis treatment: a service review of success rates involving patients requiring intravenous antibiotics for resolution of their osteomyelitis and wound healing. 4 factors were specifically studied: location of osteomyelitis, presence/absence of peripheral vascular disease, failure of prior oral antibiotics and number and type of organisms cultured

Susan Snape, Hannah Bond, Lisa Metcalf, Ravikanth Gouni
Nottingham University Hospitals NHS Trust

Introduction: The majority of patients with diabetic foot osteomyelitis (OM) receive oral (po) antibiotics with good outcomes. Those who either fail po antibiotics, are systemically unwell or unable to take po antibiotics often require intravenous (IV) antibiotics. We wanted to better inform patients regarding prognosis and hence the primary aim of our service review was to determine outcomes at 12 months of wound healing and perceived resolution of OM without any further surgery, IV antibiotics or amputation. The secondary aim was to examine outcomes in the presence of a variety of factors including: location of OM; the presence of peripheral arterial disease (PAD); previous failure of oral antibiotics; single or multiple organisms isolated from samples and nature of organisms isolated.

Methods: A retrospective service review was conducted on all patients admitted to Nottingham University Hospitals (NUH) with diabetic foot OM that received po as outpatients or IV antibiotic therapy either in hospital, post-dialysis or through the outpatient parenteral antibiotic therapy (OPAT) service in 2015 and 2016. Patients were excluded that previously had IV treatment for a related OM. Patients were identified by examining hospital records. All patients received ongoing review, wound debridement, off-loading and vascular interventions where appropriate. Hospital notes, letters and photos, along with radiological investigations were used to assess the outcome measures at 12 months. Presence of PAD was determined by duplex scan results.

Results: 266 patients with OM were identified who were treated with po antibiotics alone. 164 of the 266 (61%) patients had successful resolution of OM with no further surgery or antibiotics. This outcome is in line with published literature on the subject and gives a baseline with which to compare patients who required IV antibiotics. 98 were treated unsuccessfully and of these 70 went onto IV treatment; 22 had amputation but no IVs; 6 had surgical debridement plus local antibiotics but not IV’s. 4 patients died.  Microbiological data on this population was not captured.

139 patients with OM were identified who had had IV antibiotics and returned for follow up over a 12 month period. They had a mean age 64 (30-92), male 74%, type 2 diabetes 84%, diabetes duration over 10 years 84%. Patients were treated with 6-12 weeks of IV antibiotics. 

119 of 139 patients had initial surgery. Antibiotic choice was guided following bone samples in 86 patients; tissue in 35 patients and wound aspirate in 6 patients.

88 of 139 patients (63.3%) had theraputic success with OM resolved requiring no further surgery or IV antibiotics at 12 months. 56 of 139 patients (47.5%) had wound healing at 12 months.

The highest success rates were in patients with a forefoot or digital OM with no PAD, without Gram-negative organisms isolated and no prior antibiotic treatment for OM; in this group there was a therapeutic success rate of 92.3% (26 patients). The patients with the poorest outcomes had PAD, Gram-negative organisms present and had received previous oral treatment for OM and had a success rate of 41.2% (17 patients).

Discussion: The success rates of treatment for diabetic foot OM varied substantially depending on the situation. It was expected that patients with hindfoot osteomyelitis or PAD would have worse outcomes. The finding that the prior use of oral antibiotics was also associated with a lower success rate is not unexpected. However, the presence of Gram-negative bacteria from deep samples also appears to be an indicator of a reduced chance of successful treatment. The combination of information from this service review will be used in clinical consultations to give patients more specific advice on their prognosis based on their individual circumstances.

134
Can whole genome sequencing and maldi-tof identify vancomycin resistant enterococci transmission events in haematology inpatients?

Abstract - 134

Poster 134

Can whole genome sequencing and maldi-tof identify vancomycin resistant enterococci transmission events in haematology inpatients?

Sarah Coleman1, Emma Boldock2, David Partridge1
1Sheffield Teaching Hospital Foundation Trust. 2Royal United Hospitals Bath NHS Foundation Trust, Bath

Introduction: Healthcare associated infections caused by multi-drug resistant enterococci cause significant morbidity and mortality. Over the past 5 years, an increased rate of colonization and infections by vancomycin-resistant enterococci (VRE) has been observed at Sheffield Teaching Hospitals Foundation Trust (STHFT) particularly in haematology patients, despite the implementation of stringent infection control practices. Current typing methods for monitoring and investigation of potential outbreak strains include multi-locus sequence typing (MLST) and pulsed-field gel electrophoresis (PFGE), both of which are time consuming and laboratory intensive. This limits clinical utility during potential outbreak situations as they can only be used as a retrospective tool rather than to clinically inform at the time. With the rapidly reducing cost of whole genome sequencing (WGS) and therefore improved accessibility for laboratories, WGS could play an increasingly important role in infection control, producing a faster and more detailed return of results.

Methods: PFGE and WGS were compared in order to assess the discriminatory ability of novel methods against existing methodology (PFGE). Prospective isolates collected from a cohort of haematology inpatients at STHFT between June and November 2017 were then analysed by WGS in order to assess the benefits of WGS for identifying cross-transmission events. Further analysis was carried out through pairing of WGS and patient movement data in order to identify specific transmission events within STHFT.

Results: Results show WGS to be comparable to current typing methodology and to be a powerful and reliable method for identifying strain relatedness of VRE isolates. When paired with patient movement data, WGS is shown to be a highly effective tool for identifying cross-transmission events. Interrogation of both WGS data and patient movement data identified the presence of 5 distinct clusters of isolates in haematology in-patients during this study and, identified 2 possible transmission events on 2 separate wards.

Discussion: With the enormous reductions in cost and increased accessibility of WGS, this once novel methodology is now much more widely available, not only for research purposes, but within clinical laboratories, heralding a new genomic era for clinical medicine. At STHFT, WGS provided an exciting opportunity for the investigation of colonisation and infection by VRE, particularly in the vulnerable cohort of haematology patients. 

WGS, particularly when paired with patient movement data is shown to be a highly powerful tool for the assessment of transmission events and in the surveillance of high risk wards. Furthermore, the wealth of information created from WGS enables in-depth analysis of data for further confirmation of results in order to clinically inform with confidence. Information such as SNP numbers is highly valuable in confirming strain relatedness alongside phylogenetic trees. These results suggest that, with further validation, WGS could be clinically utilised in STHFT. Not only would this enable full in-house assessment of VRE strains within STHFT, negating the requirement for referral to reference laboratories, but it could also be used for analysis of other organisms of importance in our trust such as MRSA.

Considerations as to the clinical requirement for such in depth analysis in STHFT must be made in light of additional costs and training required for implementation into a routine laboratory particularly when current methodology is shown to be comparable. Introduction of such methodology may be considered to be more relevant upon development of infection surveillance of high risk wards, the improvement of systems to enable easy collection of patient movement data and in the move to in-house typing for multiple organisms of significance.

Major Sponsors