Abstract
Monday
1CP Improving checklist compliance through an inter-professional simulation-based training programme
R Tuffin, O Ross-Smith, D Hogg, S Cook and M Issott
2CP The development of a new nationally accredited training programme in pointof- care ultrasound for intensive care medicine
A Wong and G Barker
3CP A standardised reporting proforma for point-of-care lung ultrasound in ICM
A Wong and G Barker
4CP CPR: Are we as good as we think we are?
J Brand, S Law and I Hunter
5CP Experience within the district general hospital setting of a bedside electronic monitoring system
R Amin, V Patle and K Murray
6CP Abstract Withdrawn
7CP Hypophosphataemia during continuous renal replacement therapy in the general ICU
C Kaye
8CP Radiation exposure in the critical care unit
L Martin and P Hersey
9CP The ‘level 1’ pathway – lessons learnt at the critical care–medicine interface
G Chapman, G Ashfield, M Oppenheimer and H Chin
10CP Sensitivity of NPSA guidance for confirmation of placement of NG feeding tubes in a critical care population
P Hersey, F Ritchie, S Wadman and A Boyle
11CP Improved compliance with lung protective ventilation using an electronic alert system in a large UK teaching hospital: A retrospective cohort analysis
N Pawley, MJ Heap and S Webber
12CP The Bottom Line
D Chambler, D Slessor and S Mathieu
13CP Emergency pre-intubation checklist, Ipswich General Hospital
L Induruwage and N Berry
14CP Critical care transfer audit: Do you know what is in your bag?
L Coleman, E Lightbody and M Wilson
15CP The acute care team path to halving intensive care unit mortality
J Comara, J McCann, T Mason and the Acute Care Team
16CP Intensive care medicine trainees' experience of percutaneous dilatational tracheostomy insertion: A UK wide survey
KE Orr, C Russell and C Philpott
17CP Outcome of patients admitted to intensive care unit with acute on chronic liver disease
PN Gunasekera, S Colley and D Ventour
18CP Planned general surgical admission to critical care. . . Who made it, who did not and what impact did the management location have on patient outcomes?
A Garland, S Winn, J Paddle and M Feldman
19CP Determining the impact of an electronic early warning score systems in a district general hospital
JB Simon, RJ Ferguson and KT Murray
20CP Filter lifespan during continuous renal replacement therapy: Audit of a revised protocol
A Sisson, M O'Connor, T Hussein and JR Prowle
21CP An audit of chest drain management in Wexham Park Hospital ICU
B Millette and D Sedgewick
22CP Management of post-intubation tracheobronchial rupture
D Whitmore and J Rodrigues
23CP Whipps Cross university hospital ICU, are we ‘on the right trach’? A multidisciplinary audit approach
A de Gea
24CP Tracheostomy airway management by junior doctors on the intensive care unit
N Simpson and N Arora
25CP Checklist lead intubations on the intensive care unit
J Hobson, K Savjani and N Robin
26CP The management of two cases of severe calcium channel antagonist overdose: The role of hyperinsulinemic euglycemia therapy
R Schofield and L Wilson
27CP An audit of tracheostomy practice and complications in a large central teaching trust: Six-month data
SL Linford, A Sharman and K Somarakis
28CP High-flow nasal oxygen therapy (AIRVO2TM) Chesterfield experience
SL Taylor and CJ Goldson
29CP Does a nurse-led non-invasive ventilation service improve patient outcomes?
H Mainman, S Chambers and H Curtis
30CP One-year survival data for patients aged over 80 admitted to a general intensive care unit
N Singh and M Street
31CP Colours for key parts of tracheostomy tubes can impact on patient safety perceptions and experiences An international survey
BA McGrath, S Wallace, J Lynch, and E Ward
32CP Experience with ward-based high-flow nasal oxygen in adults with acute hypoxaemic non-hypercapnic respiratory failure in a tertiary centre
E Price, P Featherstone, J Martin, S Petty, S Precious and M Trivedi
33CP Review of HIV-positive patients admitted to Critical Care at a district general hospital in London
H Manukonda and P Gandre
34CP ITU nurses confidence assisting in airway situations and knowledge about difficult airway equipment
J Paige, L Conway, C Bassford and R Townsend
35CP A nationwide survey of consultant working practices on intensive care units in the United Kingdom
LC Evans and O Boyd
36CP The role of the occupational therapist in a major trauma critical care team: Findings from current clinical practice
R Gaunt, S Brady and R Moses
37CP A nationwide survey of trainee cover on intensive care units
LC Evans and O Boyd
38CP A feasibility study to investigate the effectiveness of using a Mechanical In-Exsufflator device for lung volume recruitment via positive pressure breathing as an alternative to the Mark 7 Respirator intermittent positive pressure device
R Moses
39CP Is Intensive care a safe environment for emergency tracheal intubation?
SS Kumar, C Christau and S Hutchinson
40CP Telemedicine in critical care echocardiography: Proof of concept
A Constantine, Y Mebrate, M Alam, G Kanaganayagam and S Price
41CP Tracheostomy practice: Changing goals?
F Donaldson, A Cormack, C Grant, W Barr and T Szakmany
42CP Measuring patient acuity in critical care CMCCN dependency scoring tool
S Clarke and M Kynaston
43CP Observational study assessing the suitability of patient referral to the critical care unit of a district general hospital
P Dhillon, J Shak, D Smith and K Turner
44CP Pulmonary complications of intravenous drug abuse: A case series
C Lynch, C Britton-Jones and E Powell
Tuesday
1CP Abstract not available
2CP ‘NICE’Practice on the ICU: Integrating NICE CG 83 into the intensive care unit daily nursing routine
N Dawson, P Taglucop, Y Li, S Etter, S Frondoza, C Torre, S Bhandari, J Beddow, DVincent, S John, K Broderick, N Findlay and M Farrow
3CP Developing a critical illness management and basic life support course in a resource poor setting
J Barnes
4CP Observational cohort study of patients admitted to a specialist centre for weaning and rehabilitation
A Warren, T Sanctuary, D Camilleri, PB Murphy, BC Creagh-Brown and N Hart
5CP Therapeutic hypothermia or targeted temperature management after cardiac arrest? An audit of current practice in intensive care units in the North West of England
A Power and R Markham
6CP Out-of-hospital cardiac arrests and emergency coronary angiography
D Slessor and A Ashton
7CP E-CPR in children: Are survival and neurological outcome predictable?
I Mennes, T Vanassche, P Vanelderen, P Skippen and A Pitfield
8CP The impact of focused intensive care echocardiography on clinical management of critically ill patients
DP Hall, H Jordan, SR Alam and MA Gillies
9CP An audit of temperature management in comatose survivors of cardiac arrest
B Millette and N Feely
10CP ICNARC defined ‘Post Unit hospital deaths’is not a reliable quality indicator
H Scott, P Tylerand and D Higgins
11CP Assessment of the impact of a new ventilator in achieving compliance with lung protective ventilation
S Brooke, R Hunt and P MacNaughton
12CP A novel method of evaluating critical care fellow performance compared to their peers in a busy multifaceted fellowship using data envelopment analysis
Avinash B Kumar and Vikram Tiwari
13CP Differing family behaviours in organ donation: A direct comparison between donations after brain death versus donation after circulatory death
N Kalsi, D Harvey, D Gardiner, C Buss, L Hogg and R Ballington
14CP A PATH to improve outcomes following ROSC after cardiac arrest?
N Madhavan, D Horner and J Mcleavey
15CP Variation in height measurement leads to excess tidal volume ventilation: A study of measurement techniques
FA Amey, MG Povey and CR Bassford
16CP Early rehabilitation of burns specific patients in critical care compared to critical care population: An audit of current practice and limitations
F Howroyd, T Lea and D McWilliams
17CP Room for improvement: An audit of intra-operative mechanical ventilation in theatres at a large district general hospital
A Rutter and R Thompson
18CP Ultrasound technology: Should it be a part of hospital induction for trainees?
K Khoo and S Nandalan
19CP Matching Warrington: Incidence of ventilator-associated pneumonia
T Bower, V Vanek, Z Qazzafi, J Little and J McCann
20CP Adults with incapacity: An intervention to improve practice
K Graham and H Tyler
21CP Improving training in end-of-life care and organ donation
DJ Haley, M Thomas and C Booth
22CP Quality Improvement ITU Patients and Relatives Booklet
S Bruemmer Smith, S Tun and L Jackman
23CP The mean arterial pressure to vasopressor dose ratio: An objective metric to assess the effect of cardiovascular interventions
DJ Stubbs, G Singleton and NM Broughton
24CP Glucose control in critical care: Compliance with hyperglycaemia guideline and effect on patient outcome
L Beard, E Beard and M Reay
25CP Evidence supporting the use of NG bridles on the ICU: Cost and clinical considerations
C Banks and A Marsh
26CP Evaluating the effectiveness of communication in ventilator-dependant tracheostomy patients utilising above cuff vocalisation The ICU Functional Communication Scale
S Wallace, J Lynch, L Nicholson, M Wilson, R Purcell and BA McGrath
27CP Early mobilisation and the barriers in intensive care units across Scotland
LG Salisbury, ME Harrold, SARWebb and GT Allison
28CP Seizures in out-of-hospital cardiac arrest survivors A poor prognostic indicator
SS Kumar, C Downes, T Leary and S Hutchinson
29CP Noise levels in ITU: A comparison between two critical care units
J Blackburn, R Hayes, V Ragothaman and N Arora
30CP An audit of discharge practice across the Greater Manchester Critical Care Network
AJ Wilson, J Ebbs, A Stevens and J Eddleston
31CP Developing a national framework for critical care nurse education
M Kynaston, A Berry and A Himsworth
32CP Meeting nutritional targets in intensive care patients: An audit of current practice
R Gale, S Harwood, P Turner and A Iyer
33CP Real-time oxygen consumption monitoring A new technology to improve critical care
PA Robbins, DP O'Neill, L Ciaffoni, J Couper, G Hancock and GAD Ritchie
34CP Clinical effectiveness project: Risk stratification using post-operative troponin
J Peters, R Sexton and M Thomas
35CP Implementing a novel gentamicin guideline on the ICU: A quality improvement project
D Palmer, T Clarke and J Roberts
36CP Palliative and intensive care A clinical paradox?
A Cran, C Davis, C Guyer, E Murphy, K Haynes and R Chambers
37CP Optimising antibiotic stewardship in suspected ICU-acquired pneumonia in two large intensive care units: A prospective audit
G Wojcik, TH Craven and TS Walsh
38CP Using high-frequency oscillation ventilation on a complex patient with refractory hypoxaemia on conventionalventilation: A case study
E Mc Sweeney
39CP Noise levels and correlation with activity and time in a district general intensive care unit
Sean J Keating, Sarah Beavis, Jeremy Windsor and Anand Padmakumar
40CP Intermittent glucose monitoring in the critically ill: A pilot survey of practice in the United Kingdom
AJ Devanesan and S Saha
41CP Restoration of speech and swallowing in dysphagic spinal cord injured patients receiving mechanical ventilation via tracheostomy – A case series
S Morgan and J McRae
42CP Multidisciplinary in-situ simulation training in Critical Care: Six-month review
EC Baker, A Beane, H Hewitt, L McWhirter and ML Kong
43CP The development of a specialist critical care outreach physiotherapist to facilitate the implementation of NICE CG83 in a regional tertiary major trauma centre and enhance rehabilitation post critical illness
S Bunting, D Brooks and R Moses
44CP Survey of patients with sickle cell disease admitted to critical care department at North Middlesex University Hospital, London
A Zuhair, O Ralley, LValco, K Malpica and P Gandre
Wednesday
1CP Abstract not available
2CP The use of enoximone in management of anaphylaxis complicated by labetolol use
S Hayhoe, V Navapurkar and A Conway Morris
3CP A pilot project to assess the feasibility and effectiveness of a mechanically ventilated manikin to simulate intensive care emergencies for junior trainees
L Christie, P Patel and M Hayes
4CP Rapid improvements in quality and safety following introductions of tracheostomy multi-disciplinary team ward rounds at a tertiary hospital
J Lynch, SWallace, B Bonvento, G Sharp, S Pieri-Davies and BA McGrath
5CP An audit and quality improvement project: Evaluating the provision of intensive care support for the critically unwell patient in the cardiac catheterisation suite undergoing primary percutaneous coronary intervention in a tertiary 24/7 centre
C Christou, C Malcolm, C Lopez-Soto and M Naik
6CP VV-ECCO2R in a patient with a traumatic brain injury and hypercarbic respiratory failure
D Kelly and B Paul
Monday 7 December 2015
1CP Improving checklist compliance through an inter-professional simulation-based training programme
R Tuffin, O Ross-Smith, D Hogg, S Cook, and M Issott
Intensive Care Unit, Harrogate and District NHS Foundation Trust, Harrogate, UK
The WHO Surgical Checklist has demonstrated the impact that checklists can have upon improving patient outcomes,1 particularly in complex clinical situations that are heavily reliant upon effective teamwork and clear lines of communication. The safe intubation of critically ill patients upon the intensive care unit (ICU) represents one such complex clinical scenario and has been shown to be an area where complications frequently occur. Our unit developed a checklist to aid team members when carrying out this task but compliance was found to be low when a survey was sent out to staff.
We therefore devised an inter-professional simulation training programme to address this issue. Importantly, this training occurred ‘in situ’ upon the ICU thus creating an ‘authentic’ learning environment for team members. After each session, a ‘debrief’ would be held to allow participants to reflect upon how the process might be improved by use of a checklist.
Throughout this process, we hoped that team members would reflect upon their practice and recognise the utility of an Emergency Rapid Sequence Induction (RSI) Checklist. We also hoped that the sessions would create a ripple effect within the unit where colleagues would informally discuss their views with others (who may not have attended the course yet) and that this might positively impact upon their clinical practice as well.
Over the 12 months that the course has been running, we have raised awareness and compliance with the Emergency RSI Checklist. Prior to the course, only 40% of staff would consider using the checklist, but after 12 months of the course starting, this had increased to 73%. Of note, 56% of these team members had not yet attended the simulation sessions, and we believe this represents evidence of the ripple effect we hoped for upon setting up the course.
We feel these results have been achieved partly through attendance upon the course but also through encouraging a community of practice to develop with our ICU that allows sharing of new ideas that will ultimately improve patient care.
Reference
2CP The development of a new nationally accredited training programme in point-of-care ultrasound for intensive care medicine
A Wong, G Barker
Oxford Critical Care Ultrasound Learning and Research (OCCULAR) Group Institute(s), Oxford University Hospitals NHS Trust, Adult Intensive Care Unit, Oxford, UK
Although championed by enthusiasts, the use of ultrasound in intensive care medicine has lagged behind that of other specialties including Emergency Medicine. This lack of formal training structure and programme is a recurring issue across the rest of Europe. It poses the question of whether scans performed have been appropriately performed and reported, and whether there exists proper clinical governance. The Intensive Care Society (UK) recently introduced the Core Ultrasound Skills in Intensive Care (CUSIC) in order to provide a formal and robust training structure to attain these competencies. We describe the experience of a university, tertiary-level hospital in developing a training programme to achieve these goals.
Through a modular system, with each module lasting three months, the training pathways comprised
Initial theoretical and practical training Supervised practice until competence demonstrated in acquiring and saving images Mentored practice with completion of log book demonstrating knowledge of an appropriate range of pathology Completion of competency assessments within the range of practice – heart, lung, abdominal and vascular Triggered assessment to investigate a clinical question
The varies modules ensure that the clinician is armed with the breadth of knowledge and skill to deal with the range of clinical situations that he/she is likely to encounter within intensive care medicine. Robust clinical governance policies are maintained through formalised working practice with all stakeholders including radiology departments. This requires a considerable degree of preparation and discussion prior to the commencement of the programme.
A formal and structured training programme ensures the highest level of competency-based training. Learning objectives and outcomes are defined from the onset for both the trainer and trainee. A modular system allows for a degree of flexibility and ensures that a balance is achieved between service-provision and training/learning periods.
The CUSIC accreditation is one of the few nationally recognised competency-based training programmes available in Europe. Through collaborations with other bodies including those from other specialties, the delivery of point-of-care ultrasound training maintains a high standard which will ultimately result in improved patient care. It is crucial that appropriate protocols are formalised to ensure clinical governance. We believe that our model for delivering point-of-care ultrasound training, although undoubtedly in its infancy and will continue to evolve, forms a good starting point for other institutes who wish to develop their own programme.
3CP A standardised reporting proforma for point-of-care lung ultrasound in ICM
A Wong, G Barker
Oxford Critical Care Ultrasound Learning and Research (OCCULAR), Oxford University Hospitals NHS Trust, Adult Intensive Care Unit, Oxford, UK
Point-of-care lung ultrasound (US) in intensive care medicine whilst fairly new was once the remit of enthusiasts. As more clinicians embrace its usefulness to evaluate and manage both the acute and more chronic dyspnoeic patient on the intensive care unit, there is a need for standardised reports. Standardisation of the report would allow consistency of information reported, audit and research projects and most importantly, maintain a high level of clinical governance. The Core Ultrasound Skills in Intensive Care (CUSIC) accreditation programme has just been introduced in the UK.1 The lung component is based on the BLU protocol2 by Liechtenstein and colleagues, ensuring a formalised, competency-based training programme.
There are no standardised national reporting proformas available in the UK or Europe. We developed our own using the template provided in the CUSIC accreditation pack. This proforma can then be freely disseminated online.
It is important that the proforma is clinical relevant and yet user friendly. Suggestions and experiences of colleagues who regularly performed and reported on lung US were used to guide its design. Several versions of the proforma were developed, and each was piloted under real world conditions on an adult intensive care unit in a tertiary-level, teaching hospital. It is a living document and will continue to evolve as our practice and knowledge in lung US continues to improve.
The final reporting proforma contains all the necessary information required when reporting lung US findings but is arranged in a more visually intuitive layout and includes a decision-making tree on the reverse side as an aide memoire.
We believe that the current version of the reporting proforma is much more user friendly whilst maintaining clinical usefulness. It is freely available on our website in the spirit of Free and Open Access to Medical Education (FOAMed). The standardised format means that audit and quality improvement projects can be undertaken with minimal interferences. Collaborative working and projects can also be conducted across several institutions. Other future plans include the development of an online document which would impart other benefit including cloud backup. We welcome suggestions from interested colleagues in order to further improve this document.
References
4CP CPR: Are we as good as we think we are?
J Brand, S Law, I Hunter
Department of Anaesthetics & Intensive Care Medicine, James Cook University Hospital, Middlesbrough, UK
Despite significant advances in resuscitation, survival from cardiac arrest remains poor.1 Numerous studies demonstrate that the performance of high-quality chest compressions as part of effective cardiopulmonary resuscitation (CPR) is vital in improving patient outcomes.2 Despite this, studies continually reveal the occurrence of poor quality compressions in both training and clinical situations.2 The aim of our study was to assess any potential difference between the perception and reality of resuscitation providers when performing basic CPR and whether feedback devices improve performance.
During an unrelated training session, anonymous questionnaires were given to a random selection of anaesthetics and intensive care medicine (ICM) trainees and operating department practitioners (ODPs), which were completed immediately. In addition to resuscitation training background, the questions aimed to identify how each participant perceived their ability to maintain the correct rate and depth of chest compressions during a 2-min cycle of CPR. The participants were then asked to complete a 2-min CPR cycle using a 3G SimMan (Laerdal Medical, UK) whilst their performance was recorded and compared with National standards.3 Immediate feedback was provided to the participants, and their performance was repeated using the inbuilt CPR feedback device in real time for a further 2-min cycle.
Forty-two questionnaires were completed, consisting of 18 trainees (CT1-ST7) and 24 ODPs. All trainees (18, 100%) held valid advanced life support (ALS) certification, whilst 16 (67%) ODPs held valid immediate life support (ILS) certification. All trainees (18, 100%) believed that their rate of CPR was in line with National standards,3 whilst 22 (92%) ODPs thought the same. In reality, only 7 (39%) trainees and 13 (54%) ODPs achieved a compression rate in line with National standards.3 Regarding depth, 17 (94%) trainees and 20 (83%) ODPs believed they achieve the correct compression depth. Our results revealed only three (17%) trainees and four (17%) ODPs achieved the correct depth of chest compression during CPR. Of the study population, 42 (100%) participants adopted the correct CPR hand position and 15 (83%) trainees and 22 (92%) ODPs allowed for the correct recoil and release of hands during compression attempts. All participants (42, 100%) achieved the correct compression rate and depth when performing CPR using a real-time feedback device.
This study demonstrates that despite adequate resuscitation training and course attendance, the perceived ability to perform effective CPR is substantially different from the reality. Unfortunately this is in keeping with other studies that note a decline in performance following course completion.4 A significant limitation of our study was because of the SimMan as participants were performing CPR on a chest with a different resistance as compared to a human and this may have altered our results relating to compression depth, this has directed us toward further study in real resuscitation scenarios. We propose that regular resuscitation updates become mandatory for all staff, and we are also planning to establish a Trust-based simulation ‘resuscitation update’ course for all theatre and ICU staff in addition to ensuring a Trust wide introduction of real-time feedback devices onto our resuscitation trolleys.
References
5CP Experience within the district general hospital setting of a bedside electronic monitoring system
R Amin, V Patle, K Murray
East Sussex Healthcare Trust, East Sussex, UK
Our Trust has recently introduced an electronic bedside monitoring system to replace the paper-based charts previously used to record patient observations.
Nursing staff using handheld devices ensure that observations are digitally recorded and National Early Warning Scores (NEWS) are automatically calculated, allowing accurate, consistent results and clear presentation of trends. These data are available at the bedside, on mobile devices and also remotely via the Trust computer system. This gives clinicians clear data both at, and remote from, the patient bedside and enables early escalation in accordance with Royal College of Physicians and NICE guidance for deteriorating patients.
The aims were to:
Assess engagement with the electronic bedside monitoring system by the junior doctors; and Assess the awareness amongst junior doctors of the Trust escalation policy for deteriorating patients, including patient review triggered by abnormal NEWS scores and the mobilisation of a Medical Emergency Team.
An electronic survey was designed and distributed to junior doctors working on wards. We received 50 completed responses.
Ninety % (45 of 50) use the electronic monitoring system daily on the ward and find the system easier to interpret than a paper-based system. The same number of trainees believed the system had improved patient care.
Seventy-four percent (37 of 50) would use a personal handheld device if offered, that would notify doctors with changing observations of patients under their care.
The majority of trainees, 64% (32 of 50) are aware of the existence of a trust escalation policy; however, 36% (18 of 50) were unaware that a NEWS of 5 required escalation.
Forty-eight percent (24 of 50) of the doctors were unaware that an Emergency Team call should be triggered at a NEWS of 9. This call would initiate senior, multi-specialty assessment and escalation to critical care if required.
Following the introduction of the bedside monitoring device, there has been clear engagement with the electronic system with 90% of junior doctors using the devices on a daily basis for ward rounds and handover of care. They found the system easy to interpret and believed that it had improved patient care. Junior doctors also indicated that they would be willing to have a personal handheld device so that they could receive updates electronically about changes in their patients’ care.
There is evidence, however, of lack of awareness of the Trust’s Escalation Policy for the deteriorating patient, and we are working on improving the standard of clinical care by starting an educational programme to improve awareness of both the Escalation Policy and the triggers for the Medical Emergency Teams. In doing so, we expect long term to see earlier escalation of the deteriorating patient to critical care and better outcomes for those patients.
References
7CP Hypophosphataemia during continuous renal replacement therapy in the general ICU
C Kaye
Department of Anaesthetics, Aberdeen Royal Infirmary, Aberdeen, Scotland
References
8CP Radiation exposure in the critical care unit
L Martin1 and P Hersey2
1Northern Schools of Anaesthesia and Intensive Care Medicine, Newcastle-Upon-Tyne, UK
2Integrated Critical Care Unit, City Hospitals Sunderland NHSFT, Sunderland, UK
Ionising radiation is frequently utilised for imaging in critical care; however, exposure to ionising radiation increases the risk of a patient subsequently developing malignancy. This retrospective observational service improvement project was designed to quantify patient exposure to ionising radiation within our facility.
The case notes and radiology archiving systems were reviewed for 100 consecutive patients in an 18-bed UK general integrated critical care unit (ICCU). The sample data included the admitting speciality, ICCU length of stay and details of any diagnostic imaging had been undertaken. All diagnostic imaging was included if it had occurred on or after the date of ICCU admission. Reference values for radiation exposure against radiological imaging were obtained, and data were collated and processed using Microsoft Excel (2013).
All 100 patients remained on the ICCU for one day or more; 30 of the 100 patients remained on the unit at day 5 and eight patients at day 10. Sixty-two patients were exposed to ionising radiation during their ICCU stay. The highest rates of exposure were observed between days 1 and 4.
The mean cumulative exposure observed was 4.1 mSv, with a range of 0–37.3 mSv. Seventy-six patients were not exposed to radiation in excess of the annual UK background exposure. Twelve patients were exposed to levels greater than the annual acceptable level for nuclear industry employees in the UK, but all patients had exposure lower than the American accepted threshold for nuclear industry workers.1,2
For each of the first six days of admission, there was no statistically significant difference in levels of exposure between medical and surgical patient subgroups (p > 0.05). Beyond six days, the sample size was insufficient for comparison, with an estimated 95% CI of the median exceeding the range of the data. Cumulative radiation exposure since day of admission demonstrated an increased trend of the median; however, a reducing sample size increased the spread of data (Figure 1).
Standard reporting form.
It would appear from our observation that the increased risk of malignancy to our patients from radiation exposure in ICCU is small; even the highest cumulative dose would see an increase in absolute risk of malignancy of only 0.186%.
Regardless of the exposure levels seen in this sample, it is essential that health professionals remain mindful of the risk posed to patients when deciding on an imaging modality, seeking alternative means of providing the answer to a clinical question where possible.
Radiation dose by days since admission.
References
9CP The ‘level 1’ pathway – lessons learnt at the critical care–medicine interface
G Chapman, G Ashfield, M Oppenheimer and H Chin
Department of Critical Care, Milton Keynes General Hospital, Milton Keynes, UK
The interface between A&E, medical wards and critical care can provide unique challenges for healthcare practitioners. The ‘level 1’ pathway is a pilot scheme introduced to select wards in Milton Keynes General Hospital in 2014. This pathway seeks to improve clinical care through greater awareness of acutely unwell patients, greater access to senior medical input and improved liaison with outreach services.
Documents pertaining to all medical patients acutely admitted to the Department of Critical Care (DoCC) in Milton Keynes University Hospital NHS Trust over a three-month period in 2013, and again in 2014, were retrospectively examined to ascertain the timing and nature of movements between the aforementioned areas, diagnoses, morbidity, mortality and markers of good clinical care and communication.
Patient demographics were comparable for the 2013 (n = 73) and 2014 (n = 68) cohorts. DoCC admissions occurred predominantly out of hours (73.5% admitted between 17:00 and 08:00). In the 24-h period preceding a critical care admission, 70% of patients had been reviewed by the Medical Consultant responsible for their care. Between 2013 and 2014, the average delay between referral to critical care and DoCC admission was reduced from 3.5 h to 2.2 h; 96% of patients referred to critical care in 2014 had a documented critical care review before DoCC admission (increased from 64% in 2013). In 2013, 93% of DoCC admissions from medical wards ‘triggered’ on two or more parameters of the modified early warning score (MEWS) before admission. The corresponding figure in 2014 was 58.5%.
Whilst the number of patients on the ‘level 1’ pathway was minimal, the approach to the medical–critical care interface appears to have improved. A greater proportion of patients admitted to DoCC are reviewed before admission, and there is some evidence that critical care input is requested at an earlier stage.
10CP Sensitivity of NPSA guidance for confirmation of placement of NG feeding tubes in a critical care population
P Hersey, F Ritchie, S Wadman and A Boyle
Integrated Critical Care Unit, Sunderland Royal Hospital, Sunderland, UK
The administration of feed or drugs through a misplaced nasogastric tube (NGT) is a never event.1 The National Patient Safety Agency (NPSA) has produced guidance related to NGT insertion to reduce the risk of this never event occurring.2
We conducted a three-cycle audit in an 18-bed general integrated critical care unit (ICCU) to test the applicability of this guidance in the critical care environment and to investigate local adherence.
The first audit was conducted in September–November 2013, with subsequent audits at 6 and 12 months. There was an ongoing process of staff education and as an outcome of the first audit cycle the NGT in use within the ICCU was changed. None of the NGT insertions sampled were conducted by or had their position confirmed by an author.
A total of 143 insertions were sampled, all of which were confirmed as being sited correctly.
Key outcomes by audit cycle.
CXR: chest X-ray.
Where pH testing was documented (n = 80), we demonstrated a sensitivity for correct site confirmation of only 27.5%. This low sensitivity was due to an inability to aspirate the NGT in 43/80 (54%) and a pH ≥ 5 in 15/80 (19%) of cases.
In the first audit sample, radiological criteria were only fully met in 20/38 (53%) of the first CXR taken to confirm placement. After changing the NGT being used, this increased in the second audit to 24/25 (96%) first CXR. In the third sample, radiological criteria were met in 100% of first CXR.
This audit cycle has shown that a significant number of CXRs can be avoided if NPSA guidance is followed. We have also demonstrated that testing aspirate pH as a first-line method to confirm correct placement within critical care is associated with a low sensitivity. Therefore, the majority of patients will still require a CXR. We have also highlighted the importance of ensuring the NGT in use is easily visible on CXR.
References
11CP Improved compliance with lung protective ventilation using an electronic alert system in a large UK teaching hospital: A retrospective cohort analysis
N Pawley, MJ Heap and S Webber
Critical Care Department, Sheffield Teaching Hospital, Sheffield, UK
Lung protective ventilation strategies (LPVS) have become increasingly accepted as best practice for invasive ventilation of patients with acute respiratory distress syndrome (ARDS).1 Previous audits of delivered tidal volumes in patients with ARDS receiving pressure-controlled mechanical ventilation on our critical care unit had shown poor compliance with LPVS. The aim of this study was to assess compliance with LPVS after the introduction of tidal volume alert warnings on a computerised clinical information system and an accompanying educational programme.
Retrospective data from 2012 and 2013–2014 were compared from patients with severe ARDS as defined by the Berlin criteria.2 Hourly tidal volume data were recorded in electronic charts (Metavision, iMDsoft), and the time spent within different tidal volume ranges calculated. Data were collected from the development of moderate ARDS until the PaO2/FiO2 ratio was >27 kPa or the patient died. Statistical significance was assessed using Student’s t-test and the chi-squared test. In 2013, tidal volume alert warnings were added to the electronic observation charts (Metavision, iMDsoft) when tidal volumes exceeded 8.0 ml kg−1 predicted body weight (PBW), accompanied by education of medical and nursing staff.
There were 28 and 31 patients in the 2012 and 2013–2014 cohorts, respectively. The pre- and post-intervention mean (SD) tidal volumes were 8.9 (3.1) vs. 7.7 (2.0) ml kg−1 (p < 0.001). The pre- and post-intervention proportion of tidal volumes >8.0 ml kg−1 were 64.7% and 35.7%, respectively (p < 0.001) (Table 1). Mortality was 57% in the 2012 cohort and 61% in the 2013–14 cohort.
The introduction of alert warnings on a computerised clinical information system when tidal volumes exceeded >8 ml kg−1 PBW, accompanied by an educational programme, was associated with a significant improvement in compliance with lung protective ventilation.
References
12CP The Bottom Line
D Chambler, D Slessor and S Mathieu
Queen Alexandra Hospital, Portsmouth, UK
There is a large time delay between the publication of new evidence and the use of this evidence in clinical practice. This delay in knowledge translation leads to patients receiving less than optimal treatment or even receiving treatments that result in harm. It also results in unnecessary expenditure for healthcare organisations. The objective of this project is to shorten this knowledge translation.
A website was developed with the aim to summarise and review trials that are relevant to Intensive Care Medicine: www.wessexics.com/The_Bottom_Line/. As new trials are reviewed, clinicians are alerted via email, a RSS feed or twitter feed: @WICSBottomLine.
The website, The Bottom Line, was launched in May 2014. Since then, ∼100 trials have been reviewed and @WICSBottomLine has ∼2600 followers on twitter. The website has had ∼150,000 page views. The email and twitter alerts have enabled clinicians to be informed about the latest evidence. By providing a succinct summary with the important ‘Bottom Line’, clinicians are able to quickly assimilate the evidence. We have been informed by our readers how this has changed their practice. The Bottom Line website has been recommended by clinicians from all over the world. The immediate past president of the Society of Critical Care Medicine has recommended The Bottom Line to ‘all of us in critical care medicine who seek disciplined thinking.’ Scott Weingart (USA) included The Bottom Line in his top 10 of 2014. ‘Life in the Fast Lane’ (Australia) regularly recommends our reviews. St Emlyn’s (Manchester) included The Bottom Line in their highlights of 2014.
The Bottom Line website has enabled thousands of clinicians from all over the world to be quickly informed of important trials that are relevant to their practice. This has led to a shorter length of time before the evidence from these trials has been used in clinical practice. By improving clinicians’ ability to practice evidenced-based medicine, the quality of patient care has improved.
13CP Emergency pre-intubation checklist, Ipswich General Hospital
L Induruwage and N Berry
Ipswich Hospitals NHS Trust, Suffolk, UK
Approximately 1.1 million intubations are performed each year1 and although most proceed without complication on the rare occasion that there is a complication this may be catastrophic in nature.
Patients who are intubated and ventilated in these settings often have not had a detailed anaesthetic assessment or peri-anaesthetic optimisation. These patients often have abnormal physiology due to the disease states that they present with and the need for rapid/emergency induction and intubation is a riskier undertaking.
The recent fourth National Audit Project (NAP4) in to airway complications showed that of all the airway complications that occurred in the study period one in four was likely to be in the intensive care or emergency department.1 This is a large proportion considering the much smaller number of intubations in comparison to the operating theatres. The NAP4 authors made the following comments in response to the findings – analysis of the cases identified gaps in care that included: poor identification of at-risk patients, poor or incomplete planning, inadequate provision of skilled staff and equipment to manage these events successfully, delayed recognition of events and failed rescue due to lack of or failure of interpretation of capnography. The project findings suggest avoidable deaths due to airway complications occur in intensive care unit and the Emergency department.1
The aim was to develop and implement a pre-intubation checklist to improve patient safety for those in need of emergency intubation in the Intensive Care or Emergency department. Our checklist ensures that all necessary equipment is available, staff are briefed and ready to intubate a patient and deal with any subsequent complications in a safe and timely manner.
We believe our checklist improves the safety of emergency intubation in three key ways:
It provides an aide memoir for staff to facilitate collection and setup of all the equipment necessary. The checklist is formally read out during pre-oxygenation of a patient in a challenge and response manner to check everything required is present and functioning. The checklist covers allocation of roles, plan for failure should this occur, who to call for help (should it be needed) and immediate post-procedure needs.
The checklist allows not only for confirmation of equipment and roles but also open dialogue within the team, the chance to clarify points and the opportunity for people to voice concerns. This makes for good CRM (crew resource management) and reduces the impact of human factors.
It is very cheap to implement and could be implemented in multiple hospitals over a very short period of time with minimal training and therefore has the potential to benefit a vast number of patients in a short period of time.
Emergency intubation checklist.
References
14CP Critical care transfer audit: Do you know what is in your bag?
L Coleman, E Lightbody and M Wilson
Manchester Royal Infirmary Critical Care Units, Manchester, UK
Transfer of a patient is a fundamental skill in critical care for patient care and patient safety. The critical care transfer bags in use were poorly organised, lacked equipment which may have been required and contained unsafe or unusable equipment. The aims were to perform a baseline audit of the contents of the critical care transfer bag for the high dependency and intensive care unit and obtain user views on the bags. Standards were derived from the Intensive Care Society,1 the NHS North West Critical Care Network2 and AAGBI3 guidelines:
One hundred percent of equipment should be checked daily and before transfer. One hundred percent of escorting personnel are familiar with the operation of all equipment.
The contents of the critical care bags were listed and a survey of the nurses in charge was performed on responsibility and timing of the bag check. User views were sought following two timed scenarios for 14 doctors where airway and circulation equipment was required to be obtained from the bags.
Five nurses in charge were surveyed on the responsibility and timing of the bag check; each nurse gave a different answer. There was no known checking procedure. The time taken to find airway and circulation equipment ranged between 67–356 s and 55–190 s, respectively. User views after the scenarios included improvement required in content organisation, familiarity with equipment and the type of equipment available. In response to this, a new transfer bag was developed with education on the bag given during the induction process to all new doctors with a detailed check occurring every Thursday and a seal check every day. The time taken to find airway equipment had a reduced range of 31–93 s and circulation equipment to 25–75 s on repeat timed scenarios.
A variety of doctors are performing critical care transfers with different backgrounds and transfer experience. The problems encountered with the transfer bags were related to content organisation, familiarity with equipment and the equipment available; this was addressed with a new transfer bag with sections for airway, breathing and circulation with a formal checking procedure. Further work is to be targeted towards a standardised regional transfer bag.
References
15CP The acute care team path to halving intensive care unit mortality
J Comara, J McCann, T Mason and the Acute Care Team
Warrington and Halton Hospitals Foundation Trust, Cheshire, UK
In 2011, we carried out an audit of unplanned admissions admitted from the wards to Warrington intensive care unit (ICU) over a four-month period. With our mortality coming back at an unnerving 58%, the need for some form of change became imperative. In 2013, Warrington launched the acute care team (ACT), a seven-day Critical Care Outreach Service. In 2014, the ACT upgraded its Modified-Early-Warning scoring system to the National Early Warning scoring system (NEWS) and introduced a two-tier nurse-led Medical Emergency Team (MET).Two important achievements but what about outcomes? The Darzi Report1 and the Keogh Report2 have challenged us to look at two key areas: increasing quality and reducing mortality. In response, our audit has set out to answer two important questions. Has Warrington’s acute care team improved quality and have we reduced mortality?
Our ACT consists of two critical care outreach nurses, four trainee assistant practitioners (TAPs), four experienced intensive care nurses, six Night Nurse Practitioners (NNPs) and the ACT Matron. Our study looked at the two-year period 2013–2014 following its introduction. We chose four mortality indicators: Trust Mortality (SHMI+HSMR), ICNARC ICU SMR and Trust weekend Mortality and four quality indicators: Critical Incidents (deteriorating patient), NCAA cardiac arrests, Unplanned ward to ICU admissions/mortality and number of Unplanned ward to ICU level 3 admissions.
During 2013–2014, our results showed that following the introduction of the ACT, Warrington Hospital’s SHMI and HSMR decreased. The ICU SMR has gone down (0.9 at present). Weekend mortality was difficult to determine. The number of critical incidents involving the deteriorating patient has reduced. We are one of the highest Trust reporters (40/1000 bed days). Unfortunately, the number of cardiac arrests has increased. Unplanned admissions increased but significantly the mortality in this group went from 58% in the 2011 audit to 27% by the end of 2014. Level 3 admissions decreased.
There is little evidence that rapid response systems (RRS) alter anything other than reducing cardiac arrest numbers. Disappointingly our cardiac arrests numbers increased but the numbers are small. On the other measures of quality and mortality, we have demonstrated improvement. Of course, there may be additional variables influencing these statistics. However, we are confident that our ACT has played a major role in this success. Our RRS, the ACT, with its emphasis on prevention as much as reaction has encouraged early referral of deteriorating patients allowing them to be admitted to ICU in a better state of health, giving them a greater chance of recovery. After all, why does ICU have to be a place where the patient has to wait until they are at their sickest point before considering admission? We recommend our RRS model – The Acute Care Team.
References
16CP Intensive care medicine trainees’ experience of percutaneous dilatational tracheostomy insertion: A UK wide survey
KE Orr, C Russell and C Philpott
Intensive Care Unit, Ulster Hospital, Dundonald, Northern Ireland
Achieving competence in the safe insertion of percutaneous dilatational tracheostomy (PDT) is considered a key skill for UK Intensive Care Medicine (ICM) trainees within the syllabus produced by the Faculty of Intensive Care Medicine.1 Under the current curriculum, at the completion of training, trainees are expected to have reached the level of independent (consultant) practice with expert knowledge and skills.1
Anecdotal evidence suggested that current training in PDT insertion might not adequately prepare all ICM trainees to reach the required competency level on completion of training. An electronic survey was sent to all UK ICM trainees and they were asked questions regarding the numbers of PDTs they had undertaken with both direct and indirect supervision, as well as their self perceived level of competence.
Out of approximately 400 trainees, there were 119 (30%) complete responses, 41 of those from trainees at ST7 level or above. 71% of respondents were training under the previous “joint” training scheme using a different curriculum which although described features of competent performance, did not require the achievement of expert level prior to training completion.2
A total of 46% of trainees had undertaken at least 10 PDT insertions in their career with direct supervision, falling to 10% with indirect supervision where a consultant was not immediately at the bed space. Amongst ST7+ trainees, the figures improved to 60% and 25%.
Trainees were asked about their self-perceived competency levels. 86% felt competent with direct supervision and 44% with indirect supervision. Again amongst ST7+ trainees, the figures were higher at 95% and 69%. Clearly some trainees felt competent to insert a PDT with indirect supervision despite not actually having done so previously.
Trainees were asked how many PDT insertions they felt were required during training to gain competence. With direct supervision, 13.74 (SD 9.73) were felt necessary on average, with 14.89 (SD12.21) under indirect supervision. The curriculum does not currently specify a minimum number.
Comments made included a perceived lack of exposure and training opportunities amongst many trainees. Some units do not perform PDT insertion as the result of previous complications and others found that consultant staff preferentially undertook PDT insertion in order to maintain their own skills in the face of low numbers. There was a variety in policy and insertion technique between units with some locally run cadaver simulation courses being highlighted as of particular benefit. Generally trainees felt that the opportunity to undertake PDT insertion with indirect supervision uncommonly occurred, as it was usual practice for a consultant to undertake bronchoscopic guidance.
Given the risks involved, perhaps PDT insertion should be developed as an additional skill by trainees and consultants with a particularly interest according to local needs, rather than it remain a key skill for all Intensivists. Local critical care networks could develop regional policies to improve both service and training provision as well as devise simulation courses perhaps as a session in an already established airway or critical care skills course.
Comparison of percentage of time spent within each tidal volume range between 2012 and 2013/14.
References
17CP Outcome of patients admitted to intensive care unit with acute on chronic liver disease
PN Gunasekera, S Colley and D Ventour
Intensive Care Unit, Hull Royal Infirmary, Hull, England
Hull has a high incidence of chronic liver disease (CLD) and admission to intensive care unit (ICU) is individualised. We wanted to identify trends that may assist us in patient care.
Recent studies have shown an ICU mortality of 38–43% and hospital mortality of 47–70%; 59% of ventilated patients died on ICU.1–3 A model for end stage liver disease (MELD) score of over 25 is associated with a higher mortality.4
We studied the case notes of all patients admitted to critical care with acute on chronic liver disease from 1 January 2012 to 1 April 2014.
Thirty-five patients were admitted. Twenty-three patients died on ICU (65.7%) and another patient died post-discharge (68.6%). Mean age of survivors was 46 years and 53 years in the non-survivors. The survivors had a median MELD score of 15 and 29 in the non-survivors. The median APACHE II score was 18 in the survivors and 26.5 in the non-survivors suggesting they were sicker. This was the first presentation of liver disease in 12 (34%) patients, while 23 had previous similar presentations.
Of the 35 patients, two were admitted for monitoring only, both survived. Four out of 10 patients admitted with single organ failure died (40%), 7 out of 10 patients with dual organ failure died (70%) and 11 out of 12 patients with three or more organ failure died (92%).
Twenty-two patients received either invasive or non-invasive ventilation of which 17 died (77%). Five patients received renal replacement therapy (RRT), all died. Nine patients received only cardiovascular support of which four died (44%). Care was limited to level 2 in 11 patients – of which 9 died (82%). RRT was restricted in two patients, both died. There was no limitation of care in 22 patients of which 10 died (45%).
Patients presenting with gastrointestinal (GI) haemorrhage (21) had a mortality of 57%, while those with sepsis (7) had a mortality of 85%. In seven patients, no aetiology was found for the acute decompensation, five died (71%).
Twelve of the deaths occurred within 48 h of admission. Treatment was withdrawn in a further nine. Mean time to withdrawal was 6.6 days. Mean length of stay for both survivors and non-survivors was 3.2 days.
Critical care patients with CLD have a very high mortality; GI bleeding has the best prognosis but this is still worse than general ICU mortality. The number of failing organs and MELD score appears to be the best predictors of survival. A requirement for respiratory support and renal support has extremely poor prognosis. Only 35 patients over 27 months and 50% of non-survivors dying within 48 h suggest that our patient selection should improve.
We have suggested our local admission policy should be based on number of failing organ systems, the support required and the admission MELD score. Patients should be seen in clinic and escalation plans made. Early admission (single/dual organ failure) is recommended. Once a patient is admitted, they should be treated aggressively for three days and if there is no improvement, treatment withdrawal considered.
References
18CP Planned general surgical admission to critical care… Who made it, who did not and what impact did the management location have on patient outcomes?
A Garland1, S Winn1, J Paddle2 and M Feldman3
1Peninsula College of Medicine and Dentistry, Knowledge Spa, Truro, UK
2Department of Critical Care, Royal Cornwall Hospital, Truro, UK
3Royal Cornwall Hospital, Truro, UK
The UK NHS is facing unprecedented pressure to deliver more efficient care pathways elective and emergency admissions. With constrained resource, we need to improve patient safety, outcomes and lower mortality rates.1,2 Whilst critical care (CC) admission following cardiac surgery is routine, provision following major general surgery is limited despite higher risks for many patients.3 Postoperative CC admission improves outcomes for high-risk patients following elective general surgery4,5 and is an essential component of emergency abdominal surgery.4,5 Despite this, we still witness elective cases being postponed due to non-availability of CC beds or continued without CC support. The aim of this paper is to assess the outcomes for patients booked for elective general surgery with critical care unit (CCU) support, differentiating patient outcomes according to location of postoperative care.
Patients at the Royal Cornwall Hospitals NHS Trust who need elective surgery supported by CC are entered into an electronic diary preoperatively. In sum, 187 patients had elective general surgery planned during the period October 2013–April 2015. Details of hospital stay, mortality, location of care and complications were retrieved from the electronic patient record (Maxims); 137 patients were planned and admitted to the CCU (it is noteworthy that the majority of upper gastrointestinal (UGI) patients are bariatric); 19 patients were patients who had been cancelled previously, as no bed was available, some were admitted on the date initially planned; 36 patients were cancelled on the day of surgery, as no CC bed was available; 28 patients attended and were told that no bed was available, but the medical team decided that the risk of ward care was acceptable compared to the risk of delaying treatment. Therefore, the group who underwent surgery without admission to the CCU were the healthier subgroup of all those not admitted to CC, the rest being sent home without surgery.
The median length of stay (LOS) for those admitted direct to CC varied depending on surgical specialty (CR = 8; UGI = 2). This was also noted in those managed on the ward (CR = 5; UGI = 2). The 30-day mortality was 0 for ward patients and 3 (2.5%) for all CCU patients admitted. The 90-day mortality for all CCU patients admitted was 13 (9.5%), whereas for all ward patients the mortality was 12.2%.
In our population those patients booked for elective general surgery who did not gain access to CC but deemed healthy enough to undergo surgery with ward care alone had a shorter LOS and lower mortality. This supports the decision making of the clinical teams in that this subgroup of patients had good outcomes and the decision to go ahead with surgery was safe.
A significant proportion of patients cancelled on the day of surgery due to a lack of CC. It may be, given the finding above, that we can judge better which patients must have CC support and which do not, in order to minimise unnecessary demand.
Further prospective investigations should focus on recording POSSUM scores. Development of an automated tool that records equivalent ICNARC scores for those planned for CC that do not make it to CC is necessary. This would give more depth to the comparator population and illness severity of those not admitted and allowing for more efficient planning of inpatient flow.
Demographics, outcomes and complications of general surgical patients who were either cancelled on the day, admitted to critical care, admitted to critical care following cancellation or those patients who received ward-based care following elective surgery at the Royal Cornwall Hospitals NHS Trust. CC: critical care; CR: colorectal; UGI: upper gastrointestinal; LOS: length of stay.
References
19CP Determining the impact of an electronic early warning score systems in a district general hospital
JB Simon, RJ Ferguson and KT Murray
Intensive Care Department, Conquest Hospital, Hastings, UK
In 2014, our hospital introduced the VitalPAC interface, an electronic system for recording of NEWS scores to replace the paper notes. It gives interactive guidance on when observations are due and allows remote viewing of patient scores. Centres that have introduced this software have demonstrated improved outcomes.3
The objective was to examine in-hospital adult cardiac arrests before the introduction of VitalPAC and determine if NEWS scoring was done appropriately and if the correct action was taken prior to arrest. Comparison would then be made to arrests post introduction of VitalPAC, allowing us to determine any differences and identify areas for further improvement.
We note that after the introduction of VitalPAC the survival rate from cardiac arrests was lower, the average age was higher, and the time in hospital prior to arrest was shorter.
In light of these findings we intend to introduce ward specific feedback, with a focus VitalPAC adherence performance. We have introduced a policy in which all cardiac arrests are analysed to determine what else could have been done, with targeted feedback. Once these changes are implemented we intend to re-audit.
References
20CP Filter lifespan during continuous renal replacement therapy: Audit of a revised protocol
A Sisson, M O’Connor, T Hussein and JR Prowle
Adult Critical Care Unit, Royal London Hospital, London, UK
A major challenge in the application of continuous renal replacement therapy (CRRT) in the intensive care unit (ICU) is balancing optimum circuit lifespan, the risks of systemic anticoagulation and the practicalities of placing central venous catheters in critically ill patients.1–4
We audited the first 72 h of therapy in consecutive patients requiring CRRT in a 44-bed adult ICU servicing a level 1 trauma centre over a six-week period six months after the introduction of a revised CRRT protocol emphasising line length and position, and optimal anticoagulant strategy. We examined site of venous catheter insertion, anticoagulation, modality of CRRT, circuit lifespan and interruptions to treatment.
Twenty-four patients (median age 61 years, 62% male) underwent CRRT during the audit period. The admitting diagnosis was medical in 62% and surgical or trauma in 38% of patients. Eleven patients (46%) remained on CRRT for the 72 h audit window; nine patients (38%) died and four (17%) recovered function during the audit period. In total, 65 filters were used with an overall median duration of CRRT of 25.5 h. The initial site of catheter insertion was internal jugular vein in 17 (71%) and femoral vein in 7 (29%) patients. Anticoagulants used were heparin 412 h (40%), prostacyclin 185 h (18%) and none 438 h (42%).
There were no significant differences observed in median filter life in hours based on catheter insertion site (femoral 9 (2–12), right internal jugular 12 (5–32) and left internal jugular 8 (6–26); p = 0.26) (Figure 1). Significantly higher effluent flow rates (ml/kg/h) were achieved with catheters inserted in jugular veins as compared to femoral veins (right internal jugular 35 (25–35), left internal jugular 35 (25–35), femoral 25 (25–25); p = 0.04).
There were no significant differences in the reasons for filter discontinuation based on catheter insertion site, or anticoagulant used. Pre-dilution was more frequently employed where no anticoagulant or prostacyclin were used as anticoagulation (heparin 21%, prostacyclin 89% and none 75%; p < 0.0001).
Twenty-nine percent of catheters violated departmental guidelines for length of Vascath in given vessel; however, this did not significantly affect average filter life.
During the first 72 h of CRRT filter, lifespan and compliance with unit protocols regarding optimal catheter length and site remain poor. In particular, many patients have contraindications to anticoagulation and use of an optimal catheter location. Improving filter lifespan could improve patient care and provide cost savings but may be challenging using current methodology in our case mix.
Filter life in hours. Green bars: median; red boxes: interquartile range; whiskers: maxima and minima.
References
21CP An audit of chest drain management in Wexham Park Hospital ICU
B Millette and D Sedgewick
Department of Intensive Care, Wexham Park Hospital, Frimley Health NHS Foundation Trust, Slough, UK
Management of chest drains has been greatly improved in the last several years following a rapid response report by the NPSA in 20081 and a national audit programme2,3 led by the British Thoracic Society (BTS). We aimed to audit practice in our institution using the 2010 BTS guidelines for management of pleural disease.4
We conducted a prospective audit over four months using an electronic proforma to capture all patients with a chest drain on the intensive care unit. The gold standard was 100% compliance to each of the criteria extracted from the 2010 BTS guideline.
In this time period, 23 chest drains were inserted in 17 patients. Nineteen of these were inserted on the intensive care unit, two in the radiology department and two in theatres; 13 (65%) were inserted for pleural effusion and two (8.7%) each for pneumothorax, haemothorax, empyema, post-surgical prophylaxis and hydropneumothorax. Written consent was only taken for the two surgical patients. Fourteen patients were unable to consent due to reduced level of consciousness but did not have consent form four completed. The remaining seven patients had capacity but were not consented in either written or verbal form. There was deficient documentation leading to no record of ultrasound use in 8 patients (34.8%) and no record of sterile technique in 12 patients (52.2%). Alarmingly, 14 drains (60.9%) were not connected to an underwater seal system, instead being connected to a drainage bag without a one-way valve. One patient (4.35%) drained more than 1.5 L in the first hour, placing this patient at risk of re-expansion pulmonary oedema. Twelve patients (52.2%) consistently had a daily check recorded. Six patients (26.1%) did not have a reason for removal of the drain recorded.
These results show local practice is in need of considerable improvement. Of particular concern is that a majority of chest drains are being connected to drainage bags with no one-way valve. This increases the risk of infection as the patient can generate a large negative pressure and aspirate bag contents back into the pleural cavity. Consent is an issue that has consistently been highlighted for improvement in national audits.2,3 We have undertaken a comprehensive overhaul of local chest drain management policy including improving chest drain documentation on our electronic health record system, writing local guidelines in conjunction with the respiratory department and multidisciplinary education of doctors and nurses focusing on best practice. We are confident that we can improve practice and patient safety with these measures. A re-audit is scheduled to ensure this anticipated improvement occurs.
References
22CP Management of post-intubation tracheobronchial rupture
D Whitmore and J Rodrigues
University Hospital Aintree, Liverpool, UK
References
23CP Whipps Cross university hospital ICU, are we ‘on the right trach’? A multi-disciplinary audit approach
A de Gea Rico, P Munro, F Murray and A Raj
Whipps Cross University Hospital, London, UK
Subsequently, a consultant-led training session including small group tutorials, practical skills and case-based discussions for junior anaesthetists, coupled with a new ward patient documentation package was implemented. This achieved improved outcomes in all groups as shown by the re-audit cycle.
Bronchoscopic view of posterior tracheal wall tear (actual patient).
References
24CP Tracheostomy airway management by junior doctors on the intensive care unit
N Simpson and N Arora
Intensive Care, Heartlands Hospital, Birmingham, England
Junior doctors working on the intensive care unit (ICU) come across patients with tracheostomies in their day to day practice. The ability to deal with an airway emergency in a patient with a tracheostomy is therefore vital for those working on the ICU. The aim of this project was to assess the knowledge of 11 junior doctors working in the ICU, regarding the management of tracheostomy airway emergencies.
A form based on the National Tracheostomy Safety Project (NTSP)1 flowchart for tracheostomy airway emergencies was used. Junior doctors where given a scenario based on a non-patent tracheostomy and deteriorating patient. The scenario guided the juniors down the emergency management flowchart, culminating in the need for tracheostomy removal and oral airway management. The doctors were asked what they would do at each step of the flowchart.
Of the 11 junior doctors assessed, only one was able to correctly state the steps at each point of the emergency pathway. 100% of junior doctors were able to identify where in the ICU they would find information regarding a patient’s tracheostomy.4 different potential places where this information could be found on the ICU were identified.
45% of junior doctors were able to correctly describe the first step of assessing an unresponsive tracheostomy patient as ‘look, listen, and feel over BOTH the mouth and tracheostomy’. 54% of junior doctors were aware that O2 was required over both the patient’s mouth and tracheostomy. 36% of junior doctors were able to correctly describe the first step in assessing a patient with tracheostomy difficulties – ‘remove the inner tube or speaking valve if applicable’. 18% of junior doctors were aware of how to manage a situation where a suction catheter could not be passed. 72% of junior doctors were aware that the final step in managing a tracheostomy airway emergency is to ‘remove the tracheostomy, cover the hole and manage with oral airway manoeuvres’.
Recommendations for future change include:
A laminated card (e.g. NTSP tracheostomy card) should be placed at the top of the bed of a tracheostomy patient, detailing the type of tracheostomy (percutaneous or surgical), date inserted, type and size of tube, laryngoscopy grade and who to call in an emergency. Implementing this would remove any ambiguity in where to look to find information about a tracheostomy in the emergency situation.
A laminated flowchart of the tracheostomy emergency airway management (e.g. NTSP flowchart) should be at the bedside of any patient with a tracheostomy. This would guide any first responders in the initial management and allow the correct steps to be taken whilst waiting for emergency help to arrive.
A simulation scenario involving the management of tracheostomy airway emergencies should be introduced as part of the induction package for new junior doctors on the ICU. This would allow doctors to put the above guidelines into practice and allow them to become familiar with the practical steps involved.
Reference
25CP Checklist lead intubations on the intensive care unit
J Hobson, K Savjani and N Robin
Intensive Care Department, Countess of Chester Hospital, Chester, UK
Adequate preparation is central to successful intubation. NAP4 showed that identifying high-risk patients, preparing correct equipment and planning in case of failure were crucial in preventing complications.
Consequently, NAP4 has recommended a pre-intubation checklist. The aim of this project was to identify whether introduction of a checklist improved the process of intubation on the ICU at the Countess of Chester Hospital.
A pre-intubation checklist was designed (Figure 1). It was read aloud while the patient was being pre-oxygenated. The questionnaire was repeated six months after launching the checklist. Eighteen responses were collected prior to and 20 after the launch.
Other aspects of intubation identified by staff to be sometimes, often or always missed were outlining a failed intubation plan 14 (78%), preparing additional equipment 12 (67%), adequate role allocation 11 (61%) and identification of difficult airways 9 (50%). These were aspects requiring communication between all team members.
Introduction of the checklist resulted in improvements in all areas. No staff reported that presence of a difficult airway trolley, CO2 monitoring and patient positioning were sometimes, often or always forgotten.
Similarly smaller proportions of staff felt that communication-dependent aspects were being missed with six (30%) staff responding that a failed intubation plan, role allocation and identification of difficult airway were sometimes, always or often omitted; and four (20%) in collecting additional equipment.
A pre-intubation checklist.
Reference
26CP The management of two cases of severe calcium channel antagonist overdose: The role of hyperinsulinemic euglycemia therapy
R Schofield and L Wilson
Department of Anaesthesia, Countess of Chester NHS Foundation Trust, Cheshire, UK
Calcium channel antagonists are used in a variety of cardiovascular conditions, notably in the treatment of hypertension and tachyarrhythmia. As a widely prescribed therapy, they can be seen in overdose with lethal consquences.1 We discuss two patients presenting sequentially to our unit with significant amlodipine overdoses and our experience of symptom management.
Patient A, a 48-year-old man with a mixed staggered overdose was referred to intensive care 12 h after taking 480 mg of amlodipine with hypotension, bradycardia, oliguria, worsening lactic acidosis and prolonged QTC. Persistent hypotension ensued with failure to responds to inotropes, glucagon infusion and a trial of 20% lipid emulsion produced only a temporary reprieve. Clinical deterioration and hypoxia necessitated invasive ventilation and hemofiltration.
Patient B, 44-year-old man with a mixed overdose including 200 mg of amlodipine, presented 5 h after patient A, with profound hypotension, bradycardia, hypoxia and metabolic acidosis. He was rapidly treated with intravenous (IV) fluids, inotropes and 20% lipid emulsion. He required invasive ventilation and hemofiltration.
Profound inotrope-resistant hypotension with metabolic acidosis prompted the use of hyperinsulinemic euglycemia therapy in both of these patients. A bolus of IV insulin at 1 unit/kg followed by 10 units/kg/h of insulin infusion in conjunction with 50% dextrose at an initial rate of 25 ml/h. Cardiovascular stability was quickly achieved and clinical improvement was observed over 5–9 days. Both patients made full recovery and were subsequently discharged.
Overdose of calcium channel antagonists amplify the normal physiological mechanism of L-type calcium channels. Effects include profound bradycardia, hypotension, metabolic acidosis and hypoperfusion resulting in end organ damage.3,4 Blockade of L-type calcium channels in the pancreatic islet cells renders a hypoinsulinemic state.3 Early recognition, aggressive treatment and supportive measures are indicated; however, conventional therapies are often inadequate.3,4 High-dose insulin administered in this context can bring haemodynamic stability and reduce mortality.5 High-dose insulin is a positive inotrope, vasodilator and promotes increased intercellular glucose transport and metabolism.4,5
With our second patient, the institution of insulin therapy and subsequent stability was significantly quicker as we had the benefit of our experience gained earlier that evening.
We have produced and present a suggested algorithm for use of high-dose insulin in subsequent patients.
Insulin is a widely available and inexpensive drug which showed a rapid effect in a deteriorating life-threatening situation and we encourage colleagues to consider it in similar cases (Figure 1).
Proposed treatment algorithm.
References
27CP An audit of tracheostomy practice and complications in a large central teaching trust: Six-month data
SL Linford, A Sharman and K Somarakis
Critical Care, Nottingham University Hospitals Trust, Nottingham, UK
The fourth National Audit Project (NAP4) attributes 50% of airway related deaths and brain damage in critical care to a tracheostomy.1 The NCEPOD report ‘On the right trach?’ highlights tracheostomy care as a key area for focus.2 These publications along with a single-centre evaluation of percutaneous tracheostomy practice3 were the driver and gold standard for an audit of tracheostomy practice at a tertiary referral trust. The audit is ongoing; here, we present the first six months data.
Data were prospectively collected for all patients undergoing percutaneous or surgical tracheostomy from October 2014. Information collected included admitting diagnosis, onset of mechanical ventilation, date of decision to and insertion of tracheostomy, procedure type, complications and time to weaning from ventilation and decannulation. Insertion complications were audited against those detailed in the Aintree study3 and other data compared to NCEPOD.2
In sum, 119 patients underwent a tracheostomy. Ninety-five (80%) were percutaneous and 24 (20%) surgical; eighty (67%) patients were male. Average time to tracheostomy decision was 7.6 days, with average time to insertion being 8.88 days; 92 (77.3%) were inserted for weaning and 25 (21%) for airway protection or toileting. Nine (7.6%) were categorised as technically difficult (vs. 26%) and 0.8% complicated (vs. 3%).3 Eight (9%) patients were decannulated in under seven days (18% in NCEPOD). Extubation was trialled in 28 (23.5%) patients (40% NCEPOD). Bronchoscopy was used during the insertion of 70 (73.6%) percutaneous tracheostomies (51.5% NCEPOD) and end tidal CO2 monitoring was used in 38.4% (54.1% NCEPOD). Complication rate during critical care stay was 24.4% compared to 23.6% in NCEPOD and 5.3% once discharged to the wards (31.3% NCEPOD). There were eight unplanned tracheostomy changes, six of which occurred on critical care.
Our complication rates at insertion compare favourably with the Aintree study.3 Reasons for this likely include sample size and an element of reporting error. However, we do perform twice the number of tracheostomies per year (95 in six months versus 572 in six years) and so would expect the trends to be born out as our sample size increases. We compare favourably with NCEPOD in terms of early decannulation, likely reflecting our experience in identifying patients requiring tracheostomy, explained in part by our status as a neuro-critical care unit. In addition, our low incidence of unplanned changes and ward-based complications is likely due to careful patient discharge, well-trained ward staff and an excellent outreach service. Our low ‘trial of extubation’ rate is likely because of our neurosurgical patient population and lack of documentation, rather than actual failure to trial. The overall critical care complication rate in our population is slightly higher than in NCEPOD. Differences in the data recorded make this comparison difficult.
Recommendations for practice include mandatory use of end tidal CO2 monitoring for all procedures and better documentation of complications and trials of extubation. The audit is ongoing and complication data will be collected to enable better comparison with NCEPOD.
References
28CP High-flow nasal oxygen therapy (AIRVO2™) – Chesterfield experience
SL Taylor and CJ Goldson
Physiotherapy Department, Chesterfield Royal Hospital (CRH) NHS Foundation Trust, Chesterfield, UK
Heated humidification of oxygen allows the airway to maintain an improved balance of heat and moisture, optimising gas exchange, lung defence and patient comfort.1,2 In September 2014, 14 high-flow nasal oxygen (HFNO) heated humidification devices (AIRVO2™) were purchased that replaced the existing cold water humidification systems. This survey aimed to evaluate the use of HFNO devices and their impact on cost and quality of care for acutely ill patients in Chesterfield Royal Hospital (CRH).
We adopted a prospective qualitative survey of HFNO use for acutely ill patients in Chesterfield. A data collection proforma was completed by the Physiotherapy staff in each clinical area. Input was sought from other clinical staff involved in the care of patients receiving HFNO via the AIRVO2™. Feedback was collected over a six-week period in January–February 2015. In total, 60 forms were returned for 49 patients from various ward areas including two adult critical care units, acute medical and surgical wards.
The majority (72%) of patients on AIRVO2™ were in a critical care area, 25% of patients were on a medical ward and 3% on a surgical ward. Criteria for starting AIRVO2™ were that patients required FiO2 > 0.35 for an expected time of >48 h. Forty-two (63%) patients met the criteria. Other reasons for starting HFNO included patients having thick secretions (n = 6, 9%) and patients who had undergone tracheostomy or laryngectomy (n = 10, 15%). For some patients, multiple reasons were listed. Senior critical care staff feedback suggested that up to 12% of patients had avoided intensive care unit admission through use of HFNO; 100% of staff responded that they would choose HFNO over other devices. Twenty-three (38%) patients who would have traditionally required CPAP therapy were successfully managed with HFNO. A cost analysis of consumables for (Continuous Positive Airway Pressure) CPAP was compared with HFNO and showed a saving of approximately £80 per patient (estimated £1600 cost saving over the six-week data collection period).
From our experience of using HFNO, there has been a positive impact on patient flow through critical care and on physical and psychological consequences associated for patients and their families by potentially avoiding critical care admissions.3 HFNO appears to have reduced the use of CPAP in our Trust by 38% (n = 23). This has had a considerable impact on the cost of therapy and also on patient comfort. From a quality perspective, staff and patients appear to like HFNO over other high-flow devices such as CPAP.
This survey demonstrates that HFNO is being used appropriately and has had a positive impact on patient and staff experience. We recommend that HFNO usage continues across the Trust and our experience is transferable to other hospitals. Training for clinical staff is important to ensure overall patient safety and governance issues are addressed. Further research and re-audit could be carried out to evaluate factors such as impact on length of stay, cost of care and whether the benefits seen within this first year are sustained and generalisable.
References
29CP Does a nurse-led non-invasive ventilation service improve patient outcomes?
H Mainman, S Chambers and H Curtis
Critical Care Department, Queen Elizabeth Hospital, Gateshead, UK
Non-invasive ventilation (NIV) has been shown to reduce in-patient mortality in acute exacerbations of COPD (AECOPD; i.e. pH between 7.25 and 35) from 20 to 10%. In 2011, national data revealed that there were multifactorial failures in effective NIV service provision, with a 26% in-patient mortality rate.1
Local audit data in 2012 showed our in-patient mortality was 40% for all patients treated with NIV compared to the national rate of 30%. There was evidence of missed patients and delays to treatments. Could a specialist nurse-led NIV service improve this situation and help prevent unnecessary intensive care unit (ITU) admissions?
In the two years preceding the implementation of this service, 71 patients were admitted to the ICU with AECOPD; 39 required NIV first line and only 22 were mechanically ventilated. We hypothesised that an efficient NIV service could have prevented 49 ITU admissions, saving ITU bed days.
With involvement from both intensive care and respiratory specialists, a dedicated 24/7 nurse-led NIV service was developed in December 2013. This service had portable NIV machines and allocated respiratory beds. The nurse would aim to be involved in all AECOPD admissions from the point of admission, caring for the patient with support from acute medical team, until discharge. All aspects of clinical care were prospectively collected. including nursing work load.
Results (Table 1) revealed that more patients received NIV with an improved success rate, reduced in-patient mortality rate and no admissions to ITU, saving beds. Mortality fell to 12% by summer 2014 and was 0% in patients with pH 7.25–7.35. Quality indicators improved, e.g. failure planning, input from respiratory team and consultants.
Although survival rates were maintained over winter 2014–2015, quality indicators slipped, due to nursing pressures across our trust. During this period the NIV-nurse had extra or alternative duties on 28% of shifts. As a consequence, six patients were treated in ITU and three operations were cancelled.
Table to show the NIV practice trends and success following the introduction of a nurse-led NIV service at the Queen Elizabeth Hospital, Gateshead.
Reference
30CP One-year survival data for patients aged over 80 admitted to a general intensive care unit
N Singh and M Street
Intensive Care Unit, Royal Sussex County Hospital, Brighton, Sussex, UK
References
31CP Colours for key parts of tracheostomy tubes can impact on patient safety perceptions and experiences – An international survey
BA McGrath1,2, S Wallace1,2, J Lynch1,2 and E Ward1,3
1Global Tracheostomy Collaborative, UK
2Acute Intensive Care Unit, University Hospital South Manchester, Wythenshawe, UK
3Global Tracheostomy Collaborative, Boston, MA, USA
Avoidable harm occurs in patients requiring tracheostomies. Most studies are performed in the hospital setting and have identified recurrent themes including staff education, training and competency, and the provision and safe use of tracheostomy-related equipment.1,2 Whilst some organisations have reported improvements in care by standardising protocols, training, environments and equipment, reports continue to highlight confusion with key elements of tracheostomy tubes.3 Examples include the failure to use appropriate inner cannulae, confusion over fenestrated inner and outer tubes and the inappropriate use of speaking valves.4 The Global Tracheostomy Collaborative (GTC, www.globaltrach.org) is a non-profit organisation committed to improving tracheostomy care for patients and families worldwide. Through the global community facilitated by the GTC, patients, families and healthcare professionals recognised variations in the types and colours of tracheostomy tubes and accessories available both worldwide and within individual countries. Anecdotal reports where confusion over key parts caused harm led us to design a survey in an attempt to define this problem in detail.
An internet-based survey was created and made available via SurveyMonkey (www.surveymonkey.com), and participants were invited by email and social media via the GTC networks. Responses were collected using check-box questions and free text. A number of statements were provided with agreement recorded on a Likert scale, bounded by 1 (strongly disagree) through 5 (neutral) to 9 (strongly agree). For simplicity, mean and mode Likert scores are reported.
There were 313 responses recorded between 5 June and 27 July 2015. Respondents were mostly North American (39.0%), from the UK (31.3%) or from Australia and New Zealand (17.7%). Carers and family members of patients comprised 21.9% of respondents, 8.0% were patients, 14.5% medical staff, 23.2% nursing staff with 26.4% Allied Health Professionals. Eighty-five percent of respondents felt that confusion over colour could lead to patient harm, with 38.8% reporting in detail actual ‘serious’ harm (emergency services or resuscitation team called). Likert scores in response to statements are summarised in Table 1.
Summary of Likert scores in response to survey questions.
Likert scale was bounded by 1 (strongly disagree) through 5 (neutral) to 9 (strongly agree). TT: tracheostomy tube; SD: standard deviation.
References
32CP Experience with ward-based high-flow nasal oxygen in adults with acute hypoxaemic non-hypercapnic respiratory failure in a tertiary centre
E Price, P Featherstone, J Martin, S Petty, S Precious and M Trivedi
John V Farman Intensive Care Unit, Addenbrooke’s Hospital, Cambridge, UK
High-flow nasal oxygen (HFNO) is increasingly being used in adults with acute hypoxaemic non-hypercapnic respiratory failure. This technique is well tolerated by patients1 and can be initiated in a ward-based setting. A recent multicentre randomised control trial demonstrated a reduction in intensive care unit (ICU) and 90-day mortality with the use of HFNO, compared with standard oxygen therapy delivered through a face mask, or non-invasive positive-pressure ventilation in a highly selected group of patients with acute respiratory failure.2 We wished to assess outcomes following initiation of HFNO in a hetereogenous, real world population.
This service evaluation was registered and approved by our institution, a 1100-bed tertiary centre with all specialties except cardiothoracics. Using retrospective electronic case note review, we identified all patients over 18 years of age with acute hypoxaemic non-hypercapnic respiratory failure and a ratio of the partial pressure of arterial oxygen to the fraction of inspired oxygen (P:F ratio) of ≤40 kPa, who were commenced on HFNO in ward areas by the Addenbrooke’s Rapid Response Team between November 2014 and March 2015. Patients with a baseline partial pressure of arterial carbon dioxide ≥6 kPa, Glasgow Coma Score ≤ 12, or haemodynamic instability were excluded.
Between November 2014 and March 2015, 85 patients (38 females; 47 males; mean age 70 years) were commenced on HFNO. Forty-six percent were current or past smokers. The commonest aetiologies of respiratory failure were community acquired pneumonia, hospital-acquired pneumonia and pneumonia related to immunosuppression. Eighteen patients (mean P:F ratio 15.5 kPa prior to initiation of therapy) were admitted to ICU after receiving HFNO for a median of two days on the ward (range one to six days). Eleven (61%) of these patients were subsequently intubated and mechanically ventilated. The 28- and 90-day mortality among patients admitted to ICU was 16.6% and 22.2%, respectively. The remaining 67 patients (mean P:F ratio 16.7 kPa) received ward-based HFNO for a median of four days (range one to twelve days) and had an overall 28-day mortality of 41.8%. Within a subgroup of 21 patients who had treatment limits in place at the time of commencement of HFNO, mortality was 71.4% at 28 days, rising to 85.7% at 90 days. These patients received a median of three days of HFNO (range one to six days).
Our results align with previously published mortality data for patients who initially receive HFNO in ICU but subsequently require intubation.1,2 The mortality rate in patients with treatment limits in place is expectedly high. This likely indicates that HFNO is being used as a palliative measure and may be appropriate for the management of symptomatic dyspnoea at the end of life. However, since the median duration of therapy is three days, it is possible that HFNO is delaying death. This requires further study. Patients in our institution commenced on HFNO represent a heterogeneous group who receive the therapy for a prolonged period. Critical care outreach is well placed to follow these patients closely for treatment efficacy and duration, be that with palliative or curative intent.
References
33CP Review of HIV-positive patients admitted to Critical Care at a district general hospital in London
H Manukonda and P Gandre
Critical Care Complex, North Middlesex University Hospital, London, UK
North Middlesex University Hospital is one of the largest specialist referral centres in North London for patients suffering from human immunodeficiency virus (HIV) infection caring for approximately 1200 patients a year. A service evaluation was carried out on patients infected with HIV and admitted to the adult Critical Care unit from August 2013 to July 2015. We reviewed the clinical course of these patients in order to extract the lessons that could be learnt from care of these patients. Data were collected retrospectively from the patients’ hospital notes and the electronic records in Critical Care. A total of 30 admissions took place during this period, eight of which were readmissions. Of the total 22 patients, 12 patients survived the hospital discharge.
The mean age of the patients was 47.3 years, minimum age being 21 years and maximum 72 years. Over 70% of the patients were males and 36% were of Afro Caribbean ethnicity. Over half of the patients were directly admitted to Critical Care from the Emergency Department. Presenting complaint of over 50% of the patients was in the form of a non-specific respiratory or gastrointestinal symptom. Five out of 22 patients were not previously known to be HIV positive until their admission to Critical Care. Two out of these five had presented with non-specific symptoms to the Emergency Department but died in Critical Care from multiple HIV-related complications. In the known HIV-positive patients duration between the diagnosis of HIV and admission to Critical Care was between one month and five years. The most common initial Critical Care diagnosis was septic shock secondary to chest infection followed by encephalopathy, acute kidney injury, acute liver failure and drug overdose and subsequently majority of the patients developed severe multi-organ failure. Some of the HIV-related complications were histoplasmosis complicated with oesophageal perforation, cerebral toxoplasmosis, pneumocystis carinii pneumonia, leukoencephalopathy leading to brain stem death, candidaemia, herpes meningitis, fatal deep vein thrombosis and B-cell lymphoma. Over 50% of the septic shock remained refractory to treatment.
Initial presentation of a new HIV infection may be non-specific and can deteriorate fast to require Critical Care admission. High level of suspicion is warranted for timely diagnosis. The Critical Care mortality and hospital mortality for HIV-positive patients requiring Critical Care is significantly higher than the rest of the patient groups and readmission to Critical Care during the same hospital stay carries even higher risk of mortality. Complications secondary to HIV infection in Critical Care are unique and early specialist input is thought to be synergistic to the Critical Care. Although the sample size is small to make definitive conclusions, our careful service evaluation indicated some specific red flags derived from social and clinical history which should be looked for, especially in high HIV prevalent areas. We believe this will be helpful in making the new HIV diagnosis quicker. Patients with good compliance with antiretroviral treatment in the community are more likely to survive the hospital discharge.
Reference
34CP ITU nurses confidence assisting in airway situations and knowledge about difficult airway equipment
J Paige, L Conway, C Bassford and R Townsend
General Critical Care Unit, University Hospital Coventry and Warwickshire, Coventry, UK
The association of anaesthetists of Great Britain and Ireland and Royal College of Anaesthetists showed during the fourth National audit project (NAP4) that intensive care units (ICUs) were high risk areas for difficult airways.1 In fact more than one in four airway events recognised by NAP4 occurred in an ICU with significantly worse outcomes than for events in theatres. It suggested that variable backgrounds of nurses may contribute to adverse events related to airway management. This was highlighted in an adverse event in our ICU where review of the case came to the same conclusion. This led us to conduct a survey of our staff.
The survey collected demographic data (grade of nurse, length of experience in ICU and specific airway training), followed by questions assessing confidence levels in a variety of airway situations scoring 0 to 10. They then completed a picture quiz of airway equipment found on the difficult airway trolley on the ICU. Respondents were asked to name the equipment and answer basic questions about it. The surveys were filled in anonymously and posted into a letter box for the authors to collect and analyse.
Our survey showed a large deficiency in the training that ICU nurses receive, with only one of 38 respondents (2.3%) receiving airway training despite 23 (60%) being involved in difficult airway situations. Confidence assisting in a difficult airway situation (Figure 1) increased as nurse seniority increased (Kruskal-Wallis p = 0.004) as it did with previous experience of difficult airway situations (4.4–6.6, p = 0.01). Recognition of equipment was poor with less than 15% recognising an airtraq, kingvision or LMA supreme; however, 86% recognised a bougie.
In conclusion, we found that the confidence of nurses in managing a difficult airway increases in line with their seniority. Our data also show that prior experience of a difficult airway situation improves nurses’ confidence in assisting physicians in difficult airway situations. This strongly suggests that simulation-based training has a role in improving the preparedness of the team. To address the issue of non-recognition of equipment, a photo checklist and e-learning package for the nurses are being developed within our trust.
Chart showing nurses self-reported confidence level when assisting with a difficult intubation according to band of nurse.
Reference
35CP A nationwide survey of consultant working practices on intensive care units in the United Kingdom
LC Evans and O Boyd
Intensive Care Department, Brighton and Sussex University Hospitals NHS Trust, Brighton, UK
Of these 10 sessions, 7.5 PAs are usually dedicated to direct clinical care (DCC). In the case of consultant intensivists a minimum of two, but ideally four or more should be dedicated to intensive care.2 In addition, consultants require time for supporting professional activities (SPA). 1.5 SPAs is the absolute minimum required to keep up to date and 2.5 is the recommendation.
Is this typical of job plans for consultant intensivists? We performed a survey to gain a better understanding of working practices amongst intensive care consultants across the UK.
Three hundred seventy-four job plans demonstrated consultants across the county working between 7 and 14 PAs. The majority (93%) work 10–12 PAs; 67 (18%) work 10 PAs, 123 (33%) 11 PAs and 157 (42%) 12 PAs.
The range of DCC sessions in intensive care for 398 consultants was 1 to 10 with an average number of 5.14; 98.5% of consultants work at least the recommended minimum of two and 79.4% work four or more.
In terms of SPAs, 344 respondents demonstrate a range of 0 to 4.5 per contract, with an average of 2.13; 46% of job plans had less than the recommended 2.5 SPAs.
The range of on-call frequencies was 1 in 3 to 1 in 14 with 20 trusts falling into the desirable medium frequency of 1 in 5 to 1 in 9. No consultants were formally rostered to be resident overnight. However, in several trusts, consultants were often present until midnight or for large proportions of the night. All except for one hospital received PA allocation for out-of-hours advice. Half of the trusts had annualised job plans, but only three had compensatory rest under the rules of the European Working Time Directive as part of their contract.3
References
36CP The role of the occupational therapist in a major trauma critical care team: Findings from current clinical practice
R Gaunt, S Brady and R Moses
Royal Preston Hospital, Lancashire Teaching Hospitals NHS Foundation Trust, Preston, UK
At Royal Preston Hospital, as a result of increasing survival rates of those suffering major trauma and critical illness, patients are demonstrating more complex cognitive rehabilitation needs. Therefore, following a departmental restructure, the decision was made to create a 0.5 FTE senior Occupational Therapist (OT) post to join the existing Critical Care Unit (CrCu) Physiotherapy team. The aim of this review is to evaluate the impact of the OT working within the CrCu team.
References
37CP A nationwide survey of trainee cover on intensive care units
LC Evans and O Boyd
Intensive Care Department, Brighton and Sussex University Hospitals NHS Trust, Brighton, UK
Overall weekday cover was good with only 15% of units usually falling short of the recommended resident to patient ratio of one to eight. However, due to expected rota variation, up to 35% of units frequently had suboptimal weekday ratios. The weekday ratio of resident to patient ranged from 1 to 1.8 to 1 to 21.
Night time cover was expectedly sparser with 77% of units consistently having insufficient trainee cover. The mean average was 1 doctor to 11 patients. Higher ratios did not reflect a higher level of trainee experience; 62.5% of units did not have dedicated trainee cover overnight and were staffed by doctors with shared responsibilities to theatre and obstetrics. These data do not reflect additional time spent away from units at emergency calls or referrals over the course of a shift.2
Regionally, there are trainee shortages and anticipated funding and training issues for the near future. Some deaneries already use ACCPs to provide cover and several others intend to recruit and train more to cope with gaps in rotas. Shortages in the numbers of residents particularly with competencies in advanced airway management were of considerable concern.
References
38CP A feasibility study to investigate the effectiveness of using a Mechanical In-Exsufflator device for lung volume recruitment via positive pressure breathing as an alternative to the Mark 7 Respirator intermittent positive pressure device
R Moses
Royal Preston Hospital, Lancashire Teaching Hospitals, Preston, UK
The aim of this study was to explore the feasibility of using an MI-E device (NIPPY Clearway) in an IPPB mode as an alternative to the Mark 7 Respirator for patients with retained secretions.
Patients with reduced lung volumes who are at risk of respiratory complications Patients with retained secretions who are unable to clear effectively secondary to poor respiratory effort Acute respiratory compromise Inability to take a deep breath Reduced Forced Vital Capacity on spirometry Poor cough secondary to poor inspiratory effort Ineffective active cycle of breathing techniques (ACBT), autogenic drainage (AD), mobilisation (if indicated) or use of lung volume recruitment bag (LVR) if appropriate SpO2 Auscultation findings Secretions cleared Reduced FiO2 Chest X-Ray (CXR) findings Improved spontaneous ability to deep breathe FVC recordings (if monitoring) Patient feedback Adverse reactions Need for Mark 7 Respirator IPPB
Patient’s with obstructive disease needed higher positive pressure (mean = 30 cm H2O) compared to restrictive patients (mean = 22 cmH2O). Those with restrictive disease had a slightly higher severity score than obstructive patients (3.6 vs. 4.1) and required slower flow profiles (53.3 vs. 37.5).
MI-E devices (NIPPY clearway) can be used as an alternative to the Mark 7 Respirator to provide IPPB to patients. This provides a cost-effective treatment strategy for patients with respiratory compromise.
39CP Is Intensive care a safe environment for emergency tracheal intubation?
SS Kumar1, C Christau2 and S Hutchinson1
1Intensive Care, Norfolk and Norwich University Hospital, Norwich, UK
2Anaesthesia, Norfolk and Norwich University Hospital, Norwich, UK
As a consequence of this study, we now conduct regular emergency tracheal intubation in-situ simulation sessions, resulting in the development of intubation checklist, education about capnography and education about equipment. We also developed educational videos demonstrating how to set up for emergency intubation, pre-oxygenation manoeuvre and the drill of safe tracheal intubation.
References
40CP Telemedicine in critical care echocardiography: Proof of concept
A Constantine, Y Mebrate, M Alam, G Kanaganayagam and S Price
Adult Intensive Care Unit, Royal Brompton & Harefield NHS Foundation Trust, London, UK
Critical care echocardiography is expanding rapidly, with concomitant development of certification programmes in focused cardiac ultrasound and comprehensive echocardiography.1 Recent guidelines emphasise the importance of echocardiography in the critical care setting, recognising the potential complexity of the patient population, suggesting that irrespective of the base specialty of the operator, they should have access to adequate supervision by an expert in the field.2,3 In an attempt to address this, we have looked at the feasibility and potential impact of telemedicine in critical care echocardiography.
A system of remote access echocardiography to allow a second, expert opinion was trialled for 30 days in a 20-bedded intensive care unit in a tertiary cardiothoracic centre, with peri-operative/critical care echocardiography performed and interpreted by a team of echocardiography-accredited intensivists and cardiologists. Images were uploaded during acquisition onto a computerised archive system. A generic, fully encrypted, secure link was available for immediate remote access review by a dedicated cardiac intensivist. Haemodynamic and treatment parameters were simultaneously accessible from the electronic patient record. Remote review was requested by the treating clinician and/or echocardiographer where there was potential discrepancy between clinical status and the echocardiographic images, and/or uncertainty regarding their implications for optimal clinical management.
Twenty-five of the 385 echocardiographic studies undertaken during the pilot were reviewed. There were no technical issues with logon/remote access, and no delays in review and discussion of patient data. In 4 of 25 studies, remote expert review provided reassurance, without leading to a change in management. In the remainder, it led to a direct change in management (Table 1).
Breakdown of the cases where remote expert echocardiography review was undertaken according to the clinical question asked.
RV: right ventricle; ACHD: adult congenital heart disease; PVR: pulmonary vascular resistance.
References
41CP Tracheostomy practice: Changing goals?
F Donaldson, A Cormack, C Grant, W Barr and T Szakmany
ACT Directorate, Royal Glamorgan Hospital, Llantrisant, UK
Change in tracheostomy practice on the ICU.
Reference
42CP Measuring patient acuity in critical care – CMCCN dependency scoring tool
S Clarke and M Kynaston
Cheshire & Mersey Critical Care Network (CMCCN), UK
At this challenging time, it is more important than ever for critical care team leaders and managers to have tools available to support their decisions in relation to registered nursing establishments and skill mix requirements. Cheshire & Mersey Critical Care Network (CMCCN) have developed a tool which has been used locally for the past 15 years to evidence and underpin safe staffing decisions on each shift, inform the development of business cases and provide a common language for discussions with senior colleagues outside of the critical care service. It provides the critical care nursing workforce with a measurement to evidence individual patients’ acuity and the associated workload of caring for them in a critical care environment. It is independent from the levels of care tool and the number of organs supported and identifies the registered nurse to patient ratio required to safely care for the individual patient based on their acuity.
The tool is systems based and includes common nursing care and interventions within each section. It is colour coded into red (care requiring 1:1 nursing), amber (care requiring 1:2 nursing) and green (care that can be delivered in the ward area). This assessment is undertaken each shift and supports decisions relating to the allocation of staff and resources. The tool has been linked by CMCCN to the National Competency Framework for Adult Critical Care Nurses and can be used to undertake a deeper analysis of skill mix requirements.
An extensive evaluation of the tool has taken place across Cheshire & Mersey, Lancs & South Cumbria and the South West, providing feedback from across over 20 sites. This poster will present the results and key finding from the evaluation and provide information from the end users on the value of using the tool in practice. The findings will outline the difference in the number of registered staff needed when using the acuity tool compared to traditional ratios, independent validation of the acuity score between nurses and professional consensus on the inclusion and exclusion of criteria used.
43CP Observational study assessing the suitability of patient referral to the critical care unit of a district general hospital
P Dhillon, J Shak, D Smith and K Turner
Critical Care Unit, Ipswich Hospital NHS Trust, Suffolk, UK P Dhillon, J Shak and D Smith are joint first authors
The Critical Care Unit (CCU) at Ipswich Hospital NHS Trust has experienced an increase in annual admissions from 901 in 2011 to 979 in 2014. This has been matched by a simultaneous increase in the number of CCU referrals but no concomitant increase in the staffing levels or bed number. CCU is a finite resource under greater pressure year on year. This observational study looked at the ‘doctors time off’ the CCU and hence reduced unit cover implication of ‘inappropriate’ referrals.
The suitability of all CCU referrals was assessed over a 28-day period in May 2015. A proforma was designed to collect information regarding patient demographics, premorbid state, presenting condition, action taken by referring team prior to referral and outcome of referral. Suitability was assessed firstly by the recording intensive care doctor (ST1 and above) and validated retrospectively by a consultant and senior registrar based on the premorbid state of the patient and their presenting condition. The referring speciality and time of referral was also recorded. Exclusion criteria included repatriations and elective surgical admissions.
Of the 96 referrals recorded, 8 were excluded due to insufficient data collection, and the remaining 88 referrals were subsequently analysed. Figure 1 shows the distribution of referrals in terms of suitability on retrospective review.
Of the 88 referrals, 26 (29.5%) were deemed inappropriate by all three professionals. Of note, 20 (22.7%) had incapacitating co-morbidities prior to hospital admission (for example, a patient of 96 years of age with advanced dementia, fully dependent on nursing home care with a DNACPR in situ referred for sepsis management). Furthermore, there were six inappropriate referrals based on presenting condition, including referrals for venous access and inadequate resuscitation by the referring team.
Suitability and outcome of referrals to Critical Care over a 28-day period in May 2015.
Overall, an average of 42 min was spent reviewing each referral (range 0–120, standard deviation 31). A total of 832 min was spent on inappropriate referrals, averaging 32 min (range 5–85, standard deviation 22).
This observational study illustrates that a third of referrals to our intensive care unit were deemed inappropriate, with significant time spent on patient assessment by the increasingly pressurised on-call doctor. This results in a significant amount of time being lost on assessing unsuitable referrals that could be spent on CCU treatment.
We propose that intensive care services can be more effectively used by advocating consultant to consultant referrals and education of referring teams about appropriate indications of intensive care.1
Reference
44CP Pulmonary complications of intravenous drug abuse: A case series
C Lynch, C Britton-Jones and E Powell
Intensive Care Unit, Morriston Hospital, Swansea, UK
Pulmonary complications are recognised as a consequence of intravenous drug abuse. We present a case series of nine patients who presented to our unit over a 10-week period with striking similarities. See Table 1 for details of cases.
Common issues in our case series include hepatitis B and C, thromboembolic disease, sepsis, pneumonia, cavitating lung lesions and pneumothoraces. All but one patient required invasive ventilation, four had intercostal drains inserted, two had inferior vena cava filters, two had tracheostomies, one required video-assisted thorascopic surgery and two required surgical evacuation of infected clots.
A variety of organisms were noted in blood, sputum and pleural fluid cultures. Of particular note was the high frequency of E. coli, anaerobes and candida. Two patients had group B streptococcal bacteraemia which is unusual outside the obstetric population and when it does occur, tends to be in elderly patients or those with underlying systemic disease such as diabetes, malignancy or liver cirrhosis.1,2
Intravenous drug abusers are a population with poor general health, at high risk of blood-borne viruses as well as other consequences of their lifestyle. They are 10 times more likely to develop community-acquired pneumonia than the general population and up to 13% of those attending hospital febrile have evidence of septic embolisation.3
Other pulmonary complications include bullous lung disease, pulmonary vascular disease and pneumothorax secondary to septic emboli or to unsuccessful injection into central veins.4 Our patients all exhibited one or more of these problems, but the clustering of cases in a short time period made us wonder if there were any other factors involved.
Local enquiries revealed no similar cases in nearby units, but the community drug and alcohol team reported a new trend of injecting methadone and noted ‘poor injection technique’. Injecting unsterile syrup, which is likely to have been spat out of someone’s mouth, could explain a sudden surge of cases of septic embolisation.
Summary of cases.
GCS: Glasgow Coma Score; DVT: Deep Vein Thrombosis; PE: Pulmonary Embolus.
References
Tuesday 8 December 2015 2CP ‘NICE’ Practice on the ICU: Integrating NICE CG 83 into the intensive care unit daily nursing routine
N Dawson, P Taglucop, Y Li, S Etter, S Frondoza, C Torre, S Bhandari, J Beddow, D Vincent, S John, K Broderick, N Findlay and M Farrow
Critical Care Unit, Royal Surrey County Hospital NHS Foundation Trust, Guildford, Surrey, UK
The deleterious effects of a prolonged stay in an intensive care unit (ICU) are well recognised with these patients presenting particular challenges.1 Not least among these was the identification of the need for intensive rehabilitation once critical illness had been overcome. Following the publication of NICE CG 83, ‘Rehabilitation after Critical Illness’2 as a means of optimising patients’ well-being after a prolonged stay in ICU critical care, physiotherapists took a leading role in adopting this pathway. These changes were embraced less enthusiastically so by the nursing staff.
In order to address this need a rehabilitation document was devised and introduced by our lead Critical Care Physiotherapist at the Royal Surrey County Hospital. However, although present at the bedside of patients in the ICU, it was regarded as a physiotherapist’s document. This became apparent when the ICU follow-up clinic highlighted the fact that the ICU nursing staff were disassociated from the rehabilitation document and made little or no active contribution to the process.
In order to address this, a working group of ICU nurses led by a senior sister was set up to examine what role the nursing staff could take in this process and how this could be achieved. The primary aim at the outset was to re-launch the rehabilitation pathway to nursing staff which encompassed:
Re-education at the bedside Time dedicated at the end of previously booked training sessions Time dedicated at staff meetings A poster promoting the pathway document was displayed in the ICU A notice board was subsequently used to update staff with compliance with the pathway usage and other issues regarding the critical care rehabilitation pathway.
Challenges identified included:
A lack of dedicated time A perception that the document did ‘belong’ to the physiotherapists A lack of understanding regarding use of the document Difficulties in identifying longer term goals particularly with respect to the patient’s needs once they have left the ICU
Identifying these changes resulted in an increase in confidence and compliance with the document and the nurses working group developed the rehabilitation pathway further and started to visit patients on the wards after their discharge from ICU. These visits were aimed at educating the ward nurses in order to ensure that the momentum of the pathway use was maintained and also to maintain the support for the patients. During the first three-month period audited, 25 patients were identified as being eligible for the rehabilitation pathway document with 12 followed up on the ward.
These visits have been invaluable in the continued development of the pathway use together with follow-up clinics, patient diary and information sheet development: all ongoing projects.
Although quantitative auditing of compliance with use of the pathway in ICU and on the general wards is carried out, the group are now looking at how its effectiveness can be audited in a meaningful and robust way.
References
3CP Developing a critical illness management and basic life support course in a resource poor setting
J Barnes1,2
1Anaesthetics and Intensive Care, Kamuzu Central Hospital, Lilongwe, Malawi
2Fistula Care Centre, Bwaila Maternity Hospital, Lilongwe, Malawi
These predominantly simulation-based training courses enable healthcare workers to become familiar with recognising and managing various critical illness scenarios and performing various lifesaving procedures, including cardio-pulmonary-resuscitation (CPR). They form a key part of training for all medical, nursing and allied healthcare professionals in the UK and evidence suggests that such procedural training improves patient care, safety and outcomes.3
The course comprised a series of lectures covering critical illness-based topics, supplemented by a half day, simulation-based, course in which delegates were asked to recognise and manage basic medical emergencies, including performing CPR. The course was taught by a visiting physician who also trained local staff to run the course in the future.
All staff in the centre attended the course, which was delivered over an eight-week period. Participants were provided with a course handbook and a series of accompanying ‘emergency management protocols’ were developed for use on the wards.
Staff feedback was extremely positive and the senior physicians in the unit felt that the programme was a valuable addition to the nurses’ current education programme. The physicians also reported that the nurses had become ‘more aware’ of sick patients on the wards, more able to recognise when patients were deteriorating and more inclined for call for help early. They also commented that the handovers being given in such situations were more concise and relevant than previously.
We developed a simulation-based education programme to fill this gap and piloted it in the unit. Uptake and feedback were extremely good; however, formal data on student performance and feedback were not collected, which is a limitation of this initial trial.
Local staff have been trained to continue to run the programme annually and we are planning to undertake formal data collection and analysis regarding student feedback and knowledge/skill acquisition in the future. If successful, we are hoping to expand the programme to other sites in Africa.
References
4CP Observational cohort study of patients admitted to a specialist centre for weaning and rehabilitation
A Warren1,2, T Sanctuary2, D Camilleri2, PB Murphy2, BC Creagh-Brown3 and N Hart2
1Guy’s, King’s and St Thomas’ School of Medicine, King’s College London, UK
2Lane Fox Respiratory Unit, Guy’s & St Thomas’ NHS Foundation Trust, London, UK
3Intensive Care Unit, Royal Surrey County Hospital, Guildford, Surrey, UK
A third of mechanically ventilated patients fail an initial spontaneous breathing trial with 7% remaining invasively ventilated for more than a week after the initiation of respiratory weaning.1 Whilst many patients are weaned in general intensive care units, such patients are increasingly being referred to specialist weaning and rehabilitation centres. This study reports eight years of experience from a tertiary specialist weaning centre that promotes a multidisciplinary approach to weaning, with the use of non-invasive ventilation as a bridge to weaning, intensive physiotherapy and personalised, patient-centred weaning strategy.
The study was performed of patients admitted for weaning and rehabilitation between February 2005 and July 2013. Demographic, clinical and outcome data were collected from paper and electronic patient records. Outcome measured were weaning success, time-to-wean, in-hospital mortality, and 12-month mortality. Weaning success was defined as being free from any form of invasive respiratory support at the time of discharge from hospital.
Two hundred and sixty-two patients were recruited: 39.7% (n = 105) had neuromuscular and chest wall disease, 21.1% (n = 55) had post-surgical weaning failure, 19.5% (n = 51) had chronic obstructive pulmonary disease (COPD) and 5.3% (n = 14) had obesity-related respiratory failure. Weaning success was observed in 168 patients (64.1%). The commonest outcome was successful weaning to self-ventilation (38.2%), followed by weaning to nocturnal non-invasive ventilation only (24.0%). Patients with obesity-related disease were more likely to be weaned successfully than other diagnostic groups (RR = 1.48, 95% CI 1.35 to 1.77; p < 0.001), but otherwise there were no effects of diagnosis on weaning success. Thirty-eight patients died during the admission (14.5%), with mortality highest in patients with COPD (RR = 2.15, 95% CI 1.19 to 3.91, p = 0.012) and lowest in those with neuromuscular or chest wall disease (RR = 0.34, 95% CI 0.16 to 075, p = 0.007), who were the group most likely to be discharged on permanent tracheostomy ventilation. The median time-to-wean was 19 days (IQR 9–33). Cumulative survival at six months was 71.7% and at 12 months was 61.8%, with no difference observed in long-term survival between the different diagnostic groups.
This is the largest cohort of difficult-to-wean patients reported to date. Patients with prolonged weaning should be considered for referral to a specialist unit with both weaning success and in-hospital mortality improving since our report over a decade ago.2 Weaning centres are central to NHS England’s plan as part of specialist complex home ventilation commissioning and indeed has the potential to provide significant cost savings.3,4
References
5CP Therapeutic hypothermia or targeted temperature management after cardiac arrest? An audit of current practice in intensive care units in the North West of England
A Power1 and R Markham2
1Lancaster Medical School, Lancaster, UK
2Royal Lancaster Infirmary, Lancaster, UK
The use of therapeutic hypothermia following cardiac arrest has been common practice in the UK for more than ten years after two multi-centre trials1,2 revealed more favourable neurological outcome in patients who were cooled. The International Liaison Committee on Resuscitation (ILCOR) currently recommends that ‘Unconscious adult patients with spontaneous circulation after out-of-hospital cardiac arrest should be cooled to 32°C to 34°C for 12 to 24 h when the initial rhythm was ventricular fibrillation (VF)’.3 However, a recent randomised clinical trial found no additional benefit in cooling patients to 33°C as compared to 36°C and suggested that ‘targeted temperature management’ may minimise some of the adverse effects of hypothermia such as immunosuppression and cardiac arrhythmias.4 Furthermore, a large proportion of patients in the standard treatment group of one of the original trials developed pyrexia2 leading researchers to propose that prevention of hyperthermia was key to improvements in neurological outcome.4
The aim of this audit was to determine whether intensive care units in our region had changed their temperature management after cardiac arrest in the light of recent evidence, in order to support a change of protocol locally. We performed a standardised telephone audit of 15 adult intensive care units across the north west of England, all of whom agreed to participate in the survey.
Three units (20%) did not have a protocol for cooling after cardiac arrest. Two of these hospitals had previously cooled patients to 32–34°C but had stopped using their protocol in light of evidence from the targeted temperature management trial. Of the 12 hospitals that used cooling therapies, seven hospitals (58%) treated patients at 32–34°C and five hospitals (42%) treated patients at 35–36°C. The five hospitals that treated at 35–36°C had previously treated at 32–34°C but had recently changed their protocol. Thus, since the publication of new guidelines, a total of 7 out of 15 hospitals (47%) had changed their cooling procedures. Of these seven, five hospitals did not have their new protocol officially published as a guideline.
This audit demonstrates a level of inconsistency in the protocols currently used by intensive care units in this region regarding post-cardiac arrest temperature management: cooling at 32–34°C, cooling at 35–36°C, or not cooling at all. This may be because current guidance does not reflect most recent evidence. Of note, three units locally did not utilise cooling therapies after cardiac arrest, potentially risking hyperthermia in patients.
More explicit guidance nationally is needed to standardise care of patients following cardiac arrest.
References
6CP Out-of-hospital cardiac arrests and emergency coronary angiography
D Slessor and A Ashton
Department of Intensive Care Medicine, North Hampshire Hospital, Basingstoke, UK
The rate of survival to discharge from an out-of-hospital cardiac arrest (OOHCA) is ∼10%. In patients with OOHCA, improved survival rates have been reported in patients who have a ST-elevation myocardial infarction (STEMI) who undergo percutaneous coronary intervention (PCI). The evidence for patients without STEMI is unclear.
The aim of this paper was to review our current management of patients following an OOHCA with ventricular fibrillation (VF), to determine how sensitive and specific our current practice is at ruling in and out stentable lesions and to determine whether staff have a consistent approach.
Patients were identified retrospectively, over a six-month period, through the Intensive Care National Audit & Research Centre database using diagnostic codes. All notes were reviewed to determine if the patient had an angiogram performed, whether this was an emergency procedure and if a PCI was performed. Sensitivity and specificity was determined for our current practice at ruling in and out stentable lesions. Survival was determined at six months.
Cardiologists and Emergency Physicians were shown the ECG of each patient, blinded to the patient identification. They were given a short clinical history and asked to decide if the patient should have undergone an immediate angiography.
Twenty-seven patients were included. Twelve patients had an immediate angiogram of whom eight had a PCI. A further four patients had a delayed angiogram, of whom three had PCI performed. The sensitivity and specificity of our current practice for ruling in and out a stentable lesion was 73% and 75%, respectively. Emergency physicians and Cardiologists were 91% and 78% sensitive, respectively, and 54% and 78% specific. There was a wide variation in practice within each group of specialists.
Less than 50% of patients who had an OOHCA from VF underwent immediate coronary angiography. In those patients who did not undergo immediate coronary angiography, a delayed angiography was performed in 27%, of whom 75% had a PCI. Current practice resulted in a sensitivity and specificity that was too low to rule in or out stentable lesions.
Emergency Physicians would arrange for more patients to have immediate angiography following cardiac arrest compared with Cardiologists. This would result in more patients with stentable lesions having PCI performed immediately but would result in more cases of immediate angiography being performed in patients who do not have stentable lesions.
7CP E-CPR in children: Are survival and neurological outcome predictable?
I Mennes1, T Vanassche2, P Vanelderen3, P Skippen4 and A Pitfield4
1Department of Anesthesiology, Algemeen Ziekenhuis Sint-Elisabeth, Herentals, Belgium
2Department of Cardiology, Universitaire Ziekenhuizen, Leuven, Belgium
3Department of Anesthesiology, Intensive Care Medicine, Emergency Care and Pain Therapy, Ziekenhuis Oost-Limburg, Genk, Belgium
4Department of Intensive Care, BC Children’s Hospital, Vancouver, Canada
There were no significant differences (p > 0.05) between the survivor- and the non-survivor group regarding the duration of CPR, the initial lactate, the initial pH, the duration of ECLS and the average NIRS on ECLS. In our series, these clinical parameters were not found to be significant outcome predictors in a logistic regression model (p > 0.05).
Reference
8CP The impact of focused intensive care echocardiography on clinical management of critically ill patients
DP Hall, H Jordan, SR Alam and MA Gillies
Department of Intensive Care Medicine, Royal Infirmary of Edinburgh, Edinburgh, UK
The utility of trans-thoracic echocardiography in the clinical assessment of critically ill patients is increasingly recognised.1 The introduction of Focused Intensive Care Echocardiography (FICE) training by the Intensive Care Society/British Society of Echocardiography provides an accelerated route to competency in conducting a focused echocardiographic assessment of the critically ill patient.2 We sought to investigate the impact of FICE studies on the management of patients in our department.
We conducted a prospective, observational study evaluating the utility of the FICE protocol in the management of critically ill patients in a 30-bedded tertiary critical care unit. Scans were carried out either by or under direct supervision of FICE-accredited physicians. Consecutive patients admitted to critical care during the study period (Aug 2014–Jun 2015) who received a FICE study as part of their management were included. All scans were trans-thoracic and carried out using a Vivid I cardiovascular ultrasound system (GE Healthcare, UK) or M-Turbo ultrasound system (Sonosite, UK) according to the FICE image set.2 Indication for performing echocardiography, image quality, number of satisfactory views, findings and changes in management as a result of the study were prospectively recorded.
Sixty FICE studies were attempted during the study period; of these, 32 were undertaken by clinicians accredited by FICE (26 by ICM consultants, 3 by ICM specialty registrars and 3 by cardiology specialty registrars) and 28 by ICM specialty registrars supervised by FICE mentors. Forty studies were undertaken on invasively ventilated patients, 10 on patients receiving either non-invasive ventilation or high-flow nasal cannula oxygen and 10 on patients receiving no respiratory support besides facemask oxygen. A complete FICE scan was obtained in 43/60 studies (all four echocardiographic windows specified by FICE), three windows were obtained in 8/60 studies, two windows in 5/60 studies and one window in 1/60 studies. The scanning clinician rated the quality of the view obtained as good or adequate 52/60 times.
The most common diagnosis was left ventricular dysfunction (16/60 studies), followed by hypovolaemia (15/60 studies), pericardial effusion (7/60 studies), pleural effusion (6/60 studies) and right ventricular dilatation (4/60 studies). FICE scans led to a change in management in 28/60 cases (47.7%); this included IV fluid therapy (11 cases), diuretic therapy (6 cases), initiation of vasopressors/inotropes (5 cases), further diagnostic procedures, e.g. CTPA (5 cases), pericardiocentesis (2 cases), treatment of acute coronary syndrome (2 cases), initiation of beta-blockers (1 case), treatment of pulmonary embolus (1 case) and commencement of antibiotic therapy (1 case). A further formal echocardiographic study by a BSE-accredited sonographer was requested following 25/60 FICE studies.
This study suggests that focused echocardiography delivered by FICE-trained critical care clinicians in our unit provided additional clinical information which led to a change in management following almost half of conducted studies. Despite this, a full BSE level study was necessary in 41% of patients, suggesting that specialist input from a suitably accredited individual is required in a significant number of situations.
References
9CP An audit of temperature management in comatose survivors of cardiac arrest
B Millette and N Feely
Department of Intensive Care, Wexham Park Hospital, Frimley Health NHS Foundation Trust, Slough, UK
The importance of controlling temperature in comatose survivors of cardiac arrest has been established by large randomised controlled trials.1,2 The target temperature (33°C vs. 36°C) is still a matter of debate, but there is general agreement that pyrexia within 48 h of arrest is associated with a worse prognosis3 and should be actively avoided. We observed variable practice in our institution and decided to audit temperature management in patients admitted to our intensive care unit following cardiac arrest.
We conducted a retrospective audit over six months collecting data using our electronic health record system and comparing our practice to local guidelines. Patient records were assessed to see if they had targeted temperature management (TTM) using the arctic sun cooling device, what target temperature was used (if any) and whether hyperpyrexia occurred in the first 48 h following cardiac arrest. A patient was deemed to be not a candidate for TTM if any exclusion criteria from the local guidelines were present. The gold standard was defined as 100% compliance to all audit criteria.
In sum, 308 patients were admitted to intensive care during the six-month audit period. Of these, 30 were post cardiac arrest (9.7%). Overall mortality was 70% in this cohort; 17 patients (56.7%) were not candidates for TTM. Two patients (6.7%) had insufficient information in the notes to determine whether or not they were candidates. Six patients (20%) received TTM using the arctic sun device. Of these, only two had a target temperature specified in the notes, five achieved target temperatures for at least 24 hand none had a temperature greater than 38°C in the first 48 h. Five patients (16.7%) who were candidates for TTM did not receive it. Of these, two had a temperature greater than 38°C in the first 48 h.
This audit demonstrates considerable room for improvement in managing the temperature of these patients. Of the five patients who did not receive TTM, two became febrile which is associated with a worse prognosis. Even in patients who received TTM there were deficiencies in specifying a target temperature in the notes and in achieving the target temperature for an appropriate length of time. We are now in the process of updating our management guidelines and rolling out a staff educational programme to address these issues and thus improve patient outcomes. We are also conducting a follow-up audit of the management of cardiac arrest to further investigate wider factors.
References
10CP ICNARC defined ‘Post Unit hospital deaths’ is not a reliable quality indicator
H Scott, P Tylerand and D Higgins
Critical Care, Southend University Hospital, Southend on Sea, UK
‘Guidelines for the Provision of Intensive Care Services’1 states that post-intensive care unit (ICU) in-hospital survival is an important indicator that may provide insights into the quality of ICU rehabilitation, appropriateness of ICU discharge, quality of care on the wards and end-of-life care decision making. Participants of the Intensive Care National Audit and Research Centre’s (ICNARC) case-mix programme receive regular ‘Quality Indicator and Outcomes’ reports which identity patients who die in-hospital after Critical Care discharge and this facilitates local audit. We have conducted an audit using this data to assess if ICNARC defined ‘post unit hospital deaths’ could be used as a reliable quality indicator.
Over a six-month period (April–September 2014), ICNARC reported 22 ‘post unit hospital deaths’ at our hospital which they defined as ‘unit survivors who die before ultimate discharge from an acute hospital excluding those discharged for palliative care’. Two clinicians reviewed each case to assess if they actually fulfilled the ICNARC definition.
Twenty-two (13 male: 9 female) of the 225 admissions (mean age 74.5) died on the wards following ICU discharge. Case reviews confirmed that 9 (41%) of these patients did not fulfil the ICNARC criteria (8 were discharged for palliative care and 1 died in a nursing home). The records of the remaining 13 patients who were discharged without palliative care in place were reviewed. None of these deaths appears to have been avoidable in this elderly population with significant comorbidities, and most deaths resulted from progressive cardiovascular, respiratory, and malignant tumour disease which is consistent with the reported frequent causes of post-ICU death in hospital.2 It was noted that 4 of these 13 patients were also commenced on palliative care by the intensive care consultants shortly after ICU discharge as part of their outreach/follow-up duties and in consultation with the responsible specialist teams. Eight of those who died on the wards also had agreed limits put in place at the time of Critical Care discharge and in all cases they include both decisions not to resuscitate and not to readmit.
Our main finding was that at least 41% of our post unit deaths were incorrectly classified despite the information being available in the patients’ notes. If we also include those where palliative care was formally initiated on the ward shortly after discharge as a result of our outreach activity, this would be as high as 59%. Data entry is clearly a problem and our conclusion is that ICNARC defined post unit deaths is an unreliable indicator of quality for our hospital. This has raised concerns regarding lack of data validation. We do not know if this is purely a local problem or whether this is more widespread involving other units participating in the case-mix programme.
References
11CP Assessment of the impact of a new ventilator in achieving compliance with lung protective ventilation
S Brooke1, R Hunt2 and P MacNaughton2
1Peninsula Medical School, University of Plymouth, UK
2Intensive Care Unit Derriford Hospital, Plymouth, UK
There is evidence that lung protective ventilation with lower tidal volumes (6–8 ml/kg predicted body weight, PBW) should be applied to all critically ill patients subjected to mechanical ventilation and not restricted to patients with the adult respiratory distress syndrome (ARDS).1 A number of studies have confirmed that compliance with low tidal volume ventilation even in patients with ARDS is difficult to achieve.2 This investigation assessed the introduction of a new ventilator (Dräger V500®) to all bed spaces in a 26-bed intensive care unit (ICU) on achieving compliance with lung protective ventilation in all ventilated patients. The V500 mandates the insertion of the patient’s height at start up resulting in the calculation of predicted body weight and allows continuous display of tidal volume in ml/kg PBW. In addition, the default start up tidal volume was set at 7 ml/kg PBW. The V500 was introduced to all bed spaces in October 2012. Before this date, the local protocol was to derive PBW from a chart attached to the side of the ventilator and to manually calculate the tidal volume with a guide target of 6–8 ml/kg PBW.
Data were collected retrospectively from the ICU electronic patient record database (Dräger Innovian®). Tidal volumes are automatically stored within the patient record at each change and at hourly intervals. The tidal volume recorded at 7 am was taken as representative for the previous 24-h period. Other data recorded included the patient’s height. Data were analysed on all patients ventilated between October 2010 and July 2014. Tidal volume data were expressed as ml/kg PBW using published formula3 and the percentage of observations within the ranges <8, 8–10 and >10 ml/kg PBW calculated (Table 1).
Data from 2192 ventilated patients were analysed and summarised.
References
12CP A novel method of evaluating critical care fellow performance compared to their peers in a busy multifaceted fellowship using data envelopment analysis
Avinash B Kumar and Vikram Tiwari
Vanderbilt University, Nashville, TN, USA
References
13CP Differing family behaviours in organ donation: A direct comparison between donations after brain death versus donation after circulatory death
N Kalsi, D Harvey, D Gardiner, C Buss, L Hogg and R Ballington
Queen’s Medical Centre, Nottingham, UK
It is well recognised that families act differently in donations after brain death (DBD) versus donation after circulatory death (DCD). One striking difference appears to be that families in DBD do not stay for the offering process and leave following their consent for donation. In effect this means they are, on average, leaving approximately 12 h before a family who stays for DCD.
This difference is believed to reflect the acceptance of when death occurs and the impact on consent for donation and potential subsequent family withdrawal of that consent.
To date, research comparing DBD and DCD has been predominantly qualitative and there is a need to consider how quantitative assessments can be made, so family behaviour can be better understood and supported in these scenarios.
Between April 2014 and March 2015 in Nottingham University Hospitals NHS Trust, we have, as a service evaluation, recorded the behaviour of all consented donor families as to whether they choose to stay or leave following their consent for organ donation.
In total during this period, 42 families consented for donation (9 DBD and 33 DCD) resulting in a total of 25 donations (9 DBD and 16 DCD); 8 out of the 9 DBD families (90%) did not stay, whereas only 6 (18%) DCD families did not stay. Two families withdrew their consent for donation; both from a potential DCD.
It is clear that there is a marked difference in the behaviour of families in DBD versus DCD. It is likely that there is an earlier acceptance of death in DBD, which allows families the opportunity to say their goodbyes and leave.
If we are to improve DCD consent and reduce withdrawal of that consent during the offering process, we need to better understand the decision we are asking of families and the impact this will have on them and their perception of the end of life care.
From April 2015, we plan on recording the behaviours of families who do not consent to donation to explore this difference further and gain a better understanding. We hope other hospitals will consider recording similar information to allow us a better understanding and in time, improve organ donation.
14CP A PATH to improve outcomes following ROSC after cardiac arrest?
N Madhavan, D Horner and J Mcleavey
Salford Royal NHS Foundation Trust, UK
Mechanically ventilated survivors of cardiac arrest account for 6% of all intensive care unit (ICU) admissions within the UK.1 Recent literature has suggested several distinct interventions may benefit these patients when robustly delivered. These measures have been nationally recommended by the ICS since 20082 and now form core recommendations in GPICS.3 However, lack of a consensus ‘bundle’ and regional variations in specialty support/clinical practice are challenges to reliable delivery.
We sought to locally assess compliance with these recommendations and ascertain institutional mortality rates for survivors of cardiac arrest. A five-year, retrospective, single-centre cohort study. We collated descriptive and physiological data (over 48 h) on all patients presenting to the Emergency Department (ED) in cardiac arrest, experiencing cardiac arrest during ED stay or admitted to intensive care as a result of cardiac arrest. Deaths in the ED were excluded. Data were extracted from online EPR systems and hard copy nursing charts.
Totally, 168 cases were retrieved; 137 patients were admitted to the intensive care unit and 97 of these cases were actively managed, with a mean age of 56.6 years. Survival to discharge and six-month follow-up were 30.9% and at 28.9%, respectively; 37.1% of actively managed cases were emergently discussed with off-site cardiology; only a single patient was taken for primary percutaneous coronary intervention (PPCI). Thrombolysis was recommended in nine additional cases and ACS medication in 25.
Following ICU admission, patients remained hyperoxic on average, with mean PO2 at 16 h approximately 20 kPa (Figure 1). Mean temperatures were well regulated between 35°C and 36°C throughout the initial 48 h of critical care, but significant variability occurred; at 16 h temperatures ranged from 31°C to 39°C in extreme cases. Mean blood sugar measurements only fell below 10 mmol/L after 8 h and significant variability again persisted throughout initial management. 24.7% of patients had witnessed seizure activity; 11.5% of these were left untreated.
Significant individual variation exists regarding delivery of evidence-based interventions for the patient following cardiac arrest. Local quality improvement measures should focus on delivery of PPCI where indicated (P) and highlight the benefits of avoiding hyperoxia (A), targeting temperature management (T) and hyperglycaemia/seizure control (H). Regional consensus and improvement science projects may achieve standardised care and have the potential to improve outcomes.
Average oxygen levels post ROSC.
References
15CP Variation in height measurement leads to excess tidal volume ventilation: A study of measurement techniques
FA Amey1, MG Povey1 and CR Bassford2
1Warwick Medical School, University of Warwick, UK
2Critical Care Unit, University Hospital Coventry, Coventry, UK
Males: PBW (kg) = 50 + 2.3 (height (in) – 60)
Females: PBW (kg) = 45.5 + 2.3 (height (in) – 60)
PBW was used to calculate 7 ml/kg tidal volumes for each method. Interclass correlation coefficient (ICC) was used to assess inter-rater reliability for each method.
References
16CP Early rehabilitation of burns specific patients in critical care compared to critical care population: An audit of current practice and limitations
F Howroyd, T Lea and D McWilliams
Queen Elizabeth Hospital Birmingham (QEHB), Birmingham, UK
To compare rehabilitation of a ‘general’ critical care to burns specific populations in critical care at QEHB. To identify limitations to rehabilitation within Burns Critical Care at QEHB.
For those patients not mobilised 25% were deemed too unwell. However, no reason was stated or apparent for the lack of mobilisation in 34% of patients, whilst 17% did not mobilise due to pain/agitation and 3% due to limited staff at weekends (Table 1).
Results.
Values are median (IQR).
Geometric means and 95% confidence intervals.
References
17CP Room for improvement: An audit of intra-operative mechanical ventilation in theatres at a large district general hospital
A Rutter and R Thompson
Department of Anaesthesia and Intensive Care Medicine, Victoria Hospital, Kirkcaldy, UK
We audited how practice in theatres at NHS Fife compared to the recommendations made by the IMPROVE group.
Thirty-seven patients were anaesthetised for intra-abdominal surgery; 13 patients received tidal volumes which were 100 ml or more above the 6 ml/kg threshold. Many patients received tidal volumes which were less than 6 ml/kg. Obese patients appear to be ventilated with larger tidal volumes than those with a normal body mass index. Ten patients did not have PEEP delivered during their intraoperative ventilation, eight of whom were undergoing intra-abdominal surgery.
References
18CP Ultrasound technology: Should it be a part of hospital induction for trainees?
K Khoo and S Nandalan
Department of Anaesthesia, Royal Albert Edward Infirmary, Wigan, UK
It is recommended that all those involved in placing CVCs using 2D ultrasound guidance should undertake appropriate training to achieve competence.1
A UK study achieved an absolute risk reduction of 5.9% by the implementation of the above NICE guidelines (pre-implementation complication rates 10.5%, versus post-implementation rate of 4.6%). The reduction in complication rates was larger for specialist registrars than for consultants (reduction of 11.2% vs. 1.6%).3 Anaesthetic trainees in the UK rotate between hospitals and may come across completely unfamiliar US machines, which they may have to use in emergencies and out of hours. We conducted a survey to map current proficiencies in US technology among anaesthetic trainees in the North West Deanery.
Frequency of use: 45% – every other day, 54% – occasionally Indications (multiple choice): 100% – CVC, 79% – peripheral nerve blockade, 36% – arterial lines, 18% – central neuraxial blockade, 13% – chest drains, 11% – transthoracic echocardiography US course: 54% – attended, 46% – not attended Other training by: 82% – consultants, 35% – senior colleagues, 10% – self taught Access to US: 95% had at all times, however Only 14% reported that US machine is part of their induction at hospital changeover US as part of induction: 30% agreed, 29% agreed but concerns with prolonging induction process, 41% disagreed
References
19CP Matching Warrington: Incidence of ventilator-associated pneumonia
T Bower, V Vanek, Z Qazzafi, J Little and J McCann
Warrington and Halton Hospitals NHS Foundation Trust, Cheshire, UK
Ventilator-associated pneumonia (VAP) is defined as pneumonia occurring >48 h after endotracheal intubation and mechanical ventilation. It is the commonest nosocomial infection encountered in the intensive care unit (ICU) with a rate of 9–28%.1 It is estimated to increase hospital stay by 7–9 days and carries a 25–50% mortality.2 However, it is difficult to accurately diagnose and no universally accepted gold standard exists. Warrington Hospital aims to prevent VAP by following a Standard Operating Procedure, which includes four interventions that should be carried out on all ventilated patients:
Elevation of head of bed to 30–45°, Sedation level assessment with daily sedation breaks, Oral hygiene with 6 hourly oral chlorhexidine gel, Tracheal tube cuff pressure to be maintained between 20 and 30 cmH2O and measured every 4 h.
We audited cases of suspected VAP on our ICU over a three-year period at Warrington ICU. Our diagnostic criteria were, new or progressive pulmonary infiltrates on chest X-ray PLUS at least two of the following three clinical features:
Purulent tracheal secretions Leukocytosis > 10,000/mm3/leukopenia < 4000 mm3 Fever greater than 38.3°C or less than 36°C.
We recorded the number of ventilated patients on ICU each day to create number of VAP’s per 1000 ventilator days. Our study also records the introduction of ventilator filters onto our ICU.
Over a three-year period, from April 2012 to April 2015, there were a total of 13 VAPs. We found the incidence of VAP on our ICU was 1.76 per 1000 ventilator days over this period, compared to a literature quoted figure of 5 per 1000 ventilator days.1 The introduction of a ventilator filter corresponded with a drop in infection rate.
The incidence of VAP on our unit is low when compared to the evidence base. We saw an improvement in the incidence of VAP on our ICU following the introduction of a SOP in 2012 for the prevention of VAP. However, in 2014–2015, we have observed an increase, from 0.8 per 1000 ventilator days in 2012–2013, to 3.78 in 2014–2015. The reason for this increase is not clear currently. We may have to look again at sub-glottic aspiration or continuous cuff pressure monitoring. Despite this rise our overall rate remains low.
References
20CP Adults with incapacity: An intervention to improve practice
K Graham and H Tyler
Intensive Care Unit, Forth Valley Royal Hospital, Larbert, Scotland
The Adults with Incapacity (Scotland) Act 2000 safeguards the welfare and finances of people who lack capacity. It protects adults (aged 16 or over) who lack capacity to take some or all decisions for themselves because of a mental disorder or an inability to communicate. All actions under the act must: benefit the adult, be the least restrictive option, take into account the wishes of the person, take account of the views of others with interest in the person’s welfare, encourage the person to exercise existing skills and develop new ones.1 In the intensive care unit (ICU), we treat patients who lack capacity on a daily basis. For the purposes of the Act, patients are deemed incapable if they cannot act upon decisions, make decisions, communicate decisions, understand decisions, or retain the memory of decisions.2,3 On issuing a certificate of incapacity, part five of the Act gives doctors authority to treat patients who are incapable of consenting to medical treatment, where no attorney or guardian has been appointed.2 Only emergency treatment should be given prior to completion of a certificate.
We performed an audit to ascertain if incapacity certificates were being used appropriately in our ICU. All critical care patients (level 1–3) were included. The capacity of the patient was assessed by the auditor and the bedside nurse, with consideration given to fluctuations with delirium. Evidence of a capacity assessment or identification of a welfare attorney was sought in the notes. The presence or absence of a valid capacity certificate was then documented.
The first cycle audited 21 patients over 10 days in October 2014. Results showed 10 (48%) patients lacked capacity, of which 6 (60%) had an incapacity certificate. Eleven patients had capacity, of which two (18%) had an active incapacity certificate.
These results were presented to the department to raise awareness and a new ward round checklist was implemented on ICU in 2015 that includes a decision box looking at whether the patient has capacity and whether a certificate is required. We subsequently performed a second cycle of audit over 10 days in July 2015, including 35 patients. Ten (29%) patients lacked capacity, of which eight (80%) had a certificate. Twenty-five patients had capacity and of those none had an active certificate (0%).
In conclusion, while capacity assessment is challenging, especially when incapacity is temporary or fluctuant, there is a legal requirement in Scotland to identify these patients and issue a certificate to authorise their ongoing treatment. Our experience showed that a simple intervention to prompt daily review of capacity improved our recognition and certification of adults with incapacity. Importantly, it also aided identification of those who regained capacity and needed their certificate retracted. We shall continue our project to raise awareness of this issue amongst staff in our ICU and aim to reach 100% appropriate certification. We also intend to create a document outlining the treatments carried out in our unit under the remit of an incapacity certificate, to provide clarity to staff and relatives.
References
21CP Improving training in end-of-life care and organ donation
DJ Haley1, M Thomas2 and C Booth3
1Intensive Care Medicine & Anaesthesia, Northwest Deanery, England
2Emergency Medicine, Salford Royal NHS Foundation Trust, Salford, England
3Intensive Care Medicine & Anaesthesia, Salford Royal NHS Foundation Trust, Salford, England
Providing adequate training in discussing end-of-life care and organ donation can be difficult due to the sensitive and emotive nature of these situations. It is often not regarded as appropriate to allow trainees to gain experience during what is already a difficult time for patients and their relatives.
We designed and implemented an end-of-life care and organ donation simulation day aimed at providing training for colleagues involved in these areas. Participants included medical and nursing staff from intensive care unit (ICU), Emergency Medicine, Theatres and NHS Blood and Transplant. The day involves high-fidelity simulation scenarios including discussions with patient relatives, played by professional actors. Thorough group-debriefing sessions are then conducted following each simulation. We have provided the course on three occasions so far and have collected and analysed feedback from each participant.
We received feedback from 58 of the 59 participants. As a result of attending the course, 44 (76%) felt their clinical skills had benefited, 56 (97%) reported improved communication skills and 56 (97%) believed the day had helped them gain a new approach to end-of-life care and organ donation. Individual feedback comments were invariably positive and highlighted the benefit of group-debriefing sessions. Pre- and post-course scores also indicated improved participant confidence in the management of several key areas of end-of-life care and organ donation (Figure 1).
The feedback we received from participants demonstrated a general improvement in knowledge and confidence in dealing with difficult situations surrounding end-of-life care and organ donation. Courses such as this one help provide training in an area that is clinically important but often difficult to gain first-hand experience.
Participant confidence (scored 1–5) in the management of each of the following areas. Blue: pre-course mean; Red: post-course mean.
22CP Quality Improvement ITU Patients and Relatives Booklet
S Bruemmer Smith1, S Tun2 and L Jackman3
1Royal Sussex County Hospital, East Sussex
2Worthing Hospital, West Sussex, UK
3Royal Sussex County Hospital, East Sussex, Uk
Reference
23CP The mean arterial pressure to vasopressor dose ratio: An objective metric to assess the effect of cardiovascular interventions
DJ Stubbs, G Singleton and NM Broughton
Intensive Care Unit, West Suffolk NHS Foundation Trust, West Suffolk Hospital, Bury St Edmunds, UK
The PaO2/FiO2 (P/F) ratio is widely used in critical care to compress multiple factors affecting oxygenation (PEEP, patient positioning, recruitment, etc.) into a convenient shorthand to track changes over time. We suggest that an analogous ratio, that of MAP to vasopressor dose (M/V ratio) may be useful in tracking patient response to cardiovascular intervention. In the case of noradrenaline (NA), the ratio is calculated by dividing the MAP (mmHg) by the dose of noradrenaline (µg/kg/min). An increasing M/V ratio represents CVS improvement by either an increased MAP for a fixed dose of NA or a maintained MAP in the face of a falling NA requirement. The opposite applies to a falling M/V ratio.
We illustrate this using hypothetical data modelled on our experience evaluating our use of methylthioninium chloride (methylene blue – Met-Blue) for refractory sepsis. In this setting, patients are invariably on NA (and potentially other vasoactive agents) prior to it being started and we wish to assess whether our intervention has been effective. We model a situation where the MAP increases after administration of Met-Blue (Figure 1(a)). The NA dose is not weaned and therefore both MAP and M/V ratio increase. We have a situation where MAP is targeted (Figure 1(b)); the NA dose is weaned to keep the MAP in the region of 65 mmHg. The M/V ratio still clearly demonstrates the effect of Met-Blue on the patient’s cardiovascular state.
We suggest the M/V ratio provides a valuable means for clinicians to objectively assess the efficacy of various interventions on the cardiovascular status of patients. It may be used in any context where a specific MAP is targeted and/or the dose of vasoactive agents adjusted; therefore, the M/V ratio may be applied to both clinical practice and also audit/research settings. In the example above, if cardiovascular data from Figure 1(a) and (b) were averaged to assess the intervention across a patient cohort, the opposite changes in MAP and NA dose may mask the efficacy of the intervention, whereas the M/V ratio response is consistent in both settings. It could also act as a patient safety trigger – preventing falsely reassuring MAP values from being seen in isolation without considering the required NA dose (akin to reviewing PaO2 without FiO2). It provides an objective shorthand for the thought processes undertaken when adjusting cardiovascular support and we believe warrants further exploration as both a clinical, research and training tool.
Theoretical data modeling two scenarios. In A the mean arterial pressure (MAP) falls despite an initial increase in noradrenaline dose (NA). At t = 0 an intervention (e. g. Met-Blue) administration occurs. There is a rapid increase in MAP but no weaning of NA dose. The M/V ratio however increases. In B there is an ongoing need for escalatihg NA doses prior to intervention at t = 0. At this point the NA dose is able to be weaned resulting in a static MAP, improvement is evidenced by the improved M/V ratio. In both situations the M/V ratio allows clear signs of effect/improvement.
24CP Glucose control in critical care: Compliance with hyperglycaemia guideline and effect on patient outcome
L Beard1, E Beard2 and M Reay1
1Intensive Care Department, Russell’s Hall Hospital, Dudley, UK
2Department of Epidemiology and Public Health, University College London, London, UK
Glycaemic control in critical care has been a topic of much debate with the optimal target range still unclear. There is no recommendation as to whether tight glucose control should be universally adopted because of conflicting data on patient outcomes coupled with the risk of severe hypoglycaemia.1–3 It is however felt that severe hyperglycaemia (blood glucose measurement (BM) > 10 mmol/l) should be avoided.1
The primary aim of the audit was to assess compliance with a recently introduced local guideline on the management of hyperglycaemia in intensive care. The audit standards were as follows:
100% patients should have their glucose checked at least every 4 h until condition stable (defined by 12 h of BM 4–8 mmol/l). 100% patients should have a laboratory glucose checked if unexpected or unusual StatStrip result. 100% patients should be started on insulin infusion if BM > 8 and known diabetic or BM > 10 and non-diabetic.
The secondary aim was to assess the effect glycaemic control had on patient outcome.
A retrospective search of the electronic notes database (ICIP) was undertaken from November 2014 (following guideline introduction), July 2015. Two patient groups, diabetics and non-diabetics were identified by a search of the pre-populated PMH and free text fields. StatStick glucose, laboratory glucose, administration of insulin infusion, mortality and APACHE II scores were obtained. Hyperglycaemia was defined as BM > 8 mmol/l, adequate control 4–8 mmol/l and hypoglycaemia < 4 mmol/l.
Data were collected on 220 patients of which, 10.5% (n = 23) had diabetes. Only 65% of patients had 4 hourly BMs recorded. Hypoglycaemia occurred in 3% (diabetic 2% vs. non-diabetic 3%) and hyperglycaemia 28% (diabetic 56% vs. non-diabetic 24%) of recordings. 94% of patients had a laboratory glucose checked when indicated.
Eighteen percent (n = 36) of non-diabetics met the criteria for insulin infusion (BM > 10 mmol/l) and had a significantly higher odds of mortality (OR 4.8) than non-diabetics who did not require insulin (p < 0.001). Diabetics who meet the criteria for insulin infusion have a 0.29 lower odds of mortality than non-diabetics (p = 0.053). Thus showing non-diabetics requiring insulin infusion to be a high-risk group.
Of the 36 non-diabetics who met the criteria for insulin infusion only 61% of these patients were treated with insulin. Those on insulin had a 0.69 times lower odds of mortality (p > 0.05). This group was adequately controlled (BM 4–8 mmol/l) for 54.5% of the time during their intensive care unit admission.
Research has shown that stress induced hyperglycaemia in non-diabetics is associated with an increase in mortality, and our data is in agreement with this.1 This group of patients should benefit most from the management of hyperglycaemia. Unfortunately, our audit showed that these patients were only adequately controlled 54.5% of the time, this is in part due to inadequate glucose monitoring and commencement of insulin infusions. This audit highlights the need for increased awareness amongst all staff to the risks of hyperglycaemia especially in non-diabetics. Departmental education and re-audit will be undertaken to increase compliance with the guideline along with a larger search and analysis of the ICIP database.
References
25CP Evidence supporting the use of NG bridles on the ICU: Cost and clinical considerations
C Banks and A Marsh
Department of Anaesthesia, North Bristol NHS Trust, Bristol, UK
Accidental removal of nasogastric (NG) tubes by confused or agitated patients in intensive care unit (ICU) is a recognised complication of enteral feeding. Previous studies have suggested that securing NG tubes with bridles helps prevent unintentional tube dislodgement in the critical care setting.1,2 This quality improvement project was designed to assess the clinical and cost effectiveness of using NG bridles for patients in ICU at our trust.
A trial sample of 10 NG bridles were purchased and placed in patients between December 2014 and January 2015. The patients selected were all those who had at least one accidental removal of an NG tube in ICU. Costs were calculated for chest X-ray (£19.80), NG tube (£11.82) and NG bridle (£103.76).
Data were collated and analysed for each patient to determine whether the bridles were effective in helping prevent tube dislodgement, with each patient followed up to discharge from ICU.
The 10 bridles were placed in nine patients with a range of diagnoses (2 × polytrauma patients, 3 × neurosurgical patients and 4 × medical patients). There were two incidences of patients removing an NG tube with the bridle still attached: in the first patient nursing staff removed the bridle when this happened, and the second occasion the bridle was kept in and re-used. There was an 89% decrease in the number and cost of NG tubes inserted per patient day pre bridle:post bridle insertion (0.45:0.05). There was an 86% decrease in the number and cost of chest X-rays per patient day pre bridle:post bridle insertion (0.35:0.05).
The total cost saving per patient day was calculated as £0.36 pre to post bridle with the cost of the bridle included in post insertion figures. This cost data does not take into account the reduction in nursing time, patient discomfort, missed feed and medication. All of which are consequences of repeated removal and re-insertion of nasogastric tubes.
Key learning points included ensuring the bridle is securely attached with a knot ≤1 cm from the nostril, bridle size and NG tube size must be matched and bridles should be reused following accidental NG tube removal.
This trial sample showed NG bridles are cost effective and clinically effective in preventing the unintentional dislodgement of NG tubes in the ICU. Subsequently, a proforma for suggested bridle use and an insertion guideline, have been developed to allow wider appropriate use of NG bridles in our ICU.
References
26CP Evaluating the effectiveness of communication in ventilator-dependant tracheostomy patients utilising above cuff vocalisation – The ICU Functional Communication Scale
S Wallace, J Lynch, L Nicholson, M Wilson, R Purcell and BA McGrath
Acute Intensive Care Unit, University Hospital South Manchester, Wythenshawe, Manchester, UK
National guidelines recommend early recognition of communication problems in intensive care unit (ICU) and involvement of Speech & Language Therapy (SLT) for rehabilitation of difficulties.1 Ventilator-dependant tracheostomy patients will usually require cuff inflation, excluding the upper airway from gas flow and thereby severely limiting the ability to communicate by vocalisation. Previous studies in ventilated patients have focussed on assessing the psychosocial effects and patient experience of communication difficulties, with frequent use of Likert scales.2 Novel ‘Talking’ Tracheostomy tubes, which do not require cuff deflation, have been studied. These focussed on assessing the acoustic aspects of voice intensity or using subjective methods for producing vocalisation, concentrating on the benefits and drawbacks of these tubes.3,4 Our service introduced above cuff vocalisation (ACV) using standard Smiths-Medical (Ashford, UK) Blue Line Ultra Suctionaid (BLUS) tracheostomy tubes to facilitate functional communication. This novel method of communication allows additional gas flow to be delivered via the subglottic suction port to exit via the larynx. Resulting speech quality is variable in our experience and existing functional assessment scales are not suitable. Existing scales are lengthy, may require detailed training and certification, and are aimed at physical function and disability rather than speech.5 Our aim was to develop and trial a simple tool to evaluate effectiveness of functional communication with (and without) ACV that could be used consistently by SLT and non-specialist staff.
Detailed literature searching confirmed that no existing tools met our requirements. A new scale was devised to evaluate the overall effectiveness of functional communication attempts by ICU patients utilising ACV. Parameters were devised by consensus amongst SLT, nursing, medical and physiotherapy staff who were experienced in communicating with ventilator-dependant tracheostomy patients. Draft versions were trialled in the same multi-disciplinary professional group, across the two ICUs in our tertiary hospital.
The final version of the scale (Table 1) provides a descriptive rating based on the parameters of absence or presence of voice or speech, use of alternative means of communication and whether attempts to communicate are ineffective or effective. The scale was found to be easy to use by all multidisciplinary staff and resulted in identical independent scores by three SLT, medical and nursing staff in five patients using ACV for periods of between two days and three weeks.
The ICU Functional Communication Scale.
Acknowledgements
The authors gratefully acknowledge the assistance of the Global Tracheostomy Collaborative (www.globaltrach.org) in sharing of experience and protocols regarding the use of ACV, especially that of Mrs Tanis Cameron and the TRAMS team, Austin Health, Melbourne, Australia (www.tracheostomyteam.org).
Funding
This work was carried out as part of a project funded by the Health Foundation’s Shine programme. The Health Foundation is an independent charity working to improve the quality of healthcare in the UK. The authors have received unrestricted funding from Smiths Medical to evaluate BLUS tubes for the purposes of ACV in a future study.
References
27CP Early mobilisation and the barriers in intensive care units across Scotland
LG Salisbury1, ME Harrold2,3, SAR Webb3,4 and GT Allison2
1Edinburgh Critical Care Research Group, University of Edinburgh, Edinburgh, UK
2School of Physiotherapy and Exercise Science, Curtin University, Perth, Australia
3Intensive Care Unit and Physiotherapy Department, Royal Perth Hospital, Perth, Australia
4Department of Medicine and Pharmacology, University of Western Australia, Perth, Australia
Early mobilisation is increasingly being advocated as a treatment modality in intensive care units (ICUs)1,2 and has been shown to be both feasible and safe.3 At present, there is little understanding of the rates of early mobilisation practise in Scotland or the barriers that prevent mobilisation being carried out.
An observational service evaluation was carried out in Scottish ICUs. Ethical approval was not required, but Caldicott Guardian approval was obtained to allow access to identifiable data. Patients were included if they received mechanical ventilation during their ICU stay. A standardised form was used to record each episode of mobilisation that occurred during each 24-h period. Mobilisation was classified as sitting, standing, tilt table or ambulation. Any adverse events that occurred during the mobilisation were recorded. If a patient was not mobilised during a 24-h period the barrier to mobilisation was recorded, more than one barrier could be recorded each day. There was a four-week inception period and patients admitted during this period were followed up for four weeks or until ICU discharge, whichever occurred first.
Data were collected in nine ICUs across Scotland. Full datasets were analysed for 167 patients. Fifty-eight percent were male, the average age was 55.6 ± 17 years, average APACHE II was 18.2 ± 7.2, median ICU length of stay was 3.7 (1.9–7.9) days and median length of hospital stay was 15 (6–29) days. In total, 68/167 (40%) participants mobilised during their ICU stay. There were a total of 263 episodes of mobilisation carried out (patient average 0.4 episodes per day) with 17 adverse events reported. On average, the median day first mobilised was 3.5 days. In total, only 2% of mobilisation episodes occurred while an endotracheal tube was in situ; this included eight episodes of sitting out of bed and two episodes of standing. For the 60% who did not mobilise, there were a range of barriers that are summarised in Table 1.
Barriers to mobilisation.
CVS: cardiovascular system; CNS: central nervous system; ETT: endotracheal tube.
References
28CP Seizures in out-of-hospital cardiac arrest survivors – A poor prognostic indicator
SS Kumar1, C Downes2, T Leary1 and S Hutchinson1
1Intensive care, Norfolk and Norwich University Hospital, Norwich, UK
2University of East Anglia, Norwich, UK
Twenty-nine (38%) demonstrated epileptiform activity during their stay ranging from first hour to 92 h post cardiac arrest. The mean and median duration to seizures was 25 and 24 h, respectively. More than 50% occurred on sedation hold and in the others sedation was prolonged because of seizures. All patients with seizures were received at least three anticonvulsants. Thirteen (45%) of them had computed tomography scan of brain, only one suggested hypoxic injury. Eleven (38%) of them had electroencephalograms of which only nine were conclusive of hypoxic brain injury. Four patients underwent MRI scans and three of these demonstrated hypoxic brain injury. The ICU mortality in this subgroup was 72% (21) and 18% (5) were discharged to ward in a vegetative state.
In the non-seizing group of 49 (62%) cardiac arrest survivors, the ICU mortality was 40% (20) and 51% (25) had good neurological recovery with cerebral performance category scale of two or less.
References
29CP Noise levels in ITU: A comparison between two critical care units
J Blackburn, R Hayes, V Ragothaman and N Arora
Heart of England NHS Trust, Birmingham, UK
Environmental factors including sleep disturbance are linked to intensive care unit (ICU) related delirium,1 which increases morbidity and mortality for patients.2 World Health Organization (WHO) guidelines recommend that overnight sound levels in hospitals should not exceed peak levels of 40 dB (decibels).3 Within this hospital trust, there are two critical care units. The first (BHH) consists of two open plan wards with 11 levels 2 and 3 beds in the intensive therapy unit (ITU) and 8 level 2 beds in the high dependancy unit (HDU). Only three of the beds on the ITU are in side rooms, the rest are in open plan wards. The second hospital (GHH) has one critical care ward with 12 mixed level 2 and 3 beds, each of them in their own separate side room. A recent study4 of five UK ICUs found that all units studied had sound levels greater than WHO recommendations.
The aims of the audit were:
To measure sound levels at patient’s bedside overnight to discern if they are above the WHO recommendations. To compare sound levels between the two hospitals, to determine if the sound levels are higher in a critical care unit with an open-plan format, compared with a unit that consists of only side rooms.
Sound levels were recorded once at each patient’s bedside between the hours of 11 pm to 3 am on three consecutive nights. These were recorded using a smartphone application (SPL Meter). These recordings were undertaken in each critical care unit.
Mean noise levels were 59.16 dB for BHH and 48.64 dB for GHH. See Table 1 for nightly means at both sites. Every sound level recorded was above the WHO recommended maximum of 40 dB, and over 70% of recordings were above 50 dB.
Our data show that the noise levels in ITU at night are higher on average than the WHO recommended maximum; this concurs with previous study findings in the UK.4 Comparing the two sites, we have found that the sound levels at night are nearly 100 times higher on average in the critical care unit with an open plan ward design (59.16 dB) compared to the unit with only side rooms (48.64 dB). This could lead to an increased risk of ICU-related delirium in the unit with an open plan, and could possibly cause an increased morbidity and mortality, which would be an interesting area for further research.
Comparison of mean noise levels in the intensive care units at two hospital sites.
References
30CP An audit of discharge practice across the Greater Manchester Critical Care Network
AJ Wilson1, J Ebbs1, A Stevens2 and J Eddleston1
1Critical Care Department, Central Manchester Foundation NHS Trust, Manchester, UK
2Greater Manchester Critical Care Network, Manchester, UK
The recently published Guidelines for the Provision of Intensive Care Services1 (GPICS) and the NHS England Service Specification for Adult Critical Care (D16)2 recommend that discharge from critical care should be timely (ideally less than 4 h) and that night-time discharges (between 10 pm and 7 am) should not occur.
NHS England’s Critical Care Clinical Reference Group has set the elimination of delayed discharges (>24 h from when the patient is ready to leave critical care) as a national priority for the service in 2015–2016. Efficient, timely discharge from critical care promotes rehabilitation and discharge planning for the patient concerned and in the wider sense, minimises the number of elective surgical patients who have surgery deferred due to lack of a critical care bed.
We have used data collected by the North West Intensive Care Bed Information Service to audit patient admissions and discharges to critical care units across the Greater Manchester Critical Care Network. Data were collected for one calendar month (10 June–9 July 2015) from 20 critical care units representing 576 ‘unit-days’.
There were 1143 admissions to critical care beds during this time period with 258 admissions occurring at night. At least one staffed, empty, critical care bed was available overnight on 54% (311/576) of the unit-days surveyed.
There were 54 night-time discharges from critical care which corresponds to a night-time discharge on 9% (52/576) of unit-days. On 30% (16/52) of these occasions, the unit concerned appeared to have sufficient capacity to accommodate night-time admissions without the need to discharge a patient.
There were 283 delayed discharges (a delay of over 24 h from when the patient was ready to leave critical care). A delayed discharge occurred on 28% (163/576) of the unit-days surveyed. At an estimate of £1000 per delayed discharge, this represents £283,000 of inappropriate costs paid for by commissioners each month or over £3.3 million per year.
We conclude that delayed discharges from critical care and inappropriate night-time discharges despite capacity remain common in our region. Further work is required to understand the barriers to timely, day-time discharge and the impact this has on elective surgical work.
Acknowledgements
We are grateful for the help with data collection provided by the North West Intensive Care Bed Information Service.
References
31CP Developing a national framework for critical care nurse education
M Kynaston1, A Berry2 and A Himsworth3
1Cheshire & Mersey Critical Care Operational Delivery Network, Cheshire and Merseyside, UK
2Greater Manchester Critical Care Network, Greater Manchester, UK
3Midlands Critical Care and Trauma Networks, The Midlands, UK
Since the emergence of intensive care units in the UK, it has been recognised that there is a need to ensure critically ill patients are cared for by professional nurses that have the appropriate specialist skills and knowledge to assess, implement and evaluate patients and the related care interventions they undertake.
Following the dissolution of the English National Board (ENB) in 1992, it became apparent that educational provision for post registration critical care courses lacked consistency across the UK. Variations included curriculum content, skills and competence acquired and academic awards gained all of which resulted in lack of transferability between critical care services.
To provide a structure able to support equity of care delivery, the Critical Care Network Lead Nurse Forum (CC3N) developed a working subgroup in 2008 to define standards required to underpin critical care nurse education.1 The membership of this group included colleagues from the British Association of Critical Care Nurses, Royal College of Nursing, Independent Sector Hospitals and Higher Education Institutes charged with the delivery of post registration critical care courses. The standards published in 2012 described the essential components of a post registration award in critical care nursing, which included a minimum academic award and assessment of competence.
A second piece of work to develop a competency framework to underpin educational courses using a three steps approach2 was completed in 2013. This framework is a collection of core competencies identified as being fundamental to the delivery of high-quality, safe and effective nursing care which reflects the typical levels of a nurse in critical care would pass through. The introduction of these step competencies provides a transferable document of evidence for each individual nurse which should ultimately act as an assurance to demonstrate their ability to work as a competent critical care nurse and should be an enabler to allow for greater movement and flexibility of the nursing workforce.
Collectively, the educational standards along with the competency framework provide a clear, cohesive and structured approach to direct the commissioning of post registration critical care courses and offers assurances that courses meet service requirements. The educational standards and competency framework have been incorporated in to the Core Standards for Intensive Care Services3 and the subsequent Guidelines for the Provision of Intensive Care Services4 which provides recognition of their importance in terms of delivering high-quality services.
The working subgroup has recently reviewed and updated the competency framework to ensure it reflects the most current evidence base and remains fit for purpose and have now embarked on updating the educational standards.
This work is widely available to all through the CC3N website and has resulted in a number of international enquires from countries including the USA, Finland and United Emirates who want to adopt or utilise the work themselves.
References
32CP Meeting nutritional targets in intensive care patients: An audit of current practice
R Gale1, S Harwood1, P Turner2 and A Iyer1
1Intensive Care Unit, Royal Liverpool University Hospital, Liverpool, UK
2Department of Nutrition and Dietetics, Royal Liverpool University Hospital, Liverpool, UK
Within 5 h of admission, 42.9% of patients had an nasogastric tube (NG) sited but 25% had a delay in excess of 16 h prior to NG insertion. Once sited, 36% of patients had a further delay of greater than 5 h before feed was started. Only 30% of these patients had a documented medical reason for this.
Seven patients achieved or exceeded their daily calculated energy requirement, based on 20 kcal/kg/day. The remaining 19 patients had an average daily energy deficit of 573 calories, range 122.1 to 1842.6 kcal. One patient had a deficit of 11,055 kcal by day 6 of ICU admission.
One break in nutritional delivery post feed initiation occurred in 82% with an average number of 2.4 breaks per patient (range 0–7). The commonest reasons for feed disruption were patient factors (30%), procedures (27%) and NG displacement (22.7%). In total, there were 117 h of missed feeding due to NG displacement.
Recommendations
Improved education of ITU staff regarding importance of calculating patient weight and a daily calorie target. To minimise delays in placement and re-insertion of NG tubes by following National Patient Safety Agency guidance and increasing use of aspirates rather than X-ray. To consider a trial of a daily calorie target approach rather than ml/hr protocol.
References
33CP Real-time oxygen consumption monitoring – A new technology to improve critical care
PA Robbins1, DP O’Neill1, L Ciaffoni2, J Couper2, G Hancock2 and GAD Ritchie2
1Department of Physiology, Anatomy and Genetics, University of Oxford, Oxford, UK
2Physical and Theoretical Chemistry Laboratory, University of Oxford, Oxford, UK
Existing technology does not permit accurate and continuous, non-invasive measurement of oxygen consumption in critical care settings where the inspired oxygen content is often elevated and not fixed. This report describes a new measurement device for analysing respiratory gas flow that enables accurate measurement of oxygen consumption under such conditions.
O2, CO2, H2O vapour and N2 tidal flows during 2 min of breathing. Overall slopes indicate consumption (up) or production (down).
We have developed an entirely new approach to respiratory gas analysis, using laser absorption spectroscopy. The compact, main-stream technology we have recently developed makes use of laser absorption spectroscopy for sensitive and selective measurements of oxygen, carbon dioxide and water vapour concentrations within the airway every 10 ms.1 The in-airway location and 10 ms response speed removes all delay. We have developed a ‘reference pneumotachograph’ which is more than an order of magnitude more accurate than commercial devices. The two technologies are combined so as to make possible the measurement of oxygen consumption in critical care in an easy, accurate, non-invasive and continuous manner.
This device offers a precision of ∼0.2% in calculating the fluxes of oxygen within the airway. The analytical performance is demonstrated by measuring the oxygen consumption and the nitrogen (balance gas) exchange rate – the latter should be zero – under the conditions of breathing air and then pure oxygen in the same subject. The measured nitrogen exchange rate is <10 ml/min under conditions of both air and oxygen breathing. The difference in O2 consumption is <5% between the measurements made during air (Figure 1) and oxygen breathing.2 Data collected from healthy volunteers as well as from ventilated post-operative patients in recovery will be presented and discussed.
We conclude that in-airway molecular flux sensing using laser absorption spectroscopy solves the problem of measuring oxygen consumption accurately, continuously and non-invasively in the setting of critical care. This opens up new opportunities for assessing immediate assessment of therapeutic manoeuvres on oxidative metabolism. Accurate calculation of oxygen consumption also allows the Fick reference method for determining pulmonary blood flow to be used with high precision.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the NIHR Biomedical Research Centre, based at Oxford University Hospitals NHS Trust, Oxford. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.
References
34CP Clinical effectiveness project: Risk stratification using post-operative troponin
J Peters, R Sexton and M Thomas
North Bristol NHS Trust, Bristol, UK
Troponin T (TnT) assay measurement in the post-operative period can be used as an independent predictor of major non-cardiac complications.1–3 We piloted use of TnT as a prognostic marker in patients admitted to the intensive care unit (ICU) of Southmead Hospital, Bristol after laparotomy to define utility in our local population.
In December 2014 and January 2015, all patients admitted to ICU after laparotomy (defined as incision into the abdominal cavity) were included. Fourth-generation TnT assays (normal range less than 14 ng/L) were taken on ICU admission, and until the third post-operative day or discharge to the ward. Results were available to the clinical teams with further investigation or treatment at their discretion.
A total of 41 laparotomies were recorded over this period; four cases were omitted due to inadequate data collection. Additional data were collected from the Intensive Care National Audit and Research Centre (ICNARC) Case Mix Programme database including Acute Physiology and Chronic Health Evaluation (APACHE) score, mortality and length of ICU stay.
Raised TnT on admission was associated with a worse outcome, with both median length of stay and mortality increased. Interestingly, on Bayesian analysis a TnT of three times the upper limit of normal (>42 ng/L) was associated with a likelihood ratio of death on ICU of 3.2 when comparing two patients with identical pre-test probability scores (Table 1).
Differences in ICU outcome with raised Troponin T.
P-POSSUM: Portsmouth – Physiological and Operative Severity Score for the enUmeration of Mortality and Morbidity4; TnT: Troponin T; ICU: intensive care unit.
References
35CP Implementing a novel gentamicin guideline on the ICU: A quality improvement project
D Palmer, T Clarke and J Roberts
Critical Care, East Lancashire Hospitals NHS Trust, Blackburn, UK
Gentamicin demonstrates ‘concentration-dependent killing’. A peak concentration to minimum inhibitory concentration (MIC) ratio of 10:1 may optimise bactericidal activity, avoid regrowth, and prevent the emergence of resistant strains.1 Gentamicin is almost completely cleared by the kidneys and dose must be adjusted for renal function to avoid sustained high levels that are associated with toxicity.
We conducted a quality improvement project to drive the adoption of a novel gentamicin protocol on a 20-bed intensive care unit (ICU) in the UK. The new protocol specified a re-dose of gentamicin was to be given when the serum level was less than 1 mg/L, and the level was to be measured twice daily. Daily 6-hourly creatinine clearance (CrCl) measurements would inform the dose: 3 mg/kg (CrCl < 20 ml/min), 5 mg/kg (CrCl 20–40 ml/min or on renal replacement therapy) or 7 mg/kg (CrCl > 40 ml/min) of ideal body weight. Measured creatinine clearance avoids the problem of using predictive nomograms in patients with unstable renal function.
Data were collected from all ICU patients receiving gentamicin over a 12-week period from May to July 2015. An interim analysis showed measurement of twice-daily levels and creatinine clearance was reliable (51/54 opportunities and 19/20 opportunities, respectively), so we elected to focus on a single process measurement – the time delay between a level reported less than 1 mg/L by the lab and the next dose administered to the patient. A run chart displaying this delay for sequential doses administered was displayed prominently on the ICU and provided real-time feedback.
Three interventions were tested using plan-do-study-act (PDSA) loops. Morning doses became the responsibility of the night doctor, detailed instructions were appended to the text accompanying the gentamicin level on the computer and a nominated ‘gentamicin nurse’ for each shift would collate all the drug charts of patients receiving gentamicin to avoid a patient being missed.
Forty-one doses of gentamicin were administered over 12 weeks. There was a downward trend in dose delay from around 6 h to 2.5 h. There was high variability at the beginning of the project – indeed, three unique doses were delayed by more than 9 h. The variability resolved after about 30 doses had been administered.
Reference
36CP Palliative and intensive care – A clinical paradox?
A Cran1, C Davis1, C Guyer1, E Murphy1, K Haynes1 and R Chambers2
1Hospital Palliative Care Team, Southampton General Hospital, Southampton, UK
2Intensive Care Unit, Southampton General Hospital, Southampton, UK
There is increasing recognition of the importance of palliative care within the intensive care unit (ICU). However, current evidence is limited, with a wide heterogeneity in both the palliative care interventions and outcomes reported.1 Palliative care interventions in the ICU have been shown to decrease both hospital and ICU length of stay with no effect on mortality.1,2 Positive outcomes seen include improved communication, reduction in family distress, decrease in the number of medical procedures performed and reduction in the time taken to initiate supportive care.1 Facilitating discharge home from ICU for the ongoing care of the dying individual is widely supported by clinicians but rarely occurs.3
Run-chart showing time delay in administration of gentamicin over sequential doses. Text annotations denote PDSA loops.
The creation of allocated time for palliative care within two intensive care consultants job plans, together with proactive visits and situational teaching by the hospital palliative care team, has resulted in a change in clinical practice in the ICU of a large teaching hospital. Following these interventions, we documented palliative care referrals from ICU and patient outcomes.
A retrospective review of healthcare records of ICU patients referred to the hospital palliative care team between April 2014 and March 2015 was conducted. A total of 46 level three ICU beds are available across general, neuro and cardiac ICUs.
In total 69 referrals, involving 67 patients, were received. In comparison, only 9 referrals were made the previous year (a 600% increase). The median age of patients was 65 years (range 18–88 years). Forty patients (58%) had a non-malignant diagnosis and 29 (41%) had a primary cancer diagnosis.
The median number of days in ICU prior to a palliative care referral was 6 days (range 0–63 days). Ten patients (14%) died before a palliative care review was possible. The majority of remaining referrals, 51 out of 59 (91%), were seen within 24 h. The mean face-to-face review time for each interaction was 42 min, which is longer than for patients not on ICU (35 min).
Nineteen percent (11 out of 59) were discharged directly from the ICU to either home (5%) or a hospice (14%). A further 12% subsequently went home (5%) or to a hospice (7%) after being discharged to a level one or two ward. Three (5%) were transferred to other hospitals nearer home. Almost half of patients (47%) died with ongoing input from the hospital palliative care team. A further 10 (17%) of referrals were discharged from follow-up.
With relatively little investment, we have managed to significantly increase the referrals to and the involvement of the hospital palliative care team in the ICU. Almost a fifth of the ICU patients seen were discharged directly from ICU to either home or a hospice for continuing care whilst dying. Referrals have continued to increase and we recognise the need for further investment in collaborative services to focus on this work.
References
37CP Optimising antibiotic stewardship in suspected ICU-acquired pneumonia in two large intensive care units: A prospective audit
G Wojcik, TH Craven and TS Walsh
Department of Anaesthetics, Critical Care, and Pain Medicine, Edinburgh University, Edinburgh, UK
Antibiotic (AB) misuse is common and the emergence of multi-drug resistant organisms is increasing.1 Improving AB stewardship is vital and is a priority for government, health policy makers and clinicians.2 Although many quality improvement initiatives have focused on reducing reported ICU-acquired infections, their impact on antibiotic use, and specifically de-escalation/discontinuation practices, is uncertain.3 We aimed to prospectively capture detailed AB use data in two large intensive care units (ICUs) to identify potential areas for quality improvement.
We collected prescription rationale, AB prescription/de-escalation, and microbiological data daily between 1 May and 31 July 2015. All patients with more than two calendar days length of stay (c.d.) were included. For the purpose of this analysis, we defined de-escalation as stopping ABs or narrowing AB spectrum within five days of prescription date. The project was considered service evaluation; ethical approval was not required.
Of 438 patients admitted to the ICUs, 210 (48%) spent more than two c.d. in ICU (mean length of stay 7.6 days). Sixty-three percent (133/210) were mechanically ventilated (mean ventilation days 4.5). Twenty-five percent (53/210) of these patients received a new course of ABs (85% of ventilated patients, 45/53). In total, 105 separate AB courses were prescribed. The most common indication was suspected/proven ICU-acquired pneumonia (ICU-AP) (57%, 60/105); these episodes occurred in 37 patients (70%, 37/53) of all patients receiving ABs). For the 60 ICU-AP episodes, 80% (48/60) were taken within two c.d. of initiating an AB). The use of diagnostic methods in the ICU-AP cases was BAL 23, endotracheal aspirate 18 and sputum sample 13. Figure 1 illustrates the outcomes of investigations and how AB decision making changed accordingly.
Rationale for antimicrobial prescriptions. AB: antibiotic; CI: chest infection; c.d.: calendar days; N: number of patients; n: number of antibiotic courses. AB course considered complete if not stopped or rationalised within five days.
Our data illustrate the complexity of understanding AB decision making, even in two ICUs within the same organisation. We have shown that suspected ICU-AP remains prevalent, despite surveillance reporting documenting low confirmed rates.4 In our cohort only 11/59 (19%) of cases had ABs de-escalated or stopped within five days. Patients with negative samples (11) or positive cultures that may have allowed rationalisation/narrowing of cover (29) comprised in total 40 (68%) cases. These cases utilised significant ABs where potential to reduce use existed through more active de-escalation.
References
38CP Using high-frequency oscillation ventilation on a complex patient with refractory hypoxaemia on conventional ventilation: A case study
E Mc Sweeney
Cardio-thoracic Intensive Care Unit, St Georges Hospital, Tooting, London, UK
Recent studies have not supported the use of high-frequency oscillation ventilation (HFOV) on intensive care units, suggesting that HFOV may actually increase mortality and complications in the modern era of lung protective ventilation strategies.1,2 This case study is of a specific complex patient who had refractory hypoxemia requiring peak pressures greater than 30 cmH2O on a conventional ventilator. Therefore, HFOV and nitric oxide (NO) were used.
The patient in question was a 52-year-old Australian woman who had a previous diagnosis of an inter-costal artery bleed 10 years prior. She had severe scoliosis and secondary restrictive lung disease.
She presented with a second inter-costal artery bleed which was cauterised. She was transferred to cardio-thoracic intensive care unit (CTICU). She was in type one and two respiratory failure with severe hypovolemic shock and subsequent Acute respiratory distress syndrome (ARDS). She had a PaO2/FiO2 ratio of 50 which is associated with 45% mortality rate according to the Berlin definition,3 not taking into account her past history and anatomy.
On day 1, she required 100% oxygen delivery with peak pressures greater than 30 cmH2O. She was hypotensive with mean arterial pressures less than 60 mmHg on moderate doses of adrenaline and noradrenaline. She had a reducing lactate and had been fluid resuscitated with three litres of crystalloids. On chest X-ray, she had a collapsed right lung that was almost completely dysfunctional.
Her ventilation requirements were such that the risks of ventilator-acquired lung injury (VALI) were thought to be significant and her hemodynamic function, despite high doses of vasopressors, was suboptimal. Chest drain insertion did not assist expansion of this right lung, and the patient was refused the chance of extracorporeal membrane oxygenation due to high risk of bleeding. Although HFOV is controversial in literature at the moment as previously mentioned, this particular unit has significant experience with HFOV and it was regarded as a safe and potentially beneficial option. For this reason, HFOV via the Care Fusion 3100B, combined with inhaled (NO) were used. The goal was to reduce risks of VALI and to improve oxygenation and hemodynamic function.
Within the first 24 h on HFOV and NO, there was no improvement in oxygenation. HFOV did however; allow ventilation at much lower peak pressure than conventional ventilation (20 cmH2O). Day 2 and 3 on HFOV and NO saw a reduction of FiO2 to .5 with PaO2 > 10 kPa. She remained on HFOV and NO for five days. After this, patient X was changed to assisted pressure release ventilation (APRV), requiring peak pressures of 25–28 cmH2O. She remained on APRV for thirty-three days and after almost three months of slow weaning was repatriated to Australia. After four months in hospital, she was discharged home.
Although the Care Fusion 3100B did not improve oxygenation in the early stages, it maintained oxygenation while reducing the risks of VALI and contributed to improving overall hemodynamic stability. Above all, this patient was ultimately discharged home ventilator free, independent with all activities of daily living.
References
39CP Noise levels and correlation with activity and time in a district general intensive care unit
Sean J Keating1, Sarah Beavis2, Jeremy Windsor2 and Anand Padmakumar2
1Critical Care Department, Royal Infirmary of Edinburgh, Edinburgh, UK
2Critical Care Department, Chesterfield Royal Hospital, Chesterfield, UK
Environmental conditions can have an adverse effect on patients in intensive care unit (ICU).1,2 Published guidelines from the World Health Organization (WHO) specify acceptable levels of background noise in hospitals (hourly average >35 dB(A) with peak values <40 dB(A)).3 However, no specific guidelines exist for those on ICU. This study aimed to quantify the noise pressure levels in a seven bedded district general ICU. In addition, we sought to investigate the reasons behind high levels of noise and propose a series of practical solutions to this problem.
Seven days were chosen at random, one for each day of the week, over a two-month period. Noise meters (calibrated Tecpel DSL-330 devices) were attached to three separate locations around the unit, which remained consistent over the course of the study. Over 24 hours, the following were recorded: mean noise pressure together with hourly minimum and maximum pressures. In addition, nursing staff recorded activities within ICU and the time they were performed. Data were analysed using Microsoft Excel and SPSS.
Over seven days, the mean background noise was found to be 55.4 dB(A), the mean hourly maximum 70.2 dB(A) and the mean hourly minimum 50.1 dB(A). For a 24-h period a side room was unoccupied with equipment in standby mode and the following results were recorded – mean 50.4 dB(A), mean hourly minimum 46.7 dB(A) and mean hourly maximum 62.4 dB(A).
We found that there was a wide range of noise pressure levels. Mean pressure was found to be higher during daytime hours, (0800–2000) than night time (2000–0800) (56.39 vs. 54.52 dB(A) p = 0.0011). Mean noise pressure was also noted to be higher during weekdays than weekends (57.25 vs. 54.28 dB(A) p < 0.0001). The movement of patients was associated with the highest pressure (56.08 dB(A)) compared with no activity (54.46 dB(A) p = 0.0024). The highest mean pressure level (58.1 dB(A)) was found nearest the nursing station (54.9 vs. 53.3 dB(A) p = 0.007).
Noise levels in our ICU consistently exceeded WHO guidelines. Even when equipment was in standby mode and a side room unoccupied noise levels were elevated. Strategies to reduce noise production should be implemented (e.g.: staff education, use of ‘night mode’ on monitor alarms and minimising patient movement overnight). However, given the high measurements obtained from unoccupied areas, it is also necessary to reduce noise from reaching the patient (e.g. the use of earplugs or noise isolating headphones, sound absorbing ceiling tiles and other sound reducing technology). Only by adopting both approaches can the impact of noise upon our patients be reduced.
A further study is being planned following completion of an education programme and the introduction of a package of noise reducing measures.
Acknowledgements
The authors thank the nursing staff at Chesterfield Royal Hospital.
References
40CP Intermittent glucose monitoring in the critically ill: A pilot survey of practice in the United Kingdom
AJ Devanesan and S Saha
Department of Intensive Care, Queens Hospital BHRUT, Romford, UK
Hyperglycaemia, hypoglycaemia and glucose variability are all associated with a worse prognosis in critical illness. As such, glucose control with intravenous insulin is a common therapy employed in intensive care units (ICUs). There is a great deal of controversy around optimal glucose targets as well as the optimal regime for glucose monitoring in the critically ill. Consensus guidelines recommend that glucose monitoring should be conducted from arterial or venous blood and not from capillary blood in patients where invasive vascular monitoring is required and analysed in the laboratory or with a blood gas analyser.1 The surviving sepsis guidelines recommends monitoring glucose 1 to 2 hourly until glucose levels are stable and then at least 4 hourly while many other guidelines do not specify a frequency.2 We aimed to conduct a pilot survey of current practice across the UK to investigate the necessity and feasibility of a larger national audit of glucose control practices in the UK.
Chart of median glucose targets for different ICUs (n = 20) comparing those which use arterial blood gas analysis only (Blue) with those which use a combination of both capillary and blood gas glucose monitoring (Red) in critically ill patients with invasive vascular monitoring.
We conducted a telephone survey of 20 ICUs in England, Wales, Northern Ireland and Scotland including both district general and teaching hospitals. We asked either the nurse in charge or the most senior doctor available what their current glucose target was, the frequency of intermittent glucose monitoring and the sampling method (arterial, central venous or capillary blood) for glucose measurement.
We had a response from all 20 units. The target glucose varied considerably, with median targets ranging from 5.5 mmol/L to 10 mmol/L. All units measured glucose 1 to 4 hourly according to a protocol. Units using a lower glucose target seem to measure glucose more frequently, but the sample size was too small for statistical significance. However, only 50% of units exclusively use arterial blood gas analysis for glucose levels in patients with invasive vascular monitoring with 50% of units using a combination of blood gas analysis and capillary glucose monitoring (Figure 1).
It is important to maintain moderate glycaemic control in critically ill patients. Currently, ICUs in the UK are using capillary glucose monitoring in excess of current guideline recommendations but are measuring glucose at an adequate frequency. We recommend a national survey of glycaemic control practices be conducted across the United Kingdom to further clarify this issue and highlight the current guidelines.
References
41CP Restoration of speech and swallowing in dysphagic spinal cord injured patients receiving mechanical ventilation via tracheostomy – A case series
S Morgan and J McRae
London Spinal Cord Injury Centre, The Royal National Orthopaedic Hospital, London, UK
Spinal cord injury (SCI) is a complex condition affecting multiple body systems. For cervical SCI (CSCI) patients, this can include paralysis of some or all respiratory muscles necessitating a period of mechanical ventilation (MV), typically via tracheostomy placement.1
Dysphagia is CSCI is under-reported and under-recognised but incidence is reported to be 40%2 with high rates of silent aspiration due to sensory impairments. The presence of tracheostomy and MV is also associated with dysphagia3 making this a particularly vulnerable group. Dysphagia can delay the weaning process, increase hospital length of stay (LOS) and reduce quality of life (QoL). These patients typically spend a considerable amount of time on intensive care units (ICUs) awaiting admission to a SCI centre (SCIC).4
CSCI patients undergo physiological changes with time. This includes resolution of cord oedema which may improve neurological level and allow use of previously paralysed respiratory muscles, resolution of pulmonary pathology, development of spasticity; allowing return of intercostal tone to improve ventilation and retraining of functioning respiratory muscles. These changes allow weaning in the majority. The weaning technique advocated by SCIC’s is ventilator free breathing (VFB).5 Although a relatively simple concept, this needs to be followed rigorously by a dedicated team to achieve ventilator independence and can take up to several months. Patients with C1/2 complete injuries will remain ventilator dependent, but the approaches discussed in this case series can be applied to allow speech and safe swallowing.
Ability to eat and speak are important patient goals and paramount to rehabilitation and reintegration. Our clinical evidence suggest that patients either spend considerable time nil by mouth (NBM) and non-verbal due to cuff inflation or resume oral intake without appropriate input resulting in aspiration events and subsequent difficulties weaning.
This case series (see table 1) will demonstrate team approaches to weaning to allow speech and safe oral intake alongside respiratory weaning attempts. This includes laryngeal weaning via cuff deflation, which helps restore pharyngeal and laryngeal reflexes, assessment with flexible nasendoscopy (FNE), targeted swallow rehabilitation, tube and ventilator choices to include subglottic suction-aid tubes and portable domiciliary ventilators.
Demographic data with patient outcomes.
References
42CP Multidisciplinary in-situ simulation training in Critical Care: Six-month review
EC Baker, A Beane, H Hewitt, L McWhirter and ML Kong
Adult Critical Care Unit, Royal London Hospital, London, UK
Patient safety in critical care depends on teams who are able to quickly recognise emerging problems, communicate effectively and deliver appropriate timely interventions. In order to develop the newly established workforce in our 44-bed tertiary referral unit in a major trauma centre, we created a programme of multidisciplinary in-situ simulation training. In response to previous local clinical governance work and national safety recommendations,1–3 the first phase of training focussed on airway emergencies. Displaced tracheostomies for example can occur relatively frequently in the critical care setting2 but individuals are unlikely to be involved in such incidents more than once or twice a year and therefore may not be familiar with recently introduced protocols for management.
Each team took part in three medium-fidelity simulated scenarios, as participants and observers. The post-simulation debrief explored clinical knowledge and relevant guidelines, practical skills, human factors and team working. Over 100 clinicians took part in the training on 12 separate days spread over four months. Six months later, we sought feedback from participants to help shape further training.
Eighty-seven percent of participants reported that they felt more confident that they knew how to manage an emergency airway situation should one occur. Only one third of participants (36%) reported that they had been in situations where they had been able to use their new skills or knowledge since their training, highlighting the infrequency of such situations. Only 36% of respondents felt that their team work had improved as a result of the training. This contrasts with feedback immediately after the session when 89% of respondents gained a better understanding of closed loop communication, role allocation, leadership and team work.
Three-quarters of respondents agreed that they would like to undergo more simulation training on the same topics (74%) while 84% would like further simulation training on other topics, including a range of topics from managing a shocked patient to dealing with families in distress. One participant commented that they had felt uncomfortable by their colleagues observing them during simulation.
This programme has shown that in-situ medium fidelity simulation is feasible in the critical care setting. It has helped develop knowledge of and confidence in key emergency airway management protocols, and familiarity with emergency airway equipment on our unit. Future training is planned for a range of critical care emergencies, and will include more explicit exploration of key Crisis Resource Management concepts during the brief and debrief, as well as a programme of interim team updates following training.
References
43CP The development of a specialist critical care outreach physiotherapist to facilitate the implementation of NICE CG83 in a regional tertiary major trauma centre and enhance rehabilitation post critical illness
S Bunting, D Brooks and R Moses
Royal Preston Hospital, Lancashire Teaching Hospitals Foundation Trust, Preston, UK
The aim of this review is to identify the role of a critical care outreach physiotherapist (CCOPT) to implement ward-based rehabilitation post ICU discharge and facilitate outpatient follow-up.
Totally, 146 patients were assessed and treated as part of an early rehabilitation pathway in ICU with 269 treatment interventions. This included early mobilisation, joint sessions with the ICU team and provision of rehabilitation prescriptions and plans. These sessions would otherwise not have been delivered.
Over the six-month period, 62 patients were seen post ICU discharge. On average, these patients received at least two follow-up sessions to implement pre-established rehabilitation programmes and facilitate a safe ICU discharge and prevent readmission. Only three patients were re-admitted to ICU and two were re-admitted as level 2 patients.
All of these patients met current NG83 rehabilitation guidance and had ongoing rehabilitation needs identified. Only a small number of patients have been followed up in post discharge clinics due to the infancy of the programme.
References
44CP Survey of patients with sickle cell disease admitted to critical care department at North Middlesex University Hospital, London
A Zuhair, O Ralley, L Valco, K Malpica and P Gandre
Department of Critical Care, North Middlesex University Hospital, London, UK
North Middlesex University Hospital (MNUH) is situated in an area with a high Afro Caribbean and Mediterranean demographic. For this reason, admission as a result of SCD is disproportionately high compared to the national average.
The aim of our survey was to assess the outcome of patients with complications of sickle cell disease (SCD) who are admitted to critical care.
Reference
Wednesday 9 December 2015 2CP The use of enoximone in management of anaphylaxis complicated by labetolol use
S Hayhoe1, V Navapurkar1 and A Conway Morris1,2
1John V Farman Intensive Care Unit, Addenbrooke’s Hospital, Cambridge, UK
2Division of Anaesthesia, University of Cambridge, Cambridge, UK
Following an intravenous infusion of piperacillin/tazobactam, the patient suffered a pulseless electrical activity cardiac arrest, with return of spontaneous circulation after 12 min of cardio-pulmonary resuscitation, and IV adrenaline (3 × 1 mg). Post-arrest systolic hypotension (64 mmHg) and bradycardia of 45 bpm was managed with boluses of 0.1 mg adrenaline, atropine and intravenous fluids, alongside 100 mg hydrocortisone and 10 mg chlorpheniramine for treatment of possible anaphylaxis. Following admission to intensive care unit, the patient remained hypotensive and bradycardic despite escalating doses of catecholamines.
Cardiac output monitoring with LiDCO demonstrated of low cardiac output (cardiac index < 1 L/min/m2), and no improvement following fluid bolus. Electrocardiogram showed sinus bradycardia with left ventricular hypertrophy. Initial arterial blood gas revealed a lactic acidosis (pH 7.27, base excess −4.8, lactate 6.3 mmoles/L, with type 2 respiratory failure PO2 13.7 kPa PCO2 6.6 kPa on FiO2 1.0). A bedside echocardiogram showed a poorly contractile, thickened left ventricle and normal-sized right ventricle.
Persistent low cardiac output (cardiac index <1 L/min/m2), hypotension and bradycardia in the presence of labetolol prompted the initiation of enoximone. Over the subsequent 2 h, there was a gradual improvement in cardiac index to an index of 2.3 L/min/m2, with blood pressure rising to 80/50 (M58), resolution of lactic acidosis and improvement in pulmonary gas exchange. There were continued improvements in haemodynamic parameters and by 12 h post-arrest all inotropes had been discontinued, patient was successfully extubated at 18 h with minimal neurological disability and subsequently discharged home.
Mast cell tryptase levels taken immediately after return of spontaneous circulation were >200 ng/ml, falling to 44 ng/ml and then 4 ng/ml at 6 and 24 h, respectively.
In the case above, it may be that the improvement seen was not solely attributable to the effect of enoximone. The half-life of oral labetalol is 6–8 h, so the full resolution in our patient’s clinical state, seen around 12–18 h after the last labetalol dose, could have been associated with a gradually increasing availability of binding sites for the adrenergic agonists as much as to the enoximone.
References
3CP A pilot project to assess the feasibility and effectiveness of a mechanically ventilated manikin to simulate intensive care emergencies for junior trainees
LE Christie1, P Patel2 and MA Hayes1,3
1Imperial School of Anaesthesia, London, UK
2Department of Anaesthesia and Critical Care, Hammersmith Hospital, London, UK
3Department of Anaesthesia and Critical Care, Chelsea and Westminster Hospital, London, UK
The aim of this project was to assess the feasibility and effectiveness of a multidisciplinary simulation session incorporating a mechanically ventilated manikin to recreate ICU emergencies for junior intensive care trainees.
Additional features to recreate the ICU environment included the use of arterial and central venous lines, a urinary catheter and intravenous sedation via syringe drivers. An ICU observation chart, arterial blood gases, blood results, chest x-ray and electrocardiogram were available for each scenario. Faculty included a multidisciplinary team of ICU staff (consultants, registrars, nurses and a physiotherapist).
Nine candidates attended the one-day session and included Foundation, Acute Common Care Stem and Core Medical trainees.
All candidates had ≤six months ICU experience (mean 2.3) and two candidates participated in each scenario. There were no technical problems encountered during the day in relation to ventilating the manikin and the measured values and alarms generated.
Regarding effectiveness, all nine junior doctors found the experience of simulation useful and all would recommend the day to others at their stage of training. Their overall confidence in managing ICU emergencies (scale 0–10) pre and post the course increased from a median of 3 (range 0–7) to 7 (range 5–8). Comments were made regarding the realism of the simulated ICU environment and how they had increased their competence regarding airway management and tracheostomies.
Acknowledgements
We are grateful to all faculty who helped facilitate this simulation session.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: We are grateful to the Lead Provider, Imperial College Healthcare NHS Trust, for funding.
Reference
4CP Rapid improvements in quality and safety following introductions of tracheostomy multi-disciplinary team ward rounds at a tertiary hospital
J Lynch, S Wallace, B Bonvento, G Sharp, S Pieri-Davies and BA McGrath
Acute Intensive Care Unit, University Hospital South Manchester, Wythenshawe, Manchester, UK
Tracheostomy multi-disciplinary team (TMDT) care involves interactions between multiple teams in complex patients. Different ward teams may subsequently manage intensive care unit (ICU) tracheostomies.1 TMDT ward rounds have shown improvements in patient outcomes such as time to decannulation, hospital length of stay and adverse events, but most published series to date took several years to demonstrate an impact.2–4 The Global Tracheostomy Collaborative (GTC) Quality Improvement (QI) programme supports rapid adoption of QI initiatives from exemplar centres, tracking patient metrics, surrogate markers of quality and adverse events. Our aim was to rapidly implement best practice TMDT ward rounds and evaluate their effectiveness.
The University Hospital of South Manchester manages around 200 tracheostomy patients annually and joined the GTC in July 2014. We collected baseline data for 20 weeks before implementing weekly TMDT ward rounds reviewing ICU and non-ENT ward-based patients. TMDT typically comprised consultant intensivist, speech and language therapist (SLT), physiotherapist, ENT specialist nurse, critical care outreach sister and a long-term ventilation specialist. We hypothesised that TMDT interventions and recommendations would directly impact upon rates of accidental decannulation, tube obstruction and patient centred outcomes. Data were entered into the GTC database and analysed using SPSS 22 (IBM systems).
Impact of the TMDT ward round on critical incidents. TBD – tracheostomy bed days.
TMDT ward rounds in our hospital rapidly impacted on frequency, nature and severity of tracheostomy-related adverse events by adopting interventions and resources from international exemplar institutions. Lengths of stay were also reduced, likely reflecting an increase in the quality of care, less fragmented care, and the ability of the TMDT to arrange interventions in a timely manner appropriate to the patient’s progress.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project is funded by the Health Foundation’s Shine programme. The Health Foundation is an independent charity working to improve the quality of healthcare in the UK.
References
5CP An audit and quality improvement project: Evaluating the provision of intensive care support for the critically unwell patient in the cardiac catheterisation suite undergoing primary percutaneous coronary intervention in a tertiary 24/7 centre
C Christou, C Malcolm, C Lopez-Soto and M Naik
Critical Care Department, Norfolk and Norwich University Hospital (NNUH), Norwich, UK
The use of primary percutaneous coronary intervention (PPCI) to treat ST elevation myocardial infarction (MI) is increasing.1 In 2014–2015, our centre performed 429 PPCIs, 59 of these patients had an out-of-hospital cardiac arrest or a perioperative arrest and 26 were recorded as needing intubation and ventilation during the procedure. This figure has increased from 21 in 2013–2014 and 10 in 2012–2013. With increasing rates of PPCI taking place, intensive care (IC) and anaesthetic involvement has also increased.
The fourth national audit project (NAP4) has highlighted that severe complications do occur during airway management. The incidence of failed intubations in the emergency setting varies from one in 300 to one in 800; this is considerably higher than in elective tracheal intubations, where a failure rate of one in 1000–2000 is reported.2
The provision of IC support to the cardiac catheterisation suite (CCS) includes airway management and cardiovascular support in acutely unwell patients during PPCIs. This has proven to be challenging with a number of issues surrounding equipment and expertise when remote site anaesthesia is needed. In our centre, both Cardiology and Intensive Care doctors were dissatisfied with the current situation.
To further investigate these issues and with the aim of improving patient care, a questionnaire was created and distributed to IC and cardiology staff involved with PPCI. The Royal College of Anaesthetists (RCOA) has guidance for the minimum standards required for remote site anaesthesia3; an audit was carried out to evaluate our compliance with these standards.
Fifteen cardiology staff and 16 IC staff responded to the questionnaire. Issues identified from the IC team were the lack of trained airway personnel available to assist in emergency intubations and lack of a separate airway trolley with equipment readily available if required, especially for more difficult intubations. In extreme cases, some anaesthetists reported to have performed a rapid sequence induction (RSI) with an untrained airway assistant. Issues identified by the cardiology team included difficulty in getting adequate airway support in an emergency.
Our audit also revealed that guidance for remote site anaesthesia was not being adhered to. NAP4 stated that RSI outside the operating theatre requires the same level of equipment and support as needed during a standard anaesthetic.2 Our audit and questionnaire demonstrated this is not the case for patients requiring RSI; end-tidal carbon dioxide monitoring (ETCO2) and a dedicated anaesthetic assistant were not always available; neither were anaesthetic record charts or access to peripheral nerve stimulators.
In response to our findings, we have instituted an emergency airway assistance call for use in situations where patients undergoing PPCI require airway support. This call means that a trained airway assistant will support the attending IC doctor. Equipment availability has been improved, a portable ventilator and monitor, including ETCO2, is now readily available to be used for these patients as well as a specialist airway trolley that has been set up in this area which includes advanced airway equipment.
To ensure these interventions are effective, we will be re-auditing their impact in the near future.
References
6CP VV-ECCO2R in a patient with a traumatic brain injury and hypercarbic respiratory failure
D Kelly and B Paul
Royal Victoria Infirmary, Newcastle-upon-Tyne Hospitals NHS Foundation Trust, Newcastle-upon-Tyne, UK
On primary survey, his airway was found to be compromised with no audible air entry and bilateral chest drains were sited. He had a heart rate of 134 and BP of 99/36. His Glasgow Coma Scale was 3/15 with reactive, equal, 2 mm pupils, and there were obvious deformities of both legs. A rapid sequence induction was performed with ketamine and he was found to be very difficult to ventilate. An arterial blood gas showed pH 7.062, pCO2 12.04, pO2 72.63 HCO3 25.1 and BE −7.1. Clinical findings were consistent with bronchospasm, achieving tidal volumes of only 300 ml with a peak pressure of 45 and PEEP of 5. This improved slightly with further ketamine, magnesium and nebulisers.
Pan-computed tomography (CT) findings included transection of the descending aorta at two levels with mediastinal haematoma and no sign of active bleeding, small pneumothoraces, right first rib and bilateral femoral fractures. CT head and neck were reported as consistent with diffuse axonal injury with no cervical spine injury identified.
The patient was transferred for cardiothoracic interventional radiology management of the aortic transection. The thoracic aorta was stented and the left subclavian artery was embolised.
The bronchospasm and hypercarbia persisted with minimal improvement on ketamine, salbutamol and aminophylline infusions. An intracranial pressure (ICP) bolt was sited 22 h after the injury with an initial pressure of 25 mmHg. Medical management of raised ICP proved difficult with resistant bronchospasm and hypercarbia, thus preventing surgical management of orthopaedic injuries and sedation hold. The decision was made to use VV-ECCO2R, accepting the risk of anticoagulation. Unfractionated heparin was used with a target APTR of 1.3–1.8.
With control of pCO2 within normal limits the ICP fell and remained consistently below 10 mmHg. This facilitated a reduction in ventilator pressures and a reduction in sedation. After five days the CO2 sweep had been weaned off, and the catheter was removed without any significant complication.
The patient has made a full neurological recovery and continues to rehabilitate after definitive management of his orthopaedic injuries.