2017 PDU Programme : simplebooklet.com

3 Welcome to Perfusion Downunder 2017 On behalf of the Perfusion Downunder Organising Committee, a very warm welcome to The Barossa and to the Novotel for this 13thh PDU winter meeting. We continue to be delighted at the support PDU receives from both the perfusion community and also our medical colleagues convening our 12th consecutive meeting. The meeting has developed an enviable reputation for excellence through its academic content and relevance to practice, an integral part of which is the Australia and New Zealand Collaborative Perfusion Registry, formerly the PDU Collaboration. This year PDU has established The Wayne Pearson Keynote Speaker to honour the unequivocal support and mentoring of the Perfusion Downunder Meeting that Wayne provided from the outset in 2004. PDU would not have developed and enjoyed the success it has achieved has Wayne’s enthusiastic and charismatic guidance not be a part of it. The inaugural WPKS is Mr Tim Jones. We look forward to an informative, friendly and relaxed conference focussing on current issues in cardiopulmonary bypass, building on previous topics and encouraging collaborative interest in perfusion research initiatives over the next few days. We are delighted to welcome Tim Jones and Donald Likosky (both well known to many of you) as well as Florian Falter from the UK as our international speakers. Our returning regional Alumni faculty include, David Sidebotham, Sara-Jane Allen, Paul Myles, Simon Mitchell Richard Newland, Rob Baker and of course Alan Merry who has been involved in every meeting to date. We are delighted to have Dr Sally Roberts - microbiologist from Auckland to join us this year. On the social side we have again arranged some great food and wine for Thursday, Friday and Saturday. If you have any special dietary requirements please let Berni Tackney know.

Page 4

4 Thursday night experience is at the Jacobs Creek Visitor Centre. Friday is a casual evening in the Barossa after the activity and Saturday night is again at historic Chateau Tanunda. The activity on Friday afternoon will be a creative outdoor experience. Dress warm. Have a great time. A reminder to companion registrants, dinner Thursday and lunch & dinner Friday including the activity and the Prof Merry Lecture, lunch and dinner Saturday, are all included in your registration. For the Organising Committee, Perfusion Downunder Tim Willcox, CCP, Green Lane Perfusion Auckland City Hospital, New Zealand. Rob Baker, PhD, Flinders Medical Centre, Adelaide, Australia. Michael McDonald, CCP, Perfusion Services Ltd, Melbourne, Australia Bernardette Tackney, Conference Co-ordinator, Perfusion Downunder, New Zealand

Page 5

5 Table of Contents The Wayne Pearson Keynote Lecture ‘Early brain development and neuroprotection’ Tim Jones p 19 Hyperbaric oxygen after perfusion accidents: Who, what, when? Simon Mitchell p 21 Sleep and cardio metabolic disorders – beware the late night cocktail. Sara-Jane Allen p 23 Patients’ experience of comfort in healthcare. Alan Merry p 24 International CPB survey. Florian Falter p 27 An old survey; does it tell us anything, the value of a new survey. Tim Willcox p 28 Is conventional bypass for coronary artery bypass graft surgery a misnomer? Rob Baker Donny Likosky p 29 Reporting CPB related morbidity – are we on the same page? Donald Likosky p 30 Are consensus papers worth publishing? David Sidebotham p 31 The Protamine conundrum. Florian Falter p 32 One year results from the ATACAS Trial. Paul Myles p 33 Prof Merry Lecture ‘Am I really trained to do what I do?’ Tim Jones p 37 Human factors in the operating room – instinct or intelligence. Tim Jones p 41

Page 6

Page 7

Page 8

Page 9

9 Keynote Speakers International Tim Jones Paediatric Surgeon Birmingham Children's Hospital Birmingham, UK Donald Likosky Associate Professor Department of Cardiac Surgery University of Michigan Medical School Michigan, USA Florian Falter Consultant in Anaesthesia and Intensive Care Papworth Hospital Cambridge, UK Invited Regional Faculty Paul Myles Professor and Director Dept of Anaesthesia and Perioperative Medicine The Alfred Hospital Melbourne, Australia Simon Mitchell Professor, Head of Department of Anaesthesiology School of Medicine University of Auckland Auckland, New Zealand Sally Roberts Doctor Microbiologist Auckland City Hospital Auckland, New Zealand Alan Merry Professor of Anaesthesia Green Lane Department of Anaesthesia Auckland City Hospital Auckland, New Zealand Sara-Jane Allen Intensivist/Anaesthetist Green Lane Anaesthesia & CVICU Auckland City Hospital Auckland, New Zealand Richard Newland Senior Perfusionist Flinders Medical Centre Adelaide, Australia David Sidebotham Anaesthetist / Intensivist Green Lane Anaesthesia & CVICU Auckland City Hospital Auckland, New Zealand

Page 10

Page 11

11 PROGRAMME Thursday 17th August 2017 08:30 – 12:30 ANZCPR Quality Meeting (all welcome, please let us know if attending) working lunch provided 13:00 REGISTRATION DESK OPENS Shiraz A & B in Conference Centre Welcome – Perfusion Downunder 2017 14:30 – 15:30 Session 1: The Wayne Pearson Keynote Lecture ‘Early brain development and neuroprotection’ Tim Jones – England 15:30 – 16:00 Afternoon Tea 16:00 - 18:00 Session 2: Moderator – Uncommonly considered Hyperbaric oxygen after perfusion accidents: who, what, when Simon Mitchell – New Zealand Sleep and cardio metabolic disorders – beware the late night cocktail Sara-Jane Allen – New Zealand Patients’ experience of comfort in healthcare Alan Merry – New Zealand Discussion 15 min 19:00 Welcome Dinner – Jacobs Creek

Page 12

Page 13

13 Friday 18th August 2017 07:30 - 08:30 Breakfast – served in Cabernet A & B 08:30 – 10:00 Session 3: Moderator: …the wisdom to know the difference: International CPB survey Florian Falter - England An old survey ; does it tell us anything, the value of a new survey Tim Willcox – New Zealand Variation ANZCPR PERForm Rob Baker – Australia, Donald Likosky – United States of America Panel discussion – Why practice varies 10:00 – 11:30 Session 4: Moderator: Getting a consensus Reporting CPB related morbidity – are we on the same page? Donald Likosky – United States of America Are consensus papers worth publishing? David Sidebotham – New Zealand 11:30 – 11:45 Break 11:45 – 13:15 Session 5: Moderator – Turning off the bleeding The Protamine conundrum Florian Falter – England 1-year follow up data from ATACAS Paul Myles - Australia

Page 14

Page 15

15 Saturday 19th August 2017 07:30 - 08:30 Breakfast – served in Cabernet A & B 08:30 – 10:00 Session 8: Moderator: Is the Cardiac OR a safe haven? ‘Human factors in the operating room – instinct or intelligence’ Tim Jones - England “Checking the checkers: patient safety in the operating room” Simon Mitchell – New Zealand Panel discussion – 10:00 – 10:30 Morning Tea 10:30 – 12:30 Session 9: Moderator Trial Architecture Outcomes : Days alive and out of hospital Alan Merry – New Zealand The diagnostic properties of current AKI consensus criteria Donald Likosky – United States of America A novel trial design Paul Myles – Australia Panel discussion -

Page 16

Page 17

Page 18

Page 19

19 SESSION 1 Thursday 14:30 - 15:30 The Wayne Pearson Keynote Lecture ‘Early brain development and neuroprotection’ – Tim Jones There have been significant advances in the management of children undergoing cardiac surgery for congenital heart disease (CHD). Consequently 80% of patients born with CHD now survive to adulthood. With mortality rates improving, the focus is changing to reduce morbidity associated with paediatric surgery and cardiopulmonary bypass (CPB). The true incidence of brain injury during paediatric cardiac surgery remains difficult to quantify. There are no widely accepted clinical outcomes measures applicable across ages. MRI assessment is increasingly used but pre operatively up to 62% of neonates with complex congenital heart diseases have evidence of intracranial haemorrhage on MRI which is worse in 48% of patients postoperatively. The clinical impact of these findings remains unknown. In utero, the foetal circulation ensures the brain receives the highest oxygenated blood compared to the rest of the body which is important for brain maturation. Recent data has suggested that in conditions such as transposition of the great arteries and hypoplastic left heart syndrome where this effect is reversed and consequently babies born at term with such conditions have a brain maturity comparable to a 32 week foetus and are very susceptible to hypoxic injury. Regardless of the presence of congenital heart disease neonates and infants have immature organ systems and altered physiology. It is unknown if the neonatal brain and particularly the premature neonate have the ability to autoregulate due to a lack of penetrating muscular arterioles which are not fully developed at term. This may have implications on cerebral perfusion and protection during CPB. Preoperatively neonates with congenital heart disease have significantly decreased cerebral blood flow and a blunted auto regulatory response. Fortunately the neonatal and infant brain has greater metabolic suppression with cooling compared to adults suggesting they may tolerate periods of no flow or low flow longer during hypothermia. It remains unknown if undergoing cardiac surgery in childhood influence brain development or is it only a problem if a specific brain injury occurs during surgery. There is little scientific evidence to support many of the current neuroprotective strategies and collaborative prospective clinical trials need to be undertaken. Full paper is available on the meeting APP

Page 20

Page 21

21 SESSION 2 Thursday 16:00 - 18:00 Uncommonly considered Hyperbaric oxygen after perfusion accidents: - Simon Mitchell Perfusion accidents such as inadvertent circuit disconnections or extremely low reservoir levels can result in arterial gas embolism, and this may involve large volumes of gas. The organ most vulnerable to harm in such events is the brain. Large bubbles entering the cerebral circulation may obstruct flow causing ischaemia, and even when they redistribute, their passage through the brain vasculature incites an inflammatory response that may subsequently compromise flow and function. [1] At a clinical level patients may suffer uni- or multifocal stroke, or a global ischaemic brain injury. Cardiac surgical teams invoke a suite of strategies if there is an apparent macro-air embolism during surgery. [2] These include cessation of cardiopulmonary bypass (CPB), isolation and repriming of the CPB circuit, a period of cardiac compressions (in closed chamber procedures) plus ventilation of the lungs with 100% oxygen, and expeditious resumption of CPB possibly with aortic root venting and with the patient in a head down position. Retrograde cerebral perfusion and hypothermia are sometimes used in the immediate aftermath of an air embolism event. Post-operative treatment with hyperbaric oxygen is often mentioned in care pathways, but its use is controversial. Hyperbaric oxygen (HBO) is the administration of oxygen (FiO2 = 1) at elevated ambient pressure. There are several putative advantages of HBO in the context of arterial gas embolism. [3,4] First, pressure reduces the size of bubbles, and will encourage redistribution of large occluding arterial bubbles into the venous system. Second, compression of an air bubble raises the partial pressure of nitrogen in the bubble and since the patient is not breathing any nitrogen, this establishes a large pressure gradient for diffusion of nitrogen out of the bubble, into blood and finally to the alveoli. The net effect is to further decrease the size of the bubble and ultimately cause it to involute completely. Third, the constant delivery of blood containing dissolved oxygen at very high pressures enhances the diffusion distance of oxygen through tissue before the PO2 falls to levels below those that are physiologically useful. This may help oxygenate tissue that is hypo-perfused due to the presence of bubbles throughout much of the microcirculation. Finally, hyperbaric oxygen has complex cell signalling effects that are (almost paradoxically) anti-inflammatory. In particular, it has been shown to suppress neutrophil activation in response to a variety of stimuli, including endothelial injury. [5,6] Expeditious HBO has long been considered a standard of care in the treatment of arterial gas embolism arising from introduction of bubbles to the pulmonary veins following pulmonary barotrauma in divers. It has not, however, ever been subject to any form of experimental trial.

Page 22

22 A contemporary review synthesised case / case series evidence and concluded that outcomes are generally better when HBO is used for diving related arterial gas embolism. [7] This experience with divers has formed the basis for recommending HBO for treatment of arterial gas embolism in perfusion accidents. There are numerous case reports / series in which HBO has been used to treat either known macro-air events during perfusion, or stroke syndrome of uncertain cause following cardiac surgery with apparent success. [8-17] It is difficult to interpret what these reports demonstrate about the efficacy of HBO in this setting because of the obvious potential for positive reporting bias, and the fact that there are at least some cases that have successfully been treated without HBO. [18] Nevertheless, HBO would not be the first medical intervention promoted on the basis of favourable first principles, animal evidence and human anecdote. A key challenge with HBO which renders the intervention more controversial than it might otherwise be is safe access. Hyperbaric oxygenation can only be achieved by placing the patient in a pressure vessel and this is associated with many challenges. At the very least it requires intra-hospital transfer of a potentially unstable patient who is intubated, ventilated and almost certainly on multiple infusions. A pressurised chamber is a unique form of remote anaesthetic environment because extra equipment and help cannot be immediately accessed. There are many very specialised considerations such as hyperbaric compatible ventilators, monitors and infusion pumps and those managing the patient inside the chamber must be familiar with these. In short, putting an intensive care patient in a hyperbaric chamber is a substantial logistic challenge which can be achieved, but only in hyperbaric units that are equipped, staffed and practiced for such eventualities. In Australia there is a major public hospital hyperbaric medicine unit in every state capital except Canberra. All of these are capable of receiving intensive care patients. In the opinion of this author, if an arterial gas embolism event occurred during cardiac surgery in one of these hospitals, serious consideration should be given to providing expeditious HBO treatment. Although there is no consensus on the matter, in the absence of a tissue inert gas load (which will often exist when divers are recompressed) the recompression protocols may be safely shortened for arterial gas embolism in post-operative patients in comparison to those typically used for divers. It is much less clear what should occur in arterial gas embolism during cardiac surgery at a hospital without a hyperbaric chamber. In this case accessing HBO would involve an inter-hospital transfer. This is obviously much more hazardous in the immediate postoperative period. One compromise would be to stabilise the patient locally, and reconsider the issue when the patient can be better assessed during or after emergence in the intensive care unit. If there are obvious signs of cerebral injury then HBO could be considered. There is some evidence that even delayed HBO may be useful in this setting. [12,16] Full referenced abstract is available on the meeting APP.

Page 23

23 Sleep and cardio metabolic disorders – beware the late night cocktail – Sara Jane Allen “The innocent sleep, Sleep that knits up the ravell’d sleave of care, The death of each day’s life, sore labour’s bath, Balm of hurt minds, great natures second course, Chief nourisher in life’s feast” Macbeth (2.2.46-51) William Shakespeare, 1606 Perhaps William Shakespeare summarised things best in this passage from Macbeth, where he muses on the myriad of positive health benefits of sleep. Whilst the understanding of sleep and the effects of sleep duration and quality has progressed since 1606, there are still unanswered questions regarding optimum sleep duration, individual variation, and the contribution of quality of sleep versus duration of sleep. Short or long duration of sleep, and disrupted or insufficient sleep are common worldwide, and may be increasing. Both short (<7 hours) and long (>9 hours) sleep duration are associated with morbidity and mortality in epidemiological studies, with cardiac (hypertension, myocardial infarction), metabolic (obesity, diabetes, dyslipidaemia), and neurologic (stroke, dementia) negative health outcomes increased in those with excess or inadequate sleep. The association between sleep and mortality is a U-shaped curve, with the lowest risk associated with those individuals with average sleep duration 7-8 hours per night. Insufficient sleep may be due to short duration, or due to poor sleep quality. Various pathophysiological mechanisms are proposed for the adverse health outcomes associated with sleep duration, however the exact mechanisms are not yet proven, and it is not clear whether different mechanisms are involved in short versus long sleep duration adverse outcomes. Indeed, it is likely that there is inter-individual variability in what constitutes insufficient sleep, as well as the susceptibility to the deleterious effects of sleep disruption. Potential mechanisms for the adverse effects of sleep duration include pro-inflammatory states (with elevations in interleukins, TNF-alpha, CRP, and cellular adhesion molecules), altered hormonal pathways (including ghrelin and leptin, growth hormone and cortisol), social and behavioural effects, and genetic susceptibility to the effects of sleep deprivation or excess. This talk and discussion will highlight the current evidence for the role of sleep disturbance in cardiac and metabolic health outcomes, as well as the emerging evidence for the role of sleep disturbance in cognitive dysfunction and decline. Full paper is available on the meeting APP

Page 24

24 Patients’ experience of comfort in healthcare – Cynthia Wensley2,4, Mari Botti3, Ann McKillop4, Alan Merry1 1Green Lane Department of Anaesthesia, Auckland City Hospital, and Department of Anaesthesiology, Faculty of Medical and Health Science, University of Auckland, Private Bag 92019, Auckland 1142, New Zealand. 2School of Nursing and Midwifery, Deakin University, 221 Burwood Highway, Burwood, VIC 3125, Australia. 3School of Nursing and Midwifery, Deakin University, Epworth Deakin Centre for Clinical Nursing Research, 221 Burwood Highway, Burwood, VIC 3125, Australia, 4School of Nursing, University of Auckland, 89-91 Grafton Rd, Grafton, Auckland 1010, New Zealand An important aim in the continuous improvement of healthcare systems is to enhance patients’ experience of care. Patient comfort is an integral, yet understudied, aspect of the patient experience. After reviewing the literature and interviewing cardiac surgery patients, we established a definition of patient comfort that recognises both its physical and emotional components. We further developed a framework in which to assess patient comfort, and strategies for enhancing patient comfort. Addressing the comfort of patients can be expected to improve all aspects of the quality and safety of healthcare. It is our hope that the realisation of comfort and techniques for its improvement will inform healthcare providers’ efforts to further improve patient experience. Funding Conflict of Interest Statement We have no material or financial or other relationship with any healthcare-related business or other entity whose products or services may be discussed in, or directly affected in the marketplace by, this manuscript. Full paper is available on the meeting APP

Page 25

Page 26

Page 27

27 SESSION 3 Friday 08:30 - 10:00 …the wisdom to know the difference International CPB survey – Florian Falter From the early prototypes constructed by Maximilian von Frey in Leipzig in 1885 to the first known cardiotomy with extracorporeal circulatory support by Clarence Dennis in 1951 and subsequent development by Gibbon, cardiopulmonary bypass has become an integral component of modern cardiac surgery. As time as passed, the complexity of bypass has evolved, with a wide range of different options for pump priming and cardioplegia available to surgeon, anaesthetist and perfusionist. Despite the seemingly ubiquitous nature of this important device there is little understanding of how different countries, regions and continents differ from one another in pump priming, anticoagulation or arrest techniques. Several surveys have been undertaken in various countries, however regional patterns in cardiopulmonary bypass practice worldwide remain poorly understood. There are many debates about the use of crystalloid versus cold blood cardioplegia, and the question of whole blood (microplegia) versus diluted blood cardioplegia, both of which have been the focus of recent metaanalyses, which do not demonstrate superiority of either approach. Recent literature concerning colloid versus crystalloid priming strategies and the consequences of Joachim Boldt’s activities add further confusion as to what constitutes best practice. In an attempt to document current practices and identify worldwide variations we undertook a global survey of cardiac anaesthetists to explore contemporary CPB practice. Full abstract is available on the meeting APP

Page 28

28 An old survey; does it tell us anything, the value of a new survey – Tim Willcox Auckland City Hospital Some 50 surveys of perfusion practice have appeared in the literature consistently over the last two and a half decades. Approximately half of these surveyed practice in North America and the remainder evenly proportioned from other international jurisdictions with two surveys of worldwide practice. Initial surveys were by mail or phone then in the following decades increasingly by email then web-based survey formats, each with their strengths and weakness. These surveys have in the main been cross sectional snapshots, targeting specific practice (e.g. anticoagulation, temperature management, blood management), regional practice by country, safety, vocational and behavioural aspects of perfusion. Paediatric practice has been prominent in the survey literature with a series by Groom and colleagues regularly tracking practice in North America over two decades. One of the most comprehensive surveys (An old Survey) was conducted in 2003 looking at practice in Australia and New Zealand. This 245 question survey of equipment, monitoring and management had a high response rate of 89% compared to an average 65% for perfusion surveys over the last quarter century, representing in excess of 20,000 cases. Unsurprisingly it revealed a broad consistency of equipment use with some wide variation in aspects of practice and in shifts in practice from a similar survey by Wajon and colleagues a decade previously. What the old survey does tell us is that this method of interrogation of practice suffers important limitations. It is retrospective relying on respondents’ accuracy of recall, is subject to sampling bias where response rates are low, respondents for a centre may quote protocol rather than actual practice and internal variation is not accounted for and there may be interpretation accuracy issues. The value of a new survey is in fact a shift from traditional surveys to prospective automated accurate data collection into registries such as the Australia New Zealand Collaborative Perfusion Registry. This gives a far more comprehensive picture of practice and a more effective tool for analysing change and benchmarking. However this is not without some of survey limitations especially in respect of representation of regional practice. Full abstract is available on the meeting APP

Page 29

29 Is conventional bypass for coronary artery bypass graft surgery a misnomer?– Robert A Baker, Donald S Likosky on behalf of the Australian New Zealand Collaborative Perfusion Registry, the PERForm Registry and the Michigan Society of Thoracic and Cardiovascular Surgeons Quality Collaborative Background: While recent trials comparing on- vs. off-pump revascularization techniques describe cardiopulmonary bypass (CPB) as “conventional”, inadequate description and evaluation of how CPB is managed often exists in the peer-reviewed literature. We identify and then describe regional- and center-level differences in the techniques and equipment used for conducting CPB in the setting of coronary artery bypass grafting (CABG) surgery. Methods: We leveraged prospectively collected data among isolated CABG procedures submitted to either the Australian and New Zealand Collaborative Perfusion Registry (ANZCPR) or Perfusion Measures and outcomes (PERForm) Registry between 1/1/2014 and 12/31/2015. Variation in equipment and management practices that reflect key areas of CPB are described across 47 centers (ANZCPR: 9; PERForm: 38). We report average usage (for categorical data) or median values (for continuous data) at the center-level, along with the minimum and maximum across centers. Results: 3,562 patients were identified in the ANZCPR and 8,450 in PERForm. Substantial variation in equipment usage and CPB management practices existed (both within and across registries). Open venous reservoirs were commonly used across both registries (nearly 100% in both registries), as were all-but-cannula biopassive surface coatings (>90% in both registries), while roller pumps were more commonly used in ANZCPR (ANZCPR: 85% vs. PERForm: 64%). ANZCPR participants had 640mL absolute higher net prime volumes, due to higher total prime volume (1462mL vs. 1217mL) and lower adoption of RAP (20% vs. 81%). ANZCPR participants had higher nadir Hct on CPB (27 vs. 25). Minimal absolute differences existed in exposure to high arterial outflow temperatures (36.6C vs. 37.0C). Conclusion: We report substantial center and registry differences in both the type of equipment utilized and CPB management strategies. These findings suggest that the term “conventional bypass” may not adequately reflect real-world experiences, and as such should not be used to describe contemporary CPB practice. Full paper is available on the meeting app

Page 30

Page 31

31 SESSION 4 Friday 10:00 – 11:30 Getting a consensus Reporting CPB related morbidity – are we on the same page ? – Donald S. Likosky, PhD1; Timothy A. Dickinson, MS, CCP, CPHQ2, for the PERForm Registry and the Michigan Society of Thoracic and Cardiovascular Surgeons Quality Collaborative Gaps remain in our understanding of the contribution of bypass-related practices associated with post-operative sequelae (clinical and financial) after cardiac surgery. Variability exists in the scope of these practices in the published literature, as well as the reported endpoints. We present a conceptual model for how the scope of practices and outcomes are related. In addition, we present variables worthy of investigation, especially for informing current and future updates to clinical registries. It is our hope that investigators might consider using this conceptual model to improve the evaluation of the relationship between bypass-related practices and cardiac surgical value. Full paper is available on the meeting APP

Page 32

32 Are consensus statements worth publishing? - David Sidebotham The highest grade of evidence on which consensus statements and practice guidelines can be based are randomised, controlled trials (RCTs). How robust are these randomised data? The answer, unfortunately, is not robust at all. When a new treatment is developed, the initial RCTs that are published typically demonstrate a large positive treatment benefit. Such RCTs are usually small and single centred. Subsequently, larger, multi-centre RCTs are published which typically show a diminished – or absent – treatment effect. This phenomenon is known as ‘regression to the truth’ and occurs as a consequence of random effects and publication bias. If no large RCTs are subsequently published, a persistent overestimation of treatment effect will exist. Alternatively, if the results of small RCTs are used as the basis of power calculations for large RCTs, the subsequent large trials will be underpowered. Small, positive RCTs are almost always statistically ‘fragile’. The fragility index is the number of patients that would have to change from ‘non-event’ to ‘event’ for the result to be no-longer statistically significant. In critical care, 40% of RCTs have a fragility index of ≤ 1. There is also emerging evidence that a much higher proportion of published RCTs are subject to major error or data fabrication than has been hitherto appreciated. Unfortunately, the majority of large, multi-centre RCTs published in the field of critical care are negative. A major contributor to this problem is that many large RCTs are underpowered, particularly when mortality is the primary outcome variable. Even if consensus statements and practice guidelines were underpinned by trustworthy, unbiased, randomised data –which they are not – such documents are subject to several problems. Recommendations typically don’t take into account patient age and prognosis. Strict guideline adherence leads to polypharmacy and high costs. The strongest evidence is often for the most expensive treatments. Authors are frequently subject to conflicts of interest. Finally, there is evidence that guideline recommendations have, at best, modest compliance. Full paper is available on the meeting APP

Page 33

33 SESSION 5 Friday 11:45 – 13:15 Turning off the bleeding The Protamine conundrum – Florian Falter Systemic, high-dose anticoagulation is fundamental to the safe conduct of cardiopulmonary bypass (CPB). Globally, unfractionated heparin remains the most widely used agent for this purpose. It is found on the WHO Model List of Essential medications, reflecting affordability, predictable behaviour when being metabolised, and the availability of an effective reversal agent: protamine. However, where heparin is lauded for its ease of use, the administration of protamine is undertaken cautiously, due to its well known side-effect profile, including systemic hypotension, pulmonary hypertension ad severe anaphylactic reactions. Where much of the early focus in the literature was on the negative cardiovascular effects of the drug, attention has recently turned to its anticoagulation properties, particularly in excessive doses, and with special implications for preventing bleeding after CPB. The continued use of relatively high, fixed protamine dosing for the reversal of heparin has been increasingly questioned in the literature due to growing recognition of substantial response variability between individuals. Unfortunately heparin concentration devices are often expensive, and the benefits of using them have been questioned. The last few years have bought the development of statistical and pharmacokinetic models for protamine administration as an alternative. Several recent publications advocate a two-compartment pharmacokinetic algorithm that can be tailored to the individual patient using ideal body weight, and the doses and time intervals of administered heparin before and during CPB. This algorithm is showing promise to safely reduce the protamine dose at the end of CPB without increased bleeding complications. Full abstract is available on the meeting APP

Page 34

34 One Year results from the ATACAS trial – Paul Myles, The risks and benefits of aspirin in cardiac surgery have been debate for many years,1 and although tranexamic acid is a potent antifibrinolytic, it was uncertain whether this results in a reduction in blood transfusion or serious bleeding complications.2 The ATACAS (aspirin and tranexamic acid in coronary artery surgery) trial evaluated two drugs using a factorial design and enrolling 4,631 patients having CABG ± valve/other surgery.3, 4 We found that continuing aspirin up until the day of surgery did not reduce thrombotic complications or death, but also did not increase bleeding, transfusion requirements or any other complications.4 We thus concluded, in part on the basis on an updated systematic review (see figure 1),5 that aspirin should be continued in near-all patients having coronary artery surgery. We similarly found that tranexamic acid had no adverse effects on risks of thrombotic complications after surgery, but did provide a clinical important reduction in bleeding complications, including the need for reoperation.3 Seizures associated with tranexamic acid were confirmed, but the long-term consequences of this remain unclear. A previous moderately sized (n=783) trial has reported on secondary post hoc analysis that preoperative aspirin decreased the long-term (3 years) risk of nonfatal coronary events (MI or repeat revascularization), hazard ratio (HR), 0.58 (95% CI: 0.33-0.99), and seemed to decrease the risk of major cardiac events, HR 0.65 (95% CI: 0.41-1.03).6 These results require confirmation. ATACAS included 31 study sites in 7 countries. We a priori established a 1-year follow study to determine whether aspirin or tranexamic acid reduced the risk of late myocardial infarction or stroke, and improved disability-free survival. The final analysis is currently underway, and we plan to present these hot off the press results at the PDU meeting in August. Funded by the Australian National Health and Medical Research Council (NHMRC) and the Australian and New Zealand College of Anaesthetists (ANZCA) Full referenced abstract is available on the meeting APP

Page 35

Page 36

Page 37

37 SESSION 7 Friday 18:00 – 19:00 off site Prof Merry Lecture ‘Am I really trained to do what I do?” – Tim Jones MD FRCS(CTh) Medicine (noun) - the science or practice of the diagnosis, treatment and prevention of disease As medical practitioners we are all trained to apply the appropriate principles and procedures to take the best care of our patients. The aims of our interventions are to diagnose, treat and take care of patients with illnesses or injury. The success of our intervention is judged by mortality and or complication rates. Fortunately both death and serious complications are increasingly rare events. Does this mean we are doing a good job and taking care of the true needs of our patients? Traditional medical training focuses on developing a good working knowledge of the human body in health and then studying the impacts of pathology and disease. This knowledge is then used to learn diagnostic skills and develop treatment plans. More recently there is increasing recognition of the importance of communication skills, team working and service delivery in providing good health care but these skills remain underrepresented in the curriculum. Providing health care for children differs in many ways to providing health care for adults. Whilst the child is the focus of our attention, healthcare must be delivered in a family centred way including and involving the parents. It is a unique relationship. The parent is not receiving the healthcare but they usually are the person agreeing to the intervention or therapy. It is as important to meet their needs as it is to meet the needs of the child. But as healthcare practitioners are we trained in recognising the needs of our patients and their families? Full abstract is available on the meeting APP

Page 38

Page 39

Page 40

Page 41

41 Saturday 20th August 2017 SESSION 8 Saturday 08:30 - 10:00 Is the cardiac OR a safe haven? ‘Human factors in the operating room – instinct or intelligence’ – Tim Jones MD FRCS(CTh) Cardiac surgery is a complex activity. During an operation there are planned and unplanned events that need to be responded to and dealt with appropriately. The success of the outcome is dependent upon a well-functioning multidisciplinary team. Increasingly the role of good communication, decision making, leadership and followership are recognised as having an important impact upon patient outcome. Almost 50% of errors in the operating room are due to poor communication. The importance of such human factors and non technical skills is recognised in many high risk practices outside of health care. There are very few medical programmes that incorporate human factor training or development as a routine. Traditionally, surgical trainees were described as having or not having ‘that factor X’. This was never defined but usually related to their ability to communicate and deal appropriately with situations when they arose. Those perceived lacking ‘factor X’ were often encouraged to pursue other career paths. Using high fidelity simulation we can providing an artificial representation of a real world process such as an unexpected event in the operating room. The aim is to provide education and training via experiential learning but are human factors something that can be learnt or are they a binary occurrence either present or not in an individual? It has been established that repeated practice in a simulated environment improves learner outcomes with regard to technical skills. Using these educational principles we have developed a developed a curriculum for human factor training that all cardiac surgical trainees now undergo in UK. We now need to develop the principles to training the wider multidisciplinary operating team. Full abstract is available on the meeting app

Page 42

42 “CHECKING THE CHECKERS: PATIENT SAFETY IN THE OPERATING ROOM -Simon Mitchell, Department of Anaesthesiology, University of Auckland. There is a long-standing awareness that many deaths and complications occurring in the peri-operative period arise from errors or omissions in care and are preventable [1,2]. In the mid-2000s the World Health Organisation (WHO) spearheaded a program to identify strategies to reduce perioperative iatrogenic harm. One initiative arising from this program was the “Surgical Safety Checklist” (SSC). This checklist advanced extant perioperative checking processes in two key ways. First, it was administered in the operating room (OR), as opposed to checks in pre-operative areas, and involved all staff actually performing the operation. Second, it was split into 3 phases or “domains” called “Sign In” (administered when the patient first arrived in the OR), “Time Out” (administered just prior to the first surgical incision), and “Sign Out” (administered prior to the patient leaving the OR). At each of these times a series of checklist “items” were read out; typically by a theatre nurse working from a paper copy of the checklist. The effect of introducing the SSC to clinical practice was first evaluated in a multicentre study where outcomes were recorded for 3733 patients undergoing surgery prior to checklist roll out, and 3955 patients after roll out. [3] In-hospital mortality and complications fell from 1.5% and 11% in the pre-SSC phase to 0.8% and 7% in the post-SSC phase. This spectacular result was controversial and prompted much discussion about the relatively weak uncontrolled methodology and the potential confounding effects of involving third world hospitals in the study. However, the following year, another study which included control hospitals and which was conducted in an entirely first-world setting demonstrated that introducing a surgical care map with use of a SSC as its centrepiece produced almost identical improvements in patient outcomes. [4] Since then there have been multiple studies (including two randomised trials) that corroborate the early indications that use of an SSC is associated with improved patient outcomes. [5-8] Only one study has been negative, [9] and it has been interpreted as demonstrating that merely adopting a checklist “in name” but not ensuring its proper assimilation into practice by consultation, education and audit is unlikely to reap the checklist’s potential benefits. [10] This issue has, in fact, shifted the community’s focus from a debate on whether checklists work, to how to ensure they are used properly. Our hospital emerged from participation in the original WHO SSC study [3] with what we perceived as an acceptably well-developed checklist culture in the operating room. However, audits of our practice published in 2011 and 2013 revealed a disturbing picture. [11,12] Our compliance with initiating the Sign In and Time Out domains of the checklist was excellent (>95% of cases) but Sign Out was usually not performed, there was variable compliance with administering many checklist items even when the domains were initiated, and true engagement in the process (a measure of the extent to which OR team members gave the checklist their full attention) was poor. Moreover, we discovered that actual compliance with administering checklist items fell well below that indicated by checklist tick-box records. For

Page 43

43 various reasons some nurses had become more focussed on having the paperwork completed than on meaningful application of the checklist. As a consequence, after using a paper-based nurse-led administration paradigm for 10 years, we adopted a paperless wall-mounted checklist with leadership of the 3 domains migrated among senior clinicians in the three OR sub-teams typically present (anaesthesia, surgery and nursing). This change was prompted by our own observation of an apparent positive effect of this paradigm on OR team engagement at another local hospital. [12] Others had also alluded to possible positive effects on compliance though use of wall-mounted posters and migration of leadership roles among the teams most engaged in the patient management process at the time of checklist administration. [13-15] Implementation of this new paradigm in our operating suite was associated with substantial improvements. Compliance with administration of the Sign Out domain improved from 22% of cases in 2013 to 84%. [16] Concomitantly, the incidence of surgical specimen labelling errors (specimen labelling is checked at Sign Out) was more than halved in the 6 months following roll out of the new paradigm. [17] In addition, compliance with administration of the individual checklist items within all domains, and engagement of the OR staff in the checklist process was vastly improved. [16] In the latter regard, it is notable that during hundreds of checklist domain observations in the 2013 study we never once recorded proper engagement of every staff member present in the OR during checklist administration. [12] Although still far from universal in the present study, ‘total staff engagement’ was seen in at least some cases. The proportion of cases in which at least one member of each sub-team was fully engaged at Time Out improved from 14% in 2013 to 93% of cases observed in 2015. [16] The cardiac OR environment is arguably more complex than most other settings in which the SSC might be used, not least because a fourth team (Perfusion) is central to the surgical process. Thus, the lines of communication are more complicated and the opportunities for errors and omissions are magnified. Committed engagement in an SSC process is an evidence-based strategy likely to improve patient outcomes in this setting (perhaps markedly so), and the paperless, wall-mounted checklist with migrated leadership is commended as an approach to the SSC that has been demonstrated to foster compliance and engagement in its use. Full referenced abstract available on the APP

Page 44

Page 45

45 SESSION 9 Saturday 10:30 – 12:30 Trial Architecture Days Alive and Out of Hospital - Alan Merry1, Doug Campbell, Harry Alexander, Adam Bartlett, Yan Chen, David Cumin, Derryn Gargiulo, Richard Hamblin, Jacqueline Hannam, Simon Mitchell, Matthew Moore, Craig Webster, Jennifer Weller. 1Department of Anaesthesiology, Faculty of Medical and Health Science, University of Auckland, Private Bag 92019, Auckland 1142, New Zealand and Green Lane Department of Anaesthesia, Auckland City Hospital, Park Rd, Auckland, New Zealand. For better or worse, research is becoming increasingly judged by its impact on concrete outcome measures. To this end a recent discussion paper proposed the following terminology in a results chain model for science: inputs, activities, outputs, outcomes and impacts. Additionally, public advocacy groups continue to demand access to data pertaining to the performance of healthcare providers. Current standard measures of morbidity and mortality do not lend themselves to either of these applications, with problems surrounding their sensitivity and ease of collection. Days alive and out of hospital (DAOH) is a measure that captures both mortality and morbidity on a continuous scale, and does so in a readily collectable fashion. Notwithstanding some mitigable implementation and statistical issues, DAOH represents a promising path forward in the operationalization of surgical outcomes. Full referenced paper available on meeting APP

Page 46

46 The diagnostic properties of current AKI consensus criteria – Donald S. Likosky, Ph.D.1; Michael Heung, M.D., M.S.2; Min Zhang, Ph.D.3; David Sturmer, CCP1 (1) Section of Health Services Research and Quality, Department of Cardiac Surgery, University of Michigan, Ann Arbor, Michigan; (2) Department of Internal Medicine, University of Michigan, Ann Arbor, MI; (3) Department of Biostatistics, University of Michigan, Ann Arbor, MI; Introduction: Patients undergoing cardiac surgery are at risk for developing post-operative morbid events, including acute kidney injury (AKI). Although widely reported, current methods for diagnosing post-operative AKI do not account for a patient’s fluid balance during cardiopulmonary bypass let alone differences in creatinine generation across patient subgroups (elderly, sex, race). Methods: We evaluated 5,365 patients ≥ 18 years of age undergoing cardiac surgery (absent circulatory arrest) at Michigan Medicine between 1/2010-7/2014. We compared differences in rates of AKI using the Kidney Disease: Improving Global Outcomes creatinine criteria relative to one that additionally adjusted for a patient’s intra-operative net fluid balance. We then evaluated differences in AKI rates across biological subgroups. Results: The difference between the net fluid balance adjusted and non-adjusted creatinine was 0.039 mg/dL (SD 0.067). Relative to a diagnosis of AKI by existing criteria, 0.82% (95% CI: 0.6%-1.1%) of patients were reclassified as not having AKI after adjusting for net fluid balance (false positive), and 2.63% (95% CI: 2.22%-3.07%) of patients were reclassified as having AKI (false negative). Non-significant differences existing in false positive and false negative rates across race. Conclusion: Neglecting a patient’s net intra-operative fluid balance potentially subjects patients to either false positive or false negative AKI diagnoses. Based on current criteria, some patients may be exposed to under or over treatment. While larger external validation studies are warranted, investigators should consider incorporation of a patient’s net fluid balance into existing AKI classification criteria to provide a more accurate and unbiased estimates of AKI. Full referenced paper available on meeting APP

Page 47

47 A Novel Clinical Trial Design: practice1 preference-randomised consent - Paul S. Myles, We are proposing novel trial design for the Dexamethasone for Cardiac Surgery Trial (DECS-II), a study comparing high-dose dexamethasone with placebo in patients undergoing cardiac surgery.1 There is a need for improved efficiencies in medical research,2 and this has led to growing interest in novel trial designs.3-6 There are three key factors that need to be overcome if further improvements are to be made in clinical research: (i) clinician burden (time, effort) and limited research assistant/nurse resources, especially when half the patients approached will not in fact receive the experimental (new) treatment, (ii) clinician discomfort with patients that highlight their therapeutic uncertainty, and (iii) patients may have a treatment preference but it is not available outside of the trial. Conventional randomization includes a need for patient consent for at least two treatment options, only one of which the patient will receive. An alternative approach, first proposed in 1979,7 is to gain consent for experimental treatment after randomization, so that only those receiving experimental treatment are notified and approached for their consent. Variations in “standard” clinical care are widespread. This variation can be harnessed. We propose a modification to the Zelen design for Phase IV multicenter clinical trials that forms two aggregates of sites (e.g. different countries) according to their current standard of practice. This can increase clinician engagement and should greatly facilitate patient enrolment, so that a large fraction of eligible patients can be easily enrolled into any particular study. In other words, it acknowledges strong clinician preference but balances this aspect by matching an equal number of sites with opposing standards of practice. We refer to this as a practice preference-randomized consent (PP-RC) approach. Whilst all patients are offered the opportunity to opt out of consideration for trial inclusion when they are booked for surgery, informed consent is only sought if they have been allocated to the alternative (non-standard) treatment. This proposed PP-RC design should enable many Phase IV clinical trials addressing existing treatment options or variations in care to be conducted in an extremely timely and cost-efficient manner. It will speed up enrolments and greatly reduce the cost of conducting trials. By facilitating a high proportion of eligible patient enrolments it will more readily demonstrate real-world effectiveness. The PP-RC design can be readily adapted to non-drug trials evaluating devices or healthcare practices such as medical versus surgical treatment. Full referenced abstract available on meeting APP

Page 48

Page 49

49 SESSION 10 Saturday 13:30 - 1500 Oxygen Delivery The Effect of Goal-Directed Perfusion on Postoperative Acute Kidney Injury in Adult Cardiac Surgery Patients. A Multicenter Randomized Controlled Trial. Marco Ranucci, MD, FESC1, Ian Johnson, CCP2, Seema Agarwal, FRCA2, Timothy Wilcox, CCP3, Rachael Parke, MD3, Robert A. Baker, PhD, CCP4, Richard F. Newland, CCP4; Christa Boer, MD, PhD5, Renard G. Haumann,CCP5, Andreas Baumann, MD6 ,Dirk Buchwald,Phd,CCP6, George A. Justison, CCP7, Nathaen Weitzel, MD7; Filip de Somer, CCP8, Paul Exton, BSc (Hon) ACP9, Rajamiyer Venkateswaran, MD FRCS(Cth)9, and Valeria Pistuddi1 1Dept. of Cardiothoracic and Vascular Anesthesia and ICU, IRCCS Policlinico San Donato, San Donato Milanese, Milan, Italy; 2Dept. of Perfusion and Dept. of Anaesthesia, Liverpool Heart & Chest Hospital, Liverpool, UK; 3 GreenLane Cardiothoracic Unit and Cardiovascular Intensive Care, Auckland City Hospital, Auckland New Zealand, and Dept of Anaesthesiology, Faculty of Medical and Health Sciences, University of Auckland, Auckland New Zealand; 4 Cardiac Research and Perfusion, Cardiac and Thoracic Surgical Unit, Flinders Medical Centre and Flinders University, Adelaide, South Australia; 5 Depts. of Anesthesiology and Cardio-thoracic surgery, Institute for Cardiovascular Research, VU University Medical Center, Amsterdam, the Netherlands; 6 Dept. of Anaesthesiology, Intensive Care, Palliative Care and Pain Medicineand Dept. of Cardiac and Thoracic Surgery, BG University Hospital Bergmannsheil, Ruhr-University Bochum, Germany; 7Dept. of Perfusion and Dept. of Anesthesiology, University of Colorado Denver, Aurora, CO, USA; 8 Heart Centre, University Hospital Ghent, Ghent, Belgium; 9 Department of Cardiothoracic Surgery, University Hospital of South Manchester NHS Foundation Trust, Manchester, UK. Importance. No cardiopulmonary bypass techniques have demonstrated the ability to reduce the risk of acute kidney injury following cardiac surgery. Objective. To determine whether a goal-directed perfusion strategy, aimed to maintain an oxygen delivery above 280 mL.min-1.m-2 reduces the incidence of acute kidney injury. Design, setting, and participants. This multicenter randomized trial enrolled 350 patients undergoing cardiac surgery in nine institutions. Patients were randomized to receive either goal-directed or conventional perfusion. Three hundred and twenty-six patients completed the study and were analyzed according to the intention-to-treat principle.

Page 50

50 Interventions. Patients in the treated arm received a pump flow during cardiopulmonary bypass aimed to reach and maintain an oxygen delivery ≥ 280 mL.min-1.m-2. Patients in the control arm were treated at a conventional pump flow of 2.4 L.min-1.m-2. Main outcome and measures. The primary endpoint was the rate of acute kidney injury defined according to the Acute Kidney Injury Network Criteria. Secondary endpoints included intensive care unit stay; major morbidity; rate of patients receiving packed red cells and number of units transfused; operative mortality (in-hospital or within 30 days from surgery after discharge). Results. Acute kidney injury stage 1 was reduced in patients treated with goal-directed perfusion (relative risk 0.45, 95% CI 0.25-0.83, P = 0.01. Acute kidney injury stage 2-3 did not differ between groups (relative risk 1.66, 95% CI 0.46-6.0, P = 0.528. There were no significant differences in secondary outcomes. In a pre-specified analysis on patients with a pump run between 1 hour and 178 minutes the differences in favor of the treatment arm were more pronounced, with a relative risk for acute kidney injury stage 1 of 0.39 (95% CI 0.21-0.75, P = 0.004) and a relative risk for acute kidney injury of any kind of 0.49 (95% CI 0.27-0.89, P = 0.017). Conclusions and relevance. A goal-directed perfusion aiming to preserve oxygen delivery during cardiopulmonary bypass is effective in reducing minor degrees of postoperative acute kidney injury. Trial registration: clinicaltrials.gov identifier: NCT02250131 Abstract available on meeting APP

Page 51

51 Supporting the RCT’s: ANZCPR – Richard Newland, Flinders Medical Centre, Adelaide, South Australia Intraoperative risk factors for acute kidney injury (AKI) for patients undergoing cardiopulmonary bypass (CPB) are not clearly defined. Two previous reports have identified minimum oxygen delivery index (DO2i) during CPB to be associated with AKI. This study aims to support these findings by evaluating a larger propensity-matched patient population. The Australian and New Zealand Collaborative Perfusion Registry provided multicentre data from 18,343 adult patients undergoing CPB. A critical threshold for low DO2i was determined using a Lowess plot of minimum DO2i and the probability of AKI and furthermore from the Youden’s index and cut off point on the receiver operating characteristic (ROC) curve that provided optimal coupling of sensitivity and specificity. The predictive value of the association of DO2i with AKI was assessed using the area under the ROC curve. Low DO2i was defined as <270ml/min/m2 based on the inflection point of the Lowess plot. Influence of DO2i was evaluated in 9156 patients matched on their propensity for AKI. Mixed effects logistic regression was used to determine the relationship between minimum DO2i and AKI with adjustment for red blood cell transfusion. Low DO2i was found to be an independent predictor of RIFLE class R or greater AKI (OR 1.21, CI (1.05-1.39), p=0.007). The ROC curve provided a cut off of 250 ml/min/m2 based on the Youden’s index, and 265 ml/min/m2 based on the optimal coupling of sensitivity and specificity (sensitivity 57%, specificity 55%). The area under the ROC curve was 0.58 (CI: 0.57-0.59). These results from multi-centre data support previous findings that low oxygen delivery during CPB is associated with AKI with a critical threshold found to be between 250-270 ml/min/m2. A targeted approach to maintaining oxygen delivery above this threshold may help decrease AKI following cardiac surgery. Full paper available on meeting APP

Page 52

Page 53

53 SESSION 11 Saturday 1530 - 17:00 Tales of the unexpected Healthcare Associated infections: The role of a mixed methods approach - Associate Professor Donald S Likosky Informing Targeted Improvement Efforts in Reducing Pneumonia After Cardiac Surgery: A Qualitative Study Background: Wide variability exists in the rates of healthcare-associated infections (HAI) following coronary artery bypass grafting (CABG) surgery. While pneumonia is the most common type of infection in this setting, much of this variation does not appear to be attributable to differences in patient case mix. A number of pneumonia prevention strategies have been reported (e.g., nasal decolonization or early mobility protocols); nonetheless, while not studied to date, differences in their use and manner of adoption may improve our understanding of hospital-level variation in pneumonia rates. We herein present a methodological approach for developing recommended pneumonia preventive practices in the setting of CABG surgery. Methods: We undertook 79 semi-structured interviews with key healthcare personnel and safety officers across 10 institutions in Michigan. Medical centers were chosen based on their pneumonia rate, and included 5 with low (1.9% - 3.4%), 2 with medium (5.9% - 6.1%), and 3 with high rates (10.0% - 19.1%). We employed a Rapid Assessment Process to: (1) identify key pneumonia prevention practices, processes and procedures, and (2) begin developing an understanding of potential determinants of center-level differences in pneumonia rates. Results: A list of recommendations for pneumonia prevention, based on the literature and on findings emerging from data collected through our qualitative work, included: frailty assessment, oral/nasal/skin preparation and treatment, pulmonary function testing, blood conservation, intensive care unit management, subglottic endotracheal tubes, ventilator management, and progressive mobility. Conclusions: Our methodological approach resulted in the identification of a number of clinical practices that may reduce a patient’s risk of pneumonia following CABG surgery. Current areas of investigation include evaluating whether adoption of these targeted recommendations is associated with lower pneumonia rates. This paper is currently under review and is available on the meeting APP and forms the basis for part of the presentation.

Page 54

54 The NTM saga – Sally Roberts Global outbreak of Mycobacterium chimaera infection associated with cardiac bypass heater unit units. Sally A Roberts FRACP, FRCPA. Department of Microbiology, Auckland District Health Board, Auckland, New Zealand In 2012 two patients at a Swiss hospital were diagnosed with invasive Mycobacterium chimaera infection. M. chimaera is a slow-growing mycobacterium that is ubiquitous in the environment and survives well in water. An in-depth investigation identified M. chimaera as the cause of a prolonged healthcare-associated outbreak of invasive disease in patients who had undergone cardiac by-pass surgery. The outbreak was traced to water in the heater cooler unit (HCU) reservoirs and an airborne transmission pathway was established. Subsequently, reports of similar infections have occurred globally. The incubation period is prolonged with a median of 19 months following surgery reported from the UK. Nearly all patients have had prosthetic valves or graft material implanted. Presentation varies from surgical site infections to disseminated disease but endocarditis being the most common presentation. The mortality rate is high due to the challenge of diagnosing these infections. Whole genome sequencing of isolates obtained from HCU from around the world showed that they were genetically indistinguishable supporting a point source outbreak. The majority of environmental isolates were associated with one specific type of HCU, the 3T HCU (Sorin, Milan Italy). Isolates from patients and individual HCU have been shown to be indistinguishable confirming these devices as the source of infection. This outbreak has highlighted the importance of water-containing medical devices as potential sources of healthcare-associated infections. Water-containing medical devices should be designed in such a way as to prevent aerosolization of water. Adherence to the manufacturer’s recommendations for cleaning and disinfection of the devices is essential; however, additional work is still required to validate the cleaning and disinfection process and to determine the required quality of the water. Abstract available on meeting APP

2017 PDU Programme

Page 1

Page 2

Page 3