2013 PDU Programme : simplebooklet.com

3 Contents Welcome to Perfusion Downunder 2013 ................................................................ 5 Keynote Speakers .................................................................................................... 7 Program 7 Sunday 1st September 2013 ................................................................................... 17 SESSION 1 Sunday 16:15 - 17:15 Keynote Address Developing and Executing Quality Improvement Projects (concept, methods and evaluation) ........................................................................................................ 17 SESSION 2 Sunday 18:00 - 19:00 The Professor Merry Lecture The Third Man ................................................................................................... 27 Monday 2nd September 2013 ................................................................................ 29 SESSION 3 Monday 09:00 - 11:00 Teambuilding PDU Survivor .................................................................................................... 31 SESSION 4 Monday 11:30 - 13:00 Teamwork Clinical Microsystems: A Critical Framework for crossing the Quality Chasm ... 33 Teamwork, Communication, Formula-One Racing and the Outcomes of Cardiac Surgery ............................................................................................................. 39 SESSION 5 Monday 14:00 - 15:30 Blood and Volume Management Patient Blood Management in Adult Cardiac Surgery ....................................... 47 Monitoring Coagulation during Cardiac Surgery ............................................... 53 SESSION 6 Monday 16:00 - 17:30 PDUC Blood Management Rationalising Red Blood Cell Transfusion in Cardiac Surgery: A Multicentre Quality Improvement Initiative of the Perfusion Downunder Collaboration. ....... 63 Tuesday 3rd September 2013 ............................................................................... 671 SESSION 7 Tuesday 08:00 - 09:30 Neurocognitive/Brain Assessment Cardiac Surgery, the Brain and Inflammation ................................................... 71 Monitoring Rare Outcomes in Cardiac Surgery ................................................. 81 SESSION 8 Tuesday 10:00 - 11:30 Outcomes Gastrointestinal Complications in Cardiac Surgery ........................................... 85 Prevention in Lung Injury in Cardiac Surgery: a Review ................................... 93 SESSION 9 Tuesday 11:45 - 12:30 Meaningful Outcome Measures in Cardiac Surgery.…………………………………113

Page 4

4 SESSION 10 Tuesday 13:30 - 15:30 Bypass to suit the Patient Minimising Prime Volumes – An Anaesthetists Perspective ........................... 119 Minimising Prime Volumes –A Perfusionists Perspective ............................... 121 Fluid Therapy and Outcomes: Balance is best ................................................ 125 Cardioplegia as a Determinant of Myocardial Damage, in Hospital Mortality, and Long Term Survival post Cardiac Surgery ...................................................... 133 SESSION 11 Tuesday 15:45 - 17:00 Free Papers Low State Entropy Scores on Cardiopulmonary Bypass and Association with Mortality and Major Morbidity. ......................................................................... 149 Comparison of EurosSCORE, EuroSCORE II and AusSCORE for Isolated Coronary Artery Bypass Grafting in New Zealand........................................... 151 Real-time Continuous Pulse Oximetry Monitoring during Normothermic Pulsatile Perfusion ......................................................................................................... 153 Does changing the Priming Fluid of the Heart-Lung Machine have Clinical Effects? ........................................................................................................... 155 SESSION 12 Tuesday 17:10 – 18:00 Influencing Change and Outcomes Formula 1 Racing, Red Dogs and the Green Lane Way ................................. 157

Page 5

5 Welcome to Perfusion Downunder 2013 It is a real pleasure to welcome you to this the 9th Perfusion Downunder Winter meeting and the 4th at the renowned Hayman Island Resort. This year we are arguably at winter’s end, having pushed the meeting back four weeks to get the best out of the Whitsunday’s weather and have a winter meeting in the tropics. This has opened opportunities to enjoy more of what Hayman has to offer. We are delighted to have a stellar faculty with keynote guests Donny Likosky from Michigan and David Scott from Melbourne who are joined by Paul Myles and Rob Young from Australia, Sara Allen and Alan Merry from New Zealand and Mike Pouliss from the UK. Richard Newland, Rob Baker and that other Kiwi perfusionist will again be at the lectern, and this year we have free papers from Keshavan Kanesalingam (Aust) Acrane Li and Matthew Haydock (NZ) and Yves Durandy (France). The academic programme covers outcomes of perfusion and cardiac surgery, the team dynamic on that patient journey and perspectives on blood management that is to be a focus of the PDU Collaboration. The PDU team has incorporated a social programme with a fantastic potpourri of casual, ethnic and al fresco dining and a challenging (and undoubtedly hilarious) activity that integrates with concepts from the classroom. For the final night we are hugely excited to have members of the up and coming Melbourne based band Winter Moon to entertain us at dinner, together with our after dinner speaker – the inimitable Dr David Sidebotham, raconteur (and intensivist). So there’s lots of opportunity to have time outside the lecture room to relax and enjoy the renowned PDU camaraderie and chat with the faculty over a glass. Rob, Michael, Richard and I would like to thank our sponsor, Cellplex Pty Ltd, for their on-going unconditional support for this meeting. Finally, thank you for attending our meeting and we look forward to an exciting few days. Without your participation we cannot hope to have this exceptional meeting. 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 Richard Newland, CCP, Flinders Medical Centre, Adelaide, Australia Wayne Pearson, Managing Director, Cellplex Pty Ltd, Melbourne, Australia Jill Futter, Sales Manager, Cellplex Pty Ltd, Melbourne, Australia Bernardette Tackney, Conference Co-ordinator, Cellplex Pty Ltd, Melbourne, Australia

Page 6

Page 7

7 Keynote Speakers Donald S. Likosky, PhD Associate Professor Section Head, Section of Health Services Research and Quality Department of Cardiac Surgery Center for Healthcare Outcomes and Policy (CHOP) University of Michigan Medical School Ann Arbor, Michigan David A. Scott, MB BS, FANZCA, PhD, FFPMANZCA Associate Professor and Director of Anaesthesia, St Vincent’s Hospital, Melbourne, Australia Associate Professor, Faculty of Medicine, Dentistry and Health Sciences, University of Melbourne, Melbourne, Australia Invited Faculty Alan F. Merry, FANZCA Professor and Head of the School of Medicine, Auckland City Hospital Auckland New Zealand Robert Young, FANZCA Specialist Anaesthetist Flinders Medical Centre Adelaide Australia Mike Poullis, BSc(Hons), MBBS, MD, MIEEE, FRCS(CTh) Consultant Cardiothoracic Surgeon Liverpool Heart and Chest Hospital Liverpool United Kingdom Paul S. Myles, MB BS, MPH, MD, FCARCSI, FANZCA, FRCA Director of Anaesthesia and Perioperative Medicine Alfred Hospital Monash University Melbourne, Australia Sara J. Allen, BHB, MBChB, FANZCA Intensivist/Anaesthetist GreenLane Department of Cardiothoracic & ORL Anaesthesia & CVICU Auckland City Hospital Auckland New Zealand PDU Faculty Assoc. Prof Rob Baker Flinders Medical Centre, Adelaide, Australia Mr Richard Newland Flinders Medical Centre, Adelaide, Australia Mr Tim Willcox Auckland City Hospital, Auckland, New Zealand Mr Michael McDonald Perfusion Services, Melbourne, Australia

Page 8

Page 9

Page 10

Page 11

Page 12

Page 13

Page 14

Page 15

Page 16

Page 17

17 Sunday 1st September 2013 SESSION 1 Sunday 16:15 - 17:15 Keynote Address Developing and Executing Quality Improvement Projects (concept, methods and evaluation) Donald S. Likosky, PhD, USA Associate Professor Section Head, Section of Health Services Research and Quality Department of Cardiac Surgery Center for Healthcare Outcomes and Policy (CHOP) University of Michigan Medical School Ann Arbor, Michigan Overview The foundation of science is reproducibility. Many of us recall being taught the fundamentals of research: the scientific method. Our teachers engrained in us that research requires us to: (1) state a testable hypothesis, (2) design and explicitly describe our methodology so that others could reproduce our results, (3) describe the results of our experiment, and (4) state the conclusions of our experiment. If followed, this recipe would allow us to use objective criteria for assessing the entire experiment, as well as archive our work in such a way that others might verify or refute our findings. Our teachers instructed us to apply this model throughout our education. For instance, in chemistry laboratories, many of us undertook titration experiments in college, whereby we produced sodium chloride crystals from the neutralization of sodium hydroxide by hydrochloric acid. While not all of us produced crystals, our teachers reviewed our laboratory handbook as a way of reviewing our experimental methods and findings. To this end, we learned that science requires rigor and documentation not only of our results, but indeed the methods. Fewer of us have been exposed to the science underlying quality improvement. While the science underlying this field is certainly evolving, it requires an equivalent amount of conviction and precision (1). Unfortunately such methods are often not taught in medical school or equivalent post-graduate education. As a consequence, we miss important opportunities to learn from each other regarding how best to transform health care, or more explicitly cardiovascular perfusion. This paper will inform the reader about how to execute quality improvement studies, including their design, execution, and evaluation. Prior publications have focused on the topic of developing research questions, synthesizing the literature, and using databases to address particular clinical questions (2–4). It is the hope that investigators interested in conducting quality improvement studies will use this paper as a guide.

Page 18

18 Description Many of us throw around terms such as research, quality assurance, and quality improvement, and have little notion of what these terms really mean (see box below). While seemingly mundane, such definitions are important as we discuss how to embark on developing and executing quality improvement studies. Research may be defined as generalizable knowledge, or knowledge that may be germane or applied outside of the population under investigation. Quality assurance may be defined as the systematic study or evaluation of a service or product to ensure its performance meets predefined standards for quality. Quality improvement may be defined as a systematic approach for reducing waste or improving efficiency, reliability and performance of a service or product. Beyond performing chemistry experiments, many of us have participated in research projects. Many of these might have been in vitro experiments (those conducted in test tubes or involving “wet labs”), while others have been in vivo experiments (those conducted on humans). For the latter, we recognize the importance of communicating with our Institutional Review Board (IRB, “ethics board”), which has the jurisdiction to determine whether proposed studies involving human subjects should be conducted or not. We strive every day in the operating room to provide the best cardiac surgical care for our patients. However, what defines quality from a perfusion standpoint? As a starting point, we might define quality based on our performance in a given clinical case: the percentage of measured ACT values that are within a centre’s targeted range. As another example, we might use a measure that reflects the care we provide over a series of patients, such as the percentage of cases in which a patient is transfused with red blood cells. We use such information to assess our own performance, with the goal in this example of having fewer patients being transfused. While we have certainly witnessed a greater level of scrutiny of perfusion practice over the last several decades, real, substantial, and long-lasting improvements in care require us to develop expertise in how to transform health care (5). It is no longer sufficient for us to simply perform perfusion as part of our job; instead, we must also lead efforts in transforming health care. Unfortunately, few of us have had real or tangible experience leading quality improvement projects. In the absence of this knowledge or experience, we are unable to tackle this new challenge. As a starting point, perfusionists should become adept at understanding how to develop a concrete study question based on a clinical concern, design a test of change, collect and feed data back to team members, and report their results. A concrete example may be informative and serve as a useful guide for those wishing to embark on these studies. Concept: As a faculty member previously at Dartmouth Medical School, I participated in a number of studies related to red blood cell (RBC) transfusions, especially as a member of the Northern New England Cardiovascular Disease Study Group (NNECDSG) (6). We noted morbidity associated with RBC transfusions, with patients exposed to 1-2 units of RBCs having significantly higher odds of short- and long-term morbidity and mortality (7). At Dartmouth-Hitchcock Medical Center (DHMC), we wished to rationalise RBC transfusion practice, so that RBC practices were standardised across our cardiac surgical program. To this end, we embarked

Page 19

19 on an 18-month study to reduce the variability in RBC transfusion practice at our institution (8). Study goal: We sought to reduce the number of perioperative transfusions associated with cardiac surgery. Where did we start? I submitted and received an intramural grant to fund this project through DHMC’s Quality Research Group Program. The project was divided into three phases: 1. Documenting the status quo (understanding which patients receive blood, indications for transfusions, how decisions are made, who makes the decisions, etc.). 2. Educating staff members regarding the risks and benefits of transfusions, developing new protocols to guide transfusion practice, and assessing the impact of transfusions. During this phase, team members received monthly reports reflecting transfusion practices. Data were plotted over time using process control charts. 3. Assessing whether the intervention resulted in lasting reductions in transfusion practices (that is, were early results just the “Hawthorne Effect”). Methods: With the study’s goal and approach identified, we embarked on developing the requisite tools for each phase of the study. First, and prior to any data collection, we submitted an institutional board review (IRB) to govern the data collection and analysis for this quality improvement study. We initiated this process at the same time that we submitted our intramural grant. We recognized that our study would require patient identifying information, although would have minimal risk to patients. N.B. IRB personnel can help you determine the appropriate type of submission for your study. Some types of studies may require less paperwork and overall review than others. It often is not as onerous as you might imagine. Lesson Learned 1: Start the IRB process earlier, rather than later. We developed a multi-disciplinary team, composed of a perfusionist, intensive care unit nurse, pathologist, surgeon, anaesthesiologist, and quality improvement experts. The team met on a routine basis to ensure that the study progressed. We recognized that we needed to first map the current process (Figure 1) for how transfusion decisions were made both in the operating room and in the post-operative setting. Interestingly, each team member provided a different perspective. We leveraged existing data sources to assess patient demographics, blood product delivery, and resource utilization; there was no need to duplicate existing data collection. However, we needed to develop a data form to assess the indications for each unit of transfusion, to document how decisions were made, and who made the decision. Lesson Learned 2: Develop and support a multi-disciplinary team, and meet often and frequently with the team. Once we had IRB approval, we next piloted our data form with our staff. We were particularly interested in determining some of the following: (1) feasibility of our data collection process: were the fields on our form actually available through the medical record?, (2) how onerous was the data collection process, and (3) could we validate that we received all of the requisite data collection forms?. Of note, the published

Page 20

Page 21

Page 22

22 Lesson Learned 3: Data collection instruments will likely need to be modified over the course of your study. In any event, information gathered through your study needs to be shared with your team on a frequent basis. The second phase of our study was focused around developing and executing our multi-faceted intervention. We started with inviting an outside speaker to present at multi-disciplinary grand rounds to discuss the evidence-base supporting blood transfusions. We shared medical literature with our intra- and post-operative team members. Following this talk, we developed a consensus (Figure 3) among our cardiac surgical team regarding our threshold for transfusing patients during the intra- and post-operative period. This consensus was informed by our prior work and the evidence base within the medical literature, although did not result in a formal protocol. We subsequently began feeding data back to the team on a monthly basis concerning the percentage of patients who were transfused according to our consensus. Lesson Learned 4: Quality improvement studies require sound methodological approaches. The third phase of our study was focused on documenting the effectiveness of our interventions in terms of reducing red blood cell transfusion practices. In non-randomized trials, there is often concern that clinicians modify their behaviour because they are being studied. As such, any measure of effectiveness is thereby contaminated in some regard. In order to address this concern, we studied the effectiveness of our intervention through our ongoing clinical registry data. During this phase, we stopped meeting as a multi-disciplinary group, and stopped sharing monthly data quality reports with team members. Lesson Learned 5: Quality improvement studies are susceptible to chance, bias, and confounding. Evaluation: Quality improvement studies are iterative in nature. By iterative, I mean that they inevitably involve conducting small tests of change. Based on data feedback coupled with generalizable and context knowledge, teams might initiate changes in care and subsequently determine whether the changes resulted in measurable improvement (N.B. not all changes will result in measurable improvement). Such studies require careful stewardship, including appropriate documentation of multi-disciplinary meetings, documenting how changes in the process of clinical care affect the local care team, estimating resources necessary for supporting local practices changes, and synthesizing the findings of your tests of change and overall project learning. Such detailed information is in part what distinguishes these studies from traditional research projects. In our own work, we found that 59% of intra-operative and 43% of post-operative transfusions were for low haematocrit, while 16% of post-operative transfusions were due to failure of a patient to adequately diurese. Our efforts were associated with a significant decrease in the rate of both intra- and post-operative transfusions (Figures 4a and 4b) not only in Phase 2, but also Phase 3, suggesting that these results were not attributed to a Hawthorne Effect.

Page 23

Page 24

Page 25

25 References 1. Nelson EC, Batalden PB, Ryer JC, Joint Commission on Accreditation of Healthcare Organizations. Clinical improvement action guide. Oakbrook Terrace, IL: Joint Commission on Accreditation of Healthcare Organizations; 1998. 2. Likosky DS. A primer on reviewing and synthesizing evidence. The Journal of extra-corporeal technology. Jun 2006;38(2):112-115. 3. Likosky DS. A primer on randomized controlled trials. The Journal of extra-corporeal technology. Mar 2006;38(1):10-13. 4. Likosky DS. Forming a research question from a multi-center database. The Journal of extra-corporeal technology. Mar 2009;41(1):P33-36. 5. Mulley AG, Jr. The Global Role of Health Care Delivery Science: Learning from Variation to Build Health Systems that Avoid Waste and Harm. Journal of general internal medicine. Jun 25 2013. 6. Likosky DS, Nugent WC, Ross CS, Northern New England Cardiovascular Disease Study G. Improving outcomes of cardiac surgery through cooperative efforts: the northern new England experience. Seminars in cardiothoracic and vascular anesthesia. Jun 2005;9(2):119-121. 7. Surgenor SD, DeFoe GR, Fillinger MP, et al. Intraoperative red blood cell transfusion during coronary artery bypass graft surgery increases the risk of postoperative low-output heart failure. Circulation. Jul 4 2006;114(1 Suppl):I43-48. 8. Likosky DS, Surgenor SD, Dacey LJ, et al. Rationalising the treatment of anaemia in cardiac surgery: short and mid-term results from a local quality improvement initiative. Quality & safety in health care. Oct 2010;19(5):392-398.

Page 26

Page 27

27 SESSION 2 Sunday 18:00 - 19:00 THE PROFESSOR MERRY LECTURE The Third Man Paul S. Myles, MB.BS, MPH, MD, FCARCSI, FANZCA, FRCA, Australia Director of Anaesthesia and Perioperative Medicine, Alfred Hospital, Monash University, Melbourne, Australia Graham Greene’s classic novel (also made into a film) deals with loyalty and love, and perhaps most of all, mistaken identity. The central plot revolves around who is the third man, the missing piece of the puzzle…Einstein reasoned there was something missing from quantum physics. He concluded there must be hidden variables, and that the quantum theories would not be complete until those hidden variables were found… Anaesthetists, surgeons and perfusionists have a longer life expectancy than the average population. Why? Pulmonary artery catheters lead to excess mortality in critically ill patients – is it time to pull the PA catheter? Aprotinin was withdrawn from the market in 2008, but has recently been reinstated in Europe and Canada – why? Nitrous oxide has been found to markedly reduce the odds of death at 30 days after surgery. Is this true? What about medical research? Who or what is the third man? References 1. Greene G. The Third Man. Penguin. 1949 2. Datta M. You cannot exclude the explanation you have not considered. Lancet 1993; 342:345-7. 3. Dalen JE, Bone R. Is it time to pull the pulmonary artery catheter? JAMA 1996; 276:916-8. 4. McMullan V, Alston RP. Aprotinin and cardiac surgery: a sorry tale of evidence misused. Br J Anaesth 2013;110:675-8. 5. Turan A, Mascha EJ, You J, et al. The association between nitrous oxide and postoperative mortality and morbidity after noncardiac surgery Anesth Analg 2013; 116:1026-33. 6. Kopyeva T, Sessler D, Weiss S, et al. Effects of volatile anesthetic choice on hospital length-of-stay: a retrospective study and a prospective trial. Anesthesiology 2013; 119:61-70.

Page 28

Page 29

Page 30

Page 31

Page 32

Page 33

33 SESSION 4 Monday 11:30 - 13:00 Teamwork Clinical Microsystems: A Critical Framework for Crossing the Quality Chasm Donald S. Likosky, PhD, USA Associate Professor Section Head, Section of Health Services Research and Quality Department of Cardiac Surgery Center for Healthcare Outcomes and Policy (CHOP) University of Michigan Medical School Ann Arbor, Michigan Overview Large-scale improvements have been realised in the outcomes for cardiac surgery over the last six decades, driven in part by both technological advances and changes in processes of care. Nonetheless, current care is less than ideal, whether attributed to gaps in our knowledge base, poor clinical decision-making, medication errors, unsafe transitions of care, or ineffective teamwork. Many of us recognise these current shortcomings in our own clinical practices, and often create patchwork-like fixes or workarounds to protect our patients from unintended harm. Some of the most common of these gaps in idealised practices are attributed to the way our clinical teams are organised and how they function and relate within our larger healthcare organisations. In its Crossing the Quality Chasm report (1), the Institute of Medicine (IOM) identified the need to address these deficiencies, in part by optimising the way small clinical teams, “microsystems”, function. The IOM recognised that those most suited to help transform the healthcare delivery system are indeed us, healthcare professionals. Using a case example from a regional quality improvement project, this article will inform the reader about clinical microsystems, and how they may be leveraged to support improvement in the delivery of care. In particular, this article will describe the key facets for leveraging the value of microsystems, including methods for engaging teams in the process of clinical redesign. Description Change is inevitable in any organisation, although improvement is not. However, we may increase the likelihood that the changes we make will result in improvements in care. First things first: definitions (Figure 1) (2).

Page 34

34 Figure 1. Micro-, Meso-, and Macrosystems Clinical Microsystem: A healthcare clinical microsystem can be defined as a small group of professionals who work together on a regular basis, or as needed, to provide care to discrete populations of patients. It has clinical and business aims, linked processes of care, a shared information environment, and produces services and care which may be measured and leveraged as performance outcomes. These systems evolve over time and are (often) embedded in larger systems or organisations. These systems and organisations are called “mesosystems” and “macrosystems”, respectively. Mesosystem: Links microsystems together to allow them to move from disparate units to those that support patients along their continuum of care. Macrosystem: The container that holds meso- and microsystems. “Microsystems thinking” has evolved over time, although it was derived from statistician and consultant W. Edwards Deming (3) and business school professor James Brian Quinn (4). Dr. Deming taught us that systems by their nature must have an aim, and their subcomponents must work synergistically to achieve the overarching aim (3). Dr. Quinn observed that top performing (in terms of profit, quality, and customer service) Fortune 500 companies were successful due to their focus on the smallest replicable units of their business (4). These “best in class” companies achieve their performance targets by empowering frontline teams with knowledge and understanding of their system and its interdependencies within the larger organisation. The corporate management recognises that their workers are the linkage between the organisation and the customer. As such, they are best suited to redesign workflow to meet the customer’s ever-changing needs. Indeed, these workers also recognise the inherent connections between their own work and their company’s other frontline workers.

Page 35

35 Drs. Paul Batalden and Eugene Nelson, professors at Dartmouth College, were pioneers in the translation of Deming and Quinn’s work to the healthcare sector (5). These investigators envisioned that clinical units, such as the operation room, are also microsystems (defined in this context as clinical microsystems) (Figure 1). The operating room team, while not traditionally labelling itself in this manner, works together based on shared data streams (e.g. hemodynamic monitors) to serve the patient’s needs given a shared business aim (e.g. to repair the patient’s mitral regurgitation). Using this framework, we might similarly define other clinical microsystems within a hospital setting, including the cardiac intensive care unit. Importantly, the operating room team transfers the patient to the intensive care unit for post-operative recovery. Many of us recognise that these handoffs are often less than ideal. Why might this be? While multi-factorial, certainly one culprit is the less-than-ideal transfer of knowledge concerning the patient’s operative course (e.g. problems sustaining adequate blood pressure during the bypass period) during the line placement. This connectivity between the operating room and intensive care unit microsystems is managed or controlled by what we might call a mesosystem. You could now imagine that there are numerable micro- and mesosystems contained within any healthcare organisation. Oversight and coordination of these mesosystems is conducted by a macrosystem, which is usually thought of as a chief operating officer, board of directors, etc. Case Example In 2002, through a grant from the Agency for Healthcare Research and Quality, a multi-disciplinary group of clinicians and quality improvement experts in northern New England embarked on an effort to evaluate the impact of operative practices on mechanisms of brain injury after cardiac surgery. The overarching goals were to: (I) document the association between processes of care and mechanisms of brain injury, and (II) redesign practices to reduce their incidence. To do so, investigators enrolled patients into a non-invasive neuro and systemic monitoring study (6). Subjects consented to be monitored continuously throughout their cardiac surgical procedure to measure: (a) embolisation and oxyhemoglobin desaturation in the brain, (b) embolisation leaving and traveling to the patients through the heart lung machine, and (c) hemodynamics. The microsystem that we focused on was the operating room team, composed of a cardiothoracic surgeon, physician assistant, anaesthesiologist, perfusionist, and scrub nurse. Team members of this microsystem met monthly and were supported by a quality improvement expert and cardiovascular epidemiologist. The study was broken into four phases: Phase I: understanding our microsystem and gathering baseline data concerning mechanisms producing neurologic injury; Phase II: developing a multi-disciplinary quality improvement team (which met monthly) and begin using the intra-operatively collected physiological parameters to make small tests of change; Phase III: using operative data to enhance the operative debriefing period; and Phase IV: making changes to the heart-lung machine to reduce emboli traveling back to the patient. Phases II–IV are particularly pertinent to the theme of this manuscript. Between Phases II–IV, the team discussed the data that was collected during the operative period and identified opportunities to use this context knowledge (along

Page 36

36 with generalisable knowledge from the literature) to make targeted quality improvement interventions (Figure 2, Table 1). For instance, our team noted during Phase I that periods of increased vacuum venous return were associated with emboli in the inflow of the heart-lung machine. These findings were in concert with a prior report from Willcox and colleagues (7). Based on this context and generalisable knowledge, our team reduced the amount of vacuum during procedures. Additional changes occurred in Phases III and IV, including using rich contextual data to inform the operative debriefing period and strategic changes in the use of oxygenators and pumps, respectively. In all cases, our team shared generalisable and context knowledge to make cogent and sound arguments to support suggested changes to our operative practices. We monitored before and after each of the interventions to determine whether the changes resulted in improvements in the quality of care. Over the course of the study, our changes resulted in 87.9% reduction in median microemboli in the outflow of the heart-lung machine, and a 77.2% reduction in microemboli detected in the brain, both p<0.001 (6). Figure 2. Formula for Quality Improvement Table 1. Examples of Practice Changes Microsystem Team Member Activity Relevant Literature Perfusionist Reduced use and amount of augmented vacuum Willcox (7); Rider (10) Surgeon Single clamp Hammon (11) Anaesthesiologist Use of ultrasound to guide imaging of aortic disease Djaiani (12)

Page 37

37 We recognised that our neuromonitoring study had implications beyond the intra-operative period. We hypothesised that microemboli leaving the heart-lung machine would be associated with increased levels of biomarkers of brain injury, including S100β. Prior work by Wandschneider and others has found higher S100β associated with CABG surgery using a heart-lung machine versus those conducted off-pump (8). With this generalisable knowledge as an underpinning, we communicated (vis-a-vis our mesosystem) with the pre- and post-operative microsystems to engage in a prospective study of our neuro-monitored patients. Our colleagues drew serum on 71 patients prior to and 48 hours after surgery. We found a significant increase in terciles of post-operative S100β associated with terciles of microemboli in the outflow of the heart-lung machine (9). These findings were shared broadly throughout Maine Medical Center and our regional quality collaborative via presentations at multi-disciplinary grand rounds and webinars. Teams require attention to maintain their interest and effectiveness. While our neuromonitoring team enjoyed great success in reducing the frequency of emboli secondary to the redesign of the heart-lung machine, we missed opportunities to turn the next chapter in the story. Instead of creatively redesigning the focus of the project and its constituents, the operating room microsystem’s attention was diverted to other important institutional projects and initiatives. While one could interpret this as a failure of the project and its leadership to retain its relevance, members of this microsystem directed their attention to other strategically important projects. Discussion Fancy terminology often impedes acceptance of useful frameworks. As one who seeks to engage clinical teams in improvement work, I admit that at times I often purposefully resist using much of the terminology embedded within this manuscript. I sense that the use of esoteric terminology is not an effective way of engaging clinicians. Nonetheless, I feel strongly that the conceptual framework derived from Drs. Deming and Quinn is extremely useful, as it provides a clinical and business case supporting the need to focus and empower frontline teams with sound data. Indeed, performance measures should reflect how these teams interact, rather than promulgate the out-dated notion of solely measuring individual performance. The challenge for each of us is to recognise and now act on the fact that we are leaders in this new era of increased accountability. We are uniquely capable and situated to make changes in the delivery of care to positively impact our patients’ outcomes. We do this not by creating more workarounds, but by effectively interacting with our other microsystem team members. I now challenge you to identify the clinical microsystem you work in, and leverage it to provide safe and effective clinical care.

Page 38

38 Sources of Funding Dr. Likosky was supported by a grant from the Agency for Healthcare Research and Quality (1K02HS015663-01A1). Sorin Group (Arvada, CO) provided an unrestricted grant to support study-related expenses. Somanetics Corporation (Troy, MI) provided funds to support a research coordinator. This work was partially funded by the Northern New England Cardiovascular Disease Study Group. References 1. Institute of Medicine (U.S.). Committee on Quality of Health Care in America. Crossing the quality chasm: A new health system for the 21st century. Washington, D.C.: National Academy Press; 2001. 2. Nelson EC, Batalden PB, Godfrey MM. Quality by design : A clinical microsystems approach. Lebanon, NH: Center for the Evaluative Clinical Sciences at Dartmouth; Jossey-Bass/Wiley; 2007. 3. Deming WE. The new economics : For industry, government, education. Cambridge, Mass.: MIT Press; 2000. 4. Quinn JB. Intelligent enterprise: A knowledge and service based paradigm for industry. New York: Free Press; Maxwell Macmillan Canada; Maxwell Macmillan International; 1992. 5. Nelson EC, Batalden PB, Huber TP, et al. Microsystems in health care: Part 1. Learning from high-performing front-line clinical units. The Joint Commission journal on quality improvement. 2002;28:472-493. 6. Groom RC, Quinn RD, Lennon P, et al. Detection and elimination of microemboli related to cardiopulmonary bypass. Circ Cardiovasc Qual Outcomes. 2009;2:191-198. 7. Willcox TW, Mitchell SJ, Gorman DF. Venous air in the bypass circuit: A source of arterial line emboli exacerbated by vacuum-assisted drainage. The Annals of thoracic surgery. 1999;68:1285-1289. 8. Wandschneider W, Thalmann M, Trampitsch E, Ziervogel G, Kobinia G. Off-pump coronary bypass operations significantly reduce s100 release: An indicator for less cerebral damage? The Annals of thoracic surgery. 2000;70:1577-1579. 9. Groom RC, Quinn RD, Lennon P, et al. Microemboli from cardiopulmonary bypass are associated with a serum marker of brain injury. The Journal of extra-corporeal technology. 2010;42:40-44. 10. Rider SP, Simon LV, Rice BJ, Poulton CC. Assisted venous drainage, venous air, and gaseous microemboli transmission into the arterial line: An in-vitro study. The Journal of extra-corporeal technology. 1998;30:160-165. 11. Hammon JW, Stump DA, Butterworth JF, et al. Single crossclamp improves 6-month cognitive outcome in high-risk coronary bypass patients: The effect of reduced aortic manipulation. J Thorac Cardiovasc Surg. 2006;131:114-121. 12. Djaiani G, Ali M, Borger MA, et al. Epiaortic scanning modifies planned intraoperative surgical management but not cerebral embolic load during coronary artery bypass surgery. Anesthesia and analgesia. 2008;106:1611-1618.

Page 39

39 Teamwork Teamwork, Communication, Formula-One Racing and the Outcomes of Cardiac Surgery Alan F. Merry, FANZCA, New Zealand Professor and Head of the School of Medicine, Auckland City Hospital, Auckland, New Zealand Authors: Alan F. Merry, Jennifer Weller and Simon J. Mitchell School of Medicine, Auckland City Hospital, Auckland, New Zealand Abstract Most cardiac units achieve excellent results today, but the risk of cardiac surgery is still relatively high, and avoidable harm is common. The story of the Green Lane Cardiothoracic Unit provides an exemplar of excellence, but also illustrates the challenges associated with changes over time and with increases in the size of a unit and the complexity of practice today. The ultimate aim of cardiac surgery should be the best outcomes for (often very sick) patients, rather than an undue focus on the prevention of error or adverse events. Measurement is fundamental to improving quality in healthcare, and the framework of structure, process and outcome is helpful in considering how best to achieve this. A combination of outcomes (including some indicators of important morbidity) with key measures of process is advocated. There is substantial evidence that failures in teamwork and communication contribute to inefficiency and avoidable harm in cardiac surgery. Minor events are as important as major ones. Six approaches to improving teamwork (and hence outcomes) in cardiac surgery are suggested. These are: 1) subspecialize and replace tribes with teams; 2) sort out the leadership while flattening the gradients of authority; 3) introduce explicit training in effective communication; 4) use checklists, briefings and debriefings… and engage in the process; 5) promote a culture of respect alongside a commitment to excellence and a focus on patients; 6) focus on the performance of the team… not on individuals.

Page 40

40 Introduction – an historical perspective The Green Lane Cardiothoracic Unit (CTSU) was established by Brian Barratt-Boyes, a charismatic and demanding leader, knighted in 1971 and (with John Kirklin) co-author of the seminal text book “Cardiac Surgery” in 1986. The CTSU was built on a small, highly sub-specialised team with a commitment to consultant-led practice at all hours and a culture of excellence. Great emphasis was placed on research, evidence and outcomes, and the unit’s results were closely monitored and regularly published.1 For many years the unit undertook vascular surgery as well as cardiac and thoracic surgery, and paediatric cardiac surgery was integrated into the adult unit. The unit developed a pre-eminent national and international reputation in the late 1960s and the 1970s through its pioneering work in the use of homografts for aortic valve surgery and deep hypothermic arrest for the repair of congenital heart defects, and through its world-leading outcomes. In the early days the team was small and every anaesthesiologist, surgeon and perfusionist participated in most aspects of the work, including the postoperative intensive care of patients. They were supported by cardiologists, nurses, technicians and others who worked exclusively within the unit, or at least in close association with it. People worked very long hours and the unit’s “lore” includes many stories of missed attendances at important family occasions (even childbirths), reflecting a strong commitment to the team as well as to patients. Gradually progress has wrought change. Change has been accentuated by a physical relocation from a relatively small, somewhat specialist hospital to a much larger institution. Paediatric cardiac surgery is now provided in a separate part of the hospital complex by a substantially separate group of practitioners. Intensive care is provided by a group of doctors who, formally or informally, have substantially specialised in intensive care. There is still some overlap between these intensivists and the cardiac anaesthesiologists, with some individuals practicing as both, but the overlap is incomplete, and there is now a separate department of cardiac intensive care. Thus there has been an evolution from a relatively small and tightly knit team to a much larger group, still loosely identifiable as a team, but with more dispersed leadership and a greater sense of division into separate “tribes”. In line with the approach widely adopted in New Zealand, some surgeons, perfusionists and anaesthesiologists within the unit have always conducted private practice in a nearby institution, in parallel to their (part time) public hospital commitments. This has tended to result in very tight teams within the private setting but has also had some influence (probably good and bad) on the culture of the public hospital. Outcomes have steadily improved (as they have in most units around the world) and are still excellent by international standards, but claims to world-leading results would be harder to sustain now than in earlier days. Teamwork is self-evidently essential for successful cardiac surgery. In the early days of the Green Lane CTSU, effective teamwork flowed from the many hours spent together by a relatively small group of practitioners who knew each other well and understood each other’s mutual expertise, skills and expectations. Today, with a larger group and more disparate roles, there is a place for considering explicit initiatives to promote successful teamwork and communication. Institutions differ in their approaches and case mix, but the underlying principles are similar everywhere. In the first part of this paper we will discuss adverse events in healthcare and summarise the evidence supporting the concept that teamwork and communication

Page 41

41 influence outcomes. Initiatives to improve performance should be accompanied by measurement and we will consider how this might be done in the context of cardiac surgery. In the second part of the paper we will outline the elements of teamwork, and outline six approaches to improving teamwork and communication in cardiac surgery. Part 1: Teamwork, Communication, and the Outcomes of Cardiac Surgery Error and preventable harm in healthcare Today it is widely accepted that too many patients suffer harm as an unintended consequence of their healthcare2-8. Much of this harm is avoidable, and attributable to error. Cardiothoracic surgery is no exception: it is self-evidently associated with substantial risk 9, and in a study involving retrospective review of charts in hospitals in Colorado and Utah, the rate of preventable adverse events was higher than average for coronary artery bypass graft and cardiac valve surgery8. Things may go wrong at any point in a patient’s pathway from primary care through the wards, operating rooms, intensive care, and back again. In the UK, a report from the National Reporting and Learning System database showed that of 4828 incidents involving cardiac, 21% occurred in the operating room (OR) and 79% outside the OR. Harm resulted in 23% of the OR and 34% of the non-OR incidents10. It is all too easy to develop an excessive focus on the reduction of error, or on the Hippocratic mantra “primum non nocere” (first, do no harm). The drawback of this is that most healthcare services today have limited resources and heavy loads of very sick patients requiring treatment, often urgently. Doing nothing, or investing excessively in safety at the expense of productivity, may mean fewer patients are harmed by healthcare, but it will often also mean that some die untreated while others remain too sick to enjoy a worthwhile quality of life. Quality in healthcare depends not only on safety, but also timeliness, efficiency, efficacy, equitability and patient-centeredness (represented by the acronym STEEEP) 11. The ultimate objective should not be to avoid error, or harm at all costs, but rather to achieve the best outcomes for the greatest number of sick patients. For the avoidance of doubt, a reasonable emphasis on avoiding error and reducing harm is of course a necessary part of achieving this objective, but, in the context of allocating limited resources, there is likely to be an optimal level of investment in error prevention beyond which the net effect on outcome may be negative. It does seem that most healthcare systems today invest relatively lightly in safety initiatives, so this point may be moot in contemporary practice. The Institute of Healthcare Improvement (IHI) has introduced the concept of a “Triple Aim” 12 (New Zealand has adopted a modification of this emphasizing value for the available resource rather than cost reduction per se 13). The Triple Aim involves the simultaneous pursuit of three objectives: • Improving the health of the population; • Enhancing the patient experience of care (including quality, access and reliability); and • Reducing, or at least controlling, the per capita cost of care. This expresses at the highest level the objectives of any healthcare system, and by implication of any part of such a system.

Page 42

42 Quality improvement and the importance of measurement Drawing from the experience of advances in anaesthesia safety, Lucian Leape has asserted that undue insistence on evidence and measurement can be counter-productive, and that there are some things that should be done simply because they make sense14. Nevertheless, there is a strong belief that measurement is essential for effective improvement in healthcare15 16. Donabedian 15 has outlined a simple framework for the measurement of quality in healthcare that has become widely accepted. The framework emphasizes three domains:  structure;  process; and  outcome Structure is relatively easy to measure, but there is an obvious attraction in focusing on outcome. For many interventions, and in many fields of healthcare, it turns out that this can be very expensive and perhaps impractical. For example, medication error features prominently in almost every study of harm arising from healthcare, and cardiac anaesthesia and perfusion involve the administration of large numbers of drugs, so this seems likely to be a fruitful area for improvement. However, finding a reliable way to count and quantify adverse drug events (ADEs) is not easy. Identifying drug administration or prescription errors is even more difficult. Trigger tools as a means of monitoring outcome In the early 1990s Classen et al described a computerized method of using triggers, or flags, to identify patients in whom an ADE was likely; a subsequent review of the relevant charts could then establish whether or not an ADE had actually occurred17. This proved much more efficient than traditional approaches based on the review of large numbers of randomly selected charts 18 19 but required appropriate electronic record systems. The Institute of Healthcare Improvement subsequently developed an adverse drug event trigger tool that could be used on the basis of a manual inspection of relatively small samples of charts20. For example, the use of naloxone suggests the possibility of an opioid overdose, and the use of vitamin k suggests the possibility of a warfarin overdose. A key point about the trigger tool approach is that no effort is expended on the question of preventability or error. The aim is to reduce harm, and the assumption is that an overall reduction of harm will reflect a reduction in those events that are preventable. The ADE trigger tool has been adopted quite widely, but still requires considerable resource and has some limitations. Other trigger tools, including a global trigger tool, are also available. Mortality statistics Cardiac surgery is unusual in that even routine cases are associated with relatively high rates of mortality. This has led to a well-established culture of monitoring and reporting mortality statistics, often corrected for known risk factors. For example, Kang et al have described an application of this approach to quality assurance in congenital heart surgery21. The Society for Cardiothoracic Surgery in Great Britain & Ireland (SCTS: www.scts.org) reports mortality data for indicator cardiac procedures in the public

Page 43

43 domain. These data are supplemented with information on case volumes and some indication of casemix. National averages are provided, and funnel plots are used to indicate acceptable limits of performance. Some hospitals now provide links on their websites to this sort of information for individual surgeons (see for example http://www.uhsm.nhs.uk/news/Pages/transp.aspx). This approach is, in our opinion, too narrow and places too much emphasis on individual practitioners from one subset of the overall cardiac team. We will return to this point in the second part of the paper. Why process should be measured as well as outcome In New Zealand, the Health Quality and Safety Commission has introduced quality and safety markers for three important aspects of quality in healthcare, falls, hospital associated infections and surgical harm. A marker for medication safety is expected later this year. All of these areas are of relevance to cardiac surgery. Each marker contains two parts – a measure of process and an indicator measure of outcome. For example one marker addresses handwashing. The process marker is the percentage of practitioners that comply with all of the World Health Organisation’s five moments of hand hygiene22. The outcome measure is the incidence of staphylococcus aureus bacteremia. This approach is predicated on the idea that it is more reasonable to set targets for process indicators than for outcome indicators. There is no particular reason why one hospital should find it more difficult than another to achieve a high level of compliance with a process. On the other hand, many factors contribute to outcomes. Nevertheless, it pays to include some measure of outcome for two reasons:  to give a clear message about the primary purpose of focusing on the particular process; and  to reduce the likelihood of gaming. The use of an indicator of outcome is simpler and less costly than trying to measure all relevant components of outcome. Thus for falls the outcome measure is the number of patients suffering a hip fracture from a fall while in hospital. Obviously other forms of injury also occur - fractures of arms, skulls and so forth. However, hip fractures are common and give a good indication of the overall picture. In cardiac surgery, the Society of Thoracic Surgeons (STS) in the US has adopted a broader approach than the SCTS. A three-star system (http://www.sts.org) is used to rate participating groups or institutions on a combination of outcome and process measures for certain indicator procedures. The outcome measures include mortality, reoperation, stroke, kidney failure, infection of the chest wound and prolonged ventilation. Process measures include the use of at least one internal mammary artery and the perioperative prescription of certain important medications. This approach has much to commend it, and also provides a sensible approach to monitoring efforts to improve safety. The outcome measures provide a good indication of the things that are likely to matter to patients. Reducing error and improving safety are one necessary element of achieving excellent outcomes, so an emphasis on these seems to me to be sufficient in everyday practice. There is of course good reason to monitor or investigate specific aspects of practice as well, and in this regard trigger tools and the quality and safety markers also have a role to play. The balance lies in finding a reasonable mix of measures to provide adequate

Page 44

44 information without becoming too onerous or expensive. The focus on optimising outcomes for the available resource should not be forgotten. Evidence that teamwork and communication matter in cardiac surgery Any unit achieving acceptable results for cardiac and thoracic surgery must already have skilled staff and reasonable teamwork. Nevertheless, there may be room for improvement. In the context of healthcare generally, many studies have identified failures in teamwork and communication as common factors in the genesis of adverse events in healthcare23-32 . Several studies have produced similar findings in the context of managing patients with cardiac disease33. A key theme that emerges, particularly from the work of de Leval et al, is that minor events, taken collectively, are as important as major ones 34 35 36. The, Flawless Operative Cardiovascular Unified Systems (FOCUS) project is an important initiative being undertaken by the Society of Cardiovascular Anaesthesiologists, The aim is harm-free cardiac surgery. Early findings emphasises the importance of a culture in which staff feel able to speak up if they see something that might be going wrong, and of teamwork and communication 37 38. There is also evidence that teamwork and communication in healthcare can be improved 39. Improving teamwork is an explicit objective of the World Health Organization (WHO) Surgical Safety Checklist (the Checklist) 40 41 42. Variation and teamwork Variation in healthcare should be predicated on differences in patients, but instead it often reflects differences in the belief and culture of different institutions or individual practioners43-45. A key point about variation in practice within an institution is that it makes teamwork more difficult – standardisation of practice has been pivotal in improving the safety of aviation and other high reliability enterprises, and it is well overdue in healthcare. This presentation is concluded on Tuesday 3rd September 2013 Session 12 17:10 – 18:00 (see page 157)

Page 45

45 References 1. Merry AF. Safer cardiac surgery. J. Extra. Corpor. Technol. 2009;41:P43-7. 2. Kohn LT, Corrigan JM, Donaldson MS, editors. To err is human: building a safer health system. Washington DC: National Academy Press, 1999. 3. Wilson R. Clinical Preceptor Conferences as a Venue for Total Quality Education. Optometric Education 1996;21:85-89. 4. Davis P, Lay-Yee R, Briant R, Ali W, Scott A, Schug S. Adverse events in New Zealand public hospitals I: occurrence and impact. N. Z. Med. J. 2002;115:U271. 5. Baker GR, Norton PG, Flintoft V, Blais R, Brown A, Cox J, Etchells E, Ghali WA, Hebert P, Majumdar SR, O'Beirne M, Palacios-Derflingher L, Reid RJ, Sheps S, Tamblyn R. The Canadian Adverse Events Study: the incidence of adverse events among hospital patients in Canada. CMAJ Canadian Medical Association Journal 2004;170:1678-86. 6. Brennan TA, Leape LL, Laird NM, Hebert L, Localio AR, Lawthers AG, Newhouse JP, Weiler PC, Hiatt HH, Harvard Medical Practice Study I. Incidence of adverse events and negligence in hospitalized patients: results of the Harvard Medical Practice Study I. 1991. Qual Saf Health Care 2004;13:145-51; discussion 51-2. 7. Vincent C, Neale G, Woloshynowych M. Adverse events in British hospitals: preliminary retrospective record review. Br. Med. J. 2001;322:517-19. 8. Gawande AA, Thomas EJ, Zinner MJ, Brennan TA. The incidence and nature of surgical adverse events in Colorado and Utah in 1992. Surgery 1999;126:66-75. 9. Shahian DM, O'Brien SM, Sheng S, Grover FL, Mayer JE, Jacobs JP, Weiss JM, Delong ER, Peterson ED, Weintraub WS, Grau-Sepulveda MV, Klein LW, Shaw RE, Garratt KN, Moussa ID, Shewan CM, Dangas GD, Edwards FH. Predictors of long-term survival after coronary artery bypass grafting surgery: results from the Society of Thoracic Surgeons Adult Cardiac Surgery Database (the ASCERT study). Circulation 2012;125:1491-500. 10. Martinez EA, Shore A, Colantuoni E, Herzer K, Thompson DA, Gurses AP, Marsteller JA, Bauer L, Goeschel CA, Cleary K, Pronovost PJ, Pham JC. Cardiac surgery errors: results from the UK National Reporting and Learning System. Int. J. Qual. Health Care 2011;23:151-8. 11. Institute of Medicine. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academy Press, 2001. 12. Berwick DM, Nolan TW, Whittington J. The triple aim: care, health, and cost. Health Aff. (Millwood). 2008;27:759-69. 13. The Triple Aim: Health Quality & Safety Commission, 2011. 14. Leape LL, Berwick DM, Bates DW. What practices will most improve safety? Evidence-based medicine meets patient safety. JAMA 2002;288:501-7. 15. Donabedian A. An Introduction to Quality Assurance in Health Care. New York: Oxford University Press, 2003. 16. Shekelle PG, Pronovost PJ, Wachter RM, Taylor SL, Dy SM, Foy R, Hempel S, McDonald KM, Ovretveit J, Rubenstein LV, Adams AS, Angood PB, Bates DW, Bickman L, Carayon P, Donaldson L, Duan N, Farley DO, Greenhalgh T, Haughom J, Lake ET, Lilford R, Lohr KN, Meyer GS, Miller MR, Neuhauser DV, Ryan G, Saint S, Shojania KG, Shortell SM, Stevens DP, Walshe K. Advancing the science of patient safety. Ann Intern Med 2011;154:693-6. 17. Classen DC, Pestotnik SL, Evans RS, Burke JP. Computerized surveillance of adverse drug events in hospital patients.[Erratum appears in JAMA 1992 Apr 8;267(14):1922]. JAMA 1991;266:2847-51. 18. Brennan TA, Leape LL, Laird NM, Hebert L, Localio AR, Lawthers AG, Newhouse JP, Weiler PC, Hiatt HH. Incidence of adverse events and negligence in hospitalized patients - results of the Harvard Medical Practice Study I. N Engl J Med 1991;324:370-76. 19. Leape LL, Brennan TA, Laird N, Lawthers AG, Localio AR, Barnes BA, Hebert L, Newhouse JP, Weiler PC, Hiatt H. The nature of adverse events in hospitalized patients. Results of the Harvard Medical Practice Study II. N Engl J Med 1991;324:377-84. 20. Rozich JD, Haraden CR, Resar RK. Adverse drug event trigger tool: a practical methodology for measuring medication related harm. Qual Saf Health Care 2003;12:194-200. 21. ang N, Tsang VT, Gallivan S, Sherlaw-Johnson C, Cole TJ, Elliott MJ, de Leval MR. Quality assurance in congenital heart surgery. Eur. J. Cardiothorac. Surg. 2006;29:693-7; discussion 97-8. 22. Chou DT, Achan P, Ramachandran M. The World Health Organization '5 moments of hand hygiene': the scientific foundation. Journal of Bone & Joint Surgery - British Volume 2012;94:441-5. 23. Bognor M. Human Error In Medicine. 1st ed. New Jersey: Lawrence Erlbaum Association Inc, 1994. 24. Threat and error in aviation and medicine: Similar and different. Special Medical Seminar, Lessons for Health Care: Applied Human Factors Research; 2000 January 2001; NSW. Australian Council of Safety and Quality in Health Care & NSW Ministerial Council for Quality in Health Care. 25. Reader TW, Flin R, Cuthbertson BH. Communication skills and error in the intensive care unit. Curr Opin Crit Care 2007;13:732-6. 26. Reason J. Human Error. New York: Cambridge University Press, 1990. 27. Manser T, Harrison TK, Gaba DM, Howard SK. Coordination patterns related to high clinical performance in a simulated anesthetic crisis. Anesth. Analg. 2009;108:1606-15. 28. Webb RK, Currie M, Morgan CA, Williamson JA, Mackay P, Russell WJ, Runciman WB. The Australian Incident Monitoring Study: An analysis of 2000 incident reports. Anaesth Intensive Care 1993;21:520-28. 29. Kunzle B, Kolbe M, Gudela G. Ensuring patient safety through effective leadership behaviour: A literature review. Safety Science 2010;48:1-17. 30. Mazzocco K, Petitti DB, Fong KT, Another, Another, Another, Another. Surgical team behaviours and patient outcomes. Am. J. Surg. 2009;197:678-85. 31. Symons NRA, Wong HWL, Manser T, Sevdalis N, Vincent CA, Moorthy K. An observational study of teamwork skills in shift handover. Int J Surg 2012;10:355-9. 32. Lingard L, Espin S, Whyte S, Regehr G, Baker GR, Reznick R, Bohnen J, Orser B, Doran D, Grober E. Communication failures in the operating room: an observational classification of recurrent types and effects. Quality and Safety in Health Care 2004;13:330-34. 33. Wiegmann DA, ElBardissi AW, Dearani JA, Daly RC, Sundt TM, 3rd. Disruptions in surgical flow and their relationship to surgical errors: an exploratory investigation. Surgery 2007;142:658-65. 34. de Leval MR, Carthey J, Wright DJ, Farewell VT, Reason JT. Human factors and cardiac surgery: a multicenter study. J. Thorac. Cardiovasc. Surg. 2000;119:661-72.

Page 46

46 35. Solis-Trapala IL, Carthey J, Farewell VT, de Leval MR. Dynamic modelling in a study of surgical error management. Stat. Med. 2007;26:5189-202. 36. Mishra A, Catchpole K, McCulloch P. The Oxford NOTECHS System: reliability and validity of a tool for measuring teamwork behaviour in the operating theatre. Quality & Safety in Health Care 2009;18:104-08. 37. Martinez EA, Thompson DA, Errett NA, Kim GR, Bauer L, Lubomski LH, Gurses AP, Marsteller JA, Mohit B, Goeschel CA, Pronovost PJ. High stakes and high risk: a focused qualitative review of hazards during cardiac surgery. Anesth. Analg. 2011;112:1061-74. 38. Gurses AP, Kim G, Martinez EA, Marsteller J, Bauer L, Lubomski LH, Pronovost PJ, Thompson D. Identifying and categorising patient safety hazards in cardiovascular operating rooms using an interdisciplinary approach: a multisite study. BMJ Qual Saf 2012;21:810-8. 39. Salas E, DiazGranados D, Klein C, Burke CS, Stagl KC, Goodwin GF, Halpin SM. Does Team Training Improve Team Performance? A Meta-Analysis Human Factors: The Journal of the Human Factors and Ergonomics Society 2008;50:903-33. 40. Haynes A, Weiser T, Berry W, Lipsitz S, Breizat A, Dellinger E, al e. A surgical safety checklist to reduce morbidity and mortality in a global population. New England Journal of Medicine 2009;360:491-9. 41. Neily J, Mills PD, Young-Xu Y, Carney BT, West P, Berger DH, Mazzia LM, Paull DE, Bagian JP. Association between implementation of a medical team training program and surgical mortality. JAMA 2010;304:1693-700. 42. de Vries EN, Prins HA, Crolla RMPH, den Outer AJ, van Andel G, van Helden SH, Schlack WS, van Putten MA, Gouma DJ, Dijkgraaf MGW, Smorenburg SM, Boermeester MA, Group SC. Effect of a comprehensive surgical safety system on patient outcomes. N Engl J Med 2010;363:1928-37. 43. Van Brabandt H, Neyt M, Hulstaert F. Transcatheter aortic valve implantation (TAVI): risky and costly. BMJ 2012;345:e4710. 44. Hannan EL, Cozzens K, Samadashvili Z, Walford G, Jacobs AK, Holmes DR, Jr., Stamato NJ, Sharma S, Venditti FJ, Fergus I, King SB, 3rd. Appropriateness of coronary revascularization for patients without acute coronary syndromes. J. Am. Coll. Cardiol. 2012;59:1870-6. 45. Birkmeyer JD, Sharp SM, Finlayson SR, Fisher ES, Wennberg JE. Variation profiles of common surgical procedures. Surgery 1998;124:917-23.

Page 47

47 SESSION 5 Monday 14:00 - 15:30 Blood and Volume Management Patient Blood Management in Adult Cardiac Surgery David A. Scott, MB BS, FANZCA, PhD, FFPMANZCA, Australia Associate Professor and Director of Anaesthesia, St Vincent’s Hospital, Melbourne, Australia, Associate Professor, Faculty of Medicine, Dentistry and Health Sciences, University of Melbourne, Melbourne, Australia Introduction The proportion of patients undergoing cardiac surgery who receive blood or blood products varies from centre to centre but ranges from 10% to 40% in elective coronary artery surgery and up to almost 100% for very complex procedures. The aim of blood management in cardiac surgical patients is to minimise exposure to blood and blood products because of the well documented adverse responses to blood transfusion. These adverse effects include reactions to immunoreactive elements in blood (eg transfusion related lung injury)1 as well as coagulopathy, infection risk (including unknown unknowns) and associations with adverse outcomes overall 2, 3. It is difficult to separate cause from effect in some situations (eg increased mortality with blood exposure4) from the surgical or perioperative events driving the need for transfusion. It should not be forgotten therefore that transfusion can be life-saving. Management of bleeding and blood loss therefore centres on increasing patient reserve, prevention of blood loss and coagulopathy, and careful management of replacement therapy and correction of coagulation abnormalities. Patient reserve in this context includes overall nutrition, iron stores and circulating haemoglobin and clotting factor levels. Preparation also involves identification of patients at increased risk of bleeding (innate, pharmacological) and also those with a genetic predisposition to thrombophilia. In the intraoperative environment, minimising blood loss relates to surgical skill and perfusion technique as much as maintaining an optimal haemostatic environment, which is not always easy. Off-pump coronary artery surgery is associated with lower blood loss overall5, with coagulopathy being less common. Finally, close monitoring of coagulation status and other clinical parameters is needed so that interventions can be proactive rather than reactive. Many centres have developed algorithms to guide the most effective management of bleeding in cardiac surgery, once it occurs 6. Evidence-based recommendations have been updated recently, and provide support for a range of patient blood management strategies to minimise blood loss in surgery and lower transfusion requirements5, 7. These recommendations recognise the risk of preoperative anaemia as a factor in increased blood transfusion, however there is little published data examining opportunities for timely preoperative intervention.

Page 48

48 Preoperative anaemia may be due to many causes, especially in elderly cardiac patients. Although some causes may be difficult to treat in the short term, iron deficiency, either absolute or functional, is very common. Treatment of these patients using newer formulations of intravenous iron (such as iron carboxymaltose) can achieve an increase in red cell mass within a few weeks at relatively low cost and very low risk8, 9. This is just one example of a potential intervention. Our own institution acknowledges the risks of anaemia but to date does not have a structured process to intervene in all but extreme cases. A project was undertaken to investigate the proportion of elective cardiac surgical patients presenting for surgery with anaemia, to identify whether early diagnosis and timely intervention might have been possible and to identify other clinical factors associated with blood transfusion in cardiac surgery. Investigation Methods & Results The data was obtained via a retrospective audit of all elective and sub-acute patients undergoing cardiac surgery at St. Vincent’s hospital in the 6 months from January to June 2012. Patient medical records were accessed to identify first contact or ‘Referral’ being the original cardiology assessment (usually coronary angiography or echocardiography) which generated the referral for cardiac surgery. Emergency patients, requiring surgery out of hours or by taking over the next elective patient session, were excluded. CPB was performed using membrane oxygenation and a centrifugal arterial pump head. ‘Mini’ circuits, when employed, used an optimised open reservoir design to minimise tubing lengths. All pump blood was returned to the patient ‘as is’ following CPB. Antegrade and retrograde tepid blood cardioplegia was used. The data reported is an interim report of 6 months data. The full investigation will include 12 months data. Over the 6 month period 189 cases were available for analysis. There was no significant change in Hb level through the clinical pathway from referral to hospital admission in elective or sub-acute patients. For elective patients, referral occurred a median of 34 days prior to Preadmission clinic attendance, with 81.3% of patients being referred at least 14 days prior to the clinic visit. Following Preadmission clinic, elective patients had surgery a median of 15 days later. Sub-acute cases had a much shorter period between referral and surgery, with a median of 3 days. Using gender-based WHO criteria (above), anaemia was present in 32 (24.1%) of elective and 13 (23.2%) of sub-acute cases (23.8% overall, p=0.90). There was no difference in preoperative days between anaemic and non-anaemic patients. On univariable analysis, preoperative anaemia was associated with increasing age and renal impairment, but on multivariable analysis preoperative anaemia was not significantly associated with gender or age, however it was related to preoperative renal impairment (eGFR) (OR 1.03, 95%CI 1.01-1.04; adjusted p<0.01). Overall, 44.4% of patients received a blood transfusion during their admission. In elective patients the transfusion incidence was 42.9%, and was lowest in elective valve surgery or CABG (37.0% and 38.3% respectively). Approximately 10% of patients were transfused during CPB, 10% in the OR following CPB, with the majority of transfusions (35% of patients) occurring in ICU. Most patients received 3 or fewer units of blood during their admission, with 27% of patients receiving 2 units.

Page 49

49 Pre-surgical factors associated with a need for blood transfusion were: lower BMI (OR 0.86, p<0.001), who had preoperative anaemia (WHO criteria) (OR 3.3, p=0.005) or who had renal impairment (OR 5.6, p<0.001). Intraoperative factors associated with blood transfusion were (OR [95% CI)]): Prime Volume (0.999 [0.998-1.000]; p=0.055); Volume on CPB (1.001 [0.999-1.001], p=0.063); Hb Pre-bypass (0.948 [0.926-0.971]; p=<0.001) and Duration post CPB (1.018 [1.002-1.033]); p=0.026). The nadir Hb on CPB was a strong predictor of transfusion events (No transfusion 95.1 ± 12.9 g/L, any transfusion 80.6 ± 13.8 g/L; p<0.001). Analysis of transfusion during CPB reveals a transfusion threshold of 65 – 70 g/L with 19.6% of patients being transfused at some point during surgery. Intraoperative red cell salvage using a cell saver was selective (ie based on indication) and occurred in 11.6% of cases. This selective use of a cell saver did not affect the need for transfusion overall (p=0.91) nor was it associated with decreased number of units transfused. Postoperatively, the factors univariably associated with blood transfusion were Hb on arrival and nadir, platelet count on arrival and length of stay in ICU. These all remained significantly associated on multivariable analysis with the exception of platelet count on arrival. Comments This study identified that 81.3% of elective cardiac surgical patients are being referred at least 14 days prior to their preoperative clinic visit and 75% have over 21 days prior to undergoing surgery. Approximately one-quarter of these patients are anaemic by WHO criteria. Homologous red blood cell transfusion was given to 44% of patients during or after surgery. We have confirmed the findings of others that there is a strong association between red blood cell transfusion and low preoperative haemoglobin, low BMI (reflecting low red cell mass) and preoperative renal impairment 10. Of these factors, correction of anaemia potentially represents the single most effective intervention that can be undertaken preoperatively to decrease allogeneic transfusion exposure. Red cell transfusion is associated with a number of adverse clinical outcomes, following both cardiac and non-cardiac surgery 2 10. A common criticism of the potential causative role of transfusion in these outcomes is the inability to properly control for the intraoperative and postoperative events that may lead to a need for transfusion. Notwithstanding this, studies suggest that there is a dose-response to transfusion-related morbidity and that exposure to even one unit has detectable adverse effects 3. In this study, 17% of those receiving transfusions received only one unit and 27% received two units of homologous blood. It seems reasonable to assume that a higher baseline red cell mass (eg Hb) would reduce the frequency of these small volume exposures to blood. Anaemia in the elderly is caused by a number of factors. In the United States, the most frequent identifiable causes of anaemia in those aged over 65 years are nutritional deficits (34.3%), with iron deficiency being present in half of these (20% overall). Other causes include chronic inflammation (chronic disease - including diabetes and cardiac failure) (19.7%), and renal insufficiency (12.5%)11. The investigation of iron deficiency has improved over recent years, and the use of Ferritin, Transferrin and Transferrin saturation has improved the sensitivity of detection of deficits even before anaemia is manifest 8. Causes of iron deficiency

Page 50

50 may be functional (eg inflammation or chronic disease, renal impairment) or absolute (eg blood loss (typically gastrointestinal)), dietary insufficiency and increased demand. Although the investigation and management of moderate to severe anaemia requires careful clinical assessment and possible specialist referral, this should not preclude interventions aimed at correcting the anaemia by increasing red cell production. Erythropoiesis may be stimulated by the provision of substrate (eg iron, vitamin B12) or by administration of exogenous erythropoietic stimulating agents (eg EPO – erythropoietin). Although EPO is effective in driving the marrow to produce more red cells, it is expensive and its use can be associated with adverse events such as worsening angina or thrombotic complications8. Iron supplementation on the other hand simply enhances the bodies’ endogenous pathways. Unfortunately oral iron is poorly absorbed, and with 1-2 mg uptake per day may take months to correct a total body deficit of over 1000mg. Fortunately, parenteral iron preparations have improved to the extent that formulations such as ferric carboxymaltose can provide 1000mg of iron via a 15 minute outpatient iv infusion which is well tolerated9. It should also be recognised that iron is an element of many cellular and metabolic components, including mitochondrial energy pathways, and therefor has benefits beyond those of haemopoiesis2. The time between referral, preadmission and surgery provides ample opportunity for diagnosis of anaemia and interventions to increase the patient’s red cell mass, and indeed to cope with a need to generate more red cells following surgical loss by ‘priming’ the marrow. However, at present only 40% of patients have their Hb documented at the time of referral. This represents a tremendous loss of opportunity which should be exploited. There are many factors that can influence blood loss and the need for homologous transfusion in cardiac surgery. A number of practice recommendations have been published to help minimise blood loss and transfusion5-7. This study confirmed that nadir Hb levels during CPB and in ICU were strongly associated with transfusion. However, whilst this measure is important, it is simply a reflection of the events which have led up to that point. Thus, intraoperatively, blood loss prior to and during CPB and the degree of haemodilution (eg CPB prime volume) are relevant factors. Strategies such as autologous priming aid in reducing prime volumes if the patient’s blood is used to displace priming fluid from the circuit. Ongoing bleeding and coagulopathy contribute to the high rates of transfusion in ICU. The optimisation of coagulation is beyond the scope of this article, but the majority of patients in this study were on preoperative aspirin (but not thienopyridenes) and given an antifibrinolytic during surgery. Cell salvage is a specific technique used to conserve autologous red blood cells. Cell salvage has been shown to decrease homologous transfusion requirements in a range of procedures 12, although the cost/benefit has been questioned for routine cardiac surgery 13. It should be remembered that non-cellular components of blood are lost in the washing process and this must be corrected in large volume transfusions. It is generally recommended for procedures where a large volume of blood loss (eg > 1000 mL) is expected5. Our selective use of cell salvage (11%) may reflect under-utilisation of the technique. The high incidence of homologous transfusion in this study (44%), even in elective patients, is likely to be due to a number of factors and suggests opportunities for improvement. These include preoperative care, and intraoperative techniques

Page 51

51 (19.6% of patients had intraoperative transfusion). In particular attention needs to be given to blood management in ICU, where 35% of patients received blood. This study has a number of limitations. Although the databases and records from which the data is drawn are comprehensive, any retrospective analysis will suffer from missing information and potentially low quality data. The transfusion ‘triggers’ were not strictly defined, as they would be for a prospective trial, and this will account for some variability in transfusion recipients. An integrated approach is needed to reduce patient exposure to blood transfusions and no single strategy is likely to have a large effect on its own. This study has identified a high incidence of transfusion in our cardiac surgical population, a high incidence of preoperative anaemia and sufficient time for intervention to improve this. The use of a range of intraoperative blood management strategies is also essential, underpinned by good operative technique. Finally, ICU planning and transfusion triggers need to be decided upon and adhered to as much as is clinically possible. Acknowledgements Dr Ben Slater and Mr Andrew Tung were co-investigators for the research component of this paper. References 1. Koch C, Li L, Figueroa P, Mihaljevic T, Svensson L, Blackstone EH. Transfusion and pulmonary morbidity after cardiac surgery. The Annals of thoracic surgery 2009; 88: 1410-8 2. Loor G, Koch CG, Sabik JF, 3rd, Li L, Blackstone EH. Implications and management of anemia in cardiac surgery: current state of knowledge. The Journal of thoracic and cardiovascular surgery 2012; 144: 538-46 3. Ferraris VA, Davenport DL, Saha SP, Austin PC, Zwischenberger JB. Surgical outcomes and transfusion of minimal amounts of blood in the operating room. Arch Surg 2012; 147: 49-55 4. Koch CG, Li L, Duncan AI, et al. Transfusion in coronary artery bypass grafting is associated with reduced long-term survival. The Annals of thoracic surgery 2006; 81: 1650-7 5. Menkis AH, Martin J, Cheng DC, et al. Drug, devices, technologies, and techniques for blood management in minimally invasive and conventional cardiothoracic surgery: a consensus statement from the International Society for Minimally Invasive Cardiothoracic Surgery (ISMICS) 2011. Innovations (Phila) 2012; 7: 229-41 6. Andreasen JJ, Sindby JE, Brocki BC, Rasmussen BS, Dethlefsen C. Efforts to change transfusion practice and reduce transfusion rates are effective in coronary artery bypass surgery. J Cardiothorac Vasc Anesth 2012; 26: 545-9 7. Ferraris VA, Brown JR, Despotis GJ, et al. 2011 update to the Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists blood conservation clinical practice guidelines. The Annals of thoracic surgery 2011; 91: 944-82 8. Qunibi WY. The efficacy and safety of current intravenous iron preparations for the management of iron-deficiency anaemia: a review. Arzneimittel-Forschung 2010; 60: 399-412 9. Onken JE, Bregman DB, Harrington RA, et al. A multicenter, randomized, active-controlled study to investigate the efficacy and safety of intravenous ferric carboxymaltose in patients with iron deficiency anemia. Transfusion 2013 10. Hung M, Besser M, Sharples LD, Nair SK, Klein AA. The prevalence and association with transfusion, intensive care unit stay and mortality of pre-operative anaemia in a cohort of cardiac surgery patients. Anaesthesia 2011; 66: 812-8 11. Patel KV. Epidemiology of anemia in older adults. Seminars in hematology 2008; 45: 210-7 12. Ashworth A, Klein AA. Cell salvage as part of a blood conservation strategy in anaesthesia. Br J Anaesth 2010; 105: 401-16 13. Klein AA, Nashef SA, Sharples L, et al. A randomized controlled trial of cell salvage in routine cardiac surgery. Anesth Analg 2008; 107: 1487-95

Page 52

Page 53

53 Blood and Volume Management Monitoring Coagulation during Cardiac Surgery Robert Young, FANZCA, Australia Specialist Anaesthetist, Flinders Medical Centre, Adelaide, Australia Introduction Blood loss and subsequent transfusion of allogenic blood products following cardiac surgery is associated with significant morbidity. The coagulopathy triggered by cardiac surgery with cardiopulmonary bypass is multifaceted and rapidly evolving. The results of standard laboratory tests are not immediately available to the clinician. As a result, the management of on-going bleeding post bypass is often empirical. Point of care testing offers the advantage of rapid analysis. Studies of the use of transfusion algorithms based on point of care coagulation assays show promise. Blood loss, transfusion and reoperation following cardiac surgery Post-operative haemorrhage remains a common complication of cardiac surgery. Karkouti and colleagues undertook a prospective study of adult patients undergoing cardiac surgery with cardiopulmonary bypass in the period 1999-2003 at the Toronto General Hospital, Canada. Of the 9215 patients included, 890 (9.7%) received at least 5 units of blood within 24 hours of surgery [1]. Following adjustment for various confounding factors, the odds of death were increased by 8.1 fold in this group. Blood transfusion has been demonstrated to increase rates of postoperative infection [2], pulmonary morbidity [3], length and cost of hospital stay [4] and to reduce long term survival [5]. Reported reoperation rates for bleeding range between 2-8%. Vivacqua and colleagues reviewed 18,891 cases of cardiac surgery between 2000 and 2009 [6]. 4566 (3%) underwent reoperation for bleeding. Technical factors were found to be the cause of the bleeding in 74%, coagulopathy in 13% and a combination of the two factors in a further 10%. The mortality in the cohort undergoing reoperation was significantly increased (8.5% v 1.8%, p< 0.0001). Both greater transfusion and reoperation were found to be independent risk factors. A number of studies have identified a further increase in morbidity and mortality with increasing time to re-exploration [7, 8, 9, 10]. It is clear that the ability to monitor coagulation accurately and in a timely fashion will lead to early, targeted intervention, a reduction in blood loss, and a lesser requirement for autologous blood transfusion. In addition, the ability to demonstrate the absence of a coagulopathy in a patient who continues to haemorrhage may prompt an earlier return to the operating theatre and attenuate the associated increase in morbidity and mortality.

Page 54

54 The Coagulopathy of Cardiac Surgery Cardiac surgery and cardiopulmonary bypass (CPB) cause derangement of coagulation by multiple mechanisms [11]. Coagulation cascades are triggered by the contact of blood with the CPB circuit and traumatised tissue leading to the consumption of coagulation factors. The inflammatory response to cardiac surgery and CPB also promotes activation of the coagulation cascade. Platelets are activated by contact with the CPB circuit, heparin and hypothermia. Activation of the coagulation system leads to increased endothelial cell production of tissue plasminogen activator (TPA) and fibrinolysis. The derangement of coagulation is exacerbated by pre-existing platelet inhibition, high dose heparin, hypothermia and the haemodilution of coagulation factors and platelets by the CPB circuit prime. Laboratory coagulation tests and their limitations. Coagulation factor assays The prothrombin time (PT) is a measure of the activity of coagulation factors II, V, VII and X, and fibrinogen. The test is performed by adding recombinant tissue factor and calcium chloride to platelet poor plasma. The PT is the time taken to the formation of clot, which is deemed to have occurred when the sample reaches a given optical density. The PT assay takes no account of the function of the cellular elements of coagulation, or the character of the clot, once it is formed. Thus, it provides limited information. The APTT is a measure of the activity of the coagulation factors of the intrinsic and common coagulation cascades. The assay is carried out in a similar manner to the PT. Platelet poor plasma is added to an artificial substance, commonly silica or elegic acid and coagulation triggered by the addition of calcium chloride. The same criticisms can be levelled at APTT as PT measurement. Neither test, when performed following cardiopulmonary bypass is a predictor of post-operative blood loss [12]. Platelet assays. The obvious limitation of a platelet count is that it is gives no indication of function. Further, even as a quantitative test, it has limitations. There is no consensus as to the level to which the platelet count must fall before coagulation is adversely affected. As a result, the decision to transfuse platelets is often arbitrary [13]. A number of platelet function assays are currently available. These monitors were initially designed to detect the effect of antiplatelet agents such as aspirin and clopidogrel on platelet function. Their utility as intraoperative monitors of platelet function is less clear. Slaughter and colleagues investigated the Platelet Function Analyzer (PFA 100, Siemens Deerfield, IL, USA) as a predictor of blood loss following cardiac surgery [14]. There was a high level of sensitivity (96%) but low specificity (18%), raising the possibility of unnecessary platelet transfusion in the treatment of false positive results. Similar findings have been reported by other authors [15, 16]. Ostrowsky and colleagues were able to demonstrate correlation between preoperative platelet function testing and post-operative blood loss using another device (Platelet Works, Helena Laboratories, Beaumont, Texas, USA) [17]. However, there was no correlation between assays performed following the reversal of heparinisation, and subsequent haemorrhage.

Page 55

55 Measurement of Heparin Effect Activated clotting time was first described By Hattersley in 1966 [18]. It is now used almost exclusively as a point of care test of the anticoagulant effect of high dose heparin in cardiac surgery. However, there is little correlation between ACT results and plasma heparin levels [19]. ACT can be prolonged by a number of factors such as thrombocytopenia, coagulation factor deficiencies, haemodilution and hypothermia. Addition of a heparinase to the assay allows differentiation of an elevated ACT due to residual heparin effect from other causes. This prevents the unnecessary and potentially harmful administration of further doses of protamine. A persistently elevated ACT is a non-specific finding and further investigation will be necessary to determine the nature of the coagulation defect. There have been a number of studies of the effect of ACT measurement on post-operative blood loss and transfusion requirements. These studies were recently reviewed by The Society of Thoracic Surgeons Blood Conservation Guideline Taskforce [20]. They identified 11 studies with only 6 showing a reduction in post-operative haemorrhage. A variation on the standard ACT measurement is ACT based heparin dose-response measurement. A sample of whole blood is mixed with known amounts of heparin and a dose response-curve generated. This allows calculation of the appropriate dose of heparin required to achieve a given level of anticoagulation in a particular patient. In the same way, following termination of cardiopulmonary bypass, blood is mixed with known amounts of protamine and a dose response curve is constructed in order to optimise dosing. There is some evidence that these assays are superior to simple ACT measurement in terms of a reduction in post-operative bleeding [21, 22]. However these assays are slower and more expensive than standard ACT measurements. They are not in widespread use [23]. Viscoelastic Assays These assays are based around the measurement of physical changes occurring in whole blood due to fibrin polymerisation. Thromboelastography (TEG®, Haemoscope Corp., IL, USA) was first described by Hartert in 1948 [24]. Whole blood is incubated in a cuvette at 37oc. A torsion wire is suspended in the sample and the cuvette oscillates around an arc of 4o45’. Fibrin forms between the cuvette and the pin causing transmission of the rotational movement from the cuvette to the pin. Rotation of the pin is detected and a graphic trace generated. Analysis of various facets of the trace gives information on the dynamics of clot formation, and its strength and durability. Thromboelastometry (ROTEM®, Pentapharm GmbH) is a more recently developed viscoelastic assay. ROTEM® differs from TEG® in that the pin rotates rather than the cuvette and movement of the pin is detected optically, (fig. 1). Various measurements are made from the ROTEM® trace (fig2). These parameters have been correlated with specific coagulation defects [25, 26].

Page 56

Page 57

57 Figure 2 Measurements derived from the ROTEM® trace. Viscoelastic assays have been in use in liver transplant surgery since 1985 [27] and in cardiac surgery since 1995 [28]. Both TEG® and ROTEM® can generate a representation of the function of coagulation factors, platelets and the fibrinolytic system within 20 minutes. However these tests have never been formerly validated and work to standardise them has only recently begun [29]. There are a few limitations to ROTEM®/TEG®. It is not possible to detect the effects of antiplatelet drugs such as aspirin and clopidogrel. Similarly, GPIIb/IIIa receptor antagonists do not affect ROTEM analysis. It is also not possible to detect von Willebrand syndrome. Supplemental platelet function assays can be used to detect these abnormalities. Point of Care testing One of the main limitations in the utility of laboratory based coagulation testing is the time taken between sampling and reporting of the result. There is obvious potential for error and delay. Turnaround times for laboratory assays are likely to be in the order of 30-60 minutes. The clinical situation can change significantly within this time. As a result, much initial treatment for ongoing haemorrhage is empirical. An obvious improvement is the use of point of care (POC) testing. It is no surprise that this has been shown to result in greatly reduced turnaround times [30]. However a number of concerns have been raised. These include issues around staff training, maintenance of equipment, maintenance of quality control and validation of results. A number of factors may cause differences in the results obtained from point of care and laboratory testing. Results will vary depending on the reagents and the instrument used [31, 32]. Whole blood is used for point of care tests. Samples sent for laboratory testing are usually citrated. There is a difference in the time lapsed between sampling and testing. All of these factors can affect the results of viscoelastic assays [33] Martin and colleagues have recently demonstrated a significant effect on ROTEM test results if the sample is transported using a pneumatic tube system [34].

Page 58

58 Viscoelastic assays with or without platelet function tests form the main stay of point of care coagulation assessment in cardiac surgery. Their utility has been investigated as predictors of post-operative blood loss, and as a guide for targeted transfusion of blood products in the treatment of coagulopathic bleeding. Prediction of blood loss. Cammerer and colleagues undertook a prospective study of 255 consecutive cardiac surgical patients in order to establish the ability of 2 POC assays – ROTEM® and the platelet function analyser (PFA-100) – to predict post-operative blood loss [35]. They found that the alpha angle measured on ROTEM® analysis post CPB was the most useful single parameter with a negative predictive value of 82% but small positive predictive value of 41%. The authors concluded that impaired coagulation as identified by these POC assays did not always lead to post-operative haemorrhage, but that a normal result in the face of ongoing blood loss should prompt early re-exploration for the treatment of surgical bleeding. Effect on Transfusion Practices. Transfusion algorithms have been constructed based on clinical experience with TEG®/ROTEM® [36]. A number of investigators have studied the potential benefits of point of POC coagulation testing together with the use of transfusion algorithms in the management of post-operative bleeding. Ak and colleagues prospectively studied 224 patients undergoing elective CABG with CPB [37]. Patients randomised to one group had clinician directed transfusion based on the results of standard coagulation tests. Those in the second group were transfused according to a TEG-based algorithm. There was no difference between the two groups in terms of blood loss, red blood cell transfusion or re-exploration. However those in the TEG based algorithm group received significantly fewer units of fresh frozen plasma (FFP) and platelets. Weber and colleagues performed a prospective randomised trial of POC testing in patients undergoing complex cardiac surgery [38]. One hundred patients who demonstrated profuse bleeding in the first 24 hours following reversal of heparinisation were randomised to a conventional or POC group. Blood from those in the conventional group was analysed in the central laboratory. PT, APTT, haemoglobin concentration, fibrinogen concentration and platelet counts were measured. Blood from those in the POC group was analysed using ROTEM® and platelet function testing. Transfusion of blood products was dictated by treatment algorithms based on the results of the various assays. There was a significant reduction in the transfusion of red blood cells, FFP and platelets in the POC group. Görlinger and colleagues undertook a retrospective cohort study of 3865 patients who underwent cardiac surgery either before or after the implementation of a transfusion algorithm based on POC coagulation testing [39]. There was a significant reduction in overall allogenic blood transfusion (52.5% v 42.2%). There was a significant reduction in the incidence of massive transfusion defined as > 10 units of packed red blood cells (2.5% v 1.3%) and in the incidence of re-exploration (4.2% v 2.2%). The use of packed red cells and FFP was reduced whilst the use of platelets, fibrinogen concentrate and prothrombin complex increased. The authors noted that over the course of the 5 years of the study (2004-2009), a number of changes occurred including the replacement of aprotinin with tranexamic acid, an increase in

Page 59

59 the use of dual antiplatelet therapy, an increase in the age of patients and the complexity of the surgery. There was also an increase in the number of emergency surgeries and a greater proportion of female patients. Effect on Treatment Costs Spalding and colleagues investigated the impact of the introduction of POC ROTEM® analysis on the costs of coagulation management in cardiac surgery [40]. Treatment costs were analysed for 729 patients prior to the introduction of ROTEM® and 693 patients afterwards. There was a change in the types and amounts of the various products used with the result that monthly expenditure reduced by 44%. The incidence of resternotomies decreased from 6.6% to 5.5%. The early mortality rate remained unchanged (5.9% v 6.0 %.) Effect on Mortality and Morbidity. A recent review by Afshari and colleagues for the Cochrane Anaesthesia Review Group found no evidence of improved morbidity or mortality with the use of ROTEM®/TEG® [41]. Studies in the review included ROTEM®/TEG® use in liver transplant surgery as well as cardiac surgery. There was a significant effect on blood loss. The review included 9 randomised controlled trials with a total of 776 participants. Only 5 trials provided information on mortality. None of the trials were powered to detect a difference in mortality. The authors concluded that there was an urgent need for adequately powered randomised controlled trials to evaluate the effect of ROTEM® and TEG®. Point of Care Coagulation Testing- use in current practice. Davies and McCorkell recently undertook a survey of cardiothoracic anaesthetists in the United Kingdom [23]. They received 116 responses with at least one respondent from each of the 37 cardiothoracic surgical centres in the country. All centres used ACT analysers in the operating theatre. A small number (7/116) used TEG® or ROTEM® for the analysis of heparin and protamine effect. Platelet function testing was available in 18 institutions (48%). Only 4 of the 37 centres had heparin/protamine dose response assays available. All but one institution had some form of viscoelastic assay available. Seventy-two respondents (62%) used point of care assays in conjunction with transfusion algorithms to guide their treatment of coagulopathic haemorrhage. Summary Clinicians are quite used to using goal directed therapy for the management of fluids and targeted titration of inotropes. It therefore seems incongruous to accept empirical use of blood products for the treatment of coagulopathic bleeding. In the current guidelines published by the Society of Thoracic Surgeons and the Society of Cardiovascular Anaesthetists, there is support for a multimodal approach to blood conservation and transfusion that includes the use of point of care coagulation monitoring together with transfusion algorithms for the treatment of coagulopathic bleeding [20].

Page 60

60 There is encouraging evidence of a reduction in blood product use with point of care testing and significant cost savings. The evidence for a reduction in morbidity is sparse. There are currently no studies adequately powered to demonstrate a reduction in mortality. However, given what we know about the adverse effects of bleeding, transfusion and re-sternotomy, it is reasonable to assume that targeted effective management of coagulopathic bleeding will reduce morbidity and mortality. References 1. Karkouti K, Wijeysundera D, Yau T, et al. The independent association between massive blood loss with mortality in cardiac surgery. Transfusion 2004; 44: 1453-1462. 2. Leal-Noval S, Rincόn-Ferrari M, García-Curiel A, et al. Transfusion of blood components and postoperative infection in patients undergoing cardiac surgery. Chest 2001; 119(5): 1461-1468. 3. Koch C, Liang L, Figueroa P, et al. Transfusion and pulmonary morbidity after cardiac surgery. Ann Thorac Surg 2009; 88: 1410-8. 4. Murphy G, Reeves B, Rogers C, et al. Increased mortality, postoperative morbidity and cost after red blood cell transfusion in patients having cardiac surgery. Circulation 2007; 116: 2544-52. 5. Engoren M, Habib R, Zacharias A, et al. effect of blood transfusion on long-term survival after cardiac operation. Ann Thorac Surg 2002; 74: 1180-6. 6. Vivacqua A, Koch C, Yousuf A, et al. Morbidity of bleeding after cardiac surgery: is it blood transfusion, reoperation for bleeding, or both? Ann Thorac Surg 2011; 91: 1780-90. 7. Kristensen K, Rauer L, Mortensen P et al. Reoperation for bleeding in cardiac surgery. Interact Cardiovasc Thorac Surg 2012; 14: 709-713 8. Karthic S, Grayson A, McCarron E, et al. Reexploration for bleeding after coronary artery bypass surgery; risk factors, outcomes, and the effect of time delay. Ann Thorac Surg 2004; 78(2): 527-34. 9. Choong C, Gerrard C, Goldsmith K, et al. Delayed re-exploration for bleeding after coronary artery bypass surgery results in adverse outcomes. J Thorac Cardiovasc Surg 2007; 31: 834-8. 10. Ranucci M, Bozzetti G, Ditta A, et al. Surgical reexploration after cardiac operations: Why a worse outcome? Ann Thorac Surg 2008; 86: 1557-62. 11. Paparella D, Brister S, Buchanan M. Coagulation disorders of cardiopulmonary bypass: a review. Intensive Care Med 2004; 30: 1873-81. 12. Gravlee G, Arora S, Lavender S, et al. Predictive value of blood clotting tests in cardiac surgical patients. Ann Thorac Surg 1994; 58(1): 216-221. 13. Vlaar A, Der Maur A, Binnekade J, et al. A survey of physicians’ reasons to transfuse plasma and platelets in the critically ill: a prospective single-centre cohort study. Transfusion Med 2009; 19(4): 207-12. 14. Slaughter T, Sreeram G, Sharma A, et al. Reversible shear-mediated platelet dysfunction during cardiac surgery as assessed by the PFA-100 platelet function analyser. Blood Coagul Fibrinolysis 2001; 12: 85-93. 15. Fattorutto M, Pradier O, Schmartz D, et al. Does the platelet function analyser (PFA-100) predict blood loss after cardiac bypass. Brit J Anaesth 2003; 90(5): 692-3. 16. Lasne D, Fiemeyer A, Chatellier G, et al. A study of platelet functions with a new analyser using high shear stress (PFA 100) in patients undergoing coronary artery bypass graft. Thromb Haemost 200; 84: 794-9. 17. Ostrowsky J, Foes J, Warchol M, et al. Plateletworks platelet function test compared to the thrombo- elastograph for prediction of postoperative outcomes. J Extra Corpor Technol 2004; 36: 149 – 52 18. Hattersley P. Activated coagulation time of whole blood. JAMA 1966; 196: 436-40. 19. Niles S, Sutton R, Ploessl J, et al. Correlation of ACT as measured with three commercially available devices with circulating heparin level during cardiac surgery. J extra Corpor Technol 1995; 27(4): 197-200. 20. Ferraris V, Ferraris S, Saha S, et al. Perioperative blood transfusion and blood conservation in cardiac surgery: The society of thoracic surgeons and the society of cardiovascular anesthesiologists clinical practice guideline. Ann Thorac Surg 2007; 83: S27-86. 21. Jobes D, Shaffer G, Aitken G. Heparin/protamine dosing guided by in vitro testing reduces blood loss and transfusion in cardiac surgery. Anesthesiol 1992; 77(3A): A137 22. Despotis G, Joist J, Hogue C, et al. The impact of heparin concentration and activated clotting time monitoring on blood conservation. A prospective, randomised evaluation in patients undergoing cardiac operation. J Thorac Cardiovasc Surg 1995; 110: 46-54. 23. Davies R, McCorkell S. Point of care coagulation testing in cardiac surgery: a UK national survey. Available online at www.acta.org.uk 24. Hartert H. Blutgerinnungsstudien mit der Thrombelastographie, einem neuen Untersuchungsverfahren. Klin Wschr 1948; 26: 577-83. 25. Ogawa S, Szlam F, Chen E, et al. A comparative evaluation of rotation thromboelastometry and standard coagulation tests in hemodilution-induced coagulation changes after cardiac surgery. Transfusion 2012; 52: 14-22. 26. Spiel A, Mayr F, Firbas C et al. Validation of rotational thromboelastometry in a model of systemic activation of fibrinolysis and coagulation in humans. J Thromb Haemost 2006; 4: 411–6. 27. Kang Y, Marquez D, Lewis J, et al. Intraoperative changes in blood coagulation and thromboelastographic monitoring in liver transplantation. Anesth Analg 1985; 64: 888-96. 28. Spiess B, Gilles B, Chandler W, et al. Changes in transfusion therapy and reexploration rate after institution of a blood management program in cardiac surgical patients. J Cardiothorac Vasc Anesth 1995; 9: 168-73. 29. Chitlur M, Sorensen B, Rivard G, et al. Standardisation of thromboelastography; a report from the TEG-ROTEM working group. Haemophilia 2011; 17(3) 532-7. 30. Toulon P, Ozier Y, Ankri A, et al. Point-of-care versus central laboratory coagulation testing during haemorrhagic surgery. A multicenter study. Thromb Haemost 2009, 101:394–401.

Page 61

61 31. Naghibi F, Han Y, Dodds J, et al. Effects of reagents and instrument on prothrombin times, activated partial thromboplastin times and patient/control ratios. Thromb Haemost 1988; 59: 455-63. 32. Theusinger O, Nürnberg J, Asmis L, et al. Rotation thromboelastometry (ROTEM®) stability and reproducibility over time. Eur J Cardio-thorac Surg 2010 37: 677-83. 33. Ogawa S, Szlam F, Chen E, et al. A comparative evaluation of rotation thromboelastometry and standard coagulation tests in haemodilution-induced coagulation changes after cardiac surgery. Transfusion 2012; 52: 14-22. 34. Martin J, Schuster T, Moessmer G, et al. Alterations in rotation thromboelastometry (ROTEM®) parameters: point-of-care testing vs analysis after pneumatic tube system transport. Brit J Anaesth 2012; 109(4): 540-5. 35. Cammerer U, Dietrich W, Rampf T, et al. The predictive value of modified computerized thromboelastography and platelet function analysis for postoperative blood loss in routine cardiac surgery. Anesth Analg 2003; 96: 51-7. 36. Görlinger K, Jambor C, Hanke A, et al. Perioperative coagulation management and control of platelet transfusion by point-of-care platelet function analysis. Transfus Med Hemother 2007; 34: 396-411. 37. Ak K, Isbir C, Tetik S, et al. Thromboelastography-based transfusion algorithm reduces blood product use after elective CABG: A prospective randomised study. J. Cardiac Surg 2009; 24(4) 404-10. 38. Weber C, Görlinger K, Meininger D, et al. Point-of-care testing. A prospective, randomised clinical trial of efficacy in coagulopathic cardiac surgery patients. Anesthesiol 2012; 117(3): 531-47. 39. Görlinger K, Dirkmann D, Hanke A, et al. First-line therapy with coagulation factor concentrates combined with point-of-care coagulation testing is associated with decreased allogenic blood transfusion in cardiovascular surgery. A retrospective, single-center cohort study. Anesth 2011; 115(6): 1179-91. 40. Spalding G, Hartrumpf M, Sierig T, et al. Cost reduction of perioperative coagulation management in cardiac surgery: value of ‘bedside’ thromboelastometry (ROTEM). Eur J Cardio-thorac Surg 2007; 31: 1052-7. 41. Afshari A, Wikkelsø A, Brok J, et al. Thromboelastography (TEG) or thromboelastometry (ROTEM) to monitor haemotherapy versus usual care in patients with massive transfusion (Review). Cochrane Database Syst Rev 2011; 16(3).

Page 62

Page 63

63 SESSION 6 Monday 16:00 - 17:30 PDUC Blood Management Rationalising Red Blood Cell Transfusion in Cardiac Surgery: A Multicentre Quality Improvement Initiative of the Perfusion Downunder Collaboration. Rob Baker, PhD, Dip Perf, CCP, BMedSci (Hons), & Richard Newland, BSc, Dip Perf, CCP, Australia Director, Cardiac Surgery Research & Perfusion, Cardio-Thoracic Surgical Unit, Flinders Medical Centre, Adelaide, Australia & Advanced Perfusionist, Cardiac Surgery Research & Perfusion, Cardio-Thoracic Surgical Unit, Flinders Medical Centre, Adelaide, Australia on behalf of the Perfusion Downunder Collaboration Introduction Although red blood cell (RBC) transfusion may be utilised during cardiac surgery to increase oxygen carrying capacity in the setting of blood loss or anaemia, the incidence of RBC transfusion has been associated with an increase in risk of bacterial infections, low-output failure, longer intensive care unit stay and increase in mortality (1-4). A single centre study focussing on reducing variation in the number of perioperative transfusions associated with cardiac surgery demonstrated reduction in transfusion rates in a three phase initiative, involving understanding current processes, implementing new protocols and monitoring progress of protocol implementation (5). The most frequent indication for transfusion was found to be anaemia, with 90% of intraoperative transfusions given for actual or predicted low haematocrit and 43% postoperative. Following the introduction of protocols, compliance was adhered to in 27% of patients intraoperatively and 36.5% postoperatively. Undertaking a similar initiative in a multicentre registry setting will provide an understanding of how institutional variation in process of care can influence the incidence of anaemia and transfusion, and direct quality improvement initiatives through benchmarking. Previously, we have reported baseline data for various cardiopulmonary bypass (CPB) process of care measures highlighting the potential for process improvement using benchmarking, through the comparison of data and understanding of process variation (6). Our current project is aimed at reducing RBC transfusion in a multicentre setting, through a reduction in the incidence of anaemia and improvement in adherence to institutional protocols in sites contributing to the Perfusion Downunder Collaboration.

Page 64

64 Methods This report employed the SQUIRE publication guidelines for healthcare quality improvement research [7]. The project will be undertaken according to the approval of local ethics committees1 Setting Data will be collected routinely from procedures performed in nine Australian and New Zealand cardiac centres currently contributing the Perfusion Downunder Collaborative Database (PDUCD) as previously described (6). Data collected from Jan 2007 – Feb 2013 will be utilised to report baseline measures of process outcome. Interventions The quality improvement initiative will be undertaken in three phases; Phase I: The initial phase will focus on understanding current protocols and processes of care in relation to perioperative blood management, the incidence of perioperative anaemia and RBC transfusion. Each centre will have the opportunity to present their current protocols and data collected since participation in the PDUCD. Presentations will be given at a dedicated session at the Perfusion Downunder Meeting, Hayman Island, September 1st – 3rd, 2013. Utilising the principles of benchmarking, discussion during this session will be focussed towards identification of best performing centres, and the association of the results of these centres with processes of care, unit culture and protocols that may lead to quality improvement though adoption of or modification in current practices in other centres. Results of the session will be summarised and provided to each centre for dissemination at unit level. Phase II: The second phase will firstly; involve dissemination of information at individual centres to guide adoption or modification in processes of care and secondly; to collect information prospectively as to indications for RBC transfusion over a three month period in order to determine adherence to institutional protocols. This phase will be implemented once approval has been granted from institutional ethics committees. Clinical interventions will be considered at each centre based on the clinical practice guidelines for blood conservation published by the Society of Thoracic Surgeons (STS) and the Society of Cardiovascular Anaesthesiologists (SCA) (8). A local quality improvement (LQI) team at each centre comprising a cardiothoracic surgeon, anaesthetist, perfusionist and nursing staff will be responsible for co-ordination of the project, and implementation of protocols for RBC transfusion. The requirement for protocol inclusion will be; criteria for addition of RBC in the CPB prime based on predicted haemoglobin, and minimum intraoperative and postoperative haemoglobin levels as indications for RBC transfusion. Specifically, at the beginning of the phase, the LQI team will aim to 1 Flinders Clinical Research Ethics Committee; Human Research Ethics Committee (TAS) Network; Northern X Regional Ethics Committee; Cabrini Health Research Ethics Committee; Western Sydney Local Health District Human Research Ethics Committee; Alfred Hospital Ethics Committee; Royal Perth Hospital Human Research Ethics Committee.

Page 65

65 educate staff with the evidence base for perioperative blood conservation according to the STS/SCA guidelines and RBC transfusion according to the Australian National Blood Authority patient blood management guidelines, provide the data summary of the benchmarking results from the PDUCD and ensure RBC transfusion protocols are implemented. At the end of the Phase, the team will feedback the individual centre results for incidence of transfusion, summary of the indications for transfusion and adherence to protocols during phase II. Phase III: The final phase will continue to monitor indications for transfusion and feedback data to staff regarding incidence of intraoperative and postoperative RBC transfusion on a monthly basis. Quarterly data integration from the PDUCD will be utilised over three month periods to track comparative progress for benchmarking. Final analysis will take place after 6 months of data collection and feedback. Measures The primary outcomes of this study will be incidence of RBC transfusion, minimum haemoglobin during CPB, and rate of adherence to institutional protocols. Secondary outcome measures will include length of postoperative stay, and mortality. Analysis Standard statistical methodology will be used to compare proportional and continuous data variables. Analyses will be performed using SPSS V20.0 References 1. Chelemer SB, Prato BS, Cox PM Jr, et al. Association of bacterial infection and red blood cell transfusion after coronary artery bypass surgery. Ann Thorac Surg 2002;73:138e42. 2. Corwin HL, Gettinger A, Pearl RG, et al. The CRIT Study: anemia and blood transfusion in the critically ill-current clinical practice in the United States. Crit Care Med 2004;32:39e52. 3. Koch CG, Li L, Duncan AI, et al. Transfusion in coronary artery bypass grafting is associated with reduced long-term survival. Ann Thorac Surg 2006;81:1650e7. 4. Surgenor SD, DeFoe GR, Fillinger MP, et al. Intraoperative red blood cell transfusion during coronary artery graft surgery increases the risk of postoperative low-output heart failure. Circulation. Jul 4 2006;114(1 Suppl):I43-48. 5. Likosky DS, Surgenor SD, Dacey Lj, et al. Rationalising the treatment of anaemia in cardiac surgery: short and mid-term results from a local quality improvement initiative. Qual Saf Health Care 2010 19: 392-398. 6. Baker RA, Newland RF, Fenton C, McDonald M, Willcox TW, Merry AF. Developing a Benchmarking Process in Perfusion: A Report of the Perfusion Downunder Collaboration. JECT. 2012;44:26–33. 7. Davidoff F, Batalden P, Stevens D, Ogrinc G, Mooney S. Publication guidelines for quality improvement in health care: evolution of the SQUIRE project. Qual Saf Health Care 2008;17[Supplement 1]:i3-i9. 8. Ferraris VA, Brown JR, Despotis GJ et al. 2011 Update to The Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists Blood Conservation Clinical Practice Guidelines. Ann Thorac Surg 2011;91:944-982.

Page 66

Page 67

Page 68

Page 69

Page 70

Page 71

71 Tuesday 3rd September 2013 SESSION 7 Tuesday 08:00 - 09:30 Neurocognitive/Brain Assessment Cardiac Surgery, the Brain and Inflammation David A. Scott, MB BS, FANZCA, PhD, FFPMANZCA, Australia Associate Professor and Director of Anaesthesia, St Vincent’s Hospital, Melbourne, Australia, Associate Professor, Faculty of Medicine, Dentistry and Health Sciences, University of Melbourne, Melbourne, Australia Authors: David A. Scott, Lisbeth A. Evered and Brendan S. Silbert Department of Anaesthesia, St Vincent’s Hospital Abstract Cognitive deterioration can reliably be measured following procedures requiring anaesthesia and surgery. Cardiac surgery has had the spotlight because of the high reported incidence of Post-operative Cognitive dysfunction (POCD) in early studies, but such effects occur following other surgical procedures as well. ’Early’ POCD should be considered as a different phenomenon, relating to acute pharmacological, physiological and stress-related recovery. The focus should be on what is affecting patients at 3 months, 12 month and 5 years later. As with many other aspects of perioperative risk, a significant element is the patient’s pre-operative cognitive status. We now know that up to a third of overtly ‘normal’ elective cardiac surgical patients enter surgery with some degree of pre-existing cognitive impairment (PreCI) or, when applying psychogeriatric measures, Mild Cognitive Impairment (MCI). The latter is a known prodrome or early stage of the amyloid associated Alzheimer’s disease dementia. Inflammatory responses during cardiac surgery have been recognised for years but our understanding of the complexity of systemic inflammatory response has grown significantly with the ability to assay neurohumoral markers such as interleukins and also with promising new leads like sub-regulatory micro-RNAs. The blood-brain barrier is made vulnerable by both pre-existing disorders (MCI/amyloid; vascular disease) and by the inflammatory response to surgery and CPB. Inflammation affecting the brain at this time may set in motion accelerated neurological and hence cognitive decline that, despite an initial recovery and even functional improvement, may proceed to further long-term decline at an accelerated rate in susceptible individuals. Clinical data is emerging from longer term studies to support this concern but evidence for effective preventive or therapeutic strategies is limited.

Page 72

72 Introduction Surgery is always undertaken with the anticipation of improving an individual’s condition. This improvement extends over a wide range of health issues, such as cosmetic procedures, providing a diagnosis for directing therapy, improving quality of life as in hip joint replacement surgery or even prolonging or saving life as in cancer resection or cardiac surgery. Any factors which adversely affect outcomes may potentially negate the purpose of the surgery and should therefore be aggressively identified and managed. Following reductions in perioperative mortality in most areas of surgery, the complications which most significantly affect outcome and quality of life are those that affect the brain. The brain is well protected from external threats anatomically by a rigid skull, behaviourally by high-level strategies to avoid physical injury, and physiologically by homeostatic mechanisms to maintain perfusion, and ensure oxygenation. Unfortunately, internal threats are less well defended, including exposure to drugs and alcohol. The brain is particularly susceptible to metabolic and pharmacologic interference, a susceptibility which is exploited when administering general anaesthesia which employs volatile or intravenous drugs to induce a deep coma – often mis-described as ‘sleep’. Over 100 years ago Savage reported that that some individuals did not return to their pre-anaesthetic cognitive state after surgery and anaesthesia[1] but despite reports in the 50’s and 60’s[2] [3] it was the description of cognitive decline after cardiac surgery which attracted major attention [4]. Even then, cognitive decline was viewed a minor complication as the major focus in cardiac surgery was to improve survival. In the first two decades of open heart surgery the main cerebral events of concern were stroke and acute confusional states (‘pump brain’). Post-operative cognitive dysfunction (POCD) following surgery began to be measured consistently throughout the 1980s. Unfortunately there was much confusion regarding testing procedures and diagnostic criteria. Many investigators included delirium in the POCD category, and testing time points after surgery ranged from days to months. A consensus statement emerged, providing some clarity regarding the selection of neuropsychological tests to be used [5] however the ‘diagnosis’ of POCD remained open to interpretation, usually relying on a change in a patient’s test performance of a specified magnitude in a specified number of cognitive tests. In an effort to improve consistency, most investigators now use a form of Reliable Change Index (RCI)[6] to compare patients to a comparable control group. POCD is not a clinical syndrome, is not defined in the ‘Bible’ of neuropsychological diagnoses, the Diagnostic and Statistical Manual of Mental Disorders, and remains a diagnosis based solely on neuropsychological test performance. This has led some investigators to question its validity as a construct, but weight of evidence supports its existence, including the consistency of POCD ‘detection’ by multiple investigators, and the adverse association of POCD with outcomes such as length of hospital stay, quality of life and mortality [7]. There remain many unanswered questions - why does POCD develop, who is susceptible, is it reversible, and how does it relate to other neurodegenerative conditions? The answers to these questions are likely to be inter-related and include recognition of pre-existing vulnerabilities, a better understanding of POCD at different stages of the patient’s recovery, and quite possibly the over-arching role of inflammation in the process. This review will explore aspects of the current status of these concepts and also relate them to emerging concerns regarding long term cognitive impairment and in particular, dementia.

Page 73

73 POCD and Cardiac Surgery POCD follows both cardiac and non-cardiac surgery. When patients are tested at hospital discharge or at one-week following the procedure, the incidence of POCD is higher following cardiac surgery than non-cardiac surgery (Table 1). Up to 43% of patients have POCD at 7 days following coronary artery bypass graft (CABG) surgery [8] .By 3 months and 12 months following the procedure however, there is no significant difference in the incidence of POCD regardless of the type of surgery or anaesthesia. In a prospective investigation using RCI to diagnose POCD in three patient populations (one group undergoing elective CABG with CPB under general anaesthesia (GA), one group having total hip joint replacement with spinal anaesthesia and GA, and the third having coronary angiography with sedation) the incidence of POCD at three months was 16%, 17% and 21% respectively [8].In spite of the great variation in the magnitude of surgery and type of anaesthesia, any difference in the incidence of POCD at 3 and 12 months was indistinguishable. When viewed in comparison with the incidence of POCD after one week in the cardiac group, this suggests that testing at one week after surgery may be identifying a phenomenon arising from a different mechanism. Thus the likelihood is that ‘early’ POCD is largely a different entity to later POCD Within the realm of cardiac surgery, it was long held that the physical or physiological ‘insult’ of CPB was the cause of postoperative delirium and POCD. It seemed highly plausible that the presence of microemboli, hypotension or the abnormal perfusion of CPB were contributing to the aetiology for POCD. With the advent of off-pump surgery, it finally became possible to compare cognitive outcomes in patients having similar operative procedures and anaesthetics with or without the heart lung machine. The initial studies of van Dijk [9] in 281 patients surprisingly showed an incidence at 3 months of 27% on-pump versus 21% off-pump (p=0.15) and these outcomes have been confirmed by others [10, 11]. This demonstrates that the process of CPB is not the primary factor in POCD. Others have investigated cerebral micro-embolic load using transcranial Doppler and found no association [12-14], further dissociating the heart-lung machine from POCD. The identification of POCD with cardiac surgery remains important. Despite it being a non-clinical diagnosis, the presence of POCD has been associated with an average of 1.2 days longer hospital stay (mean (SD) for no POCD 7.1 (3.4) d vs. POCD 8.3 (4.1) d; p=0.02) [15]. Quality of life scores were found to be lower at 12 months in those with POCD [16], and one-year mortality was higher [7]. Finally, the identification of cognitive decline 5 years post cardiac surgery was associated with POCD at 7 days [17]. Risk factors for development of POCD are consistently identified as increasing age and lower IQ (or education levels) [15, 18]. These may both relate to the degree of ‘cognitive reserve’ that an individual has [REF]. Baseline cognitive test performance has also been analysed to indicate the presence of pre-existing cognitive impairment (PreCI) or a more formal diagnosis of Mild Cognitive Impairment (MCI) which is a precursor to Alzheimer’s disease dementia (AD) [19]. The latter requires specific information which is not part of the ‘routine’ test batteries for POCD, in particular both subjective and objective reports of memory impairment. MCI is an important consideration because of the association with neuronal injury and its possible exacerbation by anaesthetic agents or the inflammatory processes associated with surgery.

Page 74

74 Delirium and Cardiac Surgery Delirium after cardiac surgery is independently associated with mortality up to 10 years postoperatively [20]. Delirium is an acute neuropsychiatric syndrome characterised by decline in attention, fluctuating conscious levels and disorganised thinking. It is of particular concern because of its high incidence, clinical and social consequences and because it is associated with physical precipitants. In the perioperative period, delirium occurs with a high incidence in surgical patients and patients in critical care environments. Following cardiac surgery the incidence is reported at up to 52% [21, 22], and in intensive care at up to 90% [23]. In cardiac surgery, many triggers are present such as a need for many days of hospitalisation, including time spent in a high dependency environment where circadian cues are absent. Further triggers include exposure to multiple drugs, general anaesthesia for many hours, postoperative sedation, analgesics for pain management and indwelling urinary catheters. Delirium has significant short and long-term consequences affecting quality of life, social needs and clinical morbidity and mortality [21, 24, 25]. Yet its diagnosis is frequently missed, [26] in part because the common hypoactive presentation may be undiagnosed. Clinical studies relying on a retrospective diagnosis of delirium as an indicator of cognitive outcomes are often flawed for this reason. The exact cause(s) of delirium is unknown but it is associated with a combination of predisposing and precipitating factors [27]. There is some likelihood that delirium is associated with increasing age (as for POCD) and pre-existing cognitive impairment or dementia [28]. The final common pathway of many of the insults associated with delirium is likely to be some degree of challenge which erodes the individual’s cognitive reserve (eg changed environment, pain, and stress), alters inhibitory and cognitive pathways (eg medications) and activates the inflammatory cascade [29]. In patients undergoing cardiac surgery it is not possible to modify many of the factors that have been identified as triggers for delirium. Therefore in these patients, despite the best of physical and psychological clinical care, there is a need to seek protective strategies for the brain, especially during the initiation period of peak neurochemical disruption i.e. during anaesthesia, surgery and CPB. Although an association between postoperative delirium and pre-existing dementia has been made, to date the contribution of postoperative delirium to long-term cognitive impairment (LTCI), or to subsequent dementia is less clear [30]. Preoperative cognition, MCI and Dementia The population of patients presenting for all forms of surgery is ageing as the population ages. Cognitive decline is highly associated with increased age. For example, the prevalence of Mild Cognitive Impairment (MCI) identified by subtle cognitive impairment increases with age. At present, approximately 16% of patients over 70y will have MCI and over 17% will progress to dementia every year [31] [32]. This is a huge problem where currently, 32% of anaesthetics are administered to those aged over 65y, which is projected to rise to 48% by 2051 [33]. In elderly patients (mean (SD) age, 69.8 (6.3) y) presenting for elective total hip joint replacement surgery, PreCI was identified in 20% and amnestic MCI in 22% [19]. This figure is even higher in patients presenting for coronary artery surgery (data not yet published). The relevance of this is that cognitive impairment may predispose to postoperative delirium, and that the processes underlying some forms of cognitive impairment may be exacerbated, accelerating clinical deterioration [34].

Page 75

75 Subtle cognitive impairment (either MCI or PreCI) is believed to represent early stages of Alzheimer’s Disease (AD)[35]. The neuronal degeneration and neuroinflammation of AD is associated with amyloid protein deposition (A40, A42) causing amyloid plaques, and tau proteinopathy resulting in neurofibrillary tangles [36].. These processes progress over time at different rates in affected individuals. The rate of cognitive decline is seldom assessed in studies of cognition and thus the effect of cardiac surgery and CPB on the trajectory of decline is unknown. There is a debate at the moment on the impact of anaesthesia on dementia [34], with conflicting epidemiological data [37-39]. However, laboratory data suggests that anaesthetic agents, especially volatile agents, may have some effect on the underlying processes of AD [40]. The key to understanding these effects, especially in cardiac surgery, may lie in the underlying inflammatory processes [29, 41]. Inflammation It is well known that cardiac surgery and CPB is associated with a systemic inflammatory response. Triggering factors include tissue injury and organ ischaemia, the neurohumoral stress response, and of course the process of CPB with the extensive foreign surface exposure and physical trauma to blood elements [42, 43]. The inflammatory responses to CPB include activation of clotting factors, platelets and fibrinolysis, elevation of a vast range of inflammatory cytokines including Interleukin (IL)-1, IL-6 and Tumor Necrosis Factor-alpha, and activation of endothelial and leukocyte responses.[43] Genetics may affect the inflammatory response or vulnerability to inflammation. Specific genetic polymorphisms affecting IL-6 and C-reactive protein have been identified and associated with an increased risk of stroke following cardiac surgery [44]. Apolipoprotein E (APOE) affects CNS acetylcholine synthesis, and individuals with the epsilon-4 subtype (APOE- -peprotein E (APOE) affects CNS acetylcholine synthesis, and individuals with the epsilon- polymorphisms affecting IL-6 and C-reactive protein have been identified and associated with an increased [45,46]. Systemic inflammation is common to the numerous medical and surgical conditions associated with cognitive change and delirium [29,47]. The inflammatory response provides a common process which unites the multitude of precipitating factors in vulnerable patients, and it has been implicated in dementia [29]. One mechanism by which inflammation may contribute to cognitive change is by pro-inflammatory cytokines produced by macrophages and monocytes increasing permeability of the blood brain barrier and altering neurotransmission [48]. Cardiac surgery thus provides a ‘tetrad’ of circumstances that may set the scene for initiating or accelerating cognitive decline. These are: (i) pre-existing cognitive impairment (PreCI, MCI or dementia) with its associated diminished cognitive reserve; (ii) pre-existing inflammatory states such as vascular disease or AD; (iii) the triggering of widespread systemic inflammation; and (iv) alteration in the blood brain barrier increasing exposure of CNS neurons to toxic or inflammatory effects. In addition, micro-embolism, regional or global hypoperfusion or hypoxia may increase vulnerability. Items (i) through (iii) are present to a greater or lesser extent in many procedural situations.

Page 76

76 Preventative Strategies In many physiologic and pathologic processes, preventive interventions are more effective than treatments. For example, prevention of the triggering event is most effective in relation to coagulation, immunologic, inflammatory or neuroexcitatory responses (such as in acute nociceptive responses). Pharmacological Steroids Steroids would seem an obvious choice to modify inflammatory responses, although few studies have been designed to investigate their efficacy in improving cognitive outcomes. Blunting of the normal diurnal variation in cortisol levels was significantly related to POCD at 1 week after cardiac surgery. A meta-analysis of perioperative steroids in cardiac surgery did not assess POCD, but showed no advantage in stroke [49]. A study in 4494 cardiac surgical patients compared high dose dexamethasone (1 mg/kg) with placebo and found dexamethasone use to be associated with reductions in postoperative infection, duration of postoperative mechanical ventilation, and duration of ICU and hospital stays. It was not associated with a reduction in the incidence of major adverse events, including stroke at 30 days, compared with placebo [50]. NSAIDs Limited animal data suggests that inhibition of microglial proliferation by meloxicam was associated with improved short-term memory function in rodents [51]. Ketamine Ketamine is a phencyclidine anaesthetic and anti-hyperalgesic agent with a range of pharmacological effects that are potentially neuroprotective [52]. As an N-methyl D-aspartate (NMDA) receptor antagonist, ketamine reduces glutamate induced calcium ion influx which has been shown to trigger neural injury and cell death [53]. In-vivo studies using ketamine protection against glutamate-induced or ischaemic neural injury resulted in less neuronal damage. In addition, ketamine possesses anti-inflammatory effects which are demonstrable in vitro [54], is protective following head trauma in rats [55] and attenuates human inflammatory responses post-surgery [56]. In sub-anaesthetic doses, ketamine appears to have a role in out-of-hospital emergency medicine to treat excited delirium [57]. As a preventive strategy, in a small study of cardiac surgical patients, ketamine administration immediately prior to the procedure has been shown to reduce the incidence of postoperative delirium (POD) [58]. If this is borne out in a more rigorous investigation, then important short and long-term effects such as progression to dementia may also be evaluated. Specific drugs Etanercept is a TNF- antagonist that has been used in some trials to modify the inflammatory response [REF]. Its effect on cognition has not yet been explored. It is unlikely, on its own, to be effective at brain ‘protection’ because it is a highly specific

Page 77

77 antagonist and there are a great many inflammatory mediators at work. Side effects include an increased risk of infectious complications. It may be useful as part of a larger integrated strategy. Many other drugs have shown some evidence of efficacy in small trials but await more rigorous investigations to determine their utility. Lignocaine is one example where its pharmacologic properties (neuronal stabilisation, anti-inflammatory effects) and laboratory data [59] have led some investigators to explore its efficacy with equivocal clinical results [60]. The investigation of lignocaine is difficult in cardiac surgery because it is frequently administered either as a component of cardioplegic solutions or in the treatment of dysrhythmias. Thiopentone is another agent which has long been used as a cerebral protective agent in a range of situations but for which evidence for efficacy in cardiac surgery is lacking [60]. Cognitive The brain is an incredibly adaptive organ, and strategies to exploit neuroplasticity to buffer cognitive insults and increase cognitive reserve offer the most practical post-operative intervention opportunities at present. In the acute phase, avoiding cognitive disruption by improving postoperative sleep patterns has been advocated as part of a multifaceted approach to perioperative care [61]. In the medium to longer term, cognitive enrichment including improved social engagement is likely to be beneficial and has been demonstrated in human and animal investigations [62-65]. This is supported by a retrospective analysis of predictors of recovery from cognitive change following cardiac surgery which identified greater activities of daily living at 6 weeks as a significant predictor of recovery (OR 0.891 (0.810 – 0.981]), along with baseline education level and baseline cognition [18]. Procedural Avoiding conditions that stress the metabolism of the brain may also be beneficial in improving cognitive outcomes. Intraoperative hyperglycaemia (> 11 mmol/L) has been associated with increased POCD in non-diabetic patients (p=0.035) [66]. Rewarming hyperthermia (> 37 degrees C nasopharyngeal temperature) may create a hypermetabolic state, and in one small study was linked to a higher incidence of POCD at 6 weeks (p=0.05) [67]. Hypotension and hypoxaemia might be expected to have an association with cognitive decline, and this is clearly the case for prolonged or extreme episodes where global cerebral insults occur. However, in the ISPOCD study, in elderly non-cardiac surgery patients, neither factor was associated with POCD[68]. It is likely that these events are more pronounced in cardiac surgery and strategies to optimise cerebral perfusion and oxygenation may lead to improved outcomes. The use of Near Infra-Red Spectroscopy (NIRS) to continually monitor cerebral oxygenation (cerebral oximetry) may lead to more effective maintenance of global cerebral state [69, 70]. Minimising macroembolism would clearly reduce focal ischaemia and stroke but the situation is less clear for microembolic load and POCD. Washing cardiotomy blood to remove contaminants prior to reinfusion was not associated with improved cognitive outcomes [71]. The incidence of cognitive dysfunction has not been consistently shown to be associated with microembolic load as detected by TCD following cardiac surgery[72, 73], possibly because the majority are gaseous [74].

Page 78

78 None-the-less it seems prudent to avoid exposure of the brain to unnecessary microembolism. Conclusion In concert with well-established cognitive support strategies, the use of neuroprotective drugs during exposure to anaesthesia and surgical stress may in the future offer clinical benefits to the incidence of delirium, POCD and possibly dementia. For preventive strategies to be most effective, with minimal side-effects, they need to be targeted at those who are identified as most at risk. This requires attention to pre-operative identification of vulnerable patients with mild or even subclinical cognitive impairment, and possibly in the future identifying relevant biomarkers or even genotypes. Care does not end at discharge and treatments, including cognitive enrichment strategies, should be pursued well into the postoperative period to maximise outcomes. References 1. Savage, G.H., Insanity following the use of anaesthetics in operations. British Medical Journal, 1887. Dec, 3: p. 1199-1200. 2. Bedford, P., Adverse cerebral effects of anaesthesia on old people. Lancet, 1955: p. 259-263. 3. Simpson, B.R., et al., The effects of anesthesia and elective surgery on old people. Lancet, 1961. 2(7208): p. 887-93. 4. Shaw, P.J., et al., Neurological complications of coronary artery bypass graft surgery: six month follow-up study. Br Med J (Clin Res Ed), 1986. 293(6540): p. 165-7. 5. Murkin, J.M., et al., Statement of consensus on assessment of neurobehavioral outcomes after cardiac surgery. Ann Thorac Surg, 1995. 59(5): p. 1289-95. 6. Rasmussen, L.S., et al., The assessment of postoperative cognitive function. Acta Anaesthesiol Scand, 2001. 45(3): p. 275-89. 7. Monk, T.G., et al., Predictors of cognitive dysfunction after major noncardiac surgery. Anesthesiology, 2008. 108(1): p. 18-30. 8. Evered, L., et al., Postoperative cognitive dysfunction is independent of type of surgery and anesthetic. Anesth Analg, 2011. 112(5): p. 1179-85. 9. Van Dijk, D., et al., Cognitive outcome after off-pump and on-pump coronary artery bypass graft surgery: a randomized trial. JAMA, 2002. 287(11): p. 1405-12. 10. Sun, J.H., et al., Cognitive dysfunction after off-pump versus on-pump coronary artery bypass surgery: a meta-analysis. J Int Med Res, 2012. 40(3): p. 852-8. 11. Kennedy, E.D., et al., Cognitive outcome after on- and off-pump coronary artery bypass grafting surgery: a systematic review and meta-analysis. J Cardiothorac Vasc Anesth, 2013. 27(2): p. 253-65. 12. Kruis, R.W., F.A. Vlasveld, and D. Van Dijk, The (un)importance of cerebral microemboli. Semin Cardiothorac Vasc Anesth, 2010. 14(2): p. 111-8. 13. Martin, K.K., et al., Intraoperative cerebral high-intensity transient signals and postoperative cognitive function: a systematic review. Am J Surg, 2009. 197(1): p. 55-63. 14. Rodriguez, R.A., et al., Cerebral emboli detected by transcranial Doppler during cardiopulmonary bypass are not correlated with postoperative cognitive deficits. Stroke, 2010. 41(10): p. 2229-35. 15. Silbert, B.S., et al., A comparison of the effect of high- and low-dose fentanyl on the incidence of postoperative cognitive dysfunction after coronary artery bypass surgery in the elderly. Anesthesiology, 2006. 104(6): p. 1137-45. 16. Phillips-Bute, B., et al., Association of neurocognitive function and quality of life 1 year after coronary artery bypass graft (CABG) surgery. Psychosom Med, 2006. 68(3): p. 369-75. 17. Newman, M.F., et al., Longitudinal assessment of neurocognitive function after coronary-artery bypass surgery. N Engl J Med, 2001. 344(6): p. 395-402. 18. Fontes, M.T., et al., Predictors of cognitive recovery after cardiac surgery. Anesth Analg, 2013. 116(2): p. 435-42. 19. Evered, L.A., et al., Preexisting cognitive impairment and mild cognitive impairment in subjects presenting for total hip joint replacement. Anesthesiology, 2011. 114(6): p. 1297-304. 20. Gottesman, R.F., et al., Delirium after coronary artery bypass graft surgery and late mortality. Ann Neurol, 2010. 67(3): p. 338-44. 21. Rudolph, J.L., et al., Delirium: an independent predictor of functional decline after cardiac surgery. J Am Geriatr Soc, 2010. 58(4): p. 643-9. 22. Lin, Y., J. Chen, and Z. Wang, Meta-analysis of factors which influence delirium following cardiac surgery. J Card Surg, 2012. 27(4): p. 481-92. 23. Jacobi, J., et al., Clinical practice guidelines for the sustained use of sedatives and analgesics in the critically ill adult. Crit Care Med, 2002. 30(1): p. 119-41. 24. Inouye, S.K., et al., Does delirium contribute to poor hospital outcomes? A three-site epidemiologic study. J Gen Intern Med, 1998. 13(4): p. 234-42. 25. Leslie, D.L., et al., Premature death associated with delirium at 1-year follow-up. Arch Intern Med, 2005. 165(14): p. 1657-62. 26. Meagher, D.J., et al., A longitudinal study of motor subtypes in delirium: frequency and stability during episodes. J Psychosom Res, 2012. 72(3): p. 236-41.

Page 79

79 27. Rudolph, J.L. and E.R. Marcantonio, Review articles: postoperative delirium: acute change with long-term implications. Anesth Analg, 2011. 112(5): p. 1202-11. 28. Inouye, S.K., Delirium in hospitalized older patients: recognition and risk factors. J Geriatr Psychiatry Neurol, 1998. 11(3): p. 118-25; discussion 157-8. 29. Simone, M.J. and Z.S. Tan, The role of inflammation in the pathogenesis of delirium and dementia in older adults: a review. CNS Neurosci Ther, 2011. 17(5): p. 506-13. 30. MacLullich, A.M., et al., Delirium and long-term cognitive impairment. Int Rev Psychiatry, 2009. 21(1): p. 30-42. 31. Petersen, R.C., et al., Prevalence of mild cognitive impairment is higher in men. The Mayo Clinic Study of Aging. Neurology, 2010. 75(10): p. 889-97. 32. Landau, S.M., et al., Comparing predictors of conversion and decline in mild cognitive impairment. Neurology, 2010. 75(3): p. 230-8. 33. Australian_Institute_of_Health_and_Welfare, Australia’s health 2010, Australian Government, Australia’s health series no. 12, AUS 122: Canberra, 2010 34. Scott, D.A., B.S. Silbert, and L.A. Evered, Anesthesia and Alzheimer's disease: time to wake up! Int Psychogeriatr, 2013. 25(3): p. 341-4. 35. Jack, C.R., Jr., et al., Introduction to the recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement, 2011. 7(3): p. 257-62. 36. Bissette, G., Mini-Forum: roles of Amyloid-beta and Tau Phosphorylation in Neuronal Repair and Protection. J Alzheimers Dis, 2009. 18: p. 369-370. 37. Bilotta, F., et al., Postoperative cognitive dysfunction: toward the Alzheimer's disease pathomechanism hypothesis. J Alzheimers Dis, 2010. 22 Suppl 3: p. 81-9. 38. Sprung, J., et al., Anesthesia and incident dementia: a population-based, nested, case-control study. Mayo Clin Proc, 2013. 88(6): p. 552-61. 39. Avidan, M.S. and A.S. Evers, Review of clinical evidence for persistent cognitive decline or incident dementia attributable to surgery or general anesthesia. J Alzheimers Dis, 2011. 24(2): p. 201-16. 40. Fodale, V., et al., Anaesthetics and postoperative cognitive dysfunction: a pathological mechanism mimicking Alzheimer's disease. Anaesthesia, 2010. 65(4): p. 388-95. 41. van Harten, A.E., T.W. Scheeren, and A.R. Absalom, A review of postoperative cognitive dysfunction and neuroinflammation associated with cardiac surgery and anaesthesia. Anaesthesia, 2012. 67(3): p. 280-93. 42. Hindman, B.J., Emboli, inflammation, and CNS impairment: an overview. Heart Surg Forum, 2002. 5(3): p. 249-53. 43. Levy, J.H. and K.A. Tanaka, Inflammatory response to cardiopulmonary bypass. Ann Thorac Surg, 2003. 75(2): p. S715-20. 44. Grocott, H.P., et al., Genetic polymorphisms and the risk of stroke after cardiac surgery. Stroke, 2005. 36(9): p. 1854-8. 45. Silbert, B.S., et al., The apolipoprotein E epsilon4 allele is not associated with cognitive dysfunction in cardiac surgery. Ann Thorac Surg, 2008. 86(3): p. 841-7. 46. Bryson, G.L., et al., A prospective cohort study evaluating associations among delirium, postoperative cognitive dysfunction, and apolipoprotein E genotype following open aortic repair. Can J Anaesth, 2011. 58(3): p. 246-55. 47. Girard, T.D., et al., Associations of markers of inflammation and coagulation with delirium during critical illness. Intensive Care Med, 2012. 38(12): p. 1965-73. 48. Inouye, S.K. and L. Ferrucci, Elucidating the pathophysiology of delirium and the interrelationship of delirium and dementia. J Gerontol A Biol Sci Med Sci, 2006. 61(12): p. 1277-80. 49. Whitlock, R.P., et al., Clinical benefit of steroid use in patients undergoing cardiopulmonary bypass: a meta-analysis of randomized trials. Eur Heart J, 2008. 29(21): p. 2592-600. 50. Dieleman, J.M., et al., Intraoperative high-dose dexamethasone for cardiac surgery: a randomized controlled trial. JAMA, 2012. 308(17): p. 1761-7. 51. Kamer, A.R., et al., Meloxicam improves object recognition memory and modulates glial activation after splenectomy in mice. Eur J Anaesthesiol, 2012. 29(7): p. 332-7. 52. Hudetz, J.A. and P.S. Pagel, Neuroprotection by ketamine: a review of the experimental and clinical evidence. J Cardiothorac Vasc Anesth, 2010. 24(1): p. 131-42. 53. Himmelseher, S., E. Pfenninger, and M. Georgieff, The effects of ketamine-isomers on neuronal injury and regeneration in rat hippocampal neurons. Anesth Analg, 1996. 83(3): p. 505-12. 54. Kawasaki, T., et al., Ketamine suppresses proinflammatory cytokine production in human whole blood in vitro. Anesth Analg, 1999. 89(3): p. 665-9. 55. Shapira, Y., A.A. Artru, and A.M. Lam, Ketamine decreases cerebral infarct volume and improves neurological outcome following experimental head trauma in rats. J Neurosurg Anesthesiol, 1992. 4(4): p. 231-40. 56. Beilin, B., et al., Low-dose ketamine affects immune responses in humans during the early postoperative period. Br J Anaesth, 2007. 99(4): p. 522-7. 57. Ho, J.D., et al., Successful Management of Excited Delirium Syndrome with Prehospital Ketamine: Two Case Examples. Prehosp Emerg Care, 2012. 58. Hudetz, J.A., et al., Ketamine attenuates delirium after cardiac surgery with cardiopulmonary bypass. J Cardiothorac Vasc Anesth, 2009. 23(5): p. 651-7. 59. Popp, S.S., et al., Intravenous antiarrhythmic doses of lidocaine increase the survival rate of CA1 neurons and improve cognitive outcome after transient global cerebral ischemia in rats. Neuroscience, 2011. 192: p. 537-49. 60. Mitchell, S.J. and A.F. Merry, Lignocaine: neuro-protective or wishful thinking? J Extra Corpor Technol, 2009. 41(1): p. P37-42. 61. Krenk, L., L.S. Rasmussen, and H. Kehlet, New insights into the pathophysiology of postoperative cognitive dysfunction. Acta Anaesthesiol Scand, 2010. 54(8): p. 951-6. 62. Galvan, V. and D.E. Bredesen, Neurogenesis in the adult brain: implications for Alzheimer's disease. CNS Neurol Disord Drug Targets, 2007. 6(5): p. 303-10. 63. Rodriguez, J.J., et al., Voluntary running and environmental enrichment restores impaired hippocampal neurogenesis in a triple transgenic mouse model of Alzheimer's disease. Curr Alzheimer Res, 2011. 8(7): p. 707-17. 64. Jeong, Y.H., et al., Environmental enrichment compensates for the effects of stress on disease progression in Tg2576 mice, an Alzheimer's disease model. J Neurochem, 2011. 119(6): p. 1282-93. 65. Hendrix, S.B., Measuring clinical progression in MCI and pre-MCI populations: enrichment and optimizing clinical outcomes over time. Alzheimers Res Ther, 2012. 4(4): p. 24.

Page 80

80 66. Puskas, F., et al., Intraoperative hyperglycemia and cognitive decline after CABG. Ann Thorac Surg, 2007. 84(5): p. 1467-73. 67. Grocott, H.P., et al., Postoperative hyperthermia is associated with cognitive dysfunction after coronary artery bypass graft surgery. Stroke, 2002. 33(2): p. 537-41. 68. Moller, J.T., et al., Long-term postoperative cognitive dysfunction in the elderly ISPOCD1 study. ISPOCD investigators. International Study of Post-Operative Cognitive Dysfunction. Lancet, 1998. 351(9106): p. 857-61. 69. Murkin, J.M., et al., Monitoring brain oxygen saturation during coronary bypass surgery: a randomized, prospective study. Anesth Analg, 2007. 104(1): p. 51-8. 70. Tang, L., et al., Reduced cerebral oxygen saturation during thoracic surgery predicts early postoperative cognitive dysfunction. Br J Anaesth, 2012. 108(4): p. 623-9. 71. Rubens, F.D., et al., The cardiotomy trial: a randomized, double-blind study to assess the effect of processing of shed blood during cardiopulmonary bypass on transfusion and neurocognitive function. Circulation, 2007. 116(11 Suppl): p. I89-97. 72. van Dijk, D. and C.J. Kalkman, Why are cerebral microemboli not associated with cognitive decline? Anesth Analg, 2009. 109(4): p. 1006-8. 73. Liu, Y.H., et al., The effects of cardiopulmonary bypass on the number of cerebral microemboli and the incidence of cognitive dysfunction after coronary artery bypass graft surgery. Anesth Analg, 2009. 109(4): p. 1013-22. 74. Zanatta, P., et al., Brain Emboli Distribution and Differentiation During Cardiopulmonary Bypass. J Cardiothorac Vasc Anesth, 2013.

Page 81

81 Neurocognitive/Brain Assessment Monitoring Rare Outcomes in Cardiac Surgery Donald S. Likosky, PhD, USA Associate Professor Section Head, Section of Health Services Research and Quality Department of Cardiac Surgery Center for Healthcare Outcomes and Policy (CHOP) University of Michigan Medical School Ann Arbor, Michigan Overview Perhaps no other setting within health care has witnessed as much scrutiny and improvement than cardiac surgery. It is safe to say that clinical improvement and innovation are the hallmarks of our profession. At the same time, clinicians and investigators have broadened the set of indicators used to assess performance. For example, as mortality rates have precipitously declined for patients undergoing coronary artery bypass grafting (CABG), increased attention has focused on other morbid events, including renal failure and stroke. These outcomes are important to all stakeholders (i.e. patients, providers, or payers). Medical centres, payers, and advocacy groups are increasingly reporting these data (along with benchmarks) on the Internet. In addition, these data are used for internal and external quality assurance activities and tied to reimbursement payments. While attention is easily and predictably paid to these data, many of us struggle with how to address gaps in performance, especially when dealing with such rare events. The objective of this article is to propose an approach towards tackling this paradigm, namely by focusing attention on process measures, rather than the clinical outcomes themselves. Description Example 1: Mortality after Cardiac Surgery Surgical outcomes have undoubtedly improved for patients undergoing coronary artery bypass grafting (CABG) or other surgical procedures. In 1987, the Healthcare Financing Administration (HCFA) originally drew focus on mortality rates following CABG surgery. As a consequence of its reporting, hospitals in northern New England reported differences in mortality rates across 5 hospitals in Maine, Vermont, and New Hampshire (1). Mortality rates in northern New England were 4.3%, although varied from 3.1% to 6.3% across the five medical centres participating in the Northern New England Cardiovascular Disease Study Group (NNECDSG). Interestingly, these differences were not attributed to variability in patient presentation, but rather discrete processes of clinical care. The NNECDSG embarked on a regional mode of death study to identify upstream events associated with mortality in this setting (2). Investigators found the principle reason for mortality in this setting was low cardiac output. In fact, the investigators found that patients with normal ejection fractions were likely being injured as a consequence of surgery. By focusing on processes of care rather than solely on mortality, investigators

Page 82

82 reported a 24% reduction in overall mortality in this setting, and 74 fewer deaths than overall would be expected (3). Centres made a number of modifications to their practices, including: a) the use of checklists by perfusionists, b) development of critical pathway for cardiac surgery, and c) cross-trained support staff. Example 2: Neurologic and Neurobehavioral Injury after Cardiac Surgery Investigators have reported the risk of brain injury subsequent to cardiac surgery for more than 5 decades (4). In 1965, the neurologist Dr. Sidney Gilman described three likely mechanisms producing neurobehavioral injuries, including embolisation, hypoperfusion, and hypotension. Although strokes occur less commonly than neurobehavioral deficits in the setting of CABG surgery, they are associated with increased risk of mortality, resource utilisation, and functional impairment (5,6). Investigators in northern New England undertook a chart review of 388 strokes occurring subsequent to isolated CABG surgery (7). Similar to Gilman, nearly 4 decades previously investigators found that the principle mechanisms producing strokes included embolisation and cerebral hypoperfusion. Importantly, these findings suggested that strokes and neurobehavioral injuries share a common set of etiologies. Take home point 1: Emboli and hypoperfusion are the principle mechanisms producing brain injury, defined as either a neurobehavioral injury or stroke. Certainly there is a shared interest in reducing strokes after cardiac surgery. Nonetheless, efforts are often stymied by the rarity (<2%) of the outcome (8). Over the last several decades, investigators from several research teams have focused attention on the relationship between processes of clinical care and risk of embolisation and cerebral hypoperfusion (9,10). While a systematic review of this literature is beyond the scope of this paper, a listing of seminal papers on the topic has been provided in the Appendix. For decades, the Wake Forest group has focused on the association between embolisation and brain injury. In one of their more seminal studies, Stump and colleagues reported associations between emboli and aortic cross clamping approaches (11). While such an association was provocative, it was perhaps insufficient without identifying whether aortic management strategies (single vs. multiple aortic cross clamps) were independently associated with fewer brain injuries. In a follow-up study, Hammon and colleagues monitored emboli in the carotid artery using the EDAC (Embolus Detection and Classification System; Raleigh, NC) device, and assessed the development of neurobehavioral deficits using a battery of neuropsychological tests. Hammon reported that the use of a multiple occlusion clamp strategy was associated with a significant (p<0.05) reduction, relative to a single aortic cross clamp approach, in the rate of neurobehavioral injuries at 6 months among patients undergoing CABG surgery: 30% vs. 57% (12). Such information affords clinicians with the opportunity for modifying practices to reduce neurologic or neurobehavioral injury in a way not otherwise possible if one strictly focused on stroke (Figure). So how might we leverage these studies to reduce brain injury in our own practices? Unfortunately, most centres do not routinely use emboli detection devices (e.g. EDAC or transcranial Doppler), nor do centres typically monitor patients outside of research protocols for the development of neurobehavioral injury. However, as noted earlier in this manuscript, prior work has related emboli to brain injury and likewise

Page 83

83 emboli to aortic management techniques. As such, those interested in this topic might monitor the surgeon’s aortic management approach as way of assessing efforts to minimise atherosclerotically-mediated brain injury. Other processes of care may be similarly leveraged to reduce other types of emboli (e.g. lipid or gaseous) (13). Take home point 2: Emboli are associated with discrete processes of clinical care in the setting of CABG surgery. These processes of care are modifiable and can be monitored for quality assurance and improvement. Discussion “John, our stroke rate over the last year is too high for our CABG population. We need to address this issue,” says Dr. Smith, Chief of Cardiothoracic Surgery. “What is our stroke rate?” asks John Kline, Chief of Cardiovascular Perfusion. “It is 2.1%. The other hospital we often compete with is reporting a 1.6% stroke rate. It is imperative that we address this issue.” Such a discussion may resonate with many of you, albeit the focus may not be stroke and instead may be mortality or deep sternal wound infection. Irrespective of the topic, the overarching issue relates to how to reduce the incidence of an outcome that may only occur once or twice over a series of 100 cases. Although you may have heard otherwise, statistics do not lie. To show a statistically significant decline from 2.1% to 1.6% in the rate of stroke would require a total of 23,594 subjects! Efforts to improve quality are likely to be more successful when one focuses on process measures. First, process measures are modifiable; clinicians may be motivated to make constructive changes in their surgical or perfusion practice based on sound generalisable knowledge coupled with a reasoned approach. Second, process measures occur more frequently than strokes and other rare outcomes. As such, the effect of changes in processes may be more readily apparent than if one were to solely focus on rare outcomes. For instance, reducing the rate of a multiple clamp approach from a base rate of 80% to a rate of 65% would only require a total of 302 subjects rather than the aforementioned 23,594 subjects if the focus were solely on reducing the stroke rate. I am not arguing that outcomes such as stroke or mortality are not important. They certainly are. However, efforts to reduce their occurrence are likely to be successful and sustainable if we identify, redesign, and track performance related to process measures that are mechanistically related to the outcome of interest.

Page 84

84 References 1. O'Connor GT, Plume SK, Olmstead EM, et al. A regional prospective study of in-hospital mortality associated with coronary artery bypass grafting. The Northern New England Cardiovascular Disease Study Group. JAMA. Aug 14 1991;266(6):803-809. 2. O'Connor GT, Birkmeyer JD, Dacey LJ, et al. Results of a regional study of modes of death associated with coronary artery bypass grafting. Northern New England Cardiovascular Disease Study Group. Ann Thorac Surg. Oct 1998;66(4):1323-1328. 3. O'Connor GT, Plume SK, Olmstead EM, et al. A regional intervention to improve the hospital mortality associated with coronary artery bypass graft surgery. The Northern New England Cardiovascular Disease Study Group. JAMA. Mar 20 1996;275(11):841-846. 4. Gilman S. Cerebral Disorders after Open-Heart Operations. N Engl J Med. Mar 11 1965;272:489-498. 5. Likosky DS, Roth RM, Saykin AJ, Eskey CJ, Ross CS, O'Connor GT. Neurologic injury associated with CABG surgery: outcomes, mechanisms, and opportunities for improvement. Heart Surg Forum. 2004;7(6):E650-662. 6. Roach GW, Kanchuger M, Mangano CM, et al. Adverse cerebral outcomes after coronary bypass surgery. Multicenter Study of Perioperative Ischemia Research Group and the Ischemia Research and Education Foundation Investigators. N Engl J Med. Dec 19 1996;335(25):1857-1863. 7. Likosky DS, Marrin CA, Caplan LR, et al. Determination of etiologic mechanisms of strokes secondary to coronary artery bypass graft surgery. Stroke. Dec 2003;34(12):2830-2834. 8. Charlesworth DC, Likosky DS, Marrin CA, et al. Development and validation of a prediction model for strokes after coronary artery bypass grafting. Ann Thorac Surg. Aug 2003;76(2):436-443. 9. Joshi B, Brady K, Lee J, et al. Impaired autoregulation of cerebral blood flow during rewarming from hypothermic cardiopulmonary bypass and its potential association with stroke. Anesthesia and analgesia. Feb 1 2010;110(2):321-328. 10. Hammon JW, Stump DA, Butterworth JF, et al. Coronary artery bypass grafting with single cross-clamp results in fewer persistent neuropsychological deficits than multiple clamp or off-pump coronary artery bypass grafting. Ann Thorac Surg. Oct 2007;84(4):1174-1178; discussion 1178-1179. 11. Stump DA, Rogers AT, Hammon JW, Newman SP. Cerebral emboli and cognitive outcome after cardiac surgery. J Cardiothorac Vasc Anesth. Jan 1996;10(1):113-118; quiz 118-119. 12. Hammon JW, Stump DA, Butterworth JF, et al. Single crossclamp improves 6-month cognitive outcome in high-risk coronary bypass patients: the effect of reduced aortic manipulation. J Thorac Cardiovasc Surg. Jan 2006;131(1):114-121. 13. Stump DA. Deformable emboli and inflammation: temporary or permanent damage? J Extra Corpor Technol. Dec 2007;39(4):289-290.

Page 85

85 SESSION 8 Tuesday 10:00 - 11:30 Outcomes Gastrointestinal Complications in Cardiac Surgery Sara J. Allen, BHB, MBChB, FANZCA, New Zealand Anaesthetist/Intensivist, Greenlane Department of Cardiothoracic and ORL, Anaesthesia/Cardiovascular ICU Auckland City Hospital, Auckland, New Zealand Introduction Gastrointestinal (GI) complications following cardiac surgery encompass a broad range of pathologies, ranging from minor GI bleeding to fulminant hepatic failure. Whilst the overall incidence of these complications is low, significant morbidity and mortality may occur in association. The diagnosis of GI complications is difficult, due to several factors including altered clinical symptoms and signs in patients, drugs affecting assessment (such as sedatives, neuromuscular blocking agents, analgesics, and immunosuppressants), and underlying patient co-morbidities. The pathogenesis of GI complications is not fully understood, and likely multifactorial. Multiple risk factors for complications have been identified, however, and may enable the use of preventative strategies, prompt early investigation, and allow early identification of complications in the peri-operative period. Early appropriate intervention is critical to optimise outcomes. Incidence The incidence of GI complications is variably reported in studies, ranging from 0.3% to 5.5% [1-8]. Reported associated mortality varies even more widely, from 0.3% to 87% [8-9]. GI complications are heterogenous, however the most common complication reported overall is GI bleeding. Other commonly reported complications include mesenteric ischaemia, pancreatitis, cholecystitis and ileus [3,8]. Intestinal obstruction and perforation, and hepatic dysfunction are less common [8-9]. Rare complications include fulminant hepatic failure, pseudomembraneous colitis, peritonitis and iatrogenic injury to intra-abdominal organs [4-5]. Pathogenesis Ischaemia is thought to be the primary cause of most GI complications, with both hypoperfusion and impaired oxygenation implicated in causing ischaemia [10-11]. Splanchnic hypoperfusion may be caused by reduced or suboptimal cardiac output, regional flow, or systemic mean arterial pressure (MAP). The optimal cardiac output and MAP for splanchnic perfusion remains unclear. A small study of 16 patients comparing normal MAP targets (60-65mmHg) with higher MAP targets (80-85mmHg) in patients undergoing cardiopulmonary bypass (CPB) showed no differences in splanchnic oxygenation, acid-base status, or cytokine production [12].

Page 86

86 Multiple other factors may contribute, including systemic inflammation and the systemic inflammatory response syndrome (SIRS), release of inflammatory mediators, non-pulsatile blood flow, hypothermia, drug therapy, and mechanical factors [10, 13, 14]. Systemic inflammation and the SIRS response occur due to surgical stress response, contact with the CPB circuit, and ischaemia itself (which may activate and sustain SIRS), along with reperfusion injury. The inflammatory and complement cascades release mediators such as thromboxane A2 and B2, leukotrienes and C5a, which all have vasoconstrictor actions. Cytokine activation is implicated in vascular endothelial dysfunction and damage [10]. All these factors contribute to maldistribution of blood flow, and impaired mucosal oxygen delivery. Non-pulsatile blood flow causes renin release and activation of the renin-angiotensin-aldosterone axis, with secretion of angiotensin II – a potent vasoconstrictor. Of note, the splanchnic circulation is unable to auto regulate perfusion at extremes of pressure or flow, and therefore is vulnerable to alterations during CPB, and post bypass haemorrhage, hypovolaemia, or arrhythmia. Hypothermia is associated with vasoconstriction and altered regional blood flow and distribution. Drug therapies such as noradrenaline and vasopressin may also be associated with hypoperfusion [1, 8]. Mechanical factors that may contribute to ischaemia and hypoperfusion include micro and macro emboli, due to air, atheroma, thrombus, or debris, and hepatic and GI congestion related to venous cannulae placement. Another proposed mechanism of hypoperfusion is sympathetic nervous system activation (as occurs in the stress response, but may be prolonged or sustained by factors such as prolonged mechanical ventilation) [15]. Non-ischaemic mechanisms of GI complications include bacterial translocation (due to altered mucosal barriers and blood flow), adverse drug reactions (e.g. over-anticoagulation, amiodarone-induced hepatotoxicity) [16], pre-existing pathology, and iatrogenic organ injury (e.g. malpositioned surgical drains). Risk Factors Patients with comorbidities and those with a prolonged or complicated post-operative course are most likely to develop GI complications. Studies have variably reported risk factors, but those consistently identified may be classified as pre, intra or post-operative. Pre-operative risk factors include advanced age (>70 years), reoperation, chronic renal failure, peripheral vascular disease, diabetes mellitus, chronic obstructive respiratory disease, gastrointestinal disease pre-existing, congestive heart failure (NYHA class III or IV), low cardiac output state, and use of inotropic support or intra-aortic balloon pump (IABP) [5, 7, 8, 13]. Intra-operative factors include prolonged CPB duration, valvular surgery, emergency surgery, increased blood transfusion, use of IABP, presence of arrhythmias. Post-operative factors associated are prolonged mechanical ventilation, acute kidney injury, deep sternal wound infection, post-operative low cardiac output state [4, 5, 8, 11, 15, 17, 18]. Several studies have found no difference in the incidence or mortality of GI complications in on-pump versus off-pump surgery [19,20]. The Gastrointestinal Complication Score (GICS) is a risk score model specific for GI complications following cardiac surgery. Developed using prospectively collected data from 5593 patients in a single centre undergoing cardiac surgical procedures, the model was subsequently validated on a further 1031 single centre cardiac surgery patients. Receiver operating curves (ROC) were used to assess the score’s predictive ability, with a ROC area under curve in the validation group of 0.83. The

Page 87

87 model uses the following risk factors in calculating a score: age >80 years, active smoker, pre-operative inotropic support, NYHA class III-IV symptoms, CPB duration >150 min, post-operative atrial fibrillation, post-operative heart failure, reoperation due to bleeding, and post-operative vascular complication. In the validation study, the probability of a GI complication at a GICS 15 or above was >20%, whilst at a GICS of 5 or below, was only <0.4% [19]. Prevention Pre-operative risk stratification with scoring systems, or risk factor identification, may allow preventative strategies to be used pre- and intra-operatively, as well as prompt earlier investigation, diagnosis and management of complications post-operatively. It is likely that early detection and intervention can improve both morbidity and mortality related to GI complications. Pre-operative optimisation of haemodynamic state with correction of hypovolaemia, and anaemia, and optimisation of cardiac output (e.g. inotrope therapy or IABP if required) is recommended. However, there are currently no large randomised controlled trials to validate this approach. Institutional practise is variable in the management of pre-operative anaemia, with recent recommendations to use iron supplementation and erythropoietin in selected elective patients [22], while in some institutions, pre-operative transfusion is considered in those requiring non-elective surgery. Intra-operative monitoring and maintenance of adequate cardiac output and oxygenation is clearly important, however as previously outlined, the exact parameters for adequate cardiac output and oxygen delivery are unknown, and likely vary between patients. Mucosal ischaemia and altered barrier function may occur despite adequate global flow and oxygen delivery. Several methods for monitoring GI perfusion, including measurement of gastric pH, ultrasound of blood flow in hepatic or mesenteric vessels, measurement of intestinal transport functions, measurement of trans-splanchnic changes in Il-6, Il-10, pH and lactate have been described [12, 23]. These methods are impractical for routine use and interpretation in the operating room, and are not used clinically. Specific intra-operative strategies are discussed below. Drug Therapies Several drug therapies have been associated with benefit in reducing GI complications, but findings have not been consistent, and evidence from large randomised controlled trials (RCTs) is still lacking. Aspirin treatment within 48 hours postoperatively has been associated with a reduction in both the incidence and mortality of GI complications in coronary artery bypass graft (CABG) surgery [24]. Milrinone infusion in CABG patients resulted in reduced gastric mucosal acidosis and lower inflammatory marker and endotoxin levels in a small RCT [25], but clinically significant effects have yet to be demonstrated. Dopamine and dobutamine can both increase cardiac output, however neither have shown clear benefit, and recent studies suggest potential harm [26-28].

Page 88

88 Vasopressin has consistently been demonstrated to have adverse effects on gastric mucosal perfusion [28]. Modification of CPB Multiple strategies have been proposed, focussing on maintaining adequate perfusion, avoiding haemodilution and severe anaemia, the use of pulsatile flow, the use of filters and strategies to reduce emboli and finally, avoidance of CPB with off-pump surgery. Maintenance of adequate cardiac output and oxygenation seems prudent, but as previously discussed, the thresholds for adequacy may be difficult to measure and are difficult to define for any individual patient. Minimising the use of pure vasoconstrictors is recommended, with augmentation via inotropes suggested if support for MAP targets is required. Haemodilution resulting in a haematocrit <0.25 is associated with reduced oxygen delivery, and in some studies with increased mortality [29]. The use of smaller bypass circuit components, and retrograde autologous priming may help with blood conservation and reduction of anaemia and transfusion requirement, and is a reasonable approach [22]. Debate has been extensive over the role of pulsatile versus non-pulsatile flow in CPB. In some trials, pulsatile flow has resulted in improved mucosal oxygenation and perfusion, but in others, no differences were demonstrated [30, 31]. Clinical outcomes with pulsatile versus non-pulsatile flow have not differed. Minimising the risk of emboli and subsequent hypoperfusion by careful selection of cannulation site, echo assessment of the aorta, avoidance of excess manipulation of the aorta, avoidance of IABP in those with severe atheroma, and de-airing techniques, along with CPB filters is warranted, but all these strategies remain unproven in reducing GI complications. CPB circuit filters have been demonstrated to reduce micro and macro gaseous emboli and atheroemboli, but have not been shown to reduce the incidence of GI complications. Off-pump CABG surgery avoids the CPB circuit entirely, and is associated with reduced bleeding complications, but not a reduction in GI complications [28]. Other unproven strategies include biocompatible surfaces in the CPB circuit, minimisation of circuit surface area and volume and reduction of blood-air interfaces. Other strategies Pharmacotherapy with gastric acid suppression using a proton pump inhibitor is recommended to reduce the risk of peptic and duodenal ulcers and GI bleeding, and has been shown to be superior to no prophylaxis, and to use of a histamine receptor antagonist in preventing these complications [32]. Diagnosis In the cardiac surgery patient, sedatives, analgesics, neuromuscular blockade, and vasoactive drugs may alter clinical symptoms and signs of GI complications. Multi-

Page 89

89 organ failure, metabolic derangement, and cardiovascular instability are non-specific signs – and may make clinical assessment even more difficult. Clinical presentation varies with pathology, and no single diagnostic test will reliably diagnose or exclude all intra-abdominal pathology, and investigation should be directed by patient history and presentation. Overall, a low threshold for investigation in those patients with non-routine post-operative progress is recommended. Initial investigations will include biochemistry and haematology blood analysis, and may be followed by abdominal radiography, ultrasound or computed tomography (CT) scanning, upper and lower endoscopy, and diagnostic laparotomy as indicated. Delayed diagnosis and management is associated with worse outcome [5, 8]. The routine diagnostic tests for the most common GI complications are summarised in the table (1.1) below. Table 1.1 Incidence, presentation, and investigations for GI complications [8,12,18,28] Haemorrhage 30-35% of GI complications Upper GIT: Duodenal or gastric ulceration Lower GIT:Diverticulitis, AV malformations Altered blood/melaena per rectum Haematochezia Haemodynamic instability Shock Hb  LDH Endoscopy Mesenteric ischaemia 14-20% complications Occlusive: emboli or thrombus Non-occlusive (NOMI): hypoperfusion Abdominal pain and distension Intolerance enteral nutrition GI bleeding Leucocytosis Lactic acidosis AXR CT scan abdomen CT mesenteric angiography Colonoscopy Laparotomy Peptic ulcer perforation 6-8% complications Abdominal pain, distension Peritonism Shock AXR CT scan abdomen Pancreatitis 0.5 – 1.0% of all patients NB: hyperamylasaemia common (25-35% of patients post cardiac surgery) Epigastric and back pain Nausea and vomiting Abdominal distension SIRS Shock Amylase Lipase CT scan abdomen Cholecystitis 6-11% of complications Calculous or acalculous Often 10-15 days post-surgery Right upper quadrant pain Fever Leucocytosis SIRS Shock LFTs deranged Ultrasound biliary tract CT abdomen Laparoscopy Hepatic dysfunction Transient up to 40% patients Hepatic failure <0.4% of patients May be asymptomatic Jaundice LFTs Abdominal ultrasound (exclude obstruction, thrombosis, collections) Screen for underlying cause (hepatitis)

Page 90

90 Management Determined by the specific condition, management is usually identical to standard management for the specific condition. It is important to emphasise that treatment should not be withheld due to recent cardiac surgery. The routine management for the most common GI complications is summarised in the table below (Table 1.2) Table 1.2 Management of GI complications [8, 28] Haemorrhage Resuscitation Correction of coagulopathy Consider proton pump inhibitors Endoscopy Clipping or sclerotherapy to bleeding vessels Angiography and Intervention/Surgery If lower GIT bleeding with AV malformation or diverticulitis Mesenteric ischaemia Resuscitation Circulatory support Antibiotic therapy Laparotomy and resection Peptic ulcer perforation Resuscitation Proton pump inhibitor high dose Laparotomy – vagotomy and oversew, or resection Pancreatitis Enteral rest and NG drainage Resuscitation Post pyloric feeding or intravenous nutrition Supportive therapy Analgesia Occasionally percutaneous drainage or surgical treatment (rare) Cholecystitis Surgery (calculous) with cholecystectomy Antibiotics +/- percutaneous drainage (acalculous) Hepatic dysfunction Stop potential hepatotoxins Supportive Summary GI complications are an uncommon occurrence following cardiac surgery, but with a high burden of morbidity and mortality. The diagnosis remains difficult, commonly leading to delay in definitive treatment, and preventative strategies are largely unproven currently. In the future, further large high-quality trials are needed to establish efficacy of preventative strategies. Ideally, the development of more sensitive and specific diagnostic tests will allow earlier definitive treatment – and improved outcomes..

Page 91

91 References 1. McSweeney ME, Garwood S, Levin J, et al. Adverse gastrointestinal complications after cardiopulmonary bypass: Can outcome be predicted from preoperative risk factors? Anesth Analg 2004;98:1610-1617. 2. Mangi AA, Christison-Lagay ER, Torchiana DF, et al. Gastrointestinal complications in patients undergoing heart operation. An analysis of 8709 consecutive cardiac surgical patients. Ann Surg 2005;241(6):895-901. 3. Dong G, Liu C, Xu B, et al. Postoperative abdominal complications after cardiopulmonary bypass. J Cardiothorac Surg 2012;7:108 4. Tsiotos GG, Mullany CJ, Zietlow S, van Heerden JA. Abdominal complications following cardiac surgery. Am J Surg 1994;167:553-557. 5. Huddy SPJ, Joyce WP, Pepper JR. Gastrointestinal complications in 4473 patients who underwent cardiopulmonary bypass surgery. Br J Surg 1991;78:293-296. 6. Ohri SK, Desai JB, Gaer JAR, et al. Intra-abdominal complications following cardiopulmonary bypass. Ann Thorac Surg 1991;52:826-31. 7. Yilmaz AT, Arslan M, Demirkilic U, et al. Gastrointestinal complications after cardiac surgery. Eur J Cardio-thorac Surg 1996;10:763-767. 8. Rodriguez R, Robich MP, Plate JF, Trooskin SZ, Sellke FW. Gastrointestinal complications following cardiac surgery: A comprehensive review. J Card Surg 2010;25:188-197. 9. Croome KP, Kiaii B, Fox S, Quantz M, McKenzie N, Novcik RJ. Comparison of gastrointestinal complications in on-pump versus off-pump coronary artery bypass grafting. Can J Surg 2009;52(2):125-128. 10. Ohri SK, Velissaris T. Gastrointestinal dysfunction following cardiac surgery. Perfusion 2006;21:215-223. 11. Moneta GL, Misbach GA, Ivey TD. Hypoperfusion as a possible risk factor in the development of gastrointestinal complications after cardiac surgery. Am J Surg 1985;149:648-650. 12. McNicol L, Lipcsey M, Bellomo R, et al. Pilot alternating treatment design study of the splanchnic metabolic effects of two mean arterial pressure targets during cardiopulmonary bypass. Br J Anaesth 2013;110(5):721-728. 13. Sakorafas GH, Tsiotos GG. Intra-abdominal complications after cardiac surgery. Eur J Surg 1999;165:820-827. 14. Asimakopoulas G, Taylor KM. The effect of cardiopulmonary bypass on neutrophil and endothelial adhesion molecules. 15. D’Ancona G, Baillot R, Poirier B, et al. Determinants of gastrointestinal complications in cardiac surgery. Tex Heart J 2003;30:280-85. 16. Ratz Bravo AE, Drewe J, Schlienger RG, Krahenbuhl S, Pargger H, Ummenhofer W. Hepatotoxicity during rapid intravenous loading with amiodarone: description of three cases and review of the literature. Crit Care Med 2005;33:128-134. 17. Perugini RA, Orr RK, Porter D, Dumas EM, Maini BS. Gastrointestinal complications after cardiac surgery. An analysis of 1477 cardiac surgery patients. Arch Surg 1997;132:352-357. 18. Andersson B, Nilsson J, Brandt J, Hoglund P, Andersson R. Gastrointestinal complications after cardiac surgery. Br J Surg 2005;92:326-333. 19. Musleh GS, Patel NC, Grayson AD, et al. Off-pump coronary artery bypass surgery does not reduce gastrointestinal complications. Eur J Cardiothorac Surg 2003;23:170-174. 20. Berson AJ, Smith JM, Woods SE, et al. Off-pump versus on-pump coronary artery bypass surgery: Does the pump influence outcome? J Am Coll Surg 2004;199:102-108. 21. Andersson B, Andersson R, Brandt J, Hoglund P, Algotsson L, Nilsson J. Gastrointestinal complications after cardiac surgery – improved risk stratification using a new scoring model. Interact Thorac Cardiovasc Surg 2010;10:366-370. 22. Menkis AH, Martin J, Cheng DC, et al. Drug, devices, technologies, and techniques for blood management in minimally invasive and conventional cardiothoracic surgery: a consensus statement from the International Society for Minimally Invasive Cardiothoracic Surgery (ISMICS) 2011. Innovations 2012;7(4):229-41. 23. Ott MJ, Buchman TG, Baumgartner WA. Postoperative abdominal complications in cardiopulmonary bypass patients: a case-controlled study. Ann Thorac Surg 1995;59:1210-3. 24. Mangano DT. Aspirin and mortality from coronary bypass surgery. NEJM 2002;347:1309-1317. 25. Mollhoff T, Loick HM, van Aken H, et al. Milrinone modulates endotoxaemia systemic inflammation and subsequent acute phase response after cardiopulmonary bypass. Anesthesiology 1999;90:72-80. 26. Parviarnen I, Ruokonen E, Takala J. Dobutamine-induced dissociation between changes in splanchnic blood flow and gastric intra-mucosal pH after cardiac surgery. Br J Anaes 1995;54:277-282. 27. Uusaro A, Ruokonen E, Takala J. Splanchnic oxygen transport after cardiac surgery: Evidence for inadequate tissue perfusion after stabilization of hemodynamics. Int Care Med 1996;22:26-33. 28. Hessel EA. Abdominal organ injury after cardiac surgery. Semin Cardiothorac Anesth 2004;8:243-263. 29. De Foe GR, Ross CS, Olmstead EM, et al. Lowest haematocrit in bypass and adverse outcomes associated with coronary artery bypass grafting. Ann Thorac Surg 2001;71:769-776. 30. Gaer JAR, Shaw ADS, Wild R, et al. Effect of cardiopulmonary bypass on gastrointestinal perfusion and function. Ann Thorac Surg 1994;57:371-375. 31. Ohri SK, Bowles CW, Mathie RT, et al. Effect of cardiopulmonary bypass perfusion protocols on gut tissue oxygenation and blood flow. Ann Thorac Surg 1997;64:163-170. 32. Patel AJ, Som R. What is the optimum prophylaxis against gastrointestinal haemorrhage for patients undergoing adult cardiac surgery: histamine receptor antagonists, or proton-pump inhibitors? Interact Cardiovasc Thorac Surg 2013;16(3):356-60.

Page 92

Page 93

93 Outcomes Prevention of Lung Injury in Cardiac Surgery: a Review Robert Young, FANZCA, Australia Specialist Anaesthetist, Flinders Medical Centre, Adelaide, Australia Introduction Patients are increasingly presenting for cardiac surgery at an advanced age. Mechanical lung function deteriorates over time and the elderly will be less tolerant of respiratory insult. In addition, pre-existing lung disease may be more advanced. Fortunately, perioperative management is evolving all the time and technological improvements have provided many new opportunities for the advancement in the care of cardiac surgical patients. This review contains a summary of recent interventions and changes of practice that may reduce lung injury following cardiac surgery together with a review of general aspects of perioperative management and how they can exacerbate such injury. Incidence of respiratory failure Lung injury is an inevitable consequence of cardiac surgery. It is predominantly inflammatory in nature [5]. The degree of injury may be sufficient to cause respiratory failure leading to a prolonged period of assisted ventilation in the post-operative period. A small minority of patients will develop adult respiratory distress syndrome (ARDS) with its attendant risk of death. Filsoufi and colleagues retrospectively analysed the New York State Department of Health database in order to determine the incidence of respiratory failure following cardiac surgery [6]. The database included 5798 patients who underwent cardiac surgery between January 1998 and December 2005. Respiratory failure was defined as pulmonary insufficiency requiring intubation and ventilation for 72 hours or more. The overall incidence of respiratory failure was 9.1%. The mortality rate amongst those patients who developed respiratory failure was 15.5% compared to 2.4% in those who did not develop respiratory failure. ARDS is a rare but catastrophic complication of cardiac surgery. It is characterised by extreme inflammatory damage to the lung parenchyma. It may occur as an isolated lung pathology or, more commonly, as one facet of a multiorgan failure. Asimakopoulos and colleagues reviewed 2464 cases of cardiac surgery performed at the Hammersmith Hospital in London, England between 1993 and 1997 [7]. They identified 12 cases (0.5%) of ARDS, of whom 11 (92%) died. All those patients who died had multiorgan failure. The survivor had isolated ARDS only. Milot and colleagues undertook a similar review of 3278 cases of cardiac surgery performed at their institution in Quebec, Canada between 1995 and 1998 [8]. The incidence of ARDS was similar at 0.4% but with a mortality of only 15%. This represented two deaths of which one had multiorgan failure and the other isolated ARDS. The overall mortality in the non-ARDS cases was 3.5%.

Page 94

94 The inflammatory process The inflammatory response is triggered by a number of factors during cardiac surgery. These include the exposure of blood to artificial surfaces during cardiopulmonary bypass (CPB), the recirculation of spilt blood via cardiotomy suction, endotoxaemia due to splanchnic hypoperfusion, ischaemia-reperfusion of the lung, surgical tissue trauma and the use of protamine. The effects of inflammatory lung damage can be further exacerbated by inappropriate oxygenation, ventilation, and fluid administration. The inflammatory response is a highly complex and dynamic one. Our understanding of it is incomplete. Key features early in the process are the generation of pro-inflammatory cytokines, and the activation of the complement system. This leads, amongst other things to the attraction, sequestration and activation of neutrophils within the pulmonary circulation [9]. Pulmonary endothelial cells are stimulated to express adhesion molecules on their luminal surface. Neutrophils bind to and migrate across the pulmonary endothelium. Activated neutrophils secrete both reactive oxygen species and various proteinases leading to structural damage and the influx of protein rich fluid into the alveolar airspaces. The lungs are susceptible to inflammatory damage during cardiac surgery for a number of reasons. First, the lungs are the only organ to receive 50% of the total cardiac output. Secondly, the pulmonary capillaries have a diameter of only 2 to 15 micrometres. Neutrophils become deformed as they pass through and overall transit time is greatly increased [9]. Finally, the lungs are the only organs, with the exception of the heart to suffer a significant ischaemia-reperfusion injury around the period of CPB [10]. Making Sense of Research Findings Attenuation of the inflammatory response to cardiac surgery and the protection of the lungs and other vital organs has excited much interest and led to the publication of a large number of studies. Studies of the effect of an intervention on the inflammatory process are often difficult to interpret. The factors assayed vary widely and the implication of a change in absolute levels is not clear. Indeed, it may be that changes in relative concentrations of the different pro- and anti-inflammatory mediators are of more consequence. Meta-analysis of studies is often hindered by the high degree of heterogeneity of the included studies. An attempt has been made to better focus the research effort by the recent publication of recommendations for study design in this field of investigation [11]. Clinical outcomes are obviously of more interest to those involved in perioperative patient management. Parameters most often measured are the oxygenation or respiratory index, calculated as the fraction of inspired oxygen divided by the arterial partial pressure (FiO2/PaO2), or pulmonary shunt fraction. In many cases, these measurements are only taken in the first few hours following surgery and the impact on functional recovery is not apparent. A better marker of effect is the duration of post-operative ventilation although it should be borne in mind that this can be affected by many factors which may cause bias. The duration of post-operative ventilation is likely to impact on the cost of surgery and thus the economic benefit of a given intervention.

Page 95

95 Finally, lung protection does not occur in isolation. The benefits or otherwise of a given intervention with regards to lung function must be assessed in the context of overall effect in other areas such as cardiac, neurological and renal function. The Impact of Cardiopulmonary Bypass The inflammatory response to cardiac surgery is driven by several factors. A key element, and one that is amenable to intervention, is the exposure of circulating blood to the artificial surfaces of the cardiopulmonary bypass circuit. This exposure leads to complement activation, triggering of the coagulation cascade, and activation of neutrophils and platelets [12]. Attempts have been made to attenuate the inflammatory response either by changing the nature of the materials within the CPB circuit, by reducing the total surface area of the circuit that is in contact with blood or by avoiding CPB altogether. Biocompatible cardiopulmonary bypass circuits Biocompatible circuits have been available for over 20 years. Early circuits were heparin bonded. Over time a variety of different molecules have been used including poly2-methoxyethylacrylate, polyethylene oxide chains and phosphorylcholine. Numerous studies have demonstrated a beneficial effect of such circuits in terms of attenuation of the inflammatory response [13, 14]. The clinical benefits including preservation of lung function have been less clear. Ranucci and colleagues undertook a meta-analysis of trials looking at various clinical outcomes with biocompatible circuits [15]. All included trials were prospective and randomised. Trials with paediatric patients were excluded. Thirty-six trials were included with a total number of 4360 patients. 78% of trials used heparin bonded circuits. Only outcome variables reported in at least 8 studies were analysed. These included the incidence of post-operative lung dysfunction as defined by the authors and mechanical ventilation time. Unfortunately, the latter measure was later excluded from the analysis due to the presence of publication bias. Analysis demonstrated a reduction in red cell transfusion, incidence of atrial fibrillation and shorter ICU stay. There was no difference in the incidence of lung dysfunction. Ranucci and colleagues assessed the quality of the studies included in their meta-analysis using the Jadad score. This is a widely accepted method, which is used to assess trials based on the nature of patient randomisation, blinding and the description of patient exclusions and withdrawals [16]. Scores range from 0-5. In a subgroup analysis of high quality studies (Jadad score 3 or greater), the difference in red cell transfusion rates was not significant. The authors pointed out a number of limitations to the review. First, a great majority of the studies involved heparin bonded circuits and so little can be surmised with regard to other types of biocompatible circuits. Secondly, the investigators were unable to stratify studies based on preoperative risk, and so no assessment could be made of respiratory outcomes in patients with pre-existing lung disease. There is currently no compelling evidence that the use of biocompatible bypass circuits reduces lung injury following cardiac surgery. Further investigation is required into the newer biocompatible circuits, particularly with regard to higher risk patients with pre-existing respiratory disease.

Page 96

96 Miniaturised Extracorporeal Circulation (MECC) In recent years there has been much interest in miniaturising cardiopulmonary bypass circuits in order to reduce the area of interface between blood and foreign surfaces and thus attenuate the inflammatory response. A variety of circuit configurations have been developed. Common features include greatly reduced tubing length, the use of biocompatible surfaces, removal of the venous reservoir and cardiotomy suction, and the use of centrifugal pumps rather than roller pumps. The blood air interface is eradicated. Circuit priming volumes are greatly reduced, being typically 450-500 mls. A majority of studies have demonstrated a reduction in various inflammatory mediators with the use of MECC when compared to conventional extracorporeal circuits (CECC) [17]. In a prospective study by Mazzei and colleagues, 300 patients were randomised to receive coronary revascularisation using either MECC or off pump surgery. Release of inflammatory markers was similar in the two groups [18]. There have been 4 meta-analyses of studies of clinical outcomes with the use of MECC when compared with CECC. The most recent, by Anastasiadis included 24 studies comprising a total of 2770 patients [19]. All studies were randomised and contained at least 40 patients in each treatment group (MECC versus CECC). Use of MECC was associated with a reduced duration of mechanical ventilation. Other significant findings were a reduction in red cell transfusion and overall mortality. Munos and colleagues compared 4 different strategies in the management of very high risk patients [20]. 214 patients with a EuroSCORE of greater than 9 were randomised to undergo coronary artery bypass grafting with CECC, MECC, off pump surgery or MECC with a beating heart. 42% of patients had COPD, dispersed evenly across the 4 treatment groups. Time to extubation was significantly reduced in the MECC with beating heart group when compared to the other three groups. The authors noted that their study was retrospective, non-randomised and with relatively small numbers in each group. The use of MECC is exciting much interest at present and results from trials to date are encouraging. Its use represents a significant change in practice and a much greater body of evidence is required before it gains mainstream acceptance. Off Pump Coronary Artery Bypass (OPCAB) Surgery The avoidance of cardiopulmonary bypass should, intuitively, lead to a reduction in the overall inflammatory response to cardiac surgery. Indeed, a number of studies have demonstrated significant reductions in levels of pro-inflammatory cytokine levels, activated complement, leucocyte numbers and neutrophil elastase production [21]. In an early study, Kochamba and colleagues investigated the effects of OPCAB on lung function [22]. Fifty-eight low risk patients undergoing 2-vessel bypass surgery were randomised to either OPCAB or conventional CPB. Investigators found no statistical difference in post-operative gas exchange immediately post-operatively. Pulmonary shunt was significantly less in the OPCAB group. There was no difference in the duration of post-operative ventilation. A number of further small studies in low risk patients have failed to show any significant benefit with regards to respiratory function in the immediate post-operative period. [23, 24].

Page 97

97 In the Surgical Management of Arterial Revascularization Therapies (SMART) trial, Staton and colleagues randomised 197 patients to receive on or off pump coronary artery bypass surgery [25]. Patients were not excluded on the basis of preoperative morbidities. Randomisation was stratified for both gender and diabetes. Arterial partial pressure of oxygen, measured immediately post-operatively with an inspired oxygen fraction (FiO2) of 1.0 was significantly higher in the OPCAB group. Significantly more patients in the OPCAB group were extubated within 4 hours of the completion of surgery. (73/100 versus 50/97, p 0.002). Post- operative spirometry, reintubation rates and readmissions for respiratory complications were not different between the two groups. Meharwal and colleagues reviewed 1075 patients undergoing OPCAB who were categorised as being high risk [26]. Inclusion criteria were age over 70 years, left main coronary artery stenosis, acute myocardial infarction, redo surgery or a left ventricular ejection fraction of less than 30%. The comparison group consisted of 2312 patients undergoing on-pump surgery over the same 5-year period (October 1996 – June 2001). Intubation time was significantly lower in the OPCAB group (19 +/- 5 hours versus 24 +/- 6 hours, p < 0.001). Incidence of prolonged post-operative ventilation, defined as greater than 48 hours was also lower (4.6% versus 7.6%, p = 0.002). Moller and colleagues performed a prospective randomised trial – the Best Bypass Surgery Trial [27]. The trial included 341 patients with EuroSCORE's of ≥ 5 and 3 vessel coronary disease, randomised to on-pump or off-pump surgery. There was no significant difference in time to extubation, incidence of prolonged ventilation, defined as greater than 24 hours, or the incidence of pneumonia. Patients with chronic lung disease (CLD) are a high-risk group for coronary artery surgery. Kerendi and colleagues undertook a retrospective analysis of 7060 patients undergoing isolated CABG in a single institution between 2002 and 2007, with a particular focus on patients with chronic lung disease [28]. Overall, CLD was associated with a greater incidence of prolonged ventilation, pneumonia and mortality. Those patients with CLD who underwent off-pump surgery had a significantly reduced incidence of all these parameters when compared with those undergoing on-pump surgery. The benefits of OPCAB in elderly patients (> 80 years) have also been studied. LaPar and colleagues undertook a retrospective analysis of 1993 patients undergoing surgery at one of 16 centres between 2003 and 2008 [29]. The incidence of prolonged ventilation, defined as longer than 24 hours, was significantly lower in the OPCAB group (11.4% V 14.7%,. P = 0.05). By contrast, Sarin and colleagues, in a retrospective analysis of 937 patients over 80 years of age, found no difference in the duration of post-operative ventilation, although 30-day mortality was significantly reduced [30]. A recently published meta-analysis of trials of OPCAB in octogenarians showed that respiratory failure requiring ventilation lasting over 24 hours was 30% less likely with OPCAB and lasting over 48 hours, 70% less likely [31]. Ultrafiltration Conventional ultrafiltration (CUF) during cardiopulmonary bypass may benefit the lungs in several ways. Removal of plasma water will maintain or improve plasma oncotic pressure, preventing interstitial and alveolar oedema [32].

Page 98

98 Haemoconcentration with the retention of plasma clotting factors has been shown to reduce post-operative blood transfusion requirements thus avoiding the associated increase in pulmonary morbidity [33]. A number of studies have demonstrated a reduction in various proinflammatory cytokines with the use of ultrafiltration [34, 35]. It is not clear, however, whether these reductions are clinically significant. Zero-balance ultrafiltration (Z-BUF) is a technique whereby blood is filtered and an equal volume of crystalloid or colloid containing physiological concentrations of various electrolytes returned to the circulation. Any benefit in terms of lung protection would be accrued by a reduction in inflammatory mediators rather than plasma volume control. A recent meta-analysis of trials of Z-BUF showed no benefit in terms of duration of post-operative ventilation or length of ICU stay in adult patients [36]. Modified ultrafiltration (MUF) is a technique used predominantly in paediatric cardiac surgery. Ultrafiltration occurs immediately after the cessation of cardiopulmonary bypass typically for a period of 20 minutes or until a target haematocrit is achieved. In a study of forty neonates undergoing primary biventricular operative repair, patients were randomised to receive either CUF or CUF with MUF. Various measures of lung function were taken at serial points throughout the surgery and up to 6 hours postoperatively. The addition of MUF led to a significant initial improvement in lung compliance and gas exchange, but this improvement was not sustained at 6 hours [37]. A recent meta-analysis of trials comparing CUF directly with MUF in paediatric practice found no difference in post-operative ventilator time [38]. A large study of MUF in 573 consecutive adult cardiac surgical cases demonstrated a significantly lower incidence of respiratory insufficiency, defined as a PaO2 < 60 mmHg or PaCO2 > 50 mmHg on room air, in the group receiving MUF (11/284, 3.9% v 20/289, 6.9%, p= 0.005). The difference in time on assisted ventilation between the 2 groups did not reach statistically significance [39]. Auto-transfusion Devices Autotransfusion of shed blood is now routinely used in many centres [40]. Blood is collected via cardiotomy suction, the red blood cells are separated by centrifugation and resuspended in 0.9% saline prior to transfusion. Residual circuit blood can be processed in the same way. The putative benefit is the preservation of autologous blood and the reduction in the requirement for allogenic blood transfusion. A recent meta-analysis of 31 randomised controlled trials demonstrated a decrease in the odds of exposure to red blood cells of 40%, and to any blood product of 37% [41]. The benefit in terms of lung function lies in the avoidance of allogenic blood [42]. There may be an additional benefit in the net removal of fluid during processing. Shed pericardial and mediastinal blood is exposed to traumatised tissue leading to triggering of both the coagulation and inflammatory cascades. Cardiotomy suction blood has been shown to contain elevated levels of various inflammatory mediators such as TNFα, IL-6 and IL-8 [43], as well as debris such as fat and air microemboli [44]. The washing of cardiotomy suction blood has been demonstrated to significantly reduce levels of various proinflammatory cytokines in shed blood [45]. In addition, cell salvage may favourably alter the balance of remaining pro- and anti-

Page 99

99 inflammatory factors in the processed blood [46]. Damgaard and colleagues randomised 29 patients undergoing CABG to receive processed or unprocessed blood from both cardiotomy suction and the residual circuit volume [47]. They demonstrated significant reductions in the levels of proinflammatory cytokines IL-6 and IL8 at 6 hours post CPB. These differences were lost at 24 hours. Activated neutrophils play a key role in lung injury. There is significant variation between devices with regard to the efficiency of leukocyte removal. Serrick and colleagues compared 5 different autotransfusion devices and found that leukocyte removal rates ranged between 30 and 78% [48]. A previous study by Perttila and colleagues demonstrated an increase in the proportion of neutrophils in the processed blood but no evidence of neutrophil activation. In a clinical trial, Boodhwani and colleagues randomised patients undergoing CABG or aortic valve surgery to receive processed or unprocessed shed blood during surgery [49]. Various parameters of lung function were measured before and immediately after cardiopulmonary bypass and 2 hours later. They found no difference between the 2 groups in mechanical function or gas exchange. The meta-analysis of trials of intraoperative cell salvage by Wang and colleagues showed no difference in post-operative ventilation time or ICU length of stay [41]. There is evidence that processing of salvaged blood reduces levels of various proinflammatory mediators. In addition levels of anti-inflammatory proteins are also altered. There is variable removal of leukocytes, platelets and debris, depending on the device used. However, there is no evidence of clinical benefit with regard to lung function. Leukocyte Depletion Activated leukocytes play a key role in inflammatory lung damage [50]. Leukocyte filters have been added to both the arterial and venous sides of the CPB circuit in an attempt to ameliorate the inflammatory response. Filters have also been used to process salvaged and residual circuit blood. More recently blood added to the cardioplegia solution has been filtered with the aim of reducing myocardial reperfusion injury. The use of leukocyte filters in cardiac surgery has been widely investigated. Study end points have included change in leukocyte counts, changes in levels of substances released from activated leukocytes such as neutrophil elastase, and changes in other pro-inflammatory proteins. In addition, there have been a few clinical trials of pulmonary function with the use of leukocyte filters. Warren and colleagues published an excellent review in 2007 of the effects of the various leukocyte filtration strategies used in cardiac surgery [51]. The reviewers identified twenty-six studies in which pre and post-operative white cell counts were recorded, with and without the use of leukocyte filtration. Of these, there was no difference in fifteen studies. In the remainder the reduction in white cell count was short lived, tending to disappear post CPB. A great majority of studies into plasma levels of various cytokines, showed no difference with systemic leukofiltration. Neutrophil elastase is an enzyme released by activated neutrophils. Six studies were identified by Warren and colleagues, in which neutrophil elastase levels were assayed. In three of these no difference was found with leukofiltration. In the remaining three, elastase levels were elevated with leukofiltration, leading to the

Page 100

100 suggestion that trapped neutrophils released increased levels of elastase into the circulation. Another approach is to assay markers of lung injury such as exhaled nitric oxide (NO) [52]. Two studies to date have been published looking at the effects of arterial line leukocyte filters on exhaled NO levels [53, 54]. A number of the investigators were involved in both studies. As a result the study designs were similar. Patients with pre-existing lung disease were excluded. Leucocyte filters were added to the arterial side of the CPB circuit. Exhaled NO was measured before and 30 minutes after CPB. In both studies, the increase in exhaled NO was significantly reduced in the filtered groups. The significance of this attenuation of NO production at one time point immediately post CPB is not clear. Finally, changes in clinical parameters of lung function such as respiratory index and duration of assisted ventilation postoperatively with leukofiltration have been assessed, in a number of trials. Generally, numbers of patients in these trials have been small and they have been of poor quality. Results are conflicting. The meta-analysis by Warren and colleagues, mentioned above, included a review 21 studies with a total of 996 patients, in which clinical measures of lung function were made. Analysis demonstrated an improvement in oxygenation index up to 12 hours post-operatively in the leukofiltration treated patients. In addition, ventilation time was significantly reduced. In a subgroup analysis of trials with greater than 25 patients per group (7 from 21 trials) no benefit was demonstrated with leukofiltration. Warren and colleagues analysed the quality of trials using the Jaded score [55]. The average Jaded score for the trials in this meta-analysis was 1.43 with only 8 trials scoring greater than 2. Analysis of these 8 trials revealed no benefit with leukofiltration either in terms of oxygenation index within the first 12 hours or the duration of post-operative ventilation. When trials including patients at high risk of post-operative respiratory complications (5 from 21) were analysed, a significant reduction in ventilation time was demonstrated. None of these trials were in the high quality group (Jadad score >2) or the group with greater than 25 patients per study group. There have been a small number of relevant trials published since Warren’s meta-analysis. Of note, Bechtel and colleagues under took the largest investigation to date [56]. The study was a retrospective one. During a 10-week period leukocyte filters were used in the arterial side of the CPB circuit and in the cardioplegia delivery line for all patients undergoing cardiac surgery at their institution. The total number of patients was 266. The control group was constructed of an equal number of patients who underwent surgery immediately before or after this period. The 2 groups were well matched for multiple characteristics including history of smoking and presence of chronic lung disease. The number of patients extubated within 12 hours was significantly greater in the leukofiltered group, [182(68.4%) versus 159(59.8%)]. The incidence of reintubation and pneumonia was similar in the 2 groups. Leukofiltration would, in theory, be an effective way to reduce lung injury in cardiac surgery, given what is known about the generation and effects of activated neutrophils in this setting. However, the studies to date are disappointing. There is a high degree of heterogeneity in studies with variation in the positioning of filters and the timing and duration of their use. Clinical trials have generally been small and poorly designed. It would seem that three factors are important in maximising any potential benefit from leukodepletion. First there needs to be further investigation into the optimal

Page 101

101 deployment of these filters for maximum effect. Secondly, attention should be paid to patient selection, as benefit may only be significant in those with pre-existing lung disease. Finally, filter design is likely to improve with time leading to greater efficiency and capacity. This may lead to improved clinical outcomes in the future. Ventilation Inappropriate positive pressure ventilation can lead to overdistension of the alveoli (volutrauma), over pressurisation (barotrauma), and repeated collapse and reinflation (atelactotrauma). Shear stress leads to the generation of cytokines and activation of neutrophils. Lung injury has been confirmed by microscopy in animal studies [57]. Ventilator-induced lung injury is now a well-recognised phenomenon. Studies of ventilation strategies in critically ill patients without lung injury, have demonstrated that larger tidal volumes lead to increased levels of proinflammatory cytokines in both bronchoalveolar lavage fluid and blood together with a significantly higher incidence of the development of lung injury [58, 59]. In the landmark paper produced by the Acute Respiratory Distress Syndrome Network (ARDSNet) in 2000, patients with existing lung injury were randomised to receive initial tidal volumes of either 12 ml/kg of predicted body titrated to maintain plateau airway pressure below 50 cm H20 or 6 ml/kg with a plateau pressure at or below 30 cmH20 [60]. The study was terminated early due to recognition of a significantly lower mortality in the low tidal volume group. The number of days without ventilator use in the first 28 days after randomisation, was also significantly lower in this group. The lungs of patients undergoing cardiac surgery are already at risk due to the inevitable systemic inflammatory response. Suboptimal ventilation may exacerbate this injury. Zupancich and colleagues compared a strategy of high tidal volumes (10-12ml/kg) and low positive end expiratory pressure -PEEP (2-3 cm H2O), with one of a low tidal volume (6-8 ml/kg) high PEEP (10 cm H2O). Inflammatory cytokines in blood and bronchoalveolar lavage fluid were significantly higher in the high tidal volume group at 6 hours post separation from CPB [61]. A similar study by Wrigge and colleagues found no difference in levels of inflammatory cytokines at 6 hours postoperatively when tidal volumes of 6 or 12 ml/kg were employed [62]. It should be noted that in the latter study the different ventilation strategies were only employed once the patient reached the ICU and not immediately upon recommencing ventilation after separation from CPB, as was the case in the Zupancich study. Chaney and colleagues performed a small prospective randomised trial of ventilation with different tidal volumes (6 ml/kg versus 12 ml/kg) in low risk patients undergoing CABG. Various parameters of lung function were measured at 60 minutes after arrival in the ICU and compared with measurements taken soon after initial intubation [63]. Both groups were ventilated with a FiO2 of 1.0 and PEEP of 5 cm H2O). The group with the higher tidal volume demonstrated significantly greater increases in peak airway and plateau pressures. The decrease in lung compliance was significantly greater in the high tidal volume group. Significant decreases in static lung compliance and increases in shunt occurred in the high tidal volume group but not the low tidal volume group. One of the concerns associated with using low tidal volumes is the tendency towards alveolar collapse or atelectasis. Significant levels of atelectasis are likely during cardiac surgery due to inflammatory changes leading to reduced pulmonary

Page 102

102 surfactant release and the period of absence of ventilation during CPB. The use of alveolar recruitment manoeuvres, that is to say the application of a brief period of high continuous airway pressure in an attempt to reopen collapsed alveoli, has been shown to reduce the levels of inflammatory cytokine interleukin-8 in the post-operative period [64]. Celebi and colleagues demonstrated improved oxygenation and lung compliance, together with reduced radiographic evidence of atelectasis when alveolar recruitment manoeuvres were applied soon after arrival of study patients in the ICU. The improvement was significant up to 4 hours post intervention. The period of absence of ventilation during CPB may be harmful to the lungs. Absence of ventilation leads to lung collapse and increased resistance to blood flow through the bronchial arteries. This reduction in flow is likely to cause ischaemic damage. Okuda and colleagues subjected rat lungs to 90 minutes of ischemia and measured lung oedema after 60 minutes of reperfusion. Continuous ventilation of the lungs during the period of ischaemia led to a significant reduction in lung oedema. This reduction was seen when ventilating with 21% oxygen or 100% nitrogen [65]. Imura and colleagues subjected pigs to cardiopulmonary bypass. Those pigs that were ventilated at low tidal volumes during CPB were better oxygenated with lower alveolar-arterial oxygen gradients and less histological evidence of lung injury [66]. A recent meta-analysis of studies of different lung protective strategies during CPB reported some benefit in oxygenation immediately post bypass with the use of CPAP or alveolar recruitment manoeuvres [67]. However, there was no evidence for sustained benefit. The authors reported that the overall quality of studies was poor and that there was a high degree of heterogeneity with regards to study protocols and measured endpoints. There are currently few trials of ventilation during CPB. John and colleagues randomised 23 low risk patients undergoing coronary artery bypass grafting to receive either ventilation or no ventilation during CPB. Ventilated patients had tidal volumes of 5 ml/kg [68]. The FiO2 was not stated in the paper. Extravascular lung water post-bypass was significantly reduced, as was time to extubation. There was no significant difference in lung compliance or alveolar-arterial oxygen difference. Gagnon and colleagues found no statistically significant improvement in oxygenation, pulmonary vascular resistance or length of hospital stay when patients were ventilated with tidal volumes of 3 ml/kg during CPB [69]. The authors noted that 5 patients in the ventilated group received blood transfusion during surgery. No patients in the control group were transfused. Lung Perfusion during CPB During cardiopulmonary bypass pulmonary blood flow is arrested. The lungs are perfused via the bronchial arteries. Schlensak and colleagues have demonstrated in animal studies that bronchial blood flow is greatly reduced during CPB [71, 72]. This finding raises the possibility of ischaemia-reperfusion injury to the lungs. This is supported by initial animal studies [73] and later investigations in patients undergoing cardiac surgery [74]. Serraf and colleagues, using neonatal piglets demonstrated that supplemental low flow pulmonary perfusion could attenuate the lung injury associated with CPB [75]. Sieppe and colleagues undertook a similar study. Pigs were randomised to a control group with no pulmonary perfusion, or to pulsatile or non-pulsatile pulmonary

Page 103

103 perfusion during 120 minutes of CPB. Following 120 minutes of reperfusion various pro-inflammatory proteins were assayed. They demonstrated a reduction in levels with pulmonary perfusion, which was generally more marked in the pulsatile perfusion group. In a clinical trial Suzuki and colleagues randomised thirty infants undergoing open cardiac surgery to standard CPB or CPB with continuous pulmonary perfusion [76]. The perfusion group received pulmonary perfusion with oxygenated blood at 30 ml/kg/min during the period of CPB. Patients were electively ventilated for 24 hours post-operatively. The oxygenation index was significantly higher in the perfusion group at all-time points up to 24 hours. The duration of ventilator support was significantly shorter in the perfusion group (67.2 hours +/- 13.8 versus 183.8 hours +/- 56.5, p= 0.049). Santini and colleagues randomised thirty low risk patients undergoing CABG to either conventional CPB or CPB with pulsatile pulmonary perfusion [77]. Oxygenated blood was infused via a cannula into the pulmonary artery and drained from the left atrium. A pulsatile pump was used at a rate of 60 beats per minute. Various measures of lung function were made on admission to ICU, at 3 hours post-surgery and following extubation. Alveolar-arterial oxygen gradient, oxygenation index and lung compliance were significantly better in the perfusion group across all 3 time points. Mean pulmonary arterial pressures and pulmonary vascular resistance were significantly lower in the perfusion group. Analysis of bronchoalveolar lavage samples collected on admission to ICU and 4 hours later showed a reduced absolute white cell count and lower percentage of neutrophils in the perfusion group. An alternative to continuous pulmonary perfusion is single shot pulmonary perfusion with a protective solution at the commencement of CPB. In a study on dogs, Liu and colleagues perfused the right lung with Ringer’s lactate solution or a protective solution containing glucose-insulin-potassium, l-arginine, aprotinin, anisodamine and phosphate buffer [78]. Both solutions were cooled to 4oc. Oxygen content of blood from the right pulmonary vein was significantly higher in the intervention group. Sequestration of white cells was reduced in the intervention group, as were levels of malondialdehyde, a marker of oxygen radical-mediated cell injury. Histological examination revealed normal lung parenchyma in the right lung of the intervention group. There was marked alveolar oedema and abundant neutrophils in the left lungs of both groups. The right lungs of the controls exhibited moderate alveolar haemorrhage and interstitial vessel congestion. Sievers and colleagues investigated the effects of single shot pulmonary perfusion at the commencement of CPB in adult patients undergoing scheduled coronary bypass or aortic valve surgery [79]. Study numbers were small, and no indication of pre-operative risk was given. Lung perfusion consisted of arterial blood cooled to 15oc at 1 l/min for 10 minutes. The lungs were ventilated for the duration of perfusion. Alpha2 macroglobulin levels increased significantly in bronchoalveolar lavage (BAL) samples from the control group but not in the perfused group. Alpha 2 macroglobulin is a large plasma protein and elevated BAL levels are taken to indicate increased leakiness of the capillary endothelium. The alveolar-arterial oxygenation gradient was significantly lower in the perfused lungs. The oxygenation index did not differ significantly. Drew and Anderson first described the use of a separate pulmonary extracorporeal circulation with ongoing ventilation in 1959 [80]. Using this system the lungs continue their function as physiological oxygenators removing the requirement for a

Page 104

104 mechanical oxygenator in the systemic bypass circuit. Richter and colleagues randomised 30 low risk patients undergoing CABG to receive conventional CPB or bilateral extracorporeal circulation as described by Drew and Anderson [81]. Peak concentrations of pro-inflammatory cytokines IL-6 and IL-8 were significantly lower in the Drew-Anderson group. Oxygenation index was significantly higher in the Drew-Anderson group at 30 minutes and 2 hours post-bypass. The difference was lost at 4 hours. A similar pattern, favouring the Drew-Anderson group, was demonstrated for the alveolar-arterial oxygen gradient. Shunt fraction was significantly greater in the standard CPB group at 30 minutes post CPB. This difference was lost at 4 hours. Time to extubation was significantly shorter in the Drew-Anderson group (5.2 hours +/- 0.4 versus 9.5 hours +/- 1.2, p = 0.0025). Oxygenation A high inspired fraction of oxygen (FiO2) leads to alveolar collapse with the extent of collapse being related to both the FiO2 and duration of exposure [82]. There is radiological evidence of basal atelectasis after only brief exposure to 100% oxygen [83]. The effect is magnified in lungs already made susceptible by the inflammatory response to CPB [84]. Santos and colleagues assessed the effects of ventilation with 100% oxygen in 8 patients with acute lung injury [85]. There was a significant increase in intrapulmonary shunt at 30 and 60 minutes. Reber and colleagues randomised 20 low risk cardiac surgical patients to receive either 100% or 35% oxygen following separation from CPB [86]. Venous admixture, a measure of pulmonary shunt, was significantly increased in the hyperoxic group when compared to the level prior to CPB. There was no such difference in the control group. Hyperoxia causes increased production of reactive oxygen species (ROS) by mitochondria [87]. Hyperoxia also increases pulmonary sequestration and activation of neutrophils with further ROS production [88]. Hyperoxia has also been shown to worsen ischaemia-reperfusion injury of the lung [89], and to exacerbate lung injury caused by high tidal volume ventilation [90]. In the setting of cardiac surgery, hyperoxia is likely to exacerbate inflammatory lung injury. Pizov and colleagues investigated the effects of high oxygen concentration on CPB induced lung injury [91]. Thirty patients were randomised to receive either 100% or 50% oxygen throughout the course of their surgery. Whilst on bypass, the lungs of patients in the 100% oxygen group were flushed with oxygen at 4 litres per minute whilst those in the 50% oxygen group had air insufflated at 4 litres per minute. The investigators demonstrated significantly elevated levels of pro-inflammatory cytokine TNFα in the BAL fluid of the hyperoxic group following CPB. Levels of IL-8 in BAL fluid were not elevated in either group. There was no significant difference in the elevation of plasma cytokine levels between the two groups. Oxygenation index was reduced in both groups post CPB. However, values had returned to pre-CPB levels by 6 hours post-operatively in the 50% group but not in the 100% group. Transfusion of blood products Transfusion related acute lung injury (TRALI) is defined as the acute onset of hypoxia with the presence of bilateral pulmonary infiltrates, occurring within hours of a blood transfusion [92]. The lung injury is thought to be due to the activation of

Page 105

105 neutrophils already sequestered within the lungs either by anti-neutrophil antibodies in the donor blood or biologically active substances such as lipids or cytokines [93]. Vlaar and colleagues, in a recent prospective study of patients undergoing cardiac surgery with CPB [94]. Of 668 patients studied, 16 developed TRALI. The mortality rate in these patients was 13%. Risk factors for the development of TRALI were patient age, duration of CPB, total amount age of blood products transfused, age of transfused red cells, total amount of plasma transfused, and presence of antibodies in donor plasma. Koch and colleagues reviewed 16,847 cases of on pump cardiac surgery undertaken over an eight-year period at the Cleveland Clinic, USA [95]. Patients who received transfusion of red cells or plasma had a significantly increased incidence of respiratory distress, respiratory failure, and ARDS. Intubation times were significantly longer, and the incidence of reintubation higher. Leal-Noval and colleagues investigated the impact of blood transfusion on the development of severe postoperative infection (SPI) in patients following cardiac surgery [96]. The investigators identified 70 cases from a cohort of 738 patients. Of the patients with SPI, 63% had nosocomial pneumonia. Transfusion of greater than 4 units of blood components was identified as a risk factor for the development of SPI. The overall mortality in those patients with SPI was 52.8% compared to 8.2% in non-SPI patients. The requirement for allogenic blood in cardiac surgical patients can be minimised by the use of various techniques for the preservation of blood intra-operatively. These techniques include cell salvage and autologous transfusion, the use of ultrafiltration, and miniaturised extracorporeal circuits. In addition, any intervention that attenuates the inflammatory response to cardiac surgery may also reduce the incidence of coagulopathy and post-operative blood loss. Further interventions include the identification and treatment of elective patients with pre-operative anaemia, the avoidance of excessive haemodilution by the injudicious use of intravenous fluids, and scrupulous surgical technique. Finally, appropriate triggers for transfusion should be used [97, 98]. Fluid Management It has long been recognised that administration of excessive volumes of intravenous crystalloid solutions leads to the development of interstitial oedema. More recently, there has been work to indicate that the administration of colloidal solutions to euvolaemic patients also leads to interstitial oedema [99]. The lungs are rendered highly susceptible to oedema during cardiac surgery due to increased capillary permeability, elevated pulmonary arterial vascular resistance and reduced myocardial diastolic function following reperfusion. The administration of excessive volumes of fluid can be avoided by using goal directed therapy. Fluids are given based on the measurement of a parameter that has been demonstrated to accurately indicate fluid responsiveness in the form of an increase in cardiac output. Numerous parameters have been assessed. Static measures of pressure such as central venous pressure (CVP) and pulmonary artery occlusion pressure (PAOP) have been shown repeatedly to be of no use in this regard [100,101]. Dynamic interactions between intrathoracic pressures during positive pressure ventilation and left ventricular output, such as pulse pressure

Page 106

106 variation and stroke volume variation accurately reflect fluid responsiveness [102,103]. The use of transoesophageal echocardiography during cardiac surgery is now widespread. Static measurements such as left ventricular end diastolic area (LVEDA) reflect cardiac preload. However, these measurements do not take into account changes in ventricular compliance. It is, therefore, not surprising that studies into LVEDA measurement as a predictor of fluid responsiveness have produced mixed results [103,104]. Possibly the most sensitive echocardiographic parameter for the prediction of fluid responsiveness is the superior vena cava (SVC) collapsibility index. Collapse of the SVC by more than 36% has been shown to predict fluid responsiveness with 90% sensitivity and 100% specificity [105]. Haemodilution of the blood at the commencement of CPB leads to a reduction in colloid osmotic pressure leading to pulmonary oedema. This effect can be attenuated either by the use of colloids in the bypass prime [106] or by the use of retrograde autologous priming [107]. Pharmacological Interventions In recent years, a large number of drugs have been shown to alter the inflammatory response during cardiac surgery. These include phosphodiesterase inhibitors [108], ketamine [109], aminophylline [110], free radical scavengers such as n-acetylcysteine [111], and antioxidants [112]. However, adequately powered clinical trials demonstrating a beneficial effect on post-operative lung function are few. Corticosteroids are a notable exception. Two recent meta-analyses of trials investigating the clinical benefits of steroids have demonstrated no reduction in the duration of post-operative ventilation [113,114]. In contrast, Dieleman and colleagues recently published the results of a large multicentre, randomised double-blind placebo controlled trial [115]. 4494 patients were randomised to receive high dose dexamethasone (1 mg/kg) or placebo intraoperatively. 10.9% of patients in the dexamethasone group and 11.9% of the placebo group were being treated for chronic lung disease prior to the surgery. The duration of mechanical ventilation was significantly shorter in the dexamethasone group. The incidence of post-operative pneumonia was also significantly reduced. Respiratory failure, defined as a period of ventilation for an uninterrupted period of greater than 48 hours, occurred in significantly fewer cases in the dexamethasone group (67 {3%} versus 97{4.3%}, p=0.02). Summary and Future Directions There has been much recent interest in the attenuation of the inflammatory response to cardiac surgery and the resulting reduction in lung injury. Studies are many but definitive answers currently few. The avoidance of allogenic blood transfusion is clearly beneficial. Scrupulous perioperative management of oxygenation, ventilation and intravenous fluids is also important. The evidence of benefit with interventions such as the use of biocompatible circuits, leukocyte filters and cell salvage is currently weak. However ongoing improvements in technology may lead to more obvious benefit.

Page 107

107 Off-pump cardiac artery bypass surgery has been associated with a reduction in inflammatory lung injury and it has become the procedure of choice for high-risk patients in many centres. However it has not been universally embraced and recent investigations have failed to show a benefit at 30 days [116]. In addition, the recently published Randomised On/Off Bypass (ROOBY) trial demonstrated less effective revascularisation in those patients randomised to off-pump surgery [117]. The adverse cardiac event rate at one year was also significantly greater (16.4% versus 5.9%, p < 0.001). There is now interest in alternative strategies to OPCAB in high-risk patients, such as CPB with concurrent ventilation or bilateral perfusion with or without ventilation. There are currently few trials, and the body of evidence will need to increase substantially before these approaches are adopted. It is clear that the greatest benefit from the various interventions discussed in this review will accrue in those patients at most risk of pulmonary morbidity. Identifying these patients is therefore important. Kor and colleagues recently published an evaluation of a scoring system for the identification of patients at risk of acute lung injury following surgery - the surgical lung injury prediction (SLIP) score [118]. Important surgical predictors included high-risk cardiac, thoracic and vascular procedures. High-risk cardiac procedures included CABG, valve replacement, multiple valve repairs, aortic repair, congenital heart repair, pericardial resection, transplantation and any reoperation. Significant comorbidities included COPD, gastro-oesophageal reflux disease and diabetes mellitus. The single modifying factor found to be significant, was alcohol abuse. The score is calculated out of a possible total of 101 with high-risk patients (SLIP score >27) having a combined risk of ALI of 12.2%. Risk estimates will only be of use if the data that is included is accurate. Adabag and colleagues performed pulmonary function tests in 1169 patients undergoing cardiac surgery at the Minneapolis Veterans Affairs Medical Centre between 2000 and 2007 [119]. They defined airway obstruction as a forced expiratory volume in 1 second to forced vital capacity ratio (FEV1: FVC) of < 0.7. 483 patients had a recorded history of COPD. Of these, 178 had no airway obstruction on testing. 186 patients without a diagnosis of COPD had evidence of airway obstruction. When the investigators analysed outcomes they found that the mortality risk was ten times higher in those patients with moderate to severe airway obstruction together with a diffusion capacity of less than 50% of predicted value. The documented causes of death were varied and included multi-organ failure, cardiogenic shock, respiratory failure, stroke and infection. Finally there is emerging evidence of genetically determined variation in susceptibility to lung injury. Dodd-o and colleagues demonstrated strain-specific differences in sensitivity to ischaemia reperfusion lung injury in mice [120]. The investigators found that multiple genes were involved and that differences were not explained by differences in the production of reactive oxygen species. Chen and colleagues demonstrated a link between polymorphisms of the gene coding for interleukin 18 and the risk of developing an acute lung injury following CPB surgery [121]. Over time, genetic screening may become an additional method for the identification of those at high risk of lung injury following cardiac surgery.

Page 108

108 References 1. Boyle E, Pohlman T, Johnson M, et al. Endothelial cell injury in cardiovascular surgery: the systemic inflammatory response. Ann Thorac Surg 1997; 63: 277-284. 2. Filsoufi F, Rahmanian P, Castillo J, et al. Predictors and early and late outcomes of respiratory failure in contemporary cardiac surgery. Chest 2008; 133: 713-21. 3. Asimakopoulos G, Taylor K, Smith P, et al. Prevalence of acute respiratory distress syndrome after cardiac surgery. J Thorac Cardiovasc Surg 1999; 68(3): 1107-15. 4. Milot J, Perron J, Lacasse Y, et al. Incidence and predictors of ARDS in cardiac surgery. Chest 2001; 119(3): 884-8. 5. Boyle E, Pohlman T, Johnson M, et al. Endothelial cell injury in cardiovascular surgery: the systemic inflammatory response. Ann Thorac Surg 1997; 63: 277-284. 6. Filsoufi F, Rahmanian P, Castillo J, et al. Predictors and early and late outcomes of respiratory failure in contemporary cardiac surgery. Chest 2008; 133: 713-21. 7. Asimakopoulos G, Taylor K, Smith P, et al. Prevalence of acute respiratory distress syndrome after cardiac surgery. J Thorac Cardiovasc Surg 1999; 68(3): 1107-15. 8. Milot J, Perron J, Lacasse Y, et al. Incidence and predictors of ARDS in cardiac surgery. Chest 2001; 119(3): 884-8. 9. Grommes J, Soehnlein O. Contribution of neutrophils to acute lung injury. Mol Med 2010; 17(3-4): 293-307. 10. Hengst W, Gielis J, Lin J, et al. Lung ischaemia-reperfusion injury: a molecular and clinical view on a complex pathophysiological process. Am J Physiol Heart Circ Physiol 2010; 299: H1283-H1299. 11. Landis C, Murkin R, Stump D, et al. Consensus statement: minimal criteria for the reporting of the systemic inflammatory response to cardiopulmonary bypass. Heart Surg Forum 2010; 13(2): E108A – 15A 12. Butler J, Rocker G, Westaby S. Inflammatory response to cardiopulmonary bypass. Ann Thorac Surg 1993; 55: 552-9. 13. Wendel H, Ziemer G. Coating –techniques to improve the hemocompatibility of artificial devices used for extracorporeal circulation. Eur J Cardiothorac Surg. 1999; 16: 342-50. 14. Sohn N, Marcoux J, Mycyk T, et al. The impact of different biocompatible coated cardiopulmonary bypass circuits on inflammatory response and oxidative stress. Perfusion 2009; 24(4): 231-7. 15. Ranucci M, Balduini A, Ditta A et al. A systematic review of biocompatible cardiopulmonary bypass circuits and clinical outcome. Ann Thorac Surg 2009; 87: 1311-9. 16. Jadad A, Moore R, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: Is blinding necessary? Controlled Clinical Trials 1996; 17(1) 1-12. 17. Hall R. Identification of inflammatory mediators and their modulation of strategies for the management of the systemic inflammatory response during cardiac surgery. J Cardiothorac Vasc anaes 2012, published online December 23rd. 18. Mazzei V, Nasso G, Salamone G, et al. Prospective randomized comparison of coronary bypass grafting with minimal extracorporeal circulation system (MECC) versus off pump coronary surgery. Circulation 2007; 116: 1761-7. 19. Anastasiadis K, Antonitsis P, Haidich A, et al. Use of minimal extracorporeal circulation improves outcome after heart surgery; a systemic review and meta analysis of randomized controlled trials. Int J Cardiol. 2013; 164: 158-69. 20. Munos E, Calderon J, Pillois X et al. Beating-heart coronary artery bypass surgery with the help of mini extracorporeal circulation for very high-risk patients. Perfusion 2011; 26(2): 123-31. 21. Raja S, Dreyfus G. Current status of off-pump coronary artery surgery. Asian Cardiovasc Thorac Ann 2008; 16: 164-78. 22. Kochamba G, Yun L, Pfeffer A, et al. Pulmonary abnormalities after coronary arterial bypass grafting operation: cardiopulmonary bypass versus mechanical stabilization. Ann Thorac Surg 2000; 69: 1466-70. 23. Vedin J, Jensen U, Ericsson A, et al. Pulmonary hemodynamics and gas exchange in off pump coronary artery bypass grafting. Interact Cardiovasc Thorac Surg. 2005; 4(5): 493-7. 24. Cimen S, Ozkul V, Ketenci B, et al. Daily comparison of respiratory functions between on-pump and off-pump patients undergoing CABG. Eur J Cardiothorac Surg 2003; 23: 589-94. 25. Staton G, Williams W, Mahoney E, et al. Pulmonary outcomes of off-pump vs on-pump coronary artery bypass surgery in a randomized trial. Chest 2005; 127: 892-901. 26. Meharwal Z, Mishra Y, Kohli V, et al. Off-pump multivessel coronary artery surgery in high-risk patients. Ann Thorac Surg 2002; 74: S1353-7. 27. Møller C, Perko M, Lund J, et al. No major differences in 30-day outcomes in high-risk patients randomized to off-pump versus on-pump coronary bypass surgery. The Best Bypass Surgery Trial. Circulation 2010; 121: 498-504. 28. Kerendi F, Halkos M, Puskas J, et al. Impact of off-pump coronary artery bypass graft surgery on postoperative pulmonary complications in patients with chronic lung disease. Ann Thorac Surg 2011; 91: 8-15. 29. LaPar D, Bhamidipati C, Reece T, et al. Is off-pump coronary artery bypass grafting superior to conventional bypass in octogenarians? J Thorac Cardiovasc Surg 2011; 141: 81-90. 30. Sarin E, Kayatta M, Kilgo P, et al. Short- and long-term outcomes in octogenarian patients undergoing off-pump coronary artery bypass grafting compared with on-pump coronary artery bypass grafting. Innovations: Technology and Techniques in cardiothoracic and vascular Surgery 2011; 6(2): 110-5. 31. Pawlaczyk R, Swietlik D, Lango R, et al. Off-pump coronary surgery may reduce stroke, respiratory failure and mortality in Octogenarians. Ann Thorac Surg 2012; 94: 29-37 32. Magilligan D, Oyama C. Ultrafiltration during cardiopulmonary bypass: Laboratory evaluation and initial clinical experience. Ann Thorac Surg 1984; 37: 33-39. 33. Boodhwani M, Williams K, Babaev A, et al. Ultrafiltration reduces blood transfusions following cardiac surgery: a meta-analysis. EJCTS 2006; 30: 892-897. 34. Journais D, Israel-Biet D, Pouard P, et al. High-volume, zero- balanced haemofiltration to reduce delayed inflammatory response to cardiopulmonary bypass in children. Anesthesiology 1996; 85: 965-976. 35. Berdat P, Eichenberger E, Ebell J, et al. Elimination of proinflammatory cytokines in pediatric cardiac surgery: analysis of ultrafiltration method and filter type. J Thorac Cardiovasc Surg 2004; 127(6): 1688-96. 36. Zhu X, Wang G, Long C. The effects of zero-balance ultrafiltration on postoperative recovery after cardiopulmonary bypass: a meta-analysis of randomized controlled trials. Perfusion 2012; 27(5): 386-392. 37. Mahmoud A, Burhani M, Hannef A, et al. Effect of modified ultrafiltration on pulmonary function after cardiopulmonary bypass. Chest 2005; 128(5): 3447-3453.

Page 109

109 38. Kuratani N, Bunsangjaroen P, Srimueang T, et al. Modified versus conventional ultrafiltration in paediatric cardiac surgery: A meta-analysis of randomized controlled trials comparing clinical outcome parameters. JTCVS 2011; 142(4): 861-867. 39. Luciani G, Menon T, Vecchi B, et al. Modified ultrafiltration after adult cardiac operations. Circulation 2001; 104[suppl I]: I-253 – I-259. 40. Belway D, Rubens F, Wozny D, et al. Are we doing everything we can to conserve blood during bypass? A national survey. Perfusion2005; 20: 237-241. 41. Wang G, Bainbridge D, Martin J, et al. The efficacy of an intraoperative cell saver during cardiac surgery: a meta-analysis of randomized trials. Anesth Analg 2009; 109: 320-330. 42. Koch C, Li L, Figueroa P, et al. Transfusion and pulmonary morbidity after cardiac surgery. Ann Thorac Surg 2009; 88: 1410-8. 43. Westerberg A, Bengtsson A, Jeppsson A. Coronary surgery without cardiac suction and autotransfusion reduces the postoperative systemic inflammatory response. Ann Thorac Surg 2004; 78; 54-59. 44. Solis R, Noon G, Beall A, et al. Particulate microembolism during cardiac operation. Ann Thorac Surg. 1974; 17: 332-44. 45. Amand T, Pincemail J, Blaffart F, et al. Levels of inflammatory markers in the blood processed by autotransfusion devices during cardiac surgery associated with cardiopulmonary bypass circuit. Perfusion 2002; 17(2): 117-23. 46. Gäbel J, Westerberg M, Bengtsson A, et al. Cell salvage of cardiotomy suction blood improves the balance between pro- and anti-inflammatory cytokines after cardiac surgery. Eur J Cardiothorac Surg, published online Feb 12th 2013. 47. Damgaard S, Nielson C, Andersen L, et al. Cell saver for on-pump coronary operations reduces systemic inflammatory markers: a randomized trial. Ann Thorac Surg 2010; 89(5): 1511-7. 48. Serrick C, Scholz M, Melo A, et al. Quality of Rd Blood Cells Using Autotransfusion devices: a comparative analysis. JECT 2003; 35: 28-34. 49. Boodhwani M, Nathan H, Mesana T, et al. effects of shed mediastinal blood on cardiovascular and pulmonary function: A randomized, double-blind study. Ann Thorac Surg 2008; 86: 1167-74. 50. Grommes J, Soehnlein O. Contribution of neutrophils to acute lung injury. Mol Med 2011; 17(3-4): 293-307. 51. Warren O, Alexiou C, Massey R, et al. The effects of various leukocyte filtration strategies in cardiac surgery. Eur J Cardiothorac Surg. 2007; 31: 665-76. 52. Kharitonov S, Barnes P. Nitric oxide in exhaled air is a new marker of airway inflammation. Arch Chest Dis. 1996; 51:533-7. 53. Alexiou C, Tang A, Sheppard S, et al. A prospective randomized study to evaluate the effect of leukodepletion on the rate of alveolar production of exhaled nitric oxide during cardiopulmonary bypass. Ann Thorac Surg 2004; 78: 2139-2145. 54. Sheppard S, Gibbs R, Smith D, et al. Does the use of leucocyte depletion during cardiopulmonary bypass affect exhaled nitric oxide production. Perfusion 2004; 19: 7-10. 55. Warren O, Tunnicliffe C, Massey R, et al. Systemic Leukofiltration does not attenuate pulmonary injury after cardiopulmonary bypass. ASAIO Journal 2008; 54:78-88. 56. Bechtel J, Muhlenbein S, Eichler W, et al. Leukocyte depletion during cardiopulmonary bypass in routine adult cardiac surgery. Interact Cardiovasc and Thorac Surg. 2011; 12: 207-212. 57. Halter J, Steinberg J, Gatto L, et al. Effect of positive end-expiratory pressure and tidal volume on lung injury induced by alveolar instability. Critical Care 2007; 11: R20. 58. De Oliveira R, Hetzel M, Silva M et al. Mechanical ventilation with high tidal volume induces inflammation in patients without lung disease. Crit Care 2010; 14: R39. 59. Determann R, Royakkers A, Wolthuis E, et al. Ventilation with lower tidal volumes as compared with conventional tidal volumes for patients without acute lung injury: a preventative randomized controlled trial. Crit Care 2010; 14:R1. 60. The Acute Respiratory Distress Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342: 1301-8. 61. Zupancich E, Paparella D, Turani F, et al. Mechanical ventilation affects inflammatory mediators in patients undergoing cardiopulmonary bypass for cardiac surgery: A randomized clinical trial. J Thorac Cardiovasc Surg. 2005; 130(2): 378-83. 62. Wigge H, Uhlig U, Baumgarten G, et al. Mechanical ventilation strategies and inflammatory responses to cardiac surgery: a prospective randomized clinical trial. Intensive Care Med 2005; 31: 1379-87. 63. Chaney M, Nikolov M, Blakeman B, et al. Protective ventilation attenuates postoperative pulmonary dysfunction in patients undergoing cardiopulmonary bypass. J Cardiothorac Vasc Anaes 2000; 14(5): 5154-518. 64. Miranda D, Gommers D, Struijs A, et al. Ventilation according to the open lung concept attenuates pulmonary inflammatory response in cardiac surgery. Eur J Cardiothorac Surg. 2005; 28: 889-95. 65. Okuda M, Furuhashi K, Muneyuki M. Decrease of ischaemia-reperfusion 66. related lung oedema by continuous ventilation and allopurinol in rat perfusion lung model. Scand J Clin Lab Invest 1993; 53(6): 625-31. 67. Imura H, Caputo M, Lim K, et al. Pulmonary injury after cardiopulmonary bypass: Beneficial effects of low frequency mechanical ventilation. J Thorac Cardiovasc Surg 2009; 137(6): 1530-7. 68. Schreiber J, Lancé M, De Korte M, et al. The effects of different lung-protective strategies in patients during cardiopulmonary bypass: A meta-analysis and semi quantitative review of randomized trials. J Cardiothorac Vasc Anesth 2012; 26 (3): 448-54. 69. John L, Ervine I. A study assessing the potential benefit of continued ventilation during cardiopulmonary bypass. Interact Cardiovasc Thorac Surg 2008; 7: 14-7. 70. Gagnon J, Laporta D, Béïque F, et al. Clinical relevance of ventilation during cardiopulmonary bypass in the prevention of postoperative lung dysfunction. Perfusion 2010; 25(4): 205-10. 71. Schlensak C, Doenst T, Preusser S et al. Bronchial artery perfusion during cardiopulmonary bypass does not prevent ischaemia of the lung in piglets: assessment of bronchial artery blood flow with fluorescent microspheres. Eur J Cardiothorac Surg 2001; 19: 326-31. 72. Schlensak C, Doenst T, Preusser S, et al. Cardiopulmonary bypass reduction of bronchial blood flow: a potential mechanism for lung injury in a neonatal pig model. J Thorac Cardiovasc Surg 2002; 123: 1199-205. 73. Serraf A, Sellak H, Hervé P, et al. Vascular endothelium viability and function after total cardiopulmonary bypass in neonatal pigs. Am J Respir Crit Care Med 1999; 159: 544-51. 74. Massoudy P, Zahler S, Becker B, et al. Evidence for inflammatory responses of the lungs during coronary artery bypass grafting with cardiopulmonary bypass. Chest 2001; 119: 31-6.

Page 110

110 75. Serraf A, Robotin M, Bonnet N, et al. Alteration of the neonatal pulmonary physiology after total cardiopulmonary bypass. J Thorac Cardiovasc Surg 1997; 114: 1061-9. 76. Suzuki T, Fukuda T, Ito T, et al. Continuous pulmonary perfusion during cardiopulmonary bypass prevents lung injury in infants. Ann Thorac Surg 2000; 69: 602-606. 77. Santini F, Onorati F, Telesca M, et al. Pulsatile pulmonary perfusion with oxygenated blood ameliorates pulmonary haemodynamic and respiratory indices in low-risk coronary artery bypass patients. Eur J Cardiothorac Surg 2011; 40: 794-803. 78. Liu Y, Wang Q, Zhu X, et al. Pulmonary artery perfusion with protective solution reduces lung injury after cardiopulmonary bypass. Ann Thorac Surg. 200; 69: 1402-7. 79. Sievers H, Freund-Kaas C, Eleftheriadis S, et al. Lung protection during total cardiopulmonary bypass by isolated lung perfusion: Preliminary results of a novel perfusion strategy. Ann Thorac Surg. 2002; 74: 1167-72. 80. Drew C, Anderson I. Profound hypothermia in cardiac surgery. Report of three cases. Lancet 1959; 1: 748-50. 81. Richter J, Meisner H, Tassani P, et al. Drew-Anderson technique attenuates systemic inflammatory response syndrome and improves respiratory function after coronary artery bypass grafting. Ann Thorac Surg 2000; 69: 77-83. 82. Joyce C, Williams A. Kinetics of absorption atelectasis during anesthesia: a mathematical model. J Appl Physiol 1999; 86: 1116-1125. 83. Edmark L, Kostova-Aherdan K, Enlund M et al. Optimal oxygen concentration during induction of general anesthesia. Anesthesiology 2003; 98(1): 28-33. 84. Magnusson L, Zemgulis V, Wicky S, et al. Atelectasis is a major cause of hypoxaemia and shunt after cardiopulmonary bypass. Anesthesiology 1997; 87(5): 1153-63. 85. Santos C, Ferrer M, Roca J, et al. Pulmonary gas exchange response to oxygen breathing in acute lung injury. Am J Respir Crit Care Med 2000; 161: 26-31. 86. Reber A, Budmiger B, Wenk M, et al. Inspired oxygen concentration after cardiopulmonary bypass: effects on pulmonary function with regard to endothelin-1 concentrations and venous admixture. Brit J Anaesth 2000; 84(5): 565-70. 87. Carvalho R, Schettino G, Maranhao B, et al. Hyperoxia and lung disease. Curr Opin Pulm Med. 1998; 4(5): 300-4. 88. Sue R, Belperio J, Burdick M, et al. CXCR2 is critical to hyperoxia-induced lung injury. J Immunol 2004; 172: 3860-8. 89. Brueckl C, Kaestle S, Kerem A, et al. Hyperoxia-induced reactive oxygen species formation in pulmonary capillary endothelial cells in situ. Am J Resp Cell Mol Bio 2006; 34(4): 453-63. 90. Sinclair S, Altemeier W, Matute-Bello G, et al. Augmented lung injury due to the interaction between hyperoxia and mechanical ventilation. Crit Care Med 2004; 32: 2496-501. 91. Pizov R, Weiss Y, Oppenheim-Eden A, et al. High oxygen concentration exacerbates cardiopulmonary bypass-induced lung injury. J Cardiothorac Vasc Anesth 2000; 14(5); 519-23. 92. Toy P, Popovsky M, Abraham E, et al. Transfusion-related acute lung injury: Definition and review. Crit Care Med 2005; 33: 721-6. 93. Bux J, Sachs U. The pathogenesis of transfusion-related acute lung injury (TRALI). Br J Haematol 2007; 136(6): 788-799. 94. Vlaar A, Hofstra J, Determann R, et al. The incidence, risk factors and outcome of transfusion-related acute lung injury in a cohort of cardiac surgery patients: a prospective nested case control study. Blood 2011; 117(16). 4218-25. 95. Koch C, Li L, Figueroa P, et al. Transfusion and pulmonary morbidity after cardiac surgery. Ann Thorac Surg 2009; 88: 1410-8. 96. Leal-Noval S, Rincón-Ferrari M, García-Curiel et al. Transfusion of blood components and postoperative infection in patients undergoing cardiac surgery. Chest 2001; 119(5): 1461-8. 97. Hebert P, Yetisir E, Martin C, et al. Is a low transfusion threshold safe in critically ill patients with cardiovascular diseases? Crit Care Med 2001: 29(2): 227-34. 98. Hajjar L, Vincent J, Galas F, et al. Transfusion requirements after cardiac surgery. JAMA 2010; 304(14): 1559-67. 99. Rehm M, Haller M, Orth V, et al. Changes in blood volume and hematocrit during acute preoperative volume loading with 5% albumin or 6% hetastarch solution in patients before radical hysterectomy. Anesthesiology 2001; 95: 849-56. 100. Osman D, Ridel C, Ray P, et al. Cardiac filling pressures are not appropriate to predict haemodynamic response to volume challenge. Crit Care Med 2007; 35: 64-8. 101. Marik P, Baram M, Vahid B. Does the central venous pressure predict fluid responsiveness? A systematic review of the literature and the tale of seven mares. Chest 2008; 134:172-8. 102. Marik P, Cavallazi R, Vasu T, et al. Stroke volume variation and fluid responsiveness. A systematic review of the literature. Crit Care Med 2009; 37: 2642-7. 103. Preisman S, Kogan S, Berkenstadt H, et al. Predicting fluid responsiveness in patients undergoing cardiac surgery: functional haemodynamic parameters including the respiratory systolic variation test and static preload indicators. Br J Anaesth 2005; 95(6): 746-55. 104. Belloni L, Pisano A, Natale A, et al. Assessment of fluid-responsiveness parameters for off-pump coronary artery bypass surgery: a comparison among LiDCO, transoesophageal echocardiography, and pulmonary artery catheter. J Cardiothorac vasc anesth 2008; 22: 243-8. 105. Vieillard-Baron A, Chergui K, Peyrouset O, et al. Superior vena caval collapsibility as a gauge of volume status in ventilated septic patients. Intensive Care Med 2004; 30: 1734-39. 106. Hoeft A, Korb H, Mehlhorn U, et al. Priming of cardiopulmonary bypass with human albumin or ringer lactate: effect on colloid osmotic pressure and extravascular lung water. Br J Anaesth 1991; 66(1): 73-80. 107. Eising G, Pfauder M, Niemeyer M, et al. Retrograde autologous priming: is it useful in elective on-pump coronary artery bypass surgery? Ann Thorac Surg 2003; 75(1) 23-7. 108. Heinze H, Rosemann C, Weber C, et al. A single dose of pentoxifylline reduces high dependency unit time in cardiac surgery – a prospective randomized and controlled study. Eur J Cardiothorac Surg 2007; 32: 83-9. 109. Welters I, Feurer M. Preiss V, et al. Continuous S-(+)-ketamine administration during elective coronary artery bypass graft surgery attenuates pro-inflammatory cytokine response during and after cardiopulmonary bypass. Br J Anaes 2011; 106(2): 172-9. 110. Luo W, Ling X, Huang R. Effects of aminophylline on cytokines and pulmonary function in patients undergoing valve replacement. Eur J Cardiothorac Surg 2004; 25: 766-71. 111. Qu X, Li Q, Wang X, et al. N-acetylcysteine attenuates cardiopulmonary bypass-induced lung injury in dogs. J Cardiothorac Surg 2013; 8: 107-13. 112. Castillo R, Rodrigo R, Perez F, et al. Antioxidant therapy reduces oxidative and inflammatory tissue damage in patients subjected to cardiac surgery with extracorporeal circulation. Basic Clin Pharm Toxicol 2010; 108: 256-62.

Page 111

111 113. Whitlock R, Chan S, Devereaux P, et al. Clinical benefit of steroid use in patients undergoing cardiopulmonary bypass: a meta-analysis of randomized trials. Eur Heart J 2008; 29: 2592-600. J Cardiothorac Vasc Anesth 2011; 25(1): 156-65. 114. Cappabianca G, Rotunno C, Schinosa L, et al. Protective effects of steroids in cardiac surgery: A meta-analysis of randomized double blind trials. 115. Dieleman J, Nierich A, Rosseel P, et al. Intraoperative high-dose dexamethasone for cardiac surgery. A randomized controlled trial. JAMA 2012; 308(17): 1761-7. 116. Møller C, Perko M, Lund J, et al. No major differences in 30-day outcomes in high-risk patients randomized to off-pump versus on-pump coronary bypass surgery. The Best Bypass Surgery Trial. Circulation 2010; 121: 498-504. 117. Hattler B, Messenger J, Shroyer L, et al. Off-pump coronary artery bypass surgery is associated with worse arterial and saphenous vein graft patency and less effective revascularization. Results from the Veterans Affairs Randomized On/Off Bypass (ROOBY) Study Group. Circulation 2012; 125: 2827-35. 118. Kor D, Warner D, Alsara A, et al. Derivation and diagnostic accuracy of the surgical lung injury prediction model. Anesthesiology 2011; 115(1): 117-28. 119. Adabag A, Wassif H, Rice K, et al. Preoperative pulmonary function and mortality after cardiac surgery. Am Heart J 2010; 159(4): 691-7. 120. Dodd-o J, Hristopoulos M, Welsh-Servinsky L, et al. Strain specific differences in sensitivity to ischaemia-reperfusion lung injury in mice. J Appl Physiol 2006; 100: 1590-5. 121. Chen S, Xu L, Tang J. Association of interleukin 18 gene polymorphism with susceptibility to the development of acute lung injury after cardiopulmonary bypass surgery. Tissue Antigens. 2010; 76: 245-9.

Page 112

Page 113

113 SESSION 9 Tuesday 11:45 – 12:30 Meaningful Outcome Measures in Cardiac Surgery Paul S. Myles, MB BS, MPH, MD, FCARCSI, FANZCA, FRCA, Australia Director of Anaesthesia and Perioperative Medicine, Alfred Hospital, Monash University, Melbourne, Australia Introduction The most common cardiac surgical procedures are coronary artery bypass graft (CABG) surgery, and valve repair or replacement. The two most common underlying conditions are coronary artery disease and heart failure, manifesting as exertional angina, dyspnoea and poor exercise tolerance. It is these symptoms, along with concerns about their own mortality, that bring patients into contact with the cardiac surgical process (1, 2). The major goals of surgery are to both alleviate patient symptoms and to improve survival. This reality, therefore, should inform the choice of outcome measures in clinical studies enrolling patients undergoing cardiac surgery. Unfortunately this is not often the case. The aim of this review is to consider the range of outcome measures typically included in clinical studies of patients undergoing cardiac surgery. The growing interest in the use of patient-centred outcomes, the relationship between these and perioperative complications, and the potential value of using disability-free survival as a primary outcome measure are addressed. Surrogate outcome measures A large proportion of clinical studies in patients undergoing cardiac surgery are focused on variables that perhaps help to explain underlying pathophysiological processes during and after surgery, and/or surrogate measures that directly or indirectly correlate with true patient outcomes (3-6). For example, measures of cardiovascular performance (cardiac output, vascular resistance, diastolic dysfunction, serum lactate), gas exchange (arterial blood gases), renal function (urine flow, creatinine flux), and inflammation (interleukins, C-reactive protein). Surrogate measures such as these often substitute for true clinical outcomes (7). Studies focussing on surrogate outcome measures are relied upon all too often (3, 7). Many are of questionable significance and often have no convincing relationship with patient outcome. For example, a drug proven to lower blood pressure may result in an increased risk of stroke (8), reduce hyperlycemia but increase the risk of death (9), or lower cholesterol but possibly increase the risk of cardiovascular events (10). Efforts to improve surrogate markers of kidney function, such as urine flow or reduce biomarkers such as neutrophil gelatinase-associated lipocalin (NGAL) may or may not be valid measurement of renal function or failure (11). Surgical recovery times and hospital stay reflect both the postoperative course, as well as administrative and patient social aspects such as a supportive home environment. Time to tracheal extubation and duration of stay in the intensive care unit (ICU) are common endpoints in cardiac surgical studies. They do provide some

Page 114

114 indication of the patient’s recovery profile and exposure to complications after surgery, and are in effect composite measures of the entire perioperative process. They also have social and health economic value - patients want to return home, and the costs associated with prolonged hospitalisation can be better spent in other areas. They are, therefore, useful surrogate outcome measures in cardiac surgery. But they are not sufficient and often don’t reflect or evaluate the underlying goals of cardiac surgery. They can also be manipulated. For CABG surgery and percutaneous coronary intervention, there is genuine interest in the rate of graft occlusion following the revascularisation procedure (6). Advances in stent technology, and greater understanding of identification of suitable patients, coronary artery anatomy (of lesions), and antiplatelet therapy, are aimed at avoiding in-stent thrombosis and early graft occlusions (12). But an occluded graft is only important if it threatens myocardial viability, and so it can be argued that such endpoints are merely surrogates for recurrence of angina, myocardial infarction (MI) and patient survival (6). Clinical Indicators and Registries Clinical indicators are process and outcome measures of both the safety and effectiveness of patient care (13-16). They may also be used to measure surgeon and hospital performance, allowing benchmarking and the identification of many aspects of quality of care. These are best done by adjusting for hospital casemix. With the growing interest in the role of public reporting of outcomes in cardiac, thoracic and vascular surgery (17-20), procedure registries such as that run by the Society for Cardiothoracic Surgery in the UK and the Society of Thoracic Surgeons (STS) National Adult Cardiac Surgery Database in the US offer a valuable resource (20, 21). But there are ongoing concerns about clinician and patient privacy, selective reporting, differences in casemix, and “gaming”. The capacity of the general population to properly evaluate performance and risk is open to question. Valuable indicators include unplanned reoperation, unplanned ICU admission, and hospital readmission rates. Such outcomes are major setbacks for the patient and have numerous adverse consequences that could include a greater likelihood of poor survival (22). But few clinical indicators consider the perspective of the patient: are their symptoms relieved and is their life improved?. Serious complications Traditional outcome measures include death and serious complications such as MI and stroke. These are uncommon after most types of cardiac surgery, making clinical research more difficult because very large numbers of patients must be enrolled into studies in order to have sufficient power to detect important differences. For this reason most researchers combine several outcomes into a single composite endpoint such as major adverse cardiac events (MACE) (23-25). The use of composite endpoints is problematic and needs to be justified (23). An influential trial in noncardiac surgery found that perioperative beta-blockers reduced the risk of MI but increased the risk of stroke and death (26). Such conflicting findings challenge the veracity of such composite endpoints, and raise a far more important question: which of these endpoints, or even others that were unmeasured, are most important to a patient recovering from surgery? As stated above, recovery times such as ICU and hospital stay mostly reflect a composite of several adverse events after surgery.

Page 115

115 Postoperative delirium and cognitive dysfunction Delirium is a common, costly and potentially serious complication after cardiac surgery. It is more common in the elderly, and is associated with an increased risk of death, institutionalisation, and possibly dementia (27). Although this is an unwanted scenario after surgery and dose demand extra nursing resources, it is mostly a short-term problem, and perhaps related more to the hospitalisation and ICU process itself rather than as a surgical-related outcome measure. Postoperative cognitive dysfunction (POCD) has been frequently studied in cardiac surgery (28-30). Contemporary thinking suggests that POCD is not unique to cardiac surgery involving cardiopulmonary bypass, but in fact occurs at comparable frequency in many types of major surgery (31, 32). It may even occur after coronary angiography or in any hospitalised patient with a major medical condition (31). Important aspects of study design for POCD include the use of a recommended test battery, an appropriate control group, and focussing on relative change indices rather than absolute values (33, 34). POCD is defined statistically, it is not based on patient symptoms or their capacity to manage activities of daily living. The subtle changes detected by neuropsychological testing are often not apparent to the patient or their family. There seems to be a relationship between test performance and longer term outcome (30), but how POCD manifests in a patient’s daily life is unclear. POCD is a surrogate measure of mild cognitive impairment and dementia. But at present there is insufficient evidence to link POCD and risk of later-onset dementia. Patient-Centred Outcomes Research Rahimi et al (35) did a systematic review of randomised trials evaluating the treatment or prevention of cardiovascular disease published in 10 leading general medical and cardiology journals from 2005 to 2008. They found that few studies used primary outcome measures that could be rated as important from the patient’s point of view. Only 93 of 413 trials (23%) used endpoints such as death or major morbidity, or any other patient-reported outcomes. A similar proportion used a composite endpoint that included one or more less-important outcomes. They recommended that patients should be directly involved in the selection of meaningful outcome measures for cardiovascular trials. More recent evidence suggests that clinicians sometimes make wrong assumptions about patient preferences and values (36). These are not new concepts. Chalmers and Clark noted that most large cardiovascular trials use patient survival/mortality as the sole primary endpoint, so-called “tombstone trials” (37). They noted that patients are not only interested in the potential benefits of new treatments on survival, but also how these treatments affect the quality of their lives. In the US, the Patient Protection and Affordable Care Act (PPACA) becomes fully implemented in January 2014. The central aim is to reduce costs and improve healthcare outcomes by focusing on quality (rather than quantity) of care. The Patient-Centered Outcomes Research Institute was established as part of this government program, with an aim to undertake more comparative effectiveness research (38). These changes place much greater emphasis on outcomes research, evidence synthesis and knowledge translation, aiming to encourage clinicians to ask

Page 116

116 about and understand patient preferences in order to properly inform clinical decision-making. Quality of Recovery and Quality of Life An overall measure of quality of recovery after surgery is useful in that it can provide a global measure of outcome from the patient’s perspective (39). The 40-item quality of recovery score (QoR-40) has undergone extensive psychometric evaluation (40), and has been used in many cardiac and other surgical studies (41). A systematic review of postoperative recovery outcome measurements found the QoR-40 was the only instrument that fulfilled all of the eight pre-specified criteria need to measure health status: appropriateness, reliability, validity, responsiveness, precision, interpretability, acceptability, and feasibility (42). Other systematic reviews have confirmed its psychometric properties and clinical utility (41, 43). Quality of life after surgery is also important (44), and is being measured more often in cardiac surgical studies (12, 44, 45). It is acknowledged that these are important patient-centred outcomes, but some aspects are unrelated to the surgery itself and they may not reflect how well patients can function in the months and years after cardiac surgery. Patients recovering from major surgery, especially the elderly and those with comorbidity, have a slow and complicated recovery course. They are at increased risk of numerous surgical and medical complications in the weeks and months after surgery. Nearly one fifth of US Medicare (elderly) patients discharged from hospital, estimated to be more than 2.5 million people, have an acute medical problem over the next 30 days leading to re-admission to hospital. Krumholz has labelled this phenomenon the post-hospital syndrome (46), and we know this is associated with very poor longer term survival (47). We simply have insufficient information about what happens to patients in the months after surgery – how many are relieved of their symptoms and go on to enjoy a healthy existence? What proportion are harmed by their surgery? It is not just survival, but the relief of symptoms, avoidance of long-term disability, and a sense of wellbeing that are likely to be the most important and highly valued outcomes for patients undergoing major surgery (44, 48, 49). Disability-free survival, therefore, seems to satisfy the key criteria for an ideal outcome measure after cardiac surgery. It addresses the primary aims of most cardiac surgery – reduced symptoms and/or improved healthy survival. It is clearly a patient-centered outcome. The question then becomes: how would our patients define disability and how should it be quantified after cardiac surgery? Are standard measures of disability such as the Katz activities of daily living (50), and the World Health Organization disability assessment scale (WHODAS) (52), valid and reliable after surgery? These important questions require urgent study.

Page 117

117 References 1. Wright RS, Anderson JL, Adams CD, Bridges CR, Casey DE, Jr., Ettinger SM, et al. 2011 ACCF/AHA focused update incorporated into the ACC/AHA 2007 Guidelines for the Management of Patients with Unstable Angina/Non-ST-Elevation Myocardial Infarction: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines developed in collaboration with the American Academy of Family Physicians, Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons. J Am Coll Cardiol. 2011 May 10;57(19):e215-367. 2. Bonow RO, Carabello BA, Chatterjee K, de Leon AC, Jr., Faxon DP, Freed MD, et al. 2008 focused update incorporated into the ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to revise the 1998 guidelines for the management of patients with valvular heart disease). Endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. J Am Coll Cardiol. 2008 Sep 23;52(13):e1-142. 3. Fisher DM. Surrogate outcomes: meaningful not! Anesthesiology. 1999 Feb;90(2):355-6. 4. Goshima KR, Mills JL, Sr., Awari K, Pike SL, Hughes JD. Measure what matters: institutional outcome data are superior to the use of surrogate markers to define "center of excellence" for abdominal aortic aneurysm repair. Ann Vasc Surg. 2008 May-Jun;22(3):328-34. 5. Yudkin JS, Lipska KJ, Montori VM. The idolatry of the surrogate. BMJ. 2011;343:d7995. 6. Harskamp RE, Williams JB, Hill RC, de Winter RJ, Alexander JH, Lopes RD. Saphenous vein graft failure and clinical outcomes: toward a surrogate end point in patients following coronary artery bypass surgery? Am Heart J. 2013 May;165(5):639-43. 7. Institute of Medicine. Evaluation of biomarkers and surrogate endpoints in chronic disease. wwwiomedu/Reports/2010/Evaluation-of-Biomarkers-and-Surrogate-Endpointsin-Chronic-Diseaseaspx. 2010. 8. Law MR, Morris JK, Wald NJ. Use of blood pressure lowering drugs in the prevention of cardiovascular disease: meta-analysis of 147 randomised trials in the context of expectations from prospective epidemiological studies. BMJ. 2009;338:b1665. 9. Finfer S, Chittock DR, Su SY, Blair D, Foster D, Dhingra V, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009 Mar 26;360(13):1283-97. 10. Kastelein JJ, van Leuven SI, Burgess L, Evans GW, Kuivenhoven JA, Barter PJ, et al. Effect of torcetrapib on carotid atherosclerosis in familial hypercholesterolemia. N Engl J Med. 2007 Apr 19;356(16):1620-30. 11. Cruz DN, Soni S, Ronco C. NGAL and cardiac surgery-associated acute kidney injury. Am J Kidney Dis. 2009 Mar;53(3):565-6; author reply 6. 12. Angelini GD, Culliford L, Smith DK, Hamilton MC, Murphy GJ, Ascione R, et al. Effects of on- and off-pump coronary artery surgery on graft patency, survival, and health-related quality of life: long-term follow-up of 2 randomized controlled trials. J Thorac Cardiovasc Surg. 2009 Feb;137(2):295-303. 13. Haller G, Stoelwinder J, Myles P, McNeil J. Quality and safety indicators in anesthesia: a systematic review. Anesthesiology. 2009;110:1158-75. 14. Majoor JW, Ibrahim JE, Cicuttini FM, Boyce NW, McNeil JJ. The extraction of quality-of-care clinical indicators from State health department administrative databases. Med J Aust. 1999 May 3;170(9):420-4. 15. Brook RH, McGlynn EA, Shekelle PG. Defining and measuring quality of care: a perspective from US researchers. Int J Qual Health Care. 2000 Aug;12(4):281-95. 16. Peterson ED, Roe MT, Mulgund J, DeLong ER, Lytle BL, Brindis RG, et al. Association between hospital process performance and outcomes among patients with acute coronary syndromes. JAMA. 2006 Apr 26;295(16):1912-20. 17. Shroyer AL, McDonald GO, Wagner BD, Johnson R, Schade LM, Bell MR, et al. Improving quality of care in cardiac surgery: evaluating risk factors, processes of care, structures of care, and outcomes. Semin Cardiothorac Vasc Anesth. 2008 Sep;12(3):140-52. 18. Hannan EL, Cozzens K, King SB, 3rd, Walford G, Shah NR. The New York State cardiac registries: history, contributions, limitations, and lessons for future efforts to assess and publicly report healthcare outcomes. J Am Coll Cardiol. 2012 Jun 19;59(25):2309-16. 19. Paul S, Sedrakyan A, Chiu YL, Nasar A, Port JL, Lee PC, et al. Outcomes after lobectomy using thoracoscopy vs thoracotomy: a comparative effectiveness analysis utilizing the Nationwide Inpatient Sample database. Eur J Cardiothorac Surg. 2012 Jul 22. 20. Bridgewater B. Cardiac registers: The adult cardiac surgery register. . Heart Failure Reviews. 2010;96:1441-3. 21. Shroyer AL, Coombs LP, Peterson ED, Eiken MC, DeLong ER, Chen A, et al. The Society of Thoracic Surgeons: 30-day operative mortality and morbidity risk models. Ann Thorac Surg. 2003 Jun;75(6):1856-64; discussion 64-5. 22. Ranucci M, Bozzetti G, Ditta A, Cotza M, Carboni G, Ballotta A. Surgical reexploration after cardiac operations: why a worse outcome? Ann Thorac Surg. 2008 Nov;86(5):1557-62. 23. Myles PS, Devereaux PJ. Pros and cons of composite endpoints in anesthesia trials. Anesthesiology. 2010 Oct;113(4):776-8. 24. Montori VM, Permanyer-Miralda G, Ferreira-Gonzalez I, Busse JW, Pacheco-Huergo V, Bryant D, et al. Validity of composite end points in clinical trials. BMJ. 2005 Mar 12;330(7491):594-6. 25. Kip KE, Hollabaugh K, Marroquin OC, Williams DO. The problem with composite end points in cardiovascular studies: the story of major adverse cardiac events and percutaneous coronary intervention. J Am Coll Cardiol. 2008 Feb 19;51(7):701-7. 26. Devereaux PJ, Yang H, Yusuf S, Guyatt G, Leslie K, Villar JC, et al. Effects of extended-release metoprolol succinate in patients undergoing non-cardiac surgery (POISE trial): a randomised controlled trial. Lancet. 2008 May 31;371(9627):1839-47. 27. Witlox J, Eurelings LS, de Jonghe JF, Kalisvaart KJ, Eikelenboom P, van Gool WA. Delirium in elderly patients and the risk of postdischarge mortality, institutionalization, and dementia: a meta-analysis. JAMA. 2010 Jul 28;304(4):443-51. 28. Silbert B, Scott D, Evered L, Lewis M, Kalpokas M, Maruff P, et al. A comparison of the effect of high- and low-dose fentanyl on the incidence of postoperative cognitive dysfunction after coronary artery bypass surgery in the elderly. Anesthesiology. 2006;104:1137-45. 29. Silbert B, Maruff P, Evered L, Scott D, Kalpokas M, Martin K, et al. Detection of cognitive decline after coronary surgery: a comparison of computerized and conventional tests. Br J Anaesth. 2004;92:814-20.

Page 118

118 30. Newman MF, Kirchner JL, Phillips-Bute B, Gaver V, Grocott H, Jones RH, et al. Longitudinal assessment of neurocognitive function after coronary-artery bypass surgery. N Engl J Med. 2001 Feb 8;344(6):395-402. 31. Evered L, Scott DA, Silbert B, Maruff P. Postoperative cognitive dysfunction is independent of type of surgery and anesthetic. Anesth Analg. 2011 May;112(5):1179-85. 32. Hakkinen A, Heinonen M, Kautiainen H, Huusko T, Sulkava R, Karppi P. Effect of cognitive impairment on basic activities of daily living in hip fracture patients: a 1-year follow-up. Aging Clin Exp Res. 2007 Apr;19(2):139-44. 33. Lewis MS, Maruff P, Silbert BS, Evered LA, Scott DA. The influence of different error estimates in the detection of postoperative cognitive dysfunction using reliable change indices with correction for practice effects. Arch Clin Neuropsychol. 2007 Feb;22(2):249-57. 34. Lewis MS, Maruff P, Silbert BS, Evered LA, Scott DA. The sensitivity and specificity of three common statistical rules for the classification of post-operative cognitive dysfunction following coronary artery bypass graft surgery. Acta Anaesthesiol Scand. 2006 Jan;50(1):50-7. 35. Rahimi K, Malhotra A, Banning AP, Jenkinson C. Outcome selection and role of patient reported outcomes in contemporary cardiovascular trials: systematic review. BMJ. 2010;341:c5707. 36. Mulley AG. Inconvenient truths about supplier induced demand and unwarranted variation in medical practice. BMJ. 2009;339:b4073. 37. Chalmers I, Clarke M. Outcomes that matter to patients in tombstone trials. Lancet. 2001 Nov 10;358(9293):1649. 38. Selby JV, Beal AC, Frank L. The Patient-Centered Outcomes Research Institute (PCORI) national priorities for research and initial research agenda. JAMA. 2012 Apr 18;307(15):1583-4. 39. Lee A, Lum M. Measuring anaesthetic outcomes. Anaesthesia and intensive care. 1996;24(6):685. 40. Myles P, Weitkamp B, Jones K, Melick J, Hensen S. Validity and reliability of a post-operative quality of recovery score: the QoR-40. Br J Anaesth. 2000;84:11-5. 41. Gornall BF, Myles PS, Smith CL, Burke JA, Leslie K, Pereira MJ, et al. Measurement of quality of recovery using the QoR-40: a quantitative systematic review. Br J Anaesth. 2013 Aug;111(2):161-9. 42. Herrera FJ, Wong J, Chung F. A systematic review of postoperative recovery outcomes measurements after ambulatory surgery. Anesth Analg. 2007 Jul;105(1):63-9. 43. Kluivers KB, Riphagen I, Vierhout ME, Brolmann HA, de Vet HC. Systematic review on recovery specific quality-of-life instruments. Surgery. 2008 Feb;143(2):206-15. 44. Cheema FN, Abraham NS, Berger DH, Albo D, Taffet GE, Naik AD. Novel Approaches to Perioperative Assessment and Intervention May Improve Long-Term Outcomes After Colorectal Cancer Resection in Older Adults. Ann Surg. 2010 Dec 22;253:867-74. 45. Myles P, Viira D, Hunt J. Quality of life at three years after cardiac surgery: relationship with preoperative status and quality of recovery. Anaesthesia and intensive care. 2006;34(2):176-83. 46. Krumholz HM. Post-hospital syndrome--an acquired, transient condition of generalized risk. N Engl J Med. 2013 Jan 10;368(2):100-2. 47. Krumholz HM, Lin Z, Keenan PS, Chen J, Ross JS, Drye EE, et al. Relationship between hospital readmission and mortality rates for patients hospitalized with acute myocardial infarction, heart failure, or pneumonia. JAMA. 2013 Feb 13;309(6):587-93. 48. Myles PS, Hunt JO, Nightingale CE, Fletcher H, Beh T, Tanil D, et al. Development and psychometric testing of a quality of recovery score after general anesthesia and surgery in adults. Anesth Analg. 1999 Jan;88(1):83-90. 49. Amemiya T, Oda K, Ando M, Kawamura T, Kitagawa Y, Okawa Y, et al. Activities of daily living and quality of life of elderly patients after elective surgery for gastric and colorectal cancers. Ann Surg. 2007 Aug;246(2):222-8. 50. Ettinger WH, Jr., Fried LP, Harris T, Shemanski L, Schulz R, Robbins J. Self-reported causes of physical disability in older people: the Cardiovascular Health Study. CHS Collaborative Research Group. J Am Geriatr Soc. 1994 Oct;42(10):1035-44. 51. Guralnik JM, LaCroix AZ, Branch LG, Kasl SV, Wallace RB. Morbidity and disability in older persons in the years prior to death. Am J Public Health. 1991 Apr;81(4):443-7. 52. Garin O, Ayuso-Mateos JL, Almansa J, Nieto M, Chatterji S, Vilagut G, et al. Validation of the "World Health Organization Disability Assessment Schedule, WHODAS-2" in patients with chronic diseases. Health Qual Life Outcomes. 2010;8:51.

Page 119

119 SESSION 10 Tuesday 13:30 – 15:30 Bypass to suit the Patient Minimising Prime Volumes – An Anaesthetists Perspective David A. Scott, MB BS, FANZCA, PhD, FFPMANZCA, Australia Associate Professor and Director of Anaesthesia, St Vincent’s Hospital, Melbourne, Australia, Associate Professor, Faculty of Medicine, Dentistry and Health Sciences, University of Melbourne, Melbourne, Australia The need for large prime volumes in the early days of CPB meant that the use of blood prime was a common event if significant haemodilution was to be avoided. With improved circuit design, the volume of fluid needed to prime the pump has significantly decreased. None-the-less, it is not uncommon for prime volumes (containing crystalloid alone or with a colloid) to still be in excess of 1500 mL. The impact of this volume on a non-anaemic adult patient with a large blood volume may be quantitatively small, but that does not apply to the 50kg elderly borderline anaemic patient. This may increase the need for blood transfusion. It’s not all about the volume either. The sudden influx of a non-blood perfusate to the brain in particular but also to the systemic circulation exposes the delicate microcirculation and endothelial glycocalyx to an unphysiologic environment that may cause transient or permanent injury. This is crudely demonstrated by the haemodynamic responses to initiation of CPB. Avoidance of these risks includes facilitation of prime fluid volume reduction and if possible prime displacement by the patient’s own blood. Prior to cannulation the patient should be euvolaemic, haemodynamic ally stable and not haemodiluted. If autologous priming is being used in addition to a small circuit volume, the perfusionist requires the patient to be supported whilst displacing blood into the circuit. This requires close communication and co-operation between the anaesthetist and perfusionist. At the end of bypass, there may be less pump blood available for volume replacement. Autologous collection of blood during the pre-bypass period is useful for this stage. It should be remembered that the prime aim of circulatory support is to maintain perfusion and oxygenation of tissues and that this should not be compromised by excessive vasopressor support or adherence to a protocol at all costs.

Page 120

Page 121

121 Bypass to suit the Patient Minimising Prime Volumes – A Perfusionists Perspective Timothy Willcox, CCP, New Zealand Chief Clinical Perfusionist, Auckland District Health Board, Green Lane Perfusion, Auckland, New Zealand In the early days of cardiopulmonary bypass (CPB) the volume of large disk oxygenators, venous reservoirs, external heat exchangers and the associated tubing could require prime volumes in excess the blood volume of both adult and paediatric patients. Fresh heparinised blood commonly drawn from donors on the day of operation was routinely used in the CPB prime for all patients to provide adequate oxygen carrying capacity. Priming volumes of extracorporeal circuits have progressively become smaller with the evolution of heart lung machines but more especially with the development of the circuit components. Notwithstanding these advances, current adult CPB circuits have priming volumes in the order of 1.5 litres [1]. Homologous donor blood has become a scarce resource and despite advances in screening techniques, allogeneic blood transfusion continues to be associated with variable morbidity including stroke, acute kidney injury, post-operative low output failure, lung injury and inflammation[2-6]. Over the last decade evidence-based consensus guidelines for avoidance of allogeneic transfusion during CPB have recommended reducing prime volume and circuit size [7-9] based, in part, on large observational studies associating increased haemodilution and nadir haematocrit during CPB with adverse outcomes [10-12]. Reduction in the foreign surface area of the CPB circuit is perceived as being associated with reduced inflammation and was included in a class IIa (level B) recommendation as “might be useful —effective in reducing the systemic inflammatory response” [7]. While attenuation of CPB mediated inflammatory markers has been demonstrated by interventions such as modified ultrafiltration [13, 14] and to a lesser extent zero balance ultrafiltration [15], reduction of “circuit size” per se has not been similarly associated. It is plausible that this is because by far the greatest component of foreign surface of the CPB circuit is the surface area of the oxygenator membrane and that until recently adult oxygenator membrane surface area has remained relatively unchanged. To a large extent the circuit size is governed by capital equipment in the form of the heart lung machine with a lifespan of 10-15 years, and by established mind-sets within a given centre. Conventional modular heart lung machines have undergone relatively little significant change in the last two and a half decades and attempts to introduce heart lungs machines that allow vertically stacked pumps such as the Medtronic Performer ™ or heart lung machines with pumps that are variably positional such as the Maquet HL30 were short lived on the market. Of note however is the development of mini CPB systems from what was arguably the premature introduction of the CorX™ mini bypass system.

Page 122

122 The focus on minimising the CPB prime in adult cardiac surgery, primarily to reduce the use of allogeneic blood, has resulted in two conceptual developments of the CPB circuit. On the one hand was the radical redesign of the CPB circuit with the advent of the mini-circuit, which consists of a closed system and removal of the venous reservoir. On the other hand, and more recently, rather than a one size fits all, has been a reconfiguration of the conventional adult CPB circuit components to reduce circuit size. This has been facilitated by the development of new design oxygenators with integration of arterial filtration and reservoirs with reduced dynamic priming volumes that are sized matched to the patient, (a concept well established in paediatric CPB). This is being referred to as the optimised circuit [16]. From a perfusion perspective, current options for managing the adult circuit prime volume (that includes the initial circuit prime and intraoperative added volume) are threefold. Firstly, retaining a one size fits all (for convenience of inventory and standard setup). Secondly, “optimising” the circuit to minimise tubing length, using size matched oxygenators and potentially integral arterial filtration, and thirdly using a closed system mini-circuit (and typically microplegia). Applying additional interventions may effectively further reduce the CPB prime, including retrograde autologous priming (RAP), cell salvage, haemofiltration throughout and following CPB, and microplegia can be used to further minimise prime volume. The introduction of the mini system with the CorX™, while innovative in the removal of the venous reservoir, was associated with air handling problems that was arguably the commercial downfall of that device [17]. The limitation to deal with air removal and the associated safety concerns, together with the complexity of using a closed system circuit for CPB contributed to suspicion of subsequent iterations of mini systems that has likely influenced its widespread adoption [18, 19]. However there is now a growing body of literature including recent meta-analyses that demonstrate the use of mini-circuits results in significantly higher haematocrit and significantly reduced blood transfusion during CPB [19, 20], although clinical outcome benefit is less well established [9]. While a recent survey of CPB practice in the United Kingdom reported use of mini system CPB at 20% [21], by contrast, and somewhat interestingly, there is no use of mini systems in Australasia at the present time. Less challenging from a technical perspective in minimising CPB prime volume is optimising the standard CPB circuit in conjunction with the application of additional interventions to manage haemodilution. New generation oxygenators with integrated filters and reservoirs with smaller dynamic prime volumes are now being released to the market [16] and there is increasing evidence for the use of RAP and cell salvage and consideration of a range of further interventions [9]. While the routine application of vacuum assisted venous drainage in adult CPB has been used to achieve reduction in cannula and tubing size contributing to a concomitant reduction in nadir haematocrit [14], this remains controversial [22, 23]. There is current attention to the importance of monitoring oxygen consumption and delivery during CPB not only as an accurate indicator of adequacy of perfusion for avoidance of renal injury, but also as a guideline to threshold for transfusion whereby increasing blood flow may avoid the need for adding donor blood during CPB [24, 25]. This has led to the recent inclusion of these parameters into new heart lung machine monitoring software (Sorin Goal Directed Perfusion (GDP), Sorin

Page 123

123 Group Deutschland – Munich Germany) with the potential to effectively ameliorate the effects of haemodilution. Isolated interventions to minimise the size of the CPB circuit are less likely to result in measurable outcome benefit than a considered multi-modal approach. As an example of an ad hoc approach, removal of autologous blood pre bypass, with no evidence of outcome advantage, may result in an unwarranted nadir haematocrit during CPB [26, 27]. An effective shift to minimising priming volumes in the CPB circuit may require an institutional shift in mind-set and culture to achieve an agreed evidenced based coordinated approach. This extends beyond the operative period. Pre-operative anaemia is associated with increased transfusion and adverse outcome [28]. Efforts to minimise prime volumes to avoid haemodilution during CPB are frustrated by elective patients arriving at the operating room with unnecessary anaemia. Identifying and treating pre-operative anaemia should be an essential part of patient workup. In the postoperative intensive care variable or liberal transfusion practice may undo the efforts to limit haemodilution during CPB. A multi-centre controlled randomised trial in critically ill non bypass patients found that a restrictive strategy of red cell transfusion (maintaining Hb 7-9 g/dL) was “at least as effective if not superior” to a more liberal strategy (maintaining Hb 10-12 g/dL)[29]. This study is planned to be repeated in cardiac surgery patients and the results will be of particular interest to the current debate on transfusion triggers. Minimising the CPB prime may require a change of mind-set not only on the part of the perfusionist but by the surgical team as a whole with reference to external regional benchmarks such as the Perfusion Downunder Collaboration. References 1. Pappalardo, F., et al., Reduction of hemodilution in small adults undergoing open heart surgery: a prospective, randomized trial. Perfusion, 2007. 22(5): p. 317-22. 2. Bahrainwala, Z.S., et al., Intraoperative hemoglobin levels and transfusion independently predict stroke after cardiac operations. The Annals of thoracic surgery, 2011. 91(4): p. 1113-8. 3. Vellinga, S., et al., Identification of modifiable risk factors for acute kidney injury after cardiac surgery. The Netherlands journal of medicine, 2012. 70(10): p. 450-4. 4. Surgenor, S.D., et al., Intraoperative red blood cell transfusion during coronary artery bypass graft surgery increases the risk of postoperative low-output heart failure. Circulation, 2006. 114(1 Suppl): p. I43-8. 5. Vlaar, A.P., et al., The incidence, risk factors, and outcome of transfusion-related acute lung injury in a cohort of cardiac surgery patients: a prospective nested case-control study. Blood, 2011. 117(16): p. 4218-25. 6. Banbury, M.K., et al., Transfusion increases the risk of postoperative infection after cardiovascular surgery. Journal of the American College of Surgeons, 2006. 202(1): p. 131-8. 7. Shann, K.G., et al., An evidence-based review of the practice of cardiopulmonary bypass in adults: A focus on neurologic injury, glycemic control, hemodilution, and the inflammatory response. Journal of Thoracic & Cardiovascular Surgery, 2006. 132(2): p. 283-290.e3. 8. Ferraris, V.A., et al., 2011 update to the Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists blood conservation clinical practice guidelines. The Annals of thoracic surgery, 2011. 91(3): p. 944-82. 9. Menkis, A.H., et al., Drug, devices, technologies, and techniques for blood management in minimally invasive and conventional cardiothoracic surgery: a consensus statement from the International Society for Minimally Invasive Cardiothoracic Surgery (ISMICS) 2011. Innovations, 2012. 7(4): p. 229-41. 10. DeFoe, G.R., et al., Lowest hematocrit on bypass and adverse outcomes associated with coronary artery bypass grafting. Northern New England Cardiovascular Disease Study Group.[see comment]. Annals of Thoracic Surgery, 2001. 71(3): p. 769-76. 11. Habib, R.H., et al., Adverse effects of low hematocrit during cardiopulmonary bypass in the adult: should current practice be changed?[see comment]. Journal of Thoracic & Cardiovascular Surgery, 2003. 125(6): p. 1438-50. 12. Karkouti, K., et al., Low hematocrit during cardiopulmonary bypass is associated with increased risk of perioperative stroke in cardiac surgery. Ann Thorac Surg, 2005. 80(4): p. 1381-7. 13. Atkins, B.Z., et al., Modified ultrafiltration attenuates pulmonary-derived inflammatory mediators in response to cardiopulmonary bypass. Interactive cardiovascular and thoracic surgery, 2010. 11(5): p. 599-603. 14. Papadopoulos, N., et al., The effect of normovolemic modified ultrafiltration on inflammatory mediators, endotoxins, terminal complement complexes and clinical outcome in high-risk cardiac surgery patients. Perfusion, 2013. 28(4): p. 306-14. 15. Tallman, R.D., M. Dumond, and D. Brown, Inflammatory mediator removal by zero-balance ultrafiltration during cardiopulmonary bypass. Perfusion, 2002. 17(2): p. 111-5.

Page 124

124 16. Starck, C., et al., Initial results of an optimized perfusion system. Perfusion, 2013. 28(4): p. 292-7. 17. Abdel-Rahman, U., et al., Initial experience with a minimized extracorporeal bypass system: is there a clinical benefit? The Annals of thoracic surgery, 2005. 80(1): p. 238-43. 18. Nollert, G., et al., Miniaturized cardiopulmonary bypass in coronary artery bypass surgery: marginal impact on inflammation and coagulation but loss of safety margins. The Annals of thoracic surgery, 2005. 80(6): p. 2326-32. 19. Ranucci, M. and S. Castelvecchio, Management of mini-cardiopulmonary bypass devices: is it worth the energy? Current opinion in anaesthesiology, 2009. 22(1): p. 78-83. 20. Zangrillo, A., et al., Miniaturized cardiopulmonary bypass improves short-term outcome in cardiac surgery: a meta-analysis of randomized controlled studies. The Journal of thoracic and cardiovascular surgery, 2010. 139(5): p. 1162-9. 21. Warren, O.J., et al., Variations in the application of various perfusion technologies in Great Britain and Ireland--a national survey. Artificial organs, 2010. 34(3): p. 200-5. 22. Durandy, Y., Vacuum-Assisted Venous Drainage, Angel or Demon: PRO? Journal of Extracorporeal Technology, 2013. 45: p. 122-127. 23. Willcox, T., Vacuum Assist: Angel or Demon CON. Journal of Extracorporeal Technology, 2013. 45: p. 128-132. 24. Ranucci, M., et al., Transfusions during cardiopulmonary bypass: better when triggered by venous oxygen saturation and oxygen extraction rate. Perfusion, 2011. 26(4): p. 327-33. 25. Ranucci, M., Perioperative renal failure: hypoperfusion during cardiopulmonary bypass? Seminars in Cardiothoracic & Vascular Anesthesia, 2007. 11(4): p. 265-8. 26. Ramnarine, I.R., et al., Autologous blood transfusion for cardiopulmonary bypass: effects of storage conditions on platelet function. Journal of Cardiothoracic & Vascular Anesthesia, 2006. 20(4): p. 541-7. 27. Virmani, S., et al., Acute normovolemic hemodilution is not beneficial in patients undergoing primary elective valve surgery. Annals of cardiac anaesthesia, 2010. 13(1): p. 34-8. 28. Spiess, B.D., Transfusion of blood products affects outcome in cardiac surgery. Seminars in cardiothoracic and vascular anesthesia, 2004. 8(4): p. 267-81. 29. Hebert, P.C., et al., A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. The New England journal of medicine, 1999. 340(6): p. 409-17.

Page 125

125 Bypass to suit the Patient Fluid Therapy and Outcomes: Balance is Best Sara J. Allen, BHB, MBChB, FANZCA, New Zealand Anaesthetist/Intensivist, Greenlane Department of Cardiothoracic and ORL, Anaesthesia/Cardiovascular ICU Auckland City Hospital, Auckland, New Zealand Introduction Administration of intravenous fluids is routine in the management of surgical and critically ill patients. Recent published randomised controlled trials and systemic reviews have examined the efficacy, morbidity and mortality associated with the use of different types, timing, and doses of intravenous fluids, and suggest that each of these factors may influence outcomes for patients1-6. However, controversies still exists with regard to the ideal type, dose, and timing of intravenous fluid for peri-operative and critically ill patients. This review re-examines the evidence to date, with particular focus on current best practice in the specialties of perfusion, anaesthesia, and critical care, and on how a balanced approach to all domains of fluid administration may positively influence clinical outcomes. Types of fluid: crystalloid versus colloid Definitions Crystalloid fluids consist of isotonic saline (e.g. 0.9% saline, normal saline) or balanced electrolyte solutions (e.g. Plasmalyte, Ringer’s lactate), and are considered to rapidly and widely distribute across the extracellular fluid compartments after administration. Colloids are fluids with a crystalloid carrier solution containing suspended large molecular weight molecules which do not freely diffuse across the extracellular fluid compartments (e.g. albumin, hydroxyethyl starch), and therefore exert a colloid oncotic pressure. Colloids have traditionally been thought to remain in the intravascular fluid compartment following administration. However, several factors may alter the distribution of colloid solutions, including patient volume status, the integrity of the vascular endothelium, systemic inflammation, and the use of alpha-adrenergic agonists. Colloid subtypes: albumin, gelatins, and synthetic starches Within colloids are several subtypes of fluid. Albumin is a fluid containing the plasma protein, most commonly in 4% solution and usually suspended in isotonic saline, derived from pooled human plasma, whilst synthetic colloids such as Voluven® and Volulyte® contain 6% hydroxyethyl starch (HES) suspended in balanced crystalloid solution, derived from plant sources. Older colloids include Haemaccel® and Gelofusin®, containing succinylated gelatin molecules derived from animal sources and suspended in balanced crystalloid solutions. Each type of colloid has potential

Page 126

126 risks associated with use – albumin as a blood product has a potential risk of infection, synthetic colloids have potential risks of coagulopathy, end organ damage and anaphylactoid reactions. Table 1 Composition of common crystalloid and colloid fluids5 Electrolyte (mmol/L) Plasmalyte Normal saline Ringer’s lactate Albumin 4% Voluven HES 6% Gelofusin Sodium 140 154 131 140 154 154 Potassium 5 0 5 0 0 0 Chloride 98 154 111 128 154 125 Calcium 0 0 2 0 0 0 Magnesium 1.5 0 1 0 0 0 Bicarbonate 0 0 0 0 0 0 Lactate 0 0 29 0 0 0 Acetate 27 0 0 0 0 0 Gluconate 23 0 0 0 0 0 Octanoate 0 0 0 6.4 0 0 Outcomes Several large randomised controlled trials have examined the use of crystalloid versus colloid solutions. Important differences according to which colloid was used have been demonstrated. In 2004, the SAFE Study, a large multicentre randomised controlled trial, investigated the use of 0.9% saline versus 4% albumin for fluid resuscitation in 6997 intensive care patients. Total fluid administration between the two groups differed over the first four days, with a ratio of volume of albumin to volume of saline of 1:1.4. This represented an approximate mean difference of 800-900mL of fluid over this time, a clinically modest amount. No differences in overall mortality or outcomes at 28 days were demonstrated, although in pre-planned sub-group analysis albumin was associated with differential outcomes (adverse outcome in traumatic brain injury, improved outcome in severe sepsis)2. The use of colloids has traditionally been recommended as a method of minimising total volume of fluid required for resuscitation in comparison with crystalloids. The expected ratio of volume of crystalloid to colloid administered for the same degree of intravascular fluid expansion is approximately 5 to 1. However, as highlighted above, trials have demonstrated that the estimated additional volume expansion with colloids in place of crystalloids is much smaller in vivo than estimated based purely on physicochemical properties and Starling’s Laws1-2. This reflects the fact that the endothelial glycocalyx layer is damaged in many critically ill and surgical patients, due to the systemic inflammatory response, and is unable to maintain full integrity. This allows more rapid diffusion of molecules and carrier fluid into the interstitium than would otherwise be predicted. The crystalloid to colloid ratio is estimated from large trials to be only 1.2-1.4 to 11-2. Additional to interstitial oedema resulting from administration of all fluids, there is significant concern regarding tissue injury with synthetic colloids, due to accumulation of synthetic molecules in the interstitium after

Page 127

127 breaching the endothelium, with toxic effects previously described in the kidney, liver and bone marrow7. In 2012, the 6S Trial Group and the Scandinavian Critical Care Trials Group reported results from a large multicentre, parallel-group, randomised, blinded trial comparing fluid resuscitation in the ICU with 6% HES versus Ringer’s acetate in 798 patients with severe sepsis7. This study demonstrated increased risk of mortality at day 90 (relative risk 1.17), and increased risk of renal-replacement therapy (relative risk 1.35) in patients treated with HES. Additionally, patients treated with HES received more allogeneic blood product transfusions. There were no differences in the total volumes of fluid used between the two groups. A further study published in 2012 compared the use of HES or 0.9% saline for resuscitation in Intensive Care1. A total of 7000 patients admitted to an intensive care unit (ICU) were randomly allocated to receive 6% HES or 0.9% saline for all fluid resuscitation until discharge from ICU. There was no significant difference in mortality between the two groups, but an increased use of renal replacement therapy in patients treated with HES. Additionally, an increase in adverse events (e.g. pruritis, skin rash) occurred in the HES treated group. There was again a modest difference in the volume of fluid administered between groups. Conclusions Overall, the use of synthetic colloids in critically ill patients confers no benefit, and is associated with significant harm and increased cost. Several organisations have issued statements and guidelines recommending the cessation of synthetic colloid use in critically ill patients8-10. The use of these fluids in non-critically ill and routine surgical patients is not as well studied, however given the lack of documented benefit, and increased cost, their use is not recommended. In addition, as albumin has been demonstrated as a safe colloid fluid it may be used in the limited number of patients where a benefit is expected in modest volume sparing and prolongation of intravascular expansion. Crystalloid solutions: balanced electrolyte solutions versus isotonic saline Recently, research regarding optimal fluid therapy has focussed on the use of balanced crystalloid solutions (e.g. Ringer’s acetate, Plasmalyte) in comparison with normal saline, due to concerns regarding the deleterious effects of administration of normal saline, particularly in relation to metabolic, gastrointestinal, renal and coagulation side effects. Isotonic 0.9% (normal) saline Normal saline contains 154mmol/L sodium and 154mmol/L chloride ions, with an osmolality of 287 mOsm/kg H2O (identical to plasma osmolality). Infusion of large volumes of isotonic saline (e.g. cardiopulmonary bypass prime, fluid replacement intraoperatively) therefore results in a chloride load and dilution of plasma anions – causing a dilutional hyperchloremic metabolic acidosis.

Page 128

128 Balanced solutions Different balanced solutions exist, however all have electrolyte compositions similar to that of plasma, as outlined in Table 1. Organic anions such as lactate, acetate and gluconate are used as buffers to provide in vitro isotonicity, but are rapidly metabolised after intravenous infusion, resulting in decreased osmolarity in vivo. Infusion of large volumes of balanced crystalloid solutions does not produce a large chloride load, and has less dilution effect (as rapid diuresis is evoked by suppression of antidiuretic hormone release which occurs due to the in vivo hypotonicity). Balanced solutions therefore have preserved acid-base homeostasis, and chloride levels remain normal. Outcomes Controversy exists regarding the metabolic acidosis induced by saline – whether the acidosis affects clinical outcomes, or is simply a side effect of saline which is transient and benign. There is currently no clear evidence to elucidate this. However, the effects of excess chloride due to saline administration are more clearly understood. Renal and gastrointestinal effects Chloride is regulated by absorption and secretion in the gastrointestinal tract and reabsorption and excretion by the kidney. Excess chloride induces renal vasoconstriction via reabsorption and tubuloglomerular feedback, decreasing glomerular filtration rate and renin activity, whilst increasing responsiveness to angiotensin II11. These effects reduce renal blood flow, diuresis and natriuresis. Chloride may also cause thromboxane release, and increased responsiveness to circulating vasoconstrictors12. A 2012 study in healthy volunteers demonstrated sustained hyperchloremia after 0.9% saline infusion, associated with reduced renal blood flow and reduced renal cortical tissue perfusion. Infusion of plasmalyte was not associated with these changes. Also, whilst blood volume changes were identical, a greater subsequent expansion of the extravascular fluid compartment was observed with saline infusions13. In a small study of elderly patients undergoing major surgery, infusion of saline solution (or 6% HES in saline solution) versus balanced solution (or 6% HES in a balanced solution) was associated with a more prolonged time to first micturition, significant metabolic acidosis, and worsened gastric mucosal tonometry14. A study of 1407 intensive care patients treated with either a chloride-liberal or a chloride-restrictive strategy demonstrated a significant increase in acute kidney injury (AKI) as defined by the RIFLE criteria, and in the use of renal replacement therapy (RRT) in the chloride-liberal group. No mortality differences were demonstrated15. Finally, a large observational study of 30,994 patients undergoing abdominal surgery who received either 0.9% saline or a balanced crystalloid solution demonstrated increased morbidity with the use of 0.9% saline, with increased infection risk, increased AKI and requirement for RRT, and increased requirement for blood transfusion16.

Page 129

129 Infection Hyperchloremia is associated with proinflammatory and procytokine effects11. In patients with hyperchloremia, altered immune responses may increase the risk of infection. Also, acute kidney injury and RRT increase infection risk. Coagulation and bleeding effects In high doses, 0.9% saline may cause coagulopathy16. Mechanisms contributing to this include dilution, acidosis, and the absence of calcium within the saline solution11. Clinical studies are limited and no difference in bleeding or blood product use has been consistently demonstrated. Colloids Some studies have reviewed colloids in balanced versus saline solutions. In a 2006 study of 81 patients undergoing cardiac surgery, HES in a balanced solution or normal saline was administered, with comparison of total volumes required, serum pH, serum chloride and base excess levels. No differences in volume of fluid required were noted between groups. Lower chloride levels and corresponding higher pH and base excess levels occurred in the group administered HES in a balanced solution carrier. No morbidity or mortality outcomes were measured in this study, however17. Future research Further large randomised trials are warranted to examine the incidence and effects of hyperchloremia in hospitalised patients, and patient sub-groups such as cardiac surgery. The chloride content is not the only difference between 0.9% saline and balanced crystalloids, and the possible beneficial effects of buffers or lactate should also be studied. Conclusions Recent research is consistent in demonstrating possible harm with the use of 0.9% saline versus balanced crystalloids in fluid therapy. The exact mechanisms and extent of possible harm are not fully elucidated, and more research is likely to be beneficial. Balance in doses: restrictive versus liberal administration Fluid therapy is often used to achieve hemodynamic targets such as cardiac output, mean arterial blood pressure, and central venous pressure; however, over administration of fluid may be as harmful as inadequate fluid resuscitation. The optimum hemodynamic targets for achieving end-organ perfusion are still not known, and are likely to vary between patients, and may vary over time for an individual patient. Excess fluid may impair tissue oxygenation, wound healing, and is associated with AKI18. The Fluid Expansion As Supportive Therapy (FEAST) trial compared bolus fluid resuscitation (saline or albumin) with standard slow rehydration therapy in more than 3000 children in sub-Saharan Africa presenting to hospital with severe infections.

Page 130

130 The trial demonstrated increased mortality with fluid bolus therapy with either saline or albumin when compared with standard rehydration19. In contrast, studies of early goal directed therapy have demonstrated improved outcomes including reduced mortality, with fluid bolus therapy directed by haemodynamic targets and protocolised treatment pathways20. Several trials are currently in progress examining optimal fluid dosing and timing regimens (e.g. Restrictive versus liberal fluid therapy in major abdominal surgery (RELIEF) trial). Summary Current evidence suggests that there is no benefit, but significant potential for harm, associated with the use of synthetic colloid solutions as intravenous fluids. As synthetic colloid solutions also have increased cost compared with crystalloid solutions, their continued use cannot be justified. There is increasing evidence in favour of the use of balanced crystalloid solutions rather than isotonic saline, with reduced metabolic acidosis, and reduced acute kidney injury in patients treated with balanced crystalloid versus isotonic saline solutions. No large randomised controlled trials yet exist to support the number of smaller studies. Not only is the type of fluid administered clinically relevant, but the dose of fluid is likely to contribute to clinical outcomes – with excess fluid administration associated with increased tissue oedema, coagulopathy, acidosis, and potential organ dysfunction, while inadequate fluid resuscitation may be associated with inadequate cardiac output and organ perfusion, with resultant dysfunction.

Page 131

131 References 1. Myburgh JA, Finfer S, Bellomo R, et al. Hydroxyethyl starch or saline for fluid resuscitation in intensive care. N Engl J Med 2012; 367:1901-1911. 2. SAFE Study Investigators. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med 2004; 350:2247-2256. 3. Shaw AD, Bagshaw SM, Goldstein SL, et al. Major complications, mortality, and resource utilisation after open abdominal surgery: 0.9% saline compared to Plasma-Lyte. ANN Surg 2012; 255:821-829. 4. Maitland K, Kiguli S, Opoka R, et al. Mortality after fluid bolus in African children with shock. N Engl J Med 2011; 364:2483-2495. 5. Guidet B, Soni N, Della Rocca G, et al. A balanced view of balanced solutions. Crit Care 2010; 14:325. 6. Raghunathan K, Shaw AD, Bagshaw SM. Fluids are drugs: type, dose and toxicity. Curr Opin Crit Care 2013; 19:290-298. 7. Perner A, Haase N, Guttormsen AB, et al. Hydroxyethyl starch 130/0.42 versus Ringer’s acetate in severe sepsis. N Engl J Med 2012; 367:124-134. 8. Reinhart K, Perner A, Sprung CL, et al. Consensus statement of the ESICM task force on colloid volume therapy in critically ill patients. Intensive Care Med 2012; 38(3):368-383. 9. Dellinger RP, Levy MM, Rhodes A, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 2013; 41:580-637. 10. Perel P, Roberts I, Ker K. Colloids versus crystalloids for fluid resuscitation in critically ill patients. Cochrane Database Syst Rev 2013; 2:CD000567. 11. Mohd Yunos N, Bellomo R, Story D, et al. Bench-to-bedside review: chloride in critical illness. Crit Care 2010; 14:226. 12. Quilley CP, Lin YS, McGIff JC. Chloride anion concentration as a determinant of renal vascular responsiveness to vasoconstrictor agents. Br J Pharmacol 1993; 108:106-110. 13. Chowdhury AH, Cox EF, Francis ST, et al. A randomised, controlled, double-blind crossover study on the effects of 2-L infusions of 0.9% saline and Plasma-Lyte 148 on renal blood flow velocity and renal cortical tissue perfusion in healthy volunteers. Ann Surg 2012; 256:18-24. 14. Wilkes NJ, Woolf R, Mutch M, et al. The effects of balanced versus saline-based hetastarch and crystalloid solutions on acid-base and electrolyte status and gastric mucosal perfusion in elderly surgical patients. Anesth Analg 2001; 93:811-816. 15. Mohd Yunos N, Bellomo R, Hegarty C, et al. Association between a chloride-liberal vs chloride-restrictive intravenous fluid administration strategy and kidney injury in critically ill adults. JAMA 2012; 308(15):1566-1572. 16. Shaw AD, Bagshaw SM, Goldstein SL, et al. Major complications, mortality, and resource utilization after open abdominal surgery. Ann Surg 2012; 255:821-829. 17. Base E, Standl T, Mahl C, et al. Comparison of 6% HES 130/0.4 in a balanced electrolyte solution versus 6% HES 130/0.4 in saline solution in cardiac surgery. Crit Care 2006; 10(Suppl1):P176. 18. Kambhampati G, Ross EA, Alsabbagh MM, et al. Perioperative fluid balance and acute kidney injury. Clin Exp Nephrol 2012; 16:730-738. 19. Maitland K, Kiguli S, Opoka RO, et al. Mortality after fluid bolus in African children with severe infection. N Engl J Med 2011; 364:2483-2495. 20. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001; 345:1368-1377.

Page 132

Page 133

133 Bypass to suit the Patient Cardioplegia as a Determinant of Myocardial Damage, in Hospital Mortality, and Long Term Survival post Cardiac Surgery Michael Poullis, BSc(Hons), MBBS, MD, MIEEE, FRCS(CTh), United Kingdom Consultant Cardiac Surgeon, Liverpool Heart and Chest Hospital, Liverpool, United Kingdom Authors: Michael Poullis, BSc(Hons), MBBS, MD, MIEEE, FRCS(CTh), John Chalmers, RDCS, Mark Pullan, FRCS(CTh) and Neeraj Mediratta, FRCS(CTh) Liverpool Heart and Chest Hospital, Liverpool, United Kingdom Abstract Background:The effect of cardioplegia technique - type, route, temperature, timing, and the use of hot shot, on post-operative creatinine kinase (CKMB) release, in hospital mortality and long term survival are unknown. Methods:Analysis of a prospective single institution cardiac surgery database was performed, N=16,806. Univariate and multivariate analysis was performed with regard to CKMB, in hospital mortality and long term survival. Results:Overall mortality was 3.9% for all cases. Univariate analysis in isolated CABG patients demonstrated that warm blood, p<0.001, crystalloid, p<0.001, absence of hot shot, p<0.001 and cross clamp fibrillation, p<0.001 were associated with significantly raised CKMB levels. Retrograde had no benefit, p= 0.76. With regard to CKMB release, warm cardioplegia and isolated antegrade cardioplegia were significant factors for all operations, and crystalloid was also identified for isolated MVR and combined AVR and CABG cases. Hot shot significantly reduced CKMB release post combined AVR and CABG. Post CABG warm cardioplegia and hot shot usage were associated with significantly reduced long term survival. Post AVR isolated antegrade cardioplegia was a significant factor determining long term survival. Post MVR isolated antegrade cardioplegia and warm cardioplegia were significant factors affecting long term survival. Post combined AVR and CABG cardioplegia technique was not identified as a significant factor determining long term survival. Cardioplegia technique was not identified as a significant factor determining in hospital mortality for CABG, AVR, and combined AVR and CABG. Crystalloid cardioplegia was a significant factor determining risk of in hospital death post MVR. Conclusion:The optimal cardioplegia technique depends on the cardiac surgery operation being performed. We would recommend against the use of warm cardioplegia, and crystalloid cardioplegia, and would encourage the use of retrograde cardioplegia as an adjunct to antegrade cardioplegia. The hot shot technique seemed to offer no significant advantages to clinical outcomes.

Page 134

134 Introduction Myocardial function post cardiac surgery is critically dependent on surgical technique and myocardial protection(1). Numerous studies exist with regard to myocardial protection, yet no consensus exists as to an optimal technique(2,3,4). A lack of consensus on cardioplegia exists due to the numerous combinations and permutations involving solution(s), route(s), temperature, timing, volumes, and outcome measures analysed. Differing operative procedures, coronary artery bypass surgery, aortic and mitral valve surgery in isolation or combination increase the difficulty in identifying a potentially optimum cardioplegia technique when meta-analysis is utilised. Numerous comparative trials exist. Few very large studies exist on the subject, and meta-analysis has resulted in conflicting results(5,6). We analysed the effect of technique, route, solution, and the effect of hot shot administration in patients undergoing isolated coronary artery surgery, isolated valve surgery and combined valve and coronary surgery. Methods Institutional review Local institutional review board permission was granted for this retrospective single center study. Study population The characteristics of the study population is shown in table 1. No off pump CABG cases were included in the analyses. Outcome measures Primary outcome was day one post-operative creatine kinase muscle-brain is enzyme (CKMB) expressed in units/L, which was utilised as a surrogate marker of myocardial damage post cardiac surgery(7,8,9), and in hospital death. Secondary outcome was long term survival. Cardioplegia variables analysed The following cardioplegia variables were analysed: cardioplegia verses cross clamp fibrillation (coronary bypass surgery only), blood verses crystalloid cardioplegia, antegrade verses combined antegrade and retrograde, cold verses warm blood, and the use of hot shot (antegrade or retrograde). Linking CKMB release, in hospital death, and long term survival A univariate and multivariate analysis on the effect of CKMB release on in hospital death and long term survival was performed.

Page 135

Page 136

136 was performed with all patients and with the in hospital deaths removed. Cox regression analysis was plotted at the mean of the covariates. All analyses were performed for isolated CABG, isolated AVR and MVR and combined AVR and CABG. Factors included in the multivariate analysis included: age, sex, diabetes, congestive cardiac failure, logistic EuroSCORE, body mass index (kg/m2), cerebrovascular disease, respiratory disease, peripheral vascular disease, dialysis, operative priority, left internal mammary artery use, cardiopulmonary bypass time, cross clamp time, retrograde, antegrade or combined cardioplegia, hot shot technique, cardioplegia temperature (warm or cold), and continuous or intermittent cardioplegia administration. Long term Survival analysis As previously described the national strategic tracing service in the United Kingdom{Fontaine, McShane, et al. 2010 10 /id} was utilised to study the effect of renal failure and impairment on long term survival. Statistical software All statistical analysis was performed with MedCalc for Windows, (version 12.2.1, MedCalc Software, Mariakerke, Belgium). Results Study population Characteristics of the population studied, N=16,806, is shown in table 1, isolated CABG 9,983, isolated AVR N=3,228, isolated MVR N=1,499, combined AVR and CABG N=2,096. Overall mortality was 3.9% (N=653 for all cases). Linking CKMB release, in hospital death, and long term survival An arbitrary cut off for CKMB of 50 was utilised for demonstration purposes. The univariate effect of CKMB release on in hospital death, p<0.0001, and long term survival, p<0.0001 is shown in figure 1A and 1B respectively for the whole study cohort. Multivariate analysis identified CKMB release as a significant factor determining in hospital death (odds ratio 3.5, 95% confidence interval 2.9 to 4.3), p<0.0001, and long term survival, (hazard ratio 1.4, 95% confidence interval 1.3 to 1.5), p<0.0001, figure 1C for the whole study cohort.

Page 137

Page 138

138 release. In hospital mortality and long term survival was not significantly different, p=0.12, and p=0.9 respectively. Cold blood verses warm blood with no hot shot This demonstrated, figure 2C, that warm blood (N=1,316) cardioplegia was associated with a significantly higher CKMB release, p<0.001, compared to cold blood (N=6,268). In hospital mortality and long term survival was not significantly different, p=0.98, and p=0.07 respectively. Cold antegrade blood verses cold antegrade with retrograde blood but no hot shot Addition of retrograde cardioplegia (N=1,500) to antegrade (N=4,827) in isolated CABG cases, with no hot shot, resulted in a significant reduction in CKMB production, p<0.0001, figure 2D. In hospital mortality and long term survival was not significantly different, p=0.2, and p=0.5 respectively. Effect of hot shot on antegrade and combined antegrade and retrograde cardioplegia. Hot shot resulted in a significant reduction in the CKMB release in isolated CABG (N=1,361), p<0.001, figure 2E. In hospital mortality and long term survival was not significantly different, p=0.3, and p=0.07 respectively. Figure 2a Comparison between cross clamp fibrillation, and cardioplegia, Figure 2b Comparison between blood and crystalloid cardioplegia CKMB05101520253035404550Cross clampfibrillationCardioplegiaCKMB05101520253035404550Blood Crytalloid

Page 139

Page 140

140 Post-operative CKMB Linear regression (adjusted R square=0.06) demonstrated that cold cardioplegia, isolated antegrade cardioplegia, diabetes, female sex, ejection fraction, logistic EuroSCORE, cardiopulmonary bypass time, LIMA non usage were significant determinates of released CKMB, table 2. Table 2.Multivariate linear analysis of post-operative CKMB release. Independent variables Coefficient Standard Error t P ISOLATED CABG Cold cardioplegia 7.42 1.71 4.34 <0.0001 Combined Antegrade and retrograde cardioplegia -10.16 1.91 -5.33 <0.0001 Diabetes -2.72 0.83 -3.28 0.0011 Female sex 9.32 1.76 5.29 <0.0001 Good ventricular -2.38 1.15 -2.07 0.04 Logistic EuroSCORE 0.37 0.11 3.30 0.0010 Cardiopulmonary bypass Time 0.33 0.021 15.94 <0.0001 Left internal mammary not used -4.54 2.27 -2.00 0.046 ISOLATED AVR Combined antegrade and retrograde cardioplegia -13.55 2.20 -6.16 <0.0001 Warm cardioplegia 35.45 3.97 8.93 <0.0001 Diabetes -3.48 1.71 -2.03 0.04 Cardiopulmonary bypass Time 0.34 0.024 14.15 <0.0001 ISOLATED MVR Crystalloid cardioplegia 24.08 7.92 3.04 0.002 Combined antegrade and retrograde cardioplegia -22.83 4.52 -5.06 <0.0001 Warm cardioplegia 18.06 5.25 3.44 0.001 Cardiopulmonary bypass time 0.28 0.083 3.32 0.001 COMBINED AVR AND CABG Isolated antegrade cardioplegia 11.24 5.09 2.21 0.03 Cold cardioplegia -31.01 6.30 -4.92 <0.0001 Crystalloid cardioplegia 10.69 4.66 2.29 0.02 Hotshot technique -35.07 4.97 -7.05 <0.0001 Cardiopulmonary bypass time 0.34 0.03 11.29 <0.0001 In hospital mortality Logistic regression of isolated CABG (ROC=0.76, Hosmer-Lemeshow, p=0.79) demonstrated that age, female sex, diabetes, ejection fraction, IABP preoperatively, and logistic EuroSCORE were significant factors determining risk of in hospital death, table 3. Cardioplegia technique was not identified as a significant factor determining in hospital mortality. Table 3. Multivariate logistic analysis of in hospital mortality Variable Odds ratio (OR) 95% CI of OR P

Page 141

Page 142

142 Table 4. Multivariate Cox regression analysis of long term survival Covariate Hazard Risk (HR) 95% CI of HR P ISOLATED CABG Warm cardioplegia 1.36 1.20 to 1.53 <0.0001 Hot shot 1.21 1.07 to 1.37 0.003 Age 1.05 1.04 to 1.06 <0.0001 Congestive cardiac failure 1.32 1.13 to 1.55 0.001 Diabetes 1.65 1.37 to 1.98 <0.0001 Renal impairment 1.37 1.13 to 1.67 0.002 LIMA non usage 1.45 1.28 to 1.64 <0.0001 Ejection Moderate 1.37 1.24 to 1.51 <0.0001 Fraction Poor 1.96 1.68 to 2.29 <0.0001 Cardiopulmonary bypass time 1.01 1.01 to 1.01 <0.0001 Aortic cross clamp time 0.99 0.99 to 1.00 <0.0001 Logistic EuroSCORE 1.02 1.02 to 1.03 <0.0001 ISOLATED AVR Isolated Antegrade 1.29 1.10 to 1.50 0.002 Age 1.03 1.03 to 1.04 <0.0001 Congestive cardiac Failure 1.49 1.19 to 1.88 0.001 Renal impairment 2.22 1.74 to 2.83 <0.0001 Cardiopulmonary time 1.01 1.01 to 1.01 <0.0001 Logistic EuroSCORE 1.01 1.01 to 1.02 <0.0001 ISOLATED MVR Isolated antegrade cardioplegia 1.76 1.40 to 2.23 <0.0001 Cold cardioplegia 0.73 0.56 to 0.97 0.03 Age 1.03 1.02 to 1.04 <0.0001 Congestive cardiac failure 1.78 1.37 to 2.30 <0.0001 Diabetes 1.90 1.12 to 3.21 0.02 Cardiopulmonary bypass time 1.01 1.01 to 1.01 <0.0001 Cross clamp time 0.99 0.99 to 1.00 0.02 Logistic EuroSCORE 1.02 1.02 to 1.03 <0.0001 COMBINED AVR AND CABG Age 1.03 1.02 to 1.04 <0.0001 Congestive cardiac failure 1.65 1.34 to 2.03 <0.0001 Diabetes 1.52 1.12 to 2.05 0.01 Renal impairment 1.30 1.02 to 1.64 0.03 Poor ejection fraction 1.28 1.03 to 1.59 0.02 Cardiopulmonary bypass time 1.01 1.00 to 1.01 <0.0001 Cross clamp time 1.00 0.99 to 1.00 0.01 Body mass index 0.97 0.95 to 0.98 0.0002 Logistic EuroSCORE 1.01 1.00 to 1.02 0.002

Page 143

Page 144

144 In hospital mortality Logistic regression of isolated CABG (ROC=074, Hosmer-Lemeshow test, p=0.06) identified dialysis and logistic EuroSCORE as significant factors determining risk of in hospital death, table 3. Cardioplegia technique was not identified as a significant factor determining in hospital mortality. Long term Survival analysis Multivariate Cox regression analysis (Harrel c-index 0.72) identified isolated antegrade cardioplegia, age, congestive heart failure, renal impairment, cardiopulmonary bypass time, and logistic EuroSCORE, as significant factors determining long term survival, table 4. The effect of cardioplegia route on outcomes is shown in figure 4. Figure 4 The effect of retrograde cardioplegia in addition to antegrade cardioplegia on risk adjusted survival post isolated AVR. Isolated MVR Post-operative CKMB Linear regression (adjusted R square=0.13) demonstrated that isolated antegrade cardioplegia, warm cardioplegia, crystalloid cardioplegia, cross clamp and cardiopulmonary bypass time were significant determinates of released CKMB, table 2. In hospital mortality Logistic regression of isolated CABG (ROC=0.76, Hosmer-Lemeshow test, p=0.65) demonstrated that crystalloid cardioplegia was a significant factor determining risk of in hospital death, table 3. Long term Survival analysis Multivariate Cox regression analysis (Harrel c-index 0.76) identified isolated antegrade cardioplegia, warm cardioplegia, age, congestive cardiac failure, diabetes, cross clamp and cardiopulmonary bypass time, and logistic EuroSCORE as significant factors affecting long term survival, table 4. 0 5 10 15 200102030405060708090100Survival (Years)Survival probability (%)Cardioplegia routeAntegradeCombined Antegrade and retrograde

Page 145

Page 146

146 in hospital death, table 3. Cardioplegia technique was not identified as a significant factor determining in hospital mortality. Long term Survival analysis Multivariate Cox regression analysis (Harrel c-index 0.63), identified age, congestive cardiac failure, diabetes, renal impairment, ejection fraction, cross clamp and cardiopulmonary bypass time, body mass index and logistic EuroSCORE as significant factors determining long term survival, table 4. Discussion Cardioplegia type, route of administration and temperature have a significant effect on CKMB release. After risk adjustment crystalloid cardioplegia was associated with a significantly higher mortality after mitral valve surgery. Warm blood cardioplegia and isolated antegrade cardioplegia administration were associated with a significantly reduced long term survival. Univariate analysis of isolated CABG cases demonstrates that cold crystalloid and warm blood cardioplegia are associated with a significantly higher CKMB release potentially identifying them as inferior techniques. Retrograde cardioplegia and the hot shot technique resulted in a significantly reduced CKMB. Univariate analysis also demonstrated cross clamp fibrillation as a significantly worse technique compared to cardioplegia for isolated CABG cases. The highly heterogeneous risk factors that exist in a cardiac surgery population mean that multivariate analysis needs to be performed. This technique has been utilised previously, however failing to account for known risk factors affecting operative mortality and long term survival means the validity of the results obtained is questionable (10). The importance of elevated CKMB post cardiac surgery being linked to long term survival confirms a recent meta-analysis (11). The heterogeneous nature of cardiac cases also remains a potential source of error in meta-analysis. To account for the heterogeneous nature of the different operations isolated CABG, isolated AVR and MVR and combined AVR and CABG were analysed separately. The finding of a statistically significant CKMB difference does not necessarily mean clinical significance. However CKMB was significantly associated with in hospital mortality, and long term survival, a finding that has been previously described(11). Multivariate analysis demonstrated that cold cardioplegia and the addition of retrograde cardioplegia to antegrade administration was associated with reduced CKMB release. Crystalloid cardioplegia was additionally found to be associated with a significantly increased CKMB release post mitral valve surgery. The hot shot technique was found to be associated with a significantly reduced CKMB release in patients undergoing combined AVR and CABG. Multivariate analysis demonstrated that with the exception of crystalloid cardioplegia for mitral valve procedures, cardioplegia technique did not affect in hospital survival. We speculate that in hospital mortality may be too crude an outcome measure to use for the assessment of the subtle effects of cardioplegia on myocardial protection.

Page 147

147 The effect of cardioplegia on long term survival post cardiac surgery has not been previously investigated. The risk factors that determine long term survival differ between the different the operative interventions CABG, isolated AVR and MVR and combined AVR and CABG, necessitating separate analysis for each procedure. With regard to isolated CABG, warm cardioplegia and the use of hot shot were associated with significantly reduced long term survival. Isolated antegrade cardioplegia was associated with significantly reduced long term survival post isolated AVR and MVR. This finding is concurrent with the CKMB results. Cold cardioplegia was associated with increased long term survival post isolated MVR, compared to warm cardioplegia. Cardioplegia was not found to be a significant factor determining long term survival post combined AVR and CABG. The speculate, based on clinical observation, that the heterogeneous nature of combined AVR and CABG patients may mask the effects of cardioplegia. Hot shot use was at the discretion of the surgeon and interpretation thus needs to be cautious. Univariate analysis suggested that hot shot is a beneficial technique with regard to CKMB release post isolated CABG, however multivariate analysis identified it as a significant factor reducing long term survival on multivariate analysis. Hot shot use is however more frequent in situations where the operating surgeon perceives the case as being a high risk and or long case. An interaction analysis identified a significant interaction with logistic EuroSCORE, but not cross clamp time. Cross clamp was popular prior to the widespread introduction of cardioplegia, however it is perceived to be an inferior technique with regard to myocardial protection. The univariate analysis identified a higher CKMB release in cases utilising this as an operative strategy, however multivariate analysis failed to identify it as a significant factor effecting CKMB release, in hospital mortality or long term survival. To date the majority of the studies on cardioplegia have been univariate in nature. This is the first multivariate analysis of CKMB release, in hospital death and long term survival in a sizeable population. CKMB release and in hospital mortality have been extensively investigated with regard to cardioplegia techniques. As CKMB is significantly associated with long term survival, we speculated that cardioplegia technique may have an impact on long term survival, via CKMB release. We were unable to identify cardioplegia technique as an important risk factor, with the exception of crystalloid cardioplegia in patients undergoing mitral valve surgery, for in hospital death. The optimal cardioplegia technique depends on the cardiac surgery operation being performed. We would recommend against the use of warm cardioplegia, and crystalloid cardioplegia, and would encourage the use of retrograde cardioplegia as an adjunct to antegrade cardioplegia. The hot shot technique seemed to offer no significant advantages to clinical outcomes. Limitations Operative difficulties secondary to patient tissue quality, and adequacy of cardioplegia administration are impossible to quantify and remain potential important confounding factors in this work, as in any other report on cardioplegia. We did not have surgical reason (planned or salvage) or route of hot shot administration recorded in this study. The low value of the R squared for the linear regression

Page 148

148 analysis indicates that the model results need to be interpreted with caution. All surgeons subjecting to this manuscript were subjected to annual CUSUM curve inspection for isolated CABG, isolated AVR, isolated MVR, and combined AVR and CABG procedures. References 1. Vahasilta T, Malmberg M, Saraste A et al. Cardiomyocyte apoptosis after antegrade and retrograde cardioplegia during aortic valve surgery. Ann.Thorac.Surg.2011;92:1351-1357. 2. Caputo M, Santo KC, Angelini GD et al. Warm-blood cardioplegia with low or high magnesium for coronary bypass surgery: a randomised controlled trial. Eur.J.Cardiothorac.Surg.2011;40:722-729. 3. Ovrum E, Tangen G, Tollofsrud S, Oystese R, Ringdal MA,Istad R. Cold blood versus cold crystalloid cardioplegia: a prospective randomised study of 345 aortic valve patients. Eur.J.Cardiothorac.Surg.2010;38:745-749. 4. Poncelet AJ, van Steenberghe M, Moniotte S et al. Cardiac and neurological assessment of normothermia/warm blood cardioplegia vs hypothermia/cold crystalloid cardioplegia in pediatric cardiac surgery: insight from a prospective randomized trial. Eur.J.Cardiothorac.Surg. 2011;40:1384-90. 5. Fan Y, Zhang AM, Xiao YB, Weng YG,Hetzer R. Warm versus cold cardioplegia for heart surgery: a meta-analysis. Eur.J.Cardiothorac.Surg.2010;37:912-919. 6. Jacob S, Kallikourdis A, Sellke F,Dunning J. Is blood cardioplegia superior to crystalloid cardioplegia? Interact.Cardiovasc.Thorac.Surg.2008;7:491-498. 7. Hedstrom E, Astrom-Olsson K, Ohlin H et al. Peak CKMB and cTnT accurately estimates myocardial infarct size after reperfusion. Scand.Cardiovasc.J.2007;41:44-50. 8. Sellgren A, Nilsson F,Jeppsson A. The relationship between ASAT, CKMB, troponin-T and mortality after cardiac surgery. Scand.Cardiovasc.J.2007;41:386-390. 9. Shand JA , Howe A. CKMB and troponin I levels post coronary artery bypass surgery show a graded association with mortality. Biomark.Med.2011;5:375. 10. Flack JE, III, Cook JR, May SJ et al. Does cardioplegia type affect outcome and survival in patients with advanced left ventricular dysfunction? Results from the CABG Patch Trial. Circulation.2000;102:III84-III89. 11. Domanski MJ, Mahaffey K, Hasselblad V et al. Association of myocardial enzyme elevation and survival following coronary artery bypass graft surgery. JAMA.2011;305:585-591.

Page 149

149 SESSION 11 Tuesday 15:45 – 17:00 Free Papers Low State Entropy Scores on Cardiopulmonary Bypass and Association with Mortality and Major Morbidity. Keshavan Kanesalingam*, FANZCA, Orison Kim**, FANZCA, and Katrina A Kanesalingam***, FANZCA. * Department of Anaesthesia, Westmead Hospital, Sydney, Australia, **Department of Anaesthesia, Westmead Hospital, Sydney, Australia, ***Department of Anaesthesia, Prince of Wales Hospital, Sydney, Australia. Background Postoperative mortality and morbidity has been associated with cumulative anaesthetic duration below a processed electroencephalographic (EEG) threshold of 45. This study sought to better quantify exposure to low processed EEG scores by looking at not only the duration the processed EEG was <45 but also the extent to which it was <45. This study specifically looked at processed EEG scores (State Entropy) whilst on cardiopulmonary bypass (CPB). Methods 211 patients who underwent cardiac surgery at Westmead Hospital in 2011 were studied until hospital discharge. Primary outcome was all cause mortality. Secondary outcomes were new onset stroke, post-operative ventilatory time > 48 hours, new renal failure requiring dialysis, reoperation and mediastinitis. Exposure to low State Entropy scores whilst on CPB was quantified by graphing the entropy scores versus time and calculating the area under the curve (AUC) below an entropy score of 45. Results When comparing median AUC values there was a statistically significant association between death (10207 v 1235 [P=0.017]), the need for ventilation >48hrs (2234 v 1202 [P= 0.025]), the need for reoperation (2645 v 1202 [P=0.022]) and the incidence of mediastinitis (5353 v 1235 [P=0.017]) with increased AUC values (an increased AUC corresponds to increased exposure to processed EEG scores < 45). There was also a statistically significant association between having any one or more of the complications and an increased AUC value (2085 v 1106 [P=0.002]) Conclusion Exposure to low processed EEG scores whilst on CPB is associated with postoperative mortality and morbidity.

Page 150

Page 151

151 Free Papers Comparison of EurosSCORE, EuroSCORE II and AusSCORE for Isolated Coronary Artery Bypass Grafting in New Zealand TKM Wang1 MBCHB, AY Li1 MBCHB, T Ramanathan1 PhD, FRACS, GD Gamble2 MSc, RAH Stewart1,2 MD, FCSANZ, HD White1,2 DSc, FCSANZ 1 Green Lane Cardiovascular Service, Auckland City Hospital, Auckland, New Zealand, 2 Department of Medicine, University of Auckland, Auckland, New Zealand Background For over a decade, EuroSCORE had been the most influential scoring system for outcome prediction following cardiac surgery. Recently, a revised EuroSCORE II and AusSCORE, a system based on an Australasian population, were developed to improve calibration. Our study compared EuroSCORE I, EuroSCORE II and AusSCORE for predicting outcomes after coronary artery bypass grafting (CABG). Methods Between July 2010 and June 2012, all isolated CABG patients at Auckland City Hospital had their EuroSCORE I, EuroSCORE II and AusSCORE retrospectively calculated. Discrimination and calibration of scores for outcomes were assessed. Results 818 patients were followed-up for 1.4+/-0.6 years. Mean EuroSCORE I, EuroSCORE II and AusSCORE were 4.5+/-5.0%, 2.6+/-3.1% and 0.9+/-1.3% respectively. 30-days mortality rate was 1.6% (n=13). The C-statistic of EuroSCORE I, EuroSCORE II and AusSCORE for 30-days mortality were 0.675 (95% confidence interval 0.531-0.819), 0.642 (0.503-0.780) and 0.661 (0.516-0.807), while the Hosmer-Lemshow tests for calibration were p=0.061, 0.150 and 0.420 respectively. Mortality rate at follow-up was 2.9% (n=24). C-statistics of the three scores for mortality at follow-up were 0.639 (0.525-0.752), 0.604 (0.483-0.752) and 0.593 (0.480-0.705). EuroSCORE I was the best model at detecting stroke (c=0.736) and ventilation>24 hours (c=0.712), and EuroSCORE II for deep sternal wound infection (c=0.720) and return to theatre (c=0.626). Conclusion EuroSCORE II and AusSCORE have slightly better calibration, but were not superior to EuroSCORE I at discriminating outcomes for isolated CABG. Revising risk models to improve calibration is important to reflect contemporary surgical outcomes. Room for improvement, however, is limited for discrimination.

Page 152

Page 153

153 Free Papers Real-time Continuous Pulse Oximetry Monitoring during Normothermic Pulsatile Perfusion Yves D Durandy, MD, France Perfusion and Intensive Care Unit Department, CCML Le Plessis-Robinson, France. Background Although there is an ongoing debate regarding the risks and benefits of pulsatile perfusion (PP) most of the experimental works demonstrated improved end-organ microcirculation when comparing pulsatile to non-pulsatile flow. Methods : Since 2010, we have used normothermic PP for every patient during non-beating heart cardiopulmonary bypass. PP was performed with a S5 heart lung machine (Sorin group, Munchen, Germany). A finger sensor using Massimo Signal Extraction Technology was used to measure pulse oximetry. Data was displayed on the multi-parameter patient monitor. Results 848 patients were treated with PP. PP proved to be safe and efficient without extra cost. In every case, PP allows us to: - monitor pulse oximetry during aortic cross-clamping period, - choose the best plethysmographic signal by modifying arterial pump PP parameters, - check plethysmographic signal quality in several sites, - use the lowest FIO2 able to maintain pulse oximetry between 97 and 99 % saturation. Discussion The next step is to demonstrate that during PP, perfusion index displayed on Sp02 monitors could be considered a surrogate for measurement of energy equivalent pressure delivered by PP. Real time monitoring of pulse oximetry has to be compared to near-infrared spectroscopy, a more expensive technique of regional tissue oxyhaemoglobin saturation. Conclusion Good peripheral perfusion is likely to be associated with good end-organ perfusion. In this current era of expensive medicine, we expect that this un-expensive technique with no need for major resource utilisation will have the potential to improve patient outcomes.

Page 154

Page 155

155 Free Papers Does changing the Priming Fluid of the Heart-Lung Machine have Clinical Effects? Matthew D Haydock* CertSci, Cornelius Kruger** FANZCA, Timothy Willcox*** CCP, David A Haydock* FRACS * Green Lane Cardiothoracic Surgical Unit, Auckland City Hospital, Auckland, New Zealand, ** Green Lane Cardiothoracic Anaesthesia, Auckland City Hospital, Auckland, New Zealand,***Green Lane Perfusion, Auckland City Hospital, Auckland, New Zealand Background Auckland Hospital Cardiothoracic unit recently removed Mannitol and Voluven from its plasma-lyte based cardiopulmonary bypass (CPB) priming fluid. Any change to practice benefits from comprehensive audit to identify positive or negative effects. The aim of this retrospective analysis is to investigate the effect of changing the CPB prime constituents on fluid balance and clinical outcome parameters. Methods Records were reviewed for 100 consecutive patients undergoing primary, isolated coronary artery bypass grafting (CABG), 50 patients before the prime change and 50 after. Data was collated into a central database for analysis. Results Mean arterial pressure on bypass was higher in the new prime group (61.5mmHg vs. 57.5mmHg, p=0.002). Peak haemoglobin during surgery was higher in the new prime group compared to the old prime group (99.7g/L vs. 93.8g/L, p=0.043), however minimum haemoglobin, haematocrit and requirement for blood products during surgery were not significantly different. There was no significant difference in serum sodium, potassium or creatinine post-operatively between groups. There were no significant differences regarding important outcomes such as post-operative weight and fluid balance, time on ventilation, length of stay in ICU or hospital and mortality. Interestingly, the new prime group spent a smaller proportion of their time in ICU on mechanical ventilation. Conclusion This study demonstrates that removing Mannitol and Voluven from priming fluid does not have detrimental effect on electrolytes, fluid status and other important outcomes in primary isolated CABG surgery. The risk-benefit balance combined with the obvious economical benefit clearly favours removing Mannitol and Voluven from priming fluids.

Page 156

Page 157

157 SESSION 12 Tuesday 17:10 – 18:00 Influencing Change and Outcomes Formula 1 Racing, Red Dogs and the Green Lane Way Alan F. Merry, FANZCA, New Zealand Professor and Head of the School of Medicine, Auckland City Hospital, Auckland, New Zealand Authors: Alan F. Merry, Jennifer Weller and Simon J. Mitchell School of Medicine, Auckland City Hospital, Auckland, New Zealand This is the conclusion to the Session 4 presentation ‘Teamwork, Communication, Formula-One Racing and the Outcomes of Cardiac Surgery’ (see page 39) Introduction Many years ago a visiting resident to Green Lane from the United States first said “Red dog!” when he meant “Let’s go onto bypass.” The phrase stuck, at least for some practitioners of that era. Its adoption as implicitly understood code reflected a well-developed approach to teamwork that may not have involved current day jargon but did follow many principles now being taught as new and innovative. For example, the “sterile cockpit” was normal when AFM began his career – little idle chatter was allowed, and full attention on the task in hand was expected from everyone. Closed loop communication has always been expected for key procedures such as giving heparin or protamine. And people did not need introductions because everyone knew each other. “The Green Lane Way” was understood by all, and there was considerable standardisation – even down to the way arms were tucked by patients’ sides. Transfer to the intensive care provides a great example of how much change has taken place. In those early days it was done without any monitoring other than a finger on the pulse, and although there was handover to a nurse and a resident, the senior surgeon and anaesthesiologist continued to be responsible for patient care so did not need to be given any new information at the point of transfer. More recently, the complexity of transferring highly monitored paediatric patients with numerous drug infusions and possibly other forms of therapy to the care of intensivists has prompted paediatric cardiac surgeon Marc de Leval and his team (in the UK) to seek the assistance of airline pilots and formula one racing teams to develop a more organised and standardised approach to hand over 46.

Page 158

158 As indicated in the introduction, in this second part of the paper we will outline the elements of teamwork and discuss five ways to improve the performance of teams in cardiac surgery. Elements of teamwork On the basis of empirical evidence from various settings, Salas et al 47 have proposed a model for teamwork, which includes five dimensions: team orientation, team leadership, mutual performance monitoring, backup behaviour and adaptability. These dimensions are underpinned by three coordinating factors: mutual trust; effective communication; and shared mental models within the team. Method 1 – subspecialise and replace tribes with teams You can’t expect teamwork if people don’t understand (and agree) that they belong to the team 48. In healthcare, teams tend to form or a particular purpose, and then disband. Teams in cardiac surgery are often more stable than in many other parts of healthcare, but still typically less stable than a hockey or basketball team. The skills and knowledge that are required to care for surgical patients come from different disciplines: surgery, anaesthesia, perfusion and nursing – supported by technical, administrative and other staff, all of whom contribute in important ways. It is not usually possible, today, for a member of one group to take over and adequately perform the duties of a person from another. This is different from the present day situation in an aeroplane cockpit where the pilot and co-pilot can do each other’s work, and where a single authority gradient applies. Health professionals typically train in silos – medical schools, nursing schools and so on. Thus the healthcare setting tends to promote a sense of “tribalism”, in which a nurse or anaesthesiologist may identify more with large departments of nursing or anaesthesiology than with a cardiothoracic unit. This may lead to differences in priorities, and in the way individuals communicate and view issues of leadership and teamwork 49 50. The extent to which tribalism of this type usurps the perception of belonging to a multidisciplinary team does vary. In the days of Sir Brian, the Green Lane CTSU had a very strong sense of identity. In part this may be because it was situated in a relatively small hospital. With the move to the much larger Auckland City Hospital, and with the other developments outlined above, this has been at least slightly diluted. One aspect of this question relates to the degree to which clinicians specialise in a particular field. For anaesthesiologists and nurses there is a tension between the flexibility provided by generalist staff, and the advantages of subspecialisation. However, in a field as complex and challenging as cardiac surgery, subspecialisation is clearly an advantage for all concerned, both because it increases skills (transoesophageal echocardiography is a clear example of this point) but also because it increase the time the team members spend together and therefore the effectiveness of teamwork. It is noteworthy that the 2001 report on the inquiry into paediatric cardiac surgery at the Bristol Royal Infirmary51 supported the concept of patient centred multi-disciplinary teams. This year (2013) a decision has been made by the department of anaesthesiology at the Green Lane unit to require a substantial minimum time commitment to the unit. This can be expected to improve teamwork in particular and performance in general.

Page 159

159 Method 2 – Sort out the leadership while flattening the gradients of authority Speaking up is pivotal to safety, and this may need the most junior person in the room to ask a question that might seem silly, such as “Aren’t we meant to be operating on the other side of this patient?” Steep authority gradients and a strongly hierarchical culture tend to inhibit speaking up, and so does tribalism because individuals may limit their perceived responsibility to the boundaries of their professional group’s area of practice. At the same time, it does need to be clear who is in charge in relation to any particular decision. Hospitals operate through various models of leadership, and there is often some degree of mistrust between the “tribes” of healthcare. The principle that the surgeon should automatically be the ’captain of the ship’” is no longer tenable. Today, every healthcare worker is held accountable for his or her actions, and as explained, skills are not necessarily interchangeable between groups. In the OR, leadership should alter depending on the decision in question. Bleakely 52 introduced the idea of democracy in healthcare teams, advocating open conversations and the encouragement of suggestions from other members of the team. However, making decisions in real time in the operating room on the basis of a vote does not appeal, and the term “consultative” may be more appropriate. Each person’s expertise needs to be respected, and all members of the team need the opportunity to comment and contribute, but if the matter is obviously anaesthesia related, the senior anaesthesiologist should make the final call, and so on. The overall leadership of a unit is a different matter, and a greater degree of democracy may well be effective in this context. A governance structure with representation from each professional subgroup working collectively under an agreed chairperson (who could come from any subgroup – particularly if a rotating fixed-term approach is adopted, so that change from group to group can be reasonably anticipated) would seem to be ideal. A more traditional leadership model with an open-ended appointment of a designated leader (often a surgeon) would also be functional (as illustrated by Green Lane in Sir Brian’s time), but perhaps harder to impose on groups that have become used to “tribal autonomy”. Surprisingly, a clear structure for overall governance is sometimes absent from cardiac and other surgical units, and instead each discipline may tend to meet and make decisions in isolation from members of other disciplines. It seems to be a matter of common sense that a multidisciplinary committee should meet regularly (at least monthly, probably weekly), make explicit decisions about policy and strategy, and deal with problems as they arise. In this way standardised approaches to common problems could be developed and the mental models for policy, direction and clinical care could become mutually agreed, and therefore shared. Method 3 – introduce explicit training in effective communication The principles of effective communication include clarity, comprehensiveness, and the principle of confirming that messages have been received (i.e., “closing the loop”). Communication should be directed to named individuals and to the room in general (a technique which can be refined to the point where it warrants the title “hint and hope”). It follows that people’s names should be used, and this is also a matter of mutual respect. This is why introductions are a key element of the WHO Surgical Safety Checklist. It is helpful for names also to be written on a whiteboard. Graded assertiveness is a skill that facilitates speaking up in particular and effective

Page 160

160 communication in general. All of these skills can and should be taught – as with canullating an aorta or intubating a trachea it is unlikely that real proficiency will be achieved by simply expecting these skills to be acquired by some form of informal absorption 53 . Simulation is now an established method of teaching teamwork and communication in health care.54-58 In a recent study simulation was used to improve the management of acute crises in cardiac surgery 59. Simulation is fairly costly, but drawing from the experience of airlines, it seems that its wider adoption by whole teams working together is overdue. Method 4 – Use checklists, briefings and debriefings… and engage in the process The WHO Surgical Safety Checklist should accepted as a standard of care 40-42 60 but its use will only be effective if all concerned engage fully in the process. Briefings at the beginning of every surgical list have huge potential to improve efficiency, set a positive and collegial tone, and avoid harm. Short debriefing sessions at the end of the day, to which all should contribute, can identify areas for improvement while reinforcing effective behaviours. Method 5 - Promote a culture of respect alongside a commitment to excellence and a focus on patients A particularly interesting study by Curry et al identified a commitment to excellence as a key element of achieving good outcomes for acute myocardial infarction 61. This is entirely aligned with the culture of Green Lane Hospital in the days of Sir Brian – for its day, the commitment to excellence was evident and exceptional. Within a culture of excellence there should also be a commitment to respect. Courtesy should not be negotiable. The basic assumption should be that everyone is skilled and knowledgeable and fully committed to working together to provide the best possible care for patients. If this proves to be untrue for a particular individual, the problem should be addressed formally rather than through verbal abuse or other forms of bullying. The latter approach is antithetical to the promotion of teamwork. In the study of Curry et al, the following illuminating comments were reported: “We don’t accept anything less than the very best.” “[Staff here have] a very, very strong work ethic. . . . If you didn’t intend to work in a similar fashion, this isn’t a good place for you. . . . They are very careful in their selection from the very beginning. Success breeds success. . . You have to fit into the culture” Method 6 – Focus on the performance of the team… not on individuals There is a strong case for emphasising and celebrating the performance of the team rather than of individuals within it. In this context we are very much opposed to the public dissemination of individual surgeon’s mortality data. The data provided by the The Society for Cardiothoracic Surgery in Great Britain & Ireland (SCTS: www.scts.org) is illustrative of one reason for this. There will always be differences

Page 161

161 between the mortality rates achieved by individual surgeons within a given unit. Funnel plots are used to make the point that most of these individuals’ results lie within acceptable limits, but in fact surgeons with low numbers of cases might well have allegedly “acceptable” rates that would not be acceptable for surgeons with high caseloads. An informed patient might well feel that it would be preferable to choose one of the latter surgeons, but few systems would accommodate choice at that level. More importantly, these data discount the contribution of the other members of the team. It has been established that individual anaesthesiologists can influence the outcomes of cardiac surgery 62 63, yet their results are seldom published. In fact, outcome depends on all members of the team working together, efficiently and effectively. The best way to promote this is surely to publicise the results of the unit as a whole, in comparison with national averages or percentile data. This should be accompanied by internal (confidential) review of the results of individuals, and an assurance to the public that all surgeons and anaesthesiologists (and perhaps perfusionists) are functioning within the acceptable limits of the funnel plot. If all concerned were focused on improving the unit’s results, then idiosyncratic behaviour, predicated on the “hero” model and the (statistically impossible) belief by each person that he or she is (or at least should be) the best, could be replaced by a collective approach to excellence. In such an approach practitioners would surely be more inclined to help each other with difficult cases and to address issues related to low volumes or poor performance from particular individuals. They are also more inclined to feel comfortable with a level of transparency that is fair and appropriate. Conclusion The story of the Green Lane CTSU provides some interesting messages about excellence and teamwork. Results have improved around the world, but the risk of cardiac surgery is still considerable. Enhanced teamwork provides considerable potential for achieving better results, and there are probably few units today in which some aspect of teamwork could not be improved. Acknowledgements We have drawn on material included in the following sources, with which there is some overlap of conceptual content: Merry AF and Weller J. Teamwork and Minimizing Error in Cardiac and Thoracic Surgery. In Myles P and Alston P (eds). Oxford Textbooks in Anaesthesia: Cardiothoracic Anaesthesia. Submitted. Merry AF, Weller J and Mitchell SJ. Patient Safety in Cardiac Anesthesia. Journal of Cardiothoracic and vascular anesthesia. Submitted.

Page 162

162 References 46. Catchpole KR, de Leval MR, McEwan A, Pigott N, Elliott MJ, McQuillan A, MacDonald C, Goldman AJ. Patient handover from surgery to intensive care: using Formula 1 pit-stop and aviation models to improve safety and quality. Paediatr. Anaesth. 2007;17:470-8. 47. Salas E, Sims DE, Burke CS. Is there a "Big Five" in teamwork? Small Group Research 2005;36:555-99. 48. Burford B. Group processes in medical education: learning from social identity theory. Medical Education 2012;46:143–52. 49. Hall P. Interprofessional teamwork: Professional cultures as barriers. Journal of Interprofessional Care 2005;19:188-96. 50. Weller J. Shedding new light on tribalism in health care. Medical Education 2012;46:134-36. 51. Inquiry T. The Inquiry into the management of care of children receiving complex heart surgery at the Bristol Royal Infirmary. Bristol, 2001. 52. Bleakley A. Social comparison, peer learning and democracy in medical education. Medical Teacher 2010;32:878-79. 53. Pian-Smith MC, Simon R, Minehart RD, Podraza M, Rudolph J, Walzer T, Raemer D. Teaching residents the two-challenge rule: a simulation-based approach to improve education and patient safety. Simulation in Healthcare: The Journal of The Society for Medical Simulation 2009;4:84-91. 54. Joy BF, Elliott E, Hardy C, Sullivan C, Backer CL, Kane JM. Standardized multidisciplinary protocol improves handover of cardiac surgery patients to the intensive care unit. Pediatr Crit Care Med 2011;12:304-8. 55. Burtscher MJ, Kolbe M, Wacker J, Manser T. Interactions of team mental models and monitoring behaviors predict team performance in simulated anesthesia inductions. J Exp Psychol Appl 2011;17:257-69. 56. Weller J, Frengley R, Torrie J, Shulruf B, Jolly B, Hopley L, Hendersdon K, Dzendrowskyj P, Yee B, Paul A. Evaluation of an instrument to measure teamwork in multidisciplinary critical care teams. Qual Saf Health Care 2011;20:216-22. 57. Frengley RW, Weller JM, Torrie J, Dzendrowskyj P, Yee B, Paul AM, Shulruf B, Henderson KM. The effect of a simulation-based training intervention on the performance of established critical care unit teams. Crit. Care Med. 2011;39:2605-11. 58. Ruel M, Labinaz M. Transcatheter aortic-valve replacement: a cardiac surgeon and cardiologist team perspective. Curr. Opin. Cardiol. 2010;25:107-13. 59. Stevens L-M, Cooper JB, Raemer DB, Schneider RC, Frankel AS, Berry WR, Agnihotri AK. Educational program in crisis management for cardiac surgery teams including high realism simulation. J. Thorac. Cardiovasc. Surg. 2012;144:17-24. 60. Birkmeyer JD. Strategies for improving surgical quality--checklists and beyond. N Engl J Med 2010;363:1963-5. 61. Curry LA, Spatz E, Cherlin E, Thompson JW, Berg D, Ting HH, Decker C, Krumholz HM, Bradley EH. What distinguishes top-performing hospitals in acute myocardial infarction mortality rates? A qualitative study. Ann. Intern. Med. 2011;154:384-90. 62. Merry AF, Ramage MC, Whitlock RML, Laycock GJA, Smith W, Stenhouse D, Wild CJ. First-time coronary artery bypass grafting: the anaesthetist as a risk factor. Br. J. Anaesth. 1992;68:6-12. 63. Slogoff S, Keats AS. Does perioperative myocardial ischemia lead to postoperative myocardial infarction? Anesthesiology 1985;62:107-14.