Critical Care Patients: Difficult Intubation, Airway Management, and Other Considerations , Apuntes de Filología Inglesa

¡Descarga Critical Care Patients: Difficult Intubation, Airway Management, and Other Considerations y más Apuntes en PDF de Filología Inglesa solo en Docsity! Edited by Fang Gao 5mith Associate editor Joyce reung y Core Topics in Critical Care Medicine This page intentionally left blank Core Topics in Critical Care Medicine Edited by Fang Gao Smith Professor in Anaesthesia, Critical Care Medicine and Pain, Academic Department of Anaesthesia, Critical Care and Pain, Heart of England NHS Foundation Trust, Clinical Trials Unit, University of Warwick, UK Associate editor Joyce Yeung Anaesthetic Specialist Registrar, Warwickshire Rotation, West Midlands Deanery and Research Fellow, Academic Department of Anaesthesia, Critical Care and Pain, Heart of England NHS Foundation Trust, UK CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi, Dubai, Tokyo Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK First published in print format ISBN-13 978-0-521-89774-7 ISBN-13 978-0-511-71311-8 © Fang Gao Smith and Joyce Yeung 2010 Every effort has been made in preparing this publication to provide accurate and up-to-date information which is in accord with accepted standards and practice at the time of publication. Although case histories are drawn from actual cases, every effort has been made to disguise the identities of the individuals involved. Nevertheless, the authors, editors and publishers can make no warranties that the information contained herein is totally free from error, not least because clinical standards are constantly changing through research and regulation. The authors, editors and publishers therefore disclaim all liability for direct or consequential damages resulting from the use of material contained in this publication. Readers are strongly advised to pay careful attention to information provided by the manufacturer of any drugs or equipment that they plan to use. 2010 Information on this title: www.cambridge.org/9780521897747 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate. Published in the United States of America by Cambridge University Press, New York www.cambridge.org eBook (NetLibrary) Hardback Contents List of contributors page vii Foreword by Julian Bion ix Preface xi Acknowledgements xii List of abbreviations xiii Section I Specific features of critical care medicine 1 1 Recognition of critical illness 1 Edwin Mitchell 2 Advanced airway management 6 Isma Quasim 3 Patient admission and discharge 16 Santhana Kannan 4 Transfer of the critically ill 21 Gavin Perkins 5 Scoring systems and outcome 27 Roger Stedman 6 Information management in critical care 34 Roger Stedman 7 Haemodynamics monitoring 40 Anil Kumar and Joyce Yeung 8 Critical care imaging modalities 49 Frances Aitchison 9 Vasoactive drugs 58 Mamta Patel 10 Nutrition 67 Yasser Tolba 11 Pain control 72 Edwin Mitchell 12 Sedation 77 Joyce Yeung 13 Ethics 85 John Bleasdale 14 Organ donation 91 Angeline Simons and Joyce Yeung Section II Systemic disorders and management 99 15 Sepsis 99 Yasser Tolba and David Thickett 16 Multiple organ failure 108 Zahid Khan 17 Immunosuppressed patients 116 Tara Quasim 18 Principles of antibiotics use 124 Edwin Mitchell 19 Fluid and electrolyte disorders 130 Prasad Bheemasenachar 20 Acid–base abnormalities 148 Prasad Bheemasenachar 21 Post-operative critical care 159 Prasad Bheemasenachar 22 Post-resuscitation care 170 Gavin Perkins Section III Organ dysfunction and management 177 23 Bleeding and clotting disorders 177 Nick Murphy 24 Acute coronary syndromes 185 Harjot Singh and Tony Whitehouse 25 Cardiac arrhythmias 194 Khai Ping Ng and George Pulikal 26 Acute heart failure 202 Harjot Singh and Tony Whitehouse v Gavin Perkins Associate Clinical Professor in Critical Care Medicine Birmingham Heartlands Hospital West Midlands Critical Care Research Network Birmingham, UK Khai Ping Ng Medical Specialist Registrar Birmingham Heartlands Hospital West Midlands Critical Care Research Network Birmingham, UK Elinor Powell Anaesthetic Specialist Registrar/Research Fellow Birmingham Heartlands Hospital West Midlands Critical Care Research Network Birmingham, UK George Pulikal Medical Specialist Registrar Derriford Hospital Plymouth, UK Isma Quasim Consultant Anaesthetist Golden Jubilee Hospital Scotland, UK Tara Quasim Senior Lecturer Glasgow Royal Infirmary Glasgow, UK Nick Sherwood Consultant Intensivist Birmingham City Hospital West Midlands Critical Care Research Network Birmingham, UK Angeline Simons Medical Specialist Registrar Birmingham Heartlands Hospital West Midlands Critical Care Research Network Birmingham, UK Harjot Singh Consultant Anaesthetist University Hospital Birmingham West Midlands Critical Care Research Network Birmingham, UK Richard Skone Paediatric Intensive Care Registrar Birmingham Children’s Hospital Birmingham, UK Catherine Snelson Medical Specialist Registrar/Advanced Trainee in Intensive Care Medicine Birmingham Heartlands Hospital West Midlands Critical Care Research Network Birmingham, UK Roger Stedman Consultant Intensivist Birmingham Heartlands Hospital West Midlands Critical Care Research Network Birmingham, UK David Thickett Senior Lecturer in Respiratory Medicine University Hospital Birmingham West Midlands Critical Care Research Network Birmingham, UK Yasser Tolba Consultant Intensivist Birmingham Heartlands Hospital West Midlands Critical Care Research Network Birmingham, UK Bill Tunnicliffe Consultant Intensivist University Hospital Birmingham West Midlands Critical Care Research Network Birmingham, UK Sandeep Walia Consultant Anaesthetist University Hospital Birmingham West Midlands Critical Care Research Network Birmingham, UK Tony Whitehouse Consultant Intensivist University Hospital Birmingham West Midlands Critical Care Research Network Birmingham, UK Joyce Yeung Anaesthetic Specialist Registrar/Research Fellow Birmingham Heartlands Hospital West Midlands Critical Care Research Network Birmingham, UK List of contributors viii Foreword The range of chapter titles in this concise and focussed textbook demonstrates how far intensive care medi- cine has travelled along the road from the first steps of providing co-located care for patients with a single disease – respiratory paralysis from polio – to becom- ing a speciality caring for patients with life-threatening disease of multiple organ systems. The outcome from the polio epidemics of the 1950s was transformed by the anaesthetist Professor Bjorn Ibsen, who reduced the mortality from 90% to 40% by combining labo- ratory science with applied physiology to change the way care was delivered – from iron lung respirator to positive pressure ventilation via a cuffed tracheostomy tube. The ‘power supply’ (medical students) was soon replaced by the development of mechanical ventila- tors, and the scientific innovation – arterial blood gas measurement – rapidly become a standard investiga- tion in any acutely ill patient. Although polio has virtually disappeared, intensive care was retained by hospitals convinced of its apparent utility for suppor- ting patients with an increasingly diverse mix of diseases. In the Western world at least, we now care for patients with a substantial chronic disease burden underlying their acute illness, and intensive care has increasingly come to resemble general medical prac- tice with its accompanying ethical issues for individuals and for society. Indeed, Professor Henry Lassen’s data describing the polio epidemic (Lancet 1953) demonstrated that although the new technique of ventilatory management saved many lives, those who eventually died did so much later: intensive care has the capacity to delay, but not always prevent, death. As a new multi-disciplinary speciality we have many challenges and opportunities ahead, from understanding the cellular mechanisms of organ dys- function and sepsis to improving the reliability and safety of care delivered across multiple transitions in time, place and staff. The modern intensivist must combine many roles: compassionate clinician, scien- tist, educator and team leader amongst them. For those wishing to participate, the experience will be demanding and rewarding. This textbook provides a sound basis for that journey. Professor Julian Bion MBBS FRCP FRCA MD Professor of Intensive Care Medicine University Department of Anaesthesia and Intensive Care Medicine Royal College of Anaesthetists Chair, Professional Standards Committee Chair, European Board of Intensive Care Medicine ix Abbreviations A&E Accident and Emergency ABC Airways, Breathing, Circulation ABCDE airways, breathing, circulation, disability, exposure ABGs arterial blood gases ABS analgesic-based sedation ACE angiotensin-converting enzyme ACE-Is angiotensin-converting enzyme inhibitors ACPE acute cardiogenic pulmonary oedema ACS acute coronary syndrome ACT activated clotting time ACTH adrenocorticotropic hormone ADH antidiuretic hormone AED anti-epileptic drug AEP auditory evoked potentials AF atrial fibrillation AFE amniotic fluid embolism AG anion gap AHA American Heart Association AHF acute heart failure AIDF acute inflammatory demyelinating polyneuropathy AIDS acquired immunodeficiency syndrome AIS abbreviated injury scoring AKI acute kidney injury ALERT™ Acute Life Threatening Events – Recognition and Treatment ALF acute liver failure ALI acute lung injury ANP atrial (A type) natriuretic peptide AoCLD acute on chronic liver disease APACHE acute physiology and chronic health evaluation APC activated protein C APH antepartum haemorrhage APRV airway pressure release ventilation APTT acitvated partial thromboplastin time ARB angiotensin II receptor antagonist ARDS acute respiratory distress syndrome ARF acute renal failure ARR absolute risk reduction ASV adaptive support ventilation AT antithrombin ATC automated tube compensation ATLS Advanced Trauma Life Support Course ATN acute tubular necrosis ATP adenosine triphosphate ATS American Thoracic Society AV atrioventricular AVNRT atrioventricular nodal re-entrant tachycardia AVPU patient is alert, responding to voice, responding to pain, unresponsive AVRT atrioventricular re-entrant tachycardia BBB bundle branch block BIS bispectral index BMI body mass index BMR basal metabolic rate BNP brain (B type) natriuretic peptide BOOP bronchiolitis obliterans organizing pneumonia bpm beats per minute CABG coronary artery bypass grafting CAD coronary artery disease CAMP cyclic adenosine monophosphate CBF cerebral blood flow CBV cerebral blood volume CCK cholecystokinin CCO critical care outreach CCRISP™ Care of the Critically Ill Surgical Patient CEMACH confidential enquiries into maternal and child health CI cardiac index CI confidence interval CIRCI critical illness related corticosteroid insufficiency CK creatine kinase CMV continuous mandatory ventilation xiii CNS central nervous system CoBaTrICE Competency Based Training in Intensive Care Medicine in Europe COPD chronic obstructive pulmonary disease COX cyclo-oxygenase CPAP continuous positive airway pressure CPFA coupled plasma filtration absorption CPP cerebral perfusion pressure CPR cardiopulmonary resuscitation CrCU Critical Care Unit CRF chronic renal failure CSF cerebrospinal fluid C-spine cervical spine CSS Canadian Society Classification of Angina CT computed tomography CVA cerebrovascular accident CVP central venous pressure CVVH continuous veno-venous haemofiltration CVVHD continuous veno-venous haemodialysis CVVHDF continuous veno-venous haemodiafiltration CXR chest X-ray DAI diffuse axonal injury DI diabetes insipidus DIC disseminated intravascular coagulation DVT deep vein thrombosis EBV Epstein–Barr virus ECCO2R extracorporeal carbon dioxide removal ECG electrocardiograph ECLA extracorporeal lung assist ECLS extracorporeal lung support ECMO extracorporeal membrane oxygenation EDH extradural haematoma EEG electroencephalograph EF enteral feeding EN enteral nutrition EPIC evidence-based practice in infection control ERCP endoscopic retrograde cholangiopancreatography ERV expiratory reserve volume ESC European Society of Cardiology ETCO2 end-tidal carbon dioxide EVLW extravascular lung water F/VT frequency/tidal volume ratio FA flow assist FAST focused assessment sonography in trauma FLAIR fluid-attenuated inversion recovery FRC functional residual capacity GABA γ-aminobutyric acid (inhibitory neurotransmitter) GCS Glasgow Coma Score GCSE generalized convulsive status epilepticus GEDV global end diastolic volume GFR glomerular filtration rate GIT gastrointestinal tract GTN glyceryl trinitrate GvsHD graft versus host disease HAART highly active antiretrovival therapy HBD heart-beating donation HBDs heart-beating donors HBS hypnotic-based sedation HCAP healthcare-associated pneumonia HCV hepatitis C virus HDU High Dependency Unit HE hepatic encephalopathy HELLP haemolysis, elevated liver enzymes and low platelets syndrome HepB hepatitis B HES hydroxyl-ethyl starch HFOV high-frequency oscillatory ventilation Hib Haemophilus influenzae type B HIT heparin-induced thrombocytopenia HIV human immunodeficiency virus HME heat and moisture exchange unit HPV hypoxic pulmonary vasoconstriction HSCT haematopoietic stem cell transplant HUS haemolytic uraemic syndrome HVHF high-volume haemofiltration IABP intra-aortic balloon pump IC inspiratory capacity ICD implantable cardiovertor– defibrillator ICF intracellular fluid ICH intracranial hypertension ICNARC Intensive Care National Audit and Research Centre ICP intracranial pressure ICS Intensive Care Society IHD intermittent haemodalysis Abbreviations xiv IJV internal jugular vein INR international normalized ratio IR infrared IRV inspiratory reserve volume IRV inverse ratio ventilation ISP increase pressure support ITBV intrathoracic blood volume IV intravenous IVF in vitro fertilization JVP jugular venous pressure LBBB left bundle branch block LDL low-density lipoprotein LED light emitting diode LMA left mentoanterior LMWH low-molecular-weight heparin LOC loss of consciousness LV liquid ventilation LVEDP left ventricular end diastolic pressure LVH left ventricular hypertrophy MAP mean airway pressure MAP mean arterial pressure MARS molecular absorbent recirculation system MCQ multiple choice questions MDMA methylenedioxymethamphetamine; ecstasy MENDS maximizing efficacy of targeted sedation and reducing neurological dysfunction MET medical emergency team MI myocardial infarction MIC minimum inhibitory concentration MIP maximum inspiratory pressure MMDS microcirculatory and mitochondrial distress syndrome MODS multiple organ dysfunction syndrome MOF multiple organ failure MOST multi-organ support therapy MPAP mean pulmonary artery pressure MPM mortality probability model MRI magnetic resonance imaging MRSA methicillin resistant Staphylococcus aureus MSBT safety of blood and tissues for transplantation MSOF multiple systems organ failure MV mechanical ventilation MV minute ventilation MVV maximal voluntary ventilation MW molecular weight NAC N-acetylcysteine NAD+ nicotinamide adenine dinucleotide NAPQI N-acetyl-p-benzoquinone-imine NAVA neurally adjusted ventilatory assistance NCSE non-convulsive status epilepticus NETI nasotracheal endotracheal intubation NHBD non-heart beating donor NHS National Health Service NICE National Institute for Clinical Excellence NICO non-invasive cardiac output NIV non-invasive ventilation NKH non-ketotic hyperglycemia NMDA N-methyl-d-aspartate NO nitric oxide NPPV non-invasive positive pressure ventilation NRTI nucleoside reverse transcriptase inhibitors NSAID non-steroidal anti-inflammatory drug NTG nitroglycerine ODTF organ donation taskforce OHSS ovarian hyperstimulation syndrome PA pulmonary artery PACS picture archiving and communication system PACT patient-centred acute care training PAE post antibiotic effect PAFC pulmonary artery flotation catheter PAV proportional assist ventilation PAWP pulmonary artery wedge pressure PC pressure control PCA patient-controlled analgesia PCI percutaneous coronary intervention PCP Pneumocystis jiroveci/carinii pneumonia PCR polymerase chain reaction PCV pressure control ventilation PCWP pulmonary capillary wedge pressure PDEIs phosphodiesterase enzyme inhibitors PE phenytoin, equivalents PE pulmonary embolus PECLA pumpless extracorporeal lung assist PEEP positive end expiratory pressure PEG percutaneous endoscopic gastrostomy PF parenteral feeding PF4 platelet factor 4 Abbreviations xv kidneys, liver and gut. Oxygen delivery depends on adequate oxygen uptake from the lungs, an adequate cardiac output to deliver the oxygen to the tissues and an adequate haemoglobin concentration to carry the oxygen. * These goals of resuscitation are usually achieved by the use of supplemental oxygen, fluid or red blood cell transfusion, inotropic support or antibiotics as needed. In certain circumstances, such as penetrating trauma, a surgical approach to limiting life-threatening bleeding is considered to be a part of the resuscitation process. * Resuscitation should begin as soon as the need for it has been identified. There is now evidence showing that early intervention (within a few hours of admission) limits the degree of organ dysfunction and improves survival. Waiting until the patient reaches the intensive care unit may be too long a delay if further deterioration in the patient’s condition is to be prevented. * In some situations, such as head injury, even single episodes of hypotension or hypoxia are associated with worsened outcomes. * Early and complete resuscitation is associated with improved outcomes. Monitoring the progress of resuscitation At present, there are only limited ways in which the function of individual tissue beds can be assessed. Assessing the adequacy of resuscitation is usually based on either global markers of oxygen supply and utilization (such as the normalizaton of mixed venous oxygen saturations and lactate concentration), or the clinical responses of the affected organs – urine output from the kidneys for example. Whilst resuscitation is ongoing, invasive monitors such as an arterial can- nula, a central venous cannula and a urinary catheter may be placed, but these additional monitors should not detract from the clinical monitoring of the patient. * Resuscitation must be tailored to the individual patient. There are now data to suggest appropriate goals or parameters for resuscitation in certain clinical states, notably sepsis (Table 1.2), acute head injury and penetrating trauma. * Over-enthusiastic attempts at resuscitation can lead to problems with fluid overload, worsening haemorrhage through dilution of clotting factors, or rapid electrolyte shifts leading to cerebral oedema. * The importance of early assessment by adequately trained staff, with regular review of clinical progress, cannot be over-emphasized. Once resuscitation is under way and the patient is stabilized, it is appropriate to begin an in-depth assess- ment of the patient. This means taking a more com- plete history, making a thorough examination and ordering clinical investigations as indicated. This phase of the process aims to establish an underlying diagnosis and guide definitive treatment. If deteriora- tion occurs over this time, the cycle of assessment and resuscitation should begin again. Physiology monitoring systems Physiology monitoring systems are systems that allow the integration of easily obtained and measured phys- iological variables into a single score or code that triggers a particular action or care pathway (see also Chapter 5: Scoring systems and outcome). * The commonly measured physiological variables are heart rate, blood pressure, respiratory rate, temperature, urine output and consciousness level, and these can be assessed at the bedside. * Action may be triggered by a single abnormality or by an aggregate score. Aggregate scoring systems are generally preferred as they may also allow a graded response depending on the score. * Physiological Scoring Systems (PSS) developed from the recognition that critically ill patients, and Table 1.1 Signs suggestive of critical illness * Obstructed/threatened airway * Respiratory rate >25 breaths/min or <8 breaths/min * Oxygen saturations <90% on air * Heart rate >120 bpm or <40 bpm * Systolic blood pressure <90mmHg * Capillary refill >3 seconds * Urine output <0.5ml/kg per hour more than last 4 hours * Glasgow coma score <15 or status epilepticus or patient not fully alert Table 1.2 Suggested goals to be achieved within 6 hours of presentation for the resuscitation of septic shock refractory to fluid therapy (after Rivers et al.) * Mean arterial pressure >65 mmHg * Central venous pressure 8–12 mmHg * Urine output >0.5 ml/kg per hour * Central venous oxygen saturation >70% Section I: Specific features of critical care medicine 2 in particular patients who suffered cardiac arrests, often had long periods (hours) of deterioration before the ‘crisis’ or medical emergency occurred. * PSS scores are often termed ‘track and trigger’ scores; they aim to identify and monitor patients whose clinical state is worsening over time, and then trigger an appropriate clinical response. * The Department of Health has recognized the need for the early identification of critically ill patients and recommends the use of track and trigger systems in all acute hospitals in the UK. The current recommendation is to use PSS to assess every patient at least every 12 hours or more frequently if they are at risk of deterioration. * PSS may have variable sensitivity and specificity for predicting hospital mortality, cardiac arrest and admission to critical care. Triggering scores may need to be set locally to maximize the benefits from these scoring systems. Typically, these scoring systems are not very sensitive but have high negative predictive power for the outcomes mentioned above. Advantages and disadvantages of PSS are summarized in Table 1.3. Medical emergency team and outreach It has been recognized that intensive care units will never have the capacity for all the patients that may benefit from some degree of critical care provision. The concept of ‘critical care without walls’ is that patients’ critical care needs may be met irrespective of their geographical location within the hospital. Medical emergency teams (METs) and critical care outreach (CCO) aim to redress the mismatch between the patient’s needs when they are critically ill and the resources available on a normal ward, in terms of manpower, skills, and equipment. * At present there is no clear consensus in the literature about the exact composition and role of these teams, nor their nomenclature. * Currently there is emphasis in teaching critical care skills to all hospital doctors via courses such as ALERT™ (Acute Life threatening Events – Recognition and Treatment) and CCRISP™ (Care of the Critically Ill Surgical Patient). * METs are usually understood to be physician-led. The team might typically consist of the duty medical registrar and intensive care registrar, a senior nurse and a variable number of other junior doctors. * METs are often formed from people who do not usually work together, coming together as a team only when the clinical need dictates. The MET has an obligation to arrive quickly, to contain the necessary skill mix in its members, to document the extent of its involvement accurately, and to liaise with the team responsible for the patient’s usual treatment. * METs are summoned to critically ill patients who have been identified either by a scoring system as outlined above, because they have attracted a particular diagnosis (e.g. status epilepticus), or because of general concerns that the nursing staff have about a patient. * METs have been shown to reduce the numbers of unexpected cardiac arrests in hospital in some observational studies, but the exact level of benefit is controversial. In some hospitals METs have replaced the traditional cardiac arrest team. Critical care outreach (CCO) teams are typically nurse-led, and have a variety of roles compared to the MET, depending on local policy (Fig. 1.1). The nurses in CCO are typically senior nurses who have been recruited from an intensive care, coronary care or acute medical background. CCO nurses are often employed full-time in this role and may perform addi- tional duties, such as following up patients on dis- charge from the intensive care unit, acute pain services, tracheostomy care and providing non- invasive ventilation advice. Table 1.3 Advantages and disadvantages of Physiological Scoring Systems Advantages Disadvantages * Rapid assessment * Facilitates communication between healthcare workers * Empowers staff * Reduces time from deterioration to action * Poor sensitivity (may not identify all critically ill patients) * Not validated in target populations * May not be appropriate for all patients (chronic health conditions, terminally ill, children, etc.) * Scores may not be calculated correctly Chapter 1: Recognition of critical illness 3 * When summoned to a critically ill patient, CCO will typically make an assessment and refer directly to intensive care services, or make suggestions to the parent medical team according to the requirements of the patients. * At present, not all CCO are staffed to provide a round-the-clock service and thus patients still often rely on junior medical staff to provide their care out of hours. In the UK, CCO is the most frequently used model, following on from Department of Health recommen- dations made in the late 1990s. Their explicit purpose is to avert ITU admissions, support discharge from ITU and to share critical care skills with the rest of the hospital. Other countries, most notably Australia, have pioneered the MET model since 1990. In some hospi- tals both systems run side by side. Currently the sys- tems are in a state of flux. The rapid introduction of MET/CCO systems in most hospitals has made the assessment of its impact on patient survival difficult. It is also difficult to assess how many patients at any one time need the input of a MET/CCO, and the implications that this may have for resource alloca- tion. At the time of writing, most of the available data suggest that the MET is under-utilized. Referral to critical care team Critical care can offer: * organ support technologies * high nurse : patient ratio * intensive/invasive monitoring * specialist expertise in managing the critically ill Patients who need these services should be referred to the critical care team. Intensive care units exist to support patients whose clinical needs outstrip the resources/manpower which can be safely provided on the general wards. The patient must also generally be in a position to benefit from the treatment, rather than simply to prolong the process of dying from an underlying condition. Chronological age alone is a poor indicator of survival from a critical illness; chronic health problems and functional limitations due to these are better predic- tors. There should be a discussion with the patient (if possible), or their family, to explain the proposed treatment and to seek their consent for escalating management. Most critical care facilities operate a ‘closed’ policy, in which the referring team temporarily devolves care to the intensive care team. The latter is led by a clinician trained in intensive care. There is evidence that this approach leads to reduced lengths of stay and increased survival rates in patients. As part of this strategy, all referrals to intensive care should be passed through the duty intensive care consultant. The referring team still has an important role to play as definitive manage- ment of a condition (e.g. surgery) is still often provided by them. Referral to the critical care team may occur via a variety of routes. The admission may be planned well in advance in the case of elective surgery, or antici- pated and discussed with the ITU consultant in the case of emergency surgery. Acute medical admissions should be referred to the ITU consultant directly from the medical consultant, but in emergencies referral may be made via the MET/CCO. The patient is usually reviewed on the ward prior to admission in order to facilitate resuscitation and safe, timely transfer to crit- ical care. Key points * Early recognition and treatment of the critically ill patient may improve outcome. * Recognition of a critically ill patient by junior or inexperienced staffmay be facilitated by a scoring system. * Physiological scoring systems are widely used, but not always well validated. * METs and CCOs aim to provide critical care skills rapidly to critically ill patients. * Referrals to the critical care services may happen from any level, but the final decision to admit a Fig. 1.1 Critical care outreach (CCO) teams often carry portable equipment to help stabilize patients in places outside of critical care areas. Section I: Specific features of critical care medicine 4 (3) Equipment needed should be checked prior to the procedure including two working Macintosh laryngoscopes, endotracheal tubes, laryngeal mask airways, working suction. Airway adjuncts such as oropharyngeal airways, longer laryngoscope blades, McCoy laryngoscope or bougie that might be required in an unexpectedly difficult intubation should also be available. (4) Trained assistance familiar with the technique should be available. (5) Preoxygenation with high flow 100% oxygen for 3–5 minutes to maximize oxygen reserves and prevent hypoxaemia until tracheal intubation is established. (6) A rapidly acting intravenous induction agent such as thiopentone and suxamethonium should be used to achieve rapid muscle relaxation and tracheal intubation. (7) Sellick’s manoeuvre or cricoid pressure should be applied just before patient loses consciousness. This involves digital pressure against the cricoid cartilage of the larynx, pushing it backwards (Fig. 2.1). This causes compression of the oesophagus between the cricoid cartilage and the C5/C6 vertebrae posteriorly, thus minimizing passive regurgitation of gastric contents (Fig. 2.2). Force applied should be 30N to 40N and should be maintained until the correct placement of endotracheal tube has been confirmed by auscultation and cuff inflated. It must be released during active vomiting, to reduce the risk of oesophageal rupture. Induction agents Most critical care units do not have fixed protocols for the drugs used to facilitate tracheal intubation, the choice of which can vary according to the personal experience, drugs availability and the patients’ pre-admission con- ditions or co-morbidities. The majority of anaesthetic induction agents are vasodilators and have cardiodepressant effects. The use of these induction agents can lead to a precipitous fall in blood pressure and cardiac output in dehydrated, septic or haemodynamically unstable patients. It is good practice to monitor patients’ cardio-respiratory param- eters closely including invasive blood pressure mon- itoring prior to induction and have fluid boluses and vasopressor drugs prepared and immediately available. It is beyond the realm of this chapter to go into the pharmacology in depth but a few of the commonly Cricoid Cartilage Fig. 2.1 Cricoid cartilage. (With permission of Update in Anaesthesia, Issue 2 (1992), Article 4: Cricoid pressure in Caesarean section.) Table 2.1 Critical care indications for intubation Indications Examples Provide long-term positive pressure ventilation Respiratory failure Protect the airway Glasgow Coma Scores <8 Secure the airway Airway obstruction, inadvertent or failed extubation Chapter 2: Advanced airway management 7 used intravenous induction agents and muscle relax- ants are outlined below: Propofol – The induction dose is 1.5–2.5 mg/kg but this should be reduced in haemodynamically unstable patients as it can cause profound hypoten- sion. It has the advantage of being able to be continued as an infusion for sedation and is a drug familiar to the majority of anaesthetists and intensivists. Etomidate – The dose of 0.3 mg/kg causes less haemodynamic instability than propofol, and has been used as drug of choice for critically ill patients. Its use has declined asmajor concerns have been raised over adrenocortical suppression even when given as a single dose at induction. Thiopentone – The classical induction agent used in RSI along with suxamethonium. It provides smooth rapid induction in a dose of 3–6mg/kg but also produces dose-related cardiac depression. Its metabolism is slow and sedation can persist for many hours afterwards. Muscle relaxants In a RSI situation, only suxamethonium should be used and it is given immediately after the IV induction agent without bag-mask ventilation. In a situation in which it is safe and appropriate to use a longer-acting muscle relaxant as the primary agent, then hand ven- tilation must be checked prior to its administration to avoid the ‘can’t intubate, can’t ventilate’ scenario. Suxamethonium – Suxamethonium is a depolariz- ing neuromuscular blocker and is the only available neuromuscular blocker with a rapid onset of effect and an ultra short duration of action of around 5 min. Given in a dose of 1–1.5 mg/kg it provides excellent intubating conditions within 60 sec but is contraindicated in con- ditions such as burns (>24 hours old), hyperkalaemia, malignant hyperpyrexia, myotonia and other neurolog- ical diseases. Suxamethonium is metabolized rapidly by plasma pseudocholinesterase and the duration of action is increased in patients who carry an atypical gene for this enzyme. In patients who are heterozygous for the atyp- ical gene duration of action is increased by 50–100%; in patients who are homozygous for the atypical gene dura- tion of action is increased to 4 hours. Rocuronium –A dose of around 0.6 mg/kg provides good intubating conditions within 2 min; however when given in a dose of 1–1.2 mg/kg it can facilitate intubation within 60 sec and can be used in a ‘modified rapid sequence induction’ when suxamethonium is contrain- dicated. It has duration of action of around 30 min. Atracurium – The initial dose is 0.5 mg/kg and provides intubating conditions within 2 min. It is useful in patients with renal or hepatic impairment as it is broken down by spontaneous Hoffmann degradation. It has a short duration of action of around 20–25min and if prolonged muscle relaxation is necessary it can be given by IV infusion at 0.5mg/kg per hour. See Table 2.2 for a summary of induction agents and muscle relaxants. Inhalational induction There will be instances when an inhalational induction may be the preferred method, e.g. upper airway Fig. 2.2 Application of cricoid pressure. (With permission of Update in Anaesthesia, Issue 2 (1992), Article 4: Cricoid pressure in Caesarean section.) Section I: Specific features of critical care medicine 8 obstruction, burns or status asthmaticus, and sevoflurane is one of the most commonly used agents within the UK for this technique. Inhalational induction is usually car- ried out by an anaesthetist and requires an anaesthetic machine and circuit. Endotracheal intubation Endotracheal intubation is usually via the oral route but historically nasal intubation was common practice. The oral endotracheal tube avoids the risk of sinusitis and allows a tube with a larger internal diameter to be used, reducing thework of breathing.Nasal endotracheal tubes, on the other hand, are better tolerated in awake patients and are used in some paediatric critical care units. During intubation, the patient should be fullymoni- tored. Equal breath sounds should be auscultated to rule out oesophageal intubation but this is not always reli- able. Capnography should be used to confirm tracheal intubation as recommended by the Royal College of Anaesthetists. Lightweight portable capnography devi- ces are available when intubation is required to be performed outside of the critical care, operating theatre or anaesthetic room environment. Chest X-rays should be performed to confirm the position of the tip of the endotracheal tube to avoid inadvertent endobronchial intubation. Difficult and failed intubation In an unanticipated difficult intubation situation, oxy- genation should be maintained with hand ventilation until appropriate help arrives. Clinicians experienced in advanced airway management should familiarize themselves with the failed intubation drill (Fig. 2.3). The Difficult Airway Society UK website (see Further reading) has the following management plans: (1) Unanticipated difficult tracheal intubation during rapid sequence induction of anaesthesia. (2) Rescue techniques for the ‘can’t intubate, can’t ventilate’ situation. Extubation/ weaning protocols In 2000, a study investigated characteristics of conven- tional mechanical ventilation in 412 medical and sur- gical ICUs involving 1638 patients across North America, South America, Spain and Portugal. The study confirmed that there was similarity between countries for the primary indications for mechanical Table 2.2 Commonly used induction agents and muscle relaxants Name Main features Doses Indications Cautions Propofol Rapid onset; unaffected by renal or hepatic disease 1.5–2.5 mg/kg Intravenous induction agent Given by infusion for sedation Anticonvulsant Vasodilatory and cardio- depressant Etomidate Rapid onset; less cardiodepressant. 0.3 mg/kg Induction agent Not for use in porphyria Adrenocortical suppression even after single dose Thiopentone Smooth rapid onset 3–6 mg/kg Induction agent; anticonvulsant Avoid in porphyria. Slow hepatic metabolism Suxamethonium Effective within 60 sec; lasts 3–5 min 1–1.5 mg/kg Depolarizing muscle relaxant Not for use in malignant hyperthermia and myotonia Caution in burns, renal failure, spinal cord injuries Atracurium Adequate muscle relaxation within 3 min; lasts 20–25 min. Safe for use in renal and hepatic failure 0.3–0.6 mg/kg Non-depolarizing muscle relaxant Histamine release may cause bronchospasm and hypotension Rocuronium Rapid onset if dose of 1–1.5 mg/kg used. Longer duration of action of 30–40 min 0.6 mg/kg gives adequate relaxation in 2 min Non-depolarizing muscle relaxant Intermediate risk of anaphylaxis Chapter 2: Advanced airway management 9 There are a number of bedside tests that may be used to assess the likelihood of extubating a patient; how- ever, they may be poor predictors of success in an individual patient. Parameters that can be used include: (1) PaO2/FiO2ratio >200 mmHg (26 kPa) (2) Respiratory rate <35 bpm (3) Tidal volume >5 ml/kg (4) Maximal negative inspiratory pressure 20–30 cmH2O (5) Respiratory rate/tidal volume < 100 min/l Tracheostomy Indications for tracheostomy Tracheostomies are commonly used in critical care units to: * Facilitate weaning after prolonged ventilation. * Secure and clear the airway in upper respiratory tract obstruction. * Facilitate removal of bronchial secretions, e.g. inadequate cough. * Protect the airway in the absence of laryngeal reflexes, e.g. bulbar palsy. * Provide an airway in patients with injuries or surgery to the head and neck. Contraindications for tracheostomy The only absolute contraindication to a surgical or percutaneous tracheostomy is local severe sepsis or uncontrollable coagulopathy. Relative contraindications to a percutaneous tracheostomy are abnormal anatomy, moderate coagulopathy, children under 12 years (due to difficulty identifying anatomical landmarks) and high FiO2 and PEEP requirements. Timing of tracheostomy The timing of when to perform a tracheostomy is still a clinical decision rather than one based on medical evidence. Some studies suggest that a tracheostomy performed earlier rather than later leads to improved weaning, a shorter ICU length of stay and a reduction in the incidence of nosocomial pneumonia. A system- atic review performed in 2005 comparing early trache- ostomy (up to 7 days after admission to ICU) with late tracheostomy (any time thereafter) suggested early tracheostomy led to a reduction in the duration of mechanical ventilation and length of ICU stay but the numbers involved in the studies were relatively small. A study investigating the effect that the timing of tracheostomy has on 30-day mortality in UK inten- sive care units (TrachMan trial) has been completed to compare early tracheostomy (days 1 to 4) with late tracheostomy (day 10 or after) in patients expected to require mechanical ventilation for 7 days or more. Early tracheostomy (n= 450) was not associated with any beneficial effect on the development of ventilator-associated pneumonia, ICU length of stay, use of sedation or 30-day mortality compared to late tracheostomy (n= 450). Interestingly, only 45% of the patients in the late group actually received a tracheostomy. Techniques of percutaneous tracheostomy The two methods of performing a tracheostomy are surgical or percutaneous. Surgical tracheostomies are best carried out by experienced surgeons and will not be discussed in any detail here. Percutaneous tracheostomy carried out using a Seldinger techni- que is rapidly becoming more established within the critical care setting. At present there are at least five different techni- ques for carrying out a percutaneous tracheostomy, but there is no evidence that one technique is superior to another (Fig. 2.4). (1) Classic Ciaglia technique using a guidewire over which a series of dilators are used (2) Forceps dilational technique (3) Retrograde translaryngeal technique (4) Single-dilator, dilational ‘Blue Rhino’ technique (5) ‘PercuTwist’ technique using rotational dilation with a self-cutting screw. Complications Complications of tracheostomy are numerous and the rates vary considerably; generally complication rates are higher for emergency tracheostomy and for airway obstruction. Complications can be divided into: Immediate – e.g. haemorrhage, pneumo- thorax, misplacement of tube. Delayed – e.g. tube blockage by secretions, infection of stoma site, infection of Section I: Specific features of critical care medicine 12 bronchial tree, mucosal ulceration due to cuff inflation. Late – e.g. persistent sinus at tracheos- tomy site, tracheal stenosis at cuff site, scar formation requiring revision. Tracheostomy tubes A variety of tracheostomy tubes are available which have different properties. A patient may require many different types of tracheostomy tubes during their critical care admission. Cuffed tracheostomy tubes – generally used for patients on positive pressure ventilation (Fig. 2.5). The cuff must be inflated to prevent air/oxygen leaking (a) (c) (d) (b) Fig. 2.4 Serial dilator percutaneous tracheostomy technique. Preparation: The patient should be positioned with neck extended and supported by pillow of a bag of fluids between the shoulders. The neck is cleansed with antiseptic solution and properly draped. The cricoid cartilage is identified, and the skin is anaesthetized with 1% lidocaine with 1: 100 000 adrenaline below the cricoid cartilage. The endotracheal tube is withdrawn so cuff is visible at the vocal cords to avoid accidental puncture. Some clinicians like to perform procedure under direct vision with fibreoptic bronchoscope. (a) A 1.5- to 2-cm transverse skin incision is made on the level of the first and second tracheal rings. A 22-gauge needle is inserted between the first and second or the second and third tracheal rings. (b) When air is aspirated into the syringe, the guidewire is introduced. (c) After the guidewire is protected, the dilators are introduced. (d) All dilators are inserted in a sequential manner from small to large diameter. The tracheostomy tube is then introduced along the dilator and guidewire. The guidewire and dilator are removed, the cuff of the tracheostomy tube is inflated, and the breathing circuit is connected. The ET tube can then be removed. (With permission from Update in Anaesthesia Issue 15 (2002), Article 16: Percutaneous tracheostomy.) Fig. 2.5 Cuffed tracheostomy tube. Chapter 2: Advanced airway management 13 backwards and reduce the chance of aspirate entering the lungs. Obstruction of a cuffed tracheostomy tube by secretions is potentially life-threatening. To mini- mize this risk, tracheostomy tubes should be changed frequently. Changing a tracheostomy tube carries the risk of misplacement and causes patient discomfort. Tracheostomy tubes with an inner cannula – allows an inner cannula to be changed frequently to reduce the risk of obstruction. Themain disadvantage of these tubes is the reduced inner diameter, which causes an increased work of breathing. Fenestrated tracheostomy tubes – allow airflow to pass through the oropharynx as well as the tracheal stoma and reduce the work of breathing provided the cuff is deflated (Fig. 2.6). These should not be used if the patient is at risk of aspiration or is on positive pressure ventilation unless the fenestrations are blocked by an inner cannula. Tracheostomy tubes with adjustable flange – useful in patients with abnormal anatomy such as obesity, neck swelling and spinal deformities (Fig. 2.7). The flanges allow the tracheostomy tube to be adjusted to a certain length. Decannulation of tracheostomy tubes Tracheostomy tubes should be removed as soon as possible once: * Resolution of the primary cause of the tracheostomy has occurred. * Patient is able to expectorate past the tube and not require regular suctioning. * Effective swallow, gag and cough reflexes are present. * Patient is comfortable with the cuff deflated. * Nutritional status is adequate. * No airway obstruction is present above the tracheostomy. Weaning often involves increasing length of time with the cuff deflated or ‘downsizing’ the tube to one with a smaller diameter allowing the patient to breathe past the tracheostomy. Speaking valves or occlusion/decannulation caps should only be used on non-cuffed tubes, cuffed tubes with the cuff deflated or fenestrated tubes with the cuff deflated. If the patient tolerates at least 4 hours with the occlusion cap and is able to effectively cough, then decannulation can take place. Cricothyroidotomy and mini tracheostomy A cricothyroidotomy is usually performed as an emer- gency procedure when a secure airway is needed and attempts at orotracheal or nasotracheal intubation have failed. The patients are likely to be profoundly hypoxic and it is important that the airway is secured quickly. This can be done either as a needle or as a surgical cricothyroidotomy. There are several ways in which a cricothyroidot- omy can be carried out. All the techniques involve an incision of the cricothyroid membrane, which is between the thyroid cartilage superiorly and cricoid cartilage inferiorly. * A small cannula (usually inserted over a needle), which needs a high-pressure gas source such as the ManuJet system (VBM, GmbH, Sulz, Germany) to Fig. 2.6 Fenestrated tracheostomy tube. Fig. 2.7 Tracheostomy tube with adjustable flange. Section I: Specific features of critical care medicine 14 it is likely that there is a window of opportunity beyond which this may lose its full benefit. There is one other problem with the scoring sys- tems. A 90% probability of survival still means that 10% of patients will die. Conversely, a 90% probability of mortality still means that 10% will survive. It is impos- sible to know which particular group the patient in question will fall under. However, the above statistics may aid decisions on issues such as treatment limita- tion and withdrawal. For example, the outcome in neutropenic sepsis requiring invasive ventilation and inotropes has a survival of around 10–15%. The majority of such patients could be in the younger age group with minimal co-morbidities. Hence, such patients are still treated in the hope that the survivor will have successful outcome with chemotherapy. Although the correlation between advancing age and mortality in hospital patients is well known, age alone should not be a criterion for admission to critical care. An 80-year-old active patient with no known co- morbidities is likely to have a better prognosis than a 55-year-old patient with history of myocardial infarc- tion, diabetes, renal failure and limited exercise tolerance when treated for respiratory failure due to pneumonia. It is also important that probability of survival to ward discharge is not used as a sole criterion for admission to critical care. Return to an acceptable quality of life should be aimed for. Defining this criterion has high potential for disagreement between patients, relatives and physicians. This can present difficult ethical and legal issues. It is important that all decisions and discus- sions with relatives are clearly documented. The decision making process should be based on evidence or accepted guidelines. The patient and relatives should be given an opportunity to clarify any doubts and also consult bodies such as Patient Advocate Liaison Service. Some interventions including certain drugs involve high direct costs. In order to ensure that cost does not become a sole criterion for provision of potentially beneficial treatment, it is essential that institutional guidelines and policies are in place. These could be in the form of a checklist where a drug is authorized if certain criteria are met. Decision to admit (Fig. 3.1) The critical care consultant in charge should have the responsibility and be informed for all admissions, discharges and transfers. The consultant may choose to decide based on input from other members of the critical care team including Critical Care Outreach. Outreach typically consists of experienced critical care nurses. The Department of Health Comp rehen sive Critical Car e docu ment (2000 ) iden- tified three main aims for outreach services: to avert admissions (or to ensure that admissions are timely) by identifying patients who are deteriorating; to enable discharges; and to share critical care skills. Management of critically ill patients when unit is full The admission criteria to the critical care unit should be based on need of the patient rather than bed avail- ability. This means that if a patient is deemed not a Table 3.1 Classification of levels of patient care Level Description Nurse to patient ratio Level 0 Patients whose needs can be met through normal ward care 1 : 6 Level 1 Patients whose needs are greater than what can provided by normal ward care, but may be met on acute wards with support from critical care team 1 : 4 Level 2 Patients needing support for a single failing organ system 1 : 2 Level 3 Patients needing advanced respiratory support alone, or basic respiratory support together with support of at least two organ systems 1 : 1 Table 3.2 Indications for admission to critical care Invasive or non-invasive ventilation for acute respiratory failure Optimization of fluid balance requiring invasive procedures Post-operative monitoring (cardiac surgery, neurosurgery, major vascular surgery, long surgical or interventional procedures, massive blood loss, multiple co-morbidities with low systemic reserve) Haemodynamic instability requiring inotropic support Potential for deterioration (e.g. airway swelling,metabolic disorders, coagulopathies, hypoxaemia, hypercarbia, hypovolaemia, intracranial events, acute arrhythmias) Interventions that cannot be performed in a general ward – continuous veno-venous haemofiltration, extra-corporeal gas exchange Chapter 3: Patient admission and discharge 17 candidate for critical care, this should be applied even if a bed is available. Conversely, if a patient is consid- ered a candidate for critical care, all measures should be taken to accommodate the patient in the critical care environment as early as possible. Methods of looking after critically ill patients without CrCU beds A number of steps could be taken if a critically ill patient presents in the absence of an available bed in the unit. The choice is often determined by the severity of illness, haemodynamic stability, ease of oxygena- tion, necessity of advanced interventions, time of the day and availability of medical staff. (1) Transfer out to another unit – usually done in daytime hours, if the patient is haemodynamically stable and/or not too difficult to oxygenate. If definitive therapy is only available in another hospital (e.g. neurosurgery), then early transfer is essential. (2) Stabilization in the theatre recovery area – after critical care nurses, the theatre recovery staff are best placed to manage a critically ill patient. Limitations include management of ventilators and limited ability to provide interventions such as continuous haemofiltration in the recovery area. This option is usually chosen out of hours or if a bed is imminently going to be available in the unit. (3) Earlier than planned discharge of eligible patients – it is a common practice to discharge patients from critical care after the morning round. However, some patients may be suitable for discharge at other times. Discharge to the ward after 2200 h should be an exception rather than the rule. (4) Manipulation of nursing staff workload – exceptionally, the number of nurse shift leaders could be reduced so that one of them could accommodate an additional critically ill patient. The risk and benefit of this intervention should be discussed between the nurse in charge and Good systemic reserve Limited organ support in patients with low systemic reserve/advanced malignancy Return to acceptable quality of life (e.g. severe brain damage) Low survival probability, poor systemic reserve and premorbid status Advance Directives, Documented treatment limitations Stable patients with a high risk of acute deterioration Interventions needing CrCU environment (e.g. CVVH, pre- optimization, ventilation) High probability of hospital discharge Factors influencing decision to accept [or refuse] admission Fig. 3.1 Factors influencing decision to accept (or refuse) admission to critical care. Section I: Specific features of critical care medicine 18 consultant so that an optimum nursing skill mix is still available. Any such measures should ensure that adequate numbers of staff are available for all future shifts. (5) Accommodate in critical care with the aid of outreach and/or medical staff – this option is chosen if the staff availability is likely to improve within a few hours or if the patient presents late at night when risk of transfer is deemed to be higher. (6) Utilize services of other monitored beds in the hospital (e.g. medical assessment unit, surgical assessment unit, coronary care unit) – these areas have their own limitations in terms of level of monitoring and nursing skills. However, in certain circumstances, and depending upon the type of patient, their services could be used as an interim measure. (7) Utilize the resuscitation room in Emergency Department – some ED nurses are trained in the management of critically ill patients. This option is least desirable since it has the potential to block ED beds. Decision not to admit Patients who would potentially benefit from critical care but have issued advance directives which prohibit the use of such interventions should have their views respected. These may still mean that they are admitted but have treatment limitations in place to take into account their wishes. Similarly, patients with a ‘Do not resuscitate’ order in place may still be admitted to critical care with treatment limitations. Patients who are deemed to have irreversible or severe organ system damage which is likely to prevent reasonable recovery should have treatment limits in place. Some patients discharged from critical care may not be suitable for readmission to critical care in the event of deterioration (e.g. advanced lung disease with prolonged weaning). Adequate communication with the patient, relatives and other teammembers is essen- tial in such situations. Admission procedure Adequate communication is vital. The priority of indi- vidual patient will vary depending upon their systemic status. For example a patient with potassium of 6.5 mmol/l unresponsive to standard measures will need urgent renal replacement therapy. However, a patient who is stable after major surgery for post- operative monitoring could wait in the recovery area for a short period before a bed is ready. All patients who are admitted to the unit should be handed over to one of the critical care doctors. This should include a summary of the history, treatment received and any planned investigations, etc. The patient should then have a relevant detailed clinical examination. Appropriate documentation regarding the treatment plans should be made. Inadequate doc- umentation and secondary errors are an important reason for weak defence in the event of litigation. It is helpful to have a ‘critical care admission template’ so that any relevant details are not missed. As soon as possible, the patient and/or the relatives should be given an explanation of current condition and plan of treatment. This will also be the ideal opportunity to address any concerns. Discharge A timely discharge from the CrCU is just as important as timely admission. Risks of delayed discharge Risks solely attributable to prolonged stay in critical care include potential cross-infection and psycholog- ical disturbances. Since a critical care bed is several times costlier than a ward bed, it is not a good use of resource if patients are kept longer than necessary. Also, with the high demand on beds, it is possible that the admission of a more deserving patient might be delayed as a result of bed blockages. Delayed dis- charge to wards also leads to cancellation of major elective surgery. It is essential that appropriate systems are in place to minimize delayed discharges. Outcomes of ‘premature’ discharge On the other hand, data also show that about 5% of patients are probably ‘prematurely’ discharged. It has been estimated that about a third of these patients have a higher mortality and might not have died if they had stayed in the critical care for 48 hours. If each patient has their bed ‘held’ in the ward, there will be no delay when a decision for their discharge is made. However, most hospitals in the UK and elsewhere do not have this facility due to resource issues. If the incidence of ‘delayed discharges’ from CrCU is low and the number of cancelled elective surgery patients due to non-availability of CrCU beds is high, it will probably be more economical to ‘hold’ a ward bed for Chapter 3: Patient admission and discharge 19 Mode of transport The optimal mode of transport selected for a patient transfer will depend upon a number of factors. These include: * the indication for, and urgency of, transfer * time to organize/mobilize transport * weather and traffic conditions * space (particularly when additional equipment is required) * cost. Road transfer by ambulance – This is the commonest method used in the UK. The advantages of road trans- fer are relatively low costs, space, rapid mobilization (in general) and less weather dependency. The disad- vantages are long journey times for transfers over long distances and unexpected delays due to traffic congestion. Air transfer, either by helicopter or fixed-wing aircraft, can be considered for longer journeys (over 50 miles or 2 hours). The time to mobilize these resources, the reduced space available, the physiolog- ical effects of flying, the costs and the potential need (and time implications) for transferring between vehicles at the beginning and end of the transfer should be taken into account when considering this mode of transport. Some of the space considerations with helicopter transfer are illustrated in Fig. 4.1. Personnel Current guidelines recommend that a minimum of two people accompany the transfer of a critically ill patient in addition to the staff required to operate the transport vehicle. The transfer team may be provided by the referring hospital or from a specialized retrieval service. The choice of team is usually dictated by local policy and the availability of a specialized transfer team. Observational data have shown that transfer by a specialist transfer team is associated with fewer adverse events and potentially better patient outcomes. The personnel usually involved in a critical care transfer include a registered medical practitioner with training and experience in the transfer of critically ill patients along with a suitably experienced nurse/para- medic/technician. Rather than defining specific per- sonnel by job title (e.g. anaesthetist, intensivist, critical care nurse, etc.), it is more important that the mem- bers of the transfer team have the relevant competen- cies to undertake the transfer. These have been defined in the competency-based training in intensive care medicine in Europe (CoBaTrICE) initiative which describes the knowledge, attitude and skills required for transporting a mechanically ventilated critically ill patient outside the intensive care unit (http://www. cobatrice.org/). A clear plan for how staff will be repatriated after the transfer should be determined prior to departure and the transfer staff base institution should ensure that adequate insurance is in place to cover staff for personal injury. Table 4.1 Indications for patient transfer * Lack of a staffed intensive care bed at the referring hospital * Need for specialist investigation * Need for specialist treatment * Repatriation Fig. 4.1 Helicopter transfers may potentially be quicker than land-based transfers. Disadvantages of helicopter transfers includes costs, limited space, noise, time to mobilize, potential need for additional transport to/from helicopter, physiological effects related to altitude. (Pictures courtesy of Tony Bleetman, Warwickshire and Northampton Air Ambulance.) Section I: Specific features of critical care medicine 22 Preparation prior to transfer The decision to transfer a critically ill patient is usually shared between the critical care consultants at the referring and receiving hospitals in collaboration with their consultant colleagues in the relevant special- ities. Clear communication should be established between the referring and receiving hospitals and the ambulance service. The patient (if possible) and rela- tives should be informed of the need for transfer at the earliest opportunity. Patients should be stabilized prior to transfer. Time spent resuscitating and stabilizing the patient before transfer can significantly reduce complications during the actual transfer. Rare exceptions may include patients with time-critical illnesses such as ruptured aortic aneurysm or expanding intracranial haemorrhage. Even in these patients, attention to the ABC principles is paramount in order to optimize the chances of the patient arriving alive at the receiving hospital. A pre-transfer check list to assess patient suitability/stability for transfer is provided in Table 4.2. Prior to departure, the transfer team should ensure that the full clinical details of the patient, the results and copies of relevant investigations (e.g. scans, blood results) are transported with the patient. The transfer team should contact named staff at the receiving hospital to provide an update on the patient’s condi- tion and estimated time of arrival. Monitoring All patients must have as aminimum continuous pulse oximetry, electrocardiographic monitoring and regu- lar measurement of blood pressure, respiratory rate and temperature. All ventilated patients should have end-tidal carbon dioxide monitoring. In addition, the oxygen supply, inspired oxygen concentration, venti- lator settings and airway pressure should ideally be monitored. Haemodynamically unstable patients (or those at risk of becoming unstable) may benefit from intra-arterial blood pressure monitoring. Intra-arterial blood pressure monitoring has the added advantage of being less sensitive to movement artefact than non- invasive techniques. Invasive cardiac output monitor- ing is usually impractical during transfer. In selected patients, intracranial pressure monitoring may be indicated. Equipment Designated lightweight, portable equipment with suf- ficient battery life to last the transfer should be used. The displays on portable monitors should be clear and capable of showing simultaneous ECG, SpO2, blood pressure and ETCO2 signals. All equipment should be Table 4.2 Patient check list prior to critical care transfer Airway * Patients with (or at risk from) airway compromise should be intubated prior to transfer * The tracheal tube should be secured and confirmed in correct position on a CXR * ETCO2 monitoring if intubated C-spine * Adequate spinal immobilization (if indicated) Circulation * Adequate intravenous access * Circulating volume optimized * Haemodynamically stable * All lines are patent and secured * Any active bleeding controlled * Long bone/pelvic fractures stabilized * Adequate haemoglobin concentration * Patient catheterized * End-organ perfusion optimized * ECG and blood pressure monitored Breathing * Patient adequately sedated and paralysed if ventilated * Ventilation established (and stable) on transport ventilator * Adequate gas exchange on transport ventilator confirmed by arterial blood gas analysis * Adequate oxygen supply on transfer vehicle * Stomach decompressed by with naso- or orogastric tube * Unclamped intercostal drain in situ if pneumothorax present (though Heimlich valve easier to manage than underwater seals) * SpO2 monitoring established Disability * No active seizures * Initial treatments for raised intracranial pressure (if indicated) * Life-threatening electrolyte disturbances corrected * Blood glucose >4 mmol/l Exposure * Patient adequately covered to prevent heat loss * Temperature monitored Chapter 4: Transfer of the critically ill 23 checked prior to use on a transfer. Alarm limits and volume should be checked and adjusted as necessary. Drugs and fluids should be administered by syringe/ infusion pumps as gravity-fed infusions are unreliable in moving vehicles. The equipment should be left on charge until immediately prior to departure. The ambulance should carry sufficient oxygen to last the duration of the transfer, plus at least another 50% to cover unexpected delays. Enough portable oxygen to complete the journey to and from the ambulance is also required. Equipment for advanced airway management, defibrillation and basic resuscitation drugs should be readily available. Drugs according to patient needs (e.g. sedatives, muscle relaxants, vasopressors and inotropes) and ample intravenous fluids should be carried. Detailed lists containing the minimum equip- ment and drug requirements for transfers are con- tained in the Intensive Care Society guidelines (see Further reading). Standard ambulance trolleys are usually inadequate for patient transfer as they lack sufficient space to safely store infusions and other equipment. A bespoke trans- fer trolley which is compliant with the relevant regula- tions (British Standard 1789/2000) is recommended (Fig. 4.2). All equipment must be securely attached to the ambulance or the trolley as in the event of a crash, heavy equipment such as monitors or oxygen cylinders can cause serious injury to staff and the patient. Balancing equipment on the patient’s body/between their legs is an unacceptable risk. Patient transfer record sheet Demographic data and clinical details about the patient’s presentation and treatment should be recorded on a dedicated patient transfer record sheet. This sheet should also collect information on haemo- dynamics, ventilatory settings, drugs and fluids administered during the transfer. The form can also be used to record any adverse event during the trans- fer. A copy of the transfer record should be left with the receiving hospital and a copy returned to the referring hospital. Figure 4.3 provides an example of a critical care transfer record sheet. Considerations during patient transfer Continuous monitoring of the ECG, SpO2 blood pres- sure and ETCO2 should be maintained throughout the transfer and recorded on the patient transfer sheet. The battery life on portable monitors and infusion pumps should be monitored and batteries changed/pumps replaced if they are running low on charge. A high level of vigilance should be maintained for the common com- plications reported to occur during inter-hospital trans- fer such as line or tube displacement/disconnection, vibration artefact and equipment failure. Patients are at risk of becoming hypothermic during transfer and this is particularly true of patients who have sustained burns. Temperature should be monitored regularly during the transfer. Devices that reduce heat loss (such as insulated blankets) and portable warm air devices can be useful in maintaining the patient’s temperature during the transfer. Fig. 4.2 A mobile intensive care trolley with equipment monitor, ventilator and infusion pumps which are fixed below the level of the patient. This provides a low centre of gravity and greater stability. The trolley fixes to the floor of the vehicle with a secure locking system and is not dependent on side-clamps or belt straps. (Picture courtesy of Reinout J. Mildner and Phil Wilson, Birmingham Children’s Hospital NHS Foundation Trust.) Section I: Specific features of critical care medicine 24 Chapter 5 Scoring systems and outcomeRoger Stedman Prediction is very difficult, especially about the future. Niels Bohr, Danish physicist (1885–1962) Introduction Scoring systems are used widely inmedicine both inside and outside the intensive care unit. They are used as an objective way of measuring and recording the severity of complex clinical conditions in order to compare patients, diseases, treatments or services. A scoring system can be specific or generic and may also be func- tional or anatomical. They can be designed simply as a measure of severity in order to create a ‘common lan- guage’ when discussing a single condition or may be a sophisticated statistical model in order to estimate probabilities of outcome or ‘risk adjust’ outcome data. Types of scoring system Specific This type of scoring system refers to a specific con- dition or organ. Examples include Glasgow Coma Scale (brain injury), Ransom Score (pancreatitis) and Child–Pugh Score (liver failure). Generic A generic scoring system is one that can be applied to a wide range of disease conditions and uses non-specific measures of severity. For example the SOFA score (Sequential Organ Failure Assessment) grades severity of dysfunction in six organ systems according to com- monly measured variables and can be used to track changes in the course of a critical illness. Other exam- ples include APACHE I, II, III and IV (Acute Physiology and Chronic Health Evaluation), SAPS 3 (Simplified Acute Physiology Score) and MPM III (Mortality Probability Model). Anatomical These are typically used in trauma and include AIS (Abbreviated Injury Scoring), the more detailed ISS (Injury Severity Score) and TRISS (Trauma and Injury Severity Score). The latter is a combination of a phys- iological score (RTS – Revised Trauma Score) and an anatomical (ISS) and is well calibrated for evaluating outcomes of trauma care. Functional Also termed physiological, this group includes most of the generic scoring systems outlined above. It also includes the oldest of the widely available methods of classifying critical illness, the Therapeutic Intervention Scoring System (TISS). TISS is a detailed measure of the level of support provided for the patient. Points are assigned for various monitoring and therapeutic inter- ventions. One of the advantages is that it does not rely on the diagnosis which is often not available at the start or even, on occasions, at the end of an intensive care period. However its long establishment has meant that it has struggled to keep up with the introduction of new mon- itoring and therapeutic interventions – each of which has to be weighted according to its influence on outcome. The applications of scoring systems Scoring systems are used for the following purposes: * outcome probabilities * clinical decision making and prognosis * quality and performance assessment * resource allocation * pre-ICU ‘at risk’ and ‘deteriorating patient’ screening * research. Core Topics in Critical Care Medicine, eds. Fang Gao Smith and Joyce Yeung. Published by Cambridge University Press. © Fang Gao Smith and Joyce Yeung 2010. Outcome probabilities Outcome measures The measurement of outcome is important as consid- erable resources are expended in providing intensive care. It is also an intervention associated with consid- erable burden of treatment (pain, suffering, loss of autonomy, disability, prolonged rehabilitation and dependence) for the individual and their relatives. It is important to be able to show it is ‘worth it’. Outcome in intensive care can be measured with respect to mortality, morbidity, disability and quality of life. Death on the intensive care unit is sufficiently common and easy to measure for it to be an important outcome. Other outcomes are also important; how- ever, they are difficult and complex to measure, and require long-term follow-up making it difficult to meaningfully inform changes in intensive care prac- tice. This has resulted in considerably less work being done on outcomes other than death, although this is changing slowly. There are many influences on outcome which can be broadly grouped into patient factors (age, co- morbidity), disease factors (diagnosis, severity) and intensive care factors (resources, staff, equipment, skill and timing). Generally scoring systems are used to ‘adjust’ measured outcomes for patient and disease factors (which can’t be influenced) in order to dem- onstrate the effect of intensive care factors (which can be influenced). Development of outcome probability model The process by which a scoring system becomes an outcome probability model is through multiple logis- tic regression. A typical example would be the use of APACHE II scoring which gives a measure of acute physiological derangement, chronic health status and age (see Table 5.1). Data are collected on a large num- ber of patients (development population) and each data point is weighted according to the strength of association with the outcome (death). When the original APACHE II model was devel- oped on a database of nearly 6000 patients it was found that each 3 point increase on the APACHE II score (0–71) was associated with a 3% increase in risk of hospital death. However, it was found that APACHE II alone was not generalizable to populations other than the development population (i.e. did not fit well when used in other settings) because the diagnosis of the acute illness had a powerful, independent influence on outcome. This led to the process of ‘case mix adjust- ment’ by which a table of possible acute diagnoses was developed and each diagnosis through the same logis- tic regression technique was given a weighting. Limitations of outcome probability model The combination of APACHE II and case mix adjust- ment provides a much more powerful tool for calculat- ing outcome probabilities in intensive care. However there remain a number of limitations which arise from differences in the way the score was developed compared to the way it is applied: (1) The population it is applied to – although the model was developed for a mixed general intensive care population its use has become widespread both in specialist intensive care units and also outside of the intensive care unit. The model is not well calibrated for these circumstances and may significantly over- or underestimate the risk of death. (2) The timing of data collection – the model is calibrated for the collection of physiological data over the first 24 hours of the intensive care stay. In critical illness the physiological deterioration and subsequent recovery is a continuum, the data is a snapshot at a point in that continuum and may not represent the nadir of physiological deterioration. This is manifest in the phenomenon of ‘lead time bias’ where efficient well resourced intensive care units may pick up critically ill patients early in the natural history of the condition, stabilize them and prevent severe physiological derangement and consequently derive low APACHE scores but the severity of the acute diagnosis goes on to produce a poor outcome (i.e. the physiological nadir occurs after the initial 24-hour period). The model then underestimates probability of death. (3) Validity, reliability and completeness of data – when developing a model, data collection will be assiduous and complete, it will be overseen by a research teamwho will apply consistent rules to the interpretation of the data (e.g. the scoring of GCS in sedated patients). In the application of themodel data collection will be incomplete, and rules for the interpretation of data will differ between units; this can have large effects on predicted outcomes, especially if a single variable with a large impact on outcome (such as GCS) is misinterpreted. Section I: Specific features of critical care medicine 28 Clinical decision making and prognosis Clinical applications The use of scoring systems to support clinical decision making is generally confined to specific scoring sys- tems, for example the use of GCS to decide when to protect the airway of a brain-injured patient (or when to perform brain imaging). Another example might be the use of sedation scoring (Ramsay or RASS) to adjust the infusion of sedative drugs. Potential drawbacks The use of generic scoring systems to support deci- sions to withdraw or withhold treatment is more con- troversial. Although a physiological scoring system that has been well calibrated through logistic regres- sion analysis and case mix adjustment can produce an accurate estimate of the risk of death a number of caveats have to be borne in mind when applying this to individual patients. (1) An estimate of the risk of death is a probability (i.e. a number anywhere between 0 and 1) whereas a prediction of outcome is dichotomous (i.e. the patient will/will not survive). To convert a probability to a prediction requires a ‘cut-off’ to be applied to the probability – such as a risk of death of 0.9. The problem then arises if you use this to decide to withdraw care then 1/10 of the patients you withdraw on would have survived if you had continued to support them. (2) These are statistical models that describe the population (and the results of the interventions applied to them) used to develop the model. This may be both remote in time and geography and Table 5.1 APACHE II scoring system Physiology points +4 +3 +2 +1 0 −1 −2 −3 −4 Rectal temperature (°C) ≥41.0 39.0 40.9 38.5 38.9 36.0 38.4 34.0 35.9 32.0 33.9 30.0 31.9 ≤29.9 Mean BP (mmHg) ≥160 130–159 110–129 70–109 50–69 ≤49 Heart rate (/min) ≥180 140–179 110–139 70–109 55–69 40–54 ≤39 Respiration rate (/min) ≥50 35–49 25–34 12–24 10–11 6–9 ≤5 Oxygenation FiO2 ≥ 0.5 A-aDO2 66.5 46.6–66.4 26.6–46.4 <26.6 FiO2<0.5 PaO2 >9.3 8.1–9.3 7.3–8.0 <7.3 Arterial pH ≥7.70 7.6–7.59 7.5–7.59 7.33–7.49 7.25–7.32 7.15–7.24 <7.15 Sodium (mmol/l) ≥180 160–179 155–159 150–154 130–149 120–129 111–119 ≤110 Potassium (mmol/l) ≥7.0 6.0–6.9 5.5–5.9 3.5–5.4 3.0–3.4 2.5–2.9 <2.5 Creatinine (µmol/l) ≥300 171–299 121–170 50–120 <50 Haematocrit (%) ≥60 50–59.9 46–49.9 30–45.9 20–29.9 <20 White cell count (109/l) ≥40 20–39.9 15–19.9 3–14.9 1–2.9 <1 Glasgow Coma Score – subtract GCS from 15. Age – <45 = 0, 45–54 = 2, 55–64 = 3, 65–75 = 5, ≥75 = 6. Chronic health points – chronic liver disease (portal hypertension, previous ALF, encephalopathy or coma), heart failure (NYHA grade 4), respiratory disease (with exercise limitation, polycythaemia or pulmonary hypertension), dialysis, immune suppression (chemo/radio therapy, high-dose steroids, leukaemia, AIDS). Additional 5 points for emergency surgery or emergency medical admission, 2 points for elective surgical admission. Source: With permission from Knaus WA, Draper EA, Wagner DP, Zimmerman JE (1985) Crit. Care Med. 13(10): 818–29. Chapter 5: Scoring systems and outcome 29 randomized controlled trials very difficult. Non- randomized and observational studies remain impor- tant and valid research methodologies in the critical care setting. However it remains important that researchers are able to describe their study populations in a way that enables clinicians to apply the findings to their own patient populations. The use of scoring systems and case mix adjustment provides a universal language for describing critical care populations and enables the fruits of research carried out in one health- care setting be applied in another. Useful data for designing a randomized clinical trial Scoring systems also provide a tool for researchers to stratify their study populations for risk of death and to calculate sample size of the trial when performing a randomized clinical trial. Scoring systems in common use in critical care There is insufficient space to cover in detail all scoring systems in common use in critical care. Here follows a list and brief description of the important ones with references for further reading at the end of the chapter. APACHE Now in its fourth incarnation, APACHE II remains the most commonly used. APACHE III refines the model both through the extension of the physiological data set to include laboratory data, and expands the number of co-morbid conditions. The development population was much larger than APACHE II (17 440 patients). The prognostic model is based on case mix adjustment and logistic regression as for APACHE II; however, in addition APACHE III man- dates daily updates of the acute physiology score in order to incorporate a measure of ‘response to therapy’ and eliminate ‘lead time bias’. APACHE IV is largely a recalibration of the APACHE III model based on a development population of 100 000 patients from US intensive care units admitted between 2003 and 2005. SAPS The Simplified Acute Physiology Score was developed as a reaction to the increasingly complicated nature of published scoring systems and the resources required to collect, process and maintain the data associated with them. The first usable version of the model devel- oped was SAPS II and consisted of 17 easily measured physiological variables – selected from 34 analysed using logistic regression. It was an attempt to create a ‘pure’ physiological outcomemodel, the premise being that critically ill patients commonly have more than one diagnosis and determining which is the more important often impossible. The result is that SAPS models require calibrating in whichever circumstance they are used (between countries and even individual units). This limits its use as a mechanism for compar- ing performance between sites although its simplicity increases its utility for tracking performance over time on an individual site. Subsets of the SAPS II model have been developed for specific diagnostic groupings (for example sepsis). The SAPS III model includes limited case mix and co-morbidity adjustment. MPM The Mortality Probability Model is, conversely, a risk model based almost purely on diagnosis. The score is based on a list of acute and chronic diagnoses which Table 5.2 Modified Early Warning Score Score 3 2 1 0 1 2 3 Respiratory rate (/min) ≤8 9 – 14 15 – 20 21–29 >29 Heart rate (/min) ≤40 41–50 51–100 101–110 11–129 >129 Systolic BP (mmHg) ≤70 71–80 81–100 101–199 ≥200 Urine output (ml/kg/hr) 0 ≤0.5 Temperature (°C) ≤35 35.1–36 36.1–38 38.1–38.5 ≥38.6 Neurological alert responds to voice responds to pain unresponsive Source:With permission from Morgan RJM, Williams F, Wright MM (1997) An early warning score for the detection of patients with impending illness. Clin. Intens. Care 8: 100. Section I: Specific features of critical care medicine 32 are either present or absent. Each contributes a weighted coefficient (derived by Bayesian inference) and multi- plied by an age-based variable to give an overall risk score. Again the coefficients are generated from a development population which needs to be represen- tative of the sample population for accurate probabi- lity estimation. In a similar fashion to SAPS the MPM is subject to calibration ‘drift’ (i.e. over time the orig- inal development population becomes less and less like the sample populations on which the model is used). MPM is currently in its third incarnation. SOFA Sequential Organ Failure Assessment is calculated by assigning a score of 0–4 for each of six organ systems according to objective and easily obtained measures of failure. It is performed daily, an increasing score, a high maximum and a large delta SOFA (change from admission to maximum SOFA) are associated with a high risk of mortality. SOFA is the only score (other than its similar sibling MODS – Multiple Organ Dysfunction Score) designed to provide a dynamic picture of disease severity. Trauma scores: TRISS and ASCOT These scores are not specific to critical care as they can be used on trauma patients in all settings. They are calculated at the time of admission to hospital. ASCOT is the most complete incorporating anatomical, physio- logical, age and diagnostic categories. It is well calibrated to the trauma population and is generalizable to health- care systems outside of which it was developed. Trauma scoring systems perform better at calculating outcome probabilities than generic scoring systems (such as APACHE) when applied to trauma patients on ITU. Key points * Scoring systems serve many functions to support clinical decision making at the level of the individual patient right through to global healthcare policy and guideline development. * Using scoring systems to assess critical care performance through the calculation of mortality ratios is a vital part of quality improvement. * Using statistical models to estimate the probability of death requires recalibration due to changes in patient populations and changes in critical care practice (improvements in technology, new therapies and improved standards of care). * Scoring systems provide a common language to describe patients with complexmedical conditions. * Scoring systems even with well-calibrated outcome models cannot predict the future. Further reading * Harrison DA, Parry GJ, Carpenter JR, Short A, Rowan K (2007) A new risk prediction model: the Intensive Care National Audit and Research Centre (ICNARC) model. Crit. Care Med. 25(4):1091–8. * Knaus WA, Draper EA, Wagner DP et al. (1991) The APACHE III prognostic system: risk prediction of hospital mortality for critically ill hospitalized adults. Chest 100(6): 1619–36. * Le Gall JR, Lemeshow S, Saulnier F (1993) A new simplified acute physiology score (SAPS II) based on a European/ North American multicentre study. J. Am. Med. Ass. 270(20): 2478–86. * Marshall JC, Cook DA, Christou NV et al. (1995) Multiple organ dysfunction score: a reliable descriptor of complex clinical outcome. Crit. Care Med. 23(10): 1638–52. * Rowan KM, Kerr JH, Major E et al. (1993) Intensive Care Society’s APACHE II study in Britain and Ireland. I: Variations in case mix of adult admissions to general intensive care units and impact on outcome. II: Outcome comparisons of intensive care units after adjustment for case mix by the American APACHE II method. Br. Med. J. 307: 972–81. * Vincent J-L, de Mendonça A, Cantraine F et al. (1998) Use of the SOFA score to assess the incidence of organ dysfunction / failure in intensive care units: results of a multicentric, prospective study. Crit. Care Med. 26(11): 1793–800. Chapter 5: Scoring systems and outcome 33 Chapter 6 Information management in critical careRoger Stedman Data is not information, information is not knowl- edge, knowledge is not wisdom. Clifford Stoll (US astronomer, computer expert and author) and Gary Schubert (Associate Professor of Art & Computer Science) Introduction The ready availability of large amounts of information about critically ill patients is one of the defining char- acteristics of a critical care unit. Good information management is essential to the process of care. Having the right information in the right place at the right time is a prerequisite for the delivery of safe and effective care. Critical care has been at the forefront of the use of technology to manage the flow of information in the clinical setting. This is undoubtedly because this tech- nology forms the core of many of the devices used to both monitor and treat critically ill patients. The integration of bedside, ward-based and hospital-wide information systems is a considerable challenge both from the technological and the clinical points of view. It is also expensive. For this reason the majority of critical care units still remain partially if not entirely dependent on paper-based systems. Functions of an information system An information system in critical care has to fulfil the following functions: (1) Bedside charting (2) Clinical record keeping (3) Electronic prescribing (physician order entry) (4) Integration with other hospital systems (5) Decision support (6) Remote access, multi-site communication (7) Data storage/archiving (8) Data access for audit, research, quality and financial management (9) Training, education and simulated environments. Bedside charting Paper chart The ‘ICU chart’ for a long time has been the heart of information management in the critical care unit. It is often a large sheet of paper (A2 or A1) on to which is manually transcribed data from physiological moni- toring, infused drugs and fluids, data from laboratory and bedside testing. It also forms part of the clinical record with a diary of events, a daily assessment of nursing needs and recording of invasive devices. It is a mechanism of communication and continuity with important clinical decisions recorded in order that all members of the multi-disciplinary team, visiting at different times, are informed. Each sheet contains 24 hours’ worth of information, although much informa- tion is transcribed from preceding sheets. The creation and maintenance of the ICU chart is a time- consuming and labour-intensive task. Once created the charts often present a storage problem as they are too large to be archived with the medical record. Any information on them that is needed for audit or quality assurance has to be subsequently extracted and man- ually entered on to a database. Basic requirements for an electronic chart The replacement of these functions with electronic information gathering is logical and technically feasi- ble. It is important to recognize that this is the ‘human interface’ of the information system. Although a lot of information gathering can be automated much still Core Topics in Critical Care Management, eds. Fang Gao Smith and Joyce Yeung. Published by Cambridge University Press. © Fang Gao Smith and Joyce Yeung 2010. themselves in institutions that have alternative systems in use already. The interoperability of different sys- tems is vital to the usability (for example not having to log on to different systems with a different password, not having to re-enter important clinical information like allergies or renal impairment) and ultimately the safety (if it is difficult to use people will take short cuts and workarounds). Integration with other hospital systems No information system exists in isolation. In recent years a plethora of hospital-wide information systems have come into existence: laboratory systems, radiol- ogy systems, electronic prescribing, patient adminis- tration systems and referral and booking systems. For them to integrate and talk to each other they need a common language. The accepted common standard for communication between healthcare information systems is the HL7 (Health Level 7). This is an open standard developed since 1987 by an international community of health informatics and healthcare pro- fessionals; it is a not-for-profit organization. The HL7 as an interoperability standard has been accredited by both the American National Standards Institute (ANSI) and the International Standards Organization (ISO). It provides a framework by which different applications and systems can exchange healthcare- related information. The use of the HL7 messaging framework is an essential feature of any information system that is not to be condemned to isolation and incompatibility. Decision support The term decision support refers to a range of func- tions of a clinical information system from providing assistance in the interaction between clinician and patient at the bedside, to providing evidence for serv- ice improvement and organizational change. At the bedside level An information system providing decision support might offer on-screen prompts or reminders, for example, if the patient has a particular allergy, or has renal impairment and requires drug dosage adjust- ment. They can be used to alert a clinician to the presence of an abnormal laboratory result, even to the extent of linking to the hospital pager system. There have been attempts at developing more sophis- ticated decision support systems that perform a diagnostic function. An example is the integration of physiological and laboratory data to trigger a ‘sepsis alert’. Information systems are an excellent way of implementing ‘care bundles’ and ‘checklists’; not only can they remind clinicians of the elements, they can also track performance through continuous audit. At the ward level Decision support means the provision of timely and relevant access to protocols, guidelines and evidence- based practice. The shelves of most intensive care units are creaking with the weight of paper documents pro- viding this type of information (and the dust gathering upon them). A paper-based guideline is generally only effective for as long as its authors and proponents maintain awareness of its existence amongst the prac- tising clinicians. A well-indexed electronic repository of guidelines and protocols could easily be linked with the electronic patient record at the bedside with the system able to bring the relevant ones to the attention of the treating clinician. At the organization level Decision support means access to information on per- formance – both process performance (compliance with protocols, care bundles, etc.) and outcome per- formance (risk-adjusted outcome). Remote access and multi-site communication The ubiquity and robustness of modern computer networks combined with the availability of compre- hensive electronic information systems means that the age of the ‘virtual ICU’ has arrived. It is possible to link, by electronic means, an intensive care bed to an intensive care clinician separated by many miles. Information systems can not only serve all the relevant clinical data to the clinician’s desktop, they can also provide voice and video link to the bedside. This enables a single clinician to provide decision making, advice and support to a very large number of critically ill patients across a wide geographical area (Fig. 6.2). The consolidation of healthcare organizations, for economic reasons, means that many hospitals cover multiple sites often incorporating a large central teach- ing or university hospital linked to a number of smaller community hospitals. Although efforts in recent decades have been made to centralize critical care resources on teaching hospital sites, there has Chapter 6: Information management in critical care 37 remained a stubborn need to maintain critical care beds at smaller community hospitals. The virtual ICU provides a means of bringing the entire critical care bed stock of a multi-site healthcare organization into one ‘virtual’ place whilst maintaining the provi- sion local at the point of need. Data storage and archiving A critical care information system generates an enor- mous amount of data. Attempting to keep all of it inevitably creates storage issues even in the age of the multilayer DVD and the terabyte hard drive. A good archiving system will perform a form of triage on the data that is generated based on the duration of useful- ness for that data. For example, a lot of the physiological data are only useful for the duration of the patient’s admission (possibly even less than that). Some data will be needed to be kept until quality and performance information is extracted from it. Research may neces- sitate the storage of other data. The temptation to store everything ‘just in case’ should be resisted as there is inevitably a trade-off between the amount of data stored and the ease and speed with which it can be accessed. As discussed earlier there is a need to provide high- quality archive storage for information that forms the clinical record. Again the longevity of the data need to be considered. A record that needs to be accessible for 25 years will need to be written to a disk with at least that lifespan – most conventional CD ROMs have a lifespan of less than 10 years. Data access for audit, research, quality and financial management In order for the huge amount of data generated by a critical care information system to continue to be useful it has to be converted to a database format. This process is called ‘data warehousing’. It is not a straightforward process and needs the input of special- ist informatics professionals. Data need to be stored in a way that enables clinicians and managers to ‘query’ the database, without having specialist computer qual- ifications themselves. The design of the database needs to be a result of a close collaboration between the clinicians (knowing and saying what they need) and the database designers (knowing what is possible and what the best way of implementing it is). Again there is a trade-off between making every conceivable query possible and the speed and reliability of the results of those queries. The technology of data warehousing and data mining is a huge area of endeavour within the infor- matics industry – not just in the healthcare arena. There are technological leaps, in the understanding and implementation of complex data structures and the software and hardware required to exploit them, occurring all of the time. These developments promise to transform the way we use and ‘see’ information in the morass of data available in the critical care environment. Training, education and simulated environments The concept of simulation has been current in critical care for some time. However the combination of the transformation of the critical care environment into an array of electronic information and the relative ease with which this can be recreated in a ‘classroom’ has meant that the fidelity (i.e. the realism) of simu- lated environments is compelling. Simulated environ- ments are in the process of revolutionizing the way doctors and nurses are trained with enormous poten- tial benefits for patient safety. Simulated environ- ments not only train how to interact with the technology but also how individuals perform in stressful situations and teach vital team skills. They build understanding of how medical errors occur and self-awareness of an individual’s role in preventing them. Implementation of a critical care information system The implementation of a critical care information system represents a major organizational change. Fig. 6.2 A critical care physician on duty at an e-icu workstation. Section I: Specific features of critical care medicine 38 (1) Successful implementation requires an examination of every aspect of the workflow of a critical care unit and how the system will impact (and improve) on it. It requires a close collaboration between clinical and IT professionals, which should persist for the life of the system as requirements evolve. All stakeholders in the system need to be involved from the start. There is danger in the enthusiasts and technophiles of a department striking out on an ambitious project that leaves their more conservative (or wisely sceptical) colleagues behind. (2) A vital aspect of the design of the system is the contingency in place if the system is down. Once dependence on an electronic information system is established then the continuous running of the system becomes ‘mission critical’. Back-up servers and un-interruptible power supplies form part of this, as well as a well ordered reversion to paper- based systems in the event of a sustained catastrophe. (3) The potential benefits need to be clearly identified and expressed as the costs are high. Costs are incurred: in the capital investment in hardware, software and structural works; for personnel for maintenance of the service; and by individuals in their personal commitment and time taken to change the way they work. However, the potential benefits of a well-designed, well-implemented and well-fitting information system in a critical care department are legion (and often unforeseeable). A step change in the quality and safety of the care delivered is possible along with a transformation of the quality of the working environment. Key points * The assimilation and integration of large amounts of clinical information is a defining characteristic of critical care medicine. * Adapting this process to the electronic information age is an important challenge for the current generation of critical care practitioners. * Critical care information systems can impact on every aspect of the workflow of a critical care department, with potential benefits in the quality and safety of the care delivered. * There are risks if important aspects of the design of the system are not considered, including the usability of the system and the interoperability with equipment and other hospital systems. * Costs are high: equipment, personnel and investment in organizational change. * The benefits are a potential step change in the quality and safety of care delivered to the patient. Further reading * Bion JF, Heffner JE (2004) Challenges in the care of the acutely ill. Lancet 363(9413): 970–7. * Breslow MJ, Rosenfeld BA, Doerfler M et al. (2004) Effect of a multiple-site intensive care unit telemedicine program on clinical and economic outcomes: an alternative paradigm for intensive staffing. Crit. Care Med. 32(1): 31–8. * Fraenkel DJ, CowieM, Daley P (2003) Quality benefits of an intensive care clinical information system. Crit. Care Med. 31(1): 120–5. * Han YY, Cacillo JA, Venkataraman ST et al. (2005) Unexpected increased mortality after implementation of a commercially sold computerized physician order entry system. Pediatrics 116(6): 1506–12. * Murphy JG, Torsher LC, Dunn WF (2007) Simulation medicine in intensive care and coronary care education. J. Crit. Care 22: 51–5. * Scurink CA, Lucas PJ, Hoepelman IM, Bonten MJ (2005) Computer assisted decision support for the diagnosis and treatment of infectious diseases in intensive care units. Lancet Infect. Dis. 5(5): 305–12. Chapter 6: Information management in critical care 39 * On hitting the artery, pulsatile flow of blood will be noticed; insert the guidewire through needle and remove the needle leaving the guidewire in the artery (Fig. 7.4). * Railroad the cannula over the guidewire, and once it is in remove the guidewire, leaving the cannula in the artery. * The cannula should be sutured in place to prevent it from being dislodged and connected to the blood pressure monitoring system. Components of the invasive blood pressure monitoring system Pressure tubing – a fluid-filled tubing system, connecting the cannula to the transducer. Ideally it should be short and stiff to minimize resonance. Disposable pressure transducer – it is used to convert patient’s pressure signal to electrical signal. There are four resistors connected as aWheatstone bridge. It should be placed at the level of the right atrium at midaxillary line. Flush device and continuous flush solution – saline is pressurized at 100 mmHg above the systolic pressure and flows at 3–4 ml/h to continuously flush the tubing and the cannula and prevent clot formation. Monitor – has amplifier, to amplify the size of the signal to display the arterial pressure trace. Calibration and damping Once the artery is cannulated and the cannula is secured in place, it should be connected to the monitoring system. Once connected, the transducer system needs to be calibrated to obtain an accurate blood pressure reading. Zero calibration causes the elimination of the effect of atmospheric pressure on the measured pres- sure and should be done with the transducer at the level of the right atrium. At times the measured trace can be affected by damping. Damping is produced by air bub- bles, clots, catheter kinking, stopcocks, and long, nar- row, compliant tubing. Overdamped trace results in under-reading of systolic pressure and over-reading of diastolic pressure and underdamping causes the oppo- site effect. In both the situations the mean arterial pressure remains accurate (Fig. 7.5). Complications of invasive blood pressure monitoring * Haemorrhage and haematoma * Thrombosis and ischaemia * Embolism * Infection * Aneurysm and AV fistula * Accidental injection of drugs. Central venous pressure monitoring Central venous pressure monitoring measures pres- sure in the great veins of the thorax usually the supe- rior vena cava and the right atrium. It involves introducing a catheter into a vein so that the tip of the catheter lies at the junction of the superior vena cava and the right atrium. Usually it is catheterized Fig. 7.3 Insertion of arterial cannula by Seldinger technique. Fig. 7.4 Once flashback is seen in hub of needle, guidewire is threaded, needle removed and cannula threaded over guidewire. Section I: Specific features of critical care medicine 42 using percutaneous technique but rarely surgical exposure ‘venous cut down’ may be required. Cannulation sites Internal jugular is the most commonly used vein. The other veins that can be cannulated are subcla- vian, antecubital and the femoral vein. In 2002, National Institute for Clinical Excellence (NICE) recommended the use of two-dimensional (2-D) imaging ultrasound guidance as the preferred method for insertion of central venous catheters into the internal jugular vein (IJV) in adults and children in elective situations. Advantages of central venous catheterization * Monitoring of central venous pressure * Acts as a guide for fluid therapy * Administration of drugs such as inotropes, which requires a big vein * Total parenteral nutrition (TPN) * Blood sampling for mixed venous saturation * Insertion of transcutaneous pacing leads. Changes in central venous pressure Table 7.2 summarizes these changes Technique for insertion of central venous catheter in internal jugular vein There are many cannulation techniques but it is catheter-over-guidewire (Seldinger’s) technique that is employedmost frequently. Usually the right internal jugular vein is cannulated as it is technically easier for the right-handed operator. The thoracic duct lies on the left side and there is potential for causing chylo- thorax if left internal jugular is cannulated. * Patient is usually placed supine in 15–30° Trendelenburg position to increase distension of the internal jugular vein. The head down position will also minimize risk of entraining air and causing air embolism. * ECG, blood pressure and oxygen saturation should be continuously monitored throughout the procedure. * Flush the central venous catheter with sterile saline solution. * Under strict aseptic conditions, the two heads of the sternomastoid muscle and clavicle are palpated. Feel for carotid artery pulsation; internal jugular vein lies just lateral to it. Two-dimensional (2-D) imaging ultrasound should be used whenever possible to locate the artery and the vein. * Once the vein is located, skin over it is infiltrated with 0.5 ml of 2% lignocaine. * Under direct ultrasound guidance the needle attached to a syringe should be advanced aiming 150 50 150 50 150 50 Normal Overdamped Underdamped Fig. 7.5 Schematic diagram of arterial trace and effects of damping. Table 7.2 Central venous pressure can be increased or decreased as a result of various conditions Increased by Raised intrathoracic pressure: intermittent positive pressure ventilation, coughing Impaired cardiac function: cardiac failure, cardiac tamponade Circulatory overload Superior vena caval obstruction Decreased by Reduced venous return: hypovolaemia Chapter 7: Haemodynamics monitoring 43 towards the ipsilateral nipple at an angle of 45° to the skin (Fig. 7.6). * When in the vein, dark and non-pulsatile blood is aspirated. * Disconnect the syringe and the insert the guidewire. It is important to monitor the ECG for cardiac arrhythmias. * Remove the needle, leaving the guidewire in the vein. * After dilating the insertion site a central venous catheter is railroaded over the guidewire, and ECG monitored for cardiac arrhythmias. * The catheter should be sutured in place to prevent it from being dislodged and connected to the central venous pressure monitoring system. The components of this system are very similar to those of the invasive arterial blood pressure monitoring system. * Correct location should be confirmed with a chest radiograph. Complications of central venous cannulation * Arterial cannulation * Embolism – air or thrombus * Pneumothorax, hydrothorax, haemothorax * Cardiac perforation * Cardiac tamponade * Injury to surrounding structures – nerves and artery * Infection. Cardiac output monitoring Clinical parameters may be used to determine cardiac output. Poor cerebral perfusion leads to agitation and confusion. A reduction in renal perfusion will lead to decreased urine output and subsequently to anuria. Skin perfusion is a clinically useful sign and can be determined using the capillary refill time. Progressive prolongation of the capillary refill time is seen with reduced cardiac output. As it is difficult to quantify cardiac output using the clinical parameters it is not a very objective method. It is widely accepted that cardiac output measure- ment is a useful adjunct in the resuscitation of critically ill patients. There are many invasive and non-invasive methods available to monitor cardiac output. Pulmonary artery (PA) catheter is inserted via a central venous catheter sheath. The PA catheter is then ‘floated’ so that the tip of pulmonary artery catheter is correctly wedged in the artery by the guide of changes in pressures from right atrium, right ven- tricle, pulmonary artery and pulmonary capillary wedge pressure (Fig. 7.7). As this technique is invasive and can cause potential risks to the patient, other less invasive methods are gaining acceptance (Table 7.3). However, it is still widely accepted as the gold standard. Thermodilution technique Intermittent thermodilution technique * 5–10 ml of cold saline is injected rapidly and smoothly through the proximal port of PA catheter Fig. 7.6 Insertion of central venous catheter using ultrasonography. Table 7.3 Commonly used techniques for cardiac output monitoring Use of PA catheter Use of CVP Use of arterial line Invasive PA catheter Yes Yes No Yes Lithium dilution No No Yes Yes PiCCO No Yes Yes Yes PulseCO No No Yes Yes Trans-oesophageal Doppler No No No Yes Trans-oesophageal echocardiography (TOE) No No No Yes NICO (Non-invasive cardiac output) No No No No Section I: Specific features of critical care medicine 44 (3) Modelflow by the mean of multiple conventional thermodilution measurements equally spread over the ventilatory cycle. Pulse-induced contour cardiac output The pulse-induced contour cardiac output (PiCCO) method uses a central venous line with a temper- ature sensor on the distal lumen and a thermodilu- tion sensor arterial catheter, placed in either the femoral or brachial artery. A thermodilution curve is produced when cold saline is injected through the arterial line. The system uses patient’s height and weight and computes left ventricular stroke volume by measuring the area under the systolic part of the waveform from the end of diastole to the end of the ejection phase and dividing the area by the aortic impedance. Formulae are then applied and give estimations of preload index such as intrathoracic blood volume and lung oedema index as extravascular lung water, and stroke volume variation as fluid responsiveness indicators (Table 7.4). Intrathoracic blood volume seems to be a good preload index but the results reported in literature are not homogeneous in all its applications. Intrathoracic blood volume Intrathoracic blood volume (ITBV) consists of the global end diastolic volume (GEDV), the volume of blood within the heart plus the pulmonary blood vol- ume. There are three volumes within the thorax: the intrathoracic blood volume, the intrathoracic gas vol- ume and the extravascular lung water. Due to limited expansion of the thorax, these volumes interact and change proportionally to each other. ITBV is a volu- metric measurement of cardiac preload (ventricular end diastolic pressure), and is a sensitive indicator of the circulatory blood volume in mechanically ventilated patients. Studies have shown that changes in intrathoracic blood volume correlate well with changes in cardiac output and can be used as a surro- gate for patient’s fluid status. Extravascular lung water The water content in the lungs increases in left heart failure, pneumonia, sepsis and burns. Extravascular lung water (EVLW) is used to determine pulmonary function and a guide to fluid resuscitation in shock. Studies have suggested that resolution of pulmonary oedema may be more rapid when EVLW is used as a therapeutic guide. Prognosis based on EVLW has indicated a higher risk of mortality with EVLW of greater than 14 ml/kg. Transoesophageal echocardiography Quantifying cardiac output with transoesophageal echocardiography (TOE) involves an accurate meas- urement of the velocity of blood flow across a specific valve with the aid of Doppler. The area under the flow velocity curve is calculated and it represents a specific distance along which the column of blood is projected during one cardiac cycle. This measurement of the length of blood in the ascending aorta in unit time andmultiplying by the cross-sectional area of the aorta can give stroke volume from which cardiac output can be derived. In addition it also allows for the assessment of ventricular function, wall motion abnormalities during myocardial ischaemia, cardiac anatomy and valve function. In experienced hands, TOE can give valuable hae- modynamic information and evaluate the ejection fraction and mechanical movement of the heart as a pump. TOE can also aid diagnosis by identifying the presence of vegetations in infective endocarditis or pericardial effusions in pericarditis. Table 7.4 Table illustrating parameters measured by the PiCCO system Parameter/ indices Normal values Notes Intrathoracic blood volume index 850–1000 ml/m2 <850 underfilled >1000 adequate/overfilled Extravascular lung water index 3.0–7.0 ml/kg >7.0 ml/kg indicates pulmonary oedema Cardiac function index 4.5–6.5 l/ min Measured independently of preload Reflects myocardial contractile function <4.5 indicates poor myocardial contractility Stroke volume variation 10–15% Measured as the mean difference between the highest and lowest stroke volume over the last 30 sec Only accurately measured in ventilated patients Inaccurate in arrhythmias <10% adequate/overfilled >15% dehydrated/underfilled Chapter 7: Haemodynamics monitoring 47 Other methods of cardiac output monitoring * Thoracic electrical bioimpedance * Induction cardiography * Electromagnetic flow measurement. Key points * Haemodynamic monitoring is most informative when it is used to supplement clinical judgement. * It may be more beneficial to look at the trend of the variable being measured rather than a single value. * One should always ensure full asepsis when inserting an invasive monitoring device into vulnerable patients. * Good anatomical knowledge and sound technique will help in reducing some of the complications from invasive monitoring devices. Further reading * Allsager CM, Swanevelder J (2003) Measuring cardiac output. Br. J. Anaesth. 3: 15–19. * Association of Anaesthetists of Great Britain and Ireland (2007) Recommendations for Standards of Monitoring during Anaesthesia and Recovery. London: The Association of Anaesthetists of Great Britain and Ireland. * Elliott TS, Faroqui MH, Armstrong RF (1994) Guidelines for good practice in central venous catheterization. J. Hosp. Infect. 28: 163–76. * Fletcher S (2005) Catheter-related bloodstream infection. Contin. Educ. Anaesth. Crit. Care Pain 5: 49–51. * Harvey S, Harrison DA, Singer M et al. (2005) Assessment of the clinical effectiveness of pulmonary artery catheters in management of patients in intensive care (PAC-Man): a randomised controlled trial. Lancet 366: 472–7. * Inweregbu K, Dave J, Pittard A (2005) Nosocomial infections. Contin. Educ. Anaesth. Crit. Care Pain 5: 14–17. * Morgan GE, Mikhail MS, Murray M (2002) Patient monitors. In Clinical Anesthesiology, 3rd edn. New York: McGraw Hill, pp. 86–125. * National Institute for Clinical Excellence (2002) Guidance on the Use of Ultrasound Locating Devices for Placing Central Venous Catheters, Technology appraisal No. 49. London: NICE. * Pratt RJ, Pellow CM, Wilson JA et al. (2007) Epic2: national evidence-based guidelines for preventing healthcare-associated infections in NHS hospitals in England. J. Hosp. Infect. 65(S): S1–S64. * Singer M, Clarke J, Bennett ED (1989) Continuous monitoring by esophageal Doppler. Crit. Care Med. 17: 447–52. Common measured haemodynamic variables Cardiac output (CO) HR × SV/1000 = 4–8 l/min Cardiac index (CI) CO/BSA = 2.5–4 l/min/m2 Stroke volume (SV) CO/HR × 1000 = 60–100 ml/ beat Stroke volume index (SVI) CI/HR × 1000 = 33–47 ml/ m2/beat Systemic vascular resistance (SVR) 80 × (MAP–RAP)/CO = 1000–1500 dyne s/cm3 Systemic vascular resistance index (SVRI) 80 × (MAP–RAP)/CI = 1970– 2390 dyne s/cm3/m2 Pulmonary vascular resistance (PVR) 80 × (MPAP–PAWP)/CO = <250 dyne s/cm3 Pulmonary vascular resistance index (PVRI) 80 × (MPAP–PAWP)/CI = 255–285 dyne s/cm3/m2 HR = heart rate BSA = body surface area RAP = right atrial pressure PAWP= pulmonary artery wedge pressure MPAP= mean pulmonary artery pressure Section I: Specific features of critical care medicine 48 Chapter 8 Critical care imaging modalitiesFrances Aitchison All critically ill patients will have imaging tests during their time in the intensive care unit. For some patients this will be a simple procedure, such as a plain chest X-ray, however for others more complex investigations such as magnetic resonance imaging will be required. General considerations * It is always important that the reason for the imaging request is clearly stated on the referral form. This will allow the imaging department to ensure that the most appropriate imaging technique is used, particularly in patients with multiple pathologies where several imaging modalities may be needed. It may well be helpful to discuss the clinical problem with a radiologist. Advice is available from referral guidelines published by the UK Royal College of Radiologists 2007 (see Further reading). * Some imaging tests are limited by considerations of patient and staff safety and thought must be given to the patient’s clinical condition when planning investigations. For more complex examinations, such as computed tomography (CT), the imaging equipment cannot be brought to the intensive care unit and the patient will require to be moved. Up to 70% of critically ill patients undergoing intra-hospital transfer will have an unexpected adverse event (see Further reading). These adverse events include both equipment malfunction and patient instability such as hypotension, arrhythmia, increased intracranial pressure, hypocapnia and hypercapnia and significant hypoxaemia. The establishment of, and adherence to, guidelines on personnel, equipment and monitoring during intra-hospital transfer is of major importance for each critical care unit. * Particularly in CT, it must be remembered that there are also significant risks to the critically ill patient associated with the radiation dose and the use of potentially nephrotoxic iodinated intravenous contrast. * Most hospitals in the UK now use a picture archiving and communication system (PACS) to store X-ray, CT, magnetic resonance imaging (MRI) and ultrasound examination images directly onto the hospital computer system. PACS systems are used in conjunction with digital or computed radiography (DR or CR) equipment for plain X-rays. These have replaced the use of conventional X-ray film cassettes with an X-ray system involving reusable photo-stimulable plates. There are many advantages – the examinations should always be immediately available, are stored in chronological order and have wider latitude of exposure since the image brightness and contrast can be manipulated on the screen during viewing. This has reduced the number of repeat films and is particularly helpful for critically ill patients. However, the PACS itself and the high-resolution monitors required for viewing are expensive. There follows a general description of each of the main imaging modalities and more detailed description of their use for particular clinical problems. X-ray imaging Basic principles * All X-ray imaging involves the use of an X-ray source to produce a beam of radiation which is passed through the patient and a detector to produce an image. The detector may be a cassette Core Topics in Critical Care Medicine, eds. Fang Gao Smith and Joyce Yeung. Published by Cambridge University Press. © Fang Gao Smith and Joyce Yeung 2010. * Tracheostomy tube – the T3 vertebral level defines the ideal position of the tube. * Central venous catheters – the ideal position of the catheter tip is within the superior vena cava. Incorrect positioning of central venous catheters occurs in up to 10% of patients (see Further reading). Cannulation of the right subclavian vein is associated with the highest risk of malposition. Haemofiltration lines are of larger calibre than standard central venous lines (Fig. 8.5). * Pulmonary artery (Swan Ganz) catheter – with an uninflated balloon, the tip of the catheter should be in the right or left pulmonary artery and within 5 cm of the midline on chest X-ray (Fig. 8.6). * Pacing wires – chest X-rays should be checked for pneumothorax and the tip of the wire should overlie the apex of the right ventricle (Fig. 8.7). * Chest drains – the drain should be placed postero- inferiorly for pleural fluid collections and antero- superiorly for pneumothorax. The side holes of the drain must be within the pleural space (Fig. 8.8). * Intra-aortic balloon – the balloon should lie in the descending aorta with its tip just above the left hilum on the chest X-ray. * Enteral tubes – the tip of nasogastric or orogastric tubes should always be below the diaphragm. Many tubes do not have a radio-opaque marker along their entire length and only the last few centimetres are visible on the X-ray. * Oesophageal Doppler probes – tip should be in mid thorax below the level of the carina. Appearance of lungs The appearance of the lungs should be carefully assessed on every chest X-ray. It is particularly helpful to make a comparison with any previous chest films. Following disease progression over time will allow more accurate distinction between infection and pul- monary oedema. The presence and size of any pleural Fig. 8.3 Bilateral large pleural effusions on erect chest X-ray. There is a diffuse increase in density in the lower chest and loss of definition of the diaphragm on each side. Note the patient has a long-term tracheostomy in situ. A C D B Fig. 8.5 Right subclavian and left internal jugular central venous lines (A and C) and haemofiltration lines (B and D) . All are satisfactorily positioned. There is also an endotracheal tube and nasogastric tube. A B C D E Fig. 8.4 Satisfactory positions for: A, endotracheal tube; B, oesophageal Doppler probe; C, right internal jugular venous line; D, ECG lead wire; E, tip of nasogastric tube. Section I: Specific features of critical care medicine 52 fluid collection, pneumomediastinum or pneumo- thorax should be documented. Abdominal X-ray The abdominal X-ray is useful for assessment of acute abdominal pathology. Critically ill patients are unable to have erect chest films taken; therefore if perforation of a hollow abdominal viscus is suspected evidence must be sought from abdominal X-rays. This will include the visualization of both sides of the bowel wall (Rigler’s sign). Plain abdominal X-rays are used for detection of small and large bowel obstruction (Figs. 8.9 and 8.10) and are useful to demonstrate abnor- mal intra-abdominal opacities such as renal calculi. CT scanning CT scans are extremely useful in management of crit- ically ill patients. * Intravenous iodinated contrast is often used during CT assessment of the mediastinum, abdomen and pelvis and in CT angiography. * Minor adverse reaction to iodinated contrast (e.g. flushing, urticaria, bronchospasm) is not uncommon. The incidence of severe reaction to modern non-ionic agents is 0.04% and very serious reaction (including death ) is 0.004%. Patients most at risk are those with a history of previous contrast reaction, asthma or previous allergic reaction requiring medical treatment. * Iodinated intravenous contrast is potentially nephrotoxic. Patients most at risk of contrast induced nephropathy are those with pre-existing renal impairment, particularly patients with GFR <30 ml/min. * Dilute oral iodinated contrast may be used during CT of abdomen or pelvis to opacify bowel and distinguish from abnormal abdominal fluid collection or abscess. In critical care patients oral contrast may be given via a nasogastric tube. Fig. 8.6 Satisfactory position for tip of Swan Ganz catheter (arrow shows tip of catheter). Fig. 8.8 Unsatisfactory position of left-sided chest drain with one set of side holes outside the lateral border of the ribs. Fig. 8.7 Satisfactorily positioned dual lead permanent pacemaker with tip of the more cranial lead in right atrium and tip of more caudal lead at apex of right ventricle. Note left-sided pneumothorax as complication of insertion. Chapter 8: Critical care imaging modalities 53 CT head The most commonly performed CT examination in this patient group is CT head scan which is particularly useful for the detection of: * Haemorrhage – after acute haemorrhage the area of abnormality will have the following appearance on CT relative to normal brain: 0–2 weeks, whiter (hyperdense); 2–4 weeks, same density (isodense); >4weeks, darker (hypodense). It is important to distinguish haemorrhage with the substance of the brain itself from haemorrhage outside the brain (extra-axial bleed). Extra-axial haemorrhage can be divided into: extradural, egg-shaped peripheral bleed; subdural, more diffuse crescentic peripheral bleed (Fig. 8.11); and subarachnoid, widespread blood within CSF involving sulci on brain surface, basal cisterns and ventricular system (Fig. 8.12). * Infarction: for 24–48 hours after infarction the affectedbrainmayhavenormaldensity. Subsequently CT will demonstrate a focal area of reduced density within vascular territory affected (Fig. 8.13). * Tumour, inflammatory disease or infection. CT cervical spine CT cervical spine is an increasingly commonly requested examination in trauma patients. Neck support collars and patient immobility often make plain films difficult to interpret in this group. CT cervical spine allows detailed assessment of bony structures but very limited assessment of the spinal cord itself. Detailed guidelines on appropriate use of CT head and CT cervical spine in trauma patients can be found at www.nice.org.uk. CT chest CT is very useful for investigation of chest disease allowing assessment of mediastinum, lung paren- chyma and chest wall. CT pulmonary angiography (CTPA) has good sensitivity and specificity for pulmo- nary embolic disease (Fig. 8.14). High-resolution CT (HRCT) is performed without intravenous contrast and uses very thin slices to assess lung parenchyma. It can be used, for example, to distinguish the acute and fibrotic phases of adult res- piratory distress syndrome. CT abdomen Critically ill patients with acute abdominal pathology are often assessed with CT for detection of bowel Fig. 8.9 Plain X-ray appearance of small bowel obstruction with multiple dilated loops in centre of abdomen demonstrating complete transverse mucosal folds. Fig. 8.10 Plain X-ray appearance of large bowel obstruction with markedly dilated loops of bowel at periphery of right and upper abdomen which have incomplete mucosal folds. Section I: Specific features of critical care medicine 54 Use in the critically ill * Ultrasound imaging is very well suited to the intensive care unit environment. There is no radiation dose to the patient or risk to staff and the equipment can brought to the patient’s bedside. Manufacturers are now able to produce easily portable, good quality ultrasound machines even smaller than a laptop computer. * The technology is particularly useful for the detection of fluid collections such as pleural effusion, infected collections/abscesses or free intraperitoneal fluid in cases of abdominal trauma. A skin site marking made at the time of ultrasound is useful for safe drainage of collections. * The liver, kidneys and gall bladder are particularly well seen on ultrasound. Ultrasound will not penetrate through air and therefore bowel and lung generally cannot be assessed with ultrasound. * Over recent years ultrasound has become increasingly used in the intensive care unit to assist in safe placement of central vascular lines. Key points * The reason for the imaging request should be stated on the referral form to allow the most appropriate imaging technique to be used. It may be helpful to discuss the clinical problem with a radiologist. * Consideration should be given to patient and staff safety and the patient’s clinical condition when planning investigations. * All patients should have a chest X-ray on arrival in the unit, after any invasive procedures such as endotracheal intubation, central vascular catheter placement or thoracocentesis and insertion of naso- or orogastric feeding tubes. * Intravenous iodinated contrast is often used during CT assessment of the mediastinum, abdomen and pelvis and in CT angiography. Risks include minor adverse reaction to iodinated contrast (e.g. flushing, urticaria, bronchospasm) and nephrotoxicity. Patients with GFR <30 ml/min are most at risk of contrast-induced nephropathy. * MRI can produce extremely detailed images of soft tissue and is more sensitive than CT for subtle pathology, for example the posterior fossa of brain or the spinal cord. * Ultrasound involves no radiation dose to the patient or risk to staff, and the equipment can be brought to the patient’s bedside. It is used to detect fluid collections such as pleural effusion, abscesses or free intraperitoneal fluid. A skin site marking can be made for safe drainage of collections. * Ultrasound has become increasingly used in the ICU to assist in the safe placement of central vascular lines. Further reading * European Society of Intensive Care Medicine (2004) Patient Centered Acute Care Training Programme: Skills and Techniques – Clinical Imaging Module. Brussels: ESICM. * Papson JP, Russell KL, Taylor DM et al. (2007) Unexpected events during the intra-hospital transport of critically ill patients. Acad. Emerg. Med. 14(6): 574–7. * Pikwer A, Baath L, Davidson B et al. (2008) The incidence and risk of central venous catheter malpositioning: a prospective cohort study in 1619 patients. Anaesth. Intens. Care 36(1): 30–7. * Royal College of Radiologists (2007)Making the Best Use of Clinical Radiology Services: Referral Guidelines, 6 edn. London: RCR. Chapter 8: Critical care imaging modalities 57 Chapter 9 Vasoactive drugsMamta Patel Shock is defined as the failure of the cardiovascular system to maintain adequate organ perfusion pres- sure. This causes inadequate oxygen delivery result- ing in tissue hypoxia, lactic acidosis and end organ damage. Causes of shock * Cardiac or ‘pump’ failure – tamponade, infarction, ischaemia, cardiomyopathy * Hypovolaemia/bleeding – haemorrhage, third space loss * Septic shock – pancreatitis, meningitis, burns * Obstructive – massive pulmonary embolism * Anaphylaxis. Oxygen delivery The cardiovascular system comprises the heart, blood vessels and the circulating blood volume. It interacts with the lungs to maintain tissue perfusion and oxygenation. The relationship between the respiratory and cardiovascular system (CVS) is illustrated by the oxy- gen delivery (DO2) equation. DO2 ¼ðSaO2 Hb 1:34þ ð0:03 PaO2ÞÞ  CO CVS Lungs CVS where: DO2 = oxygen delivery (ml/min) SaO2 = % saturation of Hb in the blood Hb = haemoglobin concentration (g/dl) PaO2 = partial pressure of oxygen in the blood (kPa) CO = cardiac output (l/min). CO is equated as: CO ¼ stroke volume  heart rate Thus substituting into the equation below: DO2 ¼ ðSaO2 Hb 1:34þ ð0:03 PaO2ÞÞ  stroke volume heart rate Abnormalities in any of the above variables will result in shock and tissue hypoxia if the compensatory mech- anisms of the body fail. Stroke volume is determined by preload, afterload and contractility. Stroke volume ¼ diastolic volume  end systolic volume Contractility is the ability of the heart to contract independently of preload and afterload: * It is increased by inotropes, e.g epinephrine, dobutamine and calcium * It is decreased by acidosis, hypoxia and hypocalcaemia. Afterload is the force that opposes ventricular contraction. * It is determined by the sympathetic tone, ventricular volume and pressure, the renin–angiotensin– aldosterone system and baroreceptor activity. Preload is related to the end diastolic volume and hence central venous pressure (CVP) (Fig. 9.1). * The Frank–Starling curve demonstrates the relationship between preload and stroke volume. * Increasing preload will follow the curve on the diagram from A to B, augmenting stroke volume up to the peak of the curve. Beyond this, an increase in preload will not further improve stroke volume. * Inotrope administration will increase myocardial contractility and on the diagram, this is demonstrated by a shift from A to C on the curve. Core Topics in Critical Care Medicine, eds. Fang Gao Smith and Joyce Yeung. Published by Cambridge University Press. © Fang Gao Smith and Joyce Yeung 2010. Vasoactive drugs Definitions For the purposes of the rest of the chapter the term vasoactive drugs refers to both inotropes and vaso- pressor drugs. There are very few pure inotropes or vasopressors. * Inotropes affect the force of myocardial contraction. A positive inotrope will increase myocardial contractility. * Vasopressors cause vasoconstriction of blood vessels (most act by α1 receptor activation) and therefore increase mean arterial blood pressure (MAP) and systemic vascular resistance (SVR). Mechanism of action Vasoactive drugs are used to support tissue perfusion and hence oxygenation. Vasoactive drugs act on various receptors in the body to produce their effects. Some drugs may act at more than one receptor to produce multiple effects. Some drugs have both inotropic and vasopressor activity (Table 9.1). Alpha-1 (α1) receptors These are present in vascular smooth muscle. * Stimulation of α1 receptors in vascular smooth muscle activates phospholipase C via Gq proteins to increase inositol triphosphate. This increases intracellular calcium, which causes vasoconstriction. * Examples of drugs that have α1 activity are phenylephrine and noradrenaline. Beta (β1 and β2) receptors Stimulation of β1 and β2 receptors via Gs proteins activates adenylate cyclase, which increases cyclic AMP (cAMP). This causes a protein kinase to phos- phorylate various proteins resulting in their action, e.g. muscle relaxation, inotropy, etc. Beta-1 (β1) receptors These are present in the myocardium. * Stimulation of β1 increases cAMP and subsequently intracellular calcium concentration. This increases myocardial contractility and heart rate. * β1 receptor activation also liberates renin from the juxtaglomerular apparatus in the kidney, releasing aldosterone and increasing reabsorption of sodium and water. Beta-2 (β2) receptors These are present in the vascular and bronchial smooth muscle. * β2 stimulation produces bronchial smooth muscle dilatation and vasodilatation of the blood vessels in skeletal muscle decreasing systemic vascular resistance. * Normally the β1: β2 receptor ratio in the heart is 3 : 1. Excessive sympathetic drive in heart failure may cause downregulation of β1 receptors to change the ratio to 3 : 2 respectively. Dopaminergic (D1 and D2) receptors D1 receptors act via Gs-coupled adenylate cyclase leading to increased cAMP and cause vasodilatation in the vascular and splanchnic smooth muscle. Centrally, D1 stimulation modulates extrapyramidal activity. D2 receptors act via Gi-coupled adenlyate cyclase leading to decreased cAMP and are located presynapti- cally. They have mainly modulatory functions and reg- ulate the neurotransmission by feedback mechanisms, Stroke volume C A B Preload CVP (mmHg) Normal heart Inotropes Heart failure Fig. 9.1 Frank–Starling curve illustrating the relationship between stroke volume and preload in a normal heart (A and B), with inotropes (C) and in heart failure. Chapter 9: Vasoactive drugs 59 * Activation of V1 receptors. * Modulation of K-ATP channels. * Modulation of nitric oxide. * Potentiation of adrenergic and other vasoconstrictors. Endogenous vasopressin levels are markedly reduced in adults with severe sepsis but not in children with meningococcal septic shock. Wenzel et al. performed a large trial of vasopressin versus epinephrine in human cardiopulmonary arrest that had promising results, but poor neurological out- comes raised controversy in introducing vasopressin into CPR guidelines. In severe sepsis and septic shock, current evidence suggested that although low doses of vasopressin may be effective in increasing blood pressure in patients refractory to other vasopressors, no outcome data were available and ‘cautious use of vasopressin pend- ing further studies’ was recommended. If vasopressin is used in advanced vasodilatory shock outside of clinical trials, it should be used as a supplementary vasopressor and in doses not exceeding 0.04U/min. Vasopressin is associated with ischaemic skin lesions in 30% of patients. Most of the skin lesions occurred on the distal limbs (68%), and 26% occurred on the tongue. Phenylephrine Phenylephrine is an α1 agonist with minimal β activity. * It causes arteriolar vasoconstriction and increases SVR and MAP. It causes a reflex bradycardia due to baroreceptor activation and cardiac output is decreased as a result. * Phenylephrine is a vasoconstrictor with minimal inotropic action. Its uses are mainly in patients with spinal shock and spinal–epidural anaesthesia. * Phenylephrine is not routinely used as a single agent in sepsis because of its tendency to reduce cardiac output. Phosphodiesterase enzyme inhibitors (PDEI) (e.g. milrinone/enoximone) The two groups of phosphodiesterase inhibitors are bipyridines (amrinone) and imidazole (e.g. enoxi- mone) derivatives. * Phosphodiesterase inhibitors (PDEI) increase cAMP and thus intracellular calcium. This increases cardiac output. PDEIs, however, unlike β1 agonists, do not require a receptor for their action. * PDEIs also reduce systemic and pulmonary vascular resistance. * PDEI drugs exhibit lusitropy, i.e. they relax the heart in diastole thus increasing coronary blood flow. * PDEIs are used in combination with other vasoactive drugs. * Enoximone may be effective in treating right heart failure and pulmonary hypertension by causing systemic and pulmonary vasodilatation and increasing right ventricular contractility. Levosimendan Levosimendan is a calcium sensitizer with some PDE III inhibitory actions similar to dobutamine. It is an inodilator that exerts its effects by combining with troponin C to enhance the calcium sensitivity of the myocardium. * A recent study in the treatment of heart failure compared levosimendan with dobutamine; levosimendan did not reduce mortality when compared with dobutamine. * There were higher incidences of atrial fibrillation, hypokalaemia and headache in the levosimendan group. Which vasoactive drug to use? * Hypovolaemic shock – fluid/blood initially * Anaphylactic shock – epinephrine * Cardiogenic shock – consider dobutamine +/− norepinephrine and +/− epinephrine * Septic shock – consider norepinephrine +/− dobutamine Management of shock When managing shock it is paramount to maintain blood pressure and tissue perfusion whilst curtailing unwanted side effects (Table 9.2). Resuscitation of a shocked patient begins with: Section I: Specific features of critical care medicine 62 * Optimizing airway, oxygenation and ventilation. * Restoring the circulatory volume. This may necessitate blood transfusion for haemorrhagic shock or crystalloid/colloid for septic shock, e.g. pancreatitis. * Supporting cardiac output. Consider vasoactive drugs, intra-aortic balloon pump, etc. * Treating the cause of the shock, e.g. drain a cardiac tamponade. * Treating factors that may affect cardiovascular function such as arrhythmias, hypocalcaemia, hypokalaemia, hypomagnesaemia, drug interactions and acidosis. * Bicarbonate should only be used for severe acidosis (pH<7.15) unresponsive to adequate resuscitation. It is best to be cautious as bicarbonate can cause inadvertent intracellular acidosis by raising intracellular pCO2. * Institute appropriate monitoring and investigations, e.g. blood cultures and cardiac output monitoring in sepsis (see Chapter 7: Haemodynamics monitoring), echocardiogram in cardiac failure, invasive blood pressure monitoring in hypovolaemic shock. Using vasoactive drugs If optimization of oxygenation, ventilation and adequate fluid resuscitation fail to restore cardiac out- put then treatment with vasoactive drugs should be considered. When considering vasoactive drugs, the adverse effects on myocardial perfusion, oxygen demand and organ perfusion should be taken into account. End organ perfusion can be impaired further due to excessive vasoconstriction. Myocardial oxygen- ation requirements can increase due to increased tachycardia in an effort to maintain cardiac output. Commencing vasoactive drugs in a volume-depleted patient (absolute or relative hyporolaemia due to sepsis) can have dire consequences for the patient. Initiating vasoactive therapy Most vasoactive drugs have mixed vasopressor and inotropic action. At high doses, the less predominant effects become more evident. A combination of vaso- active drugs with differing pharmacological actions may be needed to provide optimal effect. * The cause of shock should be determined, e.g. does the patient have low cardiac output state or septic shock (low SVR/MAP)? Monitoring considerations Appropriate parameters should be used to monitor for response and side effects e.g. ECG, ABP, CO, SV and urine output. Usually, vasoactive management is indi- cated to sustain a MAP greater than 60mmHg in the fluid resuscitated patient to sustain organ perfusion. * Patients with atherosclerosis and hypertension may need a higher MAP to maintain blood flow. ‘Adequate’ blood pressure is no guarantee of blood flow. * There is no ‘normal’ CVP. Serial trends of CVP and its response to therapy and fluid boluses should be monitored. * Tricuspid regurgitation, pulmonary hypertension and excessive ventilatory pressures can all give false CVP readings. * Very high doses of some vasopressors can also cause venoconstriction. This may cause a falsely elevated CVP. Response Therapy may need to be started with minute-by- minute assessment of the patient’s response. Table 9.2 Shock and vasoactive drugs Type of shock Heart rate CVP Cardiac output Stroke volume Systemic vascular resistance Considered vasopressor/ treatment initially Hypovolaemic ↑ ↓ ↓ ↓ ↑ Fluid/blood Septic shock ↑ ↓ ↑ ↓/↑ ↓ Norepinephrine Cardiogenic ↑ ↑ ↓ ↓ ↑ Dobutamine Chapter 9: Vasoactive drugs 63 The response to vasoactive drugs is often unpre- dictable and dependent on the cause of shock and baseline circulation. The same drug may produce dif- ferent responses depending on the patient’s myocar- dial compliance and volume status. The dosages of vasoactive agents should be reviewed frequently in the face of varying response. For patients who are on high doses of vasoactive drugs, steroid supplementation may be considered as steroids can have a facilitatory effect on the action of inotropes. In the past, high dose steroids use was asso- ciated with higher mortality in septic patients but a recent meta-analysis by Minneci revealed a consistent and significant beneficial effect of glucocorticoids on survival and shock reversal regardless of adrenal func- tion. The relationship between steroid dose and survival was found to be linear, characterized by benefit at low doses and increasing harm at higher doses. It is now recommended that a 5- to 7-day course of physiological hydrocortisone (150–200mg per day in divided doses) with subsequent tapering should be given to patients with vasopressor-dependent septic shock. Is the resuscitative therapy working? This is assessed by collectively and serially monitoring clinical response and measured variables. * Improved parameters, e.g. mean arterial blood pressure, stroke volume and pulse rate. * Improved thermoregulation (e.g. cardiogenic shock patient warms up). * Improved Urine output. * Improved PaO2. * Improved base deficit. * Reduced serum lactate. * Return of mixed venous saturations towards normal. * Cardiac index/output returns towards normal. Monitoring Serial trends and responses to monitoring are more important than isolated readings. * ECG – to monitor for arrhythmias and signs of myocardial ischaemia. * Arterial line – for beat-to-beat monitoring of blood pressure. A ‘swing’ in the blood pressure trace may indicate that the patient requires more filling. It also allows for serial sampling of arterial blood gases, base excess and lactate measurements. * CVP – to help guide right heart ‘filling’ or preload. It is not very useful for monitoring individual organ perfusion. If the patient has a history of ischaemic heart disease or tricuspid regurgitation the CVP may not give any indication of left heart pressures. Its other uses include administration of vasoactive drugs and to obtain mixed venous saturation sampling if a pulmonary artery flotation catheter (PAFC) is unavailable. * Hourly urine output. * Echocardiogram – to determine/exclude cardiac pathology. * Oesophageal Doppler – a non-invasive method of measuring stroke volume and cardiac output and is subject to variability. * Pulmonary artery catheter (if indicated) – its use is not without complications and is debatable. It is an invasive method of obtaining: (1) an indication the left heart filling pressure and volume (2) measurement of cardiac output (3) mixed venous oxygen saturation (SvO2) and oxygen consumption (DO2). * Mixed venous oxygen saturation (SvO2) – this is determined using a blood sample obtained from the pulmonary artery. If a sample from PAFC is not available then a CVP line sample may be used. Normal SvO2 is around 70%. * A low SvO2 indicates excessive extraction of oxygen from the blood or inadequate delivery. * A high SvO2 is difficult to interpret as it may be due to increased oxygen delivery or it may be due to the inability of the tissues to extract oxygen from the blood. * However, care should be taken with interpretation of SvO2. It is a very globalmeasurement. If normal, it may give no indication of individual organ ischaemia. Inotropes complications and considerations * Excessive stimulation of α1 receptors can lead to extreme vasoconstriction and may impair tissue perfusion. * Increased afterload may lead to impaired left ventricular function. Section I: Specific features of critical care medicine 64 Step 1: Calculate resting energy expenditure for caloric requirements This requires sophisticated equipment so it is more often estimated using formulae. One such formula is the Harris Benedict equation, which estimates basal metabolic rate (BMR) in kcal/day: BMR for male: 66 + (13.7 × W) + (5 × H) − (6.8 × A) = kcal/day. BMR for female: 655 + (9.6 × W) + (1.8 × H) − (4.7 × A)= kcal/day. Where W = body weight in kg, A = age in years, H = height in cm, ‘small’ calorie = 4.184 J, ‘large’ calorie = 1 kilocalorie (kcal). This will usually give a result of around 25 kcal/ day. The equation estimates BMR in afebrile healthy individuals. So it needs to be multiplied by the stress level: surgery = 1.2, starvation = 0.85–1, trauma = 1.35, sepsis = 1.6, severe burn = 2.1 Fever increase BMR by 10% for each 1°C above 37° C (up to max of 40°C) daily energy required for maintenance = BMR × stress factor × 1.25 (an additional 25% for hospital activity, not added if paralysed on a ventilator or heavily sedated). daily energy requirements for weight gain = maintenance + 750 kcal. Step 2: Calculate protein requirements * Normal: 0.8–1 g/kg/day protein (up to 60–70 g/day). Moderate depletion/stress: 1–1.5. Severe: 1.5–2. Less protein is needed in patients with renal failure before dialysis and hepatic encephalopathy. Step 3: Calculate non-protein (carbohydrates + lipids) component * Fat calories help decrease the risk of carbohydrate overload and keep total amount of fluid down. * Fat requirements should be less than 40% calories as fat may reduce the immune response. * Aminimum of 4% of total calories as essential fatty acids (linoleic). * Give the remaining energy requirements as carbohydrates. Calorie value of macronutrients (kcal/g) Fat 9, protein 4, carbohydrates 4, IV dextrose 3.4 and 1ml of 10% fat emulsion 1.1. Step 4: Calculate micronutrients (vitamins, electrolytes and trace elements) Daily requirements for vitamins A 3300 IU, D 200 IU, E 10 IU, B1 3 mg, B2 3.6 mg, B3 40 mg, B5 15 mg, B6 4 mg, B7 60 mg, B9 0.4 mg, B12 5 mg, C 100 mg, K 2–4 mg/week. Patients with sepsis have been shown to have large vitamin A losses. Daily electrolyte requirements (in mmol/kg per day) Na+ 1.0–2.0, K+ 0.7–1.0, Ca++ 0.1, M++ 0.1, Cl− 1.0–2.0, PO4 − 0.4. Catabolism and loss of lean body mass can occur if low in K, Mg, Zn, P and S. Daily requirements for trace elements (in micrograms) Chromium 10–15, copper 500–1500, manganese 150– 800, selenium 30–60, zinc 2500–4000. Burns patients lose selenium, zinc and copper via their exudates and trauma patients lose selenium and zinc through their drains. Routes of administration Nutritional support can be given through one of two routes: enteral feeding (EF) (via the gastrointestinal tract) or parenteral feeding (PN) – intravenous (via either peripheral or central vein). Enteral feeding (EF) * Enteral feed involves using the gastrointestinal tract for the delivery of nutrients; this includes eating food, oral supplements and all types of tube feeding. * If at all possible enteral is the preferred route for nutritional support. It is cheaper, more physiological, reduces the risk of peptic ulceration and minimizes mucosal atrophy. * If the patient can eat orally then this should be encouraged. It is important to know how much the patient is eating to see whether they are receiving adequate nutrition. If not they will need supplementation either orally or enterally. Chapter 10: Nutrition 67 * Early enteral nutrition could lead to lower stress hormone concentrations, lower infection rates, shorter hospital stays. Better survivals have been shown in some, but not all, studies in which nutrition was started within 4–24 hours. * Lack of enteral intake can lead to small intestinal villus atrophy, decreased villus cell count and reduced mucosal thickness. Mucosal surface patterns change from finger-like to leaf-like microvilli. Intestinal permeability is increased. These changes can be reversed with enteral feeding. * These changes in gut integrity could cause intestinal translocation of bacteria. This can activate the gut’s immune inflammatory system (Peyer’s patches and hepatic Kupffer cells). The released cytokines and other mediators then exacerbate the already existing systemic inflammatory response syndrome leading to multiple organ failure. This is called the ‘gut hypothesis of multiple organ failure’. * Situations previously thought to preclude enteral nutrition, including major gastrointestinal surgery or acute pancreatitis, have now been shown to be best treated with enteral nutrition. Requirements for enteral feeding Patients should be haemodynamically stable with no history of massive GI bleeds, no intestinal obstruction, no severe diarrhoea, no high-output enteric fistula or abdominal distension. Assessment of gut function Gut output should be less than 600ml/day. Presence of bowel sounds (does not necessarily correlate with peri- stalsis and passage of flatus or stool is a better marker of gut function. Access for enteral feeding Gastric route The route of EFmost often used in critical care is naso- gastric tubes. Some patients may already have percuta- neous endoscopic gastrostomy (PEG) tubes in situ, if it is known they will not be able to feed for some time after surgery. Jejunal route Nasojejunal tube feeding is considered ideal so as to avoid the often present gastric ileus and possibly prevent aspiration. A nasogastric tube can be passed during surgery and manually manipulated into the jejunum. In patients without surgical placement, nasogastric tubes may be placed. It is often difficult to pass such tubes from the stomach into the jej- unum. Right lateral positioning and prokinetic drugs may be attempted, but often placement has to be performed under radiological or endoscopic guidance. Jejunostomy catheter can be placed during lapa- rotomy to start feeding within hours of bowel surgery. Complication rates are comparatively low (1.5%). The most common complications are occlusion or dis- lodgement of feeding catheters. Types of feeds Polymeric preparations Enteral feeding solutions usually contain homogen- ized substrates similar to those found in normal feeds and are termed polymeric solutions. Elemental or semi-elemental solutions contain free amino acids or hydrolysed proteins, glucose, or oligosaccharides and medium-chain triglycerides to facilitate digestion and absorption in patients with altered digestive func- tion. However, most patients can be fed with poly- meric solutions, and these should be preferred to elemental formulae, since they are less likely to induce diarrhoea and are associated with improved nitrogen retention and improved gut trophicity, while being considerably less costly. * Polymeric preparations contain normal proteins, fats and carbohydrates, which require digestion prior to absorption, and electrolytes, trace elements, vitamins and fibre. * Commonly used ingredients include the protein casein (from milk), soya protein, maize and soya oils and the carbohydrate maltodextrin. * The different preparations vary in their osmolality, calorie:nitrogen ratios and carbohydrate:lipid ratios. They provide energy between 0.5 and 2 kcal/ml although most are 1 kcal/ml. * Standard polymeric feeds (generally 1 kcal/ml) are most often used, although there are higher- energy alternatives (1.2 or 1.5 kcal/ml) available for patients who need more calories in a shorter period of time, or who do not tolerate large volumes. Section I: Specific features of critical care medicine 68 Disease-specific formulae * Liver disease: low sodium and altered amino acid content to minimize hepatic encephalopathy. * Renal disease: low phosphate, potassium and high energy (2 kcal/ml) to reduce volume of fluid intake. * Respiratory disease: high fat content to reduce CO2 production. Elemental preparations These preparations contain the macronutrients in a readily absorbable form, i.e. proteins as peptides or amino acids, lipids as medium-chain triglycerides and carbohydrates as mono- and disaccharides. These prep- arations are expensive and only indicated for patients with severe malabsorption or pancreatic insufficiency. Parenteral nutrition (PN) Special considerations Intensive care clinicians should make all the efforts to improve tolerance of enteral feeding before consider- ing parenteral nutrition. It is unclear, however, about the duration of failure in enteral feeding before PN should start, which mostly depends on pre-existing nutritional status and the disease process. * Patients receiving less than 25% of their predicted nutrition requirements are at increased risk of sepsis. Most units believe that PN is better than no nutritional support and will start PN in patients who are expected not to tolerate adequate enteral feeding for 7 days. * The only absolute indication for PN is gastrointestinal failure. * PN can be used to supplement enteral nutrition, for example in short gut syndrome. * PN can be used as the sole source of nutrition as total parenteral nutrition (TPN). Components of parenteral nutrition Parenteral nutrition is commonly administrated as a sterile emulsion of water, protein, lipid, carbohydrate, electrolytes, vitamins and trace elements. Components and volumes of PN are based on the recommendations discussed earlier on nutritional requirements. However, PN can be given as separate components. Protein Protein is given as amino acids which include essential amino acids – histidine, leucine, isoleucine, lysine, threonine, methionine, phenylalanine, tryptophan and valine and most of the non-essential amino acids. Lipid Lipid provides a source of essential fatty acids – linolenic acid (an omega-3 fatty acid) and linoleic acid (an omega- 6 fatty acid). Lipid is also important for absorption of fat-soluble vitamins. * Commonly given as intralipid, an emulsion made from soya with chylomicron-sized particles. * Propofol is made of 10% intralipid and it should be included in calculation of energy requirement to avoid over-feeding. * Intralipid can be used alone. Carbohydrate Commonly given as glucose. It can be given alone initially for its protein-sparing effect in starvation. Electrolytes and micronutrients These should usually be included in the emulsion. Otherwise, they need to be given separately. Pharmaco/immunonutrition This is a relatively new concept in critical care feeding for which there is a growing body of evidence report- ing benefits. Despite adequate nutritional support, nosocomial infection still remains a major problem in critically ill patients. Therefore, addition of specialized nutrients to the standard diet has been suggested in order to decrease patients’ susceptibility to infection by enhancing the immune response. The target of the so-called ‘immunonutrients’ can be the gastrointestinal tract (i.e. the enterocytes or the immune cells of the intestinal wall) in order to prevent or diminish the translocation of bacteria or bacterial products. During critical illness, the amount of fuel for these rapidly renewed cells may become the rate-limiting step of an appropriate immune response. The function of circulating immune cells (mainly lymphocytes and macrophages) may also be influenced by the dietary constituents. Chapter 10: Nutrition 69 Chapter 11Pain controlEdwin Mitchell The importance of pain relief Pain is a common finding amongst patients on the intensive care unit, and by definition is unpleasant expe- rience. The principal reason for treating pain is human- itarian, to relieve the suffering of a fellow human being, but there are other potential benefits as well. Effective analgesia is associated with * improved ability to cough, clear secretions and a reduced rate of respiratory failure following major abdominal and thoracic surgery * reduced duration of intestinal stasis following abdominal surgery * reduced sedation requirements. * improved compliance with physical rehabilitation, with positive outcomes on joint mobility, and lower rates of deep venous thrombosis * minimized surgical stress response * reduced incidence of chronic pain syndromes. Sources of pain Pain may be considered to originate from either the underlying pathological process, e.g. trauma, pancrea- titis or myocardial infarction, or be iatrogenic from medical procedures, e.g. surgery, line insertion points and poor patient positioning. Approximately 60% of patients on general inten- sive care units in the United Kingdom are suffering from surgical conditions, most of these have had an operation. Post-operative pain is common and often severe; 70–90% of patients report pain of severe/ unbearable intensity following surgery. Types and severity of pain Most of the pain experienced on the intensive care unit is acute pain that results directly from traumatized tissue. The severity of the pain is often proportional to the amount of tissue trauma, but it must be borne in mind that patients may suffer from several painful areas at once, and it is often insufficient to treat any one painful area alone, no matter its size. The nature of surgery is often another important factor to determine the severity of acute pain. For example, thoracic open surgery brings more severe acute pain than extensive head and neck surgery. Assessment of severity of pain Assessing the severity of pain in intensive care patients is challenging. Patients may be sedated, or unable to communicate due to the presence of endotracheal tubes, dressings and other impediments. Methods such as the visual analogue scale, or pain score, are frequently used to assess the severity of pain (Fig. 11.1). These are quick, easy, reproducible and sensitive to changes in the patient’s condition, but can be difficult to administer in patients who have limited ability to communicate. Alternative forms of assessing pain relief that score pain behaviours (e.g. grimacing) or score physiological changes (e.g. heart rate and blood pressure changes), are avail- able, but patients who recover from critical illness frequently recall pain of greater severity than their physiological changes have suggested. Modalities of pain relief Analgesia may be provided either systemically or regionally (Table 11.1). Multimodality approaches are often employed, with multiple techniques and drugs used to treat a particular pain state. This approach has the benefits of minimizing the amount of potentially toxic drugs, such as opioids, as well as providing better pain relief. Core Topics in Critical Care Medicine, eds. Fang Gao Smith and Joyce Yeung. Published by Cambridge University Press. © Fang Gao Smith and Joyce Yeung 2010. Systemic analgesia Paracetamol A simple analgesic that is very useful for soft tissue pain and as an opioid sparing drug inmore severe pain states. Paracetamol is available in an oral, rectal or intravenous preparation. Oral bioavailability is high, with absorption occurring rapidly from the duode- num. Paracetamol may cause mild derangement of liver function tests in normal doses, and caution should be exercised in patients with severe hepatic impairment. Paracetamol has anti-pyretic properties and has been associated with mild hypotension in critical care patients. Non-steroidal anti-inflammatory drugs Non-steroidal anti-inflammatory drugs (NSAIDs) are powerful analgesics, particularly useful for bony pain. They have an opioid sparing effect when used in combi- nation with opioids. All NSAIDs work by the inhibition of the cyclooxygenase (COX) enzyme, and this may lead to adverse effects such as bronchospasm, renal impair- ment, gastric irritation, bowel perforation and platelet inhibition. The elderly are at particular risk from the adverse effects of NSAIDs. Critically ill patients who are at risk from stress ulceration and renal impairment must be carefully assessed before using NSAIDs. Cyclooxygenase II (COX-II) inhibitors were devel- oped in an attempt to remove the adverse side effects of traditional NSAIDs. COX-II inhibitors work by inhib- iting only the induced form of cycloxygenase which is produced by tissue trauma. They have been associated with an increased risk of cardiovascular events in patients who have used these drugs for some time and their role in the critical care population is uncertain. Tramadol Tramadol is an atypical analgesic with antagonist actions at morphine receptors and inhibitory effects on the reuptake of serotonin and norepinephrine from the synaptic cleft. Potentiation of serotinergic neurons may be important in activating the descending pain control neuronal pathways. It is useful for moderate pain and may be administered orally or by intravenous injection. Oral bioavailability is high, approaching 100% with repeated dosing. The major adverse reaction is nausea and vomiting, but it is also associated with sedation, confusion and hallucinations, particularly in the elderly. Table 11.1 Types of analgesia Systemic Regional Paracetamol Neuraxial blockade NSAIDs Peripheral nerve blocks Tramadol Opioids Ketamine Inhaled – nitrous oxide MODERATE 0 NO PAIN Verbal Descriptor Scale WONG-BAKER FACIAL GRIMACE SCALE ACTIVITY TOLERANCE SCALE SPANISH TAGALOG CHINESE KOREAN PERSIAN (FARSI) VIETNAMESE JAPANESE Alert Smiling No humor serious flat Furrowed brow pursed lips breath holding Wrinkled nose raised upper lips rapid breathing Slow blink open mouth Eyes closed moaning crying NO PAIN NADA DE DOLOR Walang Sakit Konting Sakit Katamtamang Sakit Matinding Sakit Pinaka-Matinding Sakit Pinaka-Malalang Sakit UNPOQUITO DE DOLOR UN DOLOR LEVE DOLOR FUERTE DOLOR DEMASIADO FUERTE UN DOLOR INSOPORTABLE CAN BE IGNORED MILD PAIN MODERATE PAIN MODERATE PAIN SEVERE PAIN WORST PAIN PASSIBLE BEDREST REQUIRED INTERFERES WITH BASIC NEEDS INTERFERES WITH CONCENTRATION INTERFERES WITH TASKS 1 2 3 4 5 6 7 8 9 10 This pain assessment tool is intended to help patient care providers assess pain according to individual patient needs. Explain and use 0–10 Scale for patient self-assessment. Use the faces or behavioral observations to interpret expressed pain when patient cannot communicate his/her pain intensity. UNIVERSAL PAIN ASSESSMENT TOOL Fig. 11.1 An example of a pain assessment tool. Chapter 11: Pain control 73 Opioids Opioids are themost potent analgesics widely available and form the basis of most critical care pain manage- ment treatments. All opioids act via opioid receptors, which are found centrally, in the spinal cord and peripherally. Opioids mimic endogenous neuropeptides, such as endorphins, which have analgesic properties. Activation of the morphine receptor causes inhibition of adenyl cyclase, hyperpolarization of the neuron, and a reduction in signal transmission. Opioids have a similar range of effects through their actions at the morphine receptors: * analgesia * sedation * respiratory depression * miosis * nausea and vomiting * reduced intestinal motility * tolerance and dependence. The first two of these properties make opioids popular choices in sedation regimes in the critically ill. The different drugs may have differing pharmacokinetic properties, however (Table 11.2). Morphine Morphine may be administered orally, subcutane- ously, intramuscularly, intravenously, epidurally or intrathecally. Its oral bioavailability is only 50%. Morphine is widely used due to the versatility in route of administration and low cost. Morphine undergoes conjugation in the liver to a number of metabolites, including morphine-6-glucuronide, a molecule which retains opioid agonist activity, and may accumulate in renal failure. Diamorphine Diamorphine is a diacylated morphine pro-drug. It has no intrinsic analgesic activity, but is rapidly deacetylated into morphine in vivo. Diamorphine hydrochloride is more soluble than morphine sulf- phate, and this allows it to be prepared in smaller volumes, an advantage when preparing small-volume syringe drivers. Fentanyl A synthetic opioid with a very high lipid solubility, fentanyl has a shorter half-life in the plasma than morphine following bolus administration, principally due to redistribution. Following multiple boluses or infusions redistribution may become slower and elim- ination half life then becomes important. Fentanyl depends on hepatic clearance. It is often used in epi- dural analgesic regimes at very low concentrations. Alfentanil A synthetic opioid similar to fentanyl, but with a lower potency due to lower lipid solubility. More alfentanil is present in the unionized state in plasma compared to fentanyl, and this means it has a more rapid onset of action than fentanyl. Compared to fentanyl and mor- phine, alfentanil has a shorter terminal half-life, mean- ing recovery is more rapid following infusions. Alfentanil is metabolized in the liver, and its clearance is relatively unaffected by renal failure. Remifentanil Remifentanil is an ester that is rapidly hydrolysed in the plasma to a virtually inactive compound. The half-life in plasma of remifentanil is less than 10 mins, resulting in a rapid recovery from its effects. Analgesia provided by infusion with remifentanil may be suitable where the painful stimulus is not expected to be long lasting, and in patients with renal failure. Codeine Codeine is a weak opioid. Approximately 10% of an orally administered dose is O-demethylated to mor- phine which accounts for the analgesic action. This depends on cytochrome CYP2D6 enzyme activity in the liver (10% of the UK population lack this enzyme). Codeine is much less efficacious for pain relief in these patients. Codeine has higher bioavailability than mor- phine when administered orally. Table 11.2 Pharmacokinetic properties of opioids Drug Vd (l/kg) Clearance (l/min) t½ (h) Alfentanil 0.8 6 1.6 Fentanyl 4.0 13 3.5 Morphine 3.5 15 3 Remifentanil 0.4 40 0.1 Tramadol 2.9 6 7 Section I: Specific features of critical care medicine 74 Chapter 12 SedationJoyce Yeung Introduction For many patients, the critical care environment can be a frightening and stressful environment and anal- gesics and sedatives are used to improve the comfort and safety of critically ill patients. Agitation is believed to be present in at least 71% of patients on intensive care. The majority of critically ill patients cannot com- municate easily the way they feel or what they need. Procedures such as tracheal intubation, ventilation, suction and physiotherapy cannot be tolerated without adequate level of sedation. However, continuous administration of sedatives can prolong the time on mechanical ventilation and ICU stay. The correct management of sedation is one of the most important and often difficult goals to achieve in critical care. Although the mainstay of therapy will be pharma- cological, other approaches are just as important and should not be neglected. Good communication and regular assurance from nursing staff can help alleviate anxiety of patients in unfamiliar surroundings. Environmental control such as temperature, noise and lighting can provide a restful environment. The man- agement of thirst, constipation and full bladder can help with the general comfort and well-being of patients. Aims The American College of Critical Care has recommen- ded that the use of analgesics, sedatives and neuro- leptics for treatment of pain, anxiety or psychiatric disturbance in the intensive care unit should be used as ‘agents to mitigate the need for restraining method and not overused as a method of chemical restraint’. The main goals should be: (1) Patients should be comfortable and provided with adequate analgesia. Studies have shown that 70% of patients in ICU remembered being in pain despite their healthcare providers believing them to be pain free. Adequate analgesia can often reduce the need for deep sedation. (2) Anxiety should be minimized as it can reduce efficiency of ventilatory support and agitated patients have higher metabolic rate and higher oxygen consumption. (3) Patients should be able to tolerate procedures and organ system support. (4) Patients should be calm, co-operative and able to sleep when undisturbed. (5) Patients should not be paralysed and awake. Levels of sedation The required level of sedation will vary according to the patient being cared for. Deep sedation will be required for status epilepticus but the modern venti- lator can work with patients who are lightly sedated and spontaneously breathing. The desired level of sedation should be documented and once sedation is started, the level of sedation should be regularly assessed. Oversedation will lead to increase time on mechanical ventilation, increased risk of nosocomial pneumonia and unnecessarily prolonged ICU stay. It may also increase the need for frequent neurological assessments such as CT scans and increase the inci- dence of long-term cognitive and psychological prob- lems in patients. Undersedation, on the other hand, can cause hypercatabolism, immunosuppression, hypercoagulability and increased sympathetic activity. Many scoring systems have been devised to pro- vide an assessment of levels of sedation. Vital signs, such as blood pressure and heart rate, are not specific or sensitive markers of sedation and not useful in assessing sedation. A reliable and practical objective method of assessment is still being developed. Core Topics in Critical Care Medicine, eds. Fang Gao Smith and Joyce Yeung. Published by Cambridge University Press. © Fang Gao Smith and Joyce Yeung 2010. Assessment tools Clinical scoring systems There are many clinical scoring systems and commonly used ones include Ramsay (Table 12.1) and Bloomsbury scales (Table 12.2). These are designed to give a quanti- tative score which can be done regularly by nursing staff and documented on observation charts. Their limita- tions include interpreter variability and the lack of dis- crimination between deeper levels of sedation. Electroencephalograms Electroencephalograms (EEG) can provide a measure of cerebral activity. This technique is complex and requires trained personnel. It is more suitable for the assessment of depth of anaesthesia and can be difficult to interpret in the encephalopathic patient. Bispectral index The bispectral index (BIS) has been used successfully to monitor the depth of anaesthesia in the operating theatre environment (Fig. 12.1). The monitor gives a quantitative value between 0 to 99, with value of 0 representing EEG silence and 100 representing a fully awake patient. Some studies have shown good corre- lation between Ramsay Score of 1–5 and BIS in critical care settings but correlation is more variable when Ramsay Score is 6. Auditory evoked potentials Evoked potential monitors measure electrical activity in certain areas of the brain in response to stimulation of specific sensory nerve pathways. Auditory evoked potentials (AEP)monitoring technique isolates the neu- rophysiological signal generated during stimulation of cranial nerve VIII using a repetitive auditory stimulus (Figs. 12.2 and 12.3). The repeated sampling allows the signal to be extracted from the background EEG noise. Studies have suggested that long latency auditory evoked potentials can provide an objective electrophy- siological analogue to the clinical assessment of sedation independent of the sedation regime used. Table 12.1 Ramsay Scale Level Response Awake 1 Patient anxious and agitated or restless or both 2 Patient co-operative, orientated and tranquil 3 Patient responds to commands only Asleep 4 Brisk response to a light glabellar tap or loud auditory stimulus 5 Sluggish response to a light glabellar tap or loud auditory stimulus 6 No response to a light glabellar tap or loud auditory stimulus Table 12.2 Bloomsbury Scale Sedation score 3 Agitated and restless 2 Awake and comfortable 1 Aware but calm 0 Roused by voice −1 Roused by touch −2 Roused by painful stimuli −3 Unrousable A Natural sleep P Paralysed Fig. 12.1 A bispectral index (BIS) monitor displaying BIS value with BIS sensor strips for attachment. (Photo courtesy of Aspect Medical Systems.) Section I: Specific features of critical care medicine 78 Pharmacological management Loading dose It takes four half-lives of a drug given by intravenous infusion to achieve steady state. It is therefore neces- sary to start with a loading dose to minimize delays to achieve adequate sedation. However, the initial high infusion rates for a loading dose are often not required to be continued once the steady state is achieved. Any increase in infusion rates should be by small incre- ments as high infusion rates will encourage tolerance to sedatives. Side effects of sedatives Currently there is no ideal sedative agent available and pharmacological agents will all share the same side effects: * Accumulation with prolonged infusion leading to delay in weaning from organ support and prolonged ICU stay. * Adverse effects on circulation and blood pressure resulting in inotropic support. * Adverse effects on pulmonary vasculature, increasing V/Q mismatch leading to increased ventilatory support and risk of nosocomial pneumonia. * Tolerance with continued use. * Withdrawal symptoms when sedation is stopped. * Sedative agents do not provide rapid eye movement (REM) sleep needed for rest. REM sleep deprivation is an important cause of ICU psychosis. * Reduced intestinal motility interfering with the absorption of enteral feed and drugs. Ideal properties of sedatives The ideal properties of sedatives are summarized in Table 12.3. Commonly used sedatives Choice of sedative will vary according to local guide- lines and cost effectiveness. Combinations of sedatives AWAKE 100 60 50 40 30 25 20 15 10 0 LIGHT ANAESTHESIA SURGICAL ANAESTHESIA DEEP ANAESTHESIA Fig. 12.3 Diagram illustrating AEPmonitor displaying index number of 21. For this particular monitor, an index of below 25 indicates adequate anaesthesia. (Photos and illustration courtesy of Danmeter, Denmark.) Fig. 12.2 AEP monitor attached to surface electrodes on patient’s head and headphones providing the auditory stimulus. (Photo courtesy of Aspect Medical Systems.) Chapter 12: Sedation 79