Sepsis-Manual-Sixth-Edition 2022.pdf

6th edition Created 2022 THE SEPSIS MANUAL Responsible management of sepsis, severe infection and antimicrobial stewardship. ...

6th edition Created 2022 THE SEPSIS MANUAL Responsible management of sepsis, severe infection and antimicrobial stewardship.  Published by: United Kingdom Sepsis Trust Email: [email protected] Website: sepsistrust.org The UK Sepsis Trust combines clinical expertise with translational knowledge to equip health professionals with the knowledge and tools to help them recognise and manage sepsis decisively and responsibly without overuse of valuable antibiotics. We also support people affected by sepsis, and work to heighten public awareness so that members of the public can access healthcare promptly when worried about sepsis. 2022 United Kingdom Sepsis Trust All rights reserved. No part of this book may be stored in a retrieval system or reproduced in any form whatsoever without prior permission in writing from the publisher. This book is sold subject to the condition that it shall not, by trade or otherwise, be lent, resold, hired out or otherwise circulated without the publishers prior permission in any form of binding or cover other than that in which published, and without a similar condition including this condition being imposed on the subsequent purchaser. ISBN: 978-1-3999-2367-5 The UK Sepsis Trust registered charity number (England & Wales) I 158843 I (Scotland) SC050277. Company Reg No 8644039. Sepsis Enterprises Ltd company number 9583335. VAT Reg No 293133408. 2 THE SEPSIS MANUAL 6th edition Edited by: Dr Ron Daniels and Professor Tim Nutbeam Publisher: United Kindgom Sepsis Trust 3  FOREWORD “ "The world must urgently step up efforts to improve data about sepsis so all countries can detect and treat this terrible condition in time. This means strengthening health information systems and ensuring access to rapid diagnostic tools, and quality care including safe and affordable medicines and vaccines.” – Dr Tedros Adhanom Ghebreyesus, WHO Director-General In 2017, the World Health Assembly, the decision-making body of the World Health Organisation (WHO), adopted a resolution to improve the prevention, diagnosis and management of sepsis. This resolution marked a new era in our fight against sepsis. All 194 member states of the United Nations are urged to develop national action plans against the condition, which is responsible for 1 in 5 lives lost worldwide. In 2020, the Institute for Health Metrics and Evaluation published a report suggesting that sepsis affects 49 million people worldwide each year, claiming 11 million lives. Many of these fatalities are in children, particularly in low and middle income countries. In the United Kingdom, there are at least 200,000 episodes of sepsis in adults each year and perhaps as many as 918,000 (Academy of Medical Royal Colleges 2022), with around 48,000 people dying as a result. Sepsis claims more lives than breast, bowel and prostate cancer put together. Many people who develop sepsis have underlying medical conditions, and a significant proportion are frail or approaching the end of natural life. We can't prevent every death from sepsis, but we need to work hard to reduce the thousands of avoidable deaths. In the context of the rising threat of antimicrobial resistance, we must do so responsibly. Antimicrobials must be preserved for the sickest patients, and used correctly – otherwise we risk the very real threat of being unable to treat our patients in the not-too-distant future. We've come a long way since we started this fight over a decade ago. We understand sepsis better and have designed effective clinical systems around it. In some countries (including the UK) these steps have resulted in gradual improvements in survival. In the wake of the COVID-19 pandemic, though, the resilience of such learning and systems has been tested, and we must strive once more to improve the quality of care. To achieve our dream of preventing any avoidable death from sepsis, we’ll need continued effort from governments, policy makers, professional bodies, the public, the media – and from you. I hope this manual will mark the start of a new and reinvigorated phase in your fight against sepsis, because this involves every one of us. This manual has been extensively revised to welcome and help implement the 2022 guidance for the UK led by the Academy of Medical Royal Colleges, paving the way for a comprehensive, joined-up approach to sepsis management across the U.K. With very best wishes Dr Ron Daniels B.E.M, FFICM, FRCA, FRCP(Ed) Vice-President - Global Sepsis Alliance CEO - UK Sepsis Trust 4  THE BURDEN OF SEPSIS AND SEVERE INFECTION CONTENTS 06 DEFINING SEPSIS 12 SOURCES OF INFECTION 29 THE PATHOPHYSIOLOGY OF SEPSIS 39 THE SEPSIS 6 53 ONGOING CARE 84 SPECIAL PATIENT GROUPS 90 MICROBIOLOGY 96 HUMAN FACTORS IN SEPSIS 104 AFTER SEPSIS - SURVIVOR ISSUES 108 THE BURDEN OF SEPSIS AND SEVERE INFECTION 01 ESTIMATING THE BURDEN OF SEPSIS & SEVERE INFECTION Sepsis and severe infection are one of the most common reasons for admission to hospital, and perhaps the most common cause of inpatient deterioration. Whilst this statement might well be true, and other than knowing that it is a significant issue, the reality is that we still don’t truly understand the burden of sepsis. This introductory chapter will start by describing how we use the best available data to estimate: i) How many cases of sepsis we see each year across the United Kingdom (section 2) ii) How many people die as a result of sepsis (section 3) iii) The economic burden to our healthcare system and to the wider economy (section 4) 02 NUMBER OF CASES Across each country, hospital coded data are collected at national level in order to examine disease trends and inform policy and commissioning of healthcare, which capture the number of 'episodes' of sepsis (not the same as the number of people, as some people may develop sepsis more than once!). A clinical coder will interpret what's written in a set of notes and translate it into a set of codes, based on the International Classification of Disease criteria. These are currently in their 11th iteration, ICD-11, which incorporates the 2016 'Sepsis-3' definition. However, this system isn't quite perfect, and might give rise to an under-estimate of the number of people with sepsis. As one example, a 2015 report, ‘Just Say Sepsis’, by the National Confidential Enquiry into Patient Outcome and Death (NCEPOD) found that, where patients with sepsis had died, it was only recorded on the death certificate in 40% of cases. CLINICAL PRACTICE TIP Accurate record keeping is a vital part of good clinical practice. What we write in the notes affects not only the care of the individual patient, but also coding. In turn, coding affects, for example, how much a hospital gets paid; and more importantly our broader societal understanding of clinical and public health issues. If we write in the notes ‘possible sepsis’, or ‘? sepsis’ and no one subsequently confirms the clinical diagnosis, the patient will not be coded as having sepsis even if they end up on Intensive Care with multi-organ failure as a result. So, if you think sepsis, remember to say ‘sepsis’, write ‘Diagnosis: sepsis’ or ‘Δ sepsis’, and assess and record the level of severity, or acuity. More about this below, but remember, coding matters! 7  In England, coded data are assimilated into ‘Hospital Episode Statistics (HES)’ data. When we look at data like these over the years, it appears that sepsis is becoming 'more common', but it's likely most of this is down to improving recognition and coding. However, our ageing population and increasing tendency to perform a greater number of invasive interventions will have a significant effect. Antimicrobial resistance plays a small, but ever-growing part. An analysis by the UK Sepsis Trust of HES data for England for the year 2017/18 showed around 200,000 admissions to hospital with sepsis. How does this estimate stack up against other evidence? Recent work in the UK and further afield estimates around 5% of Emergency Admissions to be due to sepsis. There were around 4.5 million Emergency Admissions during 2017-18 – this would mean around 226,000(1) cases of sepsis in the UK. per year International data on the incidence of sepsis vary widely. A 2001 United States study suggested an incidence of 300 episodes per 100,000 population, whilst a 2016 population- based study from Sweden identified an incidence of 780 per 100,000 per annum. With a population of just over 67 million, these incidence figures would suggest an annual 202,500(2) and 526,500(3) cases of sepsis in the UK respectively In January 2020, the Global Burden of Disease team estimated that the UK sees 245,000(4) cases of sepsis with 48,000 deaths (IHME, 2020). However, the study acknowledged imperfections in the use of coded data alone In 2022, the Academy of Medical Royal Colleges estimated that there are 918,000(5) admissions to hospital with 'suspicion of sepsis' across the UK each year (extrapolated from Inada-Kim 2017) Applying these sense checks to the estimate yielded by coded data would therefore seem to support that there are at least 200,000 cases of sepsis each year in the UK, more likely 250,000, and possibly quite a few more! To put this into context, latest figures from the British Heart Foundation estimate there to have been 100,000 admissions with heart attack in the UK per year. It's important to acknowledge that it's often difficult to distinguish clinically between sepsis and severe infection, particularly at first presentation. There are well over 1.5 million episodes of the most common sources of infection giving rise to sepsis each year in England alone- it is likely that at least a proportion of these will also have sepsis but not be coded as such. 8 03 NUMBER OF DEATHS Between 2015 and 2020, governments in Scotland and Wales reported national mortality rates of 20% and 24% respectively. In 2018, Professor Sir Brian Jarman reported mortality rates in England to be just below 20%. If we apply a 20% mortality rate across the estimated 250,000 people developing sepsis annually in the UK, we would estimate that we see 50,000 deaths each year. The Global Burden of Disease team estimate from 2020 for the UK was similar, suggesting 48,500 deaths in the UK per year. In 2022, the Academy of Medical Royal Colleges suggested a higher number of deaths at 66,096 each year. It seems highly likely that, across the UK, sepsis claims 48,500 lives per year. SEPSIS CLAIMS MORE LIVES THAN LUNG CANCER, AND MORE THAN BOWEL, BREAST AND PROSTATE CANCER COMBINED 9  04 THE COST OF SEPSIS In 2017, the UK Sepsis Trust commissioned an independent piece of work from the York Health Economics Consortium (YHEC) to estimate the cost burden of sepsis to the NHS, and to our wider economy. YHEC estimated direct costs to the NHS based upon the use of consumables, drugs, clinical time and bed days in hospital, together with the need for rehabilitation, ongoing organ support and other access to healthcare. The group also estimated indirect costs, primarily due to lost productivity, but also in litigation. Clearly, if a patient suffers a sepsis-related death, they are unable to return to productive life, and they will not be able to pay taxes – the same is true for many survivors. For example, 22% of survivors of sepsis who have needed Intensive Care have post-traumatic stress disorder; and 17% of survivors have moderate-to-severe cognitive decline. Even if we do save a life, and particularly if we delay diagnosis and treatment, survivors might struggle to return to their previous quality of life for some time.. YHEC estimated, given that there are at least 200,000 cases of sepsis every year, that sepsis costs the NHS between £1.5 and £2 billion each year, and our wider economy at least £11 billion and possibly as high as £15.6 billion. The direct cost of sepsis to the NHS accounts for a full 1% of NHS budget. To put these figures into context, the Asthma UK Centre in Applied Research estimates the annual cost to the NHS of treating Asthma to be £1.1 billion. SEPSIS COSTS THE NHS MORE THAN ASTHMA 10 CONCLUSION Whilst we have improved our recording of the number of cases of sepsis and understand better its impact on the NHS and society, we still have to estimate figures based on the best available data. Conservative estimates would suggest that we see at least 200,000 cases of sepsis in the UK each year, with around 48,500 deaths and a direct cost to the NHS of at least £1.5 billion. Sepsis costs our society as much as £15.6 billion every year. It is likely that even these numbers are under-estimates, since a proportion of the more than 1.5 million patients suffering severe infection in England every year are likely to have uncoded sepsis. The Academy of Medical Royal Colleges in 2022 estimated that we may see as many as 66,096 deaths each year from sepsis. Whichever way we cut it, sepsis is huge. 11 DEFINING SEPSIS INTRODUCTION The definition of sepsis has changed over time, and will continue to do so. These changes have, at times, created confusion, but it is hoped that from the time of writing there will be a period of stability for some years while we continue to advance improvements in clinical systems and understanding. There are various purposes to a definition for any condition, including: The use of a common language to improve communication between health professionals, and between healthcare systems and their patients The use of language suitable to educate the well public about the condition The establishment of criteria and thresholds beyond which intervention is recommended Provision of criteria to determine eligibility for inclusion in a clinical trial, audit or assessment A single description can't fulfil all of these purposes. In a complex condition like sepsis (which can affect multiple organ systems, can strike at any age and can occur as a result of almost any infection caused by a vast range of pathogens) it's likely that any ‘official’ and necessarily precise definition using a wide array of criteria would be operationally challenging to deliver at the bedside. Thus, for sepsis, we have multiple components to our definition. This chapter will describe the definitions of sepsis in non-pregnant adults, and will draw on the recommendations of the Task Force for the Third International Consensus Definitions for sepsis and septic shock (known as ‘Sepsis-3’ and published in 2016), together with operational ‘bedside’ solutions proposed by the Academy of Medical Royal Colleges in 2022. No definition is currently perfect, and we do not yet enjoy the routine adoption of any one set of criteria to prompt either a screen for sepsis or treatment for sepsis. Where it is felt it will add clarity, we make reference to now historic aspects of sepsis definitions. For example, 'Red Flag Sepsis', an operational tool put forward by the UK Sepsis Trust in 2015 and endorsed by the National Institute for Health and Care Excellence (NICE) in 2016, applied individual 'red flag' clinical criteria to a patient with a NEWS2 score of 5 or above (or who appeared unwell to a health professional) to empower clinicians to treat as sepsis. In line with the 2022 Academy of Medical Royal Colleges guidance, we have now simplified Red Flag Sepsis to empower action in patients with a NEWS2 score of 7 or higher. Precision is not always possible. From a patient’s perspective, and often that of an organisation, the difference between sepsis and a severe infection requiring hospital admission for intravenous antibiotics is somewhat semantic! Where it is felt it will add clarity, we make reference to now historic aspects of sepsis definitions. “ There follows a fairly detailed description of how we’ve arrived at where we are now: detail is included as many will have existing knowledge, and some might feel confused as to the various terms and definitions around sepsis. 13  01 NARRATIVE DEFINITIONS The 2015 NCEPOD study ‘Just Say Sepsis’ found around 80% of episodes of sepsis in the UK to occur in response to community-acquired infections. That same study also found that patients delayed accessing healthcare, often by two days or longer. For this reason, it is essential that we have a narrative definition, using accessible language, which can be used to describe sepsis to the public. In 2010 in New Jersey, the Global Sepsis Alliance penned what is now accepted by all parties as the best way to encapsulate what we know about sepsis in such communication. This definition, termed the ‘Me- rinoff definition’ after the family who sponsored the meeting, was considered by the Sepsis-3 Task Force to be the most suitable for current use: Lay definition of sepsis: the Merinoff definition ‘Sepsis is a life-threatening condition that arises when the body’s response to an infection injures its own tissues and organs.’ However, the Task Force considered it appropriate to modify this slightly for use by health professionals to reinforce the fact that sepsis is used to describe only those patients who have organ dysfunction: Professional narrative definition of Sepsis: ‘Sepsis-3’ ‘Sepsis is characterised by a life-threatening organ dysfunction due to a dysregulated host response to infection.’ Importantly, both describe sepsis not as ‘a bad infection’, but as the body’s response to infection. This is helpful in order for us and our patients to understand that antibiotics alone will not fix the problem. Septic shock is a subset of sepsis. In Sepsis-3, septic shock was redefined: Definition of septic shock: ‘Sepsis-3’ ‘Septic shock is a subset of sepsis where particularly profound circulatory, cellular and metabolic abnormalities substantially increase mortality.’ 14 02 SCREENING PROMPTS Now that we know that sepsis is characterised by organ dysfunction as a result of an infection, we need to know in which patients we should start looking for sepsis. When we consider if a patient has sepsis or not and make a decision, this process is called 'screening'. Where possible, we should record that screening has occurred. Risk factors for sepsis should always prompt a high index of suspicion for sepsis – health professionals should always ‘think sepsis’. But in a resource-constrained, busy healthcare system, this is not always 100% reliable. It is important to have a set of criteria which indicate potential acute illness or deterioration, and which in the context of infection should prompt a health professional to actively look for organ dysfunction. Of course, though patients with risk factors are more prone to developing sepsis, it is important not to rely upon risk factors alone. NICE, in NG51, also recommended the application of clinical acumen – to ‘think sepsis’ if a patient looks unwell, if they are deteriorating unexpectedly or failing to improve as expected. It is particularly important to listen to the concerns of colleagues, the patient, and their advocates, carers or family. Subtle cues such as ‘she’s not normally like this’ and ‘I’ve never seen him so unwell’ should not be ignored. Risk factors according to NG51 are detailed in the box opposite. You might have heard (or read) about qSOFA. This was a tool proposed as a screening prompt by the Sepsis-3 Task Force to aid in the identification of patients with infection who have a high risk of death (‘SOFA’ is an acronym derived from the Sequential (or Sepsis-related) Organ Failure Assessment (SOFA) score). A qSOFA was positive if the patient was found to have 2 or more of the following: Respiratory rate of 22/min or greater Altered mentation (Glasgow Coma Scale of less than 15) Systolic blood pressure of 100 mmHg or less In 2021, the most recent Surviving Sepsis Campaign academic guidelines strongly recommended against the use of qSOFA as a screening prompt for sepsis, particularly in systems already using track- and-trigger scoring systems such as the National Early Warning Score (NEWS2). So let's assume that we will no longer be using qSOFA in NHS practice! 15  RISK FACTORS FOR SEPSIS (adapted from NICE guideline [NG51], Sepsis: recognition, diagnosis and early management, 2016) The very young (under one year) and older people (over 75 years) or people who are very frail People who have impaired immune systems because of illness or drugs, including: people being treated for cancer with chemotherapy people who have impaired immune function (for example, people with diabetes, people who have had a splenectomy, or people with sickle cell disease) people taking long-term steroids people taking immunosuppressant drugs to treat non-malignant disorders such as rheumatoid arthritis people who have had surgery, or other invasive procedures, in the past 6 weeks people with any breach of skin integrity (for example, cuts, burns, blisters or skin infections) people who misuse drugs intravenously people with indwelling lines or catheters Women who are pregnant, have given birth or had a termination of pregnancy or miscarriage in the past 6 weeks are in a high-risk group for sepsis. In particular, women in this group who: have impaired immune systems because of illness or drugs have gestational diabetes or diabetes or other comorbidities have needed invasive procedures (for example, Caesarean section, forceps delivery, removal of retained products of conception) had a prolonged rupture of membranes have or have been in close contact with people with group A streptococcal infection, for example, scarlet fever have continued vaginal bleeding or an offensive vaginal discharge For neonates, risk factors include: invasive group B streptococcal infection in a previous baby maternal group B streptococcal colonisation, bacteriuria or infection in the current pregnancy premature rupture of membranes preterm birth following spontaneous labour (before 37 weeks’ gestation) suspected or confirmed rupture of membranes for more than 18 hours in a preterm birth intrapartum fever higher than 38°C, or confirmed or suspected chorioamnionitis parenteral antibiotic treatment given to the woman for confirmed or suspected invasive bacterial infection at any time during labour, or in the 24-hour periods before and after the birth (this does not refer to intrapartum antibiotic prophylaxis) suspected or confirmed infection in another baby in the case of a multiple pregnancy 16 The National Early Warning Score (NEWS2), from the Royal College of Physicians 17  NEWS2 In late 2017, the Royal College of Physicians launched the second incarnation of NEWS (the National Early Warning Score) for national roll out. As well as being useful for identifying unwell patients from all causes, NEWS2 has been found to perform well in patients with sepsis and time dependent infection. We recommend that a patient be screened for sepsis when in the context of presumed or confirmed infection: a clinician or carer is worried about their patient the NEWS2 score is 5 or more the patient is at risk of neutropenia there's evidence of organ dysfunction e.g. lactate 2mmol/l or above 03 THE NEED FOR SCREENING – CONFIRMING INFECTION SUSPECTED Using our chart or screening tool we have identified a patient who requires screening for sepsis: START SEPSIS SCREEN IF THE PATIENT: LOOKS UNWELL HAS HAD RECENT CHEMOTHERAPY HAS A LACTATE 2mmol/L OR ABOVE HAS A NEWS 2 IS 5 OR ABOVE It is important to be mindful that other things can cause deterioration. Before we move on to look for signs of organ dysfunction (and therefore ‘diagnose sepsis’), we need to confirm we are on the right track. We need to consider carefully, using clinical examination and history-taking supported (later) by investigations, whether the patient is likely to have a new underlying infection or whether we need to look for other, equally important, diagnoses. As always, the UK Sepsis Trust seeks to make such definitions usable at the bedside – often most needed at first point of contact with a health professional in hospital – we are conscious that we rarely have rapid access to a neutrophil count, thus have changed this within our tools to a patient who has recently undergone chemotherapy. 18 All that is needed is a reasonable clinical suspicion of infection, so a chesty cough with green sputum, or pain on passing offensive-smelling urine in someone who’s been feeling unwell are as good as a chest X-ray, and arguably better than a urine dipstick! Sometimes, of course, you might think a patient has an infection but have no idea (at first) where. Such a patient might clearly describe a history of fever, they might be running a high (or low) temperature, or show other signs of infection such as sweating or looking flushed. That’s fine – clinical suspicion of an infection is all that’s needed at this stage. We will discuss the importance of identifying the source and if necessary 'controlling' it later in the manual. If you’re really unsure whether this is an infective or non-infective cause of illness, it’s always best to check. Ask a senior, ensure someone orders tests such as a chest X-ray and blood tests, and revisit the diagnosis once you have more information. It’s not good practice to proceed to looking for organ dysfunction and treating with broad-spectrum antibiotics ‘just in case’, and it might lead the entire team down the path of wrongly assuming the patient has sepsis and failing to treat another condition. 04 LOOKING FOR ORGAN DYSFUNCTION: ‘DIAGNOSING’ SEPSIS, AND DETERMINING ITS SEVERITY Remember that the narrative definition of sepsis requires the patient to have one or more ‘dysfunctional’, or failing organs. We are going to consider two routes by which organ dysfunction can be identified: change in SOFA score, and the new Red Flag Sepsis (RFS) updated by recent guidance from the Academy of Royal Colleges. You need to be familiar with which method is is use in your organisation: though we expect and encourage movement toward the Academy's approach. i. SOFA score Sepsis-3 recommends the use of an increase in a patient’s Sequential (or Sepsis-related) Organ Failure As- sessment Score (SOFA) of two points (or a score of two where a patient presents for the first time and the baseline isn’t known) as the ‘official’ definition of sepsis, and it is likely that this score is the most appropri-ate measure available at present to formally identify organ dysfunction- for example, for use to identify patients for inclusion in research. 19  The SOFA score Measurement Score Measurement Score Respiratory Liver PaO2/FiO2 (mmHg) 5. If treatment is dependent on the return of blood results, this can lead to significant delays, and these patients who are at great risk of sepsis will tend to deteriorate more rapidly than their counterparts without neutropenia. Therefore, rapid assessment and escalation onto the Sepsis 6 pathway as soon as neutropenic sepsis is suspected is recommended. Febrile neutropenic patients are usually recognised and their treatment started early. Patients with non-fe- brile neutropenia will often deteriorate further before being recognised, highlighting the importance of a standardised, graduated response system to deteriorating patients even in specialist areas. 02 THE PREGNANT PATIENT Maternal mortality from sepsis varies hugely depending on access to safe and affordable healthcare. Maternal mortality remains extremely high at around 400 per 100,000 live births in low middle income countries (LMICs) as compared with developed countries, where the mortality is lower- for example 8 per 100,000 live births in the UK. This discrepancy is unacceptable, and there is evidence to suggest that ma- ternal sepsis is on the increase, with at least 50,000 women dying from sepsis each year globally. 91  Sepsis that occurs during pregnancy is termed ‘maternal sepsis’. If it develops within six weeks of delivery it is termed postpartum, or ‘puerperal’ sepsis. Sepsis is one of the leading causes of direct maternal death in the UK, and is the leading cause globally. This is partly because the immunological changes naturally oc- curring during pregnancy together with the increased exposure to healthcare, and additional risks such as with premature rupture of membranes or gestational diabetes, mean a pregnant woman is more suscep- tible to infection than her non-pregnant counterpart. The natural adaptations to the body with pregnancy may mask the signs and symptoms of infection or an acute abdomen until the woman deteriorates. Risk factors for the development of sepsis in pregnancy Sepsis can be as a direct result of the pregnancy or an indirect cause unrelated to the pregnancy, for in- stance pneumonia or a urinary infection. Following a number of maternal deaths from the H1N1 influenza pandemic, the flu vaccine is now routinely offered to pregnant women in most industrialised countries. The commonest sources for sepsis are urinary tract prenatally and genital tract postnatally. E. coli ac- counts for one third of episodes of sepsis, and infection with group A streptococcus can rapidly progress to septic shock. Due to the physiological changes in pregnancy, the National Early Warning Score (NEWS) is not designed for use in pregnant patients. Use of a modified obstetric early warning score (MEOWS) alongside the Ma- ternal Sepsis screening tool is recommended to facilitate the early recognition and escalation of deterio- rating maternal patients. The maternal sepsis screening tools are not only for use in patients who are currently pregnant, but also for those who have recently been pregnant and are within the post-partum period. A sepsis screening tool may also consider foetal distress. A foetal heart rate >160 bpm is of significant concern and is considered as an equal trigger when screening for sepsis as the woman looking sick. The Red Flags for a pregnant woman are the same as the Red Flags in their non-pregnant coun- terpart. Lactate levels should be interpreted with caution in women in active labour, as a rise is normal. Any pregnant woman with suspicion of sepsis requires an urgent senior review and multidisciplinary care. It is highly possible that the timing of delivery may need to be influ- enced by this diagnosis. Consideration should be given to- ward prophylactic treatment of the new-born if particularly at risk of neonatal sepsis, such as in women identified as having Group B Streptococci in their genital tract- some centres offer screening for this pathogen, which is the leading cause of severe infection in newborns. Specific guidance on managing sepsis in the pregnancy and the puerperium are available from the Royal College of Obstetricians and Gynaecologists’ Green Top series. 94 92 03 THE CHILD OR INFANT Sepsis is a major cause of death in the under-five population worldwide, particularly in Sub-Saharan Africa and Asia where many sepsis-related deaths are preventable. The 2020 Global Burden of Disease study found that almost half of cases of sepsis globally occur in early childhood in resource-poor countries. This group of patients is vulnerable, and they often present with atypical or vague signs and symptoms, po- tentially resulting in delayed or inappropriate treatment. You should maintain a high index of suspicion in children, and have a low threshold for admission and observation. It is important to take a detailed history and to listen to the concerns of the parent or carer as they know their child best. In young children and infants, language and understanding can be a communication barrier. You may need to take a collateral history from a parent or relative and use other means to communicate. If dis- charging a child or infant from your care, ensure that verbal and preferably written safety-netting advice has been given and that the care givers know the warning signs of sepsis and when they should seek medical help. Due to the nature of childhood illnesses, a fever can be quite common. Screening should take place for all infants and children who look unwell or are feverish, particularly with a temperature greater than 39˚C, but remembering that in those infants younger than three months a temperature of just 38˚C or more is a Red Flag. A low temperature (of 2 mmol/l. If the lactate is above 4, PICU should be involved early. Similarly, if the infant or child remains decompensated after two initial 20ml/kg fluid boluses (or 10ml/kg in neonates), Critical Care advice regarding inotropic support should be sought – usually using dopamine or epineph- rine. The fluid of choice is usually sodium chloride 0.9% for the initial 10ml/kg boluses in paediatrics, although if blood is being lost this will also need replacing. Whilst the management of the neonate with sepsis is beyond the scope of this book, special consideration is needed for this group due to their immature immune systems making them more susceptible to infec- tion – particularly respiratory, urinary or line-based infections. Any underlying disease process or low birth weight contributes to a higher mortality; any suspected or proven infection in the mother during the third trimester will often indicate a need for prophylactic treatment in the new-born as well as in twin pregnan- cies where one twin develops an infection or sepsis shortly after birth. If a previous child developed an invasive strep B infection, this also puts a subsequent newborn at greater risk of developing the same. Management of intravenous access, fluids and antibiotics in these patients is a specialised field. Metabolic changes including lactic acidosis and increased glucose requirements are recognised early responses to sepsis in the neonate, and other differences such as a depletion in vitamin B compounds and glutamine have also been noted. More research in these areas is required. 94 SUMMARY The treatment principles for patients with sepsis are identical regardless of the cause. Initial assessment and resuscitation should follow the ABCDE format with the application of the appropriate Sepsis Screening Tool. Patients should be managed using the Sepsis 6 approach. Liaison with Critical Care should be timely, particularly in the presence of septic shock or multi-organ failure. Patients with pneumonia represent the largest group of patients with sepsis. Common causes of sepsis aside from pneumonia include gastrointestinal pathology, urinary tract, biliary tract and skin infections. Sources will vary in the pregnant patient. Remember to keep an open mind when assessing a patient presenting with sepsis. The importance of consultation with microbiologists locally who will be aware of pathogens and resistance patterns in your own institutions cannot be over emphasised. Most organisations now have their recommended first-line empiric treatments for common infections on their intranet sites. 95 MICROBIOLOGY The management of infections includes multiple facets, but in essence centres around identification of the pathogen, control of any source of infection including judicious use of antimicrobials where necessary, and management of any sequelae of infection including sepsis. Source control is, therefore, an essential part of managing sepsis where this is practicable. In this context, source control means physical removal of the source, such as drainage of abdominal collections, removal of invasive lines or surgical removal of infected tissue. Source control will also include re-estab- lishing flow of fluid which has become obstructed – for example relief of biliary or urinary obstruction. To achieve effective and rapid source control may therefore demand close liaison with colleagues in surgery and radiology. However, for some conditions (such as pneumonia) where there is neither a collection of infected material amenable to drainage nor a presence of prosthetic material which can be removed, source control is not possible. Here, antimicrobial therapy, usually considered as an adjunct to source control, becomes the only way of controlling the trigger for sepsis. Initial antibiotic choice (assuming, as in the majority of cases, the likely pathogen is bacterial) is usually based on the suspected focus of infection, determined through clinical suspicion supported by radiolog- ical and microbiological evidence. It is vital that the right antimicrobials are given to control the infection and fight the organisms present, and this will often mean initially using broad spectrum ‘best guess’ agents with a later focus on a narrower spectrum when (if) the organism becomes known (known as ‘Start smart then focus’). Organisms take a while to grow; therefore taking the right sample and sending it in the right container as soon as possible following the diagnosis of sepsis can help to identify the likely patho- gens in a timely fashion. Taking samples A sterile technique should be adopted when taking samples for microbiological investigation. Because a plethora of microorganisms are ubiquitous within our environment, they can easily contaminate samples, resulting in the predominant organism isolated from a culture being an environmental contaminant rath- er than a true pathogen. In the case of blood cultures, commensal skin flora can be picked up instead of true pathogens. If not correctly interpreted, such false positives can result in inappropriate antimicrobial prescribing, which could leave a patient undertreated, and/or put them at risk of acquisition of a multid- rug resistant organism or C. difficile infection. Advice on taking appropriate specimens can usually be obtained from local microbiology or infectious diseases teams. Once a specimen has been taken, it must be placed in a container that maintains viability of any pathogenic microorganisms during transit. For example, formalin kills organisms, so placing any samples in formalin-containing specimen pots is unlikely to yield any pathogens. Likewise, some viruses are easily inactivated by detergents (for example the influenza virus), swab sticks or the transport material (e.g. gel, activated charcoal) contained within wound swab containers. Using the right container for the suspected organism is essential. Urgent samples Many laboratories operate an on-call system for urgent microbiological specimens (e.g. tissues taken in theatre, CSF) – if any samples are urgent the laboratory team must be called to come in and process the samples. This is usually not necessary for blood cultures, but it is always better to check with your local laboratory. Many laboratories don’t place blood cultures in the incubator immediately if received over- night – we would encourage this practice to change. 97  Labelling of samples All samples must be labelled correctly to avoid rejection once they reach the laboratory. There are UK standards for specimens, required to ensure that the sample has come from the patient stated on the request form and to ensure traceability back to the requestor and/or person taking the sample in case of queries. The minimum information required on the sample container is: Patient’s forename and surname Location (i.e. ward or department) Date and time of sampling Type of specimen The request form must also contain the same information as well as: Important clinical findings e.g. prosthetic heart valve in situ, known infectious condition such as TB or HIV Working diagnosis e.g. pneumonia Travel history over past 12 months Printed (legible) name and registration number of both requestor and person taking sample Recent and current antimicrobial therapy 98 These details are important, as different organisms tend to affect and infect different parts of the body, and require different conditions in order to grow. Prior knowledge of the type of infection suspected and the site affected helps the laboratory scientists and clinicians to determine which type of organisms they need to look out for, which can mean using different types of agar plates and techniques such as molecu- lar PCR (polymerase chain reaction) and serological tests. Recent antimicrobial therapy can affect the ability to detect organisms which are susceptible to the antibiotics given, though modern blood culture media attempt to bind antibiotics present to reduce their masking effect. Sepsis can occur in patients that have travelled abroad; common examples of travel associated infections that can cause sepsis include malaria, TB and typhoid fever. Some of these infections are extremely infec- tious e.g. typhoid (due to Salmonella typhi and Salmonella paratyphi), TB, viral haemorrhagic fevers (e.g. Ebola) and MERS-CoV infections. These can result in outbreaks, so it is always wise to consult local policies and liaise with your infection specialists – the microbiologist or infectious diseases clinician. Providing a travel history not only allows the infection specialist to advise on the most appropriate tests, it can also help to prevent ongoing transmission of infection to laboratory staff, other patients and other staff, in- cluding yourself. Identifying pathogens Isolating pathogens Methods of identifying pathogens depend on what is trying to be identified. Most bacterial organisms will grow on standard culture media. Anaerobes, fastidious bacteria, slow growing organisms and most myco- bacteria (including TB) require different media and special growing conditions such as temperature and atmospheric conditions. Other organisms are so difficult to grow they need to be sent away to a specialist laboratory for testing, or require that alternative techniques such as serology and PCR be used. Often this can add to the time taken to identify an organism, hence the importance of clinical suspicion of the source of infection and likely pathogens affecting the patient being conveyed to the infection teams. If the incubator containing a blood culture sample detects the production of carbon dioxide by bacteria, it will flag the sample as positive. Specimens are then processed and usually spread onto agar plates, or put into special liquid media for harder to grow organisms. When organisms grow, they appear as “colonies” on the plate, which can then be tested for identification and antimicrobial susceptibility testing. Sometimes there is more than one organism present, in which case the individual organisms need to be “picked off” and cultured again to ensure that there is a pure growth of organism. This helps to ensure that we do not get false results with regards to identification and antimicrobial susceptibility testing, which can lead to inappropriate antibiotic treatment. Mixed E. coli (pink colonies) and Klebsiella pneumoniae Pure growth of Pseudomonas putida on a blood agar plate (blue colonies) on a chromogenic agar plate 99  Blood cultures Identification of microorganisms that are causing a blood stream infection is made by incubating blood taken from an affected person and incubated. In most institutions this is done using automated contin- uous monitoring incubation systems. The blood culture bottles contain mixtures of a culture medium, an anticoagulant and resin or charcoal mixtures to reduce the effects of antimicrobial agents and oth- er toxic compounds. In adults, there are two bottles in a blood culture “set”; an aerobic bottle and an anaerobic one. To optimise recovery of microorganisms an adequate volume of blood is required; this is approximately 8-10 ml blood per bottle. This volume helps optimise recovery of microorganisms from the blood even when there are very low numbers of organism (