Human factors is an evidence-based scientific discipline used in safety critical industries to improve safety and worker well-being. The implementation of human factors strategies in anaesthesia has the potential to reduce the reliance on exceptional personal and team performance to provide safe and high-quality patient care. To encourage the adoption of human factors science in anaesthesia, the Difficult Airway Society and the Association of Anaesthetists established a Working Party, including anaesthetists and operating theatre team members with human factors expertise and/or interest, plus a human factors scientist, an industrial psychologist and an experimental psychologist/implementation scientist. A three-stage Delphi process was used to formulate a set of 12 recommendations: these are described using a 'hierarchy of controls' model and classified into design, barriers, mitigations and education and training strategies. Although most anaesthetic knowledge of human factors concerns non-technical skills, such as teamwork and communication, human factors is a broad-based scientific discipline with many other additional aspects that are just as important. Indeed, the human factors strategies most likely to have the greatest impact are those related to the design of safe working environments, equipment and systems. While our recommendations are primarily provided for anaesthetists and the teams they work with, there are likely to be lessons for others working in healthcare beyond the speciality of anaesthesia.

Healthcare relies on high levels of human performance, as described by the 'human as the hero' concept. However, human performance varies and is recognised to fall in high-pressure situations, meaning that it is not a reliable method of ensuring safety. Other safety-critical industries embed human factors principles into all aspects of their organisations to improve safety and reduce reliance on exceptional human performance; there is potential to do the same in anaesthesia. Human factors is a broad-based scientific discipline which aims to make it as easy as possible for workers to do things correctly. The human factors strategies most likely to be effective are those which 'design out' the chance of an error or adverse event occurring. When errors or adverse events do happen, barriers are in place to trap them and reduce the risk of progression to patient and/or worker harm. If errors or adverse events are not trapped by these barriers, mitigations are in place to minimise the consequences. Non-technical skills form an important part of human factors barriers and mitigation strategies and include: situation awareness; decision-making; task management; and team working. Human factors principles are not a substitute for proper investment and appropriate staffing levels. Although applying human factors science has the potential to save money in the long term, its proper implementation may require investment before reward can be reaped. This narrative review describes what is known about human factors in anaesthesia to date.

Peri-operative medication safety is complex. Avoidance of medication errors is both system- and practitioner-based, and many departments within the hospital contribute to safe and effective systems. For the individual anaesthetist, drawing up, labelling and then the correct administration of medications are key components in a patient's peri-operative journey. These guidelines aim to provide pragmatic safety steps for the practitioner and other individuals within the operative environment, as well as short- to long-term goals for development of a collaborative approach to reducing errors. The aim is that they will be used as a basis for instilling good practice.

Contemporary guidance takes a patient-centred approach and recommends discussing and planning treatments that should be considered, not just those that should be withheld. Although some organisations and communities still use specific DNACPR (do not attempt cardiopulmonary resuscitation) forms to recommend that cardiopulmonary resuscitation is not attempted, this approach has been shown to have disadvantages and is no longer regarded as best practice. The following guidelines have been produced in response to this change. They are designed to help anaesthetists, as part of the wider healthcare team, to implement and respond to advance care planning documents before and during procedures. The guidelines apply to all procedures, however minor and low risk they are considered to be, and the same ethical and legal principles apply to procedures carried out under local or regional anaesthesia and/or conscious sedation, as well as to those under general anaesthesia.

References 1. Gammell C. War veteran died after receiving wrong drug in operation. The Telegraph 2008. https://www.telegraph.co.uk/ news/uknews/2075206/War-veteran-died-after-receiving-wrongdrug-in-operation.html (accessed02/12/2021). 2. Toft B. External Inquiry into the adverse incident that occurred at Queen’s Medical Centre, Nottingham. 2001. https://www.who. int/patientsafety/news/Queens%20Medical%20Centre% 20report%20(Toft).pdf (accessed 02/12/2021). 3. National Patient Safety Agency. Safer spinal (intrathecal), epidural and regional devices. 2009. https://www.oaa-anaes.ac.uk/assets/ _managed/cms/files/NPSA/2009%20NPSA%20_Safer-spinal% 20supporting%20info_10215.pdf (accessed 02/12/2021).

References 1. Klein AA, Meek T, Allcock E, et al. Recommendations for standards of monitoring during anaesthesia and recovery 2021. Anaesthesia 2021;76: 1212–23. 2. Audrey De Jong A, Pardo E, Rolle A, Bodin-Lario S, Pouzeratte Y, Jaber S. Airway management for COVID-19: a move towards universal videolaryngoscope?. Lancet Respiratory Medicine 2020;8: 555. 3. Sullivan EH, Gibson LE, Berra L, et al. In-hospital airway management of COVID-19 patients.Critical Care 2020;24: 292.

This guideline updates and replaces the 5th edition of the Standards of Monitoring published in 2015. The aim of this document is to provide guidance on the minimum standards for monitoring of any patient undergoing anaesthesia or sedation under the care of an anaesthetist. The recommendations are primarily aimed at anaesthetists practising in the UK and Ireland, but it is recognised that these guidelines may also be of use in other areas of the world. Minimum standards for monitoring patients during anaesthesia and in the recovery phase are included. There is also guidance on monitoring patients undergoing sedation and during transfer. There are new sections specifically discussing capnography, sedation and regional anaesthesia. In addition, the indications for processed electroencephalogram and neuromuscular monitoring have been updated.

1Consultant, Department of Anaesthesia, Guy’s and St. Thomas’NHS Foundation Trust, London, UK, Council Member, Association of Anaesthetists andCo-Chair of theWorking Party 2Consultant, Department of Anaesthesia, University Hospitals of Coventry andWarwickshire, Coventry, UK, Difficult Airway Society representative andCo-Chair of theWorking Party 3Consultant andHonoraryClinical Senior Lecturer, Imperial College London, London, UK 4Chair, TheAssociation for Anaesthetic and Respiratory Device Suppliers, Bromley, UK 5 Specialist Trainee, South East Scotland School of Anaesthesia, UK andMember of theAssociation of Anaesthetists TrainingCommittee 6 Ergonomics expert, Link Ergonomics, Nottingham, UK 7Consultant, Department of Anaesthesia, Guy’s and St. Thomas’NHS Foundation Trust, London, UK and President of Regional Anaesthesia (RA), UK 8Consultant, Department of Anaesthesia, Imperial CollegeHealthcareNHS Trust, London, UK, andCouncilMember, Royal College of Anaesthetists 9 Consultant inMedical Devices, Edinburgh, UK

We thank Pollard et al. for observations on our letter 1 and agree with the Health Service Investigation Branch (HSIB) concerns about practice variability in stop before you block. The HSIB chose as an index case an example of wrong side block from a Trust that used a sticker, but where this was not applied 2. The HSIB might equally have used an index case where a sticker was applied on the wrong side, which led to wrong-side block. Pollard et al. readily concede that stickers have been ineffective in their experience, so it is unclear if they are defending this practice. Additional marks/stickers have caused wrong side blocks and are contrary to other guidance cited, which states the only mark should be the surgical site mark 3. Pandit et al. have speculated (before the most recent discussions, and to embrace a wide range of views) about the use of a sign or a sticker, but, crucially, only if these are applied within the sterile field, after skin prep and immediately (just seconds) before needle injection 4. The alternative described by Pollard et al. involves non-sterile stickers applied long (many minutes) before injection. This cannot serve as a stop moment, but rather only as a reminder. As Hopping et al. discovered, it is not just a case of what forces the stop moment, but when this is applied 5. Pollard et al. refer to some international literature, but since practices differ, extrapolation is difficult, especially as Never Events are unique to NHS England 6. The radiology and other society guidance referred to concerns situations where the block is the sole intervention. That the correspondents extrapolate this to blocks for surgery represents another source of confusion. The HSIB has now asked the Safe Anaesthesia Liaison Group to review the guidance to resolve inconsistencies. We cannot prejudge the outcome of that review, but checklist fatigue will be one consideration 7. Mock Before You Block (a personal behavioural barrier) or ultrasound screens (aides memoire) are harmless, at least 1. It is more difficult to identify those measures that are both harmless and proven effective 8.

We are grateful for the opportunity to reply to Dr White's letter about the genotoxic effects of sleep deprivation. The profession's interest in the deleterious effects of fatigue on clinicians has been stirred by several high-profile tragic deaths of young doctors on the way home from a night shift. The ‘Fight Fatigue’ campaign organised by the Association of Anaesthetists, the Royal College of Anaesthetists and the Faculty of Intensive Care Medicine is prompting a national conversation about the impact of fatigue and shift working. The campaign aims to raise awareness, change attitudes and improve the working environment and has already garnered prominent backers in the medico-political world and from parliamentarians in England, Scotland, Wales and Ireland. We have asked front-line doctors in anaesthesia and intensive care about the causes and impact of work-related fatigue on their lives in three separate online surveys targeting trainees, Specialty and Associate Specialist (SAS) doctors and consultants. The responses are adding to our understanding of the problem. The results of the trainee survey have already been published 1 and we hope to publish the results from the consultant survey in the near future with the SAS doctor survey following later. The Association of Anaesthetists has created an online resource at www.aagbi.org/letsfightfatigue, containing up-to-date information on fatigue and practical advice on simple steps that can mitigate its effects. Standards for rest facilities and rest culture offer individuals and organisations alike opportunity to review current practices and thinking helping to identify areas for improvement. As we wait for the scientific case for the adverse effects of sleep deprivation and potential pharmacological solutions to build, the Association of Anaesthetists will of course continue to support its members by developing its wellbeing resources, adding to the low-tech integrative self-care approaches already highlighted.

I agree with Pandit et al. that putting an additional mark on a patient presenting for elective surgery, who already has a surgical site mark, is unlikely to improve safety and could even lead to wrong-site surgery 1. However, I think that there are certain circumstances in which marking the block site should be encouraged, particularly if nerve catheters are inserted, and invite the authors to comment on the role of ‘anaesthetic’ block site marking in these. Some trauma patients, for example, require a block but are unlikely to require subsequent surgery, such as those presenting with fractured ribs, for whom paravertebral catheters may be inserted. In these cases, ‘anaesthetic’ marking of the block site is mandatory to comply with the ‘stop before you block’ process, involving verification of the site of any pain, the radiological findings or report and emergency department or parent team documentation. Secondly, some trauma patients may require a block before surgical site marking has occurred for subsequent surgery, such as those presenting with fractured neck of femur, for whom fascia iliaca catheters may be inserted. Again, ‘anaesthetic’ marking of the block site may be required. Finally, surgical drapes may cover any surgical marks during nerve block catheter insertion 2, necessitating further ‘anaesthetic’ marking of the block site to reduce the risk of wrong-site block.

A ‘stop’ moment should be undertaken before performing a local anaesthetic or regional block to prevent wrong-side block. This guidance was produced by the Safe Anaesthesia Liaison Group (SALG) and then included in the National Safety Standards for Invasive Procedures (NatSSIPs) 1. There are several measures proposed to force the ‘stop’ moment, from checklists to signs on ultrasound machines 2, or the novel ‘mock before you block’ technique 3, 4. In this context, we write to express concern about the recent correspondence that advocates either making additional ‘anaesthetic’ marks 5, or placing anaesthetic labels 6 on the side to be blocked, as a means of forcing the ‘stop’ moment. The NHS instructions on skin marking for surgery can be found in the same NatSSIPs document 7 and in a previous NPSA document 8 and have been further explained 9. The National Patient Safety Agency Patient Safety Alert ‘World Health Organization (WHO) surgical safety checklist’ (2009) determines “that the overall responsibility for the site marking for regional blocks lies with the operating surgeon. The anaesthetist should only proceed with a regional block after confirming that the site for surgery has been marked” 10. Therefore, the national consensus is that the only mark on the patient should be a surgical mark, made as an arrow near the incision site. On behalf of the national organisations representing patient safety in anaesthesia, we believe that the use of any additional marks, including labels, is contrary to the NHS guidance. Moreover, use of such additional marks has the potential to cause error, including perhaps wrong-side surgery if the label is not removed. Indeed, we are now aware of several incidents of wrong-side block emanating from Trusts where such sticker label policies are in force. It is possible that when the label is accidently placed on the side opposite to be blocked, an ‘invisible gorilla’ effect can occur, wherein the original surgical site mark becomes invisible as the practitioner then concentrates solely on the (misplaced) sticker. We encourage all anaesthetists to follow the national guidance on operative site marking, and refrain from introducing local policies relying on additional marks or labels to the surgical side. There are many proposed interventions such as signs on ultrasound machines 2, ‘mock before you block’ 3, 4 or electronic alerts 11 that do not contravene the guidance. We remind colleagues that the only mark that should be used is the surgical site mark, and this should be the sole focus of attention when identifying the side to be blocked.

The celebrated Stonehenge scene in Rob Reiner’s classic 1984 spoof rock band documentary is an apt portrayal of what happens when an end-user’s expectation differs from the manufacturer’s delivery of a standard (and those uninitiated should head immediately to their nearest streaming site for illumination). In this issue of Anaesthesia, Thomas et al. assess the performance of the adjustable pressure limiting (APL) valves of two widely-used anaesthesia workstations from different manufacturers [1]. In short, the researchers found that the APL valves were not precise (in terms of circuit pressure generated versus dialled in pressure), and that the valves performed very differently from each other. Their conclusion is especially critical of one of the APL valves’ performance characteristics: it increased system pressure more-orless linearly, but only after it had been turned approximately 60 degrees; before that, it generated no increase in pressure. The other valve generated a pressure that increased more or less linearly beginning from zero degrees of turn. The response from the manufacturer of the criticised valve, as well as being critical of the study methodology, is adamant that the valve performs within the relevant international standard [2]. So who is right? Perhaps inevitably, both parties are ‘right’. The characteristics of APL valves (and indeed all parts of the anaesthesia workstation) are contained in the 117-page standard ISO 80601-2-13:2011, ‘Particular requirements for basic safety and essential performance of an anaesthetic workstation’ [3]. The exact wording is as follows:

Understanding the working principles of anaesthetic equipment and drugs helps us compensate when they don't work as expected. It is presumed to be a foundation of safety, and has therefore become a fundamental part of the postgraduate syllabus. Yet, there are exceptions. The most baffling, perhaps, are the very notions of consciousness and its loss 1. But that, practical clinicians would say, is philosophy. So, how about lipid emulsion as an antidote to local anaesthetic systemic toxicity (LAST)? Nothing is philosophical about fat – and, at first glance, its use might seem amenable to hard empirical graft. Yet, life-threatening LAST is unpredictable, potentially lethal and rare. This makes it impossible to study prospectively and ethically in humans. And so, in 2007, when the AAGBI first recommended the use of lipid in LAST - as a last resort- its mechanism had not been studied in our species. Indeed, it wasn't clear to what degree it worked at all. It just seemed not to do harm, and to have saved some lives in extremis 2, 3. Since then, two putative mechanisms have attracted most speculation. The first has a confusing array of labels: ‘physical’; ‘partitioning’; ‘pharmacokinetic’; and ‘the lipid sink’- they all suggest that the hydrophobic local anaesthetic settles in the suddenly increased lipid phase of the circulating volume, and the concentration in the aqueous phase falls safely away. The second possible set of mechanisms is more biological, or pharmacodynamic: lipid may have direct effects on pertinent tissues, above all the heart and brain. Studied in isolation, in the simplest of models, evidence has accrued for both mechanisms. For example, lipophilic drugs such as bupivacaine will indeed partition into lipid added to a phial of plasma, and work with dyes provides visually striking support for this mechanism 4-6. Similarly, isolated intoxicated hearts are given a fillip when lipid is added to their perfusate 7. But, of course, the price of such simplicity in the laboratory is potential clinical irrelevance. An intoxicated patient dying fast is a far cry from a dangling heart in Langendorff preparation. So, investigators have tackled more complex models, making different compromises. In some work, toxic doses have been studied in animal models; in other work, smaller doses have been studied in humans 8-13. There has also been more work done in vitro and also, more recently, in silico- that is, with computer modelling 14. Over time, lipid's action as an antidote has become somewhat clearer. One group of studies on whole animals is particularly instructive. In the first work on lipid's beneficial effect in LAST, rats were studied 8. From there, investigators scaled up to dogs and rabbits 9, 11. Again, lipid seemed beneficial in LAST. But work on lipid in LAST conducted in pigs has been equivocal 10, 12. These mixed results have sustained sceptics. In the UK, lipid's contribution to resuscitation in LAST is currently accepted. But in Finland, for example, there is no national guideline, and protocols for treatment of LAST include lipid emulsion in only 47% of hospitals 15. It transpires that pigs have an idiosyncratic reaction to lipid emulsion infusion: increases in systolic and pulmonary artery pressures correlate with dose, while heart rate falls. Exactly what mediates the reaction is unclear, though thromboxane is implicated 16. But whatever the reason, the reaction makes pigs a poor model for any studies of lipid as an antidote, and provides a salutary lesson in the dangers of presuming one species approximates another. Researchers studying sub-toxic doses of local anaesthetic in humans have taken another tack. Professor Rosenberg's group, for example, studied healthy male volunteers who were administered a low dose of bupivacaine, and then either saline or lipid 13. Blood samples taken five minutes later yielded fascinating results. If the lipid functions as a sink, it should absorb lipophilic bupivacaine, and this in turn should reduce the concentration of bupivacaine in the aqueous phase of plasma. Amongst Professor Rosenberg's volunteers, those given lipid should have had, in the aqueous phase of their plasma, lower concentrations of bupivacaine. And yet they didn't. Between the two groups, there was little difference five minutes after lipid administration. The only discrepancy will be relevant in a moment: the total bupivacaine concentration in circulating blood decayed faster in the volunteers who got lipid; there was a statistically significant difference 20 and 30 minutes after the lipid was given. So, is the pharmacokinetic hypothesis debunked? Or, if you'll forgive it, is the sink sunk? In short, the answer is ‘yes’. But the damage hasn't been done only by Rosenberg et al.; Professor Weinberg's group has also contributed 14. Looking at rats, Weinberg's group zoomed in on the first minutes after lipid was administered. In their model, action was very fast; five minutes after the lipid had been given as an antidote to intoxication by local anaesthetic, much of relevance was over. In the first minutes after intoxicated rats were given lipid, the antidote made a substantial difference to the distribution of the local anaesthetic through the rats’ bodies; after two minutes, the rats who got lipid had more bupivacaine in their blood, but less in their cardiac tissue and less in their cerebellum. Bupivacaine also washed faster out of the lungs and kidneys of rats treated with lipid. In other words, lipid didn't seem to act as a lipid sink; instead, lipid appeared to serve first as a vehicle. It appeared to scavenge local anaesthetic from the heart, brain and other well-perfused organs, carrying it to other more poorly perfused tissues. Five minutes after lipid had been given to the rats, it scarcely affected the concentration of bupivacaine in whole blood. All this is consistent with Rosenberg et al's results: no evidence for a long-lasting sink, but acceleration of the clearance of local anaesthetic. Weinberg et al's results also add detail to a pharmacodynamic effect. They imply that lipid has little effect on cardiac function while concentrations of local anaesthetic are high. But as the lipid vehicle relieves cardiac tissue of its local anaesthetic burden, then, as a threshold is passed, the lipid begins to boost cardiac output. So the whole process may be summarised thus: lipid carries local anaesthetic from heart and brain to less well-perfused organs, and then, as the concentration of local anaesthetic falls far enough, lipid also acts as a tonic to the depressed hear. One reasonable response to all this is to ask: ‘does it matter’? Three strands braid together to form the predictable answer: ‘yes, with reservations’. First, a better understanding of lipid's action as an antidote will guide design of therapeutic emulsions. After all, Intralipid® has been widely used and studied largely because, in many countries, it is more familiar than alternatives as a foundation for total parenteral nutrition. However, it may not be optimal as an antidote. In the past, researchers have compared different emulsions to identify those which bind more local anaesthetic when mixed up with local anaesthetic and something like plasma 4, 6. Others have gone further to engineer pegylated liposomes, all intended especially to absorb toxins 17. Tinkering further to acidify the liposomes’ interior significantly boosts the nano-carriers’ capacity to capture local anaesthetic 18. But, now it seems that such absorption into circulating lipid may not be what the intoxicated patient needs. Just as haemoglobin is an excellent transport molecule because it not only binds blood gases, but also releases them appropriately, so the affinity of lipid for local anaesthetic may not be the ultimate test. What may matter more is how effective a shuttle an antidote is. Second, the armamentarium of the resuscitator may change. In the latest UK Resuscitation Council Guidelines, epinephrine and amiodarone hold their places, confounding the winds of change, despite a remarkably shallow evidence base 19. Resuscitation in LAST was notoriously difficult before the use of lipid, and nothing has changed in this regard. But it might: levosimendan has its proponents, for example 20. Used in LAST, it may simply complement lipid emulsion;(the sole published study was underpowered 21). But, in the more distant future, some drug may so surpass lipid as to render it redundant. Third, LAST's incidence may be falling, though the causes of this welcome development are unclear. Changes in training, staffing and in the presentation and storage of local anaesthetic may have contributed. Changes in anaesthetic technique may also have played a part. Recognition of the danger of single-shot epidural injections of large volumes of local anaesthetic led to wider adoption of catheters, incremental and test doses. In addition, wider use of ultrasound to guide perineural and fascial blocks may be leading to smaller doses of local anaesthetic, and above all, to operators moving the needle as the dose is delivered 22. A moving needle may cause more physical trauma, but intravascular injections should be smaller. Some have suggested that the danger of LAST is now more ghost than bogeyman 23. Local anaesthetic systemic toxicity lies between the two extremes: it still happens, but not so frequently as to be familiar. So it behoves us to beware, and to incur the small costs of remaining ready to treat it. Complacency is dangerous to all; at least in some settings, the risk of being sued for malpractice rises with the anaesthetist's age. Perhaps the delusion of safety and supranormal ability gathers with success through time, until the very worst occurs 24. No external funding or conflicts of interest declared.

Purpose of review Early warning scores, early warning systems and rapid response systems, were established in 1999. In the UK, a National Early Warning Score was launched in 2013 and is now used throughout the National Health Service. In 2007, a firm recommendation was made by the maternal confidential death enquiry that maternity units should incorporate a modified early obstetric warning score chart into clinical practice. Although there was enthusiastic uptake of this recommendation, local recording systems vary throughout the country and there is now a need to revisit revise and standardize an obstetric early warning system (ObsEWS). Recent project The intercollegiate Maternal Critical Care group of the Obstetric Anaesthetists’ Association have produced an ObsEWS in line with the aggregate UK National Early Warning Score. Six physiological parameters are incorporated: respiratory rate, oxygen saturations, temperature, systolic blood pressure, diastolic blood pressure, and pulse rate. However, robust physiological thresholds for the measured parameters are currently lacking but required for a more sensitive and specific ObsEWS. Summary A greater focus and study on the management of maternal morbidity (in addition to mortality data) and the development of better systems within and across the multidisciplinary team to detect early deterioration should improve management of serious illness in obstetrics. It is imperative that we undertake robust ObsEWS and data collection, including electronic systems with research and evidence-based recommendations to underpin this system. This should improve patient safety and result in more efficient, cost-effective management of sicker patients in our complex modern healthcare systems.