Nationalauditprojects.org.uk

Drug errors and awake paralysis Jonathan H Mackay headline
13.1. This chapter describes cases of brief awake paralysis reported to NAP5, as a result of drug errors that led to a
neuromuscular drug being administered without prior anaesthesia. Although it can be argued that these cases are not technically ‘accidental awareness during general anaesthesia' the experiences and consequences for the patient are similar to AAGA. NAP5 received reports of 17 such cases. It is notable that the distress during the patient experiences and the subsequent psychological distress was of a greater severity than all other cases of AAGA.
Background
13.2 An early landmark study of human error as a cause
odds, it is almost inevitable that an anaesthetist of untoward anaesthetic outcomes was published will make drug errors during their career; yet many by Cooper et al. (1978). At that time, a syringe practitioners remain in a state of denial that they swap or the unintended administration of an could make such an error, preferring to believe that incorrect drug was the third most common cause of they are less fallible than their colleagues (Evley et anaesthetic critical incident (human error involving disconnection of circuit or inadvertent changes 13.4 Many errors are due to slips or lapses in in gas flow being the two commonest). Syringe concentration that occur in the multi-tasked setting swaps now account for an even higher proportion in which anaesthetists work. It is important that the of critical incidents in anaesthetic practice because, broader environment in which anaesthetists work, is over time, the latter two have been virtually not forgotten as a source of contributory factors to eliminated. Osborne et al. (2005) reported that of drug error: the likelihood of a final slip or lapse may 4,000 reports received by the Australian Incident be increased by many ‘latent factors' (see Chapter Monitoring Study (AIMS), there were almost as 23, Human Factors). At an individual level, haste, many cases of awake paralysis due to syringe swaps inattention and distraction are likely to increase as awareness during anaesthesia. the risk of drug errors. The practice of anaesthesia 13.3 More recent incident reporting studies suggest involves continuous vigilance, and that may be a rate of drug error of 1 every 140 anaesthetics impaired by the effects of fatigue. (Webster et al., 2001; Zhang et al., 2013). This 13.5 Reason's classic ‘Swiss cheese model' of human is almost certainly an underestimate as many error in medical care explains that the coincidental unrecognized errors are not reported. Given these lining up of ‘holes' or faults in the protective Report and findings of the 5th National Audit Project
Drug errors and awake paralysis
barriers in the environment is what allows errors to have been trialled with evidence of modest benefit manifest as patient harm (Reason, 2000). in reducing drug error but no clear evidence of patient benefit (Merry et al., 2011).
13.6 During syringe swaps patients are likely to have the distressing experience of total paralysis (perhaps 13.10 At the time of writing neither two-person checking including painful fasciculations with suxamethonium) nor scanning-based systems are in routine use, nor in the absence of any anaesthetic agent that reduces widely recommended in anaesthetic practice.
consciousness. Patient experiences include awake- 13.11 While litigation as a result of drug errors causing paralysis, distress, fear of dying and that paralysis permanent harm in anaesthesia appears rare, may be permanent. PTSD may follow. As both drug errors from syringe swaps leading to awake a feeling of paralysis and distress at the time of paralysis is prominent in these claims (Mihai, et al. awareness are associated with worse psychological 2009). Such claims are almost invariably judged to sequelae after AAGA (Samuelsson et al., 2007, represent sub-standard care and litigation is almost Ghonheim, 2009) it is not surprising that these invariably successful (see Chapter 22, Medicolegal).
cases are associated with a high rate of severe 13.12 There are several separate problems: psychological sequelae (Mihai et al., 2009).
(a) a syringe swap occurs when a drug error 13.7 There have been several solutions suggested to occurs because drug from the wrong syringe is reduce the incidence of drug errors in anaesthesia and other branches of medicine. These include (b) a drug labelling error occurs when the contents checking drugs with another person before of the syringe are different to that indicated on administration (‘two person checking' or ‘double the label, either because drug was drawn up checking') and also the use of technology (bar from the wrong ampoule or the wrong label was code scanning). Both systems have been trialled in anaesthesia (Evley et al., 2010). (c) a drug omission occurs when the intended drug 13.8 While double checking of drugs has an appeal, is omitted due to failure to draw up a drug in a there are recognised problems with it as a solution. To be effective, the double-checking must include all phases of drug administration (drawing up, drug selection and drug administration). A recent UK naP5 case review and
study found it was impractical due to the inability to ensure two individuals were present whenever drug administration was required (Evley et al., 2010). The 13.13 We used Class G of the AAGA reporting system evidence from other areas of medicine that double as a miscellaneous category. This rapidly filled checking reduces drug error is limited. A systematic predominantly with syringe swaps and drug error; of review identified a single RCT which reported that which there were 17. These cases equal 1 in 8 of all it reduced ward-based drug error from 2.98 to 2.12 definite and probable cases reported to NAP5. The per 1000 drug administrations (one error prevented 17 UK cases comprised 11 syringe swaps, five drug in every 1,162 drug administrations, Alsulami et al., labelling errors and one omission error. There was 2012), which the authors described as of ‘unclear suspicion of omission errors (either no drug given, clinical advantage'. Toft has described ‘involuntary or partial mixing) in several other cases not included automaticity' as an explanation of why double (or here. Fifteen of 17 drug errors occurred at induction of even multiple) checking may still enable errors general anaesthesia; two occurred due to accidental to occur (Toft & Mascie-Taylor, 2005). There is a injection of neuromuscular blocking drug or local tendency to ‘see what you expect to see' and while anaesthetic during intended regional anaesthesia. there may be mechanisms to reduce its effect it 13.14 Thus, three difficult-to-classify cases originally in may not be entirely avoidable. this class are not considered further in this chapter. 13.9 Bar-code scanning also appears to be a reliable One was an awareness of inhalational induction in solution but previous studies have identified many a child; one was awareness of cricoid pressure and shortcomings with currently available systems and one was likely partial paralysis in recovery.
there are important cost implications (Evley et al., 13.15 The demographic characteristics of the patients in 2010). In order to prevent scanning of a syringe that this group were similar to the patients in the Activity in fact contains the wrong drug, bar-code scanning Survey: median age 36-40, median weight 70kg, systems for drugs ideally need to be combined with median BMI 26kg/m2, and this suggested that all systematic use of pre-filled syringes. Such systems types of patient were susceptibleto syringe swaps.
NAP5 Report and findings of the 5th National Audit Project
Drug errors and awake paralysis
13.16 Most cases were ASA 1 or 2, and most events 65%) experienced distress at the time of the event. occurred during daytime hours. Thus it did not Distress was more common during brief awake appear to be the case that these events were paralysis than in definite and probable cases of related to out-of-hours or emergency surgery.
AAGA (Table 13.1).
13.17 Most events were reported immediately, except Table 13.1. Comparison of the immediate impact (Michigan
one case which was reported after several years. D denoting distress) and longer term impact (Wang scale and modified NPSA score >2; i.e. moderate or severe) for Class A and 13.18 The median perceived duration of the paralysis was B (Certain/probable) versus G (awake paralysis). In all categories, very short, 60 (10-180 [5-900]) sec. One case where the impact of the last appears more adverse the experience was very long did not appear to NAP5 Class
Michigan D
NPSA >2
have been administered any anaesthetic during Definite/probable the episode, perhaps because the syringe swap was not recognised and the diagnosis was initially A young, anxious patient was undergoing elective orthopaedic surgery. To alleviate anxiety, the anaesthetist planned to give midazolam 2mg but the patient became 13.21 Three (18%) of the drug error cases led to a formal unresponsive and was hand ventilated via a face mask. complaint or initiation of legal action at the time Two consultant colleagues arrived to help and it was later the case was reported to NAP5, a little higher than observed that the patient was behaving similarly to an was the case with Certain/probable cases (16% in inadequately reversed patient. Reversal was given and the Class G vs 11% of all Class A/B cases).
patient started responding again. The patient was later able to give a detailed description of being paralysed and unable 13.22 The Panel judged that all cases of awake paralysis to respond to the anaesthetist's commands (to take deep caused by drug error were preventable, and breaths and opening eyes). There was fear of death. The therefore, the quality of clinical care was generally episode lasted 15 min. The patient developed unpleasant deemed to be poor in the period leading up to dreams, nightmares and flashbacks, and symptoms of PTSD. AAGA. In contrast, quality of care after the event The patient received counselling for this. A formal complaint was frequently good (77% cases), largely because was received by the trust.
the event was promptly recognised and well managed (Table 13.2).
13.23 The majority of syringe swaps that led to AAGA in 13.19 All cases except one occurred on induction, before NAP5 were due to events that led to administration surgery started.
of a neuromuscular blockade without being 13.20 Most patients (15, 88%) experienced paralysis but preceded by a hypnotic agent (Table 13.3). In one two patients did not experience this sensation case lidocaine was given instead of an antibiotic despite the drug error and experienced only which led to cardiovascular and respiratory collapse tactile or auditory sensations. Pain was uncommon, and need for resuscitation. The patient recalled (1 of 17, 6%) arising only once and that was in events during the resuscitation.
conjunction with paralysis. The majority (11 of 17; Table 13.2. Panel judgements on quality of care and preventability for each of the Class A and B (certain/probable) versus Class G (awake
paralysis). Notwithstanding the inherent difficulties of the judgement (discussed in Chapter 5, Methods), quality of care before AAGA was always judged poor in Class G and always judged preventable Quality of care before AAGA
Quality of care after AAGA
Certain/probable, Class A Possible, Class B Awake paralysis, Class G Report and findings of the 5th National Audit Project
Drug errors and awake paralysis
Table 13.3. Drugs involved and psychological impact of ten
Swaps involving larger syringes, such as in induction syringe swaps. (NMBD: unidentified neuromuscular blocking agent and antibiotic, also led to paralysis and AAGA, as the antibiotic was mistaken for induction Drug Given
agent. Perhaps understandably, this did occur with thiopental and antibiotic (but just one case). Equally understandably, no drug error arose with propofol. In some of these cases poor communication within the team involving more than one anaesthetist led to these errors. Identifying and agreeing the roles of each anaesthetist in such teams is likely to reduce error.
A patient undergoing an urgent laparotomy for bowel obstruction was under the care of three anaesthetists on an emergency list; the plan was to administer fentanyl followed by thiopental and suxamethonium. Unfortunately, cefuroxime was mistaken for thiopental and administered instead. The patient's trachea was intubated but the patient became markedly tachycardic and hypertensive. The error was then 13.24 The psychological sequelae of AAGA for the patient realized and thiopental was administered. Post-operatively in this setting can be particularly severe. Of note: the the patient recalled the sensation of being unable to breath, severity does not appear to be related to duration of the discomfort of cricoid pressure and an unpleasant experience and even a few seconds of unintended sensation of a tube being passed into the back of their throat. This experience lasted for a maximum of two minutes. paralysis can lead to prolonged psychological The patient was not overly concerned about this event and sequelae (also see Chapter 7, Patient Experience). overall hospital experience was very positive. General anaesthesia was planned for a middle-aged The similarity of appearance of thiopental and cefuroxime in close obese patient for drainage of an abscess. The anaesthetist intended to give an anti-emetic before the induction dose of propofol, but mistakenly gave suxamethonium. The error was recognised immediately. The patient was aware for 30 seconds. The patient was extremely distressed in recovery and reported to staff that they had been paralysed, unable to breathe and felt they were going to die. In the post-operative period the patient was very angry and litigation was started.
13.25 The risk of a drug error is logically reduced by avoiding giving unnecessary drugs at the time of induction.
A young patient undergoing emergency surgery was anaesthetised out-of-hours by two trainees planning to undertake a rapid sequence induction. Suxamethonium 13.26 Seven drug preparation errors were reported (six was given instead of fentanyl while the patient was awake. of labelling error and one drug omission): and all The mistake was recognised quickly and the patient was led to awake paralysis and severe psychological anaesthetised with propofol. The patient had recall for a sequelae. (Table 13.4).
few seconds but no pain or discomfort and was generally unconcerned by the whole event. NAP5 Report and findings of the 5th National Audit Project
Drug errors and awake paralysis
Table 13.4. Drugs involved and psychological impact of six
Illustration of the similarity of labelled syringes for midazolam and ampoule-labelling and one drug-omission error. (*there was a atracurium left in close proximity suggestion that parexocib was also intended) Drug Given
13.27 The fundamental cause of most cases of wrong An anxious young patient was due to undergo general labelling or incorrect preparation appeared task- anaesthesia for minor surgery. Spontaneous respiration was related. Distractions and perceived time pressures planned. The patient remembered the anaesthetist's reassuring during the drawing up of drugs may lead to errors. words that they would soon be asleep, then remembered their arm ‘dropping' and being unable to hear their breathing. The consultant anaesthetist immediately realised that A middle aged patient was due to undergo elective shoulder suxamethonium had been given instead of fentanyl, and surgery. The anaesthetist intended to sedate the patient administered a dose of propofol whilst continuing to reassure before performing an interscalene block and then induce the patient. A single loose ampoule of suxamethonium had general anaesthesia. Atracurium 10mg was injected instead been placed lying close to the fentanyl and other induction of the intended midazolam. The patient recalled feelings drugs in the tray. This arose because the hospital had of panic, acute distress and the awareness of a very rapid instituted a policy preventing the entire box of suxamethonium heart rate. The anaesthetist quickly recognised that a being removed from the fridge (to avoid room temperature muscle relaxant had been administered, and anaesthesia degradation). Instead, the ODP had placed a single ampoule was induced within a few minutes. Whilst drawing up drugs of suxamethonium on the tray. The patient was supported, a in preparation for the case, the anaesthetist had been full explanation offered, and they suffered no long term impact. distracted by the ODP's request to leave the anaesthetic room to fetch equipment from a nearby store room. On return to the original task, atracurium was inadvertently An example of a single loose ampoule of suxamethonium and how drawn up into the syringe labelled as midazolam. Both easily it might be drawn instead of (in this example) ondansetron ampoules were of similar size and nearly similar colour. The anaesthetist's explanation to the patient in recovery post-operatively was graciously accepted, and no formal psychological support or treatment was required.
13.28 Preparation error accounted for a minority of the drug error cases reported to NAP5. A common thread between them was pre-existing organisational elements that were likely to have increased the chance for error to be introduced (i.e. latent errors).
Report and findings of the 5th National Audit Project
Drug errors and awake paralysis
13.29 The practice of having a delay between drawing 13.33 Recurring themes in the details of the cases up a dilutant into a pre-labelled syringes and then were mention of staff shortages, a pressured later mixing/adding the active drug led to AAGA environment with ‘busy' lists. Some hospital policies through drug omission.
for the storage and preparation of drugs appeared misguided and themselves were contributory to A middle-aged patient required a general anaesthetic for error (see Chapter 23, Human Factors).
expedited surgery. After induction the anaesthetist noticed 13.34 Distractions during critical moments can have very greater than expected fasciculations in the patient. Following serious consequences. Jothiraj el al. (2013) reported intubation, a volatile agent was immediately commenced. that other anaesthetists and circulating nurses are At this point the anaesthetist realised that no induction the most common causes of distractions. In terms agent had been administered, only suxamethonium. In that of individual conduct, it seemed that a lack of hospital, thiopental was kept in a central store, so was not vigilance and having several similar sized syringes immediately available for mixing. After finishing the previous on the same drug tray may be contributory.
case, the anaesthetist forgot that the thiopental had not been mixed and proceeded with a rapid sequence induction. 13.35 Although checking ampoules and labels with The patient was induced with a syringe containing only a second person is theoretically attractive, the water (but presumably labelled as thiopental). In recovery, evidence base for checking with a second person the patient reported experiencing paralysis and was clearly before drawing up or giving a drug is weak. afraid: ‘'I thought I might not make it through the operation". Although double-checking is accepted as necessary The patient was aware of being intubated and was unsure in other familiar settings (e.g. the administration how long it would last but soon after lost consciousness. of blood products), the value of checking routinely The patient developed a new anxiety state, flashbacks and administered drugs in the anaesthetic context is possible PTSD. The patient subsequently had meetings with more controversial. the clinical director and counselling was arranged. 13.36 When two people are responsible for the same task, neither person is truly responsible. There are several examples of this phenomenon in this report, 13.30 The cases in this chapter are perhaps more where two anaesthetists have been present during accurately termed ‘unintended awake paralysis', a case, yet perhaps nobody was truly leading the but are perceived by the patient as ‘accidental team. Paradoxically, the introduction of double- awareness'. The adverse impact is commonly checking for routine drug administration could severe. This underlines the reality that paralysis worsen ‘involuntary automaticity' and reduce, rather whilst conscious is a potentially harmful experience. than increase, patient safety. Of note: the impact of paralysis in generating 13.37 A technical solution to the problem would involve distress and longer-term harm, which is also use of pre-prepared drug syringes and use of emphasised elsewhere – Chapters 6, Results; 8, scanning technology to ‘check' drugs before Induction; 9, Maintenance; 10, Emergence; and 19 administration. Any method would need to accommodate the need for rapid response to 13.31 The majority of drug errors causing awareness in a changing situation during surgery, and hence this category are due to simple syringe-swaps of the need to have a range of drugs immediately similar sized syringes, or similar coloured fluids, such available whose use was not anticipated.
as suxamethonium vs. fentanyl or ondansetron (all 13.38 Short of such technology, anaesthetists need to normally drawn in 2ml syringes); non-depolarising accept that they are all prone to making errors drugs vs midazolam (both normally in 5 ml syringes); and should therefore, develop robust individual or antibiotics vs thiopental (both usually in 20 ml mechanisms to protect themselves. The anaesthetist syringes). Indeed, not a single error was reported for needs to recognise their vulnerability to these dissimilar sized syringes (Tables 13.3 and 13.4).
potentially very serious incidents, and develop 13.32 However, the overall incidence of drug error related layers of defence to prevent drug errors; particularly to neuromuscular blockade must be regarded those involving the unintended administration of as low. The Activity Survey indicates 2.8 million neuromuscular blocking drugs. In this context the general anaesthetics per year, with 44.8% ( 1.25 NAP5 data suggests several strategies that could million) involving neuromuscular blockade. This reduce error.
represents one report of accidental paralysis 13.39 Anaesthetic departments should work with for every 70,000 general anaesthetics involving pharmacy departments to take ampoule NAP5 Report and findings of the 5th National Audit Project
Drug errors and awake paralysis
appearance into proper consideration when iMPlicaTions For research
choosing suppliers and should avoid frequent, changes of drug suppliers. If this is unavoidable, Research Implication 13.1
then it must appear on the hospital risk register. Further research is needed into issues relating to the cause 13.40 Individual strategies that may be helpful include and prevention of drug error in anaesthesia. Relevant reserving 5ml syringes for neuromuscular blockade questions include: Whether errors are more frequent when only, double-labelling of these syringes or, if drugs are prepared by anaesthetists vs assistants vs double available, using coloured syringes or different checking? Which strategies for double checking might reduce error? What sort of psychology is involved when teams double-check drugs? 13.41 Although often relegated to being a routine, perhaps subconscious task, anaesthetists should Research Implication 13.2
appreciate that preparing drugs is a potentially high The design of technical solutions to minimise drug error risk activity and so be careful to avoid all distractions offers large scope for further research, to establish how the during this period. The need to read all ampoules right drug is given at the right time to the right patient. This and use labels is self-evident, but any doubt or might include further analysis of interventions involving concern or distraction should lead to consideration barcoding, or pre-prepared drugs, or drugs released from that the wrong drug may have been prepared.
fridges or cupboards only on specific request.
13.42 Perhaps greater attention is also needed to organising the anaesthetic workspace, with attention to detail on where and how the most potentially ‘dangerous' drugs (i.e. the neuromuscular blocking drugs) are kept and handled (e.g. in separate trays). Part of this is the need to avoid unnecessarily complicated anaesthetic techniques and avoid the Hospitals should take ampoule appearance into administration at induction of drugs not directly account to avoid multiple drugs of similar appearance. necessary (e.g. anti-emetics, which can often safely Hospital policies should direct how this risk is managed. be administered later).
This may require sourcing from different suppliers. 13.43 After an error had happened, the patient experience appeared greatly influenced by The relevant anaesthetic organisations should anaesthetic conduct. In some cases hurried efforts engage with industry to seek solutions to the were made to reverse paralysis without attending to problem of similar drug packaging and presentation.
the patient's level of consciousness, while in others reassurance of the patient and ensuring comfort was prioritised. In the latter group, it seemed that patients, on understanding events, appeared to Anaesthetists should develop clear personal strategies have considerably more benign experiences and in the preparation of drugs that minimise or avoid fewer or no sequelae. scope for drug error. This includes the recognition that preparation of drugs for use is a potentially high risk 13.44 Where a drug error leading to accidental paralysis activity, in which distractions should be avoided. This has occurred there are three priorities, in order: applies particularly to neuromuscular blocking drugs.
first, immediately reassuring the patient that they are safe, whilst second, inducing anaesthesia promptly to mitigate continued adverse impact Where a drug error leading to accidental paralysis (including airway management) and last, to has occurred there are three priorities, in order: first, consider reversing the paralysis at an appropriate immediately reassuring the patient that they are time (e.g. guided by nerve stimulator monitoring). safe, whilst second, inducing anaesthesia promptly to mitigate continued adverse impact (including airway management) and last, to consider reversing the paralysis at an appropriate time (e.g. guided by nerve stimulator monitoring). Report and findings of the 5th National Audit Project
Drug errors and awake paralysis
reFerences
Alsulami Z, Conroy S, Choonara I. Double checking the administration
of medicines: what is the evidence? A systematic review. Archives of
Disease in Childhood 2012;97:833–37.
Cooper JB, Newbower RS, Long CD, McPeek B. Preventable
anesthesia mishaps: a study of human factors. Anesthesiology
1978;49:399–406.
Evley R Russell J, Mathew D, Hall R, Gemmell L, Mahajan RP.
Confirming the drugs administered during anaesthesia: a feasibility
study in the pilot National Health Service sites, UK. British Journal of
Anaesthesia 2010;105:289–96.
Ghoneim MM1, Block RI, Haffarnan M, Mathews MJ. Awareness during
anesthesia: risk factors, causes and sequelae: a review of reported
cases in the literature. Anesthesia & Analgesia 2009;108:527–35.
Jothiraj H, Howland-Harris J, Evley R and Moppett IK. Distractions
and the anaesthetist: a qualitative study of context and direction of
distraction. British Journal of Anaesthesia 2013;111:477–82.
Merry AF, Webster CS, Hannam J, et al. Multimodal system
designed to reduce errors in recording and administration of drugs
in anaesthesia: prospective randomised clinical evaluation. British
Medical Journal 2011;343:d5543.
Mihai R, Scott S, Cook TM. Litigation related to inadequate
anaesthesia: an analysis of claims against the NHS in England 1995-
2007. Anaesthesia 2009;64:829–35).
Osborne GA, Bacon AK, Runciman WB, Helps SC. Crisis management
during anaesthesia: awareness and anaesthesia. Quality & Safety in
Health Care 2005;14:e16.
Reason J. Human error: models and management. British Medical
Journal 2000;320:768–70.
Samuelsson P, Brudin L, Sandin RH. Late psychological symptoms
after awareness among consecutively included surgical patients.
Anesthesiology 2007;106:26–32.
Toft B, Mascie-Taylor H. Involuntary automaticity: a work-system
induced risk to safe health care. Health Services Management
Research 2005;18:211–16.
Webster CS, Merry AF, Larsson L, McGrath KA, Weller J. The
frequency and nature of drug administration error during anaesthesia.
Anaesthesia and Intensive Care 2001;29:494–500.
Zhang Y, Dong YJ, Webster CS, et al. The frequency and nature of drug
administration error during anaesthesia in a Chinese hospital. Acta
Anaesthesiologica Scandinavica 2013;57:158–64.
NAP5 Report and findings of the 5th National Audit Project

Source: http://www.nationalauditprojects.org.uk/NAP5Doc_Drug_errors_and_awake_paralysis