Leonie Watterson

Crisis management during anaesthesia: the development of an anaesthetic crisis management manual

Background: All anaesthetists have to handle life threatening crises with little or no warning. However, some cognitive strategies and work practices that are appropriate for speed and efficiency under normal circumstances may become maladaptive in a crisis. It was judged in a previous study that the use of a structured “core” algorithm (based on the mnemonic COVER ABCD–A SWIFT CHECK) would diagnose and correct the problem in 60% of cases and provide a functional diagnosis in virtually all of the remaining 40%. It was recommended that specific sub-algorithms be developed for managing the problems underlying the remaining 40% of crises and assembled in an easy-to-use manual. Sub-algorithms were therefore developed for these problems so that they could be checked for applicability and validity against the first 4000 anaesthesia incidents reported to the Australian Incident Monitoring Study (AIMS). Methods: The need for 24 specific sub-algorithms was identified. Teams of practising anaesthetists were assembled and sets of incidents relevant to each sub-algorithm were identified from the first 4000 reported to AIMS. Based largely on successful strategies identified in these reports, a set of 24 specific sub-algorithms was developed for trial against the 4000 AIMS reports and assembled into an easy-to-use manual. A process was developed for applying each component of the core algorithm COVER at one of four levels (scan-check-alert/ready-emergency) according to the degree of perceived urgency, and incorporated into the manual. The manual was disseminated at a World Congress and feedback was obtained. Results: Each of the 24 specific crisis management sub-algorithms was tested against the relevant incidents among the first 4000 reported to AIMS and compared with the actual management by the anaesthetist at the time. It was judged that, if the core algorithm had been correctly applied, the appropriate sub-algorithm would have been resolved better and/or faster in one in eight of all incidents, and would have been unlikely to have caused harm to any patient. The descriptions of the validation of each of the 24 sub-algorithms constitute the remaining 24 papers in this set. Feedback from five meetings each attended by 60–100 anaesthetists was then collated and is included. Conclusion: The 24 sub-algorithms developed form the basis for developing a rational evidence-based approach to crisis management during anaesthesia. The COVER component has been found to be satisfactory in real life resuscitation situations and the sub-algorithms have been used successfully for several years. It would now be desirable for carefully designed simulator based studies, using naive trainees at the start of their training, to systematically examine the merits and demerits of various aspects of the sub-algorithms. It would seem prudent that these sub-algorithms be regarded, for the moment, as decision aids to support and back up clinicians’ natural responses to a crisis when all is not progressing as expected.

Crisis management during anaesthesia: hypotension

Background: Hypotension is commonly encountered in association with anaesthesia and surgery. Uncorrected and sustained it puts the brain, heart, kidneys, and the fetus in pregnancy at risk of permanent or even fatal damage. Its recognition and correction is time critical, especially in patients with pre-existing disease that compromises organ perfusion. Objectives: To examine the role of a previously described core algorithm “COVER ABCD–A SWIFT CHECK”, supplemented by a specific sub-algorithm for hypotension, in the management of hypotension when it occurs in association with anaesthesia. Methods: Reports of hypotension during anaesthesia were extracted and studied from the first 4000 incidents reported to the Australian Incident Monitoring Study (AIMS). The potential performance of the COVER ABCD algorithm and the sub-algorithm for hypotension was compared with the actual management as reported by the anaesthetist involved. Results: There were 438 reports that mentioned hypotension, cardiovascular collapse, or cardiac arrest. In 17% of reports more than one cause was attributed and 550 causative events were identified overall. The most common causes identified were drugs (26%), regional anaesthesia (14%), and hypovolaemia (9%). Concomitant changes were reported in heart rate or rhythm in 39% and oxygen saturation or ventilation in 21% of reports. Cardiac arrest was documented in 25% of reports. As hypotension was frequently associated with abnormalities of other vital signs, it could not always be adequately addressed by a single algorithm. The sub-algorithm for hypotension is adequate when hypotension occurs in association with sinus tachycardia. However, when it occurs in association with bradycardia, non-sinus tachycardia, desaturation or signs of anaphylaxis or other problems, the sub-algorithm for hypotension recommends cross referencing to other relevant sub-algorithms. It was considered that, correctly applied, the core algorithm COVER ABCD would have diagnosed 18% of cases and led to resolution in two thirds of these. It was further estimated that completion of this followed by the specific sub-algorithm for hypotension would have led to earlier recognition of the problem and/or better management in 6% of cases compared with actual management reported. Conclusion: Pattern recognition in most cases enables anaesthetists to determine the cause and manage hypotension. However, an algorithm based approach is likely to improve the management of a small proportion of atypical but potentially life threatening cases. While an algorithm based approach will facilitate crisis management, the frequency of co-existing abnormalities in other vital signs means that all cases of hypotension cannot be dealt with using a single algorithm. Diagnosis, in particular, may potentially be assisted by cross referencing to the specific sub-algorithms for these.

Crisis management during anaesthesia: bradycardia

Background: Bradycardia in association with anaesthesia may have many potential causes and associated conditions, some rare and/or obscure. A prompt appropriate response is important as some homeostatic mechanisms may be impaired under anaesthesia. Objectives: To examine the role of a previously described core algorithm “COVER ABCD–A SWIFT CHECK”, supplemented by a specific sub-algorithm for bradycardia, in the management of bradycardia occurring in association with anaesthesia. Methods: The potential performance of this structured approach for each of the relevant incidents among the first 4000 reported to the Australian Incident Monitoring Study (AIMS) was compared with the actual management as reported by the anaesthetists involved. Results: From the first 4000 incidents reported to AIMS, 265 reports which described bradycardia during anaesthesia were extracted and studied. Bradycardia was associated with hypotension in 51% of cases, cardiac arrest in 25% of cases and hypertension in one case. In 22% of reports apparent desaturation or an abnormality of ventilation was described. Bradycardia was caused by drug events (28%), airway related events (16%), autonomic reflexes (14%), and regional anaesthesia (9%). Airway and drug events caused 75% of cases involving children. It was considered that, correctly applied, the core algorithm COVER would have diagnosed 53 cases (20%) and led to corrective management in 45 (85%) of these; this included an important subset of airway and ventilation problems. Completion of COVER ABCD–A SWIFT CHECK followed by the specific sub-algorithm for bradycardia would have resulted in diagnosis and appropriate management in all but two cases. It would have led to earlier recognition of the problem and/or better management in 11 cases (4%) when compared with the actual management described in the reports. Conclusion: Steps should be taken to manage bradycardia whilst associated conditions are managed concurrently. Analysis of cardiac rhythm should not be pursued to the exclusion of supportive therapy. The use of a structured approach in the management of bradycardia associated with anaesthesia is likely to improve management in the small percentage of cases in which the diagnosis of the cause may be missed or delayed.

Crisis management during anaesthesia: cardiac arrest

Background: Cardiac arrest attributable to anaesthesia occurs at the rate of between 0.5 and 1 case per 10 000 cases, tends to have a different profile to that of cardiac arrest occurring elsewhere, and has an in-hospital mortality of 20%. However, as individual practitioners encounter cardiac arrest rarely, the rapidity with which the diagnosis is made and the consistency of appropriate management varies considerably. Objectives: To examine the role of a previously described core algorithm “COVER ABCD–A SWIFT CHECK”, supplemented by a sub-algorithm for cardiac arrest, in the management of cardiac arrest occurring in association with anaesthesia. Methods: The potential performance of this structured approach for each the relevant incidents among the first 4000 reported to the Australian Incident Monitoring Study (AIMS) was compared with the actual management as reported by the anaesthetists involved. Results: There were 129 reports of cardiac arrest associated with anaesthesia among the first 4000 AIMS incident reports. Identified aetiological factors were grouped into five categories: (1) anaesthetic technique (11 cases with this category alone; 32 with this and one or more of the other categories, representing 25% of all 129 cardiac arrests); (2) drug related (16; 32, 25%); (3) associated with surgical procedure (9; 29, 22%); (4) associated with pre-existing medical or surgical disease (30; 82, 64%); (5) unknown (8; 14, 11%). The “real life” presentation and management of cardiac arrest in association with anaesthesia differs substantially from that detailed in general published guidelines. Cardiac rhythms at the time were sinus bradycardia (23%); asystole (22%); tachycardia/ventricular tachycardia/ventricular fibrillation (14%); and normal (7%), with a further third unknown. Details of treatment were recorded in 110 reports; modalities employed included cardiac compression (72%); adrenaline (61%); 100% oxygen (58%); atropine (38%); intravenous fluids (25%), and electrical defibrillation (17%). There were no deaths or permanent morbidity in the 11 cases due solely to anaesthetic technique. 24 of the 25 deaths occurred in patients with significant pre-existing medical or surgical disease. Conclusion: Because there are often multiple contributing factors to a cardiac arrest under anaesthesia, a complete systematic assessment of the patient, equipment, and drugs should be completed. The “COVER ABCD–A SWIFT CHECK” algorithm was judged to be a satisfactory process in this context and should be carried out even if the cause of the cardiac arrest is already thought to have been found. The diagnosis and management of cardiac arrest in association with anaesthesia differs considerably from that encountered elsewhere. The outcome is generally good, with most patients leaving hospital alive and apparently well.

Effective Management of Anaesthetic Crises: Development and Evaluation of a College-accredited Simulation-based Course for Anaesthesia Education in Australia and New Zealand

The Effective Management of Anesthetic Crises (EMAC) course is a joint initiative between the Australian and New Zealand College of Anesthetists (ANZCA) and simulation centers. This standardized 2.5-day course has become an integral component of training for Fellowship of ANZCA and as such is an innovative development on the global anesthesia scene. Since its inception in 2002, over 600 anesthetists, with equal numbers of specialists and trainees, have attended EMAC throughout Australia, New Zealand, and Hong Kong. Course evaluations from 499 anesthetists and a follow-up survey showed strong support for the course and its relevance to clinical practice. The course is perceived by participants as changing their practice and improving their management of anesthetic crises. Exposure to the concepts of effective crisis management is now widespread in the anesthetic community in the region and should contribute to improved patient safety.

Observer roles that optimise learning in healthcare simulation education: a systematic review

BackgroundSimulation is widely used in health professional education. The convention that learners are actively involved may limit access to this educational method. The aim of this paper is to review the evidence for learning methods that employ directed observation as an alternative to hands-on participation in scenario-based simulation training. We sought studies that included either direct comparison of the learning outcomes of observers with those of active participants or identified factors important for the engagement of observers in simulation. We systematically searched health and education databases and reviewed journals and bibliographies for studies investigating or referring to observer roles in simulation using mannequins, simulated patients or role play simulations. A quality framework was used to rate the studies.MethodsWe sought studies that included either direct comparison of the learning outcomes of observers with those of active participants or identified factors important for the engagement of observers in simulation. We systematically searched health and education databases and reviewed journals and bibliographies for studies investigating or referring to observer roles in simulation using mannequins, simulated patients or role play simulations. A quality framework was used to rate the studies.ResultsNine studies met the inclusion criteria. Five studies suggest learning outcomes in observer roles are as good or better than hands-on roles in simulation. Four studies document learner satisfaction in observer roles. Five studies used a tool to guide observers. Eight studies involved observers in the debrief. Learning and satisfaction in observer roles is closely associated with observer tools, learner engagement, role clarity and contribution to the debrief. Learners that valued observer roles described them as affording an overarching view, examination of details from a distance, and meaningful feedback during the debrief. Learners who did not value observer roles described them as passive, or boring when compared to hands-on engagement in the simulation encounter.ConclusionsLearning outcomes and role satisfaction for observers is improved through learner engagement and the use of observer tools. The value that students attach to observer roles appear contingent on role clarity, use of observer tools, and inclusion of observers’ perspectives in the debrief.

Contextual information influences diagnosis accuracy and decision making in simulated emergency medicine emergencies

Background It is well documented that adaptations in cognitive processes with increasing skill levels support decision making in multiple domains. We examined skill-based differences in cognitive processes in emergency medicine physicians, and whether performance was significantly influenced by the removal of contextual information related to a patients medical history. Method Skilled (n=9) and less skilled (n=9) emergency medicine physicians responded to high-fidelity simulated scenarios under high- and low-context information conditions. Results Skilled physicians demonstrated higher diagnostic accuracy irrespective of condition, and were less affected by the removal of context-specific information compared with less skilled physicians. The skilled physicians generated more options, and selected better quality options during diagnostic reasoning compared with less skilled counterparts. These cognitive processes were active irrespective of the level of context-specific information presented, although high-context information enhanced understanding of the patients’ symptoms resulting in higher diagnostic accuracy. Conclusions Our findings have implications for scenario design and the manipulation of contextual information during simulation training.

Crisis management during anaesthesia: myocardial ischaemia and infarction

Background: Myocardial ischaemia and infarction are significant perioperative complications which are associated with poor patient outcome. Anaesthetic practice should therefore focus, particularly in the at risk patient, on their prevention, their accurate detection, on the identification of precipitating factors, and on rapid effective management. Objectives: To examine the role of a previously described core algorithm “COVER ABCD–A SWIFT CHECK” supplemented by a specific sub-algorithm for myocardial ischaemia and infarction in the management of myocardial ischaemia and/or infarction occurring in association with anaesthesia. Methods: The potential performance of this structured approach for each of the relevant incidents among the first 4000 reported to the Australian Incident Monitoring Study (AIMS) was compared with the actual management as reported by the anaesthetists involved. Results: Of the 125 incidents retrieved from the 4000 reports, 40 (1%) were considered to demonstrate myocardial infarction or ischaemia. The use of the structured approach described in this paper would have led to appropriate management in 90% of cases, with the remaining 10% requiring other sub-algorithms. It was considered that the application of this structured approach would have led to earlier recognition and/or better management of the problem in 45% of cases. Conclusion: Close and continuous monitoring of patients at risk of myocardial ischaemia during anaesthesia is necessary, using optimal ECG lead configurations, but sensitivity of this monitoring is not 100%. Coronary vasodilatation with glyceryl trinitrate (GTN) should not be withheld when indicated and the early use of beta blocking drugs should be considered even with normal blood pressures and heart rates.

Crisis management during anaesthesia: tachycardia

Background: Tachycardia during anaesthesia is a common event. In most cases the cause is easily identified and the problem promptly resolved. However, in some the cause may be rare or obscure. Under such circumstances, attempting to initiate appropriate supportive therapy and to consider a large differential diagnosis in a comprehensive manner may lead to delays which can put a patient at risk. Objectives: To examine the role of a previously described core algorithm “COVER ABCD–A SWIFT CHECK”, supplemented by a specific sub-algorithm for tachycardia, in the management of tachycardia developing in association with anaesthesia. Methods: The potential performance of this structured approach for each of the relevant incidents among the first 4000 reported to the Australian Incident Monitoring Study (AIMS) was compared with the actual management as reported by the anaesthetists involved. Results: There were 145 causative events identified in 123 reports of tachycardia during anaesthesia which were extracted and studied from the first 4000 incidents reported to AIMS. Subgroups were identified based on blood pressure at the time of presentation. Of the 145 causes, tachycardia was associated with hypotension (33%), normotension (27%), hypertension (26%), and cardiac arrest (17%). For simplicity it is recommended that other cardiovascular sub-algorithms are followed when the blood pressure is also abnormal. This includes cardiac arrest and hypotension. In hypotensive states the tachycardia sub-algorithm should be followed until the cardiac rhythm is diagnosed. Sinus tachycardia and hypotension should be managed as hypotension. It was considered that, correctly applied, the core algorithm COVER would have diagnosed 35% of cases and led to resolution in 70% of these. It was estimated that completion of COVER followed by the sub-algorithm for tachycardia would have led to earlier recognition of the problem and/or better management in four cases when compared with actual management reported. Conclusion: Tachycardia during anaesthesia is frequently associated with a simultaneous change in other monitored vital signs. The differential diagnosis is large. Addressing it in a comprehensive fashion requires a structured approach. A specific sub-algorithm treatment for tachycardia based on the associated blood pressure and on the prevailing heart rhythm in the case of hypotension offers a systematic guide which complements the benefits obtained by employing the core algorithm COVER ABCD.

A cadaver study comparing three fibreoptic‐assisted techniques for converting a supraglottic airway to a cuffed tracheal tube

After rescuing an airway with a supraglottic airway device, a method to convert it to a cuffed tracheal tube is often needed. The best method to do this has never been directly studied. We compared three techniques for conversion of a standard LMA® Unique airway to a cuffed endotracheal tube using a fibrescope. The primary endpoint was time to intubation, with secondary endpoints of success rate, perceived difficulty and preferred technique. We also investigated the relationship between level of training and prior training and experience with the techniques on the primary outcome. The mean (95% CI) time to intubation using a direct tracheal tube technique of 37 (31–42) s was significantly shorter than either the Aintree intubation catheter technique at 70 (60–80) s, or a guidewire technique at 126 (110–141) s (p < 0.001). Most (13/24) participants rated the tracheal tube as their preferred technique, while 11/24 preferred the Aintree technique. In terms of perceived difficulty, 23/24, 21/24 and 9/24 participants rated the tracheal tube technique, Aintree technique and guidewire technique, respectively, as either very easy or easy. There was no relationship between prior training, prior experience or level of training on time to completion of any of the techniques. We conclude the tracheal tube and Aintree techniques both provide a rapid and easy method for conversion of a supraglottic airway device to a cuffed tracheal tube. The guidewire technique cannot be recommended.