A prospective, cohort evaluation of major and ... - Wiley Online Library

Anaesthesia 2017, 72, 42–48

doi:10.1111/anae.13640

Original Article A prospective, cohort evaluation of major and minor airway management complications during routine anaesthetic care at an academic medical centre J. M. Huitink,1 P. P. Lie,2 I. Heideman,3 E. P. Jansma,4 R. Greif,5 N. van Schagen6 and A. Schauer1 1 Consultant Anaesthetist, 2 Physician, 3 Student Health and Life Sciences, 6 Computer Technician, Department of Anesthesiology, VU University Medical Center, Amsterdam, the Netherlands 4 Medical Information Specialist, Department of Epidemiology and Biostatistics, VU University Medical Center, Amsterdam, the Netherlands 5 Professor, Anaesthetist, Department of Anesthesiology and Pain Therapy, Bern University Hospital, University of Bern, Bern, Switzerland

Summary The aim of this study was to develop an audit tool to identify prospectively all peri-operative adverse events during airway management in a cost-effective and reproducible way. All patients at VU University Medical Center who required general anaesthesia for elective and emergency surgical procedures were included during a period of 8 weeks. Daily questionnaires and interviews were taken from anaesthesia trainees and anaesthetic department staff members. A total of 2803 patients underwent general anaesthesia, 1384 men and 1419 women, including 2232 elective patients and 571 emergency procedures, 697 paediatric and 2106 adult surgical procedures. A total of 168 airway-related events were reported. The incidence of severe airway management-related events was 24/2803 (0.86%). There were 12 (0.42%) unanticipated ICU admissions, two patients (0.07%) required a surgical airway. There was one (0.04%) death, one cannot intubate cannot oxygenate (0.04%), one aspiration (0.04%) and eight (0.29%) severe desaturations < SpO2 50%. We suggest that our method to determine and investigate airway management-related adverse events could be adopted by other hospitals. .................................................................................................................................................................

Correspondence to: J. M. Huitink Email: [email protected] Accepted: 19 July 2016 Keywords: airway management; brain damage; complications; emergency surgical airway; general anaesthesia; serious adverse events Presented in part at Euroanaesthesia, Berlin, Germany, May 2015.

Introduction Although airway management complications are rare, they can result in severe consequences such as brain damage and death [1, 2]. Despite their importance, airway management complications remain generally poorly monitored, so that reliable and reproducible information on the incidence and nature of events is 42

often inadequate, leaving litigation-based analyses as an estimate of severity and frequency [3–6]. The Fourth National Audit Project (NAP4) of the Royal College of Anaesthetists and the Difficult Airway Society tried to estimate the incidence of major complications in airway management by performing both quantitative and qualitative analyses in the UK. Data © 2016 The Association of Anaesthetists of Great Britain and Ireland

Huitink et al. | Determination of airway management-related adverse events

on airway management complications during anaesthesia and in the emergency department were prospectively studied and collected in NHS hospitals over a 12-month period [1, 2]. Inspired by this, we aimed to identify all peri-operative airway management complications and problems at our academic hospital in an easy prospective and cost-effective way.

Methods This was a prospective, observational study of perioperative airway management-related complications and problems performed over 8 weeks (60 days) at the VU University Medical Center (VUmc), the Netherlands, following Institutional Review Board approval. The study period was from 08.00 hours April 14th to 08.00 hours June 15th 2014. All anaesthesia trainees and staff anaesthetists consented to participate. Patient consent was waived because this was an audit of clinical practice. All patients at VUmc who required general anaesthesia for elective and emergency surgical procedures were included. Procedures with sedation or regional anaesthesia as the sole anaesthetic technique were excluded from analysis. A medical student undertook daily questionnaires and interviews from anaesthetic trainees and staff members of the anaesthetic department at the end of each working day to obtain data concerning airway-related complications or problems. Nurse anaesthetists were not interviewed to prevent duplicates (in the Netherlands, nurse anaesthetists work under close supervision and never manage cases themselves). This was done personally by a telephone call or, more often, a face to face meeting in a manner that was appropriate and convenient for the doctors so as to not disturb their busy daily clinical routine. We asked directly if problems had occurred and also if voluntary reporting was possible. Physicians were contacted by one investigator (IH). Enquiries concerning problems or complications during airway management were made. All other problems, difficulties or events referring to airway management were also documented and included in the database. Data were anonymised for patients, doctors and assisting medical teams. The anaesthetist and trainee who worked as a team were interviewed together, if that was possible, to prevent duplication. If it was not possible, for example, due to © 2016 The Association of Anaesthetists of Great Britain and Ireland


clinical work, one member of the team was interviewed. Patient numbers were noted and, if duplicates were found, this was corrected in the database. In our department, it is already standard for anaesthestic trainees working weekend and night shifts to routinely inform all colleagues about relevant patient information by giving presentations the following morning or after the weekend shift on Monday morning. In addition, this trainee was personally consulted by IH the following morning to discuss and include all relevant airway-related complications and problems. Besides being invited to actively report, perioperative desaturations during anaesthesia were also monitored by an automated computer analysis program, VUmc’s Anaesthesia Information System. All episodes of peri-operative desaturation during induction of anaesthesia or emergence or in the recovery area (SpO2 < 93%) were monitored and collected in a database. An automated proprietary computer program (VUmc data management system, not commercially available) was used. This system is available in all operating and recovery rooms. Airway-related problems were defined as clinical situations that could not be avoided during daily clinical practice, but did not cause irreversible damage. Problems reported were: episodes of hypoxia (SpO2 < 93%); need for additional airway devices, for example, intubation catheters; accidental oesophageal intubation; and difficulty with tracheal intubation. Difficulty with tracheal intubation was caused by many reasons, including: choosing a wrong sized tracheal tube; cervical spine immobilisation collars that caused difficulty during airway management; difficult upper airway access (e.g. post-traumatic anatomy, small mouth opening, Mallampati ≥ 3, difficult mask ventilation); technical problems (ventilator failure, accidental tracheal tube disconnection); accidental extubation or supraglottic airway device (SAD) dislodgement; and other problems that were reported by medical staff. Complications were defined as any adverse events that could have been avoided and could have, or did cause adverse clinical outcome for a patient [7]. We used the same definitions as had been used during NAP4: death; permanent brain damage; need for an unanticipated surgical airway; and unanticipated ICU admittance. 43



Complications were categorised into severity according to their clinical outcome: 1 no consequence; 2 mild consequence (e.g. lip laceration, tooth damage); 3 moderate consequence (e.g. airway obstruction, aspiration); 4 severe consequence (serious adverse events such as unanticipated ICU admission, irreversible brain damage, death, emergency surgical airway). Emergency surgical airway was defined as any form of unplanned invasive tracheal access as part of managing the airway in case of emergency, for example, surgical tracheostomy or hollow needle insertion. Hypoxia was defined as SpO2 < 93% because that is the deflection point for the oxygen haemoglobin desaturation curve. Values below 93% can become rapidly dangerous because they may further drop very quickly at this point. Difficult tracheal intubation was defined as the need for more than one attempt to successfully intubate the trachea with a tracheal tube [6]. Difficult supraglottic airway device (SAD) insertion was defined as the need for more than one attempt to insert a SAD [6]. The following patient characteristics were retrieved from the electronic patient files: age; ASA physical status; sex; body mass index (BMI); Mallampati score; and Cormack and Lehane intubation grade. Any previous airway-related comments in the patient’s medical history or findings during pre-operative screening were also included. This study was performed at the VU University Medical Center Amsterdam which is an academic teaching hospital and level-1 trauma centre. VUmc serves a population of about 3 million people and approximately 18,000 general anaesthetics are administered every year. It has 16 operating theatres located on two different hospital floor levels. The emergency department is a busy metropolitan unit with four resuscitation bays. The anaesthesia department manages the recovery area which has 24 beds, of which between two and four are used for 24-hour care. Patients in the resuscitation rooms are treated by multidisciplinary teams which always include an anaesthetist. A helicopter emergency medical system staffed with a team of trauma doctors and paramedics and a helicopter landing area is located on the upper level and roof of the hospital. We calculated mean scores of patients’ characteristics and the incidence of adverse events. 44

Confidence rates were calculated for incidences. MedCalc Software© 16.4.3 was used for statistical analysis. After data collection by the interviewer, all events were categorized by three authors (JH, PL, IH) into complications and problems according to our definition and by consensus.

Results A total of 2803 patients underwent general anaesthesia, 1384 men and 1419 women, including 2232 elective and 571 emergency procedures. There were 697 paediatric (younger than 18 years) and 2106 adult surgical procedures. A total of 172 airway-related complications and problems were reported during these 2803 procedures of which 168 were considered relevant for analysis. Grounds for exclusion were events were not caused by general anaesthesia or not actually an airway-related event. The incidence of severe events was 24 (0.86%). In one patient, an emergency surgical airway contributed to death in a palliative setting. Unanticipated ICU admission occurred in 12 patients (0.42%), one death 1 (0.04%), two surgical airways (0.07%), no irreversible brain damage, one resuscitation from hypoxia (0.04%), one ‘cannot intubate cannot oxygenate’ (0.04%), one aspiration (0.04%), 44 had SpO2 < 93% (1.5%) and eight had SpO2 < 50% (0.29%) (Table 1). The 168 reported airway-related adverse events, consisted of 64 (38%) airway management complications. The majority of airway management-related events in patients undergoing general anaesthesia occurred in men (105/168; 62.5%); in ASA 1 and 2 (127/168; 75.6%) patients, and in the age range of < 10 years or between 40 years and 80 years (106/168; 63%). SpO2 < 93% was the most reported event (Table 1). The peri-operative severe events occurred in the operating theatre (n = 18), recovery room (n = 4), and hallways during patient transfer from the operating theatre to the recovery room (n = 2). In adults, the majority of airway-related complications occurred in adults with BMI > 28 kg.m 2 (Table 2). The phases of anaesthesia was induction in 115 patients (68.5%), maintenance in 20 (11.9%), © 2016 The Association of Anaesthetists of Great Britain and Ireland


Table 1 Number and aetiology of airway managementrelated complications and problems. Events Accidental extubation Airway obstruction/stridor Pulmonary aspiration Blood clots in airway Bronchospasm Cannot mask ventilate Cannot intubate/cannot oxygenate Death Dental injury SpO2 < 93% during induction Emergency surgical airway Epistaxis Oesophageal intubation Laryngospasm Lip injury No seal with SAD Nasal intubation impossible Neurological damage Mucus plug in airway Pneumothorax Resuscitation Postoperative upper airway bleeding Respiratory failure/re-intubation Copious salivation Gastric tube misplaced in trachea Technical problem with device Unanticipated intubation problems Unanticipated ICU admission Total

Table 2 Patient adverse events.

characteristics

2 5 3 2 12 5 1 1 1 44 2 1 5 11 1 6 3 0 6 1 1 3 5 1 3 5 29 9 168

and

(1.20)% (2.98)% (1.79)% (1.20)% (7.14)% (2.98)% (0.60)% (0.60)% (0.60)% (26.00)% (1.20)% (0.60)% (2.98)% (6.55)% (0.60)% (3.57)% (1.79)% (0.00)% (3.57)% (0.60)% (0.60)% (1.79)% (2.98)% (0.60)% (1.79)% (2.98)% (17.26)% (5.36)% (100.00)%

airway-related

General anaesthesia n = 168 ASA 1 2 3 4 5 Not recorded Age; years < 10 11–20 21–40 41–60 61–80 > 80 BMI (adult patients); kg.m ≤ 20 21–25 26–30 31–35 ≥ 36 Not recorded

39 88 34 6 – 1

emergence in 13 (7.7%) and during the recovery phase in 11 (6.5%) (Fig. 1). Nine reported cases (5.4%) had airway management-related events during two or more phases of anaesthesia. In 33 (19%) cases, the medical teams consulted another anaesthetist to assist in airway management. There were 29 (0.1%) cases of unexpected difficulties with tracheal intubation. A successful attempt was made 15 times with a videolaryngoscope (Glidescope, Verathon Medical Europe; C MAC Karl, Storz, Germany), four times with a flexible fibreoptic scope (Olympus LF-TP; Olympus Medical, Zoeterwoude, The Netherlands) and three times with a combination of a videolaryngoscope and fibrescope. Two emergency surgical airways were performed, and a different direct laryngoscope blade Macintosh 3 or 4 (Heine Medical) was used in five patients. A Frova Intubation Catheter (Cook Medical) was used in 44 (0.2%) patients during airway management. There were five (0.02%) cases of ‘cannot mask ventilate’. In one patient, cricoid puncture with a 16G Ravussin needle and Manujet (VBM Medizintechnik GmbH, Sulz, Germany) emergency oxygenation was used during resuscitation. This was a female patient who had acute airway obstruction postoperatively after a head and neck cancer diagnostic procedure and who had a difficult primary awake flexible fibreoptic intubation. Resuscitation with chest compressions and epinephrine because of hypoxic bradycardia and near-arrest and invasive tracheal access to the front of the neck was necessary. The number of peri-operative desaturations during induction or emergence detected with our background computer system correlated 100% with the number of

Timing of airway management related adverse events (n = 168)

30 4 21 54 52 7

140 120 100 80 60

2

11 53 32 22 11 6


(8.1%) (39.2%) (23.7%) (16.2%) (8.1%) (4.4%)

© 2016 The Association of Anaesthetists of Great Britain and Ireland

40 20 0 Inducon

Maintanance

Emergence

Postoperavely

Figure 1 Number and timing of airway-related adverse events. 45



peri-operative adverse events reported by the medical staff.

Discussion The incidence of severe airway management-related complications during anaesthetic care at our hospital was 0.86% and one patient died during the audit period. The incidence of a serious complication of airway management in our operating theatres was 1:117; unplanned ICU admission was 1:234, surgical airway was 1:1400 and death was 1:2803. A ‘cannot intubate cannot oxygenate’ situation occurred in 1:2803. A total of 168 airway management-related events occurred, which is an average of 2.5 per day. As well as a major cause of morbidity and mortality, these events are an economic burden for hospitals and should be prevented if possible. It is difficult to compare our data with other studies because no similar prospective studies have been done and every hospital has its own complication rate dependent on patient load and probably medical staffing and financial resources, but it appears to be quite high. It may be interesting for other hospitals to use our study format to compare their data. Our incidence of surgical airway was 1:1400. Our hospital has a large head and neck surgery department which has many complex cancer cases. This may explain why the incidence of surgical airway is this high. If we extrapolate this number, we estimate that we perform, on average, six emergency surgical airways in the operating theatres annually. Identification of problems can allow steps to be taken to improve safety. For example, we have interviewed all team members involved in the CICO head and neck cancer case resulting in death and use these interviews as clinical lesson for the department. This is the first time that we have prospectively collected patient data during airway management perioperatively. Interestingly, all doctors willingly participated and, after a few weeks, some started reporting events spontaneously before being contacted at the end of their shift. This has been observed before in other studies [1, 2]. Most airway events occurred during the induction phase of anaesthesia in our centre. It is also of note that most complications occurred in patients with a BMI > 28 kg.m 2. This is in concordance with other studies 46

[1, 2]. Children < 10 years of age and adults over 40 years old appear to be more at risk of adverse events. Bronchospasm and laryngospasm at extubation occurred relatively frequently, similar to previous reports [8]. In 19% of cases with adverse events during airway management, a colleague was consulted to help. This number is higher than reported in the NAP4 study. This may have prevented complications, although we have no evidence. At our centre, help will probably arrive within 3–5 min which we think is relatively fast because it is a busy academic hospital. We do not have data from other centres. Our clinical setting may be totally different from other hospitals. Other studies have defined oxygen desaturation as SpO2 < 90, but we wanted to find all potentially dangerous situations and, therefore, decided to use a cutoff at 93%. The strength of our study is characterised by its clear format and by the prospective nature of data collection, personal interviews and automated database desaturation computer program. We estimate that 95% of all events and problems were found, and we think that our study can be reproduced by any other hospital. Every hospital is unique, has its own patient caseload and procedures. Each hospital should be aware of its own complications before measures can be applied to improve clinical outcomes. Reproduction or extrapolation of the NAP4 data, which was performed in the UK, to the Netherlands and other countries would be difficult because of the following differences between the countries. Nurse anaesthetists play an important role in the Netherlands, allowing anaesthetists to supervise two operating rooms contemporaneously. This system differs from other countries and, therefore, the results of NAP 4 would also be difficult to extrapolate to our country. As a result of high medical costs in the Netherlands in addition to the high costs of such a nationwide research project, we think that NAP4 could not be easily repeated here. There have been attempts and plans to repeat NAP4 (personal communication, Dr Tim Cook, Consultant Anaesthetist Royal United Hospital, Bath; Joint Lead for NAP4) in various countries but, to our knowledge, no national initiatives have been taken thus far. In contrast to NAP4, where statistical analysis was inconclusive and it was deemed consistent with © 2016 The Association of Anaesthetists of Great Britain and Ireland


capturing all the cases or just a quarter of them because the distribution of the data met a Poisson distribution, we think that we have found 95% of all events at our hospital because of our investigation method combining interviews with a computer monitoring system. The NAP4 investigators indicated that they may have only found the ‘tip of the iceberg’ of all complications [1, 2]. Unlike a secure website where UK physicians could report their complications, our study protocol was based on a personal approach, contacting each physician on a daily basis, which probably led to our high response rate because it required minimal effort from clinicians and confidentially was assured. The daily interviews were short and not very time-consuming. It is known that voluntary reporting of complications is not very reliable and may miss many events [4]. Our study could be easily repeated in any department given the necessary manpower to collect and analyse the data. This study shows that airway management-related complications can be easily studied within a study period of two months when patient caseload and procedures are evenly distributed. It can be less burdensome in terms of financial costs and manpower than described by the NAP4 study. Such a study does depend on the compliance of medical staff and their willingness to share potentially sensitive information, emphasising the importance of processing data anonymously. This method offers a powerful audit tool for benchmarking and recognition of risk in our everyday medical practice, potentially improving patient safety and medical care. As we showed, it can be done in a cost- and time-effective manner (the medical student conducting the audit was not paid) and the findings of such a study may increase patient safety. For example, at our centre, some changes to daily practice were made directly after the study. After the case of an obstructed airway in the recovery ward, we developed an emergency drill for the management of future similar cases. With our data, we also have a tool to estimate how often videolaryngoscopes and other devices are being used for airway rescue in order to ensure equipment availability where most often needed. We also know how often colleagues are consulted during © 2016 The Association of Anaesthetists of Great Britain and Ireland


airway management problems and that knowledge gives us information how we can improve the system to ensure that help will arrive in time. This study did not include patients in the ICU or emergency room in need of airway management procedures as ICU is not part of the Department of Anesthesiology at VUmc and we decided not to include emergency room admissions because an electronic data capture system is not available and we would probably have missed many desaturations. It would, however, be very interesting to do a similar study in the ICU and other resuscitation sites [9–11]. Our intent was not to find risk factors for adverse events, as more clinical factors would have been needed for analysis and this group of patients with adverse events would need to be compared with a group of patients without adverse events. We did not study procedural sedation because these cases are not routinely undertaken by anaesthetists in the Netherlands. However, this would be an interesting area for future research, since many airwayrelated incidents occur outside the operating theatre [1]. Another limitation may have been the voluntary reporting approach, leading to accidental loss of information when doctors underestimate the significance of an event or just forget about it after a busy clinical day. We think that some events were under-reported, for example, if multiple events occurred on the same day in different patients. However, we think that all moderate and severe airway-related events were included. A student handled the interviews, so we anticipate that data about some events could have been more detailed if the interviewer had more medical experience. On the other hand, students are cheap to employ and can participate in such a project with the supervision of an experienced anaesthetist. We cannot comment on preventability, quality of care and contributory factors of airway-related events because this database is not large enough and that was not the primary goal of our study. Although this study is just a ‘snapshot’ of 8 weeks, our caseload is evenly distributed over the year so it should offer a sufficient and appropriate outline. This may not be true for other hospitals and a longer study period may well be needed. When we compared the first half to the second half of our study, no major changes in surgical patients were found. The surgical time for the different users of the operating rooms are 47



evenly distributed over the months at VUmc. We acknowledge the fact that we may have missed complications that occur very infrequently because of the short study period.

Acknowledgements We thank all medical staff of the Department of Anesthesiology of VUmc for participating in this study. This study was funded by departmental resources.

Competing interests JH has received airway management equipment from various companies for airway management courses and clinical studies. RG has received equipment from several companies for airway management courses and research. R.G. is the Chief Editor of Trends in Anaesthesia and Critical Care Journal and chairs the European Airway Management Society. Other authors have no conflict of interest to declare. JH is a member of the International Advisory Board of Anaesthesia and this article has therefore undergone additional external review.

References 1. Cook TM, Woodall N, Frerk C; Fourth National Audit Project. Major complications of airway management in the UK: results of the Fourth National Audit Project of the Royal College of Anaesthetists and the Difficult Airway Society. Part 1: Anaesthesia. British Journal of Anaesthesia 2011; 106: 617–31.

48

2. Cook TM, Woodall N, Harper J, Benger J. Major complications of airway management in the UK: result of the Fourth National Audit Project of the Royal College of Anaesthetists and the Difficult Airway Society. Part 2: intensive care and emergency departments. British Journal of Anaesthesia 2011; 106: 632–42. 3. Cook T, Behringer EC, Benger J. Airway management outside the operating room: hazardous and incompletely studies. Current Opinion in Anaesthesiology 2012; 25: 461–9. 4. Cook TM, Scott S, Mihai R. Litigation related to airway and respiratory complications of anaesthesia: an analysis of claims against the NHS in England. Anaesthesia 2010; 65: 556–63. 5. Cook TM, MacDougall-Davis SR. Complications and failure of airway management. British Journal of Anaesthesia 2012; 109: i68–85. 6. Apfelbaum JL, Hagberg CA, Caplan RA, et al. Practice guidelines for management of the difficult airway: an updated report by the American Society of Anesthesiologists Task Force on Management of the Difficult Airway. Anesthesiology 2013; 118: 251–70. 7. US Food and Drug Administration. http://www.fda.gov/ Safety/MedWatch/HowToReport/ucm053087.htm (accessed 05/01/2016). 8. Holzman RS, Mancuso TJ, Polaner DM. Practical Approach to Pediatric Anesthesia, 2nd revised edn. Alphen aan den Rijn: LWW/Lippincott Williams & Wilkins, 2015. 9. Schwartz DE, Matthay MA, Cohen NH. Death and other complications of emergency airway management in critically ill patients: a prospective investigation of 297 tracheal intubations. Anesthesiology 1995; 82: 367–76. 10. Jaber S, Amraoui J, Lefrant JY. Clinical practice and risk factors for immediate complications of endotracheal intubation in the intensive care unit: a prospective, multi-center study. Critical Care Medicine 2006; 34: 2355–61. 11. Simpson GD, Ross MJ, McKeown DW, Ray DC. Tracheal intubation in the critically ill: a multi-centre national study of practice and complications. British Journal of Anaesthesia 2012; 105: 792–9.

© 2016 The Association of Anaesthetists of Great Britain and Ireland