Introduction

The specialty of anesthesiology has a long history of improving patient safety. Much of this progress has been achieved by evaluating patient outcomes to determine risks to safety, and by deploying new strategies to counter these risks. Reviews of critical incidents (including those resulting in unintended harm during care), whether done locally or nationally, have led to the development and adoption of new technologies, protocols, clinical practice guidelines, and cognitive aids intended to improve care. Critical incidents can be identified by such means as national databases and registries, prospective time-limited large-scale data gathering (e.g., the National Audit Projects in the United Kingdom), smaller scale surveys, and institutional critical incident reviews.1,2,3,4 Critical incidents can also be captured by analyzing medicolegal data. The most expansive and authoritative ongoing analysis of closed civil legal cases in anesthesiology is the American Society of Anesthesiologists Closed Claims Project (ASACCP). Analyses of airway-related closed legal claims were published by the ASACCP in 1993, 2005, and 2019.5,6,7 Reviews of closed civil legal cases associated with airway management have also been performed in England and Canada.8,9

The Canadian Medical Protective Association (CMPA) is a not-for-profit mutual defense organization that represents over 95% of Canadian physicians. The mission of the CMPA is to protect the professional integrity of physicians and promote the safety of medical care in Canada. Unlike other medical malpractice insurance companies in the United States or Europe, the CMPA is not an insurance company. It provides a broad range of services to its members outside of the constraints of an insurance contract, which typically limits the scope of assistance offered to the policyholder and the indemnification that might be paid to the policy limit. The CMPA maintains a national repository of coded medicolegal cases relating to both civil legal actions and complaints to physician regulatory authorities (e.g., provincial Colleges of Physicians and Surgeons or territorial medical regulatory authorities). Ongoing analysis of these cases is integral to the multi-level education services and programs that the CMPA provides to its members.

The last CMPA airway-related anesthesiologist-based closed civil legal case report encompassed 1993-2003.9 Airway management in anesthesia practice has substantially changed since the last report. In particular, there is evidence that the further development and widespread deployment of new technologies such as video laryngoscopes and flexible bronchoscopes has had a positive impact on outcomes of care.3,10 As well, new iterations of guidance documents from specialty societies and interest groups have provided anesthesiologists with updated strategies to assess the patient airway and safely manage both unanticipated and anticipated difficulties.11,12

Given that the last study of Canadian airway-related anesthesiologist medicolegal cases preceded these advances, and to assess contemporary anesthesia practice, we analyzed the nature of closed airway-related medicolegal cases for the most recent ten-year period for which data were fully available. We conducted a retrospective descriptive analysis of airway-related anesthesiology civil legal cases that were closed in 2007-2016, looking for potential patterns of healthcare-related patient harm that could inform patient safety initiatives.

Methods

Study design and case selection

This study was a retrospective descriptive analysis of CMPA aggregate data derived from civil legal cases closed from 1 January 2007 to 31 December 2016, and which involved anesthesiologists, where airway management was the central focus of the case. At the CMPA, case closure indicates that a final medicolegal outcome was determined by the court or regulatory authority or by a mutual agreement between the parties to resolve the action. The most recent ten-year period for which data were available was chosen as the authors wanted to focus on contemporary anesthesia practice. These closed cases were in the public domain and are presented here in aggregated data form. The Advarra Institutional Review Board approved the study (Protocol #Pro00020829; most recently approved with amendments # MOD00587430, on 24 February 2020).

The files were searched according to the CMPA type of work, and only files that included anesthesiologists who were Fellows of the Royal College of Physicians of Canada (FRCPC) were included; family physician anesthetists were not included in this review. We identified files that included anesthesiologists who were CMPA members named in civil legal cases involving either the act of securing an airway, and/or a complication arising from airway management (e.g., respiratory distress/failure) in any clinical care location. To avoid misrepresenting the frequency of certain outcomes, class action cases were excluded.

Data repository and medicolegal coding

Canadian Medical Protective Association records are organized by “case”, representing instances in which a physician contacted the CMPA regarding a medicolegal matter involving that physician. Medical analysts (registered nurses extensively trained in medicolegal research) review each case and code specific clinical details using the Canadian enhancement to the International Statistical Classification of Diseases and Related Health Problems, 10th revision, Canada and the Canadian Classification of Health Interventions (CCI).13 Analysts use an in-house CMPA coding framework to categorize patient safety incidents and the contributing factors (categorized as provider-, team-, or system-related) based on peer expert opinions (see Appendix A for glossary of medicolegal terms).14

In this context, we defined peer experts as physicians retained by the parties in a legal action to interpret and provide their opinion on clinical, scientific, or technical issues surrounding the care provided. They are typically of similar training and experience as the physicians whose care they are reviewing. Severity of patient harm is determined using a modified version of the American Society for Healthcare Risk Management’s Healthcare Associated Preventable Harm Classification.15 This allows patient harm that is related to healthcare to be identified rather than patient harm from inherent risks of care or near misses, or no harm, as defined in Appendix A. For example, undetected esophageal intubation resulting in patient anoxic brain injury is a healthcare-related harm. In comparison, a case where a patient with end-stage emphysematous bullous disease who, despite being counselled preoperatively by the anesthesiologist on the risks of positive-pressure ventilation, nonetheless has a pneumothorax while undergoing appendectomy, would be an example of the inherent risk of care.

Data collection

Variables abstracted from the cases included patient characteristics such as age and American Society of Anesthesiologists (ASA) physical status. Details of patient care were also abstracted including type of airway management, and location and urgency of patient care. Location of care was divided into hospital versus non-hospital (e.g., freestanding clinic) locations. Hospital location was further divided into operative areas (preoperative, intraoperative, and postoperative locations) and non-operative areas (e.g., emergency department, critical care areas). In addition, the perioperative phase of care (i.e., pre-, intra-, or postoperative) when patient harm occurred was abstracted. Analysts also abstracted predictors of difficult airway management and degree of patient harm. The analysts sought out themes in peer expert criticisms and recorded the medicolegal outcome including any costs arising from court awards or settlements.

Statistical analysis

We report all variables with freqencies and proportions and calculated medians [interquartile range (IQR)] using SAS software, version 9.4 for all statistical analyses (SAS® Enterprise Guide® software, Version 9.4. Cary, NC, USA: SAS Institute Inc.; 2013).

Results

In the period from 2007 to 2016, the CMPA closed 406 legal cases involving FRCPC anesthesiologists. Of these, 46 (11%) cases involving 47 patients identified complications related to airway management (one case involved two patients). Anesthesiologists were identified as the only specialty involved in 30 (65%) cases. Residents or other trainees were involved in three cases.

Patient characteristics

The patient characteristics and clinical circumstances are detailed in Table 1. Twenty-six (57%) involved elective surgery. In 16 of 26 (62%) of these elective cases, patients either sustained brain injury or died. In 13 (50%) of these elective cases, patients were classified as ASA physical status I-II, 8 (31%) as ASA physical status > II, and five (19%) had no ASA assigned. Table 2 identifies the predictors of difficult airway as determined by CMPA analyst manual review of the health records. In 27 (59%) cases, no predictors were identified and/or no airway evaluation was documented.

Table 1 Patient and case characteristics for CMPA closed cases, 2007–2016 (n = 46 cases)
Full size table
Table 2 Difficult airway predictors identified in CMPA closed cases, 2007–2016 (n = 46 cases)
Full size table

Location of care and airway management events

Of the 46 cases, 41 (89%) had care provided in a hospital setting and five (11%) in an out-of-hospital setting. In 35 of the 41 (85%) hospital-setting cases, care was provided in an operative area. Specific airway management events (one component of care provided) took place in operative settings in 27 cases, critical care locations in eight cases, and non-critical care hospital locations in three cases. Some cases included more than one airway management event. Three of five out-of-hospital cases with airway management events occurred in an out-of-hospital clinical setting. Not all locations were clearly specified in the medicolegal record. The urgency of the airway event is reported, stratified by location, in Table 3.

Table 3 Urgency of airway by location for CMPA closed cases, 2007–2016 (n = 46 cases)
Full size table

Patient safety incidents

Analysts reviewing the files concluded that 33 of the 46 (72%) patient safety incidents resulted from the provision of healthcare, 12 (26%) were inherent risks of the care provided, and one patient outcome was unrelated to the healthcare experience and was a result of natural disease progression (Fig. 1).

Fig. 1
figure 1

Patient harm flowchart

Full size image

Peer expert criticisms of care provided

There were 260 peer experts who reviewed the 46 cases and offered assessments of the care provided. Peer experts were critical of some aspects of the care in the majority of cases (39/46, 85%); these included inappropriate management decisions and clinical judgement failure. Specific criticisms are detailed in Table 4. The most common deficit in clinical care cited by the peer experts was inadequate preoperative patient evaluation, including information gathering and failure to record an airway evaluation in 27 (59%) of the claims. Deviations from Canadian Anesthesiologists’ Society standards of care at the time that care was delivered (e.g., not using standard equipment such as pulse oximetry) were present in ten (22%) of the cases. These deficits were deemed by the peer experts to have negatively impacted the anesthesiologist’s ability to diagnose and develop alternative plans when difficulties with airway management were encountered. Communication issues were also identified by peer experts as contributing to poor patient outcomes and involved failed interactions between the principal care provider and other members of the team, as well as between trainees and supervising physicians (five cases) (Fig. 2).

Table 4 Peer expert and analyst-identified judgement failures for CMPA closed cases, 2007–2016 (n = 46 cases)
Full size table
Fig. 2
figure 2

Factors contributing to the 46 medicolegal cases involving airway management

Full size image

Concerns identified by peer experts arose during various phases of anesthesia care. Care during the pre-induction phase was most frequently criticised, with inadequate preoperative evaluation of risk factors such as difficult airway history or comorbid conditions (e.g., sleep apnea, obesity) and/or inadequate airway examination or documentation identified in 27 cases (59%). A complete lack of a documented preoperative airway examination was identified in ten (26%) of the operative cases.

In the induction, maintenance, and extubation phases, we identified peer expert criticisms in 20 of 39 cases (51%), whereas the standard of care was met in the remaining seven cases. Specific issues cited by peer experts during the induction and maintenance phases included failure to recognize esophageal intubation (three cases), failure to ensure endotracheal tube patency, failure to manage a deteriorating or difficult-to-ventilate patient, and not meeting the current Canadian Anesthesiologists’ Society standards for monitoring (e.g., oxygen saturation or carbon dioxide monitoring). Other concerns cited by peer experts included the anesthesiologist setting alarms to a low (or silent) mode and failure to use neuromuscular monitoring to verify the adequacy of recovery and/or reversal of muscle relaxants.

In the post-extubation phase of care, peer experts identified concerns in 11 of 39 (28%) cases, which occurred in a variety of care locations. Specific concerns included lack of pulse oximetry monitoring, failure to consider the impact of opioids on the potential for respiratory insufficiency/arrest, and a lack of documentation of specific patient monitoring or follow-up communication to guide subsequent care.

In both in-hospital (but non-operative) and out-of-hospital cases, peer expert concerns were related primarily to clinical decision-making, such as failure to use pulse oximetry or waveform capnography to confirm endotracheal tube placement as well as failure to diagnose issues with endotracheal tube placement and/or patency in patients who became difficult to ventilate/oxygenate.

Finally, in the three cases involving residents, peer experts criticised the lack of supervision by staff physicians, residents failing to anticipate problems, deficient skill sets for managing higher acuity patients, delays in recognizing a deteriorating clinical situation, and delays in communicating with staff physicians when difficulties were encountered.

Patient harm

Of the 46 patients, seven (15%) had mild harm (e.g., deficits in concentration), 14 (30%) had moderate to severe harm (of which 11 [24%] had anoxic brain damage—five [45%] elective cases, four [36%] urgent, and two [18%] unknown), and 24 (52%) died (19 [79%] deaths were related to airway management—11 [58%] elective cases, three [16%] urgent, and five [26%] emergent) (Table 5). In one instance, the patient outcome was considered to be unrelated to the healthcare provided and is not included here.

Table 5 Patient outcomes and payment amounts for CMPA closed cases, 2007–2016 (n = 46 cases)
Full size table

Medicolegal outcomes

The medicolegal outcome was favourable for the plaintiff (see Appendix A for glossary) in 27 (59%) cases with settlement in 26 of these and a judgement for the plaintiff in one. Eighteen (39%) cases were dismissed, and one (2%) judgement was for the anesthesiologist (Table 5). Two of the six remote (from the operative setting) cases had medicolegal outcomes that were favourable for the plaintiff. In four of five cases that involved an out-of-hospital setting, the anesthesiologist was assigned sole responsibility for the care and all four cases had an unfavourable outcome for the anesthesiologist.

Discussion

Patient care issues that were identified by peer expert reviewers in this series of closed civil legal cases can be categorized according to the phase of anesthesia care: preoperative assessment, induction and initial airway management, maintenance of anesthesia, and postoperative care. Inadequate or undocumented preoperative airway assessment occurred in the majority (59%) of cases. Although the utility of the preoperative airway exam in predicting difficulties with airway management has been scrutinized and found wanting, its most important role may be as a “cognitive-forcing” strategy to encourage the creation of an airway strategy for every patient.16,17,18 Although 59% of patients in this series did not have any documented predictors of anatomic airway difficulty, in many cases where airway assessment was documented, it was found to be lacking or incomplete; 41% of patients had predictors of potential airway difficulty. Lack of difficult airway anticipation and failure to consider the implications of identifiable risk factors for difficulty were common themes noted and criticized by peer experts reviewing the care provided.

Peer experts were also critical of specific decisions taken once difficulties were encountered with airway management. Perseveration with techniques that were failing to achieve the intended outcome, rather than pursuing alternate strategies, was identified in five cases (11%). Perseveration was also identified as an important factor leading to adverse patient outcomes in a quarter of the cases contained in the most recent report of the ASA Closed Claims Project (ASACCP) on the management of difficult tracheal intubation.7 Perseveration was defined by these authors as the consistent application of any airway management technique or tool for ≥ three attempts without deviation or change of technique, or the return to a technique or tool that had previously been unsuccessful. Perseveration is a judgement error related to loss of situational awareness in the potential setting of lack of alternative equipment or the skills to use alternative equipment.7 We cannot be certain in our series as to which of these, if any, explained the behaviours criticized by the peer experts. Failure to use a supraglottic device as a bridge for oxygenation was also cited by experts as inappropriate in one case in our series, and again was identified as the most common judgement failure in the recent report from the ASACCP (occurring in 26% of difficult airway claims).7 Finally, a reluctance to move to an emergency surgical airway when other strategies had failed occurred in two of our series’ cases. This is again consistent with the finding in the recent ASACCP report that, even when a decision to move to surgical airway was made, it was delayed 40% of the time.7 A similar delay to move to a surgical airway in the event of a cannot intubate, cannot oxygenate scenario was also noted in the report of the Fourth National Audit Project (NAP4) of the Royal College of Anaesthetists on major complications of airway management in the United Kingdom.2

The majority of the operative patients were ASA physical status I-II presenting for elective surgery. Peer expert review found that a delay or failure to recognize a deteriorating clinical situation and/or failure to initiate effective interventions in a timely fashion were common. Once a difficulty was encountered, perseveration with failing techniques without modification and non-use of recommended monitoring equipment was also common. Extubation is often difficult during anesthesia care. Poor anticipation of difficulties in the post-extubation phase of care, with no preparations in place to monitor the patient for deterioration or to manage difficulties, was identified in 11 of our reported cases.

A previous CMPA report published in 2005 analyzed airway-related closed civil legal cases between 1993 and 2003.9 Sixteen airway-related civil legal cases against anesthesiologists were found. These cases involved largely younger patients undergoing elective surgery: 13 (81%) of these patients suffered moderate to severe brain damage or death and case outcomes were commonly unfavourable to anesthesiologists. In this update, which includes 46 cases from 2007 to 2016, more than half (57%) of the cases involved elective surgeries, two-thirds of the operative patients were ASA physical status I-II patients. Moderate to severe injuries or death occurred in 82% of cases, and medicolegal outcome was unfavourable to the anesthesiologist in 59% of cases.

Analysis of closed civil legal cases across a number of domains, as well as large-scale incident reports such as the NAP4 project in the United Kingdom, show findings consistent with those noted in these reviews of closed CMPA civil legal cases.2,5,6,7,8 Poor patient outcomes are often related to failure to perform or document a preoperative airway evaluation and the failure to plan for failure. The clinical setting for many of these cases was a reasonably straight-forward elective clinical scenario rather than an extreme patient circumstance in which the applied technologies or guidance strategies failed.

We do not know how often some of the practice issues identified by the peer experts in this series occur during the daily provision of anesthesia care, but they might not be as uncommon as we would like to think. For example, failure to either conduct a preoperative airway assessment or to modify the management plan accordingly is a pattern of practice identified in multiple analyses of closed civil legal cases as well as in audits of practice.2,5,6,7,8 It is also inconsistent with recommendations offered by any and all authorities providing guidance for airway management in anesthesia practice. The phrase “normalization of deviance” was coined by Diane Vaughan to describe behaviours that deviate from safe practice but which may be supported by some rationale and may be commonly encountered.19,20,21 The behaviours might be justified as necessary in that they allow for the timely completion of tasks, such as moving the operative schedule along more efficiently by not pausing to prepare for plausible but low probability difficulties. The result is that safety boundaries may cease to be consciously recognized over time and, if negative consequences to these violations have not yet been experienced by the individual practitioner, practices may actually be deemed to be safe. The lack of bad outcomes can reinforce practices, because past “successes” negate the objective assessment of risk and what began as a deviation from safe practice eventually becomes “normalized” practice. If and when a poor outcome occurs, it may be regarded as the inevitable consequence of care rather than the result of risk-enhancing behaviours.22 Many of the behaviours identified in this review could be described by this construct and it may be that they persist in practice because practitioners typically get away with them. For example, it is acknowledged that, in many instances, airways that are deemed to be possibly difficult to manage after an assessment is done can turn out to be relatively easily dealt with. Thus, ignoring the results of a concerning assessment may have no consequences for a consecutive series of patients and may lead an anesthesiologist to conclude that doing so is a safe practice—until that strategy fails. It is possible to perform a surprisingly large number of interventions before a plausible adverse event occurs.23

Many of the errors identified by peer review in this analysis would also be classified as errors of situational awareness. Situational awareness in acute care medicine has been described as the ability to accurately perceive relevant information in a dynamic environment, comprehend its meaning, and create appropriate strategies to care for the patient.24 Conversely, situational awareness error has been defined as the failure to perceive relevant clinical information, failure to comprehend the meaning of available information, or failure to project, anticipate, or plan. Situational awareness errors are commonly identified in the analysis of adverse outcomes arising from airway management in both closed claims and audits. When the ASA Anesthesia Closed Claims Project database was analyzed to determine the prevalence of situational awareness errors in claims involving death and brain damage, airway and respiratory events were the most common category of clinical events associated with such errors.25 In multiple cases when difficult intubation or inadequate oxygenation or ventilation occurred, the anesthesiologists, while recognizing the difficulty encountered, did not adequately comprehend the seriousness of the clinical situation or recognize the deterioration of the patient’s condition.

There are important limitations to this review and analysis of closed civil legal cases assessing healthcare-related patient harm. Patients’ motivations for filing complaints or legal claims are complex, although some combination of poor outcome and dissatisfaction with care tends to predominate. It is unknown how many patients are harmed by the provision of healthcare; therefore, closed claims cannot be assumed to be an accurate sampling of healthcare-related patient harm, and there are no denominator data available to assess representativeness. There is typically a considerable time delay between an event happening and the case being filed and closed. Therefore, practices being assessed in closed civil legal cases may be historical in nature; experts might be challenged to appropriately assess care from the past, and current practice expectations may be imposed on temporally distant cases.26 There were 260 peer experts involved in these medicolegal cases and there may have been inter-expert variability in opinions that we did not measure or quantify because not all cases were reviewed by multiple experts. The high severity of injury common in many claims also leads to harsher expert reviews, but severity of injury is not necessarily correlated to the provision of substandard care.27 This hindsight bias may also be encountered in the peer expert case reviews, as more information may become available after the fact that may have influenced care decisions had it been available at the time of the patient encounter.26 There is evidence, however, that in the majority of instances where patients are compensated, both healthcare-related harm and negligent care were present.28 Finally, the documentation reviewed for this study was not collected for research purposes and as such was not always complete, which limited the ability to report on certain specific findings.

Two main conclusions may be drawn regarding this analysis of closed civil legal cases focused on the consequences of airway management in Canadian anesthesiology practice. First, healthcare-related patient harm often arose as a result of failure to manage relatively straight-forward clinical circumstances. Proceeding with care without adequately assessing the patients (including an airway exam), particularly in the setting of otherwise healthy patients presenting for elective surgery, was a common theme. Once airway management difficulties were encountered and the clinical situation began to deteriorate, there was a failure to either recognize or intervene early with an effective salvage strategy. This often resulted in considerable patient harm. Canadian guidance statements directed to the appropriate management of such events are published and their consistent implementation even in the face of well patients presenting for elective surgery is encouraged.11,12