Anaesthesia‐related equine fatalities: How good are we?

Safety is a routine concern in anaesthesiology as it has the potential to induce physiological changes that may lead to morbidity and mortality. As a result, it is commonly regarded as a high-risk activity. With accordance to most dictionaries, 'risk' implies future uncertainty about deviation from expected earnings or expected outcome. Risk measures the uncertainty that an investor is willing to take for a potential gain from an investment. In other words: How likely is it that the outcome will not meet my expectations – and can I (or my patient) live with this risk. Some pro- and retrospective, single- or multi-centre studies try to give an answer to this when it comes to equine anaesthesia. Anaesthesia is a medical specialty focused on risk management and patient safety, and the mortality risk attributable to anaesthesia itself has dropped over the last years. Although it is stated that the trends in the frequency of equine anaesthesia-related morbidity and mortality remain clear, the numbers reported in the literature are still controversial and show large variability. The reported wide range of perioperative mortality rates is probably due to the large variability in methodologies and differences in operational definitions and reporting sources, as well as a lack of appropriate risk stratification. One of the earliest reports of anaesthesia-related morbidity in horses from 1969 looked at mortality rates during and after elective and emergency procedures in horses. They reported a risk of 1.5% with more than 50% of the mortalities solely related to anaesthetic problems such as overdosing or other human errors 1. In this survey, Mitchell looked at the anaesthesia outcome of over 470 horses of various breeds, ages and body weights as well as the drugs and techniques used for premedication and anaesthesia. The purpose was to outline techniques and conditions associated with better outcomes and to provide an overview on fatalities and the current standing in equine anaesthesia. Almost 35 years later, Johnston reported a very similar mortality rate of 1.9% 2. The goal of this report was to 'document the equine perioperative mortality rate and to highlight any factor associated with an increased risk of death up to 7 days after anaesthesia' 2. In this multicentre study, data were collected from more than 40,000 cases over 6 years. Several potential contributors to the high risk of anaesthetic-related mortality included age, procedure, duration of surgery and the protocol chosen for specific cases. When looking at the data reported for emergency cases only, the reported numbers become even more inconsistent. While the mortality rate in horses undergoing anaesthesia for colic surgery decreased from 35.3% 3 and 31.4% 4 to 7.8% 2 within 20 years, an increase in mortality of 29.7% is described in a recent study from 2005 5. One problem with these data is that the number of analysed cases is much smaller compared with the reported risks for elective cases. Pascoe et al. 3 conducted a retrospective survey of 300 horses undergoing exploratory celiotomy for colic involving 341 procedures. In this single-centre study, survival was significantly influenced by the lesion, the length of the surgical procedure, and the preoperative packed cell volume and total plasma protein 3; all factors that even the best anaesthesiologist can hardly influence. In Mee's retrospective study evaluating 995 emergency equine general anaesthetics, the total mortality for emergency procedures was 31.4% 4. In this study, horses anaesthetised for surgical colic had a risk of death or euthanasia of 35.5% compared to horses anaesthetised for non-colic related problems with a total mortality of 15.3% 4. However, looking closer into these numbers, 157 of the included 792 colic cases were subjected to euthanasia during surgery due to the owner's request based on a poor prognosis. Therefore, the number of anaesthetised colic cases that died during general anaesthesia should be corrected to 34 out of the remaining 635 cases, changing the intraoperative fatality rate to 5.3% 4 which is much closer to the fatality rates reported by Johnston 2. Consistent with the results from the previous studies, Mair showed most recently that the short-term survival (discharge from hospital) of post-operative colic patients is mainly influenced by the underlying pathology, and that the most common reasons for death/euthanasia in the post-operative period after a single laparotomy were persistent pain/colic, post-operative ileus and grass sickness 5. It should be noted that the number of anaesthetised horses that were subjected to euthanasia during surgery is significantly higher in emergency cases compared with elective cases. Although unstable cardiovascular conditions might contribute to higher fatality rates during anaesthesia in these cases, the number of horses subjected to euthanasia due to a poor prognosis because of the underlying disease must be considered. There is no doubt that veterinary anaesthesia has advanced considerably over the past 50 years, but why do we not see a clear down-trending of anaesthesia-related mortality rates as reported in human medicine? Our human colleagues were able to reduce the mortality risk attributed to anaesthesia itself from about one death in 1000 anaesthetic events in the 1940s, to one death in 100,000 anaesthetic events in the early 2000s 6, 7. Also, why is there so much variation between the different studies evaluating risks for equine anaesthesia when all we are looking for are simple facts? One reason is that there is no consistency in data recording or inclusion/exclusion criteria between the different studies. The definitions for death in which anaesthesia was the primary or contributing cause varied widely, as did the time-frame for the considered perioperative period. The perioperative period has been defined as intraoperative only or intraoperative and recovery from anaesthesia, up to multiple days after recovery or even until hospital discharge. Another reason for the variation is that most reports do not distinguish between the perioperative mortality incidence and the anaesthesia-related incidence. When some of these reports did so, a positive trend was noticed. A retrospective single-centre study on equine fatalities including horses that were subjected to euthanasia or died within an undefined post-operative period 1. The fatalities directly related to anaesthesia were about 50% of all fatalities, including cardiac arrest during or after anaesthesia as well as fatal fractures or myopathies during the recovery period 1. Although cardiac arrest and death rarely occur during the perioperative period, these events represent the worst patient outcomes that are directly related to the anaesthetic event. In newer studies, anaesthesia was blamed for less than 15% of all fatalities 4. This decrease has been attributed to a variety of safety improvements including better monitoring techniques, the development and widespread adoption of practice guidelines, and other systematic approaches to error reduction. Even with these improvements, 70% of equine fatalities occur during anaesthetic recovery and are mainly due to catastrophic fractures, myopathies and neuropathies. Compared with other species, horses often have unpredictable recoveries that can lead to dramatic injuries. In the end, what can we learn from these reports? These studies help document the anaesthetic mortality rate, highlight factors associated with an increased risk of death and help to identify major and minor risks, risk groups, and maybe even events contributing to risks and fatalities. Unfortunately, they usually fail to identify causes, and remedy of these risks can sometimes lead to misleading interpretations. Many reports identify the age of the horse as a risk factor (very young foals or old horses). Assuming that age-related coexisting diseases might play an important role 2, we need to keep in mind that most elective procedures (arthroscopies and castrations) are also performed in younger horses. Another example is that different drug protocols are regularly associated with changes in risk. Acepromazine, when used as a premedication in horses, is useful from a pharmacodynamic standpoint, but this drug is rarely used in anaemic horses or colic cases which also have a higher complication rate per se. What can we do in the future to further improve safety in equine anaesthesia? As pointed out by Hartnack and colleagues, a critical incidence reporting system in equine anaesthesia might serve as a practical tool to improve safety in addition to retrospective mortality studies 8. Critical incident reporting systems refer to the structured reporting, collation and analysis of incidents. These incident investigations can be performed systematically and can also lead to the development of risk grading techniques. This would allow us to determine the impact of an incident and categorically define what constitutes a minor event all the way up to a serious incident. In the end we need to ask ourselves: How good are we and are we good enough? Why are our fatality rates so high and what is necessary to achieve more clear improvement? Maybe it is also time to blame surgery as Irwin and Kong pointed out so well: Anaesthesia itself might be relatively safe these days, but surgery is not 9.