4: Equipment and Technical Errors in Veterinary Anesthesia

Equipment and Technical Errors in Veterinary Anesthesia

Knowledge rests not upon truth alone, but upon error also.

Carl Gustav Jung (1875–1961)

Equipment and our interactions with it play a central role in modern anesthetic practice. Anesthetic machines and vaporizers provide measured flows of anesthetic gases and vapors while airway equipment and breathing systems allow these gases to be transported to and from the patient. Ventilators support respiration whilst pumps and syringe drivers administer intravenous fluids and drugs. Electronic monitoring equipment gives measurements and readings that provide information on the patient’s physiological function.

Equipment can be even more fundamental. For example, rarely do we think of catheters as equipment; more often they are thought of as supply items in part because they are disposable. Nonetheless, catheters used for intravascular cannulation are an important feature of anesthetic management and, although they have the potential for great good, they also have the potential to cause harm (Hofmeister et al. 2014; Singleton et al. 2005). Consequently, the proper use of catheters requires eye-hand coordination, knowledge of how the catheter is designed and meant to be used, and knowledge of the patient’s anatomy. Knowledge of the hazards associated with intravascular catheters is gained through education and training, the same elements that make it possible for us to use other pieces of anesthetic equipment.

So the roles of equipment in anesthesia are legion, but as technology and technical skills have become more and more integrated into the process of anesthesia the potential for misuse, malfunction, and failure of each piece of equipment remains. But how often does equipment actually malfunction or fail? Before answering that question we should first try to define what equipment failure is. This sounds like a simple task, but as Webb has stated:

It is difficult to define “true” equipment failure as almost every aspect of equipment design, manufacture, supply, installation, maintenance, testing and use involves humans and thus anything which goes wrong has the potential to be due to a human error of some sort.

Webb et al. (1993).

For the purposes of this chapter we have basically considered equipment error as errors that center around the use, misuse, or malfunction of a piece of equipment (anything that is not a person or patient).

Even when ignoring the lack of a clear-cut definition, the frequency with which equipment or technical error occurs in anesthetic practice is not clear even in human anesthesia; like many error-related issues it is under-reported. In one retrospective study of 83,000 anesthetics performed over a period of 4 years (1996–2000) in a Norwegian hospital, the incidence of reported “equipment or technical” problems was 0.23% for general anesthetics and 0.05% for locoregional anesthetics (with a total of 157 problems being reported) (Fasting & Gisvold 2002). Most of these problems were considered “trivial,” having little effect on the patients or their care, but almost 30% (45/157) caused some harm to patients, for example, a period of hypoxemia, hypercapnia, or hypoperfusion. None of the problems were considered to have caused lasting harm. About one-third (49/157) of the problems were associated with anesthetic machines, and in about one-quarter (40/157) of these events “human error” was considered a causal factor, and almost half of these events were associated with anesthetists not adequately following pre-anesthetic checks (Fasting & Gisvold 2002).

Other studies have generally analyzed incident reports but could not give an estimate of incidence due to the lack of a denominator. Cooper used a modified critical incident technique (see “Critical incident technique (CIT)” in Chapter 3) to gather reports of human error and equipment “failure” from 139 anesthesia providers (anesthesiologists, residents, and nurse-anesthetists) (Cooper et al. 1984). Of the 1089 descriptions of “preventable incidents” that were collected, only 11% represented true equipment failure while another 13% involved disconnection of the patient from the breathing system or disconnection of the breathing system from the anesthetic machine. Table 4.1 shows the distribution of equipment failures based upon the type of equipment used. Interestingly, Cooper found that equipment failure was less likely to be involved in an adverse patient outcome than human error. This led to him to conclude “perhaps people have more difficulty detecting their own errors than failures of their equipment” (Cooper et al. 1984).

Table 4.1 Distribution of equipment failures according to type of equipment involved.

Number of incidents Percentage of all equipment failures
(rounded to whole numbers)
Breathing circuit
Monitoring device
Anesthesia machine
Airway device

From: Cooper, J.B., et al. (1984) An analysis of major errors and equipment failures in anesthesia management: considerations for prevention and detection. Anesthesiology 60(1): 34–42. Reprinted with permission of the publisher.

Webb et al. (1993) analyzed and reported on equipment failures identified in the first 2000 incident reports submitted to the Australian Incident Monitoring System (AIMS) (see Chapter 3). This yielded 177 incidents of equipment failure (just under 9% of the incidents reported). Problems associated with failure of unidirectional valves (46 in total), monitoring equipment (42), and ventilators (32) were the most commonly reported equipment failures. Of these 177 incidents 97 (55%) were considered to be potentially life threatening, with 62 detectable by standard anesthetic monitoring.

More recently Cassidy et al. (2011) reported on anesthetic equipment incidents reported to the UK National Health Service’s National Reporting and Learning System between the years 1996 and 2000. Of the 195,812 incidents reported from the anesthetic and surgical specialties, 1029 incidents of anesthetic equipment failure were identified. Of these about 40% (410) were associated with monitoring equipment, 18% (185) with ventilators, 10% (99) with leaks, and 5% (53) associated with fluid pumps. The large majority of incidents (89%) did not cause patient harm, but 2.9% (30 incidents) led to moderate or severe harm. Most reports were associated with equipment faults or failure, but a small proportion were clearly or most likely the result of user error. Unfamiliarity with equipment, failure to follow checklists, and failure to act on reports of temperamental equipment were recurrently cited causal factors. It is worth noting that an additional 215 airway equipment reports were identified but not analyzed.

The most recent assessment of equipment failure in the United States was published by the American Society of Anesthesiologists’ Closed Claim Project. Mehta et al. (2013) reviewed just over 6000 closed claim reports associated with anesthesia care that were filed between 1970 and 2011, and more specifically those reports associated with anesthesia gas delivery equipment. This subset of cases was analyzed further and classified as primarily due to: (1) equipment failure (unexpected failure despite routine maintenance and previous uneventful use); (2) provider error (faults associated with maintenance, preparation or deployment of a device); or (3) failure to follow appropriate pre-anesthesia check-out procedures (faults that would have been detected had procedures been adhered to). One hundred and fifteen claims were identified, with 80% of those occurring between 1990 and 2011, and involving vaporizers, supplemental oxygen delivery equipment, and breathing systems. True equipment failure occurred in only 5% of cases, of which one-third were considered preventable had pre-anesthesia check-outs been properly performed (Mehta et al. 2013).The remaining significant issues were attributed to provider error, including inadequate setting of alarms, improvised oxygen delivery systems, and misdiagnosis or mistreatment of breathing circuit events.

So what about in veterinary anesthesia? Unfortunately the picture is likely to be far worse. A number of factors are likely to increase the incidence of equipment and technical error in veterinary anesthesia, including much less education and training compared to human anesthesia, less stringent procedural guidelines (such as pre-anesthesia checks), lack of standardization for anesthesia equipment such as anesthesia machines, and lack of policies regarding maintenance and servicing of equipment. The one mitigating factor may be reduced complexity of equipment in the veterinary sector and potentially a lower reliance on technology, especially in the general practice arena.

What follows are some examples of equipment error in veterinary anesthesia.


Case 4.1

A 6-year-old gelding weighing 514 kg was brought to a referral center for laryngeal surgery to correct airway problems that were causing poor performance. On the day of surgery the horse’s vital signs were: heart and respiratory rates 36 beats per minute and 16 breaths per minute, respectively; capillary refill time of less than 2 seconds; and moist and pink mucous membranes. Rectal temperature was 38 °C, hematocrit was 35%, and total protein was 62 g L−1. Physical examination was unremarkable and all other blood work was within normal limits for the referral hospital’s clinical laboratory.

While the horse was in its stall, the anesthetist inserted a catheter into its left jugular vein and secured it in position. Thirty minutes later the horse was walked to the anesthesia induction area where its mouth was rinsed with water to flush out food debris in preparation for orotracheal intubation. The horse was then injected with detomidine (3 mg) via the jugular vein catheter. As soon as the catheter was flushed with heparinized saline the horse stumbled, dropped to its knees, and then quickly stood and shook its head.

Thirteen minutes later, during which the horse seemed normal for a sedated horse, it was injected with diazepam (15 mg, intravenously) followed 8 minutes later by ketamine (1.2 g, intravenously). The induction was described as rough in that the horse fell suddenly and atypically to the floor. It was intubated (using a 30-mm internal diameter endotracheal tube) then hoisted by his legs onto the surgical table. The endotracheal tube was attached to a large animal anesthesia machine, and mechanical ventilation was initiated (7 breaths min−1, tidal volume 7 L) delivering halothane (3%) in oxygen (7 L min−1). At this time the anesthetist considered the possibility that the catheter was in the carotid artery and that all drugs had been inadvertently injected into it. As a consequence no additional injections were made through that catheter and another catheter was inserted into the right jugular vein through which all subsequent drugs and fluids were administered.

Throughout the course of anesthesia the horse’s heart rate ranged between 32 and 35 beats per minute and mean arterial blood pressure ranged between 60 and 80 mmHg. Results of arterial blood gas analysis at 30 and 60 minutes after induction were acceptable for an anesthetized horse in lateral recumbency. Sixty-seven minutes after induction, an infusion of dobutamine was started to treat arterial hypotension (mean arterial blood pressure 65 mmHg). After 95 minutes of anesthesia, all monitoring devices were disconnected from the horse and he was turned from lateral recumbency to dorsal recumbency for the last stage of the surgical procedure. During repositioning the horse was administered ketamine to maintain anesthesia (two 200-mg boluses). Thirty minutes after being positioned in dorsal recumbency the horse was extubated for 10 minutes so as to facilitate surgical exploration of the larynx. During this time anesthesia was again augmented with ketamine (one 400-mg bolus), and ventilation was assisted intermittently as the surgical procedure allowed. At the end of surgery the horse was moved to a recovery stall where positive pressure ventilation was continued with a Hudson demand valve until the horse started breathing spontaneously.

By 30 minutes after discontinuing anesthesia the horse started to breathe spontaneously, but recovery seemed slow for the type of anesthesia and surgery. After 70 minutes in the recovery stall the horse started to show signs of seizure activity with extensor rigidity. The horse was re-intubated and positive pressure ventilation was commenced with the Hudson demand valve. Although arterial blood gas analysis indicated normocarbemia (PaCO2 44 mmHg), the horse was hypoxemic (PaO2 47 mmHg). Diazepam was administered to control seizure activity, but its effect lasted only 10 to 20 minutes. It was decided to induce anesthesia using thiamylal (2 g) in glycerol guaifenesin (GG; 5% solution) administered to effect to stop seizure activity, and then maintain anesthesia and control seizure activity with pentobarbital (3.8 g, intravenously). Furosemide was administered to reduce cerebral edema, and dimethyl sulfoxide (DMSO) was given for its anti-inflammatory and oxygen radical scavenging effects. Despite these efforts the horse’s condition deteriorated over time. After 13 hours in the recovery stall the horse was euthanized at the owner’s request.

At necropsy a 2 × 2 cm area of yellow discoloration on the surface of the left occipital cortex was found as well as diffuse vascular congestion of the left cerebral hemisphere. A cross-section of the brain revealed bilateral yellow discoloration and malacia of the entire hippocampus and yellow discoloration of the deeper aspects of the left occipital cortex in the region supplied by the caudal cerebral artery (Figure 4.1).

Image described by caption.

Figure 4.1 Cross-section of the horse’s brain at necropsy showing injury to the left occipital lobe after accidental injection into the left carotid artery of detomidine as a premedicant and ketamine plus diazepam for induction of anesthesia. Black arrows indicate the boundaries of the lesion consisting of discoloration and malacia of the entire hippocampus and discoloration of the deeper aspects of the left occipital cortex.

Initial analysis of the case

Upon initial analysis this case seemed to be primarily an example of a skill-based error in that the act of catheterizing a horse relates to technical performance and proper execution of the task. Such errors suggest inadequate training and experience in performing this type of procedure. However, focusing only on the anesthetist ignores other factors that may have contributed to this error and warrant deeper investigation. Indeed, subsequent assessment of this case revealed that the horse was catheterized under conditions that would have made this task challenging for anyone regardless of training and experience.

The stall was dimly lit, which made it difficult for the anesthetist to see the jugular vein and the color of blood flowing out of the catheter. The anesthetist, in trying to comply with the surgeon’s request that catheters be inserted into the jugular vein as far from the surgical site as possible, inserted it close to the thoracic inlet, a location that imposed some anatomic constraints on catheterization. Inserting the catheter with its tip directed cranially also made it difficult to distinguish venous blood flow from arterial blood flow. These latter two factors posed two challenges for the anesthetist: the hand holding the catheter was forced by the point of the horse’s shoulder to direct the catheter at an angle that made it more likely it would be inserted through the jugular vein and into the carotid artery. Furthermore, inserting the catheter parallel to and in the direction of carotid arterial blood flow increased the likelihood that arterial blood would not pulse out of the catheter thus making it appear more like venous blood flow and giving a false sense of having inserted the catheter into the jugular vein.

This case is also an example of a rule-based error in that there was misapplication of a rule that went something like: “when anesthetizing a horse for surgery of the larynx, insert the catheter at a location on the jugular vein that is far from the surgical site.” But there was no matching of the conditional prerequisite—if —with the action portion—then—of the rule that goes something like: “if a drug is injected into a horse’s jugular vein and the horse immediately shows central nervous system (CNS) signs, e.g., stumbles or collapses, then one must suspect the injection was into the carotid artery and not the jugular vein.” This latter rule, had it even been made explicit in the first place and properly followed, would have forced the operator to check the location of the catheter. Eventually this check did happen, but only after the induction drugs had been injected through the catheter and caused a very rough, atypical induction. All of this suggests that the anesthetist was solely responsible for this error, but as usual there are other factors to consider.

Investigation of the case

For this particular type of laryngeal surgery the surgeons had a standing order that all catheters inserted into the jugular vein were to be inserted as far distally as possible on the horse’s neck so as not to interfere with the surgical site; this was a departure from the standard catheterization practice that was in effect at this equine hospital. The usual practice was to shave hair from the junction of the upper and middle thirds of the cervical jugular vein, wash the area until clean, and then insert the catheter with its tip directed toward the heart (Box 4.1).

Aug 14, 2022 | Posted by in SUGERY, ORTHOPEDICS & ANESTHESIA | Comments Off on 4: Equipment and Technical Errors in Veterinary Anesthesia
Premium Wordpress Themes by UFO Themes