5: Medication Errors in Veterinary Anesthesia

Medication Errors in Veterinary Anesthesia

…systemic medical errors are hard to stop because their genesis is hard to spot.

Karl E. Weick (2002)

Medication errors are among the most commonly encountered errors in human medicine, with reports suggesting they may occur in a staggering 2–14% of all hospital admissions (Cullen et al. 2000; Leape 1994; Schiff & Leape 2012). They can occur in many guises and involve any part of the prescription process including choosing the drug itself, calculating the dose, writing up orders, and administering the drug. In fact preparing and administering a drug, such as an intravenous injection as part of an anaesthetic, is a surprisingly complex task involving some 30 or more steps (Woods 2005). Recognizing this complexity makes it easy to see how, when an anesthetist is under pressure or is distracted, a significant error may occur (Woods 2005).

In simple terms medication errors can be assigned to the following categories: “wrong patient,” “wrong drug,” “wrong dose,” “wrong route,” and “wrong time”. However, these are just labels for a “technical error” made by an individual. By highlighting in this manner only the sharp end of the incident ignores the many system-associated conditions that set the stage for the medication error.

Literature from human medicine has provided us with most of what we know about medication errors and their root causes. But what about in veterinary medicine? In truth we really do not know the extent of the problem in veterinary medicine because there are so few data available to us (Alcott & Wong 2010; Mellanby & Herrtage 2004). Reports on medication errors have been published in a variety of journals, involving a variety of species (Alcott & Wong 2010; Kaplan et al. 2011; Kennedy & Smith 2014; Love et al. 2011; McClanahan et al. 1998; Means 2002; Paul et al. 2008; Piperisova et al. 2009; Smith et al. 1999; Wells et al. 2014). These articles cite the errors but their underlying causes often are not identified and the incidence in the larger patient population is unknown. In one of the only published studies to include data on the incidence of medication errors in veterinary practice, Hofmeister et al. found that medication error was reported in 1.2% of anesthetics prior to instituting specific error reduction strategies (Hofmeister et al. 2014). In that study, none of the patients died and there was little if any obvious harm to the patients, but the potential for harm was real. Unpublished data collected by the Cornell University Veterinary Medical Teaching Hospital revealed that 41 medication errors were reported over a 6-month period (see “Which incidents and errors should be analyzed?” in Chapter 3). Both of these examples involve teaching hospitals with unique environments, so the errors may seem of little relevance to general private practice. The environment of a private practice does differ from that of a teaching hospital, but despite the differences the latent conditions for medication errors are present in both. Indeed, the results of a study involving private practices for which claims were submitted to the leading veterinary indemnity insurer in the UK between January 2009 and December 2013, indicated that drug-related errors were common; the most common types of error within this category were due to incorrect choice of drug and overdose (Oxtoby et al. 2015). Some of the latent conditions that set the stage for medication errors are highlighted in the following cases and near miss vignettes.


Case 5.1

As preamble to the following case, we recognize that the processes described in this case are slightly historical in nature. However, we present it because it shows how system-related factors can and do set the stage for errors. The exact error described is unlikely to occur in veterinary practice today, but even now we veterinarians make modifications to processes in order to make them more efficient or economical––we describe a number of them throughout this book––and in doing so we unintentionally set in place error-generating conditions. This case also demonstrates how, if actions had been directed toward the person who seemingly caused the error as a solution to the problem, future error prevention would not have been achieved; the latent conditions that enabled the error to occur in the first place would still be present and set the system up to fail again in the future.

On a Tuesday morning a young, healthy bitch weighing 7 kg is scheduled for an ovariohysterectomy. To facilitate intravenous catheterization and induction of anesthesia she was administered acepromazine, oxymorphone, and atropine, all injected intramuscularly. Thirty minutes later she was catheterized intravenously and then anesthesia was induced with thiamylal. The catheter was flushed with heparinized saline, the trachea intubated, and the endotracheal tube was attached to a circle system to which only oxygen was delivered. A check of the patient at this time could not detect a peripheral pulse nor could heart sounds be auscultated. Cardiopulmonary resuscitation (CPR) was started promptly, but the dog did not respond to standard resuscitation efforts. An assessment of the anesthesia procedure did not reveal any clues as to the cause of death nor did a necropsy later that day.

The following day in the early afternoon a 13-kg dog scheduled for an ophthalmic procedure had a cardiac arrest several minutes after induction of anesthesia. Cardiopulmonary resuscitation was ineffective. Drug doses and procedures were in accordance with the practice’s standard operating guidelines. Over the next 2 hours, two dogs and a cat were uneventfully anesthetized, but another dog weighing 6 kg and scheduled for abdominal exploratory surgery, had a cardiac arrest after intravenous catheterization and before induction of anesthesia. This dog, too, could not be resuscitated. The anesthetist reported that the dog was appropriately sedated after premedication, that its oral mucous membranes were pink and moist with a normal capillary refill time, and that its pulse was within normal parameters for rate and rhythm. According to the anesthetist the arrest occurred after the intravenous catheter had been inserted and flushed with heparinized saline.

Investigation of the incident

Inspection of the two 250 mL bottles of heparinized saline being used in the induction area revealed that, despite both bottles being clearly labeled as heparinized saline, the label on one bottle covered another label. The original label indicated that the bottle contained a potassium chloride solution (4 mmol mL−1); the second label indicated the technician who had added heparin to the bottle.

Analysis of the incident

There are several questions that immediately spring to mind in this case. Was the technician responsible for the error? Was the organization—the practice––in any way responsible for the error? How could this error be prevented in the future? However, before answering these questions, contextual details are needed.

All fluids used in this practice, on both the small and large animal sides of the hospital, were made in-house at a central location because the owners believed the size of the hospital made this a cost-effective practice. For those with direct patient care responsibilities it was never clear from week to week who in the central production facility actually produced the fluids or filled orders that were submitted by the various hospital sections. Fluid volumes of one liter or less were dispensed in glass bottles with rubber stoppers and were distributed from the central location in cardboard boxes.

Every Monday and Friday, late in the afternoon, technicians throughout the hospital restocked shelves with various supplies that had been ordered earlier in the morning, including fluids. Those technicians working in sections that used heparinized saline added heparin to bottles containing 250 mL of normal saline (0.9% NaCl) to make a saline solution containing 4 units mL−1 of heparin. Another label was then affixed to the bottle indicating that heparin had been added, the date it was added, and by whom.

On any given day, the anesthesia service was responsible for anesthetizing 12 to 19 small animal patients, of which 80% were dogs, 15% were cats, and the remainder were birds, pocket pets, and reptiles. In this practice the anesthetic technique and drugs chosen for anesthesia were selected based on each patient’s health (physical status) and the procedure to be performed. In general, anesthesia consisted of premedicating patients prior to intravenous catheterization. After the premedication drugs had taken effect a catheter was inserted into a peripheral vein, usually a cephalic vein, the catheter was capped with an injection cap, and the catheter flushed with sterile heparinized saline usually administered with a 6- or 12-mL syringe.

Prior to induction of anesthesia, each patient’s heart was checked to ascertain rate and rhythm, and mucous membranes were assessed as to color and capillary refill time. Depending on the health status of each patient, anesthesia was induced with an injectable drug administered to achieve a plane of anesthesia suitable for orotracheal intubation. After intubating and securing the endotracheal tube, it would be attached to a breathing circuit, and a flow of oxygen started. At this point in the induction process the anesthetist would again check the patient’s heart rate and rhythm. If the patient’s condition was as expected, the anesthetist would turn on the vaporizer so as to start the delivery of inhalant anesthetic. The intravenous catheter was often flushed again at this time and then fluids were started to maintain intravascular volume during anesthesia.

Returning to the specific analysis of this case: the organization’s operating procedures, specifically those for the production and delivery of fluids, created latent conditions that set the technician up as the final common pathway for this error. To the technician’s credit he fully recognized his role in the error and accepted responsibility for it; there was no equivocation or attempt to place responsibility elsewhere. Of crucial importance, the technician explained how the error occurred.

Normal saline was the only fluid that had been ordered on the Monday morning prior to this incident. The technician picked up the order in its cardboard box at the central fluid production facility and took it to the anesthesia induction room. Here, as a matter of routine, every Monday and Friday afternoon, in the same location and at about the same time of day, the anesthesia technicians added heparin to bottles of normal saline to make heparinized saline. Concentrated potassium chloride was also kept in this location, but it had not been ordered that morning and the technician did not expect it to be in the box. That Monday afternoon the two anesthesia technicians were working together to make heparinized saline, but they were having a conversation about an issue unrelated to the task at hand. This scenario describes a skill-based error due to a slip, specifically a capture slip in that the technician’s attention was focused elsewhere and there was a failure of a timely check to make sure it was normal saline to which heparin was being added. It was also assumed that the box of fluids contained only normal saline because no other types of fluids had been ordered that morning. Adding heparin to normal saline was also a task performed routinely in the same familiar location twice a week.

There was no evidence that the technician intended to cause harm and the facts clearly showed that there were multiple factors that contributed to this error. In addition, the technician did not have a prior history of violations and he was intensely remorseful (see “Analysis of the person(s) at the sharp end: accountability” in Chapter 3 and Figure 3.5). Yes, the technician was responsible for the final step in the pathway to this error, but mitigating circumstances, especially the manner in which the practice made, labeled, and distributed fluids in-house (more specifically concentrated potassium chloride), and the failure at the central fluid production facility to correctly fill fluid orders, pointed to a failure of the system.

The practice owners acknowledged the mitigating factors inherent in the practice’s standard operating procedures, and recognized that any person tasked with the responsibility of making up heparinized saline within the environment of this practice, was just as likely to commit the same error. No one could predict when or how such an error would occur again, but the traps—the latent conditions—were all in place waiting to generate another fluid-related error. These facts demanded a closer look at the practice’s protocols and procedures concerning its fluid production and distribution processes.

As a consequence of this error, an evaluation of the system was undertaken. One immediate decision was to stop making and bottling potassium chloride solution and purchase it commercially so that there would be no chance of confusing it with normal saline. The commercial potassium chloride vials were then stocked in wards or service areas separate from all other fluids. In addition, the distinct packaging of the vials also helped prevent them from being confused with other fluids in use in the practice. For a short time the practice continued to produce normal saline and lactated Ringer’s solution, especially for use in large animal patients. However, as a result of this series of adverse incidents, and because of other complaints about in-house fluids (complaints that had been ignored up to then but that were becoming more prevalent) the practice pursued an in-depth analysis of the fluid production system itself. This found that there was a lack of quality control at a number of steps in the production process. Soon thereafter the practice switched to stocking only commercially produced fluids.

A cautionary note is necessary here. Switching from in-house produced fluids to those produced commercially does not assure safe practices or patient safety. Pharmaceutical houses, in an effort to develop brand identity and market loyalty, frequently use product labeling as a marketing strategy to achieve those goals. It is not unusual for a manufacturer to produce two very different classes of drugs and label them similarly. As a result, solutions such as potassium chloride continue to be administered to human patients when the injection of other drugs was intended (Charpiat et al. 2016). Figure 5.1 graphically demonstrates how similar packaging of three different types of fluids used for flushing intravenous lines or reconstituting antibiotics, can increase the likelihood that at some unknown point in time the wrong solution will be administered to a patient, possibly with fatal consequences (Lankshear et al. 2005).

Photos of three similarly packed fluids: sodium chloride (left), water for injections (middle), and sterile potassium chloride concentrate (right).

Figure 5.1 Similar packaging of various fluids used for flushing IV lines or reconstituting antibiotics, increases the likelihood that at some unknown point in time the wrong solution will be administered to a patient, possibly with fatal consequences.

From: Lankshear, A.J., et al. (2005) Quality and Safety in Health Care 14(3): 196–201. With permission of the publisher.

As already mentioned, but worth repeating, had the practice approached the error as a technician-only problem and reprimanded or terminated him as the solution to the problem, there would almost certainly have been more fluid-associated patient harms or deaths at some point in the future, at a time no one could predict. However, by taking a systems approach to the problem the practice eventually discovered the root causes of this and other fluid-related problems and effectively solved the latent problems that placed all patients at risk of harm.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Aug 14, 2022 | Posted by in SUGERY, ORTHOPEDICS & ANESTHESIA | Comments Off on 5: Medication Errors in Veterinary Anesthesia

Full access? Get Clinical Tree

Get Clinical Tree app for offline access