Case 2021-4: Measure Twice, Cut Once

Consensus Recommendations for Improving Medication Safety in the Operating Room (APSF Newsletter 2010;25).

Consensus Recommendations for Improving Medication Safety in the Operating Room (APSF Newsletter 2010;25).

Laboring patient received epidural for labor analgesia. After epidural placement and dosing, patient became hypotensive. Ephedrine and NeoSynephrine were then used to treat the hypotension. After NeoSynephrine administration, the patient became hypertensive and was reporting chest and arm pain and headache. Also, during this time period fetal distress was noted and patient underwent emergent cesarean section under general anesthesia. Patient and infant did well. Apgars were 8,9 but cord gas did show acidosis. After event, it was noted that NeoSynephrine was only diluted to 1 mg/ml instead of 100 mcg/ml.

Contributing factors:NeoSynephrine must be diluted twice – to 1 mg/ml, then to 100 mcg/ml.

Hurrying due to symptomatic hypotension in laboring parturient.

Multiple health care workers and family at bedside leading to distraction.

Lessons learned:Slow down when diluting medication and double check dilution.

Minimize distractions during preparation and administration of medication.

Double dilution of vasoconstrictors has a long history in anesthesia, with virtually every anesthesia provider at some point performing this task. Like many tasks in anesthesia, this one is not complex. Taking a 10 mg/mL to a 0.1 mg/mL solution (100 micrograms per mL) via a two-step dilution of 1 mg/mL to 0.1 mg/mL and then, with nine more mL of saline, to a 0.01 mg/mL (or 100 mcg/mL) is straightforward. The steps are linear and the same every time, and the error of omitting the second dilution falls into what James Reason defines as a skill-based error, with a slip or lapse in a routine that we have done many times before (Human Error. 1990). In Dr. Reason's framework of cognitive errors, skill-based errors are joined by rules-based errors and knowledge-based errors.

As shown in the table, Kahneman divides our day-to-day cognitive processes (and actions) into fast and slow thinking (Thinking, Fast and Slow. 2011), while Stanovich and West use the terms System I (fast) and System II (slow) (Behav Brain Sci 2000;23:645-726). System I thinking and actions are associative, unconscious, effortless, and come after many repetitions of a task or multiple exposures to a certain pattern (if bradycardic, give atropine); System I includes Reason's skill based and rule-based errors. System II thinking or actions are slow and reflective, deductive, conscious and effortful (first attempts at intubation). It is no wonder that humans much prefer to live in the world where our stored schema run subconsciously, even as we carry on conversations, or monitor vital signs. However, each of these basic types of thinking are inherently at risk for specific errors. For System I thinking common errors include that we run a schema at the wrong time (give atropine for bradycardia caused by electrocautery interference of a pacemaker), or that the schema is interrupted mid-stream, and when we return to it, we either enter a step too early or too late, omitting a step. In the case of severe hypertension noted above, the second step, that of diluting the 1 mg/mL solution once more, was omitted.

Although there is not a plethora of case reports of this type of error in the literature, virtually every provider of some experience knows of a local case or two of errors in dilution or has been called to assist when an error has occurred and the team must manage severe hypertension or, in this case, significant fetal distress. In pediatrics, the dangers are more prevalent and often more serious due to the need for dilution of nearly all medications and the narrow therapeutic range in our tiniest patients. In a study by Avidan, faculty and residents were asked to calculate the correct dilution for a number of medication infusions: only 15% got all dilutions correct (faculty were not better than residents), and incorrect calculations ranged from a drug concentration 50 times too low and up to 56 times too high (J Clin Anesth 2014;26:276-80).

Errors of dilution are, in fact, so common that virtually all medication safety guidelines call for the removal of all concentrated medications from the anesthesia medication trays (or Pyxis) and suggest that only the appropriate dilutions are provided. In 2010, the Anesthesia Patient Safety Foundation included a solution to dilution errors as one of their Consensus Recommendations for Improving Medication Safety in the Operating Room, stating: “High alert drugs (such as phenylephrine and epinephrine) should be available in standardized concentrations/dilutions prepared by pharmacy in a ready-to-use (bolus or infusion) form that is appropriate for both adult and pediatric patients” (APSF Newsletter 2010;25). The same consensus recommendations state that “routine provider-prepared medications should be discontinued whenever possible,” and “no concentrated versions of any potentially lethal agents [should be] in the operating room.” A report of survey results from 34 pediatric hospitals in 2014 found that 27 of the 34 did supply vasopressors in prefilled syringes – but clearly seven hospitals still were asking their anesthesia teams to prepare dilutions of these high-risk medications, and all of the 34 required at least one medication to be prepared by the anesthesia provider (Jt Comm J Qual Patient Saf 2014;40:471-5). In 2017, Wahr and colleagues published a review of literature and expert-based recommendations for medication safety in the OR, which included that “pharmacy provides diluted high-risk drugs” and “no concentrated drugs are in cart inventory” (Br J Anaesth 2017;118:32-43).

“At this point, there appears to be little good reason to continue the practice of having anesthesia providers perform dilution of high-risk vasoactive agents.”

As we can see from this report, a case recently submitted to the AIRS database, anesthesia providers at some hospitals continue to be required to prepare dilutions of high-risk medications on a day-to-day basis, despite the risk. The reporters(s) in this case cited, appropriately, the contributing causes of time pressure (hurrying) and distractions (multiple people at the bedside) and concluded that the lessons to be learned are to “slow down” and “avoid distractions.” These are nice sentiments, but unlikely to be possible in any anesthesia setting, where multiple medications are administered in the midst of distractions and under considerable time pressure. In addition, the comments essentially represent a “try harder” approach, which we know is useless when it comes to errors that grow out of the subconscious – the conscious mind does not keep a leash on the subconscious, as anyone who has fought to recall the name of a familiar song or book. By definition, the subconscious exists out of control of the conscious mind. Therefore, although “thinking about how we think” or metacognition can help alleviate some cognitive errors, in general, a conscious effort to never make an error in a System I task is a fool's errand.

Table

The two systems that describe the ways in which we think, and their relationship to the terminology used by James Reason (Reprinted with permission from Merry and Wahr) (Medication Safety during Anesthesia and the Perioperative Period. In press, 2021)

The two systems that describe the ways in which we think, and their relationship to the terminology used by James Reason (Reprinted with permission from Merry and Wahr) (Medication Safety during Anesthesia and the Perioperative Period. In press, 2021)
The two systems that describe the ways in which we think, and their relationship to the terminology used by James Reason (Reprinted with permission from Merry and Wahr) (Medication Safety during Anesthesia and the Perioperative Period. In press, 2021)

The task cited in this report very neatly represents what Dr. Reason called error traps: “The same situation keeps producing the same errors ... even though quite different people are involved. That surely indicates we are dealing with error prone circumstances rather than error prone people. We are dealing with error traps” (The Human Contribution: Unsafe Acts, Accidents and Heroic Recoveries. 2008). It should be clear that, if the same situation keeps producing the same error, we surely should take a systems approach to eliminating that situation, such as providing prefilled syringes of the appropriate dilution. Multiple manufacturers offer prefilled syringes like this; in addition, virtually any pharmacy can prepare these dilutions of high-risk medications in a quiet environment where a double check is available (and often mandated by policy).

Why then do hospital leaders continue to ask our anesthesia providers to undertake this risky business? Cost is the most commonly cited reason, although perhaps an unfounded one. The difference between a vial of concentrated phenylephrine is about $2.80, a prefilled syringe providing a concentration of 100 mcg/mL is $4.80. The pharmacy director may see a 2X cost, but the additional cost really disappears when one considers the labor costs associated with anesthesia providers performing the dilutions and the waste involved in drawing up a medication that may not be needed. One study found that savings associated with waste alone made prefilled syringe costs comparable to provider prepared syringes (Ann Fr Anesth Reanim 2009;28:211-4); another compared the system vulnerabilities in provider prepared versus prefilled and identified eight system vulnerabilities in using prefilled syringes versus 21 in using provider prepared (Anesthesiology 2016;124:795-803). At this point, there appears to be little good reason to continue the practice of having anesthesia providers perform dilution of high-risk vasoactive agents.

The event detailed above is clearly an error on the part of the anesthesia provider, but the refusal of hospital leadership to provide prefilled syringes of vasoconstrictors is more a violation than it is an error. As detailed by Merry and Wahr, “the distinguishing feature between an error and a violation is that in an error, one tries to do the right thing but actually does the wrong thing, whereas in a violation, one deliberately does the wrong thing (although without malevolent intent)” (Medication Safety during Anesthesia and the Perioperative Period. In press, 2021). In this case, the decision to not provide appropriate syringes represents a deviation from practices that medication safety experts have deemed necessary to maintain safety. As Merry and Wahr point out, people who violate generally believe that the violation will cause no harm – clearly leaders who choose to “save costs” by requiring providers to mix their own medications believe, although against all evidence, that the providers will do so perfectly every time. It becomes our responsibility, then, to vigorously put forth the argument that we cannot be expected to do this dangerous practice perfectly and that great harm lurks. This case put a mother and a baby at unnecessary risk; it is time to implement what we know to be those practices “deemed necessary to maintain the safe operation of a potentially hazardous system” (Medication Safety during Anesthesia and the Perioperative Period. In press, 2021).

Review of unusual patient care experiences is a cornerstone of medical education. Each month, the AQI-AIRS Steering Committee abstracts a patient history submitted to the Anesthesia Incident Reporting System (AIRS) and authors a discussion of the safety and human factors challenges involved. Real-life case histories often include multiple clinical decisions, only some of which can be discussed in the space available. Absence of commentary should not be construed as agreement with the clinical decisions described. Feedback regarding this article can be sent by email to airs@asahq.org. Report incidents or download the AIRS mobile app at www.aqiairs.org.