Nurse Gives Lethal Dose of Vecuronium Instead of Versed

On December 26, 2017, a tragic and preventable death occurred when a patient at Vanderbilt Hospital was sent for a Positron Emission Tomography (PET) scan and received a lethal dose of Vecuronium instead of Versed. Nurses General Nursing Article

PET scans are typically performed in the outpatient setting and not inpatient. They are not emergent. This was a full body inpatient PET scan which might be said is inappropriate resource utilization.

If it had been done in the outpatient setting, it's unlikely that Versed would have been administered. Essentially a patient died because a dangerous medication that was never ordered was given.

The patient, a 75 yr old female, was admitted to neuro ICU on 12/24/2017 with intraparenchymal hematoma of the brain, headache, homonymous hemianopia (loss of visual field bilaterally), atrial fib, and hypertension. She was alert and oriented and doing well, about to be transferred to SDU. While the patient was waiting for her scan, she asked for medication to prevent claustrophobia. The provider ordered 2 mg of Versed, and the pt's nurse was asked to come down and administer it.

At the time, the pt's nurse was covering a lunch break for another nurse and asked the "help all nurse" to go down and give the Versed so the procedure would not be delayed or rescheduled. The "help all nurse" performs different tasks, but there was no job description for a "help all nurse". It is not clear if the "help all nurse" was an ICU nurse, or a nurse floated from elsewhere, such as Med Surg. Clearly, the "help all nurse" was not trained in conscious sedation, and was seemingly unfamiliar with both Versed and Vecuronium.

The "help all nurse" searched for Versed under the pt's profile in the Pyxis but it did not come up. So she chose override and typed in "VE". The first drug to come up was Vecuronium, which she selected. She then proceeded to reconstitute the drug, not recognizing that Versed does not need to be reconstituted, and not taking heed of the red cap on the vial or the words "paralytic agent". Next, she administered 2 mg of the Vecuronium IVP and left the patient.

Vecuronium and Versed

Vecuronium is a neuromuscular blocking agent that causes paralysis and death if patients are not monitored.

Basically, the patient will be paralyzed, conscious and unable to draw a breath. All sensation will be experienced- panic, pain, fear, but the patient is unable to cry out. It's truly the stuff of nightmares.

Versed is often given for procedural sedation but is not a paralytic. Versed can affect respirations, and patients must be closely monitored for hypoventilation.

Timeline

Here is a timeline of events, where the time was documented.

  • The doctor ordered Versed 2 mg IV at 1447 on 12/26/2017.
  • The Pharmacy reviewed the order at 1449.
  • Vecuronium 10 mg was withdrawn from the Pyxis at 1459 using override.
  • Vecuronium administered but not documented. Patient left alone, unmonitored.
  • Rapid response was called at 1529.
  • Return of spontaneous circulation (ROSC) was restored after 2 rounds of ACLS
  • Patient was intubated and returned to neuro ICU
  • Patient displayed myoclonic jerks with posturing in ICU
  • Anoxic brain injury is documented
  • Extubation was performed at 1257 on 12/27/2017
  • Death was pronounced at 1307 due to pulselessness

The Findings

The Department of Health and Human Services Centers for Medicare and Medicaid (CMS) did not investigate the event until October, 2018 as the death was not reported to them at the time.

According to CMS standards, patients have a right to receive care in a safe setting and hospitals have a responsibility to mitigate potentially fatal mistakes.

CMS ruled that Vanderbilt failed to provide safe care and protect patients. Vanderbilt was placed in serious and immediate jeopardy by CMS.

Since that ruling, the CMS has accepted Vanderbilt's plan of action.

What Went Wrong

  • No documentation of the Vecuronium being administered
  • Patient was not monitored
  • Autopsy was not conducted as should have been for an unusual death such as one caused by a medication error
  • Death certificate said the cause of death was a cerebral bleed, not a medication error.

Contributing Factors

  • Pyxis override functionality - a necessary function, but contributed to the error
  • Failure of the hospital to ensure that only nurses trained in conscious sedation can administer Versed
  • Failure of hospital policy to address the manner and frequency of monitoring
  • Concern about convenience over safety- pressure to not reschedule a test rather than taking the time to safely prepare the patient
  • Staffing- Covering another nurse's patients results in an unsafe workload

Blaming

The simplest thing is to blame the nurse for 100% of the error. Her failure to follow basic medication safety steps is not defensible.

But stopping there prevents us from learning how it happened and preventing future mistakes. We need to learn more about the science of mistakes. We all make them. Something in our brains allows us to see what we expect to see and not always what is there, as in running a red light. Or not registering red caps and cautionary labels on a high-alert medication.

The "help all nurse" was distracted in that she had an orientee with her, and immediately after giving the medication, went to the ED to perform a swallow screen test. She was performing a series of tasks.

The patient's nurse did not delegate appropriately as she did not confirm the "help all nurses " knowledge and skills. Tasks should be delegated to the right person, and this was not the right nurse to administer Versed. Maybe nurses are not all interchangeable as administrations sometimes like to think.

Vanderbilt did not clearly define role expectations, or patient monitoring in conscious sedation.

What do you think were the causes, and would have prevented this from occurring?

Related post When Nurses Make Fatal Mistakes

I've done something horrifically stupid at least twice over the past nine years. I couldn't even explain why ...my brain just didn't pull the pieces together and there you go.

There was no harm done, but under even slightly different circumstances there could have been great harm done.

I got lucky and I feel awful for this nurse who didn't ...but does she even want to be a nurse anymore? I'm guessing not. And as much as I believe in grace in these types of situations, in some cases, it just can't be.

Specializes in MPCU.

Punishment only teaches people to avoid getting caught.

The rules are guidelines, not written in stone. Many "rules" are nothing more than ways to assign blame if something goes wrong. Best practice is to understand the "why" for the rules. Understanding "why" allows the nurse to know when the rule does not apply. The nurse in this situation did not know the rules and / or did not know "why." Punishment makes "why" = so you do not get in trouble. This culture is common place in hospitals here in the U.S.A. Remediation such as a restricted license could be helpful to the nurse and the culture. Losing her license only addresses one specific nurse in one specific situation.

Specializes in Critical Care.

A bit off topic, but some have questioned why midazolam (Versed) was ordered and not lorazepam (Ativan), suggesting lorazepam would have been safer. Between the two, midazolam is generally considered the safer procedural anxiolytic. It's shorter acting, so the CNS depressive effects will at least be of shorter duration, and even though both lorazepam and midazolam fall generally under benzodiazepines, midazolam has an added dissociative-like effect which for the same amount of procedure-tolerance effects will produce less respiratory depression and overall CNS depression.

Specializes in EMS, ED, Trauma, CEN, CPEN, TCRN.
Can't tell you the number of times that I've been able to calm down "problem" patients by sitting down and talking to them for a few minutes. I always try to go that route first, even though it's devilishly difficult due to the time constraints we're all familiar with. It really is remarkable how far a little TLC can go with an anxious patient. Just drawing up a chair, sitting down, giving them your full attention for five minutes, speaking kindly, slowly, and calmly, offering reassurance, then coming back in 15 minutes to check in on them and see what else you can help with. I mean... this strategy is like magic for your run-of-the-mill "omg I'm in the hospital and no one will talk to me and I don't know what's going on" anxiety. It's a different story when you have someone with psych issues who needs their PRN valium, of course. But taking a minute to talk and connect is, IMHO, way safer and kinder than just throwing anxiolytics at the patient and walking away. The problem is that it takes some time, which is at an absolute premium for any healthcare worker.

The kind of anxiety produced by the claustrophobia of a scanner tube is not something that can be shushed away by some hand-holding and kind words. I went for a cardiac MRI this summer and discovered that I have some lingering PTSD from my experience with congestive heart failure last year, during which I had to lie flat for studies while drowning in my own lungs. I had no idea I would react as I did! I even had an MRI of my injured hip when I came home from deployment in 2014, so I was really surprised by my utter panic. I had to halt the scan - super embarrassing, but there was no way I could talk myself through enduring an hour of what is an interactive scan in that MRI tube (open MRIs are not suitable for cardiac MRIs, unfortunately). I had to get a dose of PO Valium prescribed by my electrophysiologist and reschedule the scan. I knew it wasn't rational, but I could not have done it without 10mg of Valium. So yeah, I totally understand why this patient needed something (definitely not vec!!!) for her study. Poor woman.

Specializes in Critical Care.

I've done adverse event review at a couple of different organizations, and one of the biggest challenges in implementing changes to prevent repeat events is the belief or desire to believe that the failure to avoid these events are due to a single, isolated error.

Current best practices for avoiding medical errors are two-tiered. The first goal is to prevent the primary error, but it's a bit ignorant to believe that those primary errors can be reliably avoided completely so the second tier is to ensure that individual errors would have to make it through a gauntlet of systemic error prevention strategies in order to make it to the patient, and this is a pretty clear case of both an individual error, as well as multiple failures to prevent or mitigate the error through systemic error prevention.

The result of viewing this is primarily an individual error is typically ineffective, we remind nurses to read the vial when removing it from the pyxis and prior to administering, the problem is that it's unlikely the nurse in question had never heard this suggestion before. This often not only the easiest (laziest) fix, there seems to be some psychological comfort for us to believe that risks to patients are simple rather than complex, so we ignore the larger problems. A systemic fix on the other hand to pulling the wrong med, which we've done at my current facility and found it effective, is to first change the predictive text to require 4 letter before any medications are shown, and we've also added decoys to the med list that results, so your med is not always the first med on the list even if you type the full name, this changes the habit from typing part of or the full name of the med and then reflexively picking the first med on the list to having to always hunt for the med. Our rate of 'wrong med' scans at the barcode scanning step has dropped from almost 2 a day to less than 2 a month. For a few years prior to this, the issue was addressed an individual error issue and should be fixed by reminding nurses to look at their meds, which had no effect on the problem. The same basic premise is true here, viewing this as primarily a problem with the actions of an individual nurse isn't likely to changing the serious systemic failure that also contributed to this patient's death.

Thank you for that dose of rationality and clear-eyed summation of real world reviews, MunroRN.

Beautiful said! @MunroRn

I've done adverse event review at a couple of different organizations, and one of the biggest challenges in implementing changes to prevent repeat events is the belief or desire to believe that the failure to avoid these events are due to a single, isolated error.

Current best practices for avoiding medical errors are two-tiered. The first goal is to prevent the primary error, but it's a bit ignorant to believe that those primary errors can be reliably avoided completely so the second tier is to ensure that individual errors would have to make it through a gauntlet of systemic error prevention strategies in order to make it to the patient, and this is a pretty clear case of both an individual error, as well as multiple failures to prevent or mitigate the error through systemic error prevention.

The result of viewing this is primarily an individual error is typically ineffective, we remind nurses to read the vial when removing it from the pyxis and prior to administering, the problem is that it's unlikely the nurse in question had never heard this suggestion before. This often not only the easiest (laziest) fix, there seems to be some psychological comfort for us to believe that risks to patients are simple rather than complex, so we ignore the larger problems. A systemic fix on the other hand to pulling the wrong med, which we've done at my current facility and found it effective, is to first change the predictive text to require 4 letter before any medications are shown, and we've also added decoys to the med list that results, so your med is not always the first med on the list even if you type the full name, this changes the habit from typing part of or the full name of the med and then reflexively picking the first med on the list to having to always hunt for the med. Our rate of 'wrong med' scans at the barcode scanning step has dropped from almost 2 a day to less than 2 a month. For a few years prior to this, the issue was addressed an individual error issue and should be fixed by reminding nurses to look at their meds, which had no effect on the problem. The same basic premise is true here, viewing this as primarily a problem with the actions of an individual nurse isn't likely to changing the serious systemic failure that also contributed to this patient's death.

I don't think anyone here would argue that when serious medical errors happen a facility doesn't have some responsibility for what happened. And I'm sure we all know about "just culture" and the practice of ascribing lethal errors in care that result in harm/death to patients to systems problems. As nurses/members of the public, we can report a facility to the Department of Health for federal and state law violations, i.e. staffing, violations of the Nurse Practice Act and related regulations that affect the ability to provide safe and appropriate care, patient safety violations, and violations of patients rights, etc. However, the point has been made that as individual nurses the buck stops with us as we are the last line of defense before an order reaches the patient, and the state Boards of Nursing hold individual nurses accountable for their practice; I believe these are the points that a number of people have made here on this thread.

Specializes in Psych, Corrections, Med-Surg, Ambulatory.
Yes.

But - - in sincere effort to not make excuses, I would ask people to forget all of that and instead look ahead. So the nurse is fired, V will make sure their little policy says that a patient receiving any kind of sedation must be monitored, and will for sure lock down the vecuronium so that it can't be accessed whether it is an emergency or not.

What are those things going to do about the next non-expert running here and there who is being timed and then ridiculed if s/he so chooses to act in a prudent manner when faced with a similar situation?

The stellar people running Vanderbilt are going to carry on having sentinel events.

Specializes in Psych, Corrections, Med-Surg, Ambulatory.
The contradiction of the statement ... Vande is run by expert knowledgable people, to I have personally investigated SEVERAL SENTINAL EVENTS is amazing to me. I need a very big googly eyed emoji for this Davey!

Cheers

Yes. As much as I can't get my head around the whole sequence of events regarding this incident, that post was the one I really couldn't get my head around.

Specializes in Prior Auth, SNF, HH, Peds Off., School Health, LTC.
To play the devil's advocate, in my PICU setting we override and give paralytics all of the time; we'd be in a real bind if paralytics came from pharmacy only, as our peds cardiac patients have a tendency to flip out, bear down, have pulmonary hypertensive crises, and actively try to die within about a 2-minute span. I'd say that on a weekly basis, we have to give paralytics (plus sedation) at a moment's notice to combat pulmonary hypertension and prevent codes; granted, many of our patient's requiring paralytics have a standing PRN paralytic ordered, and they're obviously all intubated...

I do agree that there are a lot of errors in this scenario (both personal and systemic); however, I don't think that preventing a paralytic override from the pyxis is necessarily the answer. Perhaps it would make sense to only be able to override paralytics in the ICU or OR suite, but honestly, if I were in CT or MRI with one of my patients, I'd want to have the option to override roc or vec at a moment's notice.

When you take your littles to CT/MRI, don't you take a med box with you? The paralytic could be included in there... maybe it already is, being that it's an RSI drug.

In any case, there are nurses every day who function in situations where they don't (or can't) rely on the ability to scan every med before giving it, or depend on a Pyxis to give them the right med. They pull meds from crash carts, med boxes, etc. and by knowing the characteristics of the med they're about to give (i.e. Versed does not require mixing a powder and a dilutant, paralytics have very obvious markings that are designed to make the person administering it take notice, and so on....) and by doing the double checks they are supposed to do, they are able to safely administer these dangerous drugs.

It's like when a nurse gives a patient a huge dose of something and it's later found out that she had to draw from 10 vials and use 4 syringes, and that never made her pause and think whether there might be something amiss... it is the nurse's responsibility to know enough about whatever meds she gives to be able to tell if what she's about to do is reasonable. And if it is something she's not familiar with, then she should look it up before she takes it upon herself to administer it.

I know mistakes are inevitable... everyone has or will make one... but I DO NOT believe that all mistakes are created equal. I think that the higher the stakes, the higher the bar should be. And the more alert we should be to the possibility of making a mistake. I think it is absolutely possible to have mistakes like this one be avoidable. If Nurse Help-all had done a double check on her own, and taken it a step farther and had Nurse Trainee double check a critical med *with* her, then instead of mistake, it would've been a near-miss. And I'll take a million near-misses to a single fatal mistake any day

because in a fast paced ICU, you cant always have a witness to grab intubation meds.

In the same way that studies show that double verification insulin and heparin does not significantly reduce errors

Not surprised.

Got a link?