Tragically, life continues to imitate art. New hospital switches Nitrous and Oxygen pipelines

[gasspasser]

When I wrote my "anesthetically-inclined" review of the 1978 movie Coma, I figured that Hollywood was extrapolating reality.

After all, things like key-index and pin-index systems, O2/N2O fail-safe mechanisms, floating-bobbin flowmeters, drop down ventilator bellows -- all of these things and more had already made it into the practice of modern anesthesia to prevent the delivery of hypoxic mixtures to patients. Additionally, mass spectrometry gas analyzing systems were available to measure exhaled gases.

In these modern times, no one would suspect that the switching of nitrous and oxygen pipeline supplies would be possible...and furthermore go unnoticed for any significant period of time. Boy was I wrong.

We often disregard things proximal to our anesthesia machines. I can't even tell you when I last did a real machine check. And today's anesthesia machines are much more of a black box than yesterday's Narkomeds, etc. This is just a reminder that our vigilance must extend even to the unlikeliest of scenarios.

---

Members don't see this ad.

 

[deleted171991]

This is 2017. Any patient on oxygen needs continuous pulse oximeter monitoring. Period. No ifs, ands, or buts.

 

[gasspasser]

This is 2017. Any patient on oxygen needs continuous pulse oximeter monitoring. Period. No ifs, ands, or buts.

I gotta believe that they were monitoring the pulse ox. And when the patient's SpO2 kept going down, they kept turning up the oxygen....except it wasn't oxygen coming out of the wall...it was nitrous.

 

[POW16]

That might not have changed anything.

Sent from my SAMSUNG-SM-G920A using Tapatalk

 

[Ronin786]

I gotta believe that they were monitoring the pulse ox. And when the patient's SpO2 kept going down, they kept turning up the oxygen....except it wasn't oxygen coming out of the wall...it was nitrous.

I can't even imagine that situation. Just think about a newly born kid with a dip in sats that keeps getting worse as you're masking. You're probably just going down the route of tubing them and switching the oxygen never comes into mind.

 

[Gern Blansten]

This is most likely a pipeline crossover inside the wall or ceiling, so none of the PISS/DISS measures prevent it. It has been identified by the ABA as a knowledge gap area in the past (the key point being that you must disconnect the wall outlet in order for the O2 tank on the back of the machine to flow). I have personally had it occur in a hospital where I have worked (quick thinking by an anesthesiologist diagnosed/solved the problem). The gas analyzer should tell you what you need to know, but some off site locations may not have that in place.
I have discussed this with a patient who was a supervisor over gas pipeline work for hospitals and he boldly stated that it was "impossible" and implied that I was mistaken to think that something like this could occur. I was not impressed with his close mindedness and ignorance on the issue.

This is one of those things that all board certified anesthesiologists must know how to deal with. It will likely never happen to you, but if it does, you have to know about it so that you can rule it in or out as the cause of a decompensation. It is easy to blindly trust your equipment (easy to assume that O2 is being delivered through an O2 outlet) when faced with a situation such as this. I have heard of it occurring enough to know that I will be ready to question what I think are truths, especially in a new or rarely used portion of a hospital.

 

[gasspasser]

You're probably just going down the route of tubing them and switching the oxygen never comes into mind.

Exactly. Sucks.

 

[gasspasser]

It has been identified by the ABA as a knowledge gap area in the past (the key point being that you must disconnect the wall outlet in order for the O2 tank on the back of the machine to flow).

I was thinking about this. And I guess if I had a desaturating pt in the pacu or some place where all they had was the wall oxygen flowmeter spigot thingy... I might initially think that my BVM ventilation was not adequate. But then again, maybe I might suspect a big mucous plug plugging up a whole mainstem bronchus, or how about a PE. I suppose intubating the patient might be in the cards. But at *some point*, I might start suspecting the wall flowmeter being faulty somehow and maybe switch to an E cylinder of oxygen. That would be the solution ultimately. But man, if I hadn't seen that COMA movie, I probably would never consider that something behind the wall was afoul. Craziness.

This is one of those things that all board certified anesthesiologists must know how to deal with. It will likely never happen to you, but if it does, you have to know about it so that you can rule it in or out as the cause of a decompensation. It is easy to blindly trust your equipment (easy to assume that O2 is being delivered through an O2 outlet) when faced with a situation such as this. I have heard of it occurring enough to know that I will be ready to question what I think are truths, especially in a new or rarely used portion of a hospital.

I guess it's like switching to TIVA when the anesthesia machine or ventilator fails in the middle of a case. Ensure a patent airway and an independent source of oxygen.

 

[WholeLottaGame7]

Also have a healthy dose of suspicion if you're opening a new hospital, wing, floor, etc. I wouldn't necessarily think of this in a building that's been open for 40 years, but I might if it were the first day the ORs were open.

 

[gasspasser]

Oy. Sigh.

From 2007
Eight deaths linked to oxygen mix-up at Italian hospital

This time in a brand new cardiac unit...8 dead over 17 days. WTF

 

[d9sccr]

This is most likely a pipeline crossover inside the wall or ceiling, so none of the PISS/DISS measures prevent it. It has been identified by the ABA as a knowledge gap area in the past (the key point being that you must disconnect the wall outlet in order for the O2 tank on the back of the machine to flow). I have personally had it occur in a hospital where I have worked (quick thinking by an anesthesiologist diagnosed/solved the problem). The gas analyzer should tell you what you need to know, but some off site locations may not have that in place.
I have discussed this with a patient who was a supervisor over gas pipeline work for hospitals and he boldly stated that it was "impossible" and implied that I was mistaken to think that something like this could occur. I was not impressed with his close mindedness and ignorance on the issue.

This is one of those things that all board certified anesthesiologists must know how to deal with. It will likely never happen to you, but if it does, you have to know about it so that you can rule it in or out as the cause of a decompensation. It is easy to blindly trust your equipment (easy to assume that O2 is being delivered through an O2 outlet) when faced with a situation such as this. I have heard of it occurring enough to know that I will be ready to question what I think are truths, especially in a new or rarely used portion of a hospital.

Just a thought, but where do they fill up the E-cylinders? Any chance it just contains a bunch of nitrous too ?

 

[deleted171991]

I can't even imagine that situation. Just think about a newly born kid with a dip in sats that keeps getting worse as you're masking. You're probably just going down the route of tubing them and switching the oxygen never comes into mind.

It should. It's actually a textbook scenario after renovations. Also, in many places, even medical nitrous has a sweet odor, to differentiate it from oxygen.

 

[Aether2000]

Oops!

 

[Mman]

I can't even tell you when I last did a real machine check.

That's the part that kinda scares me. In the OR it's kinda hard to miss something like that since we have multiple gas analyzers picking up the FiO2.

 

[Aether2000]

Old school here...machine check when I start every day..... Adage from one of my fellow residents: If you have suction, oxygen, and positive pressure you most likely won't kill anybody.

 

[gasspasser]

in many places, even medical nitrous has a sweet odor, to differentiate it from oxygen.

If you google "Nitrous smell" you'll see lots of car people who soup up their engines with nitrous burning superthrusters or what not. I gather that it's illegal in some places? They talk about how the nitrous smells sweet and how it's a dead giveaway. I saw people wishing that they could get ahold of medical grade nitrous because it has no smell. Interesting how the different points of view exist in completely different circles.

 

[Mad Jack]

Our non-anesthesia ventilators have continuous oxygen monitoring to protect against just such a scenario. You'll get an O2 failure notice and a big red warning that never stops. The downside is, it is dependent upon the oxygen sensor in the machine to function properly, and they failed like crazy, so even if you got an O2 sensor warning, you would realistically figure the sensor was shot, replace the vent, and bag in the meantime off of the faulty outlet. A fail-safe that frequently falls is unfortunately as bad or worse than no fail-safe at all.

I'm honestly surprised anesthesia vents don't have similar O2 failure warnings though, it seems like a common sense thing to have continuous monitoring of your input gas in case of a line failure.

 

[Mman]

I'm honestly surprised anesthesia vents don't have similar O2 failure warnings though, it seems like a common sense thing to have continuous monitoring of your input gas in case of a line failure.

They do have warnings. We get a warning from the ventilator itself about the O2 inspired content and then the gas analyzer separately analyzes O2, CO2, and volatile anesthetic gas concentrations. You'd get multiple warnings if you were not delivering oxygen.

 

[caligas]

They do have warnings. We get a warning from the ventilator itself about the O2 inspired content and then the gas analyzer separately analyzes O2, CO2, and volatile anesthetic gas concentrations. You'd get multiple warnings if you were not delivering oxygen.

Yes but this is assuming you are in OR. Would not have helped in PACU.

 

[dhb]

Happened in Belgium a couple of years ago: new OR, peds mask down with 100% "O2" , pipeline error resulted in brain dead child. Crazy that thing like that can happen. Legally, pharmacy is supposed to check the nature of the gases at the outlets.

 

[Mman]

Yes but this is assuming you are in OR. Would not have helped in PACU.

well sure, it wouldn't help anywhere that you are just using wall supply for nasal cannula or ambu bag when you aren't monitoring the gas. The OP was discussing this in the context of anesthesia machines (that he doesn't really check) and how it could be a problem. I'm just pointing out that there are 2 independent oxygen sensors that would notice it on an anesthesia machine.

 

[gasspasser]

I got some time this morning...Did a writeup on my blog to sort of spread the word. It's not all-encompassing by any means, but I just wanted to put down some ideas so that we all might get something beneficial out of the senseless deaths of these people. Hope it never happens again. I'm certainly going to be revisiting the machine check for my Fisher Price anesthesia machine. Thanks y'all.

Life Imitating Art in Death

 

[DrZzZz]

Just a thought, but where do they fill up the E-cylinders? Any chance it just contains a bunch of nitrous too ?

Exactly what I was thinking. I was taught that if you have any concern of a pipeline crossover (like others said, probably very low on the differential), you don't switch to a tank because that could have theoretically been filled from the same crossed-over pipeline. You should bag the patient with room air and, if mid-operation, get a TIVA going to keep the patient asleep while you sort that all out.

 

[Winged Scapula]

Users are reminded that links to personal blogs/vlogs outside of a single Sig File notation is a violation of the Terms of Service.

 

[Iso4ane]

I know the textbook answer/test is to switch unplug the o2 line from the wall. But, what about just turning up the Air (yellow knob) flows or changing to 25% FiO2 on the digital machines and increasing FGF. After he air we breath is only 21% fiO2 and it keeps us alive.

 

[deleted171991]

I know the textbook answer/test is to switch unplug the o2 line from the wall. But, what about just turning up the Air (yellow knob) flows or changing to 25% FiO2 on the digital machines and increasing FGF. After he air we breath is only 21% fiO2 and it keeps us alive.

My textbook answer is: when in doubt, grab an oxygen cylinder and an ambubag, and go from there.

 

[deleted162650]

Just go mouth to tube.

 

[Gern Blansten]

Exactly what I was thinking. I was taught that if you have any concern of a pipeline crossover (like others said, probably very low on the differential), you don't switch to a tank because that could have theoretically been filled from the same crossed-over pipeline. You should bag the patient with room air and, if mid-operation, get a TIVA going to keep the patient asleep while you sort that all out.

I cannot think of any situation where O2 tanks would be filled on site from a crossed pipeline. Those tanks are filled off site and delivered to the hospital.

 

[anbuitachi]

Pipeline switch (O2 And nitrous) was one of our first simulations in CA1 year of residency! It was a fun one

 

[gasspasser]

Pipeline switch (O2 And nitrous) was one of our first simulations in CA1 year of residency! It was a fun one

I'm really glad to hear this! Hopefully this sort of thing doesn't happen in real life ever again!

 

[deleted171991]

Pipeline switch (O2 And nitrous) was one of our first simulations in CA1 year of residency! It was a fun one

It was one of the questions on my CA-1 machine exam. But then the guy who taught us was one of the experts in the field.

 

[BLADEMDA]

General anaesthesia should never be administered without an oxygen analyser in the breathing circuit. This is because, despite all the safety features of the machine, the final confirmation is the percentage of oxygen delivered to the patient as intended. There are many types of oxygen analysers, the commonest one used on most of the machines being the galvanic cell type. The other types are paramagnetic analysers, and the polarographic (Clark's electrode) type. Oxygen analysers have to be calibrated at regular intervals, most often as part of the machine start-up check procedure. ASTM standards do require that low oxygen alarm level cannot be set below 21%.[5]

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3821264/

 

[BLADEMDA]

Fault Algorithms

Proposed Hypoxic "O2" Pipeline Algorithm
Copyright 2002 University of Florida Department of Anesthesiology
Version 1.1
May 13, 2002

Rationale and discussion of the observations and consequences

Recommended Algorithm
If a hypoxic O2 pipeline condition is suspected because of a low FiO2 alarm and/or reading that does not match the set FiO2:

Note that this implies that an O2 analyzer is being used

https://vam.anest.ufl.edu/algorithms.html

 

[BLADEMDA]

From the UF website:

we have been informed of such a case in a military hospital where the central O2 supply was actually a bank of O2 H-cylinders in the cellar. A medical orderly was tasked with monitoring the pressure gauges on the O2 cylinders and opening a new H-cylinder when the one being used was nearing exhaustion. The new "O2 cylinder" that was opened while cases were proceeding was misfilled with Argon and multiple fatalities occurred at different anesthetizing locations.

Most developed countries have guidelines for testing O2 outlets after construction or renovation and before they are used. This system too has been reported to fail. In one reported instance, the O2 outlets after the addition of 3 new ORs were correctly plumbed and passed when tested. Subsequently, the O2 outlets were repositioned and during repositioning of the outlets, the gas pipes were crossed. The O2 outlets were not tested again after the repositioning and patient fatalities occurred in different anesthetizing locations.

https://vam.anest.ufl.edu/hypoxicpipediscussion1.html

 

[gasspasser]

From the UF website:

we have been informed of such a case in a military hospital where the central O2 supply was actually a bank of O2 H-cylinders in the cellar. A medical orderly was tasked with monitoring the pressure gauges on the O2 cylinders and opening a new H-cylinder when the one being used was nearing exhaustion. The new "O2 cylinder" that was opened while cases were proceeding was misfilled with Argon and multiple fatalities occurred at different anesthetizing locations.

Most developed countries have guidelines for testing O2 outlets after construction or renovation and before they are used. This system too has been reported to fail. In one reported instance, the O2 outlets after the addition of 3 new ORs were correctly plumbed and passed when tested. Subsequently, the O2 outlets were repositioned and during repositioning of the outlets, the gas pipes were crossed. The O2 outlets were not tested again after the repositioning and patient fatalities occurred in different anesthetizing locations.

https://vam.anest.ufl.edu/hypoxicpipediscussion1.html

So tanks with all their pin index systems get filled wrong too...man oh man

Thanks for the link!

 

[MaximusD]

Our non-anesthesia ventilators have continuous oxygen monitoring to protect against just such a scenario. You'll get an O2 failure notice and a big red warning that never stops. The downside is, it is dependent upon the oxygen sensor in the machine to function properly, and they failed like crazy, so even if you got an O2 sensor warning, you would realistically figure the sensor was shot, replace the vent, and bag in the meantime off of the faulty outlet. A fail-safe that frequently falls is unfortunately as bad or worse than no fail-safe at all.

I'm honestly surprised anesthesia vents don't have similar O2 failure warnings though, it seems like a common sense thing to have continuous monitoring of your input gas in case of a line failure.

There is an o2 analyzer at the furthest point downstream.

 

[Mad Jack]

There is an o2 analyzer at the furthest point downstream.

So basically the only way for this to happen would be to ignore your O2 analyzer alarm? Seems pretty damn preventable.

 

[MaximusD]

O2 analyzer can fail or be miscalibrated, I suppose.

 

[nimbus]

I believe the newborns died outside of an anesthetizing location. Probably a birthing suite, newborn nursery or nicu. Doubt an O2 analyzer was in use.

 

[gasspasser]

I believe the newborns died outside of an anesthetizing location. Probably a birthing suite, newborn nursery or nicu. Doubt an O2 analyzer was in use.

Yes that's correct...the pipeline crossing affected the wall oxygen in a neonatal resuscitation unit where there were flowmeters but no oxygen analyzers in use....I imagine it's like the baby warmers that we use in the nursery. I check at my hospital...they can either hook into the key indexed quick plug on the wall or run off the tanks in the back. Tanks are pin indexed so that you can only attach oxygen tanks. But there's nothing keeping the unit from delivering a different gas if that's what is being supplied out of the wall outlet

 

[Mad Jack]

Yes that's correct...the pipeline crossing affected the wall oxygen in a neonatal resuscitation unit where there were flowmeters but no oxygen analyzers in use....I imagine it's like the baby warmers that we use in the nursery. I check at my hospital...they can either hook into the key indexed quick plug on the wall or run off the tanks in the back. Tanks are pin indexed so that you can only attach oxygen tanks. But there's nothing keeping the unit from delivering a different gas if that's what is being supplied out of the wall outlet

Oh, yeah. That's pretty damn awful, and really hard to plan for... I mean, when an infant's sats are dropping after delivery, your last thought is going to be a bad source gas... By the time you've hit that low on the differential, the kid is probably done for.

 

[nimbus]

For once they can't blame anesthesia.

On a lighter note,

http://gomerblog.com/2017/02/warren-beauty-la-la-land/

 

[gasspasser]

For once they can't blame anesthesia.

On a lighter note,

http://gomerblog.com/2017/02/warren-beauty-la-la-land/

Lol just saw this...I think GomerBlog has a whole section on anesthesia stuff