Pulmonary Artery Catheters

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

DocMcCoy

Full Member
10+ Year Member
Joined
Sep 4, 2009
Messages
1,064
Reaction score
734
CA1 here giving a lecture to the department next week on Swans. My N=3 and pretty much everyone in the room is going to know more about them than me. Was curious from those with experience and still use them what types of patients are you using these in and what types of cases?

Members don't see this ad.
 
The stock answer is PA Caths are used for patients with severe cardiac disease for non-cardiac surgery where one could expect major fluid shifts and/or resuscitation. Also used in cardiac surgeries.
 
They are worthless. Haven't used one in over 15 years.
 
  • Like
Reactions: 2 users
Members don't see this ad :)
Figure that might be the case. Have a great faculty mentor that hasn't used one in years challenged me to find someone outside my residency bubble still using them.

Interestingly I find our guys that do hearts put them in more often than not for any valve work. It's not just the old guys, it's the younger guys that are super certified with TEE too.

I feel like my institution burns through a lot of them. I'm happy about it because Im learning how to use something new. Now I'm wondering if I'll ever use one outside of residency. Especially since I'm not that interested in doing hearts. Just curious what its like in the real world.
 
2014 ACC/AHA Cardiac pt for Noncardiac sx Guideline
 

Attachments

  • Screen Shot 2016-05-05 at 8.20.32 PM.png
    Screen Shot 2016-05-05 at 8.20.32 PM.png
    265.3 KB · Views: 110
  • Screen Shot 2016-05-05 at 8.20.12 PM.png
    Screen Shot 2016-05-05 at 8.20.12 PM.png
    410.3 KB · Views: 123
  • Like
Reactions: 1 user
PAC still have their place in medicine. I have placed well over a thousand in my career but haven't placed one in 10 years. Not because they are useless or out gunned but because my practice environment has changed. The TEE has chased everyone away from the PAC much like the US has chased people away from nerve stimulators. But you can't place a TEE in a pt and continuously monitor that pt for hours in the ICU. When I did the sickest of the sick hearts (just like the rest of you) the PAC was a great tool in the ICU for the following 12-24 hours. I could place one just about as fast as someone could drop a TEE probe and start to decipher the data. But nobody has that skill anymore.

So one big advantage is that you can continue to monitor pts for hours on end with a PAC. The ideal anesthesiologist can use both TEE and PAC.
 
PAC still have their place in medicine. I have placed well over a thousand in my career but haven't placed one in 10 years. Not because they are useless or out gunned but because my practice environment has changed. The TEE has chased everyone away from the PAC much like the US has chased people away from nerve stimulators. But you can't place a TEE in a pt and continuously monitor that pt for hours in the ICU. When I did the sickest of the sick hearts (just like the rest of you) the PAC was a great tool in the ICU for the following 12-24 hours. I could place one just about as fast as someone could drop a TEE probe and start to decipher the data. But nobody has that skill anymore.

So one big advantage is that you can continue to monitor pts for hours on end with a PAC. The ideal anesthesiologist can use both TEE and PAC.
I don't have your experience, but couldn't arterial waveform analysis replace the need for PAC in the ICU. They give pretty similar data and many studies have shown their equivalency. Sure, there is limitations to the data provided with the A-line, but there are limitations to the data provided by the PAC as well.
 
I don't have your experience, but couldn't arterial waveform analysis replace the need for PAC in the ICU. They give pretty similar data and many studies have shown their equivalency. Sure, there is limitations to the data provided with the A-line, but there are limitations to the data provided by the PAC as well.

SVV has so many limitations though. Not to mention the entire algorithm the SVV is based on is A.) calculated using a bunch of assumptions creating a constant for arterial compliance and B.) was initially tested/formulated using >8cc/kg TV which literally nobody uses in the ICU anymore.
 
  • Like
Reactions: 2 users
When titrating ionotropes
Inotropes, not Ionotropes. Sorry one of my many pet peeves.

I put in swans for many valves still, and the sicker cabgs. Doesn't change much in the OR but does help a lot in the ICU. It's by no means an obsolete tool.

The problem with the swan is that you have to properly interpret the data. My ICU used to have a nasty habit of starting inotropes whenever the CI was low, even if I said the EF was normal. It turns out that fluid is the solution to hypovolemia, not epi. So the monitor is only as good as the person interpreting it.

The pulse contour technology is only OK IMO, but the Flotrac especially is garbage if the patient is on pressors. It will falsely tell you things are better than they are, which is a big problem. In my limited experience the Lidco was a little better than the Flotrac.
 
  • Like
Reactions: 5 users
SVV is inaccurate on pressors, with arrhythmia, anything other than 100% controlled mechanical ventilation, pneumoperitoneum, and obesity (I think BMI>40 but don't hold me to that) among others. Basically it's pretty difficult to find a periop scenario where you should expect it to be accurate little lone actually lead to improved outcomes.
 
SVV is inaccurate on pressors, with arrhythmia, anything other than 100% controlled mechanical ventilation, pneumoperitoneum, and obesity (I think BMI>40 but don't hold me to that) among others. Basically it's pretty difficult to find a periop scenario where you should expect it to be accurate little lone actually lead to improved outcomes.
I wasn't argueing that SVV or arterial contour analysis is a great monitor. PAC isn't that great of a monitor either with errors found in a multitude of condition. E.g. PAOP is an indirect measurement of LVED volume and any diastolic dysfunction will lead to incorrect correlation.

Ultimately, I don't think you can find any consistent data to show that either monitor improves outcome. PAC however is associated with more adverse events that can increase your liability.
 
Members don't see this ad :)
I wasn't argueing that SVV or arterial contour analysis is a great monitor. PAC isn't that great of a monitor either with errors found in a multitude of condition. E.g. PAOP is an indirect measurement of LVED volume and any diastolic dysfunction will lead to incorrect correlation.

Ultimately, I don't think you can find any consistent data to show that either monitor improves outcome. PAC however is associated with more adverse events that can increase your liability.

Understood, my intentions were not to come off as attacking, only to share the limitations of the device. I see the vigileo being used in all sorts of scenarios these days where it just isn't appropriate so it's one of my pet peeves. And I agree that all monitors have limitations, and if anything PACs are the poster child for that. My main point is to know the limitations of each monitor and to know how the numbers you're looking at are derived. At least with a PAC the values are actually measured, not amalgamated from various theoretical mathematical constructs and assumptions before being generalized to all patients.....
 
  • Like
Reactions: 1 user
I wasn't argueing that SVV or arterial contour analysis is a great monitor. PAC isn't that great of a monitor either with errors found in a multitude of condition. E.g. PAOP is an indirect measurement of LVED volume and any diastolic dysfunction will lead to incorrect correlation.

Ultimately, I don't think you can find any consistent data to show that either monitor improves outcome. PAC however is associated with more adverse events that can increase your liability.
i don't completely subscribe to this notion. Problems with PAC occur, as with any monitor, we people how are unfamiliar with the monitor try to use it. Inaccurate data also falls in this catagory. Many studies are performed is institutions with people less likely to use the monitor. But someone who has shed it for years and has the "n" to show for it can make great use out of something others feel are limited.
The complications with PAC are avoidable. You just need to know how to avoid them. Like not wedging every catheter. It isn't necessary especially in someone with PHTN. and if it is necessary get your measurement and then pull back 2cm. Don't leave it there. I like the PAC mostly because I can follow trends more easily. But I trained with them so I am familiar. How many new grads and current sediments get many opportunities to place one? I'm guessing very few.
 
  • Like
Reactions: 1 user
i don't completely subscribe to this notion. Problems with PAC occur, as with any monitor, we people how are unfamiliar with the monitor try to use it. Inaccurate data also falls in this catagory. Many studies are performed is institutions with people less likely to use the monitor. But someone who has shed it for years and has the "n" to show for it can make great use out of something others feel are limited.
The complications with PAC are avoidable. You just need to know how to avoid them. Like not wedging every catheter. It isn't necessary especially in someone with PHTN. and if it is necessary get your measurement and then pull back 2cm. Don't leave it there. I like the PAC mostly because I can follow trends more easily. But I trained with them so I am familiar. How many new grads and current sediments get many opportunities to place one? I'm guessing very few.


Where I trained (just graduated in 2015), I think EVERY cardiac case i.e. CABG/Valve/Transplant/VAD got a PAC either Swann or CCO. Overkill? Probably... but I ended up placing a good 30-40 of them in residency with 1-2 of them in non-cardiac cases in really sick patients. Whether this was the right thing to do... probably not but not my idea and the new guidelines weren't out yet. I'm not sure if they've made changes though in the year I've left with the new guidelines. But I do feel comfortable placing them at least.
 
Also, these guidelines are for non-cardiac surgeries. What are the guidelines for cardiac cases? I feel like I heard a lot of places don't place central lines in routine valves and CABGs. They just do 2 large IVs, a-line and TEE. Like I mentioned we did all this plus the PAC.
 
CA1 here giving a lecture to the department next week on Swans. My N=3 and pretty much everyone in the room is going to know more about them than me. Was curious from those with experience and still use them what types of patients are you using these in and what types of cases?
Their only clear indication nowadays is to monitor pulmonary arterial pressures, usually for pulmonary or right-sided heart pathology. Anything else, and it's a lot of speculation (there is a bunch of stuff between that balloon and the left ventricle, and even if it were in the LV, what do increased pressures mean?). They are not even good for following serial cardiac outputs, because their error margin is comparable with the 10-15% magnitude of change we use to look for response. A good echo machine and bedside echocardiographist (intensivist) will beat the Swan 9 times out of 10.

P.S. I trained with Swans, so my bias should be for, not against them. I have never had a complication in the 50 I have placed, I just feel their numbers are sometimes about as relevant as the CVP.
 
Last edited by a moderator:
  • Like
Reactions: 1 user
SVV has so many limitations though. Not to mention the entire algorithm the SVV is based on is A.) calculated using a bunch of assumptions creating a constant for arterial compliance and B.) was initially tested/formulated using >8cc/kg TV which literally nobody uses in the ICU anymore.
Only about 8% of ICU patients qualify for SVV. The Swan is useless for fluid challenge, too.
 
Last edited by a moderator:
  • Like
Reactions: 1 user
Inotropes, not Ionotropes. Sorry one of my many pet peeves.

I put in swans for many valves still, and the sicker cabgs. Doesn't change much in the OR but does help a lot in the ICU. It's by no means an obsolete tool.

The problem with the swan is that you have to properly interpret the data. My ICU used to have a nasty habit of starting inotropes whenever the CI was low, even if I said the EF was normal. It turns out that fluid is the solution to hypovolemia, not epi. So the monitor is only as good as the person interpreting it.

The pulse contour technology is only OK IMO, but the Flotrac especially is garbage if the patient is on pressors. It will falsely tell you things are better than they are, which is a big problem. In my limited experience the Lidco was a little better than the Flotrac.
Tell your ICU that inotropes correlate with increased mortality in every recent study I have seen them in.

I regularly transfer cardiac patients to our cardiac ICU, stable on pressors, just to see them get killed (literally) with inotropes in a few days, while on a Swan. ;)
 
I use them on most hearts. Mostly for post op ICU mgmt, but there's utility in the OR. Occasionally hemodynamic data will subtly show concerning signs before the TEE. Then you can carefully watch TEE to confirm or refute what you think is going on. It can be very subtle so I'm not surprised that studies aren't particularly supportive. I have not put any in non-cardiac cases except in random C-section patients with Eisenmengers, or whatever.

Check out Paul Marino's ICU book. He has a full chapter on PA catheters that would be great for a CA1. You eventually might not agree with everything, but it's a good first read. Here's something from it that stuck with me over the years -

"There are two fundamental problems in the criticism of the PA catheter based on mortality data. The first is the simple fact that the PA catheter is a monitoring device, not a therapy. If a PA catheter is placed to evaluate a problem, and it uncovers a disorder that is untreatable and fatal (e.g., cardiogenic shock), the problem is not the catheter, it is the lack of effective therapy. Mortality rates should be used to evaluate therapies, not measurements.

The second problem is the seemingly prevalent notion that everything we do in the ICU must save lives to be of value. Mortality should not be the dominant outcome measure in ICUs because there are too many variables that can influence mortality in critically ill patients, and also because mortality is an eventual outcome in all patients admitted to the ICU. Management decisions should be based on the scientific rationale for an intervention—those who expect their management decisions to consistently save lives are doomed to failure. "

I interpret this to mean there's also a lot of variability in what anesthesiologists/intensivists do with the PAC data that affects how useful they are. Having the data is one thing., knowing what to do with it is entirely different. Sort of like EKG, EtCO2, pulse ox, cerebral ox, TEE, etc etc etc...
 
  • Like
Reactions: 2 users
CA1 here giving a lecture to the department next week on Swans. My N=3 and pretty much everyone in the room is going to know more about them than me. Was curious from those with experience and still use them what types of patients are you using these in and what types of cases?
@PainDrain's answer is what you should remember, at this stage of your training. It's more important to know the limitations of the device, than what it's supposed to be used for.
 
Also, these guidelines are for non-cardiac surgeries. What are the guidelines for cardiac cases? I feel like I heard a lot of places don't place central lines in routine valves and CABGs. They just do 2 large IVs, a-line and TEE. Like I mentioned we did all this plus the PAC.
wow, that's new! first time i've ever heard of someone doing a cardiac case without a central line.

in my residency program we placed them very rarely. i work at that same program and we cover a private practice hospital where every cardiac case gets PAC, so that the surgeons have the info when they're being called in the middle of the night
 
One can give pressors safely for 24-48 hours on a well-running peripheral, especially if large bore. So there is really no significant indication for the central line, if the PAC is skipped and there are two large bore IVs.
 
Check out Paul Marino's ICU book. He has a full chapter on PA catheters that would be great for a CA1. You eventually might not agree with everything, but it's a good first read. Here's something from it that stuck with me over the years -

"There are two fundamental problems in the criticism of the PA catheter based on mortality data. The first is the simple fact that the PA catheter is a monitoring device, not a therapy. If a PA catheter is placed to evaluate a problem, and it uncovers a disorder that is untreatable and fatal (e.g., cardiogenic shock), the problem is not the catheter, it is the lack of effective therapy. Mortality rates should be used to evaluate therapies, not measurements.

While all that sounds nice, it's not exactly true. If a measurement has potentially fatal complications, you kinda need to include mortality in your analysis. It'd be like bombarding everybody with massive radiation doses to detect cancer and then ignoring the increased incidence of cancer from all the radiation in your "measurement"
 
PA CATHS do the most good by forcing people to pay attention to the patient. Having said that in the CV ICU no one can seem to care for a patient without one and then in the Medical ICU with our cardiogenic shock patients we never use them. A number is easy to describe over the phone, pulling together enough of a feeling about a patient to render a decision without numbers takes presence at the bedside.

What i hate is the patient who is peeing, is out of bed, with normal BP but is placed on Inotrope for CI of 1.8. Drives me insane. If i make it to the patient first, i tell the nurse to pull the gosh darn thing!!

Never Fail to understand the assumption that pressure does not always correlate with Volume, and that PAD doesn't always reflect LA pressure,which doesn't always reflect LVEDP which doesn't always correlate with LV ED volume. Even if they all do correlate the PA cath tells you nothing about who the patient should respond to therapy, i.e. its static not a dynamic measure.
 
One can give pressors safely for 24-48 hours on a well-running peripheral, especially if large bore. So there is really no significant indication for the central line, if the PAC is skipped and there are two large bore IVs.

I have no problem running epi/levo on a (>=18g) non-hand peripheral that's nicely flowing, but I'd be very sure to hook the infusions directly to the catheter (not piggybacked on the IV set) using baxter tubing and tape in such a way that it would be essentially impossible to get kinked. Not having access to the arms if something goes awry when going on or coming off would be disastrous. Might be better served placing a 16g 2" IV into an EJ.
 
  • Like
Reactions: 1 user
Mman, fair point but I don't think it negates the overall message that a monitor alone shouldn't be expected to show a reduction in mortality before it's called worthless. There are numerous studies that have failed to show mortality benefits of pulse-ox. It's hard to draw conclusions from studies that investigate the use of a monitor but don't/can't control for the subsequent therapies.
 
PACs aren't totally useless. I am doing fellowship in an institution that puts PACs in everyone (and it drives me crazy). However we do a ton of heart and lung transplant, high risk triple valves, ECMO (VA, VV, VAV, VVV, VAA) and VADs. As others have stated, PACs are extremely helpful when trying to wean inotropes but also helpful when trying to wean mechanical support like IABP (though not critical), right sided impellas, RVAD centrimags etc. It's the interpretation part that gets people. I recommend you start off your presentation on the clearcut uses of PACs (filling pressures, MVO2, calculating SVR/PVR, monitoring PA pressures and getting a PCWP).
 
  • Like
Reactions: 1 user
Fwiw I don't wedge anyone anymore. Pretty much never. Doesn't change management and swan in wedge position confers substantial risk.

I think the SvO2 is the most useful number the swan gives you.

I think inotropes after vastly overused. They're bad for you. Like, really bad for you. Sometimes you need them. But if you don't NEED them, don't start them.

Most of my hearts come off on norepi alone.

Less is usually more.
 
  • Like
Reactions: 1 user
PACs aren't totally useless. I am doing fellowship in an institution that puts PACs in everyone (and it drives me crazy). However we do a ton of heart and lung transplant, high risk triple valves, ECMO (VA, VV, VAV, VVV, VAA) and VADs. As others have stated, PACs are extremely helpful when trying to wean inotropes but also helpful when trying to wean mechanical support like IABP (though not critical), right sided impellas, RVAD centrimags etc. It's the interpretation part that gets people. I recommend you start off your presentation on the clearcut uses of PACs (filling pressures, MVO2, calculating SVR/PVR, monitoring PA pressures and getting a PCWP).
I put them in if there is history of severe pulmonary hypertension or if the EF is very low because the surgeon and the ICU people want it. I don't really use it other than looking at the PA pressure. It leads to over treatment compared to just a central line either because the PA pressure is too high and someone wants milrinone or nitric oxide, or because the cardiac index is "low" and they want to make it over 2 regardless how the patient is doing. If it were up to me I would never put one in.
 
Last edited:
Also, these guidelines are for non-cardiac surgeries. What are the guidelines for cardiac cases? I feel like I heard a lot of places don't place central lines in routine valves and CABGs. They just do 2 large IVs, a-line and TEE. Like I mentioned we did all this plus the PAC.

Our perfusionists rely on cvp to ensure adequate venous drainage.
 
I have no problem running epi/levo on a (>=18g) non-hand peripheral that's nicely flowing, but I'd be very sure to hook the infusions directly to the catheter (not piggybacked on the IV set) using baxter tubing and tape in such a way that it would be essentially impossible to get kinked. Not having access to the arms if something goes awry when going on or coming off would be disastrous. Might be better served placing a 16g 2" IV into an EJ.

I think that, more important than the size of the catheter, flowing well or not, is the proximity of the catheter tip to insertion site with regard to agents such as epi or norepi...vessel trauma from insertion is commonly what causes iv's to eventually infiltrate, so the more distance you can put between insertion site and catheter tip, the better. Thus the utility of what have now come to be known as "midlines" for peripheral vasoactive/inotropic agents. Pretty useful, very reliable.
 
  • Like
Reactions: 1 users
Our perfusionists rely on cvp to ensure adequate venous drainage.

Not arguing against a central line, but can't they just look at the reservoir?
 
I think that, more important than the size of the catheter, flowing well or not, is the proximity of the catheter tip to insertion site with regard to agents such as epi or norepi...vessel trauma from insertion is commonly what causes iv's to eventually infiltrate, so the more distance you can put between insertion site and catheter tip, the better. Thus the utility of what have now come to be known as "midlines" for peripheral vasoactive/inotropic agents. Pretty useful, very reliable.
This is a very good and important post. This is exactly what studies say. But it doesn't have to be a midline; any 2 inch-catheter, 20G or larger, in a big vein above the antecubital and popliteal folds (so they won't reverse the flow if it's obstructed by limb flexion), will do.
 
This is a very good and important post. This is exactly what studies say. But it doesn't have to be a midline; any 2 inch-catheter, 20G or larger, in a big vein above the antecubital and popliteal folds (so they won't reverse the flow if it's obstructed by limb flexion), will do.

Not to put too fine a point on the topic...placing a tourniquet on an upper extremity and looking at a large deep vein in the long axis with u/s will occasionally show spontaneous retrograde "echo" contrast flow. For what its worth...
 
Not arguing against a central line, but can't they just look at the reservoir?
They would have to know the blood volume of the patient pretty accurately all through out the case for that to work. The guys ai work with are good, but that is a little too much.

It would get more tricky with bicaval cannulation. The moment they lose volume they would not be able to tell if it is svc cannula, ivc cannula, or just generalized bleeding.
 
  • Like
Reactions: 1 user
Top