PA Catheters in the ICU

[surg]

Anyone catch the early release article on the NEJM website on CVC vs. PA catheter in Acute Lung Injury? http://content.nejm.org/cgi/reprint/NEJMoa061895v1.pdf

The conclusion is basically that PACs placed within the 1st 48 hrs of ALI (in patients that didn't get one immediately) and which had no reason that they shouldn't be weanable, without CHF or COPD, don't reduce mortality and cause more arrhythmias. This is consistent with retrospective studies and prior smaller randomized studies previously published that were heavily criticized.

It's hard to know what to make of a study that of 11,511 patients screened only managed to enroll 1001. The thing that bothers me the most is that of the screen failures 20.8% were not enrolled because they already had PACs and another 15.9% had the physician decline. It seems like this already pre-selects out all the people who the physicians thought would definitely benefit from a PAC, leaving the trial to figure out whether a PAC was helpful in borderline cases where the physician was already indifferent between a CVC and a PAC.

Also, of note, they did not use mixed venous oxygenation for management relying on PAOP and Cardiac index.

On the other hand, despite my training in using PACs, I would have a hard time finding a good study that supports its use either, beyond my own personal experience in knowing that in some patients I really like knowing the numbers, especially the SvO2. Not necessarily really a good enough reason to put on in, I grant you, but it's what I've got for now.

Any thoughts? Would be interested in what the reaction to this is around the country.

---

Members don't see this ad.

 

[militarymd]

I think the PAC is slowly changing its role in the ICU. It is, afterall, only a monitor...and we all know that monitors don't make you better...just like having a ECG leads on all the patients....the info doesn't make you better, it is changes in therapy that the info causes that affects outcome.

It was a end all be all monitor....and slowly we're figuring out that a lot of the therapeutic decisions that we make based on the data doesn't make patients' better. ....

I think it will always have a place in the care of critically ill patients, but its role will be much smaller than it was a decade ago....I know my personal use of the PAC is slowly withering away.

 

[Eidolon6]

It's hard to know what to make of a study that of 11,511 patients screened only managed to enroll 1001. The thing that bothers me the most is that of the screen failures 20.8% were not enrolled because they already had PACs and another 15.9% had the physician decline. It seems like this already pre-selects out all the people who the physicians thought would definitely benefit from a PAC, leaving the trial to figure out whether a PAC was helpful in borderline cases where the physician was already indifferent between a CVC and a PAC.

This issue of enrollment was brought up at a review of the study design and preliminary data at my instititution. I had a part in enrolling a few patients both at my current training program and my residency training program. The fact that 21% patients already had PACs at time of enrollment is definitely indicative of the current practice trend (and perhaps the reason this study needed to be performed) as mounting smaller studies have failed to demonstrate a definite benefit to PAC and a trend towards increased morbidity in some. Additionally, the protocol to follow in the study was quite involved and understandably, a physician may feel a sense of relinquishing control (though they had power over the protocol should they choose to use it)...which can be difficult to accept when enrolling a critically ill patient in a study of the unknown.

The study reached its enrollment and power goal of 1000 patients and in that, the statistics are pretty sound...no benefit to be had. That being said, certain camps will malign the findings of this study because the largest proportion of enrollment were medical intensive care patients. The fact of the matter is that the study was extended beyond the MICU, but many physicians were unwilling to make clinical decisions without the PAC....despite little strong evidence to support its use in clinical decision making. Its hard to relinquish that warm blanket of information that one thinks the PAC provides because many can't understand why use of so much information doesn't lead to better outcomes. Thus the impact of this study in say surgical critical care managment of ALI/ARDS remains to be seen, but as with other well designed evidence based approaches in the ICU, will not make the cross over to some disciplines (lung protective ventilation, PROWESS). This is perhaps unfortunate and indicative of how hard it is to break practice habits (much like drug habits) in the face of strong contrary evidence. I can see the trauma camp, for example, who believe their patients somehow behave "differently" physiologically, being uninspired to accept this data and will in kind refuse to refine their PAC practice.

A strong aspect of the study was that is was truly a big multicenter randomized control trial for a modality that is extremely hard to study. All studies have there limits, but even with those, this study is really a monumental feat in logistics when you consider all the centers involved and the details addressed.

I can't say this study will change my current practice much as I am not a big user of the PAC (though I believe it is imperative to understand it) and have found it (much like this study) generally unhelpful (and often offering conflicting information) in management of my ALI/ARDS patients. But as MilitaryMD, implies, its a tool thats probably too complex for its users but will maintain a role in critical illness...its just not as sexy as it once was and we're finally proving it and for some patients...that will be a good thing.

 

[bullard]

Reading through the Surviving Sepsis guidelines when the PAC article came out. Would you routinely put PACs in septic patients or is CVP, MAP, and urine output enough to go by? Can you use inotropes on these patients without the PAC?

I think the PAC is slowly changing its role in the ICU. It is, afterall, only a monitor...and we all know that monitors don't make you better...just like having a ECG leads on all the patients....the info doesn't make you better, it is changes in therapy that the info causes that affects outcome.

It was a end all be all monitor....and slowly we're figuring out that a lot of the therapeutic decisions that we make based on the data doesn't make patients' better. ....

I think it will always have a place in the care of critically ill patients, but its role will be much smaller than it was a decade ago....I know my personal use of the PAC is slowly withering away.

 

[southerndoc]

Reading through the Surviving Sepsis guidelines when the PAC article came out. Would you routinely put PACs in septic patients or is CVP, MAP, and urine output enough to go by? Can you use inotropes on these patients without the PAC?

Most places use central venous pressures and central venous O2 saturations instead of wedge pressures and mixed venous O2 saturations in their sepsis protocols.

 

[TofuBalls]

In our SICU we use mostly LiDCO monitors which give almost everything the PAC does except the wedge. We seemed to get pretty good data from them. If we had the central venous SvO2 catheters it would probably be even better. Anyone else out there have any experiences with the LiDCO vs. PAC?

 

[Eidolon6]

In our SICU we use mostly LiDCO monitors which give almost everything the PAC does except the wedge. We seemed to get pretty good data from them. If we had the central venous SvO2 catheters it would probably be even better. Anyone else out there have any experiences with the LiDCO vs. PAC?

The Lidco has little evidence based validation for use and given that the evidence is slanted towards the information gained from PAC use does not benefit clinical care...it is likely that this device, although fancy, will demonstrate similar therapeutic neutrality. The editorial in NEJM accompanying the PAC trial brings up an interesting point that PAC use caught on quickly in the 70s and 80s without any validation of its worth and now physicians use it with blind faith and make the assumption that it assists in the care of the patient. I fear the LiDCO may develop a similar following.

Personally, in the NCU and SICU, I have found the LiDCO neat to look at but the information was marginally helpful at best.

 

[RuralMedicine]

Reading through the Surviving Sepsis guidelines when the PAC article came out. Would you routinely put PACs in septic patients or is CVP, MAP, and urine output enough to go by? Can you use inotropes on these patients without the PAC?

I think your practice is shaped both by those who trained you and the resources of the institution you currently practice at. At the institution we (IM/Pulm/Critical Care) did not empirically use PA catheters. We tended to use them in critically ill patients with underlying heart or lung morbidities more as an "additional data point" if things were not improving. We did use CVP and MAP extensively and that is my practice now. In sepsis I initiate/titrate pressors off of MAP and use CVP to gauge volume status. With CVP trends are more helpful than absolutes, and I've found it helpful to check CVPs before and after bolusing to gauge what the patients normal volume replete CVP is.