Third spacing and fluids

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

Radetzky

Full Member
7+ Year Member
Joined
Nov 9, 2016
Messages
423
Reaction score
637
Wondering how you all approach those complicated patients who are total body overloaded but intravascularly dry, as there doesn't seem to be a great deal of evidence based medicine to guide us.

Let me give an example- patient with bad pancreatitis and hypoalbuminemic with an oxygen requirement and maybe a small effusion but otherwise not too terrible an Xray, who is persistently oliguric and has worsening renal function.

Common practice seems to be to keep giving fluids with the aim of maintaining a semblance of intravascular volume but this never seems to work IMO. Sometimes we give albumin but again I've never really seen this make a dramatic difference, and there's not too much literature out there.

Then there's the pieces out there like this one PulmCrit- Killer resuscitation: Abdominal hypertension as an occult driver of multiorgan failure that suggest these patients might actually be developing elevated intra-abdominal pressures that are actually responsible for the renal dysfunction which we are therefore making worse with our fluids. As much as this makes sense, diuresing these patients somehow feels wrong.

Curious as to your approaches. In my part of the world ICUs often won't accept 80 year olds with multiple comorbidities who often end up being exactly the population that develops these problems, so we are often left managing them on the wards.

Members don't see this ad.
 
albumin ^^ .

Sent from my Pixel using Tapatalk
 
  • Like
Reactions: 1 user
Not that I haven't done it myself but...

*citation needed*

Have no evidence for that or for some of the other things I do. If patients seem to do ok I continue if they don't I stop
 
Not really a fan of albumin. Don't find it's much use unless it's <1 which is not that common. Just ends up biting you when it seeps into the tissues and screws up whatever sort of oncotic gradient you're trying to create.
 
You need to be doing better physical exam if you are missing IAH and ACS as the cause of your renal dysfunction. Unless you are talking about the dituation where you have identified the IAH and can't decide if it is ACS or not. That said, sometimes you need need to flood the patient (will let you all debate colloid vs crystalloid) and buy a vent for a bit. Just depends on the underlying cause. I think it is one of those areas that fall into the art of medicine versus the science.
 
  • Like
Reactions: 1 users
Thanks for replies. Albumin seems to be what we mainly do as well. But often they've received litres upon litres of crystalloid before that's considered.

I guess the article is talking mainly about IAH from excessive fluid resuscitation. Given they comment on its degree of underrecognition I'm not sure that it's simply a case of doing a better physical exam.
 
I was actually going to start a new thread about third-spacing in the medical students section (MS4). But I thought it would be okay to add to this one. My question is:
So third-spacing is a cause of hypovolemic hyponatremia. But, cirrhosis, nephrosis, and CHF are causes of hypervolemic hyponatremia. From my understanding cirrhosis, nephrosis, and CHF all cause third-spacing. Cirrhosis (increased hydrostatic/portal pressure, and decreased oncotic pressure), nephrosis (decreased oncotic pressure), and CHF (increased hydrostatic pressure). I was wondering if someone could be so kind as to explain to me the difference. Many thanks in advance!
Edit- I understand that cirrhosis, nephrosis, and CHF would all cause decreased perfusion of the kidneys, resulting in activation of RAS system and thus increasing intravascular volume. But wouldn't third-spacing do the same thing?
 
Where is the evidence that this mysterious "third space" even exists? Some people use that term to mean the interstitium, but the classical understanding (from old-school anesthesiology) is entirely different.

What you are talking about is essentially distributive shock. Septic shock is a type of destributive shock, which people often seen to forget or misunderstand. The majority (~55-60%) of septic patients are not volume-depleted as a general rule. They are profoundly venodilated and have a degree of capillary leak, but intrinsically are not depleted. Hence why giving every septic patient a 30cc per kg bolus is idiotic, because in most of these patients it goes right into the interstitium causing tissue edema and further damaging the endothelial glycocalyx and interfering with microcirculatory flow from back pressure.

I'm not entirely sure how much albumin

Albumin is not a volume expander in the sense taught in medical school, because there is no colloid-osmotic pressure gradient between the intravascular space and the interstitium (its actually between the intravascular space and the sub-glycocalyceal space). Glycocalyx in Critical Illness

Here is a good lecture from the PI of the SAFE trial: http://intensivecarenetwork.com/myburgh-fluids-2015/

Sent from my SM-N910P using SDN mobile
 
Last edited by a moderator:
Where is the evidence that this mysterious "third space" even exists? Some people use that term to mean the interstitium, but the classical understanding (from old-school anesthesiology) is entirely different.

What you are talking about is essentially distributive shock. Septic shock is a type of destributive shock, which people often seen to forget or misunderstand. The majority (~55-60%) of septic patients are not volume-depleted as a general rule. They are profoundly venodilated and have a degree of capillary leak, but intrinsically are not depleted. Hence why giving every septic patient a 30cc per kg bolus is idiotic, because in most of these patients it goes right into the interstitium causing tissue edema and further damaging the endothelial glycocalyx and interfering with microcirculatory flow from back pressure.

I'm not entirely sure how much albumin

Albumin is not a volume expander in the sense taught in medical school, because there is no colloid-osmotic pressure gradient between the intravascular space and the interstitium (its actually between the intravascular space and the sub-glycocalyceal space). Glycocalyx in Critical Illness

Here is a good lecture from the PI of the SAFE trial: http://intensivecarenetwork.com/myburgh-fluids-2015/

Sent from my SM-N910P using SDN mobile

Looking around google, the "consensus" definition for third-spacing is loss of volume that cannot participlate in ICV/ECV equilibrium, things like sepsis, trauma, fractures... I just don't understand how this depletion of volume wouldn't lead to activation of RAS and "reflective" hypervolemia.

In response to your discussion about septic shock, do you personally still give the bolus then? Your argument makes complete sense. I guess maybe the bolus would temporarily increase the intravascular volume and thus BP before it does get extravasated, and so it gives you a small amount of time for other stuff to kick in that would give a longer lasting effect on the BP? Interested to hear your opinion.

Edit: again, I'm still only MS4 so just trying to learn from you guys, the more seasoned veterans. :)
 
I was actually going to start a new thread about third-spacing in the medical students section (MS4). But I thought it would be okay to add to this one. My question is:
So third-spacing is a cause of hypovolemic hyponatremia. But, cirrhosis, nephrosis, and CHF are causes of hypervolemic hyponatremia. From my understanding cirrhosis, nephrosis, and CHF all cause third-spacing. Cirrhosis (increased hydrostatic/portal pressure, and decreased oncotic pressure), nephrosis (decreased oncotic pressure), and CHF (increased hydrostatic pressure). I was wondering if someone could be so kind as to explain to me the difference. Many thanks in advance!
Edit- I understand that cirrhosis, nephrosis, and CHF would all cause decreased perfusion of the kidneys, resulting in activation of RAS system and thus increasing intravascular volume. But wouldn't third-spacing do the same thing?


You are thinking about everything from a compensated point of view. Realize that a very sick patient who is anasarcic has seriously decompensated. This is different that your average person with swollen legs.

These patients are by far the hardest to treat, because you are trying to treat one thing, but making something else worse at the same time. A fine balance tends to prevail, and these patients linger in the hospital for a long time. They are intravascularly dry, but usually have very large + fluid balances. Total nightmare.

I agree with above, we tend to treat these people with pressers to maintain MAP as if they were in shock, and fluid restrict them (especially if they are on the vent, getting people off the vent is a major priority). The albumin or albumin +lasix push is not backed up by solid evidence, but almost everyone has done it before. It falls under the "tell me another idea and I'm all for it" approach.

Seriously though, hardest patients to treat......
 
Looking around google, the "consensus" definition for third-spacing is loss of volume that cannot participlate in ICV/ECV equilibrium, things like sepsis, trauma, fractures... I just don't understand how this depletion of volume wouldn't lead to activation of RAS and "reflective" hypervolemia.

In response to your discussion about septic shock, do you personally still give the bolus then? Your argument makes complete sense. I guess maybe the bolus would temporarily increase the intravascular volume and thus BP before it does get extravasated, and so it gives you a small amount of time for other stuff to kick in that would give a longer lasting effect on the BP? Interested to hear your opinion.

Edit: again, I'm still only MS4 so just trying to learn from you guys, the more seasoned veterans. :)

These patients arn't actually in septic shock (i mean some of them are, but thats not what we are talking about here.) They have characteristics of someone in distributive shock (of which septic is ONE kind, anaphylactic would be another). Think large expansion that the volume distributes to, loss of PVR, and loss of pressure.

The bolus for sepsis is one of the most debated about things in all of medicine, especially since CMS has mandated a 30cc/kg bolus in patients with septic shock as part of their bundled payment. They are threatening to withold money if they compliance rates are not high enough, so admin is quit literally halfway up our A** about it.

According the the bundles SIRS+Source+ANY BP less than 90 systolic = MANDATORY 30cc/kg bolus. Patient has EF of 15% and pulmonary edema, they say VENT them to get the bolus (facepalm)

There has been no less than 10-15 solid articles disputing such a one size fits all approach since then, citing lots of evidence that positive fluid balances are harmful for patients, and that a more moderate approach should be to attempt to determine if a patient is even responsive to a fluid bolus, then providing small boluses (250-1000cc) at a time, with frequent reevaluations, to guide therapy.

Marik et al published a great review article in Critical Care last year titled "Fluid responsiveness and the six guiding principles of resuscitation" which I think is an excellent place to start.
 
According the the bundles SIRS+Source+ANY BP less than 90 systolic = MANDATORY 30cc/kg bolus. Patient has EF of 15% and pulmonary edema, they say VENT them to get the bolus (facepalm)
Yeah when I first heard that was when I started questioning everything about the sepsis paradigm



45:30 is the most savage takedown of this mandate
 
Yeah when I first heard that was when I started questioning everything about the sepsis paradigm



45:30 is the most savage takedown of this mandate


I have seen this! Excellent piece. Weingart in all his weingarty glory.

Most mandated things really are good medicine, but the sepsis bundle is just plain bad medicine, and it's all admin talks about because so much money is on the line. You can't blame the admin's, were talking millions and millions of dollars on the line.

With so much backlash in the academic medicine community, and the large range of evidence that is emerging, I predict a retraction of the sepsis bundle, or at least a major change that eliminates the bolus within 2 years.
 
  • Like
Reactions: 1 user
With so much backlash in the academic medicine community, and the large range of evidence that is emerging, I predict a retraction of the sepsis bundle, or at least a major change that eliminates the bolus within 2 years.

I doubt it. I mean, we still give kayexalate in hyperkalemia and epinephrine in cardiac arrest despite a lack of evidence in both and the hospital wouldn't be penalized if we stopped doing those. 2-3L NS bolus? Funding and quality measures.
 
I was actually going to start a new thread about third-spacing in the medical students section (MS4). But I thought it would be okay to add to this one. My question is:
So third-spacing is a cause of hypovolemic hyponatremia. But, cirrhosis, nephrosis, and CHF are causes of hypervolemic hyponatremia. From my understanding cirrhosis, nephrosis, and CHF all cause third-spacing. Cirrhosis (increased hydrostatic/portal pressure, and decreased oncotic pressure), nephrosis (decreased oncotic pressure), and CHF (increased hydrostatic pressure). I was wondering if someone could be so kind as to explain to me the difference. Many thanks in advance!
Edit- I understand that cirrhosis, nephrosis, and CHF would all cause decreased perfusion of the kidneys, resulting in activation of RAS system and thus increasing intravascular volume. But wouldn't third-spacing do the same thing?

I've always used third spacing to refer to the distributive situation described above- where fluid is lost from intravascular space into interstitium. CHF is not a cause of third spacing as both intravascular and interstitial compartments are overloaded.

You are correct in saying that cirrhotic/nephrotic patients are often total body water overloaded. Your confusion I think lies in the variable definition of volemia- some people use it to describe the intravascular space while others use it to describe total body water.
 
I've always used third spacing to refer to the distributive situation described above- where fluid is lost from intravascular space into interstitium. CHF is not a cause of third spacing as both intravascular and interstitial compartments are overloaded.

You are correct in saying that cirrhotic/nephrotic patients are often total body water overloaded. Your confusion I think lies in the variable definition of volemia- some people use it to describe the intravascular space while others use it to describe total body water.

I think I tend to consider volemia as total body water. When answering questions or seeing patients, if I notice dry mucosal membranes, decreased skin turgor, increasedcapillary refill time, I would describe the patient as in a state of hypovolemia, or volume-depleted, which I think is less controversial. On the otherhand, if I notice (pitting) edema (perhaps nonpitting too?), elevated JVD, ascites, or bilateral crackling, I will consider the patient as hypervolemic or in a state of volume-excess. Would you say that this is fair, or can you think of exceptions in which I would be incorrect in doing so?

Edit: Also, thank you everyone for posting such a discussion about this topic and sepsis. I never knew that the management of sepsis can be so controversial, and for helping me think a bit more deeply about such issues.
 
I doubt it. I mean, we still give kayexalate in hyperkalemia and epinephrine in cardiac arrest despite a lack of evidence in both and the hospital wouldn't be penalized if we stopped doing those. 2-3L NS bolus? Funding and quality measures.


Those things are on a completely lower level than the Sepsis bundled payments. Your examples cite a lack of evidence of efficacy, about things that are not tied to money (ie. the hospital doesn't care if you do or don't do them). It falls into the larger category of many treatments that probably don't work as well as we hope, but really what's the harm?

The sepsis bundled payments by CMS are an entirely different beast, and frankly, a complete travesty. 30cc/kg bolus is mandated by CMS as a quality measure (one of many) for anyone with sepsis and a systolic BP less than 90 at ANY time during their hospital stay. If you don't do it, your hospital will be docked millions of dollars, the pressure is on.

There is a considerable and growing body of evidence that the 30cc/kg bolus across the board for every septic shock patient does considerable HARM, and we are being forced to HARM many of our patient's by the government or face dire financial consequences.
 
Those things are on a completely lower level than the Sepsis bundled payments. Your examples cite a lack of evidence of efficacy, about things that are not tied to money (ie. the hospital doesn't care if you do or don't do them). It falls into the larger category of many treatments that probably don't work as well as we hope, but really what's the harm?

The sepsis bundled payments by CMS are an entirely different beast, and frankly, a complete travesty. 30cc/kg bolus is mandated by CMS as a quality measure (one of many) for anyone with sepsis and a systolic BP less than 90 at ANY time during their hospital stay. If you don't do it, your hospital will be docked millions of dollars, the pressure is on.

There is a considerable and growing body of evidence that the 30cc/kg bolus across the board for every septic shock patient does considerable HARM, and we are being forced to HARM many of our patient's by the government or face dire financial consequences.

There's a growing body of evidence that Epi is harmful and Kayexalate is treated like a God for hyper-K (despite official FDA warnings of necrotic bowel from kayexalate/sorbital combinations) . If we can't get rid of treatments that cause harm that aren't quality measures, why would we even lift a finger for things with quality measures?

By the way, what would your response be if someone suggested not giving Kayexalate to a hyperkalemia patient or not giving epi to asystole?
 
There's a growing body of evidence that Epi is harmful and Kayexalate is treated like a God for hyper-K (despite official FDA warnings of necrotic bowel from kayexalate/sorbital combinations) . If we can't get rid of treatments that cause harm that aren't quality measures, why would we even lift a finger for things with quality measures?

By the way, what would your response be if someone suggested not giving Kayexalate to a hyperkalemia patient or not giving epi to asystole?


I dont see how Epi can be harmful in cardiac arrest.......they are currently dead. Doesn't get worse than that....Unless your hypothesis is those patients with ROSC who got Epi do worse in the long term, which would be impossible to actually establish, there are too many variables. I'm afraid I could not intelligently discuss Kayexalate in hyperK, it seems that many admitting physicians don't really care too much about it, i suppose it all depends on why the patient has hyperkalemia. I would not say it is a huge part of my practice. I have given it, and I have not given it.

Either way, your points detract from the conversation. Aggressive uniform mandated large volume resuscitation to ALL patients in septic shock is BAD medicine. It is bad medicine whether or not we use Kayexalate in hyperK or Epi in cardiac arrest. Those things are irrelevant.

It's like saying "you should really stop smoking, it's bad for you", and your response is "Well, I shoot heroin and that's bad for me, so why would I stop smoking" Really, you should stop all of those things.
 
I dont see how Epi can be harmful in cardiac arrest.......they are currently dead. Doesn't get worse than that....Unless your hypothesis is those patients with ROSC who got Epi do worse in the long term, which would be impossible to actually establish, there are too many variables.

Not to derail the thread any further but here's the current evidence:
1. Systematic review from 2016 showing that the current aggregate evidence does not indicate any strong long term survival or neurological recovery benefit from the use of epinephrine.

2. Some (this and this) observational studies and and perhaps this RCT, all of which are referenced in the above meta analysis, and roughly say that while ROSC is achieved, long term outcomes are not significantly better.

3. Cohort study showing that for non-shockable rhythms out-of-hospital, a time <20 min to first epi dose improved outcomes both ROSC-wise as well as at 1 month.

It's interesting that the data seems to indicate that PEA/asystole do well with epinephrine, perhaps it supports the hypothesis that increasing epicardial perfusion with the use of epinephrine allows for more responsiveness of the myocardium to a shock when the rhythm does become shockable.

Anyway, you're right, there are a lot of variables and this is a difficult subject to study rigorously in humans. The focus is also largely on out-of-hospital arrests where the mortality is higher and there is more variability in the implementation of ACLS due to practical reasons.

The SPS intestinal necrosis is seen largely due to co-administration with sorbitol (rather than water, which can be used just fine) as a cathartic due to its propensity to cause constipation. It may actually be the sorbitol that drives the adverse effect (0.2-1.8% incidence). It also takes a long time to act, so there needs to be a good reason to want to reduce the intestinal absorption of potassium that can't be done through the implementation of a strict renal diet.

Back to the topic of the bolus. Yes, it is a poorly thought out solution to the staggeringly variable implementation of the early care of septic patients. But the bundle got the ball rolling on the topic, so perhaps we will eventually have the body of evidence required to go back to saying "think through the clinical situation".
 
Top