Fluids in critical care - What does euvolaemia mean anyway?

Chris Poynter on 12-12-2013

Recently, I seem to be learning more and more about how wrong everything that I ever thought about fluids in ICU is.  In fact, I’m way more confused now than I was as a first year house officer patrolling the wards confidently prescribing “fluid challenges” in response to requests for an assessment of the patient’s fluid status.  The more I learn, the less I know.


Why is this? What are the issues that vex us? What do we actually know? And if we know so little, how should we manage fluids in ICU?


Fluid management is an extremely complex area of medicine. There are some extraordinarily strong opinions shown by most doctors based on very little hard evidence and poorly understood physiological processes with a resultant huge variation in clinical practice.  


Firstly, I’m not entirely sure whether euvolaemia has been well defined.  This appears to be because it is a moving target which depends on dynamic pathophysiology.  We have multiple ways of examining and testing for fluid status but none of them definitively indicate euvolaemia.  Most of the hameodynamic monitors seem to measure pressures or flows, which are poor surrogates for volume.  Urine output and other clinical markers have myriad different factors which affect them.  Some dynamic tests (such as the straight leg raise) can predict fluid responsiveness, but this still doesn’t indicate whether the patient actually needs fluid.  The risk-benefit analysis of giving fluid is often very uncertain. All we can say is that we are now more aware of risks and that the old adage that fluid is good for the kidneys and bad for the lungs may not hold true and oedema may be bad for all organs and this may have something to do with the endovascular glycocalyx. Ouch! My head hurts!


Then there is the question of what to give. The crystalloid or colloid debate has recently swung more in favour of crystalloids, while the next question appears to be whether chloride has a deleterious effect compared to balanced solutions.  What is the role of albumin, if any?


Is there any evidence for dividing up maintenance and replacement fluids or can we safely say that no one in ICU needs maintenance fluids? I’m not sure that the evidence is there to answer this question.  What about insensible losses and the mysterious “3rd space” that we learnt at medical school which we have never been able to find?  Where does the fluid go and is that important to how much we give and what type?


I’m not sure I have the answer to most of these questions and I can’t seem to convince myself that we have an evidence base which is able to add any certainty.  Unfortunately, the more studies that come out, the more questions that seem to arise.


So, what is the good news in all of this? It appears that ICU mortality continues to fall over time so we can’t be doing too badly.  I suspect that despite the uncertainty that still exists, our understanding is slowly improving.  Having listened to the recent Crit-IQ podcasts on fluid related topics by Ian Seppelt, Rinaldo Bellomo and John Kellum, there is plenty of good information out there to guide us and to help form an opinion.


For what it is worth, here is my approach when deliberating the question of whether to give fluid to a patient in ICU:


I ask 4 questions:

Why am I giving fluid?

Is it improving the patient’s condition?

Are there other options?

What are the potential consequences?


Then I continue to ask those questions as I regularly reassess the patient. The aim is to follow the goldilocks adage of not too much, not too little, but just enough.  This can only be done with a close eye on the patient, a clear diagnosis and a global view of the patient’s trajectory rather than a narrow focus on a few easy markers of wellbeing.  At the very least, this approach is likely to stop the issue of 10+L of fluid challenge in the hope that the problem will go away.


I am interested in your views on the matter. Does anyone have any clearer views than I? What evidence has changed your practice? Do you have a different approach than I? If so, what do you base it on? 


I suspect that Ian Seppelt is right in his assertion that “the best fluid is study fluid” and can only hope that we can gather more evidence over time so that by the end of my career I can regain the clarity I had at the beginning...


Got something to say?

login below or Click here to create new account

Log in to your account

Forgotten your password?

Anita from Australia wrote 12-14-2013 03:54:57 am
Great post Chris.

Fluids are like black magic. The more I try to understand them the more confused I get. The major points are fine - saline and albumin are about the same; starches and gelatine hurt your kidneys and don't have other benefits etc etc.

But the fundamental questions are complete mysteries to me - what target do I titrate against? How much fluid should I give? When are they full? Is it better to use pressers or fluids to get a "stressed volume"?

I think its likely to be more art than science for some time to come.

Alasdair from Australia wrote 12-19-2013 10:48:12 pm
Agree Anita, I can't see us making much further progress on research in fluids, or inotropes and pressers for that matter, until we have a reliable measure of global tissue perfusion that actually correlates with outcome. Until then, the protocols used in these trials would seem flawed from the start.

Harold Hastings from United States Of America wrote 12-31-2013 11:39:09 am
Interesting and insightful, Chris.
Let me suggest that we back up a bit - the multi-faced nature of hemodynamic instability, questions about liberal versus restricted fluids, and multitude of studies with a variety of outcomes seem to imply that the question is not simply "what is euvolemia?"
Instead let's take a more patient- and situation-focused approach to resuscitation. Let's assess critical factors affecting the patient's hemodynamic and perfusion status, including cardiac function as well as filling. As Phillipe Rola MD states when discussing resuscitation in "I think that blind (e.g. no echo assessment) is absurd ... ." I agree that echo is needed here, preferably on-demand TEE; see Greemhalgh and Patrick's 2012 editorial on peri-operative TEE in Anesthesia,

Alasdair from Australia wrote 12-31-2013 07:51:26 pm
In some ways I agree Harold, but in others I don't. Any marker you're getting from the Echo is still a surrogate marker. We don't know that improving indices of the macro circulation (for example heart rate, cardiac index or MAP) has any correlation with improving tissue perfusion. And that only holds if you assume (as we do) that improving tissue perfusion improves outcomes (and it may not in all cases). And even then, we still don't know that using volume to achieve this is better than using inoconstrictors to meet the same endpoints.

So we're still putting the cart before the proverbial horse...

Christopher from New Zealand wrote 01-10-2014 12:59:07 pm
Great discussion thanks.

I agree with Alasdair. I'm not sure that echo answers the critical question of "how much is enough?". I'm not saying that echo is not a useful tool. It is. It can add much to the diagnostic picture and give a dynamic picture of the pump in action.

However, the key point Alasdair makes is that most of the very strong opinions relating to fluids relate to research on surrogate measures rather than real world demonstrations of differences in meaningful outcomes.

Sid from Australia wrote 01-20-2014 12:46:45 am
If 2-3 lts are not working in sepsis,its unlikely another 2-3 lts will work, better to try vasoactive agents along with- considering hypovolemic shock is excluded and treated.- at least the use of fluids in major trials is reducing ( from 13 lts in first 72 hrs in EGDT to~ 3 lts in first 24 hrs in CHEST).
Its consistently being shown in ICU trials that when we overdo we kill- be it tight sugar control, be it higher dose of dialysis, large tidal volume or extra fluids.

regarding euvolemia- I dont know- why do we need to maintain euvolemia- I am not aware that it changes any outcome- again, normal physiology does not mean improved outcomes( Tidal volume, ICP etc).

regarding crystalloids-6S trial had shown improved survival with Hartman while CHEST, when using saline, had same mortality outcome- I often wonder, if they had compared plasmalyte or Hartman with starch -they would have shown the outcome benefit as well.

I think "abnormal "saline is bad- and should be abandoned- it anyways was never intended to be used in humans when discovered in first instance.

Jason from Australia wrote 01-25-2014 03:31:17 pm
I agree completely. The more one reads in regard to fluid therapy the more and more esoteric it becomes. I think most of the indicators we use are largely unhelpful, given there are a countless of reasons for physiological perturbation aside from a lack of fluid. Aside from the blindingly obvious hypovolaemic patient (eg. had copious diarrhoea for 3 days) determining the need for fluid is very difficult. Even more so in the patient who has been in ICU for a few days and already likely been adequately resuscitated. I often hear at the bedside "she looks dry" and wonder how accurate these assertations are. My guess they are probably right about 50% of the time. How can we do it better? I think defining what euvolaemia is would be great starting point. It may give an indication if a patient actually needs fluid, as pointed out a different concept to fluid responsiveness. Perhaps trying something relatively simple like body weight might be as good a marker as any.

ICU baby from Australia wrote 01-26-2014 07:58:31 pm
I think any attempt at defining euvolaemia is a waste of time. There is no such thing in an ICU patient.

The key issue is whether or not the patient will do better if we give them fluids. The evidence at present suggests we overdo it. So, a population wide approach to reducing our reliance on fluids to achieve "eu-numberaemia" is important. How we get there remains to be determined.




Stay Tuned


Recent Posts


Recent Comments