An observation on fluids and heart failure

[Radetzky]

I have been thinking recently about fluid therapy and heart failure and wonder whether we are sometimes overly cautious. Let me explain my thinking and see if you agree.

Let's suppose we have a healthy bloke who has gastro and has come in dry but hemodynamically stable, and we prescribe him 1L of saline over 30 minutes to start off with. Now take the same situation but make it someone with a poor EF. We might be inclined to go much more slowly. But is this really necessary? Or take the "infamous" 250ml bolus.

First of all, the total volume given should not matter. If someone is down 3L, they need 3L regardless of the pump function.

Second is the rate. We might be cautious of providing a rate that is too much for the failing heart to handle. But consider the average cardiac output; 6L/min. Since CO = venous return, venous return is 6L/min in a healthy person. Lets halve this for someone with a bad pump and make it 3L/min. A failing heart is still coping with 3L/min of "preload". If we take 1000ml/30 minutes this is 30mls/minute of saline in addition to this. Therefore venous return has been increased from 3L/min to 3.03L/min with our "fast" fluid therapy. Hardly seems significant.

This is not to mention that of the 1000mls we have given, only about 200mls will actually stay intravascular.

Thoughts?

---

Members don't see this ad.

 

[Psai]

Well I think its more like 250 to 300 and it redistributes over a few hours but I totally agree. I'm so sick of im people treating people like they're going to have cardiovascular collapse from a 500 bolus but don't seem to realize that the gram of vanc they're ordering also comes in 500 mL bags.

Anyone who gives a 250 bolus should have their medical degree revoked.

 

[Crayola227]

I didn't follow all that math. Pearl I was taught: volume status can only be clinically determined at bedside, and it ain't easy.

For most patients, figuring out with that level of precision that they're down 3L, is just silly and not realistic
for all that math, determining hypo, eu, or hypervolemic, is still a bedside thing.

Fluids go in faster than diuretics make them piss them out when you flood the lungs.
I assume we're not talking a sepsis/shock scenario, because in that case you wouldn't have calculated so precisely what they're "down", and resuscitation takes precedence within reason. In that case I was told MAP takes precedence, flood them to keep it up, you can put them on a vent and worry about getting the fluid out later if that's what it takes. But you don't want to do what I just said to someone not in shock, but just "down" fluid.

You point out, there's all sorts of other fluids that are easy to forget about, even the saline flushes from the nurses add up. Couple that with the fact that depending where you work, whole bags of fluids are being given in the ED and no record of it. Patients are dumping the hat of piss and flushing. Incontinence. Even an institution with great strict I/O procedures, if you rely too much on those numbers, you'll get burnt one day.

So I/Os are not a good way to keep track. Brings us back to bedside.

250 mL is like homeopathic, and I agree a silly waste of time that I never ordered.
If I wanted the patient to have that much I would have the nurse bring them H2O or juice if they're not totally NPO,
if someone was insisting I give that amount, 250 mL, I would insist they don't need IVF.
IVF are for when you need to give fluids faster than is reasonable by mouth, things that can only be given with a drip going, for when the patient is NPO AND needs the fluids. (not everyone NPO overnight needs fluids, humans can go a day without fluids even when ill depending on what's wrong).

1L is more than you think, especially in a trainwreck you just got and you have no idea how their physiology likes to space things. The numbers only tell you so much. 1L won't break most people especially when you have good cause to think they're down.

However, I've seen trainwrecks that did need to be managed on the order of 500 mL boluses beyond that. I've done 500 mL, seen how it was tolerated, and added another. When we give IVF rate not just total matters too. This is one way to gauge how fast to be giving that 1L, when you break it into 500 mL. Doing this in multiples of 500 mL really can make sense, because sometimes how much someone needs really can't be measured on the order of liters. Sometimes 2.5 L is the sweetspot, 2L would have been fine in that case too, but 2.5 L was better, but 3L would have been too much.

It depends too, if I was being ginger like this with someone's fluids, I'm constantly reassessing, and the only one being bothered by this is me getting the pages and the nurse getting the orders, the patient if they're being woke up to hang a bag, assuming I have the luxury with their condition of dragging my feet a little on IVF and doing this by 500 mL amounts rather than L.

This is also a way to give fluids when you haven't seen the patient right before (fluid status is determined bedside) but you're giving fluid sight unseen, and going by the numbers to tell you they need a little more. If you aren't reassessing bedside after each bolus, than it makes sense to go a little slower, because if they're doing OK but you think they need a little fluid, going slow is going to get you into less trouble, going fast can get you in trouble quick.

TLDR:
whatever you do, don't just do some math to decide how much/how fast with IVF (except for fluid resuscitation guidelines in sepsis, even then, keep your head)
go bedside, go slow, see how tolerated
250 mL is bull****, give the patient some damn ice chips instead
500 mL boluses are sometimes warranted to find the sweet spot
I've seen people flooded with no good way to get the fluid off, don't always assume you're gonna be able to squeeze the kidneys as much as the heart/lungs need later, it's sad to have hurt someone's already hurt heart or beans, or now they're in the ICU on a vent with VAP, because you weren't a little more patient with fluids or reassessing bedside
Don't get cocky with fluids, or trusting those I/O #s, it will bite you hard one day

 

[Raryn]

The new surviving sepsis guidelines (just out in the last few weeks) explicitly state even patients with CHF and CKD should start out with a 30cc/kg bolus if they're presenting with sepsis. It makes sense too, because at the time of treatment for sepsis, their chronic comorbidity isn't their most acute problem.

 

[Druggernaut]

I agree that people tend to be too conservative with fluid resuscitation in CHF and CKD, and I don't think I've given a 250 mL bolus other than in pediatrics. I'd still advise you to not be too cavalier. You don't magically know that they're down three liters, and they're not a bucket that you just pour fluids in until it's just short of flowing over the top. I finally overdid it with a patient in my second year of residency, when a single liter of IVF in a tachycardic abdominal pain patient with only a known medical history of hypertension ended up with her admitted for pulmonary edema a few hours later when it redistributed, her sats dropped, and she had a fluffy chest x-ray (but negative abdominal CT). I've given untold volumes of fluid before that ever happened without adverse outcome, as far as I know, but it's not impossible to do harm, and this lady was not your 5-10% EF patient.

This is not to mention that of the 1000mls we have given, only about 200mls will actually stay intravascular.

That's the concern. You don't get edema or shortness of breath because the fluids are intravascular, you get it because it's into the soft tissue and alveoli.

The new surviving sepsis guidelines (just out in the last few weeks) explicitly state even patients with CHF and CKD should start out with a 30cc/kg bolus if they're presenting with sepsis. It makes sense too, because at the time of treatment for sepsis, their chronic comorbidity isn't their most acute problem.

From the guidelines: "1. Sepsis and septic shock are medical emergencies, and we recommend that treatment and resuscitation begin immediately (BPS). 2. We recommend that, in the resuscitation from sepsis induced hypoperfusion, at least 30mL/kg of IV crystalloid fluid be given within the first 3 hours (strong recommendation, low quality of evidence). 3. We recommend that, following initial fluid resuscitation, additional fluids be guided by frequent reassessment of hemodynamic status (BPS)."

You have three hours to get it in, so you can do it in a stepwise fashion with frequent reassessments rather than when they hit the door. This is also a measure that many disagree with, including our faculty who were authors of the guidelines. Unfortunately, it's also tied to CMS payment now, and as much as our hospital could use the money, we've been advised not to drown our 200 kg CHF patients in 6 liters of fluid and to instead document why we'd rather not commit murder. Guidelines like this are good for those who don't deal with sepsis that often and would otherwise stick the patient in a corner with 100 mL/hr of fluids like many did pre-Rivers, but still should be thought through before following blindly.

Here's some other interesting points to consider about the guidelines: https://emcrit.org/pulmcrit/sepsis-myths/

 

[Radetzky]

Thanks for the quality replies. I feel I need to clarify a few points. The 3l comment was more a theoretical visualization. I agree that it is impossible to determine the exact amount. My point exactly is that it is frequent reassessment which is important, not some arbitrary difference between running it in over 1hr vs 4hrs.

I realise that redistribution causes edema. However the reason hf patients redistribute more is secondary to intravascular overload. And when we talk about frank starling and overwhelming pump function it is intravascular volume that matters.

 

[cbrons]

The new surviving sepsis guidelines (just out in the last few weeks) explicitly state even patients with CHF and CKD should start out with a 30cc/kg bolus if they're presenting with sepsis. It makes sense too, because at the time of treatment for sepsis, their chronic comorbidity isn't their most acute problem.

They are also apparently "weakly recommending" following dynamic measures of fluid responsiveness, e.g. arterial PP variation.

The 30cc/kg bolus for everyone doesn't really make sense to me even after trying to research the issue quite a bit. I've heard from an online lecture recently that 50% or more of patients who are septic aren't fluid responsive at all. In that case, IVFs aren't doing anything beneficial. Am I wrong?

EDIT: Here is the video I was talking about from a conference with Drs. John Myburgh and Paul Marik about the concept of "iatrogenic salt-water drowning"



45:20 is the relevant part about fluids

EDIT 2: found the transcript of the relevant quote if you don't want to load the video:

I think it's important to realize and we need to say this clearly: sepsis is not a fluid-depleted state. People with sepsis do not have diarrhea, they do not have cholera, they do not have diabetic ketoacidosis. They are not volume-depleted as a general rule. They are profoundly venodilated with some degree of capillary leak. But intrinsically, it is not a volume-depleted state. Secondly, most patients who are septic are not volume responsive. The only reason to give fluid is to increase stroke volume. If you do not increase stroke volume, it serves no purpose. And only about 30 to 40 percent of patients who present with sepsis will respond to fluid. What's even more shocking is the duration of the hemodynamic effect is very short-lived. What you're doing is shifting the fluid in the bag through the circulation into the interstitium and causing profound tissue edema. It's a condition called iatrogenic salt-water drowning.

Yes, there are some patients who will be volume-responsive, you know, a little old lady, she’s septic with poor oral intake, may be a little volume depleted. You can give her a little fluid. But this idea of a giving everyone a thirty-cc per kilogram "volume challenge" is outrageous and defies understanding of human physiology. Because in fact we know above a CVP of 6, the diastolic compliance of your RV goes up significantly, you increase diastolic distention, you increase release of atrial natriuretic peptide, and you screw with your glycocalyx. A high CVP actually impairs capillary and microcirculatory flow from back-pressure. Volume overload is a bad thing, and I think we actually kill more people with fluids than we help.

 

[tiedyeddog]

Anyone who gives a 250 bolus should have their medical degree revoked.

eh, I do it frequently to get nurses to leave me alone.

 

[Apollyon]

eh, I do it frequently to get nurses to leave me alone.

I smirk at the people that aren't normally meatheads that gave the person that posted that a "like".

 

[BoardingDoc]

eh, I do it frequently to get nurses to leave me alone.

A 250cc bolus isn't medicine, it's social engineering. Sometimes it pays to keep everyone happy.

 

[SurfingDoctor]

2 points:
1) Anyone who claims to fully understand the pathophysiologic mechanisms of sepsis is lying.
2) Leg raising and external hepatic compression will give you insight into fluid-responsiveness without giving fluid.

 

[Radetzky]

I realised the flaw in my initial hypothesis, which was not taking into account what happens when you get your fluid assessment wrong; much better to be running at a reduced rate when they start displaying signs of overload than just have dumped another litre.

2 points:
1) Anyone who claims to fully understand the pathophysiologic mechanisms of sepsis is lying.
2) Leg raising and external hepatic compression will give you insight into fluid-responsiveness without giving fluid.

Although the initial post was focussed more on "true" volume depletion, interesting discussion around fluids in sepsis. My question to the above is how do we incorporate the leg raise into assessment on the floor? Assessments of fluid responsiveness with PLR have all been studied in the ICU setting where they have the skills and technology to assess things such as cardiac output, PPV, carotid doppler etc. However the initial fluid resuscitation will be happening on the floor. All I have at my disposal is physical exam and, on the occasions when the US machine is working, bedside IVC assessment. Can I use BP response as a proxy? Fattening of the IVC on passive leg raise?

 

[SurfingDoctor]

I realised the flaw in my initial hypothesis, which was not taking into account what happens when you get your fluid assessment wrong; much better to be running at a reduced rate when they start displaying signs of overload than just have dumped another litre.



Although the initial post was focussed more on "true" volume depletion, interesting discussion around fluids in sepsis. My question to the above is how do we incorporate the leg raise into assessment on the floor? Assessments of fluid responsiveness with PLR have all been studied in the ICU setting where they have the skills and technology to assess things such as cardiac output, PPV, carotid doppler etc. However the initial fluid resuscitation will be happening on the floor. All I have at my disposal is physical exam and, on the occasions when the US machine is working, bedside IVC assessment. Can I use BP response as a proxy? Fattening of the IVC on passive leg raise?

Leg goes up, heart rate goes down.

 

[littlejuan]

Ultrasound > IVC. If it is collapsing you have room to go, even with CHF