Podcast 86 – IVC Ultrasound for Fluid Tolerance in Spontaneously Breathing Patients – EAT IT STONE

Can the Inferior Vena Cava Ultrasound guide our fluid administration in the ED? Of course it can!

So I was getting on the plane to Las Vegas for Essentials 2012, on my iphone was the latest from Mike and Matt of the Ultrasound Podcast. Up pops Mike “the Rock” Stone interviewing my buddy, Haney Mallemat; these two ultrasound gurus discuss some ultrasound soundbites, but then… They both state that IVC ultrasound is useless for determining fluid responsiveness. It is worth taking a listen to that episode if you have a moment. So how can two brilliant guys get it so wrong? They just had their focus knob turned all the way to the right. Lets optimize their settings with an EMCrit Podcast.

Mechanically Vented Patients

Now, most of the podcast bashed IVC in spont breathing patients, but there was some overflow disparaging of IVC in mech vented patients, so let’s get that out of the way first. There is plenty of literature for these patients. Put them on a temporary, high tidal volume (10 ml/kg). Get an IVC shot and if it increases in size by 15-18% (depending on the study), the patient is fluid responsive.

  • Intensive Care Med. 2004 Sep;30(9):1740
  • Intensive Care Med. 2004 Sep;30(9):1834
  • Neurocrit Care. 2010;13:3
  • J Trauma. 2007;63:495
  • J Intensive Care Med. 2011 Mar-Apr;26(2):116

Spontaneously Breathing Patients

Now as the two ultrasound masters allude to, there have been a few studies showing IVC ultrasound assessment merely correlates with CVP  (it actually correlates with respirophasic CVP) and then use that fact to write off the IVC. Now we have maligned CVP as a marker of fluid responsiveness so IVC is crap as well, right?

There is evidence for the use of IVC as a marker of fluid status. In patients with ultrafiltration for congestive heart failure (Intensive Care Med. 2010 Apr;36(4):692-6) as well as fluid removal during hemodialysis (Clin J Am Soc Nephrol 2006;1:749 and Nephrol Dial Trans 1989;4:563). There was also a trauma study showing that fluid resuscitated patients with IVC collapse were more likely than those without to have recurrent hypotension (J Trauma. 2007 Dec;63(6):1245).

There was also a study just published in the Aussie EM Journal. This study was severely limited by the fact that none of these patients had any significant IVC collapse and the criterion standard is not a test any of us consider useful for cardiac index measurements; further, looking at the tables, some of the responder group did not seem to have any sig. increase in their CI in response to fluid. (Emerg Med Aust 2012;24:534).

And a meta-analysis study showing IVC’s relation to fluid status (AJEM 2012;30:1414).

Luckily, there was also a recently published study with the table below (Crit Care 2012;16:R188).

Here is figure 1 from the study

Here is the area under the cure (AUC); you notice there are points with much higher specificity.


Hot off the presses, this study is more reassuring: (Shock 2013;39(2):155)

It lends additional credence to the use of dynamic IVC for fluid responsiveness.

It’s not Fluid Responsiveness, It’s Fluid Tolerance!

This is the crux of the matter. In the ED, we want to give a bunch of fluid, but not if we are going to cause pulmonary edema. The term, fluid tolerance, is a perfect description of this idea. Responsiveness is great, but all we want to make sure of is that we are not going to do harm with additional fluid. This term was introduced to me by an amazing ED Intensivist named Chad Meyers. His lecture on this stuff from the ALLNYCEM Conference will go up very soon. IVC ultrasound is the perfect guide to fluid tolerance. CVP is fine as well, but why expose the patient to a central line.


Mike discussed a study he was part of that showed where you measure the IVC matters (Acad Emerg Med. 2010 Jan;17(1):96-9). Make sure you don’t measure at the diaphragm; measure distal to the hepatic veins or at least 3 cm past the diaphragm.

The Predicting Fluid Responsiveness Chapter at Crashingpatient.com has a bunch more info on all of this

Here is the video for how to perform the IVC ultrasound:

Now Mike is an amazing doc–he offhandedly mentions throughout the podcast, none of this crap matters until you have already given 6 liters of fluid or so. Amen brother. But a lot of folks are too scared to do that, hence the need for IVC ultrasound. Now there are unfortunately also references to the fluid depleted patient showing signs like cracked lips and a generally sere appearance; yeah, not so much. If we could assess volume status on vasodilated septic patients, that would be swell, but all of sepsis literature stands against our ability to do so.

Want to know why these studies may show results all over the map? It’s probably b/c the IVC is tough to measure unless you are good as demonstrated here.

Passive Leg Raise

Instead of IVC, Mike and Haney recommend using the passive leg raise with echo assessment of cardiac output before and after. You only need to do this once to realize its utility exists merely on paper. Much better is simply…

Giving a Fluid Challenge

Check cardiac index using echo, give a liter of fluid, and then recheck cardiac index.

Up until this week, I would have advocated considering using NICOM bioreactance, but a study just published makes me a bit leery until more data are in.

Did you like this episode? Then tweet the hell out of it:

Now, on to the podcast…


You finished the 'cast,
Now get CME credit

Already an EMCrit CME Subscriber?
Click Here to Get CME Credit for the Episode

Not a subcriber yet? Why the heck not?
By subscribing, you can...

  • Get CME hours
  • Support the show
  • Write it off on your taxes or get reimbursed by your department

Sign Up Today!


Subscribe Now

If you enjoyed this post, you will almost certainly enjoy our others. Subscribe to our email list to keep informed on all of the ED Critical Care goodness.

This Post was by , MD, published 3 years ago. We never spam; we hate spammers! Spammers probably work for the Joint Commission.


  1. haney mallemat says

    Great podcast Scott! You know how much I love this stuff.

    I got to read carefully read through some of the papers before I start accurately debating, but a couple of points for now:

    1) IVC distensibility (and pulse pressure variation) are great and well validated tests in the mechanically ventilated patient. I don’t recall our discussion overflowing to bashing this, but for the record I’m down with IVC Distensibiltiy in the mechanical vent patients and it should be encouraged more.

    2) The IVC would be a great tool at the bedside because it has a steep learning curve and is easy to visualize. However, there are a few examples where strict measurement of the IVC will fail you; a) cardiac tamponade (large IVC with little variation) b) acute RV failure c) Severe tricuspid regurgitation or tricuspid stenosis. All these problems occur on the right side of the heart which may be hard for novice to intermediate users to accurately image and assess (except for tamponade). IVC in these cases does not correlate with what we would necessarily do for the patient (i.e., the acute RV failure from MI is getting fluids despite a full IVC). So without a thorough and accurate cardiac assessment, I still don’t think you can look at just the IVC.

    3) Finally, I like the term of fluid tolerance,…I do. That is good when you successfully give volume and the patient is normotensive…all ends well. The problem is, however, that some patients will still be hypotensive when they appear fluid tolerant and then what? Give more fluids? ..and then what? When can you be sure that their tank is full and it’s time to switch to pressors? How long will you let them be hypotensive when you are unsure about their preload status? That is why I like the objective left sided measures that look at stroke volume such as looking at VTI with. It doesn’t have to be passive leg raise but give a bolus and then look at VTI to assess if there is increased output from the left side of the heart. That’s why I like the VTI measurements, the fluid coming out of the aorta is the actual stroke volume the body will see.

    I’m gonna look at these papers and maybe we can do a google hangout battle royale. I know the Matt and Mike wanted in on this.

    All the best Scott,
    Haney (@criticalcarenow)

    • says

      Great points brother, here and on the hangout. To respond to your last point: If the patient is still showing collapse, I have no problem giving them 6-8 liters. At that point, adding on a low dose of norepi, ~2 mcg, will cause preload constriction without afterload constriction. Nice thing about the IVC is it will not be affected by the vasopressors. So then recheck your IVC.

  2. says

    As always a great podcast, Scott. I look forward to getting to talk about this ad nauseum once I get a real internet connection (thanks for nothing, Amtrak…)

    Until then, my thoughts:

    1) Like Haney, I don’t remember going too far into ventilated patients and IVC measurements. It appears to work in experienced hands, but let’s face it, there’s a lot of ways to skin that cat with PPV, SVV, etc etc. Spontaneously breathing patients are the real issue.

    2) I’ll admit (as I did in the podcast) that a flat or collapsible IVC likely predicts the need for volume. In the initial assessment of an acute circulatory failure patient in the ED, we should assume they need fluid, though, right? Maybe a quick echo to make sure it’s not cardiogenic shock if there’s a question clinically, but we know most undifferentiated hypotensive patients need volume (whether crystalloid, blood or both). My point being, I wouldn’t bother looking until after I’d given a healthy dose of volume (typically 4-6L but sometimes even more if suspicious of DKA w sepsis or other severe volume depletion). I hear you have data that most ED docs under-resuscitate wrt volume, but if that’s the case, shouldn’t we have sepsis algorithms that push fluids, rather than pushing IVC ultrasound to push fluids? Do we need permission from the IVC to give that initial fluid bolus?

    3) If I give 4-6L of fluids to my septic patient and they’re still hypotensive and in acute circulatory failure, I might make the case that they need to be intubated for their sepsis alone, in order to maximize tissue oxygenation and free up all that ATP their respiratory muscles are burning. Then I’ve got a ventilated patient and IVC, SVV, PPV, etc etc are all good options.

    4) There are way too many inter-rater reliability issues with IVC measurement, ranging from location to M/B Mode to simple interoperator variation and oblique imaging planes. Even with experienced sonographers (fellowship trained), their agreement for visually estimated IVC collapse was 0.60, with 95% CI=0.36 to 0.76 (http://www.ncbi.nlm.nih.gov/pubmed/21414063). All these issues make me more likely to protocolize my initial fluid challenges without regard for IVC respirophasic findings.

    Looking forward to working through this evidence as a group and thanks as always for sparking the conversation…

    • says

      Love debating you, you are at least as aggressive a doc as I am which is a rarity.

      So most of the assessments even in intubated patients suck. PPV/SVV all require circumstances that rarely exist.

      There is little wrong with giving 5-6 liters empirically, but some of your patients will be in vasodilatory shock and actually would have benefited from just getting put on some norepi at the 2 liter mark. In our collaborative, if folks can’t/won’t monitor we ask for 4 liters empiric fluid bolus.

      All patients with something like sepsis should get a 2 liter bolus. At that point, just as you mention you probably should do some undiff hypotension ultrasound assessment that invovles the heart, the IVC and the rest and go from there. The IVC at that moment is going to give me a bunch of info. Big collapse, I can give them another 2-3 liters before I come back and check again. Sometimes we’ll see minimal collapse after those 2 liters.

      But the real need for the IVC is in the patients who are not hypotensive after 2 liters, but have an elevated lactate. These patients really benefit from IVC, b/c most folks will not give enough fluids to these patients without it.

      I love to intubate, but unless the patient is showing signs of inadequate tissue oxygenation I will try to avoid it in young septic patients. In old ones, I have a hair-trigger for plastic.

      Definitely a bunch of issues both inter-rater, intra-rater, and technique; but a lot less than a moderate to expert level exam such as VTI for CI.

  3. says

    Hi Scott
    Sorry I missed the hangout. Haney and I had a quick chat and I completely agree with you. I work as one of those smaller ED docs and we are often flying blind in the early resuscitation of sepsis. IVC is a quick and easy test that works well over on the left side of the Frank-Starling curve.

    There is a quantum leap between IVC assessment and doing an accurate LVOT doppler (if you can even get the Apical 4 on the screen #%!$) – so this is a huge technological block to what they are suggesting.
    One can argue about evidence all day – but none of the studies can be extrapolated to an “occasional ECHO” practitioner. We need something that is more robust – I think the eyeball / M-mode IVC assessment is achievable – I know, I do it with pretty crappy US raining and skills!

    I think the problem is that Haney and Mike S are arguing way over on the right of the Frank-Starling. Mike says he just gives 5 – 6 L, however that is not a practice most small town docs are ‘comfy’ with. I am constantly having to prod my trainees to go past 2 L (of saline -ehhhhghk). I know this cos I work there!

    I am less interested in what happens after the IVC becomes plethoric / non-collapsing – I know I am probably getting close. Then I can look at what has happened to the lactate (did it drop by 20%, or plateau?), if I have a CVC in then do an SvcO2, is there any urine in the bag, if the patient mentating better? None of these are “gold medals” but they make a pretty handy relay team to extend your metaphor.

    I am keen to hear more about the NICOM – it was looking like a great option for our unit – no lines, easier to use. Fingers crossed it comes good!


  4. says

    G’day lads

    My comment takes a more philosophical perspective.

    Just because giving fluid increases cardiac index – just because a patient is fluid responsive – doesn’t mean fluid should be given. In many critical illness situations we don’t know what the appropriate cardiac index should be.

    The critical care pendulum – particularly from the ICU perspective – seems to be swinging back towards avoiding the administration of excessive fluid. A wet patient causes problems that become apparent beyound the resus room. I think we do need ways of assessing fluid status rather than empiric 4-6L fluid boluses – exactly how to do this remains an evolving puzzle. I like the concept of fluid tolerance/ safety.

    Great work and discussion by all!


  5. Ben Ho says

    Wow Scott, it brings a tear to my eye every time I think of all the good stuff you’ve been giving us with your podcasts. I really like this concept of fluid tolerance – is there any literature supporting the safety of giving fluid until the ivc stops collapsing (i.e. not causing pulmonary edema or any long term problems with our ICU collegues downstream)? I wasn’t aware of this concept previously and have been using ivc up to a point where there was still some respiratory collapse by eyeballing it then switching to percent change in VTI in the LVOT with rapid 500 cc boluses.

    • says

      Here’s what we know. IVC correlates with CVP (we at least all agree on that). Elevated CVP occurs before the onset of clinical pulmonary edema; you would have to go back a few decades to find that literature.

  6. says

    Love the post – and love the “fluid tolerance” wording – semantics, semantics… I bet the IVC is stoked for being such a topic of debate as it had been ignored for far too long while the aorta was being loved.
    Ill just say this, and, yes, I may sound like a radiologist here, but we need to take our ultrasound images as one data point when assessing the patient and “correlate clinically”…. while knowing the difference in IVC physiology with intubated versus non-intubated patients and various cardiac diagonoses – – -but that comes with clinical correlation. Im going to read through those studies you mention and Im familiar with a couple of them…. To me, the IVC is a volume assessment ‘usually’ correlating with CVP (particularly in the extremes of measurements), some saying better to evaluate in transverse view than longitudinal view, which ‘generally’ provides me with knowing whether the patient can “tolerate” fluids as a resuscitative tool in shock. I rarely do an IVC study on its own though, particularly if the patient is elderly – Ill add Echo – which can answer quite a few questions and clinical correlates discussed above.
    Does the IVC being low mean the patient is fluid responsive? No, it just means that I can use fluids depending on my other assessments of the patient – and re-evaluate ..

    My 2 cents….

    • says

      Stone is so mean to this poor tube, I thought it deserved some love : )
      Absolutely agree about looking at the heart along with the ivc, it is only a few cm away!

      We still need to set up our FAST Cast; can’t wait.

  7. Pelle Staffan says

    Hi Scott,
    Thank you for your fantastic podcasts, like having a bolus of good ideas. I followed the debate on IVC and was a bit confused. Excellent concept “fluid tolerance”. I doesn’t trust any single ultrasound measurement for fluid responsiveness or tolerance, what I do is a combination off just eye ball EF and myocardial contractility in standard projections, IVC and lung ultrasound. Then I might add VTI and LVOT before and after bolus. I think this is a useful approach particularly to CHF patients with sepsis or pneumonia. Not uncommonly these patients have an MI on top of everything else. For the relatively healthy septic patient bronze medal is probable good enough, but I´ll do ultrasound just for the sake speed my skills up if I have the time. Maby it will land in some protocol or algorithm.

    Looking forward to more discussion!
    Thank you all,

    • says

      Yep your practice sounds like mine and most of the ultrasound folks out there. Get a gestalt and go. We had to protocolize for the sepsis collab.


  8. says

    Hi Scott!

    I have to completely agree with you. That being said, there are a number of factors that explain why the “traditional” ultrasound assessment of the IVC may be less-than-ideal.

    The first and I think the most important is that the AP diameter in long-axis, 3cm below was (arbitrarily?) chosen as THE MEASURE. For anyone who has been doing bedside ultrasound for a while, you’ve certainly noticed that not all IVC’s are shaped the same. Some are really round, some quite elliptoid – and they can be so in either axis or diagonally. Furthermore, their collapse may be also quite asymetrical. This makes a two-point analysis, well, suboptimal in my opinion. Since we are looking in fact for a volumetric variation (eg how much blood is being sucked in and out for the chest for a given change in pleural pressure) it makes a lot more sense to look at a short axis surface area of the IVC… Where should we measure this? Who knows. That’s why in my clinical practice, I look at a fair segment of the IVC in both long and short axes to get a sense of the total volumetric variation. There is absolutely no data to support this but it makes a lot more physiological sense to me.

    The second is that, short of having an esophageal balloon, it is difficult to measure the effect of the ventilation, spontaneous or mechanical. If I breathe really slow and smooth, there is minimal variation in my IVC – and I’m pretty sure I’m mostly fluid-responsive and -tolerant. The sniff test of a young athlete and an octagenarian are unlikely to produce the same change in pleural pressure and hence, variation in IVC…

    Having said that, I think IVC variation is an extremely useful tool, you just have to integrate it along with your other observations, both echographic and visual. That’s why there’s still some art involved in medicine…

    Haney, no offense! Hope to see you soon buddy!


  9. Justin Bowra says

    Nice work all round on ultrasound of IVC.

    However, Scott’s original post approvingly quotes that 2010 study by Wallace that came up with the gem ‘Clinicians should avoid measuring IVC collapsibility index at the junction of the right atrium and the IVC’, because IVC measurements here didn’t correlate with measurements further caudally.

    This was an amazing & gutsy conclusion to draw when you look at the gold standard Wallace used: there wasn’t one.

    So in fact for all that anyone knows [including Wallace], perhaps the IVC/RA junction was the correct site and all the others are crap.

    Yanagawa found a correlation between RA pressure & IVC diameter measured just below diaphragm. (Yanagawa Y, Nishi K, Sakamoto T, Okada Y. Early diagnosis of hypovolemic shock by sonographic measurement of inferior vena cava in trauma patients. J Trauma 2005; 58(4): 825-829.)
    AND Yanagawa Y, Sakamoto T, Okada Y. Hypovolemic shock evaluated by sonographic measurement of the inferior vena cava during resuscitation in trauma patients. J Trauma 2007; 63(6): 1245-1248.

    Charron found a correlation between RA pressure & IVC diameter measured just below diaphragm <2cm from RA. (Charron C, Caille V, Jardin F, Viellard-Baron A. Echocardiographic measurement of fluid responsiveness. Curr Op Crit Care 2006; 12(3): 249-54.)

    I’ll admit it, I’m selectively quoting studies to prove a point. There are other studies that have found that the more caudal sites seem to correlate with RA pressures as well.

    But I think the real point here is that we should be very, very careful about drawing conclusions that aren’t necessarily justified by the evidence. And much of the ‘IVC ultrasound for fluid status’ evidence out there is conflicting and based on small studies.

    We also need to be a little bit careful about extrapolating results of sonographer studies when we scan… which is analogous to the point already made that the casual IVC scanner may come up with different results to the ultrasound nut who scans 24/7 in his/her ED.

    Randazzo et al found that IVC ‘central venous pressure measurements’ resulted in only 70.2% overall raw agreement between EP (trained for 3h in focused cardiac US) and formal echocardiograms performed within 4h.
    (Randazzo MR, Snoey ER, Levitt MA, et al. Accuracy of emergency physician assessment of left ventricular ejection fraction and central venous pressure using echocardiography. Acad Emerg Med 2003;10:973–7.)

    I'm deeply suspicious about absolute IVC measurements because [a] the IVC narrows caudally and changes dramatically with body position and patient size, so absolute measurements are not as exact a science as we think; and [b] the ASA guidelines that many people rely on are taken from patients in the left decubitus position and measured within 2cm of the RA [precisely where Wallace et al don’t want us to measure] … and [c] as Randazzo noted, it's possible there isn;t much correlation between clinician and sonographer IVC measurements anyway.
    (Lang RM, Bierig M, Devereux F et al Recommendations for chamber quantification: a report from the American Society of Echocardiography’s guidelines and standards committee and the chamber quantification writing group, developed in conjunction with the European Association of Echocardiography, ad branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005; 18: 1440-63.)

    How does the following sound?

    An eyeball assessment of the IVC probably does help at extremes (fat & full versus flat & collapsing)

    Don’t sweat the absolute measurements.

    Avoid M-mode [it’s too open to error].

    And remember the lungs: if they are still dry after a big fluid challenge, the shocked patient can handle more fluid resuscitation. The lungs are much easier to find than the IVC as well.

    Cheers from Down Under, Justin Bowra

    • Pelle Staffan says


      Agree on lungs. So, just fill until you start to se some B-lines? Sounds simple enough. Easy to learn an fast to perform. Maby a “harder” endpoint to stop filling at. Most of us do IVC and echo as well, the real question is probably what does IVC really add to lung ultrasound and viseverse. Did you come across any papers?

      • says


        What you describe there is essentially Daniel Lichtenstein’s FALLS protocol published in Chest in 2009. Certainly whe diffuse B lines start to appear its time to stop, but the real question is, especially when using crystalloids, should we not stop earlier? Fluid balance and morbidity studies certainly suggest it. I don’t believe an evidence-based answer to that exists yet!



    • says

      So many good points. I think we are in agreement on the following:
      -Absolute measurements send us astray
      -M-Mode is prob more likely to send us astray as the IVC moves with ventilation and your measurements may be very inaccurate as a result (see article just pub in Crit Ultrasound Journal)
      -Eyeball of collapse is the way to go
      -If the lung is dry on ultrasound, one more reason to push forward. If the lung shows signs of interstitial edema, this could be capillary leak syndrome and should not be used to withhold fluid.

  10. Longboard says

    Can I make a comment about the last article you mention in the show notes- “Cardiac output measurements using the bioreactance technique in critically ill patients”.
    The study tries to correlate absolute values for CO between a bioreactance-based technique and thermodilution by PA catheter. The correlation was found to be poor (in 11 patients).
    The real utility of continuous non-invasive CO monitoring is that it can be used to measure the change in CO over time, and therefore the patients response to therapy. The absolute value of cardiac output is less important; will will never know if a certain CO is adequate for a given patient as we have no way of measuring their oxygen demand. We can only use surrogates like lactate, ScvO2, and perhaps tissue oxygenation as markers of inadequate perfusion.
    So I don’t know if this new study is such a big blow to bioractance based continuous CO measurement. The validation study by Squara (ICM 2007) seems a bit more robust, and demonstrated that bioreactance had good test characteristics for detecting changes in CO.

  11. Minh Le Cong says

    thanks Scott et al for the debate!
    made me reflect on what I do in a prehospital setting as well as in ED.
    To be honest I cant say one approach is better than another and what you suggest sounds reasonable but to equate IVC USS directed sepsis fluid resus to what Rivers did etc, is a bit of a stretch in my view. logically it makes sense but like DSI I think you have to collect enough cases to prove the theory at least does not harm vs benefit.

    I am sure we all have seen overzealous fluid admin lead to pulmonary edema and until there is more experience and evidence, a cautious approach to IVC USS directed resus is reasonable.

    • says

      One thing all three of us agree on is the IVC equates to CVP, so you absolutely can make the jump to replacing the CVP element of EGDT with the existing evidence. We also are presenting our first 6000 patients from the NYC sepsis collaborative at SCCM in Jan. 2000 had IVC ultrasound as their fluid assessment.

  12. Seth Trueger says

    Great podcast & discussion!

    I won’t wade into whether or not passive leg raise actually gives you useful information, but you can do it without awkwardly lifting the legs.

    start with:
    -bed all the way up
    -HOB 45 degrees

    -put the bed into extreme Trendelenberg

    Now the torso should be flat with the legs up.

    Now whether or not it gives you useful information is another question…

  13. Dave says


    Not so much about the USS technicalities but with regards to significant fluid resuscitation, if you look at the FEAST study (http://www.nejm.org/doi/full/10.1056/NEJMoa1101549) you have to wonder about the down side of high volume fluid resuscitation.

    The study has issues in terms of generalisability and is in a paeds largely sick malaria population BUT was initially planned to show a benefit to high volume fluid resus and instead showed higher mortality with this. It at least raises questions that we haven’t yet answered properly

    What this means IMHO is that we don’t necessarily understand the pathophys of severe sepsis as much as we think and that maybe lots of wet stuff has harm associated with it that isn’t initially apparent.


  14. Minh Le Cong says

    thanks Dave. Glad you cited FEAST as I was not sure if bringing up a paediatric paper here in an essentially adult critical care forum was appropriate or not. There are a few issues with FEAST but yes it does raise the question of harm with aggressive fluid resus in septic kids in Africa.

    Boyd et al did this retrospective review in the VASST trial

    It calls into question the CVP targets for sepsis resus, ergo even if IVC and CVP correlate, are we promoting the right thing here?

    NYC Sepsis collaborative data will be useful to review as Scott has stated.

    • says

      ummmm not so much. The review of the VASST showed us that pt’s with high cvps (like the sepsis-induced cardiomyopathy and the sepsis-induced renal failure patients) had worse outcomes, not patients who were fluid resuscitated to a CVP goal did worse. As to what FEAST can tell us about the resuscitation of adult septic patients in the 1st world, I have absolutely no idea.

  15. says

    Sorry for my non perfect english (I work in a medium hospital in Italy). I read with great interest your post.
    I would like to give my contribution:
    First problem: CVP must not be more than our benchmark!
    Second: the single measurement of the IVC is not a viable solution but only monitoring the collapse index gives us important information.
    Third: only ecographic study of heart, lung and IVC (together) provide the information we need.
    Fourth: the study of the lung must be our point of reference. The appearance of B lines is the benchmark: B lines = stop fluid and begin vasopressor. (the problem is when the patient presents at the outset with the B lines: in this case only the study of the monitoring of the collapse index can guide us). B lines appear before a possible worsening of breathing.
    Fluid tollerance: great ideas (I’ll use it in my courses)

    • says

      thanks for commenting! We are in agreement, no point in measuring IVC in spont. breathing pt. Agree always echo. Sorry, can’t agree about the lungs. If they you have an A-line profile, then all is well; but interstitial syndrome (mult. b-lines) can be from many reasons in a severe septic pt like alveolar capillary leak.

      • says

        I agree with you that the only problem is the patient with a B or B’ profile at the outset. In this case the lung is not useful and the continuous control of the IVC is our only weapon.

  16. James French says

    Loved the podcast and I think the concept of fluid tolerance is a great one….

    Im not sure how I could practice EM without echo now and realise that I was totally crap at assessing volume status before hand….!
    I think IVC assessment combined with functional echo assessment is a great way to characterise the nature of the hypoperfusion state, including sepsis, cardiac failure and the various cocktails of both that seems to present. Most of the patients I see are not in what you would call septic shock (obviously some are), but many (given the increasing proportion of failing ventricles walking around out there) need some careful tweaking….some need fluid resuscitation, some with diastolic failure may benefit from negative chronotropy, some may even need preload reduction….I have realised that the fixed treatment paradigms that I used to practice with were probably rather simplistic…..I think being a resuscitationist is partly about nailing the components of the blood pressure equation to restore normal perfusion. What needs increasing or decreasing? Rate? End diastolic volume? SVR? Inotropy? Personally I cannot do that without echo added to the rest of my assessment….

    In my limited experience though I have found just assessing the apical PW stroke volume at the aortic root to be technically challenging in some patients….sometimes you just cant get a view! I fully admit this may well be because Im crap with an echo probe! So I also use the follow views:
    1. Apical Stroke volume at the aortic root. I think this is what we normally do…I just tend to calculate the velocity time integral and see if it increases with preload/afterload optimisation…. Im not sure you have to use the root diameter, as if the SV is going to increase by a proportion of 10% or greater then, it will regardless of the multiplication of the Aortic root diameter…..
    2. Suprasternal notch aorta view of the descending aorta…You can use the above PW method to calculate STI or SV of blood flow through the descending aorta. This is rather similar to oesophageal doppler..? The only problem with this method is that you can easily take measurements at different points in the aorta, so I guess that could effect the results …..and sometimes, like the rest of echo, you cant get a good view.
    2. M Mode fractional shortening…This estimates ejection fraction and stroke volume with one M mode section taken across the left ventricle. The good thing about this method is that usually I can get a left parasternal long axis view, however it is a bit time consuming and you need to have the correct functions set up on the machine. The other problem is that the method uses loads of assumptions. Also if there is basal hypokinesia you might not see an appropriate increase with fluid loading. Essentially it measures the change in the internal diameter of the LV using M mode in the left parasternal long axis view. The internal diameter is then cubed to give a volume. The change in the volume can give you an estimate of ejection fraction, the volume being ejected (the difference) being the stroke volume….well an estimate at one point of a change in volume……

    Not sure there is any evidence for any of the above….! Thanks! JF

    • says


      Amazing comments! Now that is truly interesting about aortic root diameter cancelling out in the equation. Is that definitely true? I must admit I’ve never done the actual calcs, just let the machine do the math. It seems intuitive that if it is a multiplicative value, then you are right, don’t need to calculate the aboslute just the relative change. v. cool.

  17. tony anagnostou says

    re: ‘is that definitely true’..

    mathematically, you can remove root diameter (and root area, the end destination for that variable) and still get a percentage change from before/after VTIs. ie, if you correctly assume that root diameter is not going to change, then SV and LVOT VTI will change proportionately.

    if the HR stays the same (probably only in paced patients, denervated/neuropathic pts, and fluid unresponsive patients, i would guess) then the CO would also change proportionately to VTI. you can calculate change in CO without root diameter, just like you can calculate change in SV without root diameter.

    (different studies on PLRs use CO or SV as their dependent variables, and those two do NOT necessarily change in proportion to each other)

  18. Sue says

    Lets simplify it for the average ED doc in a non academic institution.On Initial Ed Eval. Pt septic. No cvp line. How much and how fast to give fluids in a healthy person.?. What about the chf or esrd patient?

  19. DocXology says

    After SMACC and particularly the enlightening presentations by Bowra and Nalos on ultrasound, and Myburgh and Harris on shock and cardiac output measurement provided a useful overview of the literature in regard to both IVC U/S, fluid management and circulatory support in shock.

    It certainly reinforced the prevailing ICU view that excessive fluid administration is deleterious (particularly in sepsis) resulting in tissue oedema which is very difficult to mobilise later. Bowra’s literature search raised sufficient doubt over the precision of IVC ultrasound to accurately guide fluid replacement and certainly anything greater than a 0.9 cm IVC or less than 75% collapse in a spontaneously breathing patient is not adequate evidence in itself to continue to push fluids

    Assuming no access to ICU or sophisticated monitoring like SVO2, PICCO or PAC, my approach to sepsis would be:

    1) Take in regard the patient’s baseline MAP if that information is available
    2) Empirically bolus 1-2 litres of fluid to assess MAP response (consider more only if IVC 75% and lung ultrasound shows no evidence of lung fluid)
    3) Commence noradrenaline early following first fluid push (or adrenaline if you are concerned about giving Norad peripherally) to obtain MAP targets (and achieve other neurohumoral benefits of catecholamines).
    4) Insert urinary catheter +/- CVC (knowing that interpretation will be tricky in pre-existing chronic renal failure or heart failure) to provide further data of the perfusion status and filling of the patient
    5) Send off lactate on an hourly basis looking for clearance and response to gradual adjustments in fluid or pressor therapy (in combination with lung ultrasound looking for fluid overload)

    • says

      Not sure these are really the take home messages I would take from those two lectures. That approach is going to put every septic pt in an ICU or HDU and the evidence presented would lead to v. different conclusions except for #1.

      • DocXology says

        A patient in septicaemic shock is likely going to need ICU anyway (or at pressors and more invasive monitoring) if they don’t respond to a couple litres of fluid. Pushing 6 litres into a septic patient purely based on IVC ultrasound be a way of keeping the patient in the ICU longer than you need. It’s not a powerful enough tool alone for monitoring haemodynamics..


  1. […] IVC Ultrasound for Fluid Tolerance in Spontaneously Breathing Patients – EAT IT STONE – Scott serves it up to Stone and Mallemat, and advocates that IVC ultrasound does have a role in assessing fluid status! I’m sure this is only the start of the debate on this topic! […]

Speak Your Mind (Along with your name, job, and affiliation)