Fluids/Vasopressors (April 2018)
AIR CAPSULE: Vasopressors
Articles:
Self et al. “Balanced Crystalloids versus Saline in Noncritically Ill Adults.” NEJM 2018.
Selmer et al. “Balanced Crystalloids versus Saline in Critically Ill Adults.” NEJM 2018.
Questions:
- Why do you think the results are different for the two patient populations studied?
- How do these studies change or confirm your management in terms of fluid administration?
12 comments
Went through these articles on Easter Sunday A side shift, thanks for sending them along Alex!
First of all, bravo to Vanderbilt who did a big well designed study and got 2 NEJM articles out of it!!! Both leads got first author on one of the studies.
Looks like for ICU patients we should consider using LR instead of NS, doesn’t appear to make as much difference in our non-ICU admissions (may be better for the elderly, those with pre-existing renal failure and surgical patients).
I want to know, does LR cost more? Do we have it available? Should we consider stocking our trauma rooms with LR and encouraging people to go to that first for these highest risk ICU patients?
Check out the cost of LR and NS in the table here:
https://canadiem.org/normal-saline-the-coke-of-crystalloid-fluids/
$1.26/L for NS, 1.44/L for LR. Not a huge diff. Though in Canadian Dollars. Probably varies dep on institution.
What do you make of the fact that in non-ICU patients Balanced crystalloids resulted in a lower incidence of major adverse kidney events within 30 days than saline (4.7% vs. 5.6%; adjusted odds ratio, 0.82; 95% CI, 0.70 to 0.95; P = 0.01)? Despite no diff in hospital free days.
Eric and Sara (Sara – family medicine) discussed these with me, if you don’t have a lot of time, go to the results chart where they show the Odds Ratio bars (Forest plots). Eric has some interesting ideas based on critical care blogs he has been following.
Great article looking at a pertinent subject. There is a great review of the reasoning behind the use of certain crystalloid substances against NS (specifically in new acute renal failure) on pulmcrit that I highly recommend called fluid selection using pH-guided resuscitation.
I think the difference noted in the two different populations is based on adequate compensatory mechanisms vs those that are fully depleted (ICU group). Those that were not critically ill were able to handle the increased chloride load (along with acidosis) without significant affects. Those who already had their reserve depleted were noted only to get worse when given a crystalloid that worsened their ongoing acidosis.
What about the idea of using LR over NS overall? It doesn’t seem to show any harm in either of the studies compared to NS. Just food for thought.
.
Obvious some cases you don’t want to give it like in patients with hypercalcemia since LR contains Ca.
I would read the Editorial on this paper.
“Caution is required in interpreting these results. The patient populations in these trials were categorized by hospital location, which may not relate to the acuity of illness or to the increased risk of adverse outcomes. Surrogate outcomes, such as short-term improvements in biochemical concentrations or physiological variables, may not translate into beneficial, patient-centered outcomes. Although composite outcomes are used to mitigate competing risks in trials with low event rates (which characterized these two trials), such scoring systems are prone to confounding, even after statistical adjustment. The composite outcome, in which death, renal-replacement therapy, and a doubling of creatinine level (subject to imputation bias) are treated as equivalent components, is a metric that is not applicable as a patient-centered outcome.
What clinicians need to consider is whether the results of an open-label trial conducted in a single, major U.S. medical center can be generalized to the ways in which their own patients survive, feel, and function. None of the currently used resuscitation fluids are “physiological,” and questions regarding their safety and efficacy will remain, despite the results of these two trials and any randomized, controlled trials that are currently recruiting participants. Considerations remain regarding the effects of different types of resuscitation fluids and the ways they are used in specific, high-risk patient populations. Assessments of longer-term, patient-centered outcomes and health economics are fundamental to informing clinicians about their choice of resuscitation fluids in critically ill patients. The trials presented here inform that thinking but do not provide unequivocal clinical direction.”
Also, as has been pointed out, many single center + studies (think EGDT and sepsis, Rivers et al single center unblended RCT) ultimately prove to be invalid.
Among commentaries (here Holligsworth et al out critical care = Balanced crystalloids versus isotonic saline in critically ill and noncritically ill patients (March 2018)
The administration of balanced crystalloids (eg, Plasma-Lyte or Ringer’s lactate) versus isotonic saline for large volume resuscitation is controversial, hampered by conflicting data and the lack of an ideal crystalloid solution. One randomized trial in over 15,000 critically ill patients [13] and another in over 13,000 noncritically ill patients in the emergency department [14] found that, compared with isotonic saline, balanced crystalloids reduced the composite outcome of death from any cause at 30 days, new renal replacement therapy, or persistent renal dysfunction. Patients who received larger volumes (eg, >2 liters) benefited the most. Because the median volumes of fluid administered in these trials were low (1 liter) and the absolute improvement in the composite response was modest (1 percentage point difference), we suggest choosing the solution based on patient-specific factors such as chemistries, estimated volume of resuscitation, and potential adverse effect of the infused solution.
another critique =Single-center study limits generalizability
Non-blinded study
No separate analysis of LR vs. Plasma-Lyte A
Patients populations determined by hospital location, composite outcomes inherently not patient-centered and susceptible to confounding
4 However, the authors state that the NIDDK recommends the adverse renal event outcome used this trial as a patient-centered outcome.
When we discussed this at Crit Care Journal club, point was raised if cost not driver and no harm from LR, go with it.
The pendulum in medicine often swings – in prior times, EMS had to carry NS for “medical” and LR for trauma/hemorrhage resuscitation.
“unblended” = unblinded auto correct sp
The two studies both showed a lower rate of kidney injury and death (agreed – seems weird to pool those outcome measures together) in the balanced solution group, which was mostly LR. The difference between them was the magnitude of this effect. I’m not sure what the mechanism of this difference is, but my guess is that the two studies differed because ICU patients tended to get more IV fluid. As the non-critically ill group usually only got 1 liter of IVF, those patients likely didn’t see large physiologic changes as a result.
My take away from these studies is that balanced solutions are likely just as safe, if not safer, than 0.9% saline for volume resuscitation in the ED. Except for markedly hyperkalemic patients and those with ICP concerns/needs, I would be more inclined to start with LR or another balanced crystalloid, especially in patients who are academic already and/or where larger crystalloid resuscitation volumes are anticipated.
I tend to agree with Druga on this one. While the primary end-point in question did not show a difference between balanced IVF and NS, the results to indicate a trend towards benefit of using LR particularly in our patient’s who we assume will need larger volumes of IVD (i.e. ICU patients). Many patients that we see in the ED who are young with normal kidney function will likely have no adverse effects from receiving a couple of liters of NS for dehydration related to gastroenteritis for example. I worry more about the patient’s highlighted in the second article who receive larger volumes of crystalloid. They are likely older with some pre-existing renal function. While there may not have been a difference in 30 day mortality in this study, If you apply the trend for the wider population you begin to wonder. These studies should clearly be repeated in multiple centers to give us a clearer understanding of whether the secondary outcomes actually are causing any realy harm to patients. Anecdotally I know that in the ICU setting in Ireland they nearly always use LR (unless as Alex pointed out there was some clear reason not to). If patient’s were being admitted to the surgical service from the ED, they would also prefer LR. Price clearly comes into play here as well. Agree that a difference of 10-20 cents/liter is not big money but when you multiply that out a few thousand times those pennies add up fairly quickly.
Interesting articles. Although the overall results differed between the two studies, I was impressed that the trends tended to be similar in both. I don’t know if we can apply these results to all patients given that fluid selection should be driven by patient presentation- underlying kidney disease, presence of a head injury, anticipated amount of fluid that will be administered, ect. These studies included a wide range of patients- all with varying co-morbidities that likely cloud the picture. ICU patients vary widely. That being said, it seems generally safe to start with either LR or NS (guided by kidney function, K, acidosis) and circle back to this once the patient needs more than 1-2 L. I think this is more useful for management of inpatients and boarding patients in the ED,
As Eric emphasized, critically pts are likely to be pre-renal, hypoperfusing so when hit with a large chloride load causing metabolic acidosis, kidneys suffer worsening perfusion and renal failure. Not as profound of an effect in compensated patients with preserved renal perfusion. I definitely try to do LR in trauma pts with elevated lactated (LR does not significantly alter lactate measurement), DKA pts or any others with profound metabolic acidosis.