SCCM Resource Library-Pro/Con: Water, Water Everywhere: IV Fluids: Where's the Pendulum Now? (includes advanced cases)

Video Transcription

Hello, and welcome to the SCCM Critical Care Congress 50th Anniversary. My name is Paul Reynolds, and I'm here to introduce the session Pro vs. Con Debate.  Water, water everywhere, IV fluids, where's the pendulum now? I am a Medical Intensive Care Unit Pharmacist at the University of Colorado, and I don't  have any conflict of interest related to this talk. For other speakers giving this debate, please see the individual speaker slides for further disclosures.  The following are our talk's learning objectives. Explain the risks and consequences of aggressive IV use, evaluate the objective of protocol  suggestions and how application of these protocols may vary depending on clinical scenario, and describe IV fluid strategies that can be clinically implemented at the bedside.  The whole reason why this debate matters is there are over 200 million liters of normal saline sold in a mystery on an annual basis.  Fluids meet every aspect of the FDA definition of a drug in terms of diagnosis, cure, mitigation, treatment, and prevention of disease.  Like any other drug, and I'm spinning this from a pharmacist's perspective, obviously, every drug exhibits a therapeutic index in which underdosing, especially in the critically  ill septic patient, can lead to malperfusion, impaired cardiac output, organ dysfunction, and ischemic digits, even mortality.  Similarly, overdosing can lead to edema, specifically in the lungs, that can lead to prolonged mechanical  ventilation in patients with acute respiratory distress syndrome, acute kidney injury, and acid-base aberrancies, depending on the fluid selected.  The purpose of this debate is to explore the literature surrounding the optimal dose of  fluid therapy in the critically ill patient, and to navigate some of the therapeutic controversies and advancements in the literature of recent.  Our speakers today include Dr. Shawn Townsend, who's the Vice President of Quality and Safety at Sutter Health at the California Pacific Medical Center in San Francisco.  He has served on the Surviving Sepsis Campaign Steering Committee, and he is a measure steward for the CMS's National Sepsis Quality Measures.  He will be arguing towards the SECM guidelines saying 30 mLs per kilogram for the septic patient. This is followed by Nenu Malbryan.  Nenu received his MD from Leveland, Belgium in 1991, and from 2017 to 2020 has been the Director and Crisis Manager of the Intensive Care Department at the University Hospital  in Brussels. He is actively engaged in the European Society of Intensive Care Medicine, in which he has  chaired the Working Group on Abdominal Problems and Raised Intraabdominal Pressure. He is also co-founder of the International Fluid Academy and has chaired several of their  annual meetings. He will be arguing to the point that patients are drowning from fluid over-resuscitation.  Roderick Fontenette is an Assistant Professor of Military Emergency Medicine in Bethsaida, Merida.  Dr. Fontenette's research interests include TBI management, resuscitation, and airway management. He'll be overviewing a variety of case vignettes before and after each debate.  So I hope that you enjoy this session. This is a picture of Skógafoss, which is Icelandic for forest waterfall. It's one of the most beautiful waterfalls in Iceland.  If you ever get an opportunity, I would invite you to travel there. If you have any questions during the session or the debate, specifically for any of the  speakers, don't hesitate to submit them to the debate chat Q&A, and I will be fielding them to their respective speakers verbally at the end of each session. Thank you and enjoy.  Good afternoon. I'm Sean Townsend, and I'll be speaking on the pro side today of the resuscitation approach  for septic shock patients, suggesting that 30 ml per kilogram of crystalloid is an appropriate resuscitation strategy.  To get started, I'll briefly speak to my qualifications to speak to this topic. I am a member of the Surviving Sepsis Campaign and have been a member since 2003.  I'm an author on the 2008, 12, and 16 guidelines, and I'm presently SEP1 measure steward.  And as you may be aware, SEP1 is the measure for sepsis quality of care in the United States by Medicare, and I serve as measure steward along with Dr. Manny Rivers from Henry Ford  Hospital. In terms of disclosures, I have no financial conflicts to disclose, and I'd like to quickly  speak to a common sense idea about this topic and, in general, about quality improvement in health care use of guidelines, bundles, and measures.  The editors of Intensive Care Medicine as well as Critical Care Medicine saw fit in 2016 after the 2016 Surviving Sepsis Campaign guidelines to publish the following statement.  They said, as clinicians, we're bound to deviate from guidelines when such deviation is reasonably expected to improve an individual patient outcome.  As clinical scientists, we are bound to evaluate the prevailing standard against emerging alternatives. These three imperatives are inseparable.  We therefore caution against any quality metric or reimbursement policy that mandates slavish adherence to any particular recommendation.  And I'd just like to say I agree with that statement completely. This is absolutely the case that it's essential that each patient receives individualized  care when they need to receive it. And we've always said that in the Surviving Sepsis Campaign as well as through Medicare  and the promulgation of SEP1, that we expect clinicians to treat their patients. Quality measures exist to raise the quality of care and raise all boats across the population.  And I'll explain a little bit more about that, but that doesn't mean that we don't pay attention to individual patients and deviate when necessary.  In terms of SEP1, it's been incorrectly described as a mandate by many people. As it turns out, SEP1 is part of a voluntary reporting program for hospitals.  Hospitals choose to participate. They do not have to. It's called the Inpatient Quality Reporting Program.  This houses all of the core measures for the Centers for Medicare and Medicaid Services. So for example, the all-cause readmission measure is part of the IQR.  And in the IQR, a measure could be tied to reimbursement, but SEP1 is not, and there's no plan to do so.  In fact, Medicare has stated that they will not do so for SEP1. So there's no penalty. However well you perform in SEP1, you don't get a bonus, and however poorly you do, you  don't get a bonus. All you need to do is report SEP1 if you participate by choice in the Inpatient Quality Reporting Program.  If there's any doubt that clinicians understand this and don't take this as a mandate from on high, take a look at compliance from the program going back to 2015.  As you can see here, the initial SEP1 compliance in the third quarter of 2015 was just 40% across the country, meaning 60% of the time, hospitals were non-compliant.  Five years later, in 2020, that compliance rate has risen to 60%, and very proud that hospitals have improved that substantially over that period of time.  And we know now that the best compliance that hospitals can do, that is the top decile performance,  has hospitals achieving 80% or better compliance with the SEP1 measure in a given quarter. But these still all demonstrate that there's room for deviation, and people take that room  and deviate as necessary. So then what's the point of having a quality measure if it's a standard that you can deviate from when it's necessary? It's as I mentioned already.  It's the idea that we want to raise all boats and try to eliminate gross harms by making sure people are reminded of the best therapies.  It's a balance between what's utilitarian, the greatest number for the greatest good, and then what we do for individuals to make sure we don't cause harm.  And this is always the challenge in quality improvement in health care, but it's the mechanism we have.  We set standards, we attempt to follow them when possible, but we don't harm our patients if we can't follow those standards.  So you might say, how did we get to a standard, a sepsis quality measure that is, at least  in the United States, that asks for resuscitation with 30 mLs per kilogram of crystalloid fluid?  Well, the answers come from looking at mortality in sepsis care prior to 2001. And so what we see is mortality rates in the published trials to the right, anywhere between  30% and 80% for patients who had septic shock. And so this is actually a very high statistic.  We can compare this to modern day early sepsis, septic shock trials, in particular if you think of PROMIS, PROCESS, and ARISE, the multi-institutional trials that were done  in 2014 and published around 2015 or so, we know that mortality was much lower. In PROCESS, for example, it was 18%.  So something has changed between the year 2001 and the year 2015. And I would suggest to you that it's that we set standards for how we are going to take  care of sepsis patients, and that we actually began to adhere to those and decrease mortality over time.  One of the things that we used to see quite commonly when under resuscitation was part of the problem when we managed patients was limb ischemia and digital ischemia in particular.  I do occasionally see this today, but certainly much less than I used to in the past when managing patients.  And part of this is to make sure that we have adequate volume on depressed patients so that we don't end up causing tissue ischemia and injury like this, which of course is horrible  for the patient, difficult to recover from if it doesn't result in amputation for certain. Certainly a very bad outcome.  I think it's important, as we begin to set those standards, that people get a sense that these numbers for resuscitation guidelines that were chosen didn't come from no place  in the early 2000s. In fact, there were guidelines, practice parameters, that had been published for hemodynamic support for adults with sepsis dating back into the 90s.  So this slide just happens to show you the 2004 update of those practice parameters. But I want you to pay attention to the right-hand panel here on the slide where the goals for  monitoring and fluid resuscitation are stated. And it states that up to 6 to 10 liters of crystalloid solutions may be required for  initial resuscitation in the first 24 hours of sepsis care. And in approximately 50% of sepsis patients who present with hypotension, fluids alone will reverse that hypotension.  And so these guidelines and numbers and studies had been done previously to suggest that there was a fluid deficit when it came to managing patients with sepsis.  And so it was not a surprise that as protocolized care came into vogue, that there was a reliance  upon these previous studies to suggest that we ought to resuscitate patients to a certain standard.  One of those studies that created a protocol for managing patients with sepsis was the Early Goal-Directed Therapy trial, which you are all likely familiar with, published  by my colleague, Dr. Rivers. And in the Early Goal-Directed Therapy trial, Dr. Rivers set out a standard of 20 to 30 cc per kilogram resuscitation for patients with septic shock.  And it turns out that the results of the EGDT trial in this single institution at Henry  Ford Hospital were remarkable, with a decrease in mortality from 47% down to 31% with compliance  with this particular protocol. Because of those results, there were multiple studies that attempted to replicate those results.  And so Dr. Malbrain here has published a statement that indicates there were over 70 observational  and RCTs done after the Early Goal-Directed Therapy trial, comprising greater than 70,000 patients that found similar outcome benefits.  Now, we know, however, that three large studies of early septic shock, PROCESS, ARISE, and PROMISE, which were finally published in 2014 and 2015, actually found, however, that in  the usual care arm, care was non-inferior to Early Goal-Directed Therapy, meaning that the mortality outcomes in usual care versus EGDT were essentially the same.  What's interesting is, looking at those multi-institution randomized control trials that were designed  and harmonized so that the results could be brought together in a meta-analysis, ultimately, if we look at the individual trials, we see that the fluids prior to randomization were  similar. 2.1 liters, 2.6 liters, 2.0 liters. If we translate that to a 70-kilogram person, it's about 30 cc's per kilogram.  When you combine the trials in the end of the patients that were involved, it's about, I think, 27, to be precise, mLs per kilogram that these patients received on average.  Speaking of the PRISM meta-analysis, here are some of the results. In the red box, I've separated out the intervention from the usual care arms, just to demonstrate  to you that the average number of mLs of fluids that patients received in the Early Goal-Directed Therapy arm was 1,950 mLs, versus usual care, it was 2,000.  The point I'm trying to score here is that the number that we've chosen to resuscitate patients with 30 mLs per kilogram is what clinicians were likely to do if they're left  to their own devices, that is usual care. You see that usual care is about 2,000 cc's, and for your average person, that's going to come out to be about 30 mLs per kilogram.  That was chosen by the people who participated in usual care. They were not instructed to resuscitate patients to that volume prior to randomization.  The other thing to point out here is that what became usual care in 2014 and 2015 was not usual care back in 2001.  And so the standard that was EGDT, which here was 9,950 cc's, became the standard of practice for usual care by 2015, where patients were getting those 2,000 cc's overall.  In terms of the quality of the evidence, we are kind of stuck, because what we don't have yet are studies that have comparatively examined different quantities and amounts of fluid  in combination with different vasopressors and rates of administration of those vasopressors. So we don't have the evidence to do strong comparisons between different approaches to  understand what the most effective approach is. That doesn't mean that we don't have a standard of care. The standard of care is defined by what people do in the community.  And what we've seen from the slides I've shown you already is that by choice, people are giving roughly 30 mLs per kilogram. And that is a de facto standard of care.  This particular paper was the ESICM Task Force on Fluid Administration for Acute Circulatory Dysfunction.  And they found just what we stated, that there was insufficient evidence to use the grade system to grade this evidence.  But they were able to rely on available evidence, that is what's published and what people do, and expert opinion. And that's the state of the field.  That doesn't mean we shouldn't measure that and include it as part of the quality bundle,  because it is, in fact, what people do. It is how people care for these patients left to their own devices.  They choose to give roughly 27 cc per kilogram if you look at the PRISM trial. And so what do we want to do then when it comes to setting standards for people and  trying to raise all boats? Well, we need to address how reliable we are as practitioners in health care delivery.  And what we know is that we don't deliver the best practices in health care reliably. There are a number of reasons for that. One of them is that people make mistakes.  And so if we just look at human error rates that are common in the course of daily human activities, insurers study this kind of thing because they pay for the mistakes that people  make. We see that misreading a label, for example, people do that three times out of 1,000.  An error of omission that you might make without a reminder in place might happen more frequently,  one time out of 100, whereas an error of omission, if you embed it in a protocol or a procedure so that you're actually reminded to do it, again, drops back down to that lower rate,  three times out of 1,000. So people make mistakes, and there are ways to prevent mistakes from occurring and remind  people to do the right thing, and that's part of why we do quality improvement in health care. Not only do people make mistakes, but people violate rules that are set up.  And this slide is essentially here for just a little bit of fun. It shows you that in virtually every circumstance, if you're going to violate a rule, men are  much more likely to do it than women. And so in the first example, compliance is unimportant. It's easy to violate the rule. There are a few inducements to comply with the rule.  Let's say, for example, this is surgical attire, wearing a skull cap. The Joint Commission doesn't like that, but my surgeons do.  And male surgeons are far more likely than the female surgeons to do it, 49% for the men and 35% for the women.  And this is true if you look at all these other examples that, depending on how unacceptable  the violation is, it's always the case that the men are more likely to violate the behavior.  If we look at studies of reliability in health care, Elizabeth McGlynn in 2003 in the New York Journal published a study of how often it is that we execute the best standards of  care for patients when they're admitted to our hospitals. So examining 7,000 charts and 400 indicators of quality of care, she was able to demonstrate  that the patients received the best scientifically indicated care just 55% of the time. And that meant that the defect rate in American health care is pretty high at 45%.  And this is probably not surprising to you. It also is a little bit of a retrospective scope, the advantage of looking to the past  to know what the patient's diagnosis was and what they actually had is the advantage of looking retrospectively. But in the heat of battle, it may be more difficult.  Nonetheless, I think this is a clear indication that we always have room to improve as we care for patients in our care. This is true in sepsis as well.  And so one of the early studies that we did on the surviving sepsis campaign looking at adjunctive and supportive therapies in sepsis was to ask clinicians, how often do you think  you provide the care you provide? And the difference between audit and interview was pretty stark.  So low tidal volume ventilation, for example, clinicians would say they did it 92% of the time.  But if you measured whether they were actually delivering low tidal volume ventilation, it was substantially lower in their charts.  And some of the times, this may be a technical violation, right? Instead of setting at 6 cc per kilogram, the patient had some other setting that the respiratory  therapist had created. The point of this is just to say that we believe that we execute care consistent with guidelines.  But when we actually audit what happens in charts and we deliver care in hospitals, we see that we're not as effective as we believe we are.  If you take these types of patterns, whether it's violation behaviors, the fact that people just make mistakes, or the fact that it's hard to remember to do the right thing, or  the heat of battle confuses us as to what the right therapies are to deliver to patients at the right time, we end up with the Swiss cheese.  This is when all the holes line up and the harm ends up reaching the patient. And so this is a common occurrence that adverse events happen in health care, and we should  seek to avoid those, obviously. This is why we have things like quality improvement approaches, bundles, guidelines, measures.  All of those words can be summarized much better in a single picture. Here if we had removed the endotracheal tube in this particular patient in the proper fashion,  I think we would have avoided this erosion that occurred. If this patient survives, they're obviously going to need plastic surgery to recover from  their illness or from their injury. And this could have been avoidable with the proper protocol to avoid leaving the endotracheal  tube in place, tied such as this, for as long as it was. In a particular sepsis case study, I don't think any of this will surprise you.  This is an actual institution that I visited as part of my work of the Surviving Sepsis Campaign.  And I'll just take you through time so that you see how many failure points occurred in this particular study.  So a patient came to the emergency department, normal intensive, a little bit hypothermic and a little bit of an elevated respiratory rate.  An hour later, an hour and a half later, the lactate came back at 3.1. Then the patient's blood pressure drops out, and that's between hours five and eight.  The patient gets some boluses at that point in time. Blood culture is not drawn until nine hours into the course of care.  The patient's admitted to the ICU, hypotensive, at the 10th hour. Antibiotics were started not until the 11th hour, way beyond current modern day standards.  Central access at 11 and a half hours, CBP, similar time, chest x-ray, and finally the initiation of vasopressors to treat the hypotension the patient's had now for some 15 hours.  And so although this may be an extreme example, this is actual care pattern, and I don't think  it would surprise you that community facilities, as well as some academic facilities, have fallouts like this that we see time for time in healthcare.  It's a precise example as to why we need to have guidelines, bundles, and appropriate quality measures to prevent errors like this from happening.  So we study the fallouts that do occur. Now when you do put something like a recommendation for 30 ml per kilogram in a quality measure,  there is the possibility of harm, and we can't disregard that. And there is evidence to suggest that there may be harm associated with fluids.  I don't want to disregard that. Here's a couple of studies that are suggestive. The FEAST trial looked at more than 3,000 septic patients in Africa, and they were resuscitated  with much more fluid than we would anticipate, 40 cc's per kilogram versus 10 cc's per kilogram. And the results favored the lesser resuscitation approach.  Is this a generalizable study to our adults in the United States or other countries that have sepsis and septic shock? I don't think so.  These are children with large amounts of fluid resuscitation greater than we're recommending in the present day.  If we look at the simplified severe sepsis protocol trial, this was a small number of patients, 112 septic shock patients in Zambia.  Again, they received more fluid than we're currently recommending, 2 liters plus 2 liters over 4 hours versus usual care.  And again, it favored the lower amount of fluid resuscitation. But the large amount of fluid that these patients received makes it difficult to say that this  applies to the current circumstance and recommendations for 30 mls per kilogram. I'm heartened, though, that large observational trials in the United States have suggested  to us that, in fact, patients have much greater tolerances than you'd expect. And so Dr. Merrick, Paul Merrick, did this study, who is of no great fan of SEP1 in particular,  and also 30 mls per kilogram resuscitation as a standard. But his study did demonstrate that in 23,000 septic adults, day one fluids, up to 5 liters  per day, are well tolerated with no evidence of increased mortality until after the fifth liter of fluid is given.  And so here, we see large tolerances in humans for initial resuscitation during day one. This is similar to the observational practice parameters that I showed you, the studies  that were done back in 2004 that suggested there were deficits between 6 and 10 liters, perhaps, in patients initially in the septic shock.  Let's just look at a couple trials that have suggested that even in the patients where you think that they may be the most harmed by receiving fluids, in fact, there either  is no harm or they benefit the most. This study was done by Vinny Liu at Kaiser in Northern California, looking at a before  and after trial of patients who were optimized to receive 30 mls per kilogram because they weren't receiving it at baseline.  And so this before and after trial looks at pre-bundle in 2012 and post-bundle in 2013. And before you look at the red boxes, look at the top line where it says all patients  with the N of 18,000. And you can see that mortality did decrease, both pre-bundle and post-bundle, in terms  of hospital mortality and 30-day mortality with a significant p-value in all comers. But if you look in the red boxes and parse those out between those with heart failure  and kidney disease versus those that did not have heart failure and kidney disease, what you quickly ascertain is that all the benefit derived by making sure that patients with  heart failure or kidney disease receive their full 30 mls per kilogram with significant p-values.  In those patients who didn't have heart failure and kidney disease, in the bottom red box, non-significant p-values, no change in mortality.  All the benefit was derived by making sure those that had the comorbid conditions received their full resuscitation.  The RIVERS trial suggested that patients who received a good resuscitation prevented further organ failure, in particular, the need for intubation.  And this is despite the fact that in the RIVERS trial, the early goal-directed therapy arm had a higher prevalence of CHF at 36.7% versus the control arm that had 30.2%.  And so what you see, the number of hours after starting early goal-directed therapy in comparing the control group to EGDT, is that there's a higher rate of intubation that  occurred in the control group in both up to the 72-hour mark compared to those who actually received more fluids in the EGDT arm, again, with significant p-values.  And if you don't believe that there's benefit, at least, from giving 30 mls per kilogram,  then perhaps you'll agree with the Cleveland Clinic study, which indicates comparing patients  who received 30 mls versus those who received less than 30 mls, that there was no difference  in intubation incidence, mechanical ventilation duration, or hours to intubation in this most recent publication in Critical Care in February of this past year.  So at least we see no particular harm. I would like to leave you with a thought as you consider these trials, which is that although  there's great hyperbole about the harm that 30 mls per kilogram can be done, I don't think  you will hear in the course of this pro-con debate the definitive study that says in adults with septic shock, those who received 30 mls per kilogram did worse in terms of any of  these parameters overall than those who did not receive 30 mls per kilogram. That evidence is lacking.  And so we either have benefit or no harm, but no studies yet which tell us in adults who receive 30 mls per kilogram that they are substantially harmed and routinely harmed  during the course of that therapy. So in conclusion, I'd like to remind you we're concerned about initial resuscitation, not  the ongoing phases of fluid management, and just the 30 mls per kilogram applies to that particular time.  I have no objections to reducing the amount of fluids patients receive after the initial resuscitation or ensuring that they eliminate unnecessary fluid after that period of time.  Excellence with the 30 mls per kilogram standard is voluntary in SEP1. Patients deserve to be looked at individually, and doctors do that all the time.  We expect that, and that is commonly done. Treat your patient. There's no penalty from SEP1 if you comply or if you don't comply.  It's simply a matter of a guideline, something that we measure as a standard of care that we believe, if properly applied to most patients, could help them.  We are unreliable in executing health care practices. That's why we do quality improvement. That's why you end up with recommendations like this that apply to most people, even  though there will be exceptions. The evidence suggests that 30 mls per kilogram is something that the field has chosen as a standard.  It's not something that's been imposed upon the field, and that evidence was clear from the PRISM trial.  And I left you, of course, with the notion that harm is unlikely, and there may be mortality benefits.  So in conclusion, I appreciate your time today, and it was a pleasure to speak with you. And I will now turn you over to the remaining participants in order to hear the con and  additional case studies. Thank you very much. Hello, ladies and gentlemen. I'm Manuel Moubrin from Brussels, Belgium, and I will give a talk regarding surviving  sepsis campaign guidelines where 30 ml per kilo of fluid may be too much. These are my conflicts of interest, my biggest bias being the co-founder of the International  Fluid Academy. So if water is the problem and fluid overload is detrimental for organ function, then de-resuscitation or fluid removal may be the solution.  I was very fortunate to have met Mathilde, Her Majesty, Queen of the Belgians, during the terrorist attack in 2016, and more recently, Philip, His Majesty, King of the Belgians,  during the first wave of the COVID pandemic. I also was fortunate to have met former Belgian Minister of Health Care, Maggie de Blok. My lecture will be on breaking bad habits.  We know that sepsis is a killer. Sepsis kills, but we need to put this into perspective. Every year, around 260,000 patients die in the United States of sepsis.  But if we look at the evolution of sepsis mortality, we see that after the introduction of the surviving sepsis campaign guidelines, the natural history evolution did not change  that much, and mortality is less high in Australia than the U.S. So the sepsis tree definitions define sepsis as an infection and organ dysfunction due  to a dysregulated host response, as can be assessed by the QUIC SOFA score. When we look at the three-hour bundle, you can see the measures there.  I'm not going into detail, but with respect to fluid resuscitation, the recommendation was 30 ml per kilo starting in the first three hours, a strong recommendation with low quality  of evidence. In fact, no evidence whatsoever. Over the years, we have given fluids to our patients, and much of our tradition came from  the blue stage of spasmodic cholera, where we saw that these patients were deeply in shock, hypovolemic, and fluids saved lives.  But fluids only have short-lasting effects, and giving fluid bolus did not result in better outcome in the FEAST trial in African children.  So 30 ml per kilo in three hours, this may result in fluid overload and peripheral edema or anasarca.  If you follow the guidelines from whatever way in the algorithm, the answer always is give fluids.  But if we perform a poll on Twitter, we found that 61% of 500 respondents do not support the SEPCIS guidelines. In 2018, there was an update, and again, there was not much evident.  Now these fluids should be given in one hour of time. It was also added that early vasopressors could be started. So 30 ml per kilo given in one hour of time.  Does this make sense, or did we end up in a complete and utter state of madness regarding  the SEP-1 mandate? So if you would be a normal American patient weighing 155 kilos, you should be given 4.6 liters in one hour.  If you don't do this and you live in the state of New York where the SEP mandate is held, you go to high court because of malpractice.  If you do give such large amounts of fluids, you may also go to high court because of fluid overload.  So fluids are drugs, and let me introduce a conceptual framework of the seven Ds of fluid therapy.  It's on definitions, D2 diagnosis, the drug, the dose, duration, de-escalation, and finally discharge.  D1 multiplied with D7 gives D to the power of 7, and this is fluid stewardship. This is what we should be doing.  So D1 is on definitions, and I'm not going to overload you with boring definitions, but let me point to some of them.  We all know the fluid balance, which is the daily sum of all intakes and outputs, and we can calculate the cumulative fluid balance.  Fluid overload is defined as a 10% increase of body weight from baseline, and it's associated with worse outcomes.  Early goal-directed fluid management is what has been advocated by the surviving sepsis guidelines.  Late conservative fluid management is two consecutive negative fluid balances within the first week, and late goal-directed fluid removal is active fluid removal by means of  diuretics, and this is what we refer to as de-resuscitation. Fluid overload results in hypervolumia, hyperhydration, and peripheral edema.  We know that hypervolumia will result in edema. On the other hand, if the patient is edematous, this does not necessarily mean that there is hypervolumia.  So maybe we should avoid the term fluid overload as suggested by Michael Pinsky and Jean-Louis Vincent.  There are a lot of definitions surrounding fluid therapy, and I'm not going to address them in detail, but they will be mentioned during my talk.  D2 is on diagnosis, and we know that hypovolumia is bad. Hypervolumia may even be worse, as there will be a diffusion problem and oxygen will not  get to the tissues because of interstitial edema. Fluid status matters. And there are a lot of myths surrounding fluid therapy. Septic patients are not always hypovolemic.  Surgical patients are not always hypovolemic. Sepsis does not mean that the patient loses large amounts of insensible losses.  Peripheral edema does not equal that we need to give diuretics, and a low preload does not equal that we need to give fluids.  Septic vasopressors may be indicated because septic patients are vasoplegic and not necessarily hypovolemic. D3 is on the drug, and we need to ask ourselves four questions.  Of course, we must not do harm to the patient. First question, when to start fluids? Don't give fluids if the patient doesn't need them.  Second question, when to stop fluids, when the patient has neural intake or shock has resolved. But many patients will accumulate fluid and may need active fluid removal.  And of course, we must stop removing fluids to avoid hypovolemia, hypoperfusion. This is referred to as the resuscitation.  Resuscitation fluids gained most of the attention in the medical literature. But if you look at this study by Niels van Regemortel, about thousands of patients, he  found that only 6% of all fluids are for resuscitation, 25% is for maintenance, 33% for nutrition, and a vast 33% is fluid creep, fluids that seep into our patients unwanted.  If you would give one liter of glucose 5%, only 10% will remain intravascularly. For crystallates, it's about 25% after one hour.  And coloids are assumed to remain 100% at least for the first four to six hours. D4 is on the dose. As Paracelsus nicely stated, it is the dose that makes the poison.  Key concept from dosing comes from antibiotic stewardship, where we look at pharmacokinetics and pharmacodynamics. D5 is on duration.  And of course, we should not give fluids to treat the numbers. We should use fluids to treat shock. D6, as with antibiotics, is on de-escalation.  And let me introduce you a conceptual model on the four phases of fluid therapy. In 2014, around the same time, two papers were published.  Looking into this, we suggested the ROSE acronym. And if you would ask Twitter, there is a clear preference to use ROSE instead of SOSD. This is a ROSE.  And if you would give water to the flower, it will blossom. This is the resuscitation.  Then we need to support organ function, stabilize the patient, de-escalate, and in some patients, we need to evacuate the excess of fluids, hence the ROSE acronym.  And this represents the four phases of fluid therapy illustrated here. After the first hit, fluid balance should be positive. Then we need to maintain neutral balance.  And then we need to get rid of the excess fluids. This is illustrated by the ROSE. And the ROSE at the end is just a dry one, not a dead patient.  I'm very happy to see that this ROSE concept has been taken over by other colleagues like the ones in neurocritical care.  But the problem is that after giving fluids and rescuing the patient with goal-directed  therapy, optimizing organ function, followed by late conservative fluid management to stabilize  organ function, followed by late goal-directed fluid removal to evacuate the excess of fluids and to save the patient.  The problem is that in the initial phase, we will get treatment based on macrohemodynamics, passive lack raising, fluid responsiveness.  We may stop fluids when these parameters normalize. And we may evacuate the patient when there is an impact of fluid accumulation and organ  function evidenced by a high lung water, a low PF ratio, or a high abdominal pressure. But the problem is that there is no one single study examining these different phases.  We need to set the targets right in order to reach our goal. And the goal is avoiding global increased permeability syndrome.  Because giving fluids in septic patients will lead to second and third space fluid accumulation.  This standard abdomen elevated intra-abdominal pressure, compression of the vena cava, and  reduced blood flow to the heart, and a reduced venous return, and a reduced flow to the organs, hence inciting a vicious cycle.  A positive fluid balance is an independent predictor of poor prognosis in patients with septis and restrictive fluid regimens improve outcome. Why is that?  Because fluid overload at the inside mirrors fluid overload at the outside. And this will result in fluid accumulation, venous congestion, and organ failure, or does  the global increased permeability syndrome. This is why late conservative fluid therapy is more important than early goal-directed  treatment, as you can see in these studies looking at early conservative, late conservative fluid management carrying a better prognosis.  So it may be time for restrictive fluid therapy in critically ill patients. Therefore, anasarca is not cosmetic, but I believe it is bad medicine.  And whenever a patient is not transgressing spontaneously from the app phase to the flow phase of shock and remains in a persistent state of GIPS, we need to de-resuscitate.  So GIPS is defined as a persistent positive cumulative fluid balance with new onset organ failure dysfunction as evidenced by lung water or abdominal pressure.  There are many clinical signs or biochemical signs, also imaging signs, to assess fluid overload.  But I would like to focus on the hemodynamics, organ function, and maybe some other techniques.  Importance is that we individualize our treatment and we pick the one patient that may benefit from fluid removal.  If we look at hemodynamic effects of fluid overload, they may result in increased volumetric preload parameters, and we need to put them into perspective of ejection fraction and  the absence of fluid responsiveness. Looking at the impact on end-organ function, my favorites are the increase in lung water  and pulmonary vascular permeability ratio and, of course, intra-abdominal pressure. In the future, we will learn more techniques to monitor fluid status, circulating blood  volume, and fluid overload. This can be done by bioelectrical impedance analysis or maybe with isotope dilution techniques like DAXOR.  But don't forget venous excess examined by ultrasound and the VEXUS score. After all this, you can read in some recently published open access papers that can be  downloaded for free, and I invite you to have a look at them. One very last more thing is the fruzomide stress test, which is well known by nephrologists  giving one milligram fruzomide per kilo to assess tubular function and reserve. If urine output is less than 200 ml in the following two hours, that may be a good predictor for AKI.  So maybe we can do a ready for de-resuscitation test where we look at the slope of urine production. If the patient qualifies de-resuscitation, you could use PAL treatment.  By putting the PEEP equal to the abdominal pressure, we squeeze the fluids out of the alveolus, out of the lungs, into the interstitium, and by adding hyper-oncotic albumin, we get  the fluids into the vascular beds and with the fruzomide out of the patient. We have looked at this in a pilot study in septic patients with acute lung injury.  PAL group had higher PEEP, higher albumin because it was substituted, and this resulted in a lower abdominal pressure, a better PF, and a lower lung water, followed by better  survival and faster weaning. So ladies and gentlemen, to wrap things up, I hope I convinced you that IV fluids are  not innocent bags of water, but IV fluids should be treated as drugs. Please take into account the seven Ts and the conceptual framework of personalized fluid  therapy and get your fluid team ready. So I would like you to go home with the following messages. Fluid overload is defined by a cutoff value of 10% increase of fluid accumulation.  It has an increased mortality and morbidity and is associated with end-organ failure by global increased permeability syndrome. How to guide de-resuscitation?  What clinical and lab science are aspecific? Use imaging. Use volumetric preload. Check the absence of fluid responsiveness and the impact on end-organ function and new  techniques will be promising in the future like VIA, EIT, and Daxor, but don't forget venous congestion, peripheral edema is bad medicine. I thank you for your attention.  Well, hello and welcome to the session on Pro Con Water, Water Everywhere, IV Fluids. Where's the pendulum now? Like where are we with this age-old question now?  I'm Rod Fontenette, I'm an emergency medicine, critical care medicine physician, active duty Air Force stationed in beautiful Northern California and working full-time clinically  at UC Davis Medical Center here in Sacramento, California. I spend time in the emergency department as well as in the SICU, CTICU, and neuro ICU.  My co-presenters are Dr. Sean Townsend. Sean is out of California Pacific Medical Center and he will be presenting on the pros, right?  He's going to be covering the guidelines, say 30 mLs per kilo, just discussing that in a little bit more detail.  Dr. Manu Malbrain, he's coming to us from Belgium, from Brussels, and he will be discussing the cons, right? So it's time to turn the faucet off, patients are drowning.  So we should have an amazing pro-con debate. The way that I will present is I will just give a very brief intro into the topics that  we'll be discussing and kind of shape the framework as to the why is this important. I'll also set the scene with the case and that case will be used to kind of guide the  resuscitation and guide the discussion from Sean and Manu. So first off, I would like to thank the Society of Critical Care Medicine for giving us this  opportunity to present in a very unusual way, right? So everything has now been converted to virtual conferences because of COVID-19, but we're  going to press on either way and still deliver exceptional content. And I would also like to thank each of you for taking this time out to attend this session  on something that we just keep discussing because it is extremely important. So without further ado, let's jump right into the slides.  So first off, as I said earlier, I'm active duty Air Force. So I always like to include this DOD statement stating that these are my views and not the  views of the United States Air Force, the Department of Defense or any other government entity. These are my views and my views alone.  The Department of Defense and the Air Force does not endorse any specific resuscitation technique, IV fluid therapy, any product. Financial disclosures. I have none.  No one's paying me for my thoughts. This is just me, right?  Not to say that I'm not willing to accept offers, but no financial disclosures to disclose. All right. So why is this important? Right. So let's set the scene.  So substance by the numbers, right? So it's about 30 million cases a year. 1.6 million diagnosis each year here in the United States. It's a 19% increase spanning from 2011 to 2019.  And it's only getting worse, right? As the number one cause of hospital readmission here in the United States, two thirds of septic  patients enter the health system through the emergency department, right? So just waiting till these patients get upstairs just isn't acceptable, right?  We have to be able to meet these patients where they are, right? And if your hospital is anything like my hospital where boarding is an issue, these patients  are staying in the emergency department that much longer. So waiting until they make it upstairs just isn't acceptable, right?  So Scott Weingart always says, bring an upstairs care downstairs to the emergency department. And so that's why this is just important, right?  And the mortality associated with this is about 25 to 30%. And that number has been staying around the same for the past few years.  So an IV fluid resuscitation and a guided resuscitation is at the forefront of any conversation when it comes to substance and substance management. All right.  And what happens is the patients enter into this vicious cycle, right? So they have hypovolemic or distributive shock, initiate fluid resuscitation.  And we'll talk about where we are with that. And my colleagues will kind of do a much deeper dive into that.  And there's interstitial edema, visual swelling, intra-abdominal hypertension, and the cycle just continues, right?  Venous hypertension, oliguria, and I'll talk very briefly on major acute kidney events and acute kidney injury and the need for dialysis. And the cycle just continues, right?  Well, how do we break that cycle? And what do we have to guide us in terms of fluid resuscitation?  This changes and goes back and forth almost every year with every iteration of updates. But I can tell you now, there's still more to come, right?  So in talking about the guidelines, right? Guidelines are driven by early goal-directed therapy with Dr. Rivers from back in the early 2000s.  CMS also kind of helps drive this just a little bit. And then the Surviving Sepsis Campaign and their guidelines in terms of IV fluid resuscitation.  And I know if you're attending this talk, you're all familiar with the Surviving Sepsis Campaign bundle in terms of 30 mLs per kilo.  And Manu will talk about what's the evidence behind this, right? We said, hey, everyone should be met with 30 mLs per kilo within the first hour,  the first three hours of resuscitation. But how strong is the data to support that, right? So there's more to come on that, right?  And this is from the Surviving Sepsis Campaign, right? So measuring lactate, just giving us guidelines and targets to reach.  And if you look at number four, it says begin rapid administration of 30 mLs per kilo of crystallized for hypotension and or serum lactate are greater than or equal to four.  Again, though, the question is, where's the data to support that, right? Show me the data. And that's what we hopefully will be able to do.  And I think we'll do a very good job at it in this talk. All right. So IV fluid resuscitation. What I like to say is that not all IV fluids are created equal, right?  Mike Winters out of Baltimore Shock Trauma always likes to say, there ain't nothing normal about normal saline. I agree with that 100%.  And I like to treat IV fluids just like any other medication that we reach for, right? When the nurses go to the PICSs and they get out these medications,  our pharmacy colleagues go to the PICSs and they get these medications out the PICSs. They always go through the right patient, right route, right dose, right?  And they use the Ds, right, to make sure that it's the right drug, right? I say that same concept and philosophy should be applied to IV fluid resuscitation, right?  And we'll talk about that more in just a minute. And the question is, but okay, I've started resuscitation. How much is enough, right? And how much is too much, right?  It's like that Goldilocks theorem, right? Not too little and not too much. But sometimes I just don't think we have the most ideal tools to help guide us and say,  okay, we can stop resuscitation now because I think we've gone over into that too much aspect, right? And so we see this quite frequently,  especially by the time the patient is a day or so with us in the intensive care unit, a few days into the intensive care unit with us,  where we've kind of gone into that fluid overload status almost, but now the patient has peripheral impeding edema.  What I would love to say is that that ain't just on the lower extremities, right? So that's bowel edema. That's edema in the lungs, right, pulmonary edema.  It's in all the other tissues, right? It's in the liver. It's in the kidneys, right? So this has very far-reaching consequences.  We're only able to see it clinically sometimes just by the pitting edema and looking at how much weight the patient has gained since their time with us here in the hospital, right?  So in terms of that sweet spot, right, where we kind of ebb and flow, right, so the very early phase of resuscitation where there's inadequate fluid, right,  so the patient is hypovolemic, right? So there's cardiopulmonary events. There's tissue hyperperfusion, the inflammation and surge-type response associated with it,  and then there's organ failure, right? So we're trying to get into that optimal fluid balance, right, in terms of now we've just done the very early resuscitation,  and then now we're going on to maintenance, and then we're going to eventually transition on to the de-resuscitation or the de-escalation phase.  And, again, my colleagues will kind of do a much more deeper dive into this. I'm just trying to whet your appetite so that we can kind of start forward thinking  in terms of what's more to come with the discussion.  Some of you guys have all seen this, right? So, again, just as I stated, yeah, we may be able to see the peripheral edema,  but what's going on in a more widespread or systemic approach, right? So there's the pulmonary edema, the pulmonary fusions, the ultra-pulmonary  and chest wall elastance with decreased lung compliance, right, hepatic congestion, GI complications, abdominal wall complications, renal failure, acute kidney injury  the cardiovascular complications that are associated with this. So I think this slide does an amazing job of kind of tying it all together  to show that those fluid bonuses that you're giving have very widespread complications and things that we all need to take into consideration when we're trying to determine  is this patient no longer fluid responsive? And if not, then maybe we need to be transitioning over to starting them on basal presses sooner than later, right?  In terms of fluid overload, I think this is a great chart, actually, I think that highlights that just in a 10% fluid overload status,  you can see that predicted adjusted odds of major acute kidney events begin to go up significantly, right?  30% to 40% fluid overload, as you can see, that number just continues to climb. And these are all things that we need to take into consideration  when it comes down to fluid resuscitating these patients. All right, so this kind of brings it home a little bit more, right? So fluid imbalance and mortality.  This was a trial done and published in Critical Care Medicine back in 2008. Just over about 3,100 patients, 198 ICUs through 24 European countries.  And what this showed clearly was that there was a positive fluid balance that was independently associated with an increase in mortality. So again, should I continue?  Is this patient still fluid responsive? I think that needs to be at the forefront of your thought process before you reach for and start any additional liter of IV fluid.  Is there going to be any additional benefit? And ROSE kind of gives us a framework in terms of resuscitating these patients. So it's the resuscitation phase, the optimization phase.  Now that I've kind of gotten the patient kind of over the hump into this sweet spot, the stabilization phase will now just kind of support them a little bit more  through whatever that initial insult that got them here with us. And then the evacuation or the de-resuscitation or de-escalation phase,  where we start looking at things like diuretics and just trying to mobilize some of that fluid to hopefully start working towards getting the patient off the ventilator  if they're on the ventilator or whatever the case may be. But I think ROSE kind of helps to, again, set the framework  and gives us something to kind of aim towards in terms of resuscitating our patients. But the question is, are we waiting too late to start vasopressors?  And I know that this is always a topic of discussion in emergency medicine because sometimes we always try to hang our hat on,  well, I've got to give them the 30 mLs per kilo before I can start the vasopressors. That is not the case at all.  Another typical mindset is, well, I don't like to give vasopressors through peripheral IVs. So maybe I need to, okay, get a central line in before I can start these vasopressors.  That also is not a good theory, right? I would say if it's a large-bore IV that I can trust, I probably wouldn't do it in the hand or on the finger, right?  If I had a large-bore, say, AC IV, I would say transitioning to vasopressors sooner than later while I am still in the midst of resuscitation is key.  It's extremely important, right? So now let's set the scene. Let's discuss a case, right?  And what I will do with this case is kind of shape the discussions for the pro and con debate that my colleagues will be talking about in the later parts of this lecture.  So who do we have? We have a 53-year-old male that presents to us in the emergency department  for evaluation of an altered mental status and progressively worsening shortness of breath. Sound familiar? He was recently diagnosed with community-acquired pneumonia,  was discharged home on an oral antibiotic, and he reports that his symptoms have progressively worsened, right?  Now to the point to where he can't even get up and walk from the sofa to the kitchen without having extreme shortness of breath, right? Again, sounds familiar?  Let's look at the vitals, right? The patient is hypoxic on Romare. He's hypotensive, right, 69 over 38. He's tachycardic, right, so the patient's shock index is greater than one.  So this patient is extremely sick, respiratory rate of 37, right? So this is a patient that we see almost once or more shift in the emergency department,  and these patients obviously make their way to us in the intensive care unit, right? So this is the chest X-ray. What do you think about the chest X-ray?  That left upper lobe looks pretty healthy. The rest of that chest X-ray looks like this patient is having issues, right?  So, again, something that we see far too often, right, in the ED and in the ICU, and you can have the debate on is a lot of this fluid in the lungs,  so pulmonary edema, or is this infiltrates, and the patient is kind of going on to moderate to severe ARDS, right?  And then that's what's going to help shape and frame this discussion and shape and frame this conversation, right? So what I would like to do is when you approach this patient,  is making sure you have the right information, right, not caught with your pants down, right? You don't want to know what to do next, right?  What my colleagues are going to discuss is when is it appropriate to give IV fluid therapy in the very early phases of the resuscitation,  and what is the literature and the data that supports that? We'll also discuss what is it too much, right? When have I given them too much IV fluid,  and now I can begin to kind of de-resuscitate or de-escalate that therapy to kind of set this patient up for success in the days ahead, right?  So without further ado, I would like to hand this over to my colleague, Manu. He will take it from there and discuss the pros of IV fluid resuscitation.  I'm sorry, discuss the cons of IV fluid resuscitation, followed by Sean Townsend, who will discuss the pros, and then at the very end, we'll come back and revisit this case  to tie it all together to make this clinically relevant. If you have any questions, again, my name is Rod Fontenette. This is my contact info.  I'll see you guys towards the end of this discussion where we'll, again, begin to wrap this up and make this clinically relevant and bring this home to you  so that way you can apply this to your next patient or your next shift. Thank you for your time.

Video Summary

In this session at the SCCM Critical Care Congress, the topic of debate is the use of intravenous fluids. The session aims to discuss the risks and consequences of aggressive use of intravenous fluids, evaluate the objective of protocol suggestions, and describe intravenous fluid strategies that can be implemented at the bedside. The debate centers around the question of whether the current guideline of administering 30 mL per kilogram of crystalloid fluids for septic patients is appropriate. The first speaker, Dr. Shawn Townsend, argues in favor of the guideline and presents his qualifications and experience in the field. He emphasizes the importance of individualized care and explains that the guideline is a standard meant to improve the quality of care, but it can be deviated from if necessary. Dr. Manuel Malbrain, the second speaker, argues against the guideline, presenting evidence that suggests fluid over-respiration can lead to harm. He advocates for a more personalized approach to fluid therapy and the consideration of factors such as fluid overload and organ dysfunction. The final speaker, Dr. Roderick Fontenette, gives an introduction to the topic and presents a case study to guide the discussion. He highlights the need for an optimal fluid balance and the potential consequences of fluid overload. Overall, the session aims to provide insights into the ongoing debate surrounding IV fluid use and offer strategies for clinical implementation.

Asset Subtitle

Pharmacology, Sepsis, 2021

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2021 Critical Care Congress held virtually from January 31-February 12, 2021.

Learning Objectives

Explain the risks and consequences of aggressive IV fluid use
Evaluate the objective of protocol suggestions and how application in clinical scenarios might vary
Describe IV fluid strategies that can be clinically implemented at the bedside

Pro: Guidelines Say 30 mL/kg
Sean R. Townsend

Con: Patients Are Drowning
Manu L. Malbrain

Case Presentations
Roderick Fontenette

Meta Tag

Content Type Presentation

Knowledge Area Pharmacology

Knowledge Area Sepsis

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Tag Crystalloids

Tag Fluids Resuscitation Management

Year 2021

Keywords

intravenous fluids

aggressive use

protocol suggestions

crystalloid fluids

septic patients

individualized care

fluid therapy

fluid overload

organ dysfunction

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.