Hello, my name is Michael Conner. I'm an intensivist and a nephrologist at Emory University School of Medicine, Emory Healthcare, and Grady Memorial Hospital in Atlanta, Georgia. I want to thank the organizers for inviting me to participate in this SCCM masterclass on saving the kidneys and preventing acute kidney injury in our intensive care unit patients. I was assigned two topics that have to do with fluid management in the intensive care unit. This masterclass was intended to be presented at the 2022 SCCM Critical Care Congress in Puerto Rico. But unfortunately, we are not there, and so we are learning together on this virtual platform. I have no significant disclosures that affect the topic that we are discussing today. I do receive some research collaboration support at my institution from Rediscovery Life Sciences, which has to do with acute kidney injury and acute kidney injury therapeutics, but is not involving volume management in any way. Importantly, my wife is on faculty at Emory and she has no conflicts of interest. At the end of this talk, I want you to be able to describe the data of fluid overload increases the risk of death and understand pathophysiology and how fluid overload may cause organ dysfunction. Whenever I give lectures, I always believe that I should start upfront with our take-home message. To summarize what we're going to talk about today, I'm going to describe that fluid overload causes organ and kidney failure, prevents organ recovery, and prevents patient recovery and survival. Fluid overload remains very common in our hospitalized patients, both in ICU and non-ICU patients. We know that in the absence of some standardized approach for volume management and volume removal in ICU patients, our total body volume and weight tends to increase steadily for at least the first seven days in the intensive care unit. When we talk about fluid overload, it's important to understand that there is multiple different ways to define this. For simplicity's sake, obviously, fluid overload is defined by fluid intake being greater than the output. This can be measured both continuously or in a dichotomous fashion. Most studies tend to define fluid overload as a percent fluid overload. How much percent fluid overloaded is this patient based on their baseline? In other words, if you are normally 70 kilos and the patient is now 77 kilos, they've accumulated seven liters of fluid during the hospitalization, that patient would be 10 percent fluid overloaded. Turning to some data and to some studies. Fluid overload decreases patient survival in the ICU. I believe very strongly that this is yes, it indeed does. This is not just an association, but as a causation, and I'm going to discuss that as we go. There are many, many studies that continue to show the same thing across all spectrum and flavors of ICU patients, in pediatrics, in adults, in surgery, in medicine. I don't have time to go through all of those studies. I'm just going to show you a few studies that I think highlight some important points. First, in adult blunt trauma patients, fluid overload increases the patient's risk of complications in the ICU and the risk of death. It increases your risk of ARDS, multi-organ failure, abdominal compartment syndrome, surgical site infections, and death as well. In septic shock, fluid overload increases the risk of death. This is the vasopressin study or the VAST study. And as you went up in quartile of fluid accumulation, there was an increased risk of death, both based on 12-hour fluid balance and four-day fluid balance. In septic shock with ARDS, we know from several studies, this is just one example, in which patients who survive tend to have less fluid accumulation compared to patients who are non-survivors. This study was also very interesting because it broke patients down into whether they received an adequate initial fluid resuscitation based on surviving sepsis guidelines when they were admitted versus inadequate fluid resuscitation. And then they further divided patients as to whether they had a conservative versus a liberal fluid strategy after the initial resuscitation. And what they showed was that whether you had an adequate resuscitation or an inadequate resuscitation, if you then had a liberal fluid strategy, your risk of death was significantly higher than if you had a conservative fluid strategy after your initial fluid resuscitation. Even more interesting is the fact that the liberal fluid strategy seems to come at a cost that decreases the benefit of having had the adequate initial fluid resuscitation. So in this cohort of patients, they were adequately fluid resuscitated, then had liberal fluid strategy, and they had a higher risk of death than patients who had an inadequate fluid resuscitation, but then people were cautious with how much fluid to give. And in this study, an inability to achieve a conservative late fluid management strategy increased your risk of death sixfold. What about in cardiac surgery? We have numerous studies consistently demonstrating that accumulation of fluid increases the risk of post-cardiac surgery complications and death across all different types of patient populations. I'm just going to show you one study here, which I think is very instructive. In this study, the investigators looked at around 90 patients and looked at their cumulative fluid balance, their peak cumulative fluid balance over the course of the first eight days, and its association with complications after an on-pump cabbage. So for those of you who don't work in a cardiac surgery ICU, on-pump cabbages are our lowest risk operations. These patients should have the lowest risk of death and the lowest risk of complications, and standard of care is that they should be extubated within six hours and generally out of the intensive care unit by the next day. As the peak cumulative fluid balance increased, the risk of complications increased significantly as well. And the risk and the ability to escape the operation without complications went down significantly. Finally, this was a fairly recent meta-analysis of many different observational studies looking at the association between fluid balance and outcomes in the intensive care unit. And regardless of what metric one used, the presence of fluid overload or cumulative fluid balance or the peak fluid overload or the peak cumulative fluid balance or any other metric or any type of patient that we were different types of flavors of patients, we can see that in every situation, there's a statistically significant increased risk of death in the presence of fluid overload. This is especially true in acute kidney injury. Fluid overload is terrible when it occurs in the setting of acute kidney injury. This is data from the PICAR database, which is a prospective database of about 620 critically ill patients with acute kidney injury in the United States across multiple academic medical centers. They defined fluid overload as a dichotomous variable as to whether or not you were greater than 10% increase in weight from baseline. And in this study, if you met this definition for fluid overload at the time that your dialysis was initiated, you were two times more likely to die. I think this is instructive because I think we all know that at times we sometimes were stalled dialysis in the setting of acute kidney injury because we don't necessarily perceive the patient to be fluid overloaded enough. But this would suggest that there is a cost for waiting too long and allowing the patient to accumulate too much fluid before initiating dialysis. And this was further highlighted in this study by whether or not these patients were dialyzed or non-dialyzed. There was a linear increase in risk of death based on how much additional fluid overload that the patient was. And additionally, this study showed that the number of days in which you met this definition that you were on dialysis was also associated with increased risk of death, suggesting that when a patient is fluid overloaded and is started on dialysis, we should endeavor to decrease their fluid balance to closer to uvulemia as soon as possible. And just to show you that this holds true in other situations, the European study called the FINACI study looked at exactly the same thing and showed exactly the same results, that when you're fluid overloaded at the time of dialysis initiation, you had a decreased chance of survival, and there was a linear relationship between the amount of fluid you accumulated and the risk of death. This is an important study in patients with acute kidney injury from the MIMIC-2 database, around 10,000 critically ill patients in the Harvard system, retrospective study that confirmed what we had been thinking about for a long time, which is that death from acute kidney injury is not really from acute kidney injury itself, but really from the complications of acute kidney injury. So when you correct for fluid overload, hyperkalemia, and metabolic acidosis, there really isn't a significant excess mortality in stage three acute kidney injury compared to stage one acute kidney injury, that really it's these three problems that are driving most of the deaths we see in patients with advanced acute kidney injury. So in summary, fluid overload causes organ and kidney failure, prevents organ recovery, and prevents patient recovery and survival. I hope I've shown this to you, that there is lots of data that supports that fluid overload prevents our patient recovery and survival. But when I give these lectures, I oftentimes get a lot of questions or complaints about the data that I've shown so far. The first being that fluid accumulation may simply be a marker of disease severity, such that the sicker patients are getting more volume. I don't have time to go into all of these studies today, but there are clearly studies that demonstrate that this is likely not entirely true. In many studies, the patients who are less severely ill tend to get more fluid than the patients that are more severely ill based on Apache or SOFA scores. People oftentimes point out, which is quite correct, that all of these studies are purely association and not causation studies. And people do question what exactly is the mechanism. If it is a causation, if fluid overload causes these problems, then there has to be a mechanism. So that's what we're gonna focus on a little bit for the rest of this talk. What are the mechanisms that surround this stuff? First of all, I wanna just do a quick review of blood flow mechanics. Obviously, you all are very smart and know a lot of this stuff, but I just want us to all be on the same page. The first is that blood pressure does not equal blood flow. I think we get this confused a lot in an ICU. Especially some of our trainees will say, well, the blood pressure is good, there must be good flow. And the goal is organ perfusion, not blood pressure. The goal is flow of blood to the organs or through the organs, not necessarily just having a certain mean arterial blood pressure. So what is the pressure and flow relationship? So we all know if we have a container of water that's just sitting on a table, it doesn't flow anywhere because we have no pressure difference here. There's gravity pulling down. There's an equal pressure pushing up this way. And then the water doesn't flow out because we have atmospheric pressure pushing down this way. And everything is at steady state. In order to have blood flow through a pipe or a vessel, we have to have a difference between high pressure and low pressure. If we don't have a difference between high pressure and low pressure, then the flow will be limited. And so in our body, our heart generates high pressure on the arterial side. Our venous pressure is low on the venous side, and therefore we have blood flow from high pressure to low pressure. This is all pretty straightforward, hopefully. Blood flow is the, or the flow of any fluid, is the volume and rate of which that fluid is in the pipe or the vessel. Blood flow and blood pressure do have very similar determinants, which is cardiac output and systemic vascular resistance. Cardiac output, we know it's determined by preload, stroke volume, heart rate. Stroke volume is determined by left ventricular and diastolic volume and contractility of the heart. But stroke volume is also determined by afterload, which is largely determined by artery constriction and dilation. So as afterload goes up, stroke volume can actually go down, especially if contractility is impaired. That's what happens in our heart failure patients. It's important to remember that systemic vascular resistance is also determined by compartmental pressures and venous capacitance and venous pressures. And it's this area that we oftentimes lose sight of, and it's exactly this area that leads to volume overload impairing organ perfusion. We know that accumulated fluid resides in multiple compartments, such as our intracellular, extracellular, intravascular, as well as our intraparenchymal and interstitial compartments. Contrary to some popular belief, we cannot infinitely expand the third space. As the third space expands, the extracellular fluid volume increases. The pressure in that third space also increases, driving more fluid into the lymphatics, which drains into the lymphatic duct into the SVC. So we cannot infinitely accumulate. And that ultimately fluid accumulation leads to venous congestion and venous hypertension. So under normal conditions, venous pressure, remember, is really quite low. The pressure at the venous end of the capillary, in other words, the capillary that's draining into the vein, is only around 12 to 20 millimeters of mercury. Right atrial pressure is minus two to four, depending on the position and the portion of the spontaneous respiratory cycle. Central venous pressure is variable, and it depends on the site of the instrument and the position of the subject. So for example, if I'm standing up and I measure in the femoral area, my central venous pressure due to gravity is gonna be different than if you're measuring it in my IJ. And that increased CVP impairs capillary drainage, which leads to organ and interstitial edema and organ failure. So why does this happen? So remember we said that there is a flow and pressure relationship. We have to have high pressure to low pressure. However, if we have hypotension on this side or volume overload leading to increased CVP on this side, or God forbid, a combination of both of these, our pressure gradient suddenly rapidly disappears and flow slows down significantly. So increased venous pressure is a form of venous afterload and it impairs capillary drainage. And therefore it decreases venous drainage of the organ, decreases organ drainage then, and therefore decreases organ function. And just to show you that this really is true, I'm gonna show you a couple of stark examples. What is this? What has happened to this patient's leg? This is something called phlegmasia. And in phlegmasia, what occurs is you get acute venous occlusion of the central vein of the limb. And as a result, you get arterial ischemia and compartment syndrome because you can't drain blood out of this area. You fill up with a huge amount of interstitial fluid. There is no longer a pressure gradient from the arterial side to the venous side because of significant venous hypertension. And therefore arterial perfusion goes down. You get arterial ischemia, compartment syndrome, and critical limb ischemia, muscle and skin necrosis and limb loss. I've seen this many times in patients. Venous congestion can therefore lead to organ dysfunction in a whole host of different organs. It can cause cerebral edema and impaired cognition and delirium. Even if it's not malignant edema, high CVP will impair drainage from the brain and can cause some mild cerebral edema and oftentimes is contributing to why our patients are delirious. Obviously pulmonary edema, we can get myocardial edema, decreased conduction abnormalities and impaired contractility. Hepatic congestion was congestive nephropathy, congestive, excuse me, congestive hepatopathy, congestive nephropathy, as well as gut edema, malabsorption, ileus, and the such. I'm gonna zoom in on the kidney for a second and just take this one step further. So as we have increased venous pressure, that leads to a backup of fluid and blood can't drain out of the kidney, which leads to interstitial edema. This ends up leading to increased interstitial pressure because the kidney is encapsulated and cannot swell very rapidly the way a liver would be able to swell. As a result, as the interstitial pressure increases, that leads to compression of the individual nephrons, raising the pressure inside the nephron, which then raises the pressure in Bowman's space and reduces the filtration gradient then from the glomerular capillary pressure into the Bowman's space pressure because Bowman's space is presenting glomerular filtration. This leads to oliguria, concentrated urine with a high specific gravity, bland urine sediment if you look under the microscope, in other words, no findings of obvious ATN. If you're inclined to check, you'd have a very low fina or fractional excretion of urea. And this for all intents and purposes looks like a very intense perfusion mediated, what historically may have been called pre-renal acute kidney injury physiology. The problem is that this physiology is oftentimes misinterpreted very frequently as total body fluid overload, but intravascular volume depletion. And in fact, that's not really the case. This is just intravascular volume overload causing a perfusion mediated congestive nephropathy or acute kidney injury. We also know that there's a linear accumulation between fluid balance and intra-abdominal pressure. And we know that intra-abdominal hypertension further causes all sorts of organ dysfunctions, including further perfusion abnormalities to the kidneys, chest and diaphragm compliance challenges, lung function issues, cardiovascular challenges. And just to show you, intra-abdominal hypertension and abdominal compartment syndrome are extremely common in our intensive care units, affecting anywhere from 50 to 80% of patients having intra-abdominal hypertension. So you combine intra-abdominal hypertension with systemic hypotension, and now you have further erosion of organ perfusion. So these are the mechanisms that I think clearly demonstrate why fluid overload causes organ and kidney failure and organ recovery and contributes and causes, therefore, our patients to have decreased survival and increased risk of death. But just to remind you that organ recovery does require organ perfusion, and that perfusion to the tissue beds requires adequate blood volume, appropriate cardiac output and appropriate venous drainage. And therefore venous hypertension leads to organ edema, decreased organ function and decreased perfusion. And perfusion pressures can be defined as arterial pressure minus compartment pressures plus venous pressures. So in conclusion, in summary, fluid overload remains extremely common in hospitalized patients. I've told you that fluid overload impairs organ recovery and increases mortality. And we need to recognize the phases of fluid management as patients progress through their hospital course, from resuscitation to stabilization and active de-resuscitation, and we'll talk more about that later. So in summary, yes, fluid overload is the problem. This is a picture of Valle Novato Ski Resort in Chile, up in the Andes. I had the incredible fortune of skiing here some in July this year, where we definitely had a significant problem with fluid overload with a giant blizzard on the first day we were there, but it led to some incredible skiing afterwards. So I hope I haven't put you to sleep so far. Thank you very much for your time and attention. Of course, you can reach out to me by email or Twitter if you have any questions or comments. I'd be happy to discuss those with you and share further information. Thank you very much. Have a wonderful day.

fluid overload

organ failure

kidney failure

patient outcomes

venous pressure

phases of fluid management