SCCM Resource Library-Mechanical 2.0: When You Have More Machines Than People in the Room

Video Transcription

Good morning, everyone. Thank you for joining us for Mechanical 2.0, a session that talks about how you manage  those complex patients that have more than one machine in the room, sometimes more than two. My name is Raj Basu, and on behalf of our speakers today, we're delighted you're going  to join us. Today's session, we're excited about specifically because it's very apropos for the pandemic  we're getting through and how we manage an emerging population that's more and more complex.  Our speakers today are from the expert list of people from around the country who do this work. Dr. Raghavan Murugan is an expert in ventilation dialysis.  He'll be talking to us about the interplay of these machines and how you manage a critically ill patient. He's from the University of Pittsburgh.  Joining him is Dr. Dan Nguyen, who will talk about ECMO and dialysis, an all too common scenario in both adults and in pediatric care.  And then I'll end the session putting everything in the lens of COVID. We don't have a ton of data, but we have enough data to tell us this is an important consideration  in very sick patients. Today's objectives are pretty simple. We're going to talk about how you manage these circuits together to manage acidemia and fluid overload.  We'll talk about the specific considerations you need to have in mind when you're connecting  multiple machines, whether it's the tubing, the flow valves, the monitoring, et cetera. We'll talk about the practical management of how you deal with a patient with more than  one organ failure, two organ failures, specific to the kidney and the lungs. And then again, we'll talk about the aspects of COVID as it relates to this, specifically  addressing this idea of shortages and allocation of therapy. All our speakers have disclosed their relevant relationships and we are ready to go.  So Dr. Murugan, I'm going to hand the stage off to you. Thank you again for joining us. Good morning. My name is Ragi Murugan.  I'm one of the intensivists at University of Pittsburgh with a special interest in mechanical ventilation as well as renal replacement therapy.  Today I'm going to talking to you about the interactions that are going to be occurring between ventilator as well as the dialysis machine during treatment of patients, especially  now given the COVID situation, most of my talk is going to be about managing COVID patients. So with that in terms of disclosures, I have several disclosures to make, but there is  no conflict of interest with anything that I'm going to be presenting today. So the objectives of this talk is to start with a COVID case vignette and then provide  an overview of ventilator management in patients with COVID followed by management of the dialysis  and then end with a conclusion. So this is a 67 year old male patient with type 2 diabetes, hypertension, obesity, who  lived in a group home, was admitted recently to ICU, and he was COVID positive and was admitted with shortness of breath and fever.  On admission he was tachycardic, febrile, and hypoxic requiring four liters of oxygen to keep oxygen saturation of about 91%.  He was promptly treated with the antiviral agent remdesivir and as well as dexamethasone. However, his hypoxemia continued to worsen over the next four days and then he ended  up on high flow nasal cannula at 100%, at 70%, 70 liters a minute. And subsequently this worsened to the point that he was intubated and was placed on a mechanical ventilator.  And so this patient had severe ARDS, secondary to COVID pneumonia, as you can see bilateral infiltrates on the chest x-ray.  And then on the same day he was intubated, you can see a profound rise in serum creatinine peaking at six milligrams per deciliter.  And at this point in time, he also developed oliguria and oliguric acute kidney injury that was not responsive to a diuretic trial that we gave to this patient.  This patient also developed significant respiratory as well as metabolic acidosis. As you can see that the PCO2 profoundly increased up to 70, and then the pH dipped quite to  significant levels of 7.1, causing this patient to have hemodynamic instability secondary to severe acidosis.  At this point, the patient was on 100% oxygen and he was prone ventilated. He was on a PEEP of 20 and he was having cardiac arrhythmias and he was also requiring vasopressors.  So by this time, six days into the hospital admission, he's also become positive primarily due to several other fluids because he was not eating, he was placed on some IV fluids  and then between the medications, he was getting on a positive fluid balance, which continued to increase after he developed oliguric acute kidney injury.  So in terms of the ventilator management, these are the guidelines that was put by the Society of Critical Care Medicine back in March and in spring.  And you can see here that in terms of managing patients with mechanical ventilation, there is no sort of a clear timing that's being suggested.  And it's an individual choice that we would have to make. One of the key things is to sort of make sure that we intubate these patients at an appropriate  time point so that these patients do not tire out. So at least, you know, in our institution that this is determined by the bed situation of the patients.  We typically tend to have, if for patients who have slowly worsening hypoxemia, we tend to have them on high flow nascent candida and followed by a brief trial of noninvasive  ventilation. If they fail those, then we certainly place them on the ventilator. On the other hand, a patient who is coming in with rapid progression of respiratory failure,  rapidly increasing oxygen levels, we intubate those patients, you know, promptly and place them on the ventilator.  So I think that certainly high flow nasal candida as well as noninvasive ventilation could be used, a trial could be used, but these patients would have to be monitored  very, very closely. So for mild ARDS, the current recommendations are used to 4 to 8 ml per kilogram tidal volume  in these patients to prevent barotrauma, to target oxygen saturation between 92% to 96%, have conservative fluid management, but also to use higher PEEP and neuromuscular blockers  in patients who we cannot oxygenate them as well as prone ventilation. One thing that we found in our institution is that many patients with COVID pneumonia  have very few infiltrates on the chest x-ray, but develop profound hypoxemia. And this is primarily due to intravascular thrombosis that is occurring in these patients.  So we find that if there is not a lot of infiltrates on the chest x-ray and if the hypoxemia is  profound, then obviously you would need to screen these patients or do a CT scan looking for a PE.  Another thing to remind is that the PEEP and prone ventilation does not work very well in these patients as well.  So I think that that's an important thing to think about when we are using PEEP. If they respond to PEEP and prone ventilation, then that certainly is encouraging.  But a lot of patients who don't have infiltrates on their x-ray don't have a good response to PEEP or prone ventilation.  Another point about ventilating these patients is that barotrauma is extremely common in COVID pneumonia.  And these are two different patients we had very recently who developed significant barotrauma  within a few hours of intubating the patient, despite having normal peak airway pressures as well as the plateau pressure.  You can see here, significant anterior pneumoideus, dynamic extensive subcutaneous emphysema. And one of these patients also had a trace pneumothorax as well.  So barotrauma is very common.  It's been reported up to 30% occurs despite normal plateau pressures. All of these patients, we've had four patients so far with barotrauma, all had plateau pressures less than 30.  And it's not clear what causes them, but there have been suggestions that steroids, non-invasive ventilation, especially high tidal volumes that these patients get while they are on  PIPAP and CPAP predisposes to barotrauma in these patients. But if they do develop barotrauma, the mortality is greater than 60%.  So in terms of acute kidney injury, more than 20 to 50% of the patients with COVID developed  AKI, and you can read the mechanisms that lead to AKI in these patients, but AKI requiring RRT has a mortality of more than 60% in these patients.  So the management of AKI is pretty much standard with guidelines, with the KDGO guidelines that have been recently put forward.  But the COVID modifications essentially are, again, very similar to the regular management  of acute kidney injury where you consider kidney replacement therapy for managing metabolic demands and fluids.  The modality must be based on the need of the patient, expertise available, the equipment  and staff, prolonged methods of kidney replacement therapy such as CRRT, PIRRT are useful in patients who are hemodynamically unstable with market fluid overload.  The prescription dose, the effluent prescription for patients on CRRT must be 20 to 25 ml per kilogram per hour, and preferably vascular access in the right age is preferred.  One thing to remember is that when you place these patients on prone, especially if they are obese and hypercoagulable, there tend to be a lot of catheter malfunction.  So it's important to make sure that the catheter is functioning adequately. So in terms of timing, this is the recent publication from the start trial showing no  difference in mortality between patients in whom kidney replacement therapy was initiated early enough as within 12 hours of meeting eligibility criteria compared to patients  in whom it was delayed up to 72 hours and initiated late. In patients in whom the kidney replacement therapy was initiated early, there was substantial  increased risk of hypotension, cardiac arrhythmias, as well as hyperphosphatemia, suggesting that early initiation of renal replacement therapy is not beneficial.  And the current recommendations are, is to initiate only when you absolutely have to. So in this patient that I showed you, this patient had severe metabolic acidosis, severe  fluid overload, and this was really crashing. And so we would have to initiate CRRT at that point.  This was about three or four days after the onset of stage three acute kidney injury. When you do initiate RRT in these patients, it's important to remember that hemodynamic  instability is extremely common in patients with COVID. And the reason for this is multifactorial.  Many of the patients with COVID have cardiac dysfunction in terms of myocardial contractility. They also tend to be vasodilated, but also these patients are receiving high P, which  increases the intrathoracic pressure and decreases the venous return to the heart. And all of these things compound hypotension in these patients.  And so it's important to make sure that we don't hurt patients by rapidly fluid removal, removing fluid.  So it is a sort of a trade-off between trying to remove fluid, manage hemodynamics, and also the optimal P on these patients to make sure that these patients are receiving, being  oxygenated adequately. Fluid overload is a very common problem. Again, this is some of the data from our own work prior to COVID, where you can see fluid  overload is associated with substantially increased risk of death that persists up to one year after ICU admission.  And observational studies show that fluid removal as little as minus 500 cc in the 24 hours is associated with improved outcome. This is data from the renal trial.  So when you're thinking about removing fluid in these patients, it's important to remember  when using CRRT is to integrate both the machine fluid balance as well as the patient fluid balance.  So when you're pulling fluid at maybe 100 mLs an hour using the CRRT machine, but if the patient is getting 200 cc an hour of fluid in the same hour with the medications, then  even though you're removing 100 mLs an hour, the patient is becoming 100 mLs positive. So if the goal is to make the patient 100 mLs negative per hour, then you would have  to remove fluid at 300 cc an hour in order to get 100 mLs negative in the patient. So that's a key thing to remember where you have to integrate the patient fluid balance  as the machine fluid balance on a periodic basis. I know one hour is really, really cumbersome, but at least some four hours, six hours, it  has to be integrated so that we tend to achieve the negative fluid balance that we want. Now recently there are studies that actually show that rate of ultrafiltration is associated  with mortality. This is a study that we conducted a couple of years ago showing slower ultrafiltration  intensity or rate over a 24-hour period is associated with higher mortality compared with a higher ultra ultrafiltration rate in patients, particularly with fluid overload,  where we compared greater than 25 mLs per kilogram per day compared with less than 20 mLs per kilogram per day, and less than 20 mLs per kilogram per day is less than one  mL per kilogram per hour, and in these patients who are fluid overloaded, slower fluid removal was associated with mortality compared to faster rates of fluid removal.  Now we then looked at the same question in the renal cohort of patients who only received  CRRT, and what we found was that even rapid fluid removal was associated with higher mortality,  and you can see here, these are the adjusted predicted mortality for increase in ultrafiltration  rate, where the higher the ultrafiltration rate, the greater the mortality, and the lowest  risk is somewhere in the middle, where moderate, you know, slow ultrafiltration rate between one to 1.7 mL per kilogram per hour was associated with lowest mortality risk.  And the study also indicated that if you pull fluid pretty rapidly, there is a higher risk  of cardiac arrhythmias, and also cardiac arrhythmias requiring treatment, suggesting that, you  know, faster rates of fluid removal are associated with adverse, adverse effects in patients.  Again, this is not particular to COVID, but this is what the data from the renal failure suggested.  So the emerging conceptual model from our work is that there is clearly a net ultrafiltration  rate outcome relationship, where in a fluid overloaded patient, if you don't remove fluid, there's a higher risk of TAP, but if you also rapidly remove fluid using CRRT, if you're  using more than 1.75, there's also a higher risk of TAP. And again, these needs to be confirmed in randomized clinical trials, but for now, the  patients who have the good outcomes are in which we do it really, really slowly and gentle so that we don't cause cardiovascular instability in these patients.  So I'd like to conclude by stating that optimal management of ventilator and kidney replacement therapy is key to improving outcomes, and, you know, whether they have COVID or not,  and I know majority of the patients are COVID currently, but the same principles would apply. Hemodynamic instability is very common in COVID patients for a variety of reasons who  are on the ventilator as well as on dialysis. Both faster and slower rates have a bidirectional relationship with mortality.  So both faster and slower might be associated with increased risk of death for varying reasons.  So if a slower rate of ultrafiltration increases exposure to fluid overload, that might increase  the risk of TAP, whereas doing it faster might cause hypotension and ischemic organ injury that might also increase the risk of TAP.  So it's important to pay attention to this, and we really need randomized clinical trials to examine the ultrafiltration rate or how fluid should be removed in these patients.  So with that, I'd like to acknowledge all the work done by some of the collaborators, excellent collaborators that I work on a daily basis, and also other collaborators from around  the world who have provided input on the work. And with this, I'd be happy to take any questions.  I'd like to thank the organizers of the SCCM for providing me this opportunity to give this talk. Thank you. Hello, everyone. My name is Dan Nagoyanu.  I'm from the Hospital of the University of Pennsylvania. I'm very grateful for the opportunity to speak today, and I'll be talking about the interaction between ECMO and dialysis.  And let's just dive right in. So really, one of my areas of interest is in the ways that the therapies that nephrologists  control interact with the therapies that intensivists control. And today, we're going to be talking about the interaction between renal replacement  therapy, RRT, also known as kidney replacement therapy. We're supposed to be saying kidney replacement therapy these days, but I can't help but say RRT most of the time.  And extracorporeal membrane oxygenation, aka ECMO, which also has its most correct terminology as extracorporeal life support, ECLS.  But we're going to say ECMO for the rest of today's talk. And I'm going to first start with the briefest review of kidney replacement therapy.  One of the things that's intimidating about talking at SCCM is there's such a wide variety of people from a wide variety of backgrounds.  So the way nephrologists think about kidney replacement therapy is we break it up into two things. We break it up into solute removal and volume removal.  And this is really the way we talk on our rounds, etc. That's why we don't get invited to too many parties. And hemodialysis is a mode of solute removal.  So anything that's in higher concentration on the blood side is basically going to cross into the dialysis and going to get washed away.  And stuff that we want to give the patient, like bicarbonate, let's say, is going to be in higher concentration on the dialysate side and is going to be given to the patient's  body through the dialysis machine. Now, if we want to remove fluid, we need really a separate modality that we call ultrafiltration,  where we apply negative pressure to the dialysate side and literally suck fluid out of the patient's body.  Now, most, quote, unquote, dialysis treatments are really two treatments in one, hemodialysis for solute removal and ultrafiltration for volume removal. Now let's turn to ECMO.  And if dialysis is kidney replacement, let's use some of that same terminology to explore ECMO. And there's really two flavors of ECMO, VV, ECMO, and VA.  And we'll start off with VV. And VV is really a form of lung replacement therapy. Nobody says lung replacement, unlike in the kidney side, where we do say kidney replacement  therapy. Nobody says lung replacement therapy, but that's what it is. VA ECMO is a form of heart replacement therapy.  You don't need to be on VA ECMO if your heart is working just fine. And some patients on VA ECMO are going to have normal lungs, are going to have lungs  that have no problem whatsoever. But some may have failing lungs that also need the component of oxygenating and ventilating the blood in order to survive.  That said, anybody that you put on VA ECMO, you better oxygenate that blood. If you don't, then you're going to pump a bunch of deoxygenated blood to the arterial  side and you'll kill even a healthy person. So let's start looking at VV ECMO and ask, what are the goals of lung replacement?  Really the goals of lung replacement are ventilation, aka CO2 excretion, and oxygenation. And we assess that by the patient's PO2 and O2 sat.  And let's explain VV ECMO using some principles from hemodialysis. So this is my slide for HD, and I can just change it to an ECMO slide just by replacing  the dialysate with sweep gas. So what is sweep gas? Where does it get its name? It just gets its name from the fact that you're literally sweeping away CO2, you're sweeping  gas over the blood. And this is a typical oxygenator, a Quadrox oxygenator. The controls on an ECMO machine are going to be sweep gas flow, sweep gas FiO2, and the blood flow.  And what's really key, as we'll get to in a couple of slides, is the ratio of blood passing through the ECMO circuit to blood not passing through the ECMO circuit.  Just to see what these controls look like in real life, different ECMO circuits will look very different, but they're actually pretty unassuming and unimpressive.  You kind of have to look around the circuit to find them in many setups. So this is a typical flow meter where you set and you can see what your sweep gas flow is.  Next to it is going to be something that controls the FiO2 of that gas. And what about blood flow?  And blood flow has got a controller on it that typically is going to control the RPMs. So you control the RPMs here, and that is going to determine, depending on the resistance  of your circuit, the flow in liters per minute. Because that's going to control a rotor in most pumps, most pumps work this way, that  is going to impart kinetic energy to the blood in the circuit. So let's look at ventilation versus oxygenation in ECMO.  So ventilation, also known as CO2 excretion, is basically going to be controlled by sweep  gas flow, and oxygenation is going to be controlled by the FiO2 of the sweep gas and the blood flow. And we're going to look at these each briefly separately.  So let's look at sweep gas flow first and foremost. CO2 excretion on ECMO should very rarely be an issue. Why?  Because you can almost always increase CO2 excretion by just going up on the sweep gas. So you should almost never, almost never, have a high PCO2 or respiratory acidosis when  on ECMO. And this works even if only a small part of the patient's blood flow is going through the ECMO circuit.  Because the ECMO can clear enough CO2, even from a small proportion of the patient's blood flow, to make up for the blood not passing through the circuit.  And so, of course, we use this to rescue hypercarbic patients who are in danger of volutroma and barotrauma in the setting of a mechanical ventilation.  So really, VVECMO in many ways is there to rescue patients from the negative effects of mechanical ventilation when patients have got very poor compliance.  Now oxygenation on VVECMO is different. The FiO2 of the sweep gas is the major determinant of O2 sat in the circuit, okay?  So the FiO2 of the gas you're using is usually 90 to 100 percent, and the saturation of the  blood exiting the circuit is usually 100 percent. So it's 100 percent saturated blood. However, that completely saturated blood can't be saturated more than 100 percent, and you're  going to then mix that blood with deoxygenated blood if the patient's lungs are not working well in the setting of VVECMO, there'd be no reason to put them on the therapy if their  lungs were working fine. And so often these patients have got an oxygenation problem. If they have a pure ventilation problem, this won't be an issue.  But they often have an oxygenation problem, and that 100 percent saturated blood is going to be mixed with deoxygenated blood.  So the systemic saturation is largely determined by the ratio of oxygenated blood passing through the ECMO circuit divided by the blood not passing through the ECMO circuit.  And this is going to become significant later on when we talk about the impact of our CRRT machine, particularly volume removal with our CRRT machine.  So let's just point out that this issue of ratio of QBECMO to total QB is philosophically similar to the concept of surface area for gas exchange in a patient's lungs.  So here you have normal lungs, normal healthy lungs, and you can imagine that here all the blood going through this patient's lungs is going to be in contact with gas, is going  to be in contact with your sort of natural gas exchange or your alveoli. So here this ratio is one, the total blood going through your natural ECMO, which is  your lungs, to your total blood flow rate is basically the same. However, if you have an injured lung or if you have some pulmonary edema, some of these  alveoli here are partially filled with fluid, some are collapsed, and so some of your blood  flow is going through, in essence, a gas exchanger, aka healthy alveoli, and some is never contacting gas.  So this ratio is really similar in concept to the surface area available for gas exchange in somebody's lungs.  So what's the major cause of this ratio being low in ECMO? And the major cause is basically  poor function of the inflow cannula. How does this manifest itself clinically? It often presents  something called chatter in the cannula. You literally put your hands on the cannula and you  can feel it kind of quivering. You may see that high RPMs are needed to achieve the desired blood  flow. And under these circumstances, you want to be cautious about aggressively removing fluid. So if your inflow cannula is performing poorly, the last thing you want is to cause  intravascular volume depletion acutely with your CRT machine. And so you may want to intervene in  some way, often changing the anatomy of your ECMO circuit or perhaps giving a unit of RBCs.  Whatever it may be, you want to talk to your intensivist or people in the audience or listen  to the talk or intensivists. You want to think through something you can do to improve your inflow cannula performance before you start aggressively removing fluid. And this is just  a figure from, this is just a picture from an ECMO course I took where they had a synthetic candaver, a syndaver. This is in real time the IVC of the synthetic candaver as they were  cranking up the RPMs on the ECMO circuit. This IVC used to look like a big plump circle and it's starting to collapse. And about a second after I took the picture, it was completely  sucked down around this inflow cannula. So what's the role of volume removal via CRT in VV ECMO?  Basically, people are going to be interested in decongesting the patient. So their lungs are going to start working again so they can get off VV ECMO. That's often the reason why  our intensivist colleagues are calling us to help out in these patients. But I think it's  important to point out that you might want to be cautious about rushing this and that it's very  rarely the idea that you're going to get a person who just got put on VV ECMO off the next day.  It may take some time. It's probably going to take a couple of days. Otherwise, you may ask  the question, did they need VV ECMO in the first place? You want to make sure that the pulmonary  edema that they have, that everybody thinks that it really is cardiogenic, that you are going to  be able to mobilize it with your CRT circuit. And most of all, you want to make sure that the flows  through your ECMO circuit are stable as you are pulling fluid off of the patient. What about  solute removal? So I said at the very beginning of the talk that nephrologists separate volume  and solute removal. What's the role of solute removal? Really, it's pretty similar to any  critically ill patient. The one caveat I will say is that CRT has an amazing ability to remove  potassium, but even mild degrees of hyperkalemia, I tell my fellows, pay attention to them, not  because the potassium itself is dangerous, but it's a sign potentially of something going wrong.  In the setting of ECMO, it may be evidence of hemolysis, which is something that can definitely  happen within the ECMO circuit. If you're needing gradually greater support with your CRT machine  to remove potassium, then you want to ask yourself, where's the potassium coming from? Is it due to  hemolysis in the setting of a surgical patient? Does the patient have a slow bleed? Are they resorbing a hematoma? Where's the potassium coming from? Even though you're able to keep  the potassium easily in the safe range using your dialysis machine. So what about weaning the ECMO?  How do you get off it? This is relatively easy. You just turn down the FiO2, and then you can turn  off the sweep gas, and then you can see what the patient's doing with their his or her own lungs.  However, what about the lung replacement component of VA ECMO? Remember we said earlier on the talk,  if you're on a VA ECMO, you got to oxygenate that blood. Otherwise, you're pumping  venous blood to the arterial side, and if you don't oxygenate it, you're going to kill a patient.  So when we as nephrologists are rounding on patients, you know, we see a patient on VA ECMO,  and we want to ask ourselves, does this patient clinically need lung replacement, quote-unquote,  or not? If they do, then weaning of that lung replacement component can be very challenging.  Like I just said, you can't just turn off the sweep gas, and so those of us who are consulting  on these patients may notice that we came in one day, they were on VA ECMO. You come in the next  day, they are on VV. Why? Why did that happen? Clinically, it's a patient where the heart  replacement component no longer is needed, but clinically, the team is worried that the lung  replacement component of the therapy may still be necessary. So once you switch them to VV, now you can wean them in the same way as you typically wean VV ECMO. I just want to point  out briefly that recirculation, not an issue on VA ECMO circuits, so never invoke recirculation as  a cause of poor oxygenation. So for VA ECMO, I'm going to very briefly discuss VA ECMO now.  The anatomy of a VA ECMO circuit can vary even more widely than the anatomy of a VV ECMO circuit,  and so what do I mean by that? Basically, where are your inflow cannula? Where are your outflow  cannula? And we definitely don't have time to discuss all the different options. I'll just show  you the most basic bifemoral cannulation, where you have an inflow cannula in the IVC. You have a  femoral cannulation that goes up into the aorta from the femoral artery,  and if your heart is not pumping at all, if the myocardium is basically not working, then you'll  have something like this on your monitor. You can see that there is actually electrical activity in  this heart, but the arterial line is completely flat, okay? So the heart is not doing anything.  All of your blood flow is coming from your ECMO circuit. All of your flow is through the VA ECMO.  So oxygenation should not be an issue. 100% of your blood flow is going through the ECMO circuit.  As long as your oxygenator is not clotting, oxygenation is not an issue, but decreased flow  is a big issue. All of the patient's perfusion is going through this circuit, so you definitely  don't want inflow cannula issues. So if the flows drop, or there's chatter, or RPMs on the blood pump  need to be increased, then you want to be cautious, if you're the one in charge of the CRT machine,  about removing fluid in that setting, and often the change that needs to happen is that  if you're still being aggressive in such a patient, is some change to the anatomy of the ECMO circuit,  so perhaps additional inflow cannula. If the flows are stable, however, if you have good flows,  and the organs seem well perfused on the therapy, well, then you can be a little more liberal with  your volume removal, because you have all of the flow going through the ECMO circuit, you should  get some signal, you should see the flows drop, or you should see chatter, you have your seat,  you're watching the blood get pumped in front of you through this ECMO circuit, so you can probably  be a little more aggressive. You should have some warning if it's going to drop. What if the heart is  working? What if the heart actually is pumping, but not enough to get by on VA ECMO? So you have a  pulsatile arterial line. In that case, oxygenation may be an issue, because the heart is working,  but the lungs, if the heart is working, but the lungs are not, then this can result in something  called North-South syndrome. So here's your outflow cannula pumping oxygenated blood up this way,  and then here is your left ventricle pumping deoxygenated blood from the lungs in the opposite  direction, and you'll have an interface basically at some point where deoxygenated blood is meeting  oxygenated blood, and if that interface is past any of the artery that's feeding the brain, you're going to be in big trouble. And so if a patient's on renal replacement therapy,  under such circumstances, an intensivist may transiently ask for help with fluid removal  in an attempt to get the lungs working better or potentially actually decrease blood return to the  right side of the heart, hoping that you'll have decreased non-ECMO blood flow, but really,  it's very unlikely that you're going to be able to, as a nephrologist, help with these problems  quickly enough, and almost always the anatomy of the VA ECMO circuit needs to be changed if  you're facing something like this, such as adding an LV vent. So that said, what is the most frequent  request I get regarding a patient on VA ECMO? And it's my colleagues invoking the RV. They  basically want the RV decompressed. The physiology and options to decompress the RV are myriad. This  is a great review article looking at that, but the bottom line is when somebody's on VA ECMO,  you often, it's worth asking your critical care colleague, you know, what's going on with the RV?  Do they need help decompressing it? And that's often why they're turning to us to remove volume,  in addition to trying to get their lungs back on track if their lungs are not working.  So how do you integrate CRT into an ECMO circuit? Should the CRT be part of the ECMO circuit versus  a separate via its own dialysis catheter? This is typically a decision regarding your ECMO program  rather than something you're individualizing patient to patient. It's a talk in itself, and we don't have time to cover it. This is a great review article on the topic,  Marlies Osterman and Michael Connor. So I just take a look at that. So in summary,  ECMO therapy produces vast changes to normal circulatory physiology and renal replacement  amongst other things can remove potassium, but always ask where that potassium is coming from  and can remove fluid, but it can compromise systemic perfusion and flow through the ECMO  circuit depending on the specifics of the anatomy of your circuit. So you just want to pay attention  to that if you're in charge of the CRT. And the key questions that consultants who are in charge  of the CRT should be asking, what's the anatomy of the circuit? Is it VV? Is it VA? Where are the  inflow or outflow canyon? I often just ask the critical care RN. I ask him or her to tell me,  you know, where it is to get an idea of what the circuit looks like. Have there been problems  with flow through that circuit? Have there been chatter in the lines, decreased blood flow,  increased RPMs to maintain that blood flow? Are there any problems with oxygenation? And does  the team want us to help with those problems? Do they think that getting the fluid off is going to  fix it? And last but not least, if the patient is on VA ECMO, what's going on with the RV?  So thank you guys very much. It's been a privilege and have a wonderful 2021.  Hello everyone and thank you for this opportunity to wrap up this session on Mechanical 2.0. I  really appreciate Dr. Murugan and Dr. Negwani's great talks and I'm going to be talking about  COVID-19 and specifically, is the new really new? Just a couple minor disclosures to tell you about.  You know, as we've gone through the past 40 minutes, what we've realized and that you know  is that AKI is a signal that matters. The biggest studies over the last five years tell us very  clearly that a significant proportion of our critically ill patients, whether they're babies,  children, young adults or old adults, significant proportion of them have severe AKI and that AKI is associated with poor outcome. It's independent of why they have the AKI.  But as you know, we have a problem. COVID and AKI have gone together and what we have figured out  quite clearly is that we need to understand this in a real way. So what do we really know? A month  ago when I put this talk together, what we figured out at that time was there were about 250  publications already about COVID and AKI. A lot of them are from prominent nephrology journals,  Jason and CJason. And a lot of the information was emerging and changing and adapting so that  if you looked at the information from March, April, May, June, it was a lot different than  the stuff that came out in the fall. But what did we figure out? Well, if you look at the data that  we had, what we understood was that AKI affected a significant proportion of critically ill patients,  more so than the proportion that occurred in otherwise critically ill patients. So up to 50% almost, and this is a study from the New York collective groups about AKI in their  COVID-19 patients in the first several months. So they looked at 4,000 patients in AKI stage and during the first seven days and found a signal rate of 46%. And in this population,  a significant proportion received dialysis, 20%. And notice I say received, not required.  When we did look at the data that exists between before and after, kind of a comparison of COVID-19 to the other, what we were able to find out is that there actually is no  phenotypic difference between the type of people that were getting AKI in the setting of COVID  and the type of people that were getting AKI without COVID, but other significant illness.  The predilections for race, gender, age, chronic kidney disease, all persisted, but were not unique to COVID. We did find that their accumulative death rate was extremely high,  especially initially in the COVID pandemic. We performed a, and we're still performing, a global assessment of pediatric AKI in this population. What we found out is that in our  preliminary analysis of 105, which is now about 450, an AKI rate that was almost 50%  compared to our one in four rate from our global epidemiology aware study, severe AKI was about 25%. And we believed initially there was an association with  comorbidities, but we're not finding that on follow-up analysis. In fact, in children,  what we're finding is that there's a low rate of CRT utilization. MIS-C or the multi-inflammatory  state in children in the UK is called pediatric inflammatory multi-system syndrome. This is,  in some sense, COVID of children in terms of the diffuse inflammatory state. High rate of AKI,  high rate of poor kind of longer-term outcome. A lot of these kids did survive, but again,  even in this population, CRT was not common and the AKI was actually just a functional severe AKI  initial. So the summary that we had initially was in the context of SARS-CoV-2, severe AKI was common. AKI requiring or receiving renal replacement therapy is common. So it begs the  question of why? What's actually unique versus say ARDS or sepsis associated AKI? Are there  unique considerations? And furthermore, what do we do? Because one of the things about COVID and AKI  that we need to discuss is all the risks that are associated with exposure of personnel and people who are helping manage these patients. Fascinatingly, a number of groups have done  biopsy reports on patients with COVID-19 and looked for specific pathogenesis in the kidney.  And what they found is quite clear. There's actually no evidence of viral infection in the  kidney of COVID-19. Instead, what is found is collapsing glomeruli, membranous glomerulopathy,  and nephritis consistent with other forms of acute tubular necrosis or critical illness in the  setting of AKI. In fact, when they pooled data together looking at multicenter, what they found  is that there weren't any criteria that could establish a direct viral cause, but they also found some high incidence of thrombotic microangiopathy, which would fit with some of  the pathophysiology we know in the hypercoagulable state associated with COVID. So the mechanism of  injury driving AKI in the setting of SARS-CoV-2 is still a little unknown, but it is potential  that one of the driving forces is actually a coagulopathy within the kidney itself and not  necessarily viral associated infection in tubular necrosis. And this becomes important when we look  at histologic specimens, but also how we actually manage these patients. So this is another kind of  paper looking at a genetic variant of APOLI L1 that is common amongst African-American patients,  but again, not unique to COVID. When we put all these things together, what we find is that,  yes, there is association with the ACE2 receptor, something that everyone has talked about from the  beginning, but what we know is that the cardinal sign of AKI in COVID happens to be proteinuria and in the setting of dehydration and volume depletion, so much so that some of the  endothelial damage and thrombotic microangiopathy may be because of volume depletion in third  space. So taking a targeted approach to COVID-19, you know, review articles have already been  published on this, have focused on this idea of optimizing bioenergetics, looking at mitochondrial  restoration and function to help deliver ATP and oxygen substrate where it needs to go.  There's inflammatory dysregulation for sure, but some of that can be abated, which is why a lot of  the therapies initially were targeted and still about monoclonal antibodies to target inflammation  and optimize repair pathways. So the why summary really is that ATN might be frequent, but TMA seems to be common, but all of this may not be entirely unique to SARS-CoV-2.  So the question is, what about kidney support therapy, what we would call renal replacement  therapy? Is it unique? What are the important considerations we need to have as we continue to  be immersed in the pandemic, and what can we do to optimize the therapy with the resources we have?  So the stop COVID investigators, a multi-center approach in the U.S., has been the cardinal group  or the main group that has really looked at renal replacement therapy in the setting of COVID. And so from 70 hospitals, and this is one of their first papers, and it was put out  mid-summer, late summer, looking at the risk factors for AKI, who these patients are, what happens to them. And they looked at 3,000 some odd patients, and in those patients,  650 almost received AKI RRT, so one in five. One of the big take-home points was the idea that there wasn't really anything unique about them other than they were sick.  So the patients who were the sickest on vasoactive support, invasive ventilation, were the ones who ended up getting renal replacement therapy. One in five patients  that they studied received renal replacement, but two-thirds of these patients died.  When looking at univariate and then multivariate odds, what they found was that if you look at  all patients, age, chronic kidney disease, being sicker was relevant to the utilization of renal  replacement therapy. But if you just look at the renal replacement therapy patients only,  the two things that really stood out were coming from a small ICU and being old in terms of what  happened to them in terms of death, not actually chronic kidney disease. And if you look at the  percentage prevalence of those patients who have 28-day mortality and a pre-hospital creatinine  available, it's almost like an even distribution of all the AKI stages in RRT. So the learning from  all this was the initiation strategy for renal replacement therapy was very rapid at first. These patients were coming in with high creatinines, oliguric, and so the reflexive  approach almost was taken in a sense of let's put them on CRT. There was less attention paid  to hydration. As the data has emerged, what we found is that the heavy CRT utilization in the  early stages was replaced by more of a judicious approach to their volume status. One of the most  important aspects that came out of the initial studies was the anticoagulation strategy and  filter life, because as many of you know, all these filters were clotting off. And so many  different approaches were tried, heparin, citrate, BiVal, a combination of those things. And so there  is data available for anticoagulation. And I'll kind of get to what the signal is at the end of  this. So the main thing I think to take home from this is that anticoagulation quickly became  relevant to how much was available and who was providing the therapy. Amidst all of this, what  people started to understand was they didn't have enough therapy. Not only was it in high use, it  was one of those that people were kind of jumping the gun to renal replacement therapy and were  running out of fluid bags, of filters, of machines. So the old became new. The use of peritoneal  dialysis. So there are several case reports, several reports from single centers about how surgeons were retrained to put in catheters, how nurses were retrained to use cyclers,  the benefit of dwell volumes, how to maintain ultrafiltration using PD. The biggest thing to  come from this is that bottom right box, the need for a multidisciplinary team for nephrologists,  intensivists, surgeons, nurses, and administrators to work together to figure out how to balance the  patient who can get PD from that patient who can get hemodialysis or CRRT. Intermittent renal replacement therapy, or SLED for AKI, in the sense of splitting therapies, potentially  splitting CRRT circuits, using 12 hours of this, 12 hours of this, necessity was the mother of  invention. So we, I think many of you who kind of were in the throes of this, had to deal with this  in real time, being able to allocate resources, but this is a real thing. So we are in this  situation where we think we provide this therapy continuously for 24 hours a day, and the real  situation is we have to think of other options. So PERT became a very unique option for providing  kidney replacement therapy. The European groups started to use more and more blood purification  filters, so heme absorption, and so this report came out of some of the Italian group and Saudi  Arabian groups looking at the improvement in chest x-ray findings with cytokine removal, and what you can see is using heme absorption filters definitely resulted in a reduction in  cytokines, and whether it was time or just the improvement in ARDS improved the patient overall.  The ADQI workgroup this year, one of the last ones, was dedicated to COVID, and what they did was summarize the available data, summarize what's old, what's new, what's old is new,  what's new is not actually new, it's actually old, and came out with potential management  strategies for COVID-19. I'm not going to list all of these. These are not necessarily unique  or novel. They're part of the CAID Ego Management Bundle. They did mention experimental strategies, such as antiviral therapy, immunomodulatory therapy many people are using, again,  immunomodulation, cytokine production or blockade, the use of systemic anticoagulation,  very common now as kind of basic therapy for COVID, but in the sense of using it for RRT, what was out there? The table that I find that I would recommend to you is Table 3 in this  publication, where they talk about renal replacement therapy management, and the first  one I think is the most important, which is the indication, and the indication changed from this,  okay, earlier might be better, to this consider a judicious and safe use of IV bicarbonate,  potassium binding resins, hydration therapy, modality choice can be varied, you can use PERT,  you can use PD, all these things that potentially were not even really considered maybe in many  advanced places prior to the COVID pandemic. So I think one of those things that necessity is the mother of invention drove these recommendations in a sense. So the COVID  and kidney support therapy, nothing is really unique to COVID. Age, coexistence, CKD, hypertension,  all those things are associated with more AKI and renal replacement therapy in depth, but that's  not unique to COVID. The initial recommendations is we are now very much habituated to the renal  replacement therapy trials, whether it be a KIKI or a LANE or the START-AKI trial, looking at earlier  versus delayed. And what we found then may be really appropriate now, and the idea of take the  time to have adequate restoration of volume status, observation of human dynamics, and use mitigated  steps to control acidosis, to control hyperkalemia. Resource and supply, so necessity is the mother of  invention. You remember PD and PERT, but the real rub of the road is who will staff these? It's time  to train. So there have been a number of articles talking about the shortages of KRT for COVID-19,  a lot of commentaries talking about we need new strategies, and perhaps the most important thing  is really to think about what the surge supply should be. You should really be thinking about  how do you coordinate a team? How do you coordinate the nurses? It's one thing to coordinate the  machines in the room, and my colleagues in earlier in this session have really done a tremendous job  explaining those things. But what COVID-19 has added to the situation is the immense pressure  to coordinate care, to coordinate care to be safe for the provider, and that involves the staff,  that involves the nurses, that involves the people who clean the room, not just the MDs. And that  oftentimes means you back away from a therapy that we're so used to using now, such as CRRT,  and you use PD and you use PERT. These things do matter. Dr. Walden-Bagshaw, the PIs for the  START-AKI trial, wrote a really nice learning from the first wave, and this is what I will end  with. We have a high prevalence of AKI in these patients, but the details of kidney support therapy  need to be considered. Early may not be better, mirroring our trials. We should really think  about our strategies. Do not have a reflexive approach. I think that the importance of fluid  and hydration, I think something pediatricians really know about firsthand, has become more and  more of a consideration in the adult management of oliguric AKI. Anti-coagulation, yes. How you  do it, try to be consistent in your place. Resources is about being thoughtful, and then the ethics. Should these be provided at all? The allocation of resources, something that  we don't want to have to admit that we have to think about, but we really do need to admit that  we have to think about. And unless we think about it going into the future, as this pandemic goes on,  we're going to be in trouble. So it's really about thinking judicious, thoughtful approach  to CRRT in the setting of COVID. So we have a new summary of challenges. Consistent team-based  approach is needed. We have to be proactive, and complexity can be simplified. And with that,  I'll close our session and open up to the Q&A. Thank you very much for joining us today.  Thank you.

Video Summary

In this video, Dr. Raghavan Murugan, Dr. Dan Nguyen, and Dr. Raj Basu discuss the management of complex patients with multiple machines in the room, specifically focusing on the interaction between mechanical ventilation, dialysis, and ECMO (extracorporeal membrane oxygenation) in the treatment of COVID-19 patients. They highlight the high prevalence of acute kidney injury (AKI) in COVID-19 patients and the need for renal replacement therapy (RRT), such as dialysis or CRRT (continuous renal replacement therapy), in these cases. They discuss the potential mechanisms of AKI in COVID-19 and the challenges in managing fluid overload and acid-base imbalances in critically ill patients. The speakers also address the importance of a multidisciplinary team approach, optimal timing of RRT initiation, and the use of anticoagulation in COVID-19 patients. They emphasize the need for judicious and safe use of RRT based on individual patient characteristics and the availability of resources. Ultimately, the session provides insights into the management of COVID-19 patients with complex medical needs and the unique considerations for coordinating care in these cases.

Asset Subtitle

Procedures, 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

Discuss the interplay of mechanical ventilator and dialysis machine in the management of life-threatening acidemia and volume overload
Identify special considerations when using extracorporeal membrane oxygenation (ECMO) and dialysis together, including circuit setup options and common pitfalls
Review practical multidisciplinary approaches for comanagement of patients with respiratory and kidney failure requiring treatment with a combination of mechanical ventilation and dialysis and, in some cases, ECMO
Summarize unique considerations for renal replacement therapy in COVID-19 patients, including infection control, anticoagulation, and contingency strategies during admission surges

Ventilation and Dialysis: Managing the Interplay
Raghavan Murugan

Circuit Overload: What to Do With ECMO Plus Dialysis
Dan Negoianu

COVID Plus RRT: Complexity Compounded
Rajit Basu

Meta Tag

Content Type Presentation

Knowledge Area Procedures

Knowledge Level Intermediate

Knowledge Level Advanced

Learning Pathway Skills and Procedures

Membership Level Select

Tag Dialysis

Tag Ventilation

Tag Extracorporeal Membrane Oxygenation ECMO

Year 2021

Keywords

COVID-19 patients

acute kidney injury

renal replacement therapy

mechanical ventilation

ECMO

dialysis

fluid overload

acid-base imbalances

multidisciplinary team approach

	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.