Good afternoon, so for my second talk of the day, I will discuss renal replacement therapy in the ICU, or I should say kidney replacement therapy in the ICU, as this is the newer accepted terminology for RRT. So my name is Anita Vijay, and I'm a professor of medicine in the Division of Nephrology at Washington University in St. Louis, and the medical director for acute dialysis services at Barnes-Jewish Hospital in St. Louis. So these are my disclosures, and the ones relevant for this talk would be my consultancy for next stage. So the objective for this talk is to describe the various forms of renal replacement therapy and how the modalities are used. We'll talk about the advantages and disadvantages of the modalities, and we'll talk about timing of renal replacement therapy for initiation of RRT, sorry, timing of initiation of renal replacement therapy, as this is a very timely topic with a lot of recent data. So these are the points of my discussion, and I will also touch base on anticoagulation during CRRT and dosing of medications, and I will also discuss prolonged intermittent renal replacement therapy, which is a hybrid modality. So let's start out with timing. When I was, a few years ago, one of the fellows who graduated talked about starting dialysis as basically from Dr. Vijay and I learned to dialyze first and ask questions later, and from one of my colleagues, I learned to never start dialysis, and there's a huge controversy in what is the appropriate time to start dialysis in a patient in the ICU with acute kidney injury. So all of us knew the general criteria of hyperkalemia, metabolic acidosis, uremic manifestations, and severe fluid overload, but there was no one number that would trigger us to start dialysis. So Dr. Mehta wrote this review article as some of the things to consider before starting anybody on renal replacement therapy, and obviously we know about the severity of illness, the ability of the patient body to handle the extra metabolic requirements from a critical illness, but we also should consider other factors, right? Is this renal replacement therapy going to significantly prolong this patient's life or is it going to improve the quality of the patient's life? And we should also consider family wishes and also, you know, we shouldn't have to do this, but especially during the COVID pandemic, we had to triage patients because of lack of availability of machines and nursing. So these are other things to consider before we even start renal replacement therapy. So let's assume that the patient and the family, or we have discussed the issue with the patient and family or primarily the family since the patient is probably incapacitated and not able to give consent and we're going to start renal replacement therapy. When should we start? So as I said, there have been few trials recently from Europe and also in large international trials. So this is the first one that was published in 2016 in the New England Journal of Medicine. This was a multi-center study from France that looked at early initiation. So patients who reached stage three AKI was considered the early AKI group, and then patients who developed some indications to start renal replacement therapy, such as hyperkalemia and metabolic acidosis, they were the delayed strategy. And as you can see, the survival, 68 survival was no different between the two groups. This was followed by the, or this around the same time, another study was published called the Elaine study. This was a single center study from Germany that looked at primarily surgical patients and it started patients on CRRT as soon as they reached stage two AKI. So doubling of serum creatinine versus delayed. And as you can see, there appears to be a benefit when renal replacement therapy was started early as far as 90-day mortality was concerned. But the question remains, how many of these patients would have recovered renal function or would have survived even if they did not reach, even if they did not need dialysis? So if you look at these two studies, Akiki and Elaine, the RRT in early group of Akiki was within six hours after inclusion criteria, that's stage three AKI. And this is within eight hours of reaching stage two AKI. And here, the late group in the Elaine was actually the early group in Akiki study. And one thing that was noted in the Akiki study was that the patients who started dialysis early were more likely to have bloodstream infections and hypophosphatemia because if you waited long enough, approximately 50% of the patients didn't even need dialysis because they started recovering. So starting early can be potentially harmful. And here also, the early group are more likely to have recovered renal function and had shorter renal replacement therapy duration, but you wonder how many of these patients would probably never have needed dialysis anyway. So as I said before, the Akiki study, the 49% of the patients in the delayed group didn't even need to start renal replacement therapy, and there was a higher incidence of catheter-related infections in the early group. This was followed by the ideal ICU study, also from France, and this was only in patients with sepsis. And similar parameters to the Akiki study, and there was no difference in 90-day mortality between the two groups. So early initiation did not benefit patients with septic shock either. So this was followed by the START-AKI trial, which is a multinational study. So it had centers in Europe, Canada, US, Australia, and Asia as well, and this was the brainchild of Dr. Bagshaw and Dr. Wald, intensivist and nephrologist in Canada. And it was a huge study, almost 3,000 patients. And the accelerated group had renal replacement therapy starting within 12 hours of a patient having doubling of serum creatinine or serum creatinine of greater than 4 milligrams per deciliter or a urine output for the preceding 12 hours. And in the standard arm, the RRT was not started unless they were hyperkalemic, acidotic, either with a pH of less than 7.2 or bicarbonate less than 12, volume overloaded or evidence of respiratory compromise with a PaA2-FiO2 ratio less than 200, or they had persistent AKI for 72 hours after randomization. And as you can see, the mortality, the 90-day mortality was no difference between the standard and the accelerated group, and there was no difference in the subgroup analysis. There's no difference in recovery of renal function. The two groups were pretty much similar. If you look at the adverse events, any adverse event was higher in the accelerated strategy group. And this is primarily because of hypotension associated with renal replacement therapy, a tendency towards arrhythmias, and also hypophosphatemia. So this year, there was another publication from the same group that published the AKIKI trial, and this study basically took the patients in the AKIKI late group as the standard strategy, and then they delayed it even further. So they didn't start dialysis unless patients had a potassium greater than 6.0, greater than 5.5 despite medical treatment, pH less than 7.15, acute pulmonary edema, or a BUN greater than 140 milligrams per deciliter. And in this, when they waited that long, you can see there's a higher mortality in the delayed group with a hazard ratio of 1.65 for mortality at day 60. So there's no difference in RRT free days between the two groups. So definitely, you don't want to delay dialysis too long either. So what is the correct strategy for starting dialysis or renal replacement therapy in ICU? By the way, if you want to have a cool schematic of all five trials that I just showed you, you can go to the NEFJC website and look at this cool infographic that shows all these five papers that I just showed you. What is the appropriate criteria for starting renal replacement therapy? I would think the standard arm of the START-AKI trial would be my recommendation to consider starting dialysis. You know, hyperkalemia, potassium greater than 6, pH of less than 7.2 or bicarbonate less than 12, PA2 slash FiO2 of less than 200, plus volume overload, persistent AKI for 72 hours after randomization. I don't think we should just sit there and watch a person's renal function steadily deteriorate, but because they're non-oliguric, hold off starting dialysis even after 72 hours. But what could be other considerations? So some patients in the surgical ICUs, especially CT surgical ICU, they may benefit from early initiation, especially if they're oliguric and they've already received a lot of fluid in the OR. Medical points like personnel to place catheters and machines may guide, also may guide our decisions to start renal replacement therapy. And like I said, I'm eagerly waiting for the subgroup analysis from the START-AKI on their oliguric and aneuric patients, because these are the patients who usually get into trouble if we don't start them in a timely manner. So let's switch to modality of renal replacement therapy. So the modalities that are available are intermittent hemodialysis, continuous renal replacement therapy, or prolonged intermittent renal replacement therapy. And this has various other names. In most institutions, it's called sustained low-efficiency dialysis. There's also known by extended daily dialysis. And then there's various other names like accelerated veno-venous hemofiltration and a bunch of other names. So one modality that I'm not mentioning right now is peritoneal dialysis. So for adults in the United States and other developed countries, PD is not used widely at all, if at all, except when we came to the COVID-19 pandemic. So it's possible that PD might make a comeback. But right now, PD is not used widely in the adult population. It is definitely a modality used in the pediatric ICUs for AKI. But we should keep in mind that PD is a RRT that can be employed if we have a shortage of these other modalities. So when do you consider one over the other? So there have been numerous studies looking at intermittent hemodialysis versus CRRT. The most famous one is published in 2006 by Vincent Nou. And he took 360 patients, randomized them to CVHDF or intermittent hemodialysis. Now the intermittent hemodialysis was a relatively long session. It was almost five hours. Most IHD sessions are anywhere from three to four hours. And so that is one of the drawbacks of this study. And there was no difference in survival between the two groups. So when I looked at all the studies recently for a book chapter that I was writing, if you look at the randomized control studies and the prospective multicenter observational trials, you can see that these two were the observational trials. There is no difference in mortality. In fact, these two studies suggested higher mortality in the CRRT group. That's probably because of randomization problems and not because CRRT is associated with higher mortality, but definitely intermittent hemodialysis has not shown to be affecting mortality in these randomized control studies. So one question that I put in here is in what population is there best evidence for CRRT or intermittent hemodialysis? Is it those with intracranial hypertension, those are homodynamically unstable, overdose with low protein bound drug, hyperkalemia, or sepsis? And I'll give you a minute, I'll give you a second to think about the answer before I go to the next slide. So the answer is those with raised intracranial pressure or intracranial hypertension. So this is a very, very old study from Davenport that looked at patients who underwent continuous arteriovenous hemofiltration or hemodialysis. So instead of CVVHD, they were doing CAVHD at that time. This is, like I said, an older study. And look at the percentage change in intracranial pressure. With IHD, it started going up. With these continuous modalities, it was relatively stable. And the cerebral perfusion pressure dropped drastically with intermittent hemodialysis. And this has been shown in some other recent studies, not the exact data, but you can tell changes in blood flow during... You can look at the difference in changes in blood flow during intermittent hemodialysis and continuous renal replacement therapies and show a similar pattern. So just a word of caution with raised intracranial pressure, dialysis dicyclerium results from rapid removal of solutes. And so there's a lot of intracellular fluid shifts. And observational trials and case reports reported increases in intracranial pressure with intermittent hemodialysis. And similarly, CT scan showed increase in brain water content after hemodialysis, but not after CRRT. Just other things, when you compare CRRT versus IHD, CRRT offers more hemodynamic instability, better fluid balance, ability to give more nutrition as the clearance over a seven-day period is much higher with CRT than three times a week intermittent hemodialysis. Stable intracranial pressure, as I said, but if you need rapid removal of poison, say somebody comes in with lithium overdose, then you need intermittent hemodialysis. So, what about outcomes? Is there any other difference in outcomes between CRT and IHD besides mortality? So, this was a meta-analysis done looking at IHD versus CRT, sorry, looking at all the studies that I've compared IHD versus CRT, there is a suggestion that renal recovery seems to favor patients starting out with CRT. But this findings did not hold out to renal recovery in over a year. And other studies that I've looked at this specific question have not shown an improvement in renal recovery with CRT as the initial modality. And if anything, it's the intensity of renal replacement therapy, how frequently are you dialyzing them or how long are they on renal replacement therapy that probably determines outcomes rather than which modality that you are starting out with. So, the K-Degal recommendations for renal replacement therapy states that we should use continuous and intermittent renal replacement therapy as complementary therapies in acute kidney injury. We suggest using CRT rather than standard intermittent renal replacement therapy for hemodynamically unstable patients primarily because it is better for fluid removal and also less likely to cause further hypotension. And we suggest using CRT than intermittent for those patients with traumatic brain injury or raised intracranial pressure or generalized brain edema for the reasons I showed you earlier. Similarly, the ADKI group, which is Acute Disease Quality Initiative, which is an AKI network, also recommended similarly that patients with hemodynamic instability starts out with CRT. Same thing with acute brain injury, start out with CRT or PD if CRT is not available. Patients with acute life-threatening complications should start out with intermittent hemodialysis. So, what about convection or diffusion, right? We talk about CVVH, we talk about CVHD, and there's also the combination convection and diffusion. Does it really matter which mode of CRT you use? So, diffusion or CVHD is good for clearing small molecules, right? Molecules move from high concentration to low concentration and it's great for removing small molecules such as urea. Convection or CVVH, it's where water is sieved through the filter and it pulls molecules with it and this allows clearance of larger molecules. And obviously, this is where you would need replacement fluid. So, when you talk about molecular weights, we talk about these small molecules, potassium, phosphorus, sodium, urea, and this can be removed with diffusion and convection. But when you talk about cytokines and maybe other larger molecules, that might be better cleared with convection. Well, does it make a difference? So, this is the AMAKI study, which is a pilot study trying to look at this exact question. Does CVVH versus CVHD, does it make a difference? And it was a feasibility study and there was a tendency for the cardiovascular SOFA score to be better in patients who were receiving CVVH over CVHD, but this was not really significant and this study has not been conducted in a large fashion to look at this difference. So, my recommendation is use whatever mode of CRT is available at your institution. There's really no difference whether you use CVH, CVHD, or CVHDF. So, how do you dose renal replacement therapy? So, there have been several studies looking at more versus standard, I should say, dosing of renal replacement therapy primarily as CRT. So, the biggest ones are the ATN and the renal study. But prior to that, we had about, you know, a few smaller studies that looked at CV, sorry, 20 mLs per kg per hour versus either 35 or 45 mLs per kg per hour. And then we had the renal study by Belomo and the ATN study by Pilevsky, which looked at, again, 25 versus 40 and 20 versus 35 mLs per kg per hour. This is the ATN study. So, basically, patients were randomized to less intensive versus intensive group. And for the CRT arm, they had 20 mLs per kg per hour versus 35 mLs per kg per hour. And the primary outcome was death at 60 days. And there was no difference. So, 53.6 versus 51.5% mortality at 60 days. Similarly, the renal study primarily conducted in Australia, 1,500 patients, 25 versus 40, there was no difference in mortality at 90 days and no difference in renal recovery. So, the current dosing recommendations for CRT started about 25 mLs per kg per hour. That way, you will at least get 20 mLs per kg per hour, even if the machine clots for, you know, two, three hours. Use the premorbid body weight or their dry weight. I mean, don't use the fluid overloaded weight to calculate the dosing. Keep the CRT downtime to less than four hours a day. So, make sure we have a good quality control, quality initiative to make sure the CRT is started back very quickly. You can use higher flow rates if their patients have uncontrollable hyperkalemia, acidosis, or hyperphosphatemia. And you can decrease flow rates if patient is becoming very hypokalemic or hypophosphatemic. You know, if the patient was started at 35 mLs per kg per hour, you can make sure you bring it back down to about 20 mLs per kg per hour once they're under control. The ATN study also had a hemodialysis arm to it, right? So, we were not comparing hemodialysis to CRT, but we were comparing hemodialysis three times a week to six times a week. And we followed the single pool KTOV urea. It's a marker of urea removal, and it's a marker of how efficiently the dialysis is going. And so, a goal is to keep this number between 1.2 and 1.4 per treatment. And that's what exactly what was delivered in both arms. And there was no difference in outcome. So, the recommendation for intermittent hemodialysis is measure urea reduction ratio pre minus post BUN divided by pre BUN times 100. And the goal urea reduction ratio is about 70%. And this translates to a KTOV urea of 1.3. And we know that urea is probably not the best marker for assessing how efficiently we are dialyzing patients. But that's the only marker we have available right now. And for want of a better marker, we should continue to monitor this. And some machines actually have online urea monitoring built into it. And not every patient needs a four-hour dialysis. So, if you monitor urea reduction ratio, you might be able to cut time of some of these patients to three hours. So, to dosing of dialysis, I would say a hemodialysis three times a week with a single pool KTOV urea of about 1.2 to 1.3, and a CRT effluent flow rate of about 20 to 25 mLs per kg per hour. And above that, we're not sure if it's beneficial. Below that, there might be some survival. It might affect survival. So, somewhere around there is, I would say, the point of maximum benefit. So, what is a PERT, or Prolonged Intermittent Renal Replacement Therapy? So, it's essentially a hybrid technique that is neither intermittent hemodialysis or a 24-hour treatment. So, we can use it as a substitute for CRT, transition therapy from CRT to intermittent hemodialysis, or as a substitute for intermittent hemodialysis. So, this is a review that I did with my fellow a few years ago, and it shows the comparisons between intermittent hemodialysis, CRT, and PERT. And the blood flow on PERT is somewhere between these two. The dialysate flow is somewhere between these two. Duration, it's usually done for about 8 to 10 hours at our institution, but you can do it anywhere from 6 to 12 hours. Frequency, anywhere from 3 to 7 days a week. We do ours at night. It can be performed without anticoagulation, like an IHD treatment can. I recommend a catheter for doing PERT, since it's mostly not monitored. And it can be done in an ICU or step-down unit if the nurses are not overwhelmed. If they have one to two nursing, you can still do PERT at night. What about anticoagulation? So, this is another question. This is a patient who is going to be started on renal replacement therapy for acute kidney injury. His platelet count is 102,000. Hemoglobin is 7.3. Which one of the following anticoagulation strategies is most likely to yield the longest filter life and lowest risk of bleeding complications for this patient? Is it citrate anticoagulation, unfractionated heparin, no anticoagulation, or regional unfractionated heparin protamine? And the correct answer is regional citrate anticoagulation. So, there have been several heparin versus citrate anticoagulation studies. These three studies were published before 2014. And as you can see, the bleeding risk is always higher with heparin when compared to citrate. But more important, but also as important, the survival of the filter is always, at least in two of these studies, is better with citrate. And more recently, this study was published in JAMA earlier this year. This is from Alex Zarbock's group in Germany. And median filter lifespan was higher with citrate compared to systemic heparin. However, starting a citrate protocol is extremely complicated. It needs involvement of pharmacy, nursing, ICU physicians, nephrologists. So, make sure that all the members of the team is involved if you're planning to start a citrate protocol. So, the advantages of citrate are less risk for bleeding, prolonged filter life, and you are more likely, because you're not going to clot the machine as often, you're more likely to deliver the prescribed dose of RRT. However, citrate is more expensive than heparin. It's associated with more electrolyte abnormalities, such as hypocalcemia, maybe metabolic alkalosis or metabolic acidosis. So, it has various electrolyte problems. And then to teach nurses the correct way of doing this, monitoring citrate, monitoring calcium, it can be very challenging, whereas heparin is very easy to use. And it's very common around the world, so everybody's familiar with it. Drug dosing, I would say make sure you check with your PharmDs every day while the patients are in renal replacement therapy, because drug dosing will change depending on the patient's renal function, but also what type of renal replacement therapy is required. So, in CRRT, for a drug that's normally cleared by the kidneys, you will need to dose it more frequently. So, for example, meropenem, usual dose is three grams per day, but depending on what type of CRRT or RRT you're using and what flow rates you're using, your dose may vary from 500 BID to a gram TID. So, to summarize, renal replacement therapy in the ICU, there's a lot of variables that need to be considered, including patient selection, timing, and modality. Timing of renal replacement therapy, I think the start AKI study, the standard group criteria is applicable for most patients. The one group that might be an exception might be the surgical ICU patients that need to be, especially the CT surgery patients that may require earlier initiation. CRRT is beneficial over IHD in patients with raised intracranial pressure. CRRT is also probably the better alternative for patients who are hemodynamically unstable. If using CRRT, citrate anticoagulation has shown benefit over heparin as far as less bleeding risk and more prolonged filter life. The limitation of citrate is primarily because it's a complex procedure and it's associated with more metabolic abnormalities. Dose of renal replacement therapy for CRRT, the dose is 20 to 25 mls per kg per hour of effluent flow rate. PERT is useful as a substitute for CRRT or IHD or as a transition therapy from CRRT to IHD. If you're doing IHD, three times a week is sufficient, provided that you're offering adequate urea clearance of 70%. Increasing IHD further, the frequency, so if you decided that you want to give dialysis four to five days a week because you think it might be offering more clearance, this may be actually associated with a delay in renal recovery. This has been shown in our post-ad hoc analysis after the ATN study. Drug dosing should be closely monitored during renal replacement therapy. So thank you for listening and I will be back to discuss cases. Thank you.

renal replacement therapy

ICU

acute dialysis services

timing of initiation

intermittent hemodialysis

continuous renal replacement therapy

dosing of RRT