SCCM is performing maintenance on its websites. For the best browsing experience, please use Microsoft Edge or Safari. Those using Chrome or Firefox may experience access issues at this time.

SCCM Pod-429 Drug Dosing Considerations in AKI and RRT

visual bubble
visual bubble

Continuous renal replacement therapy (CRRT) is complex, and dosing varies among institutions. Host Michael Smith, MD, is joined by Melissa L. Thompson Bastin, PharmD, BS, BCPS, to review dosing regimens for CRRT, studies of CRRT antibiotic dosing, and CRRT utilization in COVID-19 patients. Dr. Thompson Bastin is a medical ICU and pulmonary critical care pharmacist and assistant adjunct professor at the University of Kentucky in Lexington, Kentucky, USA. This podcast is sponsored by Baxter Healthcare.

*If you are unable to play the podcast please click here to download the file.

Category: Other


Dr. Michael Smith (Host):  Hello and welcome to the Society of Critical Care Medicine’s iCriticalCare Podcast. I’m your host, Dr. Mike Smith. and today, we’re going to be talking about drug dosing considerations in acute kidney injury and renal replacement therapies. I’m joined by Dr. Melissa Thompson Bastin. She is a Pulmonary Critical Care Pharmacist and Assistant Adjunct Professor at the University of Kentucky. Dr. Bastin, welcome to the show. I wanted to start off because obviously this is a very complex topic, so I wanted to just give you a moment to maybe give us a synopsis of the true complexities around dosing and renal replacement therapy.

Melissa Thompson Bastin, PharmD, BS, BCPS (Guest):  As you suggest, an extraordinarily complex actually for a lot of reasons. I think one of the main reasons is as a critical care practitioner, we strive to be evidence based practitioners and a lot of what we do in the ICU has evidence or has a physiological rationale if the evidence is lacking. But with CRT, we really don’t have a lot of evidence to guide dosing adjustments and empiric dosing regimens for our critically ill patients. The best evidence for medication dosing comes from the FDA approval process where healthy patients are given the drug and levels are obtained and the pharmacokinetic parameters are calculated. And a lot of that, all of that, is extrapolated into the ICU. So, drug dosing in the ICU in general, is incredibly complex. But then you layer in something like CRT where we’ve only had about 25 to 30 years of experience being a rather new therapy for renal replacement and the evidence are limited even more.

So, we don’t have a lot of evidence to guide us here. and the nephrology practices have evolved and changed over the years to accommodate the complexity of our patients and the growing body of evidence on how to treat these patients. For example, with regards to volume status et cetera. It is extremely complex, and we don’t have a lot of evidence to guide us. So, I think there are three main components to the dosing that need to be kind of broken down one at a time. So, you have patient-specific factors so recognizing that these patients are critically ill and if they are getting a prescription for CRT; probably quite a few things have happened to them prior to that point, right. So, my practice is in the medical ICU. We see a lot of sepsis and septic shock. The majority of our indications are for volume overload and acidosis. So, these patients are generally volume resuscitated when they come in and that in combination with vasoactive agents to increase blood pressure and the physiologic changes that happen within the patient in a septic patient with endothelial damage and capillary leak. So all of that contributes to a wide variation in the volume of distribution for our hydrophilic drugs and drugs with a low volume of distribution.

And those drugs are our antibiotics. So, most of our antibiotics fit the criteria that needs to be really looked at for drug dosing in CRT and in fact, when you’re talking about a septic shock patient, that’s the only drug that’s going to save that patient’s life. So, getting the dose correct is really important. Some of these patients come into the unit having being fluid resuscitated at an outside hospital, possibly receiving five, ten, fifteen liters of fluid and being anuric for a period of time prior to CRT initiation. So, that patient’s volume distribution is much greater than the patient – the 20 year old male who was studied in the cefepime trials back in the day. So, recognizing that this is not the same patient population and I think it’s easy to recognize it. I think what’s difficult is – and where we really haven’t advanced evidence that much is what do we do about it.

So, if a patient normally would have received two grams of cefepime, now this critically ill patient with a volume distribution that’s twofold or higher greater should probably get a twofold or higher greater dose. Yet we don’t have the tools right now readily available to first of all measure that change in the volume of distribution and safely provide that higher dose thinking about the therapeutic index of these drugs. Which granted, for the beta lactam class of antibiotics, is generally considered very safe. However, there’s a lot of data emerging and folks around the world are doing pharmacokinetic studies looking at some of the adverse effects of these beta lactam drugs and probably realizing that the toxicities are a little bit higher than we once thought and the therapeutic index which is the range of serum concentrations that can be efficacious without being toxic is probably more narrow that we originally thought as well.

So, we know these patients are different as they come in and they require CRT. But we don’t have the data right now to support changing an empiric regimen. Additionally, knowing – you have to know the medication. So, not only knowing the patient, who is this patient receiving CRT? Who needs this medication? What is this medication that I am giving? There are – if you do a PubMed search, you’ll find a lot of literature looking at small pharmacokinetic studies that would suggest alternative or optimized dosing regimens and that’s one of the first things I encourage people to do is check PubMed because this is a hot area of research and it has been for a while and I think it will continue to be until we have larger trials that really help us understand what dose to pick out for these patients.

But reviewing your medications. So, there are some physiochemical properties that are really important for CRT. Well one of the first things you can look at, is does it have renal dose adjustments. And if it does, that kind of one of your eyeball check I’ll say to say okay this drug probably will get removed through CRT, let me continue further looking into it. The physiochemical properties that are important are number one, volume of distribution. So, like as we said, that patient with septic shock who is ten liters volume overloaded has an extraordinarily high volume of distribution. So for a drug like vancomycin for example, that has a typical volume of 0.7 liters per kilo; you would expect a higher volume in that patient. Now again, it’s difficult to measure that. It’s difficult to know exactly what to do with that.

Vancomycin is a little bit of an exception to many of our other antibiotics in general. Because we do have the ability to perform therapeutic drug monitoring which is a really important tool for critical care pharmacists and physicians and practitioners to optimize these drugs but it’s not available for all the drugs that we possibly need it for.

The second thing to look at is the protein binding of your drug. So, drugs that are highly protein bound are not readily available in the serum. So, they are bound to their binding sites like albumin for example and a molecule like albumin does not get removed through any modality of CRT. Now if you’re talking about plasmapheresis, that’s different but we’re focusing on CRT today and so albumin does not get removed through CRT. So, drugs that are bound to albumin will also not get removed through CRT. But one thing to keep in mind is that there is somewhat of a rebound effect for highly protein bound drugs which goes counter to everything that we think about for a drug being highly protein bound generally it doesn’t require renal dose adjustments but there are some data that look at some of these protein bound drugs and as you are removing the low amount in the serum, there is redistribution from the protein site into the serum. And so, you can see somewhat of an additional removal through CRT.

But in general, drugs that are less than 90% protein bound will potentially be removed through CRT. And through a modality such as convectioner CVVH modalities, the removal is inversely related to the protein binding. And there are some equations that we’re not going to get into today because we just don’t have time, that you can actually calculate out what that looks like from a clearance standpoint.

And then volume of distribution is the third physiochemical property that I really have people focus on when considering medications. And that goes back to a little bit of our third point which is the CRT itself and that – a molecular weight, I’m sorry relates to the CRT settings because most CRT filters effectively remove molecules into the 20,000 Dalton range. However many of them if you look in the package insert say they go up to 45,000 to 50,000 Daltons. But if you ask most nephrologists, they’ll say maybe into the low 20s you’ll actually get effective removal, well somewhat effective removal at the beginning of your CRT session and then that saturates over time. So, in general, you get better removal of solute when your CRT starts or if you have a new filter change for example. Which for many centers probably around every one to three days and for drugs that have a large molecular weight, you’ll see more removal at the beginning.

But what we generally think of for drugs that are effectively removed through CRT are drugs with a medium to low molecular weight. So, a low molecular weight anywhere between zero and 250 to 500 Daltons and within that range almost all of our important antibiotics fit within that range. A lot of our antiepileptic medications fit within that range as well so, these are very important drugs that we need to be dosing appropriately.

Host:  I wanted to ask a quick question that when you’re talking about some of these drugs being cleared through continuous renal replacement therapy, does the amount of time somebody’s on CRRT, does that ever play a role in the clearance of some of these drugs, meaning two days versus two weeks for some like COVID-19 patients for instance?

Dr. Bastin:  Yeah, absolutely. It does matter. It matters how long your patient has been on the therapy and that can be a few things. So, duration of therapy and then the effective dose delivered. So, you have the initial soluble removal is always going to be greatest at the beginning of therapy. So, right when you start your patient on CRT, and when you change your set. And like I said earlier, the set gets changed every one to three days. Many of the manufacturers will say change at 72 hours no matter what however, as we know, practicing in the ICU, these filters go down, they get clogged, they get clotted, things happen. A normal effective filter life is around 24 to 36 hours. So, basically, every day your patient is getting a new filter and new set. Right at the beginning of that, you’re going to have the best removal. In general, CRT is a slow therapy. It takes a few days for it to reach its what I’ll call kind of loosely steady state removal and when you’re watching these patients labs, it’s easy to watch labs as they go down through CRT. Like you can’t watch cefepime levels because we don’t have cefepime levels but you are going to remove probably the maximum amount in the first couple of days and then, ideally, you are just keeping patients at a steady state removal.

This also depends on the prescription and this is very nephrologist dependent and center dependent I would say. We do have national guidelines that suggest an effective dose of 20 to 25 mills per kilo per hour. However, I do know that in some centers, a higher dose is targeted because when you talk about duration, you also want to look at how much is your patient actually receiving. Because the dose prescribed does not always equal the dose delivered. And what that means is maybe you’re prescribing for a daily dose of 30 mills per kilo per hour, but you know there are probably two to three hours where that patient is off therapy totally. And so you need to count that into a 22 hour period not a 24 hour period.

So, most of the machines actually can give you that data if you’re looking at the card. But as for dosing purposes, that’s the dose prescribed and that’s usually what we empirically will go off of.

Host:  At this stage of it, at this time period, you’re basically making a lot of decisions about dosing in multiple therapeutics going on in critically ill patients with not a lot of evidence. You had mentioned that at the beginning. And so you are taking your knowledge of in this case, renal physiology, renal replacement therapy, the pharmacokinetics of these drugs, clearance, all that kind of stuff and you’re making what are good decisions I’m sure most of the time. So, what are we missing then? What would you like to see – things to be studied here? Like are there clinical trials in the works to help you make better decisions with this? If not, what trials do we need to look into to make this a more evidence based approach?

Dr. Bastin:  So, there are some studies that help us here and I think the most useful studies I’ll say are with drugs that we don’t have the ability to measure. Most drugs, all antibiotics have a pharmacodynamic target and as it relates to the patient. So, we call this the pharmacokinetic dynamic target. And this is the concentration that you want in the patient’s serum as it relates to the bacteria that you’re treating. And this varies from drug to bug, but these are known targets in the literature and they’re well-established and it’s what we target in normal antibiotic pharmacokinetic studies.

So, knowing that, you can go into a situation if you don’t have a bunch of evidence to tell you what to do; if you have the ability to monitor this patient’s antibiotic level, you can optimize that antibiotic to the patient’s pharmacokinetic dynamic target. A good example of that, I don’t want to get into too many details, but one class of drugs that I will keep harping on are the beta lactam class of antibiotics and cephalosporins. So, these are our backbone drugs for sepsis. We have a lot of smaller pharmacokinetic studies that suggest standard dosing in CRT patients is inadequate. So, when I say standard dosing, I mean high dose, normal dose, something that you would give a patient with a GFR of over 50 for example. We know that those regimens are inadequate. But what we don’t have right now is the ability to safely say well let’s increase the dose or by decreasing the dosing interval to increase the total daily dose. We don’t have that data.

And I think as – if there are more studies that used a personalized medication dosing approach for these patients and possibly compared it to a standard approach, which is what I think most centers do now; and provided some safety data and provided their pharmacokinetic data that suggested these are the concentrations we achieved and this is what we were targeting et cetera; I think people would be more comfortable. Because once upon a time, everyone got a gram q 12 of vancomycin, right. It was just through practice and experience and publication of those experiences that we were able to say well you know, that’s not adequate for most of our patients, let’s optimize this and personalize this to the patient and measure their levels. And vancomycin has been around for 60 years or so. We’ve increased it’s safety by doing that a lot especially with AEC based dosing which is in the newest guidelines.

So, what’s missing I guess to answer your question is the comfort and the safety data and the comfort from the physicians and pharmacists to go outside of the normal dosing with these drugs. I think it’s coming. There are a lot of groups around the world working on this. But the more data we have to help with this in CRT and then this doesn’t just apply to CRT, it applies to critically ill patients as well. Because again, like that patient we described in the beginning, that patient might not get CRT and what do you do then? Still I think the pharmacokinetic alterations still exist even if they don’t get put on therapy.

Host:  Yeah. Well Dr. Bastin, I have to tell you, it is crystal clear to me that you are very passionate about what you do, and I think that’s fantastic and we need more Dr. Bastins on the frontline to help figure this stuff out. Let’s just end, and again I know this is complex. There’s so many details. This is just a podcast so, in summary, what would you like the audience to know then about dosing considerations in say somebody on renal replacement therapy?

Dr. Bastin:  Yeah absolutely. That’s a big loaded question but I’ll try to summarize quickly. So, this is an interprofessional multidisciplinary issue. It’s not just the ICU team, it’s not just the nephrology team or the pharmacy team. I think all these groups including nursing and nursing is extraordinarily important here and we need to work together to optimize therapy for these patients. So, this includes communication for when patients start therapy and stop therapy. There’s a lot of times where there’s a lag window there where doses need to be held et cetera. Communicating that to pharmacy. We need at institutions I would encourage them to have dosing guidelines so, three are some things that we can do and some carte blanc things that we know need to be done for these patients and CRT with regards to dosing. And there are some review articles that can help guide institutional guidelines as well. But that needs to be in place in your institution. As well as taking into consideration the individual practice.

So, nephrology practice is heterogeneous at best. And sometimes you have multiple practice partners in the same hospital using different variations of CRT, using different doses and so education of everybody to know what does this mean for my patient and what does that translate into the medication dose. Monitoring of these patients is important too. One thing we didn’t talk about was residual renal function. And so, some of these patients get put on therapy for non-acute kidney injury reasons. And there’s a list of indications that we could go through but effectively, if they still have good urine output, and they’re on therapy; that is like a double whammy GFR for that patient. So, we need to be monitoring that. We need nursing to say oh, this patient’s urine output is picking up. Let me communicate that to the team. Let me make sure pharmacy knows about that et cetera.
And I think ultimately, use of therapeutic drug monitoring if you have it. So, almost all centers should have vancomycin and aminoglycoside therapeutic drug monitoring and hopefully we can start advancing the practice to other antibiotics that we know have a therapeutic window to target. And that can help us even improve the dosing strategies for CRT patients.

Host:  Well Dr. Bastin, that was a pretty good summary considering that was a loaded question I gave you at the end there. Excellent job. This concludes another edition of the iCriticalCare Podcast. For the iCriticalCare Podcast, I’m Dr. Mike. Thanks for listening.



Knowledge Area: Pharmacology Renal