Dialysis vs Conservative Management for Older Adults: Manju Kurella Tamura, Susan Wong, & Maria Montez-Rath

Eric 00:11

Welcome to the GeriPal podcast. This is Eric Widera.

Alex 00:12

This is Alex Smith.

Eric 00:13

And, Alex, I am very excited today because there was an awesome article on Annals on conservative management versus dialysis. Kind of how we think about that. This is a topic I love to talk about. So we got a lot to cover. Who do we have with us today?

Alex 00:30

Got a lot to cover. Delighted to welcome Susan Wong, who’s a nephrologist and she’s associate professor of medicine at the University of Washington, though she’s joining us from Leeds, England, where she is a distinguished Fulbright scholar. Susan, welcome to the GeriPal Podcast.

Susan 00:46

Thank you for having me.

Alex 00:48

And we’re delighted to welcome Maria Montez-Rath, who is a biostatistician and director of the Biostatistics Corps at Stanford in the division of nephrology. Maria, welcome to GeriPal.

Maria 01:00

Thank you for having me.

Alex 01:02

And we’re delighted to welcome Manju Kurella Tamura, who is a nephrologist and professor of medicine at Stanford. Manju, welcome to GeriPal.

Manju 01:12

Thank you, Alex and Eric, Nice to be here.

Eric 01:15

So we have a lot to cover. Again, we’re going to be talking about conservative management versus dialysis in older adults. But before we do, Susan, I think all the way from Leeds, we have a song request.

Susan 01:27

Yes, maybe appropriate because it’s a British band. Let It Be from the Beatles.

Eric 01:32

Why did you choose Let It Be?

Susan 01:36

To reflect continuing medical management versus starting dialysis, which is what Maria’s wonderful paper goes through.

Eric 01:46

Great, Alex.

Alex 01:50

(singing)

Eric 02:34

Wow.

Susan 02:35

That was impressive.

Eric 02:38

We will see at the end of the podcast if let it be is the appropriate, appropriate message we’ll be talking about with conservative management. Well, I got. I got a lot to cover because I’m. I’m really interested in this topic, especially around older adults, because on the inpatient side, we see this all the time. So, for example, you know, we see somebody coming in from the nursing. A nursing home. Maybe they have, you know, advanced ckd, not yet on dialysis. Their function’s kind of decreasing. Maybe they’re having some issues with cognition.

And the question always is, hey, what if we like started dialysis, will any of that improve? And man, it feels like every year those GFR numbers just keep on ticking higher and higher when we’re starting to dialysis. No longer in the less than 10s, but maybe 15s or 16s because we’re having these quasi uremic symptoms. So I wonder if I could turn to either of our nephrologists. Is what I’m seeing real? Are we seeing people like more older adults, more nursing home patients starting dialysis, and even at a higher gfr?

Manju 03:56

You want me to take this, Susan? What you’re describing, Eric, was probably what we saw starting about in the 2000s. There was definitely this idea in nephrology of what we called healthy start dialysis, which is this idea that we should start dialysis before someone develops really florid uremic symptoms. By doing so, we might prevent the mortality associated with kidney failure.

There was definitely a practice shift towards starting dialysis at higher levels of EGFR and with less severe symptoms. I would say that that practice has now shifted back the other way now that we have some randomized trial evidence that suggests that that actually probably does not benefit patients. I think the tide is turning, but.

Eric 04:59

May not be actually happening in real life. Or do we just have the evidence that it doesn’t help but dialysis centers are still dial.

Manju 05:09

I think the tide is shifting, but probably not enough just to throw some statistics out there for you. About half of patients who are over the age of 65 start dialysis with a GFR of over 9, which is sort of the level below which we have clinical trial evidence for suggesting that there may not be a benefit to starting at that level of GFR. So we’re still seeing about half of the older population starts at that level of GFR.

And I think the question for our field is, is that because the estimates that we have of GFR based on creatinine provide misrepresentation or an inaccurate representation of someone’s kidney function. Or is it just this lingering practice of starting dialysis early before really severe complications develop?

Alex 06:10

Or having watched that John Oliver episode about DaVita, we will have a link.

Eric 06:16

To it in our show. Notes.

Alex 06:18

Is it because of tremendous financial incentives to get people on dialysis and into dialysis centers?

Susan 06:25

Well, I think that’s always a worry of whether or not financial incentives play into decisions. I think if you asked any nephrologist, they would deny it, and I think they truly believe it. But there’s been really compelling data to suggest that, you know, your financial model does have some influence and Manju did an amazing study on this looking at different populations. Let me back. Actually, Manju, you probably could talk about yourself, but I cite it all the time of veterans who are equally eligible to get their care and Medicare fee for service system versus the VA and that patients who get their care in Medicare are more likely to start dialysis.

And patient subgroups who I think were most worried about their prognosis on dialysis or getting started are more likely to get started in Medicare if they get their care in Medicare than if they get their care in the VA. So patients who are over the age of 80, patients with metastatic cancer, all the folks that we get, we’re really worried about. And then there’s also another study done by Margaret Yu, who’s also at Stanford now, who looked at timing of dialysis initiation. Again, veterans who are equally eligible to get their care and Medicare or in the VA, those who get their care and Medicare start dialysis at a higher GFR than if they were getting their care in the va. So there seems to be signal but think these just kind of are food for thought right now. But we couldn’t say proof or causal.

Alex 08:01

And my understanding is that the policy and reimbursement rates have shifted to encourage less facility dialysis and more dialysis at home. For example, with peritoneal dialysis and conservative management. Is that fair to say?

Manju 08:16

Yeah. Both Susan and I practice in the va, so we’re a little bit shielded, I think from sort of the true practice change that’s happening outside of the va. But you’re right, there has been a shift in policy to incentivize patients and providers to choose home therapies. To what extent it’s incentivizing conservative management, I would say it’s probably negligible. But maybe that’s just my own VA centric view.

Eric 08:48

Well, the devil’s advocate. So you have, I mean versus a younger, healthier person with worsening ckd. That’s their main thing. They’re having some functional decline. I feel like that’s different than the older adult with multimorbidity. A lot of things may be contributing to their functional decline. A lot of things may be contributing to these symptoms that we may be defining as uremic symptoms, but they also could just be non uremic symptoms from all these other comorbidities. I think that is a big challenge of what we know. If we see somebody’s functional status declining, like how much of that is from their CKD and how Much of that would be reversed from dialysis. Isn’t that a big challenge in older adults?

Manju 09:33

Yeah, Eric, you’re giving me this great segue into how I got interested in this area in the first place. So that was exactly the dilemma that I saw when I was a trainee in nephrology. And we would see the patients that you described at the start of the podcast, older adults with functional decline, cognitive impairment, and kidney failure or advanced ckd. And the question is, is this uremia or is this something else? Is it delirium from something else? And is dialysis the right treatment in this situation? I felt like more often than not, we erred towards dialysis because there it is. It’s right there. It’s easy. That was the lingering question I had was, well, what is it actually doing? The first question I got started with is, what does it do to cognitive function? It got bigger and bigger from there. Well, not just cognitive function. What is actually dialysis? How is it benefiting patients?

Eric 10:42

What did it do to cognitive function?

Manju 10:45

That’s a really difficult question to answer, but it’s probably. We’re probably mixing up two or three different conditions here. There’s definitely uremic encephalopathy, and dialysis does improve uremic encephalopathy. But the truth is that most of us don’t see uremic encephalopathy anymore because we start dialysis before anybody actually develops uremic encephalopathy, except maybe in the acute setting. But then there are all of these older patients who have microvascular brain disease, which you guys could talk about better than, than we can.

And dialysis, at least hemodialysis probably harms that. There’s evidence that suggests that there’s cognitive decline and an increased incidence of dementia after patients start hemodialysis. And that’s corroborated by some magnetic resonance imaging studies as well. It’s a mixed bag, and it probably depends on what the underlying cause of cognitive dysfunction is.

Eric 11:50

I also have to ask, because we will have a link to this article. One of my favorite articles. I talk about it all. This is my favorite article, Susan, that I talk about of Manju is a New England Journal article that came out 2009. 2009, yeah, 2009. Looking at what happens to older adults in nursing homes when they initiate dialysis. What happens to their function? How many of them are alive and dead at one year? How many of them have improved function at one year? Would you mind just describing what you did in that article?

Manju 12:28

Just briefly, thanks for giving me an opportunity to go back to that.

Eric 12:34

Sorry if you have to go way back.

Manju 12:37

No, no worries. I’ll share maybe a small personal anecdote before I tell you about the story. I was while we were waiting to hear back from the New England Journal about this paper, I was nine months pregnant and actually I went into labor about 30 minutes after I got the email that said that our paper had been accepted. And so when my co authors were emailing me and I was of course oblivious because I was in the maternity ward by then. So by the time I finally got a chance to respond to them, I said, well, it’s the second best thing that’s happened to me today,

Manju 13:25

so it will always be linked to my daughter who’s named England.

Susan 13:30

No, I’m kidding.

Eric 13:31

I was going to say who’s named Midjum. Midjum. That’s a really interesting first name.

Manju 13:38

Yeah. Going back to your question, what we did was we had an opportunity through one of my mentors, Glenn Church, who had access to the US Renal Data System, which, as I think all of you know, is this registry of patients in the US who are treated with dialysis or transplant that had been linked to a registry called the Minimum Data Set, which I think you guys know and probably most of your listeners know, is this registry of patients in nursing homes. By linking the two, we had an opportunity to trace out the trajectory of patients functional status relative to when they started dialysis. I remember when I was first just exploring the data and looking at it, I was seeing this dramatic drop and I was showing it to people and I was like, what do you think about this? This looks really surprising to me.

Eric 14:43

Drop in function when dialysis was initiated. So you saw function before and after dialysis and what you were seeing was this drop?

Manju 14:50

Yeah, there was a dramatic drop. So people, it starts to decline a couple of months before they start dialysis, which I think resonates with Mike. Clinical experience. But then after starting dialysis, they don’t recover. Most patients don’t recover. And I think that’s what was so surprising to people because our clinical instinct is to intervene when we see someone declining with the hope that we will prevent that dramatic decline in function and rescue them from that. And we weren’t seeing that. And I think that’s probably why the paper made it into the New England Journal is. So it went counter to what our instincts, which was to intervene.

Eric 15:38

Yeah. I personally loved it because again, you saw this thing. I have face validity. We see people, they’re having some worsening functional decline. Hey, as doctors, we got a fix for that. We got dialysis. Let’s do dialysis and let’s see if their function improves. What I remember from the article is that at one year, nobody improved, but some people maintained. Not a lot, but most people either had a decrease in functional status or were dead at one year.

Alex 16:08

And I’ll just say a couple meta comments. One is people remember figures. Nobody remembers a table. The figure of the functional status decline and then minimal recovery after initiation of dialysis is just a figure that’s seared into my brain.

Eric 16:28

Mine is the bar chart looking at 3 months, 6 months, 9 months, 12 months. Seeing like at 12 months.

Manju 16:33

Well, that figure is a hat tip to Chuck McCulloch, who is still at UCSF. Without his help, I wouldn’t have been able to make that.

Eric 16:42

That was awesome. I love this entire paper because I do think it started making me question what is the role of dialysis as far as making people’s function improve, helping with their symptoms, their quality of life, and starting to question some of this stuff. So I guess the other question is what do we know about, if we think about other things that we use it for? And I think, Susan, I’m going to turn to you on this one. What do we know around quality of life? Because you’ve, you also did a, you did a study looking at this, right?

Susan 17:18

I guess where I would kind of piggyback from Manju’s work here is that I think she, as well as others, started to gather the evidence that to at least bring us to take pause about how beneficial dialysis really was for older patients with multimorbidity and frailty. And I think that myth first had to be debunked before people could be open to the idea of, well, is there something else? Is there something else we can offer, we should offer? And I think that needing to build that evidence was so important to set the stage to be able to even publish studies that would even counter the idea that actually you could maintain people fairly well by continuing medical management or even adding kidney supportive care in a bigger bucket. We call this, you know, conservative kidney management.

So in terms of what we know about conservative kidney management, I would say in the U.S. we, we don’t, we don’t do it well. And so it’s almost like a lost art. You know, before dialysis was invented, ev, you know, everybody was doing their very best at conservative kidney management. And then once dialysis came about in the 1970s, it’s a practice that’s kind of all but disappeared, and yet we’ve developed so many more services, we have so many more medications, we know so much more about the illness that we could provide better support to patients. But this infrastructure hasn’t been built. Most of what we know about conservative kidney management comes from other countries, the UK being one of the most prolific countries in terms of producing research on this.

And so what they’ve been showing here is that when you have an established conservative kidney management pathway, you can maintain patients for quite some time, for several years, even after they’ve made the decision to forego dialysis, and that they’ve seen pretty comparable survival rates between those who are conservatively managed and those who receive dialysis. So it’s been a real inspiration, I think, for many clinicians and researchers in the US to figure out, well, how can we do this better in the US and to make kidney management an option for patients? And I apologize if I’m driving the conversation to this Annals paper, but what I really liked about this paper.

So we’ve been talking about for some. Or I’ve been talking about it for some time, like, look at what they’re doing around the world, this conservative kidney management. We should really look into it. And there’s always these. There’s a lot of skeptics out there. They say, well, you know, it was done in another country. Is it generalizable to the U.S. they have dedicated programs that have been in existence for a couple of decades. We don’t have that here. You know, can we. You know, none of these studies are random. None of the studies are randomized. They’re all observational studies. You know, how trustworthy is this data? And so, you know, Manju and Maria’s paper, it really addresses many of the points of skepticism that people have about all of the data that’s been published in other countries.

Eric 20:42

Well, let’s get to that then. Let’s get to Maria first author Anil’s paper. I absolutely love this paper, got a lot to talk about because it has this interesting study design. But before we do, what got you interested in doing this particular paper?

Maria 21:01

Again, it’s very interesting because it came working with Manju. So I’ve been at the Division of Nephrology for a few years now. And so I’m getting deeper and deeper into pretending to be a nephrologist, even though I’m a statistician. But I do like to dive into the science and learn and understand. And I want to understand not only from the point of view of the doctors that I work with, but also the patients that ultimately we can provide some help. And even though I’m not a medical doctor, I feel like I can help in a way if I can show with data, with statistics, some of these things that are questionable to confirm.

And this idea, like, so far, this paper of combining survival and home time actually came from personal experience with my dad, not because of dialysis, but his surgeon wanted to do an amputation. And my dad was really afraid of that and not wanting to do it. And I went into the literature and I couldn’t find anything that explained really what happens. All of the papers are always about survival. And in the case of. My dad was like, he really. He hated being in the hospital, but he was getting all of these infections that basically were putting him in the hospital all the time. It started like every month, and then, like, was.

He would go home, and then three weeks later he would be back in the hospital. And then this kept on going back and forth. And then, like I said, like, okay, is there anything that actually can tell me whether it’s going to be beneficial in terms of home time, the time that he’s going to be spending at home, which is what he really wants. And then from there, that’s when talking with Manju and the other doctors that I work with, then we said, why don’t we do this?

Eric 23:13

In particular, looking at dialysis patients versus those who are getting dialysis versus those who are getting conservative management, how much time are they spending alive, and not just alive, but also just those days that they’re doing dialysis to, like the. But the institutionalized days, if you will.

Maria 23:35

Yeah, exactly. We knew, because I don’t know if you noticed, we have a paper that also, I’m first off with, where we actually look at what happens on patients on dialysis in the year after they start dialysis and a little bit before. And we have these. I’m a visual person. So we created these heat maps where you could see for each individual what actually happened in the days and you. And see how much time they were alive, how much time they spent in the hospital, in the nursing home and when you start dialysis.

So we had yellow was time spent in a hospital or in the nursing home. And just the amount of time that these patients are spending in an institution was just crazy for me. But then there was always the question, but, okay, what happens if you are not on dialysis? And so this really led to then us working more on this paper and trying to figure that out. I haven’t recreated those heat maps, which I still hope I will be able to do. But at least we were able to put the statistics together and really compare a survival and home time in this Annals paper.

Eric 24:57

Okay, but we’re going to do a little bit different than your usual like journal club. We’re going to jump to results and then I’m going to go back to methods. What did you find in this Annals paper?

Maria 25:09

Mendu, why don’t you go? You’ll do a better job than me.

Manju 25:13

Yeah, I have the paper here, so I’m going to just refer, I’m going to cheat and refer back to it. What we found when we did two different analyses and intention to treat analysis and a per protocol analysis. In the intention to treat analysis, we’re comparing people who started dialysis within 30 days of meeting the eligibility criteria for this hypothetical trial to everyone else. So that includes patients who did later go on to start dialysis and some who never started dialysis.

Eric 25:47

And eligibility was older than 65 and a GFR less than 12.

Manju 25:53

GFR. That’s right. Chronic kidney disease and a GFR less than 12 and people who had not been referred for a transplant evaluation. So selecting people who were probably not.

Eric 26:04

Eligible for a transplant, not acute kidney injury. These are people with chronic kidney and that’s right.

Manju 26:09

Took out people with acute kidney injury. And so in that comparison, the patients who started dialysis had longer survival by nine days but spent two weeks fewer at home.

Eric 26:23

Nine days, Nine days.

Manju 26:28

That wasn’t statistic. Actually nine days.

Eric 26:30

Out of how many days, like how long did they spend?

Manju 26:32

Over three years. Thank you for reminding me. So over three years of follow up, they spent nine additional. They lived longer by nine additional days but spent two weeks less at home. Then we did what we called the per protocol analysis which is comparing people who started dialysis within 30 days of meeting eligibility to those who never started dialysis.

Eric 26:56

Oh, because the other group actually could have started dialysis after 30 days in 3 year, 9 day difference group.

Manju 27:03

Okay, that’s right.

Maria 27:03

So in this comparison, not necessarily that they did not start dialysis, but that we only follow them when they are on dialysis or they haven’t started dialysis yet. Right. So it’s the comparison. There’s no mixing like in the intention to treat analysis. Yeah. If they don’t start dialysis at the beginning of the trial, we follow them as if they never dialysis. It’s just like, okay, but they could have.

Eric 27:32

They could have, but they’re, they’re. Now the analysis is saying that we are, this is our intention, that they were not going to do it, but they could have just like any other randomized controlled trial.

Maria 27:43

Exactly, exactly, exactly. But I mean, the per protocol, right. The control group now is one where we really only look at the time when they were still on medical management. And if they initiate dialysis afterwards, that time is not included in the analysis.

Eric 28:04

What did you find? Go ahead, Alex.

Alex 28:06

Just to clarify for our listeners, just want to say one more time because we’re getting a little researchy here for our largely clinical audience. Comes from like a time when you know, you’re on the protocol and the protocols here are either you’re on the dialysis protocol or you’re on the medical management protocol. And so per protocol analysis means you only are counted in this study for the time that you’re on that protocol or the other protocol, the dialysis protocol or the medical management protocol. Okay, continue.

Eric 28:37

And what did you find in per protocol?

Manju 28:40

So in that analysis, we found that the patients who started dialysis live longer, by 77 days, but still over three years, but still spent two weeks fewer at home.

Alex 28:55

So kind of difference between the intention to treat analysis, meaning like where you started is where you’re counted versus the per protocol analysis, which is is only while you’re in that group in which we assigned you at the beginning. How do you explain the difference?

Manju 29:12

Do you want to take that, Maria?

Maria 29:14

We’ll be more statistical than clinical, I think. But I think even before that, for me, the biggest question here between the intention to treat in the per protocol and I’m a big fan of the intention to treat analysis, and most probably you will have even other statisticians not agreeing with me. But in a lot of things that we do, we really try to get at what happens in the real world. And the intention to treat analysis is really giving us that. Because from what I learned working with Manju and Susan and all the other nephrologists that I work with, it’s just like nothing is set in stone. And so you have patience and the patient comes to you and you’re trying to make a decision right now, like, should I start dialysis or not?

And let’s say you’re going to decide like, no, actually now, I don’t know. I do not want to start dialysis. But I hear from them just like, yeah, we don’t really plan for dialysis. And then six months later, they decided something happens and they decided they want to start dialysis. Right. And that’s what happens. That’s the real world and that you only get in the intention to treat. Because if you are, when you’re trying to help the patient, if you want to use these results. It’s about the question that you’re asking right now. If I make a decision right now, what is it that I can expect? And right now you don’t know whether that patient is going to start dialysis in the future or not.

Eric 30:51

Well, I guess the other question too though is like if this was a drug like we use intention to treat because the people who, let’s say don’t take the drug are very different than the people that do take the drug. The people who are adherent are generally very different than those who don’t. So if you use like a per protocol analysis in a randomized controlled trial of a drug, it’s going to look different than an intention to treat analysis, isn’t it, Maria?

Maria 31:16

Yes, exactly. Exactly. And then the question then is like what is, what are the results generalizable to?

Alex 31:24

Right? Yeah.

Maria 31:24

Right.

Alex 31:25

We’re going to force you to stay on the dialysis pathway from this point on where no matter what you say or we’re going to force you to stay in the medicine medical management pathway. No matter what you say. That doesn’t make any sense.

Maria 31:37

That doesn’t make any sense to me. Exactly. Exactly. Yeah.

Eric 31:41

So let me ask you, going a little wonky question. So you did this. Was this a randomized controlled trial?

Maria 31:48

No, it wasn’t.

Eric 31:50

You said you emulated like what was this? This is a. The Annals loves this stuff.

Alex 31:57

Right?

Eric 31:57

Now this is a. What’s the official title?

Maria 32:01

It’s a target trial emulation.

Eric 32:04

Target trial emulation. Is that just an observational study with a fancy name that makes it sound more like a randomized control trial or are you doing something different than just the normal observational study?

Maria 32:20

Yes and no. Because I think that what the target trial emulation does is puts you in a place, is when you are designing the study, right. It puts you in a place where, okay, let’s just pretend that this is a clinical trial, right? And when you’re thinking of a clinical trial, we are in the mindset that, okay, we have protocols that we have to follow, we have standards that we need to follow, like we want to keep those.

So in the target trial emulation, that’s let’s basically puts us in that mindset that even though it’s observational data, even though the data has already been collected and we are just repurposing it for this study, we can still follow these strict Protocols that make then the results more reliable. And so for me, it’s a good strategy and it helps me when I discuss these with the researchers that I work with to really say, look, let’s just pretend that this is a clinical trial, let’s follow these protocols and let’s be strict with what we do to make sure that we can trust the results that we get at the end.

Eric 33:40

Yeah, it seems like you’re putting all the best practices for an observational trial.

Maria 33:44

Exactly.

Eric 33:45

Meshing that together, maybe branding it a little bit.

Manju 33:48

I would say that there are two big pitfalls that you see with studies that don’t use target trial emulation and use an observational framework when they’re trying to do these comparisons. One is they’re defining the intervention using future information, which Maria always tells me is a big no, no. You can’t define the groups by who got dialysis in the future. That’s a big no, no. And two, related to that, a lot of these earlier studies are affected by immortal time bias. And so target trial emulation is an approach to try to handle those design issues that we get with observational studies.

Eric 34:36

And in one sentence, if you can, in one sentence, anybody, what is a mortal time bias?

Manju 34:41

I got to turn that over to Maria again.

Eric 34:43

Maria.

Maria 34:46

So is when. Let’s go back thinking of the dialysis, right? You don’t really know. You don’t have a starting time for medical management. And so if you’re only waiting until someone starts the medication, then that waiting time is not counted for or medication or treatment. Right. And so like that’s the immortal time bias, where someone would not be eligible to really start the treatment or is eligible, but you don’t know and you are counting that or not counting it in towards your analysis. And so that your follow up time then at the end is basically incorrect.

Eric 35:30

Go ahead, Alex.

Alex 35:31

Well, I was just going to try to take a stab to answer your question. Take a stab, Alex, in one sentence, I think more total time bias is when the start time for one group doesn’t begin until some event happens and therefore they have to survive until that event. And you’re counting that survival time in your analysis. And that biases them because they’re immortal in a sense, until that event happens.

Maria 36:03

Yes.

Eric 36:04

Is that fair? All right. Yes, that’s fair. My last wonky question, fundamentally though, you’re not randomizing two groups. There’s no emulation for randomization. The biggest issue here is there’s going to be a difference between people who start dialysis and don’t stop dialysis no matter what fancy statistics we use. We. Right. Can we really ignore the need for randomization or like I was asking you Susan earlier, like a randomized controlled trial. Thoughts?

Susan 36:39

So, yeah, a couple of thoughts. So why I like this target emulation trial design is that one tries their very best to account for that kind of selection bias. There’s all this propensity score matching and versus probability waiting and all the, the fancy stuff that Maria does. Is there still residual compounding? Absolutely. And I think that’s always going to be the case. Observational studies, but it gets us as close as one can possibly get to the answer, you know, to answering the question, you know, should there be a clinical trial? There is actually a couple ongoing.

Eric 37:15

Oh really?

Susan 37:16

Patients. Yeah. So you know, for a long time people have thought that doing a clinical trial, randomizing patients to dialysis or to conservative kidney management was, was unethical. Others have argued now that there’s so much observational data suggesting clinical equipoise between the two, it’s now unethical to not explore this possibility. So in the UK at the University of Bristol they have, they’re maybe three or four years now into a clinical trial randomizing patients to prepare for dialysis and or to prepare for conservative kidney management. They have at least 380 patients now. I think the study closes to enrollment sometime next year.

So we’ll learn. And then right now there’s a Dale Lupu has and I think Woody Moss is maybe, I think they’re copias on that. A PCORI study looking at randomizing patients with late stage 4 or 5 kidney disease to receive kidney supportive care, which is, which is the major element to conservative kidney management and randomizing them to see how that might support their decision making and then kind of long term follow up following I think survival as well as quality of life measures and things. It’s an indirect study of what happens when you do provide patients support for when you want to do dial. It’s as close as we can get I think in the US to try quickly.

Alex 38:43

Just a defense of observational studies because this Annals paper, Maria first author on is an observational study. But observational studies like they’re more real world in some sense when you’re capturing like what happens in the real world and following patients. Patients in this case I believe this was VA patients. Right. So this is like all va. Right. As opposed to randomized trial, you get highly select patients like there are some patients who like you can say it’s equipoise and they’re. But they’re going to say, like, no, you’re not going to tell me, randomized mel dialysis or conservative management. I’m not enrolling in that study. And so you’re going to lose all of these patients. And I. So I worry about that. From randomized trials.

Eric 39:27

Do you do observational research, Alex? Conflicts of interest.

Alex 39:32

He’s saying I’m biased.

Eric 39:35

I want to acknowledge. We got a couple minutes left. I want to go back to Maria. Maria, the reason you did this, right. You wanted to be able to. Be able to tell patients what to expect. And I’m wondering if we can just take two minutes to think about if you were going to have a discussion about what’s the difference between conservative management and starting dialysis for these people. Let’s say this group of people, older adults, maybe some multimorbidity, GFR is less than 12. How would you talk to them about survival, quality of life, or any other benefits or harms when you’re thinking between these two groups?

Maria 40:17

Yeah. So I’m going to answer not from the medical doctor perspective, because that’s Mendew enthusiasm. But I. Again, going back to my dad, if I had to have a discussion with my dad, how would I argue this? I think what we really want is for people to just don’t assume that dialysis is the answer for these elderly patients. Of course, if they are young, there’s really no question. But the question comes when they are older and what is it that they really want? And if we just stay with survival, then the question is, yes, the patients. If you ask the patient, do you want to live longer? They’re going to say, like, yes, I want to live longer.

Do you want to live longer? Nine days for me, that’s really the question. Not even 70 days out of three years with the idea that, okay, I can give you three months of extra survival, but you’re going to spend more time in the hospital, and you’re going to be spending the time going to a dialysis facility pretty much every other day.

Eric 41:26

And most of that extra time was dialysis daytime.

Maria 41:31

A lot of it is dialysis time. Right. Because they have to go to a dialysis facility for, you know, three times a week for the most part. And so are the patients. Are you okay? Like, as. As the caregiver or as a family member of the patient? Is that really in the best interest of the patient?

Eric 41:53

Great. I love that. How about for you, Susan? How would you think about having this discussion?

Susan 42:01

I think maybe I’m walking Away with the Annals paper with too simplistic an interpretation. But I take the intention to treat analysis dialysis is confirming that we really should delay dialysis initiation for as long as possible.

Eric 42:18

There’s no reason to rush it.

Susan 42:19

There’s no reason to rush. And Manju referred to this ideal trial that randomized patients to early versus late start. And there were some criticism about it. But I think this study locks it in that really, you know, you don’t need a clinical trial. If just anybody walking into your clinic with the resources that we can provide, we can. We should probably try to delay this as much as we can. And then the per protocol analysis, I see that as an opportunity that indeed patients who were started on dialysis did live a little longer. 77 days, but that’s not very long at all, you know, at least in my book.

Eric 42:59

77 out of three years.

Susan 43:01

Yeah, three years. How can we close the gap? And that’s where I think this is where kidney supportive care can close that gap. You know, I think we were very measured in saying it’s continuing medical management because we didn’t know exactly the kind of care that patients were getting. Maybe they were getting kidney supportive care or conserved kidney management as a whole, but maybe not. But even if you didn’t have these formal structures in place like they have in the uk, you know, they. There was not that. The distance was not that wide. So it gives me hope that we can close that distance if we can provide the right infrastructure of care for them. So, yeah, that’s what I’m walking away with this.

Eric 43:41

Oh, lovely.

Susan 43:42

My patients. Yeah. It’s just a far more measured framing of what dialysis can offer somebody. It’s not a panacea.

Eric 43:50

Manju, how about you?

Manju 43:53

I think Maria and Susan expressed what to say to patients so well. So maybe I’ll express what to say to our colleagues in nephrology and in the policy world, which is that we have equipoise here. There is enough evidence here to suggest that we actually don’t know which path is beneficial and we really need more trials. Although I’m an observational study sort of person, I really do think that the way to move the needle in practice is with trials and I think that we. There’s so many unanswered questions now about how to deliver really good conservative kidney management that we should not wait for just England to do these trials, we should do them here as well.

Eric 44:41

Yeah. And I also love. I’m going to go back to that new paper from 2009 which we’ll have linked to. Just don’t assume that function is going to improve post dialysis because they’re in dialysis for three days a week, they’re sleepy afterwards. They’re just function is going to likely get worse for many of these patients. And this is not the fix. So I thought, Manju, I thought your last phrase was going to be let it be, which I was going to argue is my line. [laughter]

Manju 45:13

I should have worked that in. I was not thinking ahead. So should I say we should just let it be? [laughter]

Eric 45:20

Oh, there you go.

Alex 45:24

(singing)

Eric 46:09

Susan, Maria, Manju, thank you for joining us on this GeriPal podcast.

Manju 46:13

Thank you so much.

Eric 46:17

And to all of our listeners. Again, we have a ton of stuff in the show notes. Check it out for the articles and thank you for your continued support.

***** Claim your CME credit for this episode! *****

Claim your CME credit for EP333 “Dialysis vs Conservative Management for Older Adults”
https://ww2.highmarksce.com/ucsf/index.cfm?do=ip.claimCreditApp&eventID=14459

Note:
If you have not already registered for the annual CME subscription (cost is $100 for a year’s worth of CME podcasts), you can register here https://cme-reg.configio.com/pd/3315?code=6PhHcL752r

For more info on the CME credit, go to https://geripal.org/cme/

Disclosures:
Moderators Drs. Widera and Smith have no relationships to disclose.  Guests Manju Kurella Tamura, Susan Wong, & Maria Montez-Rath have no relationships to disclose.

Accreditation
In support of improving patient care, UCSF Office of CME is jointly accredited by the Accreditation Council for Continuing Medical Education (ACCME), the Accreditation Council for Pharmacy Education (ACPE), and the American Nurses Credentialing Center (ANCC), to provide continuing education for the healthcare team.

Designation
University of California, San Francisco, designates this enduring material for a maximum of 0.75 AMA PRA Category 1 credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

MOC
Successful completion of this CME activity, which includes participation in the evaluation component, enables the participant to earn up to 0.75 MOC points per podcast in the American Board of Internal Medicine’s (ABIM) Maintenance of Certification (MOC) program. It is the CME activity provider’s responsibility to submit participant completion information to ACCME for the purpose of granting ABIM MOC credit.

ABIM MOC credit will be offered to subscribers in November, 2024.  Subscribers will claim MOC credit by completing an evaluation with self-reflection questions. For any MOC questions, please email moc@ucsf.edu.