The future of kidney dialysis
33:43 Share

This is Stanford Engineering's The Future of Everything, and I'm your host, Russ Altman. As we start the new year, I thought it would be good to revisit the original intent of this show. In 2017, when we started, we wanted to create a forum to dive into and discuss the motivations and the research that my colleagues do across the campus in science, technology, engineering, medicine, and other topics.

Stanford University and all universities, for the most part, have a long history of doing important work that impacts the world. And it's a joy to share with you how this work is motivated by humans who are working hard to create a better future for everybody. In that spirit, I hope you will walk away from every episode with a deeper understanding of the work that's in progress here and that you'll share it with your friends, family, neighbors, co-workers as well.

So in that analysis, we found that the people who started dialysis early lived on average longer by nine days. Wow. Yeah, but they spent two weeks fewer at home. So that number is startling. A very modest extension of life advantage. That's right. This is Stanford Engineering's The Future of Everything, and I'm your host, Russ Altman. If you're enjoying the podcast, please consider rating and reviewing it.

Give us a five if you love it. But we really appreciate it. It helps us get feedback from the community and improve the show. Today, Manju Tamora from Stanford University will tell us that the decision to do dialysis is much more complicated than some people realized. It has a lot of side effects and it doesn't always lead to a better quality of life. She and her colleagues are re-evaluating when's the right time and the right way to do dialysis. It's the future of kidney dialysis.

Before we get started, another reminder to rate and review the show so that we can find out how we're doing and improve it.

So kidneys are important. Kidneys filter out bad chemicals from your blood. You have two of them and you need them both throughout your life. Unfortunately, there are a bunch of disease processes including diabetes, hypertension, and other diseases that can make kidney function go down, sometimes leading to complete kidney failure. In the setting of kidney failure, then you need to filter the blood some other way.

You can get a transplanted kidney or you can do dialysis. Many people know somebody who's done dialysis where they go to the clinic and spend three or four hours having their blood filtered a few times a week. Well,

The history of kidney dialysis has been long. We've been doing it routinely since the 1970s. And we're starting to think critically about when we should start it, how long it should last, and is it really delivering those benefits to patients? Well, Munju Tamura is a nephrologist kidney expert at Stanford University and a professor of medicine and nephrology at Stanford University Medical School.

She recently did a study that showed that dialysis may not have all the benefits we expected and that when you start it, might need a reconsideration. She's also thinking about alternatives to dialysis, either using it in different ways or not even using it at all.

Manju, before we talk about your recent work on kidney dialysis and some of the ways we should think about it and maybe change our use of it, let's start with a basic tutorial about why the kidney is a very important organ and what it does for us. So the kidney is like a filter. And so whenever we eat, take medications or exercise, waste products build up in our body.

And the kidney's job is to remove those waste products.

And evolution has engineered our kidneys to be both multifunctional in that the kidneys can clear solutes or waste products with many different chemical properties. So small, large, protein bound, not protein bound, but also wonderfully precise. And so that it can balance many, you know, the levels of many different electrolytes and solutes at the same time.

So that's what the kidney does. Now, dialysis treatment, that's anticipating perhaps your next question, is meant to be a replacement for the kidneys. But our current therapy hasn't reached that level of sophistication of our kidneys. So it's not as targeted and not as multifunctional.

As our kidneys are. Thank you. And as you know, I have medical training and I remember very clearly in medical school, my kidney professor saying that the dumbest kidney is smarter than the smartest person.

medical students. And that was in response to your comment about precision, that it really is exquisitely precise in getting the levels of all these chemicals to be just right. Why does the kidney fail? Because we know dialysis is something that happens at the end when basically the kidney isn't doing its job anymore. What are the main reasons either in the U.S. or worldwide for kidney failure? Dr.

Yeah, so in the U.S. and in most developed countries, the leading cause of kidney failure is far and away diabetes. And following diabetes, there's a host of other conditions that can lead to kidney failure. We often attribute high blood pressure to kidney failure, though now we're learning more about the genetic conditions

conditions that underpin perhaps what we've previously called hypertensive kidney disease, as well as a host of immune-related conditions, things like lupus and other conditions, genetic conditions like polycystic kidney disease.

and many others. So there's a, you know, after diabetes, there's a range of different causes of kidney failure. Okay. And unfortunately, in some of these cases, it leads to not just slight damage to the kidneys, but it can lead to, my understanding is, to complete like lack of function. And what happens to the patient when they don't have a functioning kidney?

Or two kidneys. I should mention there are two kidneys, and I presume they both go down together. That's right. And most people who have a systemic cause of kidney failure both are usually failing or affected by the same condition at the same time.

So we call the clinical syndrome that results from kidney failure uremia. And those symptoms for people who have chronic kidney disease, those symptoms tend to develop gradually. So there's not a specific time point at which they appear. And the symptoms can be things like fluid accumulation that you might see in your legs or in your lungs and causes shortness of breath.

But it can also be symptoms like nausea, poor appetite, a peculiar taste that people have in their mouth that leads them to have poor appetite, as well as itchy skin. And many of the symptoms are nonspecific. So certainly kidney failure can cause these symptoms, but there are a lot of other medical problems that can also cause these symptoms, which can make it difficult for clinicians to distinguish between

whether symptoms are truly from kidney failure or due to another condition, particularly if someone has multiple health conditions. Great. And so as you can tell, I'm headed towards dialysis, as you knew I would be. But before we go to dialysis, which is the treatment for this end stage, one of the treatments, I want to mention briefly kidney transplantation, because that's another thing we hear about all the time.

Is kidney transplantation a viable option for most of the people who have kidney failure, or is it only really relevant for a subset of people with kidney failure? Kidney transplantation is usually the first treatment option that we think about when people are approaching kidney failure.

There are a number of potential barriers, both logistic and clinical. The clinical barriers can be things like coronary artery disease that can't be treated with things like stents or surgery, other serious organ failure like heart failure or liver failure, and

frailty, really advanced frailty. So those sorts of things can present a barrier to obtaining a kidney transplant. And I know there are also supply issues, is my understanding, is that we don't have as many kidneys as we would like for these transplants. That's right. So for people who don't have a living donor...

they have an option for a deceased donor transplant. But right now, there are not enough deceased donors to match the number of people who have kidney failure. So what that means is that the waiting time for a transplant exceeds four years. And where you and I live in California, it's much more than that, depending on your blood type.

So that's right. A lot of people just can't access a transplant because of the waiting time. Great. So just a public message from the future of everything to please consider whether you would like to be a kidney donor, but we won't go any further into that. So now we get to dialysis. Can you describe what dialysis is and how it works? What's the experience of the patients? And then we can get into some of your recent research about the appropriateness and how effective dialysis is looking to be.

Yeah. So dialysis, we generally think of it as coming in two flavors.

Hemodialysis, which is done through the bloodstream, and peritoneal dialysis, which is done through the peritoneal cavity. That's the abdominal cavity lining your abdominal organs. Great. And the basic goal of dialysis therapy is to clean the bloodstream or filter out the waste products that accumulate in the body when the kidneys are not working properly. Right.

So taking dialysis as an example, that's the predominant mode that we deliver dialysis therapy in the United States. Blood is removed from a patient and passed through a filter. That filter takes out waste products that accumulate and then infuses or diffuses, if you will, clean blood back to the patient.

And how do they access the blood? How do they get access to the large volume and how long does it take? What is the patient experience of dialysis?

So most patients who receive hemodialysis do so through one of the veins in their forearm. We call that a fistula or a graft. What that is is a surgical connection between a small artery and a vein in the forearm. And that's done to enlarge one of the veins in the forearm so that the vein can be repeatedly accessed for dialysis treatment.

In circumstances where patients don't have a fistula or graft, sometimes called a shunt, placed, we can access the bloodstream through a tube that's placed into a vein in the chest. Okay.

And do they come, what's the frequency? And they come to a center. Just give a brief description of what that's like, because I know that's very germane to a lot of your research on how good this experience is. Yeah. So for most patients who receive hemodialysis, they receive these treatments in a clinic or an outpatient center. Typically, these are done three times a week.

And the typical duration of treatment can be anywhere from three and a half to four hours. The exact amount will depend on each patient.

That's the typical way of delivering dialysis treatment. Having said that, there are also options for home dialysis treatment, and that can be both home hemodialysis. So patients can have a hemodialysis machine in their home and be trained or their partner is trained to deliver the dialysis treatment at home.

or peritoneal dialysis, which I mentioned earlier. So this is a type of dialysis where the goals are the same to filter the blood, but that's done rather than by directly accessing the bloodstream by infusing fluid into the abdominal cavity, letting that fluid dwell for a few hours. And then the dirty fluid, if you will, is drained out of the abdomen and clean fluid is placed back in.

And usually home dialysis therapies are done more frequently, in part because peritoneal dialysis is less efficient, but also because patients can have a better sense of well-being by more frequent treatments in some cases.

So both of these home modalities offer patients more flexibility and more choice in terms of the timing, whereas hemodialysis treatments, there's a schedule. And you come in at a certain time with a certain, you know, schedule for your treatments.

Great. So now let's go to your recent, you've published some papers recently that got some attention because you were kind of taking a critical look at how effective these things are for the patients, for their families. So can you tell me about that work and what motivated it and what you found?

Yeah. Maybe to start, I can tell you a little bit about how dialysis treatment evolved, if that's okay, and how we got to this point. Yeah. So I think your listeners are probably familiar with the idea that dialysis

Dialysis became widely accessible to patients with kidney failure in 1972 after Congress passed an amendment to the Medicare Act, which expanded the definition of who was considered disabled and therefore eligible for Medicare to patients with kidney failure. And following that, there was broad expansion of dialysis therapy in the U.S.,

So once this question of scarcity was solved by identifying a payment mechanism for dialysis, then critical clinical questions came to the forefront, like when should we start dialysis, how much dialysis, and who to treat.

And, you know, with the excitement around more access, there was this movement to start dialysis earlier and earlier in the course of kidney disease.

with the idea being that if we start dialysis when patients are not feeling as sick, they'll do better than if we wait until they develop more severe symptoms. And that's a reasonable hypothesis, of course. But the consequence of that is that a lot more people get started on dialysis, many of whom are older and frail and have other health conditions.

And so over time, we realized that there is this trade-off that we may be starting people on dialysis who might not really be benefiting from the therapy. And so that led to a series of studies, you know, really getting at this question of when to start and who to start.

So what clinicians are confronted with is, I'll take a hypothetical example to help your listeners sort of understand. Imagine we have Mr. Jones who comes to your clinic and he's 84.

and he has kidney failure. He has a GFR that's less than 12. GFR is one of the ways we measure the function, and it normally is around 100, so 12 would be very low. Yep, exactly. So he's got low kidney function, but he's also got other health conditions. He has advanced cognitive impairment. He has heart failure.

He's experiencing some symptoms. He feels short of breath on many days. His appetite is diminished. He's lost some weight. It could be his kidney failure, but it could be his other medical conditions as

And he's not decided, does he want dialysis or not? His family doesn't know whether it's the right thing to do. So how do we decide? What do we advise him? That was sort of the motivating question behind our study. What is the right thing to do for Mr. Jones? A patient like Mr. Jones is typically not eligible for a transplant.

Due to his anticipated life expectancy, he may not live long enough to reach the top of the waiting list and his other medical problems. So the question is when and whether to start dialysis for Mr. Jones. This is The Future of Everything with Russ Altman. More with Manju Tamora next.

Welcome back to The Future of Everything. I'm Russ Altman, and I'm speaking with Manju Tamura from Stanford University. In the last segment, we learned about the basics of kidney function, why they fail, and what are the treatment options. We learned that dialysis is a complicated process where when to start it and how long to do it is not always clear. And Manju was about to describe an exciting paper that she and colleagues recently published on the issue of when and if to use hemodialysis for kidney failure.

So I guess the first question is, how did you design the study? And then you can tell us what you found. Sure. So I'm a clinician at the Veterans Affairs in Palo Alto. And one of the resources, wonderful resources that we have as scientists at the VA is the VA's electronic health record, which has information on millions of patients and their health care encounters.

So using VA records, my colleague, Dr. Montez Rath and I designed a study where we emulated a clinical, a hypothetical clinical trial. And what that means is

Because the question we're trying to answer, some might say, might not be ethical to randomize patients to dialysis or no dialysis or even feasible. Right. And that makes sense. Yeah. So what we did is we emulated a trial using electronic health record data. And this is an approach put forth first by Miguel Hernan at Brigham and Women's Hospital. And

And so what we do is we outline a series of inclusion and exclusion criteria as if we were going to conduct a trial. And then we implement it using the electronic health records. So in this case, we first asked ourselves, who are the patients in which this question is clinically relevant? So we identified veterans who were over the age of 65.

who had advanced kidney failure, which we defined as having a GFR level less than 12, who hadn't received dialysis before and didn't have evidence of acute kidney damage. Then we assigned them, if you will, to start dialysis within 30 days versus wait at least 30 days. And then they could potentially start dialysis after those 30 days, which

And then we compared their outcomes. And we focused on two important- And just to be clear, this is actually not actually happening. You're not sending these- This is like a simulation. Is that a fair word? Yes, a simulation or an emulation. We took the records of patients who had treatments that followed those patterns. Yes. And then we compared the outcomes of these patients who started dialysis within 30 days

versus those that didn't. Right, and this is the advantage of having your medical record with millions of experiences. You're pretty good at knowing what will happen to somebody if you give them dialysis, so you almost don't have to really do it because you've seen so many patients who you have done it with that you have a pretty good model for what happens to them in the ensuing weeks and months. That's right. Okay. And so the other advantage of having the electronic health record is that we have lots of information about patients

not all, but many of the potential factors that might influence a clinician's decision to start dialysis versus wait. So that includes things like their laboratory records, their medications, their clinical conditions, whether they had been hospitalized in the past, et cetera, et cetera. So we can account for many of the things that we would say in epidemiology are confounders that might influence the decision.

And we looked at two outcomes, survival and the time that people spend at home. And we focused on the second because we

We saw that as an important factor for older patients, especially when they're making decisions about therapies that are potentially life-extending, is that dialysis can be life-extending, but it is potentially incredibly burdensome. It's a huge change to someone's lifestyle. Yes. So that was why we really, really honed in on the time that people spent at home.

So what we found is that over the course of three years, we followed people for three years. The group who was assigned or followed a path of waiting for dialysis for at least three years, about half of those patients ended up starting dialysis. Whereas by design, everybody in the early group started dialysis. Right.

And we found, we did two different types of analysis, which I'll talk about for a few minutes, an intention to treat analysis and

And then what we call a per-protocol analysis. In an intention to treat analysis, that means we analyze the patients in the groups that they were originally assigned to. And so we ignore the fact that the patients in the continued medical management arm, some of them did go on to start dialysis.

So in that analysis, we found that the people who started dialysis early lived on average longer by nine days. Wow. Yeah, but they spent two weeks fewer at home. So that number is...

A very modest extension of life advantage. That's right. Now, some might say, well, that's because a lot of the people in the control group, if you will, crossed over to dialysis. And that's a fair possibility. So that's why we did a per protocol analysis where we...

basically are comparing people who we censor people when they cross over. So we're not including their data after they cross over to dialysis. And in that analysis, we found that the people who started dialysis early lived longer by 77 days. So longer as we would expect, but they still spent about two weeks fewer at home.

So that's like the overall finding. And then we did some additional analyses where we tried to look at different subgroups to see if there were differences in the benefits of dialysis there.

by things like age or level of kidney function. And in fact, we did see that there were differences, though not always in the direction that we expected. So we did find that people with more advanced kidney failure had a larger survival benefit from dialysis. And that's as you would expect, of course.

But we also found a larger benefit in older versus younger people. And by younger, I mean the group that was 65 to 79 versus the group that was over 80.

And that struck people as a little unusual. And I would offer a few potential explanations. Yeah, so just to get this clear, because it's important, you found a more benefit of dialysis in the older population, even though one might expect that a 65 to 75-year-old is a little bit more healthy, a little bit more active, and would benefit from the dialysis more. That is not what you found. I just want to make sure I got that right. Yeah, so we found a larger benefit over a three-year horizon. Gotcha.

And I think there are a couple of reasons for this. First, we were looking over a relatively short horizon, only three years. And we did that because we were really focused on

short-term events because this is an older population. And it's quite possible that we would have found a different result if we had extended this to five or 10 years, say, for example. So that's an important distinction. The other important part about this is that the way that we estimate kidney function, the GFR or eGFR,

is not entirely precise. And so its accuracy probably is influenced by age. And so a GFR of 12 in an 80-year-old is probably not the same as a GFR of 12 in a 65-year-old. And so it may reflect a more advanced level of kidney failure for an 80-year-old. And so we might be comparing different stages of the disease.

And so that's why I would just caution the interpretation of that piece of the study. Did you look at things like side effects of either the kidney disease or the dialysis process? And were there any differences in kind of, I guess, what you would call the quality of life under the two different scenarios?

Yeah, this is a great question. So because we were using electronic health records, we didn't have information about quality of life, and we couldn't look at whether there were differences in quality of life between the group that started dialysis and the group that didn't receive dialysis. But I think that's a very important question.

That brings up one thing I want to mention here is the other thing. While we accounted for many different factors that might be different between the groups of people who start dialysis and those who don't, one big thing that we did not have information about is symptoms.

And symptoms we know plays a role in determining when someone starts dialysis. But that's just something that we can't get from an electronic health record, something that requires a prospective study. Gotcha. So this sounds like a big deal, especially because there are some non-intuitive findings. How did the professional colleagues and the world at large respond to the publication of your study?

I think it's generated a lot of great conversation there. The paper was accompanied by an editorial in the annals, which is wonderful. And this is the annals of internal medicine, which is one of the premier internal medicine journals. Yeah. And picked up by the New York times, which was exciting. And, you know, I think it's continued or brought to the forefront and important conversation for both clinicians and patients. And, and,

I think that crosstalk is always wonderful, even though sometimes in communicating complicated and sometimes non-intuitive results like ours, things can be simplified and some of the nuance can be lost. And that's where this is a starting point, I think, for patients to come back. What was this? If I may ask, what was the spin? What was the headline of the Times article? What did they take away from it?

I don't remember the exact spin, but I do think they often lead, they're intended to pull people into the conversation of, is this treatment beneficial? Right. And I can appreciate that, you know, the way to pull people into these conversations is to question conventional wisdom and

At the same time, we don't want to be alarmist and communicate the idea that dialysis is not beneficial for everyone. That's not at all what we're, you know, the message we want to send. More so, it's that these decisions that are clearly life-altering should be really made in advance whenever possible and with considered deliberation of people

You know, at that stage in someone's life, what truly are their goals? What can dialysis accomplish? And at what trade-off is that? And different people will have different answers to that question of whether those trade-offs are worth it for them.

Great. So in the final minute, I just wanted to ask you, let's say that we're now, clearly we are reconsidering our position and use of dialysis. And that sounds like a healthy thing in terms of giving nephrologists the information they need to make the best decisions. Are there new things coming down the pike, new alternatives, new treatments that might help the situation if we find out that one of the things we've been doing might not be the most useful thing in the world all the time? Yeah.

What's your kind of forecast for the future? Yeah, great question. I think there are a couple of different developments coming down the pike, and they span the spectrum of patients with kidney failure. So focused on the patient who may not be eligible for a transplant,

There is a lot of buzz in our field about expanding choice for these patients in terms of incorporating more palliative care, potentially giving patients access to hospice without having to decline dialysis. So what is called concurrent dialysis and hospice.

and what some call incremental start of dialysis. So we talked about the fact that dialysis is often three times a week, and some are testing. There are now at least two trials that I'm aware of testing start of twice a week dialysis as a way to reduce the burden on patients when they're starting. So those are developments on the supportive palliative care spectrum.

And then on the other end of the spectrum, there are also some really exciting developments. 2024 marked the first pig xenotransplantation in two patients, one at the MGH and one at NYU Langone. So that's a huge milestone for our field, and certainly more is coming on that front. And then there's progress being made all the time on what's known as a wearable artificial kidney, which is a kidney that is

which is an implantable kidney that patients may have that allows for continuous dialysis. Wow. And then in addition to all that, you don't mention it, but I know because I've looked at some of your recent papers that preventing kidney failure by treating the underlying causes and trying to keep as many people away from kidney failure as possible would be a fourth arm to this approach. Yeah, absolutely.

Thanks to Manju Tamora. That was the future of kidney dialysis. Thanks for tuning into this episode. With over 250 other episodes in our back catalog, you can listen to a wide variety of topics on the future of everything.

Please remember to hit follow on whatever app you're listening to so you always get notified of the new episodes and you never miss the future of anything. You can connect with me on social media like Blue Sky, Mastodon, and Threads at RB Altman or at Russ B. Altman. And you can follow Stanford Engineering at StanfordENG.