Thank you very much, and I appreciate the opportunity to speak to a full room. It's exciting for all of us who do AKI stuff. So just a couple of minor disclosures, but as you heard, I'm a pediatrician, which is not to say I'm going to be talking about playing nice in the playground in the sandbox, but you never know. Dr. Shishadi, as an intensivist, probably gave you the impression that he likes to keep the lungs dry or that kind of perspective. Dr. Prada, as a nephrologist, likes to keep things wet. I might take the dermatologist approach and use steroids or a punch biopsy, we'll see. So what is a nexus? This term that is the title for this session, it really is a connection or a link between people and things, and it can be causal, associative, or formal. So my talk has a bunch of slides that you've probably seen already in the first two speakers, but that's intentional because I'm supposed to be the link between them. So the vision for integrated care really is about why are we doing this, who, what, how, when, and where, and I'll get to each of these. So you heard some of this data before, that the concomitant injuries of AKI and ALI are quite bad for you. These are pulled from data that exists in the literature that you could probably pull lots of data sets from, but what we know is that lung and kidney injury occur frequently together and they synergistically worsen outcome. This is not a surprise, right? But you have to ask yourself why. And we heard already, well, maybe it's about the fluid, and if you were to believe the ARDSnet study, which is now 15 years old, you would say to yourself, well, yeah, the proof is there. It's because if you are volume overloaded and you have ARDS, you do worse, you have pulmonary edema, but it's got to be more than that, right? And the truth is there is more to it than just pulmonary edema. If you look at the cellular mechanisms involved, what we know in models of AKI and ALI is completely aberrant sodium channel signaling in both organ beds, a systemic inflammatory response, and as you heard, massive apoptosis in both organ systems. And this has been proven time and time again. This article that I pulled from a journal that I've never read before, the Chinese Journal of Traumatology, indicates that there's a nephrogenic ARDS. Now, this is probably a term you've never heard before, but if you're looking for something more than just, oh, it's wet lungs and pulmonary edema, there's actually quite a bit of basic science data about the exocrine and paracrine effects of AKI, the exocrine and paracrine effects of ALI, and they come nicely with figures that demonstrate the systemic inflammatory mediation of both when you have them in isolation. But I draw your attention to the histology here, and maybe nobody really wants to look at histology, but we're going to. So if you see the two x-rays, they both are demonstrating what we would call at the bedside pulmonary edema. But the middle histogram, or the middle slide of the lungs, shows evidence of some fluid accumulation denoted by the red arrows, but inside F, you see proteination material in the alveoli, and you also have a blowup shown in G of neutrophilic infiltration in the pulmonary capillary beds. So this is different, right? So the difference really is the bottom panel are patients who have kidney injury, and the top patients have pulmonary edema from non-kidney injury sources. So there is a difference, not just from fluid, but from inflammation as well. You saw this picture before, but I draw it to your attention again, and this is a good read, and maybe you just notice the reference. It's in the Blue Journal now seven years ago, or eight years ago, but it talks to this bi-directional modality of AKI and ALI. So why would we do this? It's not just because patients are doing poorly. It's because we have unseen blind spots, and as my first boss and mentor, Hector Wong, would say, if you don't know the biology, you don't know anything. And so you have to understand the importance of renal dysthoxia, endotheliopathies, nephrogenic pulmonary edema, and the dual paracrine inflammation that occurs when you have AKI and ALI together. So who do we do this for? Well, this is the problem, right? The picture that you see that we described all too often is bad. This is probably too late. So it's not necessarily I'm up here telling you how to manage this patient. It's about something else. This is a great book, if you've never read it, by Nate Silver, about signal and noise. Why do I bring this up? It's because we're always trying to find stuff, right? If you look at pictures of the solar system, people are trying to find galaxies, trying to find planets where there's light. How do they do this? How do they find this? It's not just some random thing. I point to a model that we try to look at in the world of kidney injury, for sure, but also exists in sepsis and lung injury. This gentleman's Thomas Dauber. The name might be familiar, but what should be familiar are the Framingham criteria. He was an epidemiologist in Massachusetts who got hired to figure out why there was so much syphilis. True story. Small community, isolated, there's lots of syphilis. I don't know that it was a real mystery. But along the way, he was able to gather epidemiological data and found, well, a lot of these people have heart disease, too, and I don't know if the two go hand in hand. But what came from that is what we use in real time to manage acute coronary syndrome. That's really important because we're treating them before their heart is dead. Now, there are models for risk prognostication and risk identification in lung injury. There's one of them. It's called the LIPS study, and it's a model that includes a number of different kind of criteria, and you add it up. You can do it at the bedside, and as we heard this morning, if any of you were at the Phoenix criteria for the sepsis definitions, the intention is not to have something that you cannot use, not something with coefficients and regression analysis. You should be able to do it at the bedside, right? So lung injury, risk stratification, you can probably guess that somebody probably needs that because they're panting a little bit, and it correlates with outcome. If you have this LIPS score. Heart attacks, too, right? Somebody shows up without a shirt on and red arrows are reading out, you're having a heart attack. But there's a problem with AKI. People don't show up to the ICU looking like this, thankfully. Thankfully not to my ICU. But we believe that there's a prodrome that matches this risk stratification system used in kidney injury, and we call it the renal angina index. It's a marriage of epidemiologic risk factors based on who you are and early signs of kidney dysfunction. It's pretty easy to calculate. And what it does is it changes the paradigm from somebody who has overt injury, and it moves the needle earlier to risk and the high-risk populations, and this is really important because now over 15,000 adult and pediatric patients, we've discovered that this system is able to identify who will have acute kidney injury three days from now. So today's Sunday. It's 4 o'clock. You guys all need a beverage. We're talking about a patient comes in now, you'd be able to determine if that patient has severe AKI on Wednesday. So that's a big deal, right? For all the intensives in the room, what are we going to do with that? If you're thinking it's not a thing, it's a thing. It's on Wikipedia. It's there because I put it there. So who is it? Who's the question? The first one is you have to be using risk stratification systems. Use the things that work for you in your place, but you have to because it's proactive and it gives you the window of opportunity before they're attached to a ton of machines. The second book, The Bomber Mafia, another great book. This is about the rise of the U.S. Air Force during World War II, and their motto was profissimus mora irritante, which means we make progress unhindered by custom. The military apparatus was wasting money, ammo, machines. People were dying because they just bombed randomly. The Bomber Mafia were a group of mavericks who used this thing called the Norden bombsight to focus their efforts on certain populations. And so we said, okay, can we apply this? So if you look through the literature, you'll find that their ventilator-induced kidney injury leads to systemic mediator release. And these things are things you can measure. These are primary signals that a lot of labs have that you can measure at the bedside. Now some people don't have, but maybe they will soon, that can be measured in urine. And if you haven't heard of VIKI, you're not alone. I'm not sure that I had heard about it real recently, but this is a thing that we need to be thinking about. You can also use machines, right? So machine learning to properly identify respiratory failure, imminent respiratory failure, or future respiratory failure. And in this precision recall curve, where the better is toward the top right, you can see that the blue precision model was better than the threshold model in both the cohorts that they studied. We are looking primarily in children also about sub-phenotyping for kidney injury using biomarker arrays. So this is being studied in adult populations, but also in children, but I think this is the future for all of what we're trying to do in precision medicine, is to take this very heterogeneous cohort of patients, and instead of just blanketly giving them bicarb, blanketly giving them a ventilator, blanketly giving them intermittent hemodialysis, you have a better identification of their phenotype, and we could do this with the biomarker arrays. The basics still apply though, right? If you don't like numbers, like I don't, you look at visuals and you can understand urine output, that's important. You can also use all the data that comes out of your system in the EMR to program in other machine learning models. This is one that was recently published about machine learning in kidney injury and being predictive of performance. Are there biomarkers that could help you do both? Yes, there are. So the future, in this decade, I'm predicting that we will have access to a number of them, and one of the big majorly studied kidney injury biomarkers, Engalogist got FDA approval. I'm not getting paid to say that, but I'm just telling you, these are things that we will have at our disposal to improve patient care, and as you see, you can actually modify the patient. So what should we be using? We should be using biomarkers, the VIKI, data architecture and machine learning platforms and diagnostics to help us integrate care. How do we think again? How do we actually go to the null hypothesis and say, maybe we were wrong, maybe we need to consider new things? Our paradigm is that we treat ALI and ARDS with the ventilator. We treat AKI maybe with kidney support therapies, but really it's different, right? We can modulate the ventilator, excuse me, modulate the ventilator to address kidney dysfunction and also vice versa. So we should be thinking in new ways. This is a study that Dr. Prada actually probably mentioned here in the sense of the causal inference between ventilator manipulation and AKI, and he alluded to this, and the authors found very clearly that modulation of PEEP and changes in the respiratory compliance system were associated with up and down in AKI. So this is something that we should be thinking about. In ARDS strategy, the strategy that people use of high PEEP or tidal volume can be deleterious for AKI. If you use lower PEEP and high tidal volume, that's bad for a lung injury. However, there's got to be a Goldilocks that you're thinking about on a regular basis to treat both. So when we talked about, Dr. Prada mentioned precision management for RRT, well, the benefit of this is not just, oh, I'm going to use RRT for fluid removal. It's how you do it. So when Shashadi calls Prada and they don't choke each other out and they say, how are we going to do this? The dialogue is really about how do you follow your filtration fraction? How do you follow your ultrafiltration? So when we think about renal replacement therapy in the ICU, it's either going to be a form of renal place continuous therapy or intermittent hemo. But there are differences between the two in the sense of how they manipulate your fluid regulation or removal. So filtration fraction is something that you can definitely pay attention to and modulate even as an intensivist to help your solute clearance, to help your fluid removal. But we don't often talk about that at the bedside, but we should be talking about that. That is something we can do. And if you're thinking about, well, why would you do that or how would you do that? It's a simple manipulation of filtration, pre-filter fluid, and post-filter replacement fluid to modulate your solute clearance. This is something that is kind of second nature to nephrologists in some ways, but it's something that we can do very clearly in the sense of if you're trying to remove fluid or regulate solute clearance. So precision is important. And it's important that you do this not just at 8 in the morning for five minutes when you run on a patient, then you come back the next day. So it has to be done repeatedly. And this is a very important concept because this dynamism that we're speaking of requires attention through the day and not just once a day. So what you're seeing here is a graph that shows the patient moving, the machine moving towards each other for goals. And again, this comes from the ADQI initiatives that Dr. Prada mentioned. There are some devices out there to do combined pulmonary and renal support. This is a pretty old report talking about a multi-filter pressure machine that has gas transfer efficiency for oxygen and decarboxylation. The modern versions of this include ECMO, so VV and VA, which I'm not really getting into the specifics of how you would do that. But again, that's my point. We're talking about earlier management versus this complex machine-oriented systems. And ECMO comes with its share of problems for both injury beds. ECOR, which is extracorporeal CO2 removal, used extensively in Europe, not as much in the United States. But this is essentially renal support therapy for the lungs. You are taking the pressure off the lungs. You're decarboxylating the blood through an external filter and replacing the patient. And the things that we think about in those patients that are on ECOR therapy include the hemolysis index and what we need to be doing for ventilatory protection. This paper is still under review, but I show this anyway, hemoabsorptive therapies. So coming to you also are a series of hemo filters that can be used to modulate the inflammatory response, modulate the neutrophilic response, the leukocytic response. And that's really important. So we should be using machines, protective ventilation, not just renal replacement therapy. It should be precision and patient-focused. The last part's pretty quick. This is a model about fluid, but it really is about our patients. We have a rescue phase, we have a maintenance phase, and a removal phase. Well, when we're talking about rapid responses, we have emergency responses to codes. They are trialing rapid responses to AKI. This is a study in Italy about a rapid response team for nephrology for earlier assessment of fluid needs and regulation of renal replacement therapy. But it gets to the same point. You have to have people that are paying attention to this before the problems really manifest themselves in an irreversible way. And when you do proper solute and fluid replacement therapy, you can get to a point where you're really at the bedside in the unit looking at quality measures of your renal replacement therapy. It's not just a system of on or off. It's about how well are you delivering the dose, how much downtime do you have, do you have a lot of filter changes, those kind of things. How do you modulate your anticoagulation? So fluid stewardship is the last part. It is a big thing. But we need to involve the people that really make a difference. NINJA is a nephrotoxic medication associated program that reduces nephrotoxic exposure in AKI driven by pharmacists. Fluid, I think, is the paradigm of the nursing staff. And really, we have to involve all of our multi-professional colleagues. We're talking about, oh, intensivists and nephrologists. No, no, no. Integrated multidisciplinary care involves the whole team. And without the whole team, we can't do what's right for the patient. And when we do that, we can have real-time markers of blood gas mirrored by what I call an AKI biomarker composite. So when should we integrate? Early, rapidly, and in real time. Remember the patient. And yes, I didn't say I was going to say sandbox, but I am a pediatrician. The concepts that we use in pediatrics are being proactive, preventative, integrating people and care and modalities and using things in real time. So where is the patient? In the center, involved, and everything like that. So I will end because we're up against time. But the nexus of care has to do a lot with who, what, when, where, how, and where. Thank you so much. Thank you.

Acute Kidney Injury

Acute Lung Injury

Renal Angina Index

multidisciplinary approach

proactive management