Well, I want to thank the organizers for inviting me. I think I was looking back as I was putting the slides together for this talk, and I had a folder on my computer that has SCCM, and I was looking back to see all the talks that have the word AKI in it, and it's either 37 or 38, I'm not entirely sure. So I feel like... I guess that means I'm old, but whatever. So I think it's nice to see that there is a bit of evolution of how acute kidney injury gets presented at meetings like this, and I think it's good that the AKI story is sort of in separate... You know, this is a talk about how we can optimize cardiac surgery patients, and I think it's important to have AKI be part of that, because, as you'll hear, it's a major aspect of patient well-being. These are my disclosures. I'm not going to talk about anything that Spectral Medical does, but I am going to mention a biomarker marketed by BioMaru toward the end of the talk. The learning objectives are here. I'm going to briefly talk about AKI definitions and implications for cardiac surgery. I'm going to trust that everyone in this room is aware of the definitions and the importance, but I'll touch on it. And I'll speak a little bit about what might be different about cardiac surgery associated with AKI. I'm going to talk about what a Kedegel bundle is for AKI and how it works, and I'm going to talk about how to identify high-risk patients. So I'm going to start... This is a session on technological advancements, and so although none of the things that I'm going to speak about are rocket science, I think we have to start with just saying, look, can we use technology to really facilitate the recognition of acute kidney injury based on Kedegel definitions? And so as crude as these definitions are, and I'm hoping that in the next few years they will improve with the addition of biomarkers, etc., right now it's urine output and creatinine. But you'd be surprised, or maybe you wouldn't be surprised, just how often acute kidney injury, particularly in its early stages, is missed in the cardiothoracic surgery population but just in the hospital in general because some of the tools that should be available to us are not. And one way to sort of tackle that is to use computer electronic medical record decision support. And you might say, well, gosh, I already have... When the creatinine is abnormal, it already turns red in my EMR. That's not what I'm talking about. The biggest problem that we have identifying acute kidney injury on the basis of creatinine is that we have a hard time figuring out what a patient's baseline creatinine is, right? And this happens, by the way, and we should be very sensitive to this, this happens more often in women than it does men. A typical female who's 40 years old might have a creatinine of 0.7. If the creatinine in the EMR is 1.4, it might not even be indicated as being abnormal, OK? And yet the creatinine is doubled, OK? A male, same age, might have a creatinine of 1, so a 1.4 might not have as much importance in terms of defining acute kidney injury or impacting outcomes. But this is a study we did where, over a decade ago, we engineered an electronic medical record support that primarily the secret sauce for this is really it just goes back and pulls all the creatinine in the medical record, inpatient, outpatient, does a little bit of analytics and serves that information up to the clinician so that you don't have to spend the 10 or 15 or 20 minutes hunting through the EMR to try to figure out what the right baseline creatinine is. And then if the patient doesn't have a baseline creatinine, the computer does a little bit of baseline analytics to sort of use the demographics from the patient, primarily age and sex, to sort of figure out what that creatinine should have been if it doesn't appear in the medical record. And so this is sort of what that alert sort of looks like. And this is the result on mortality. This is on mortality. Now, this is a before and after study. We didn't randomize patients. This is in half a million patients across a 20-hospital medical system in southwest Pennsylvania. So we weren't going to randomize these patients. We weren't going to do a cluster randomization. This is a before and after study. But you can see that when we initiated this system, mortality, all-cause mortality for patients with acute kidney injury decreased significantly. Now, there's not a huge decrease here, but it decreased significantly. This is not just cardiac surgery. This includes cardiac surgery. But this is all patients in the hospital. And you can see there's no change in outcome for patients without AKI. You could also appreciate, I hope, that there's a massive difference between patients who develop acute kidney injury in terms of hospital mortality versus those patients who don't develop acute kidney injury. There was also an effect on hospital length of stay. This decreased by about a third of a day. And there was a major decrease, although the numbers are small to begin with, in the rates of renal replacement therapy. Now, this is a small effect, but if it were generalizable across the entire United States, it would actually save about 17,000 patients a year and save $1.2 billion annually if it were generalizable across the entire... Now, I don't know if it is generalizable, but we ran it on 20 hospitals in our system, and it works. We've done follow-up studies showing that mortality continues to come down for the acute kidney injury population, and it doesn't change for the patients without AKI. Your hospital should do this. It's not hard. It required about a year of lobbying by me to get it done, but it's not hard. What do we do? What do clinicians do inside and outside the cardiothoracic ICU? But cardiac surgeons and intensivists and anesthesiologists working in the cardiothoracic ICU are the biggest offenders of using nephrotoxic antibiotics. Not to say that we don't often use those drugs because we have to, but still the use of vancomycin in our cardiothoracic surgery ICU, it has dropped now from about 70% of all antibiotics used in the cardiothoracic ICU to only about 45%, but it's still 45% of anyone who gets started in antibiotics gets at least one dose of vancomycin. And guess what the largest association between acute kidney injury and any identifiable modifiable risk factor in our cardiothoracic ICU is? Yes, vancomycin. Pipercil and tazobactam isn't much better, and there is some evidence to suggest the combination of the two drive even higher rates of AKI. By identifying AKI early in these patients, there is a statistically significant decrease Other drugs that are even more controversial, such as ACEs and ARBs, no doubt associated with benefit in patients, but there are risks associated with these drugs in regard to acute kidney injury. What about saline? What about saline? It turns out that there's now very good evidence, again, not just in the cardiothoracic ICU, but holding up for sub-analysis in cardiac surgery patients that suggest that there is a small but consistent effect on major adverse kidney events, which is death, dialysis, and persistent renal dysfunction, about a 1% absolute risk reduction when you move from saline to a balanced crystalloid solution. And that's also true for patients just admitted to the wards. Now, you might argue, well, haven't there been other studies published more recently that have failed to detect this? But when you put all these studies together in a meta-analysis, you can see that actually there's a consistent, though not quite statistically significant at the traditional level, effect on all-cause mortality. The preponderance of the evidence at this point suggests that avoiding saline is associated with improvement in survival. Now, most of this stuff was actually in the KDGO bundle. So this is the KDGO bundle. A little history about this. Some people know about this, some people don't. But the history of this was I dreamt this up in the... I was sitting in the audience, like, somewhere around there, as we were at the KDGO, and I wrote this down and I presented. I was co-chairing the KDGO meeting that concluded with the guideline in 2012. And this was the most hated... This was the most hated suggestion. People said, there's no evidence for any of this stuff. There's no evidence for any of this stuff. And then people said, it's all just common sense. And what I really wanted was I wanted a list of things that we could put on a little card and the interns could carry it around, OK? Because you would be amazed just how often... You'd be amazed on the low end how infrequent people don't pay attention to these issues. And you know, because these patients come to the ICU and the first thing you do is say, well, I have acute kidney injury, and you look through their med list and you see they're on two nephrotoxic drugs, neither of which the patient really needs, and you wonder, why didn't this stuff get stopped earlier? And it goes on from there. You already heard about functioning hemodynamic monitoring and this is a very, very solid study showing that when you do the KDGO bundle, which includes optimization of volume status and hemodynamics, but also avoiding nephrotoxins and hyperglycemia, it is associated... This is the functional hemodynamic monitoring. And at this point, we could argue about this, maybe at this point of discussion, I don't necessarily care what your functional hemodynamic monitoring protocol is, but if you utilize one, you have better outcomes for the kidney and possibly for the patient in general than if you don't. And this is what Alex Arbach used. And when you combine all of this stuff, it's associated with a substantial reduction in acute kidney injury. Now, wait a minute. You might say, well, wait a minute. Why do you have a rate of acute kidney injury that's so high? And that's what's so important here is you have to say, well, this is not all patients undergoing cardiac surgery. This is the subset of patients who were biomarker positive. So Alex used a biomarker, he used TINP2IGFPP7, it's commercially marketed as nephrocheck. He studied patients first by measuring the biomarker, said, okay, now this is the high-risk population, I'm now going to use a functional hemodynamic monitoring approach along with the rest of the bundle and resulted in a significant decrease in AKI. Ivan Gosch, in general surgery, showed very similar results. The primary endpoint here was length of stay, but as a secondary endpoint, he also showed a substantial reduction in stage 2-3 AKI. Dan Engelman from Bay State has practically abolished severe acute kidney injury by doing exactly the same thing, using a biomarker to identify patients with acute kidney injury and then bringing in a SWAT team to essentially administer a bundle of care to reduce acute kidney injury. So with that, I thank you for your attention. I think together we can stop kidney attack. And that should be our goal. Thank you.

acute kidney injury

cardiac surgery outcomes

technology for AKI recognition

electronic medical record decision support

AKI prevention