Good morning, everyone, and thank you for joining us today. For my first talk, I will be discussing acute kidney injury diagnosis and management. My name is Anita Vajayan, and I'm a professor of medicine in the Division of Nephrology at Washington University in St. Louis and the medical director for acute dialysis at Barnes-Jewish Hospital. These are my disclosures, and the one relevant to this talk would be Astute, which makes biomarkers. So I will review the epidemiology and definition of acute kidney injury, review the diagnoses and risk factors, go over management and prevention. I will touch upon the biomarkers, the newer biomarkers that have been available for the last few years, and then my next talk will be a renal replacement therapy. So acute kidney injury is a huge global issue. It's not just in the developed countries, but in the developing countries for different reasons. Acute kidney injury is a huge problem. It adds to probably $9 billion a year to hospital costs, healthcare costs in the countries, and in the U.S., about 1.2 million people probably get acute kidney injury during a hospital stay, and about 300,000 might die. Your risk of, your length of hospital stay goes up by about four days, and your odds of death are significantly higher if you get acute kidney injury versus not getting acute kidney injury, and AKI, you know, mortality is much higher than those related to other chronic conditions or some cancers. So how do we define acute kidney injury? So since 2004, we have some standardized definitions, and these have evolved over the years. So the original one was risk, injury, failure, loss, ESRD, or RIFLE criteria. This changed to acute kidney injury network criteria in 2007, and now we have the KDGO, which is Kidney Disease Improving Global Outcomes, which is a consortium of intensivists and AKI, nephrologists interested in AKI, and their definition came out in 2012. It has a serum creatinine as well as a urine output criteria. Urine output, while easy to track in the ICU, it's extremely challenging in the floors, but it is part of the definition of acute kidney injury, and we probably should be paying much more attention to the urine output than we do at this time. So rise in serum creatinine by 1.5 to 1.9 from the baseline over the course of seven days, or a rise in serum creatinine by 0.3 is stage 1 AKI, and the worse the rise in serum creatinine or the worse the urine output, then it goes from stage 1 to 2 to 3, and the worse the stage, the worse the prognosis for these patients. As the definitions change, so has the incidence of acute kidney injury in the ICU, and it basically has to do with the fact that prior to any of these definitions being available or standardized, researchers had varying definitions for acute kidney injury, and they also probably had more stringent definitions of acute kidney injury where the rise in serum creatinine had to be greater than 3.5 for it to be called AKI, and so the incidence was, anyway, 6% to 7%, but after these newer definitions came about, the incidence has been noted to be anywhere from 12% or 11% to 67%, so still a wide variety. And this is from the International Conference reviewing studies from across the world with varying definitions used, or I shouldn't say varying, with the definitions used in Akin, Rifle, and KDGO, and if you just look at the KDGO urine output and creatinine criteria, the incidence of AKI is anywhere from 55% to 74%. So why do we worry about acute kidney injury? So acute kidney injury has systemic effects, if I might say that. So animal studies have shown that just by clamping renal arteries and inducing AKI, you can see neutrophil trafficking in the lungs within hours of the damage happening in the kidney, and so if you look at our patients in the ICU or on the floor who develop acute kidney injury, they start having respiratory distress or respiratory failure, they start having mental status changes, maybe GI bleeding, myocardial dysfunction, etc., and so it really is a systemic effect that we are seeing from renal damage. Acute kidney injury accelerates the rate of progression to end-stage renal disease, so if you have somebody in, or if I have somebody in my clinic with stage 3 CKD who is doing fairly well in the outpatient setting, if they come into the hospital and develop acute kidney injury, for whatever reason, that rate of progression to ESRD is clearly accelerated, and this has been shown in other studies as well. So what constitutes acute kidney injury? So kidney is an extremely complicated organ, and if you remember, the cortex is where the glomeruli are, and this is where most of the blood supply goes to, right? 25% of the cardiac output goes to the kidneys, but most of it is going through the cortex. So the medulla is kept relatively hypoxic, and this is where the tubules are, specifically the proximal tubules and loop of Henle, and this is where a lot of the damage will happen but having said that, you know, we shouldn't forget the glomerular processes that, or glomerular injury that can occur, or even the interstitial nephritis that can occur in some patients as well. So this is a patient we had in the ICU with vasculitis that we ended up biopsying while the patient was still on a ventilator in the ICU and turned out to have anconegative vasculitis. This is a patient who has interstitial nephritis, either from medications or sarcoidosis. So glomeruli and interstitium can still become damaged, and this is something that we should look out for in our patients if things are not fitting into a, you know, neat picture as to why they have an AKI, because these patients will respond to steroids or other immunosuppression. So the classic AKI syndromes that we see in the ICU, we have the pre-renal, the volume depletion, not as common in the ICU as we once thought it was, maybe more on the patients coming into the hospital from home or nursing home. We can also have obstruction, right? And this is more common outside the ICU unless your Foley gets blocked. So most of it has to do with multifactorial injury to the tubules, either mediated through sepsis, so sepsis can lead to inflammation, which can affect renal blood flow, renal perfusion, cause damage to the tubules, and maybe even some glomerular disease as well. We can have various toxins, nephrotoxins, that lead to acute kidney injury, primarily tubular injury. Cardiorenal syndrome starts out with severe heart failure and affecting renal blood flow, renal perfusion. Hepatorenal syndrome, again, in the setting of severe liver failure. We are seeing this more and more in the ICU as younger patients are getting alcoholic liver disease. And one of the most common causes of acute kidney injury in the ICU is related to cardiac surgery. Again, maybe the bypass and also the nephrotoxins and hypotension associated with cardiac surgery leading to tubular injury. So how do we prevent acute kidney injury? So volume resuscitation is key, right? So we do need to volume resuscitate early if the patient is hypotensive, volume depleted, and deemed to be responding to fluids. So we know volume overload is associated with adverse outcomes, so volume resuscitation has to be done very judiciously. Recent studies have shown that early goal-directed therapy in sepsis is not as beneficial as once thought, and volume resuscitation should be reserved for patients who are hypotensive and deemed to be volume responsive. Normal saline used to be given very commonly in the emergency room in the ICU, but it should be considered as a drug, and we know normal saline is what we call abnormal saline because of the high sodium and chloride content, and this can further affect renal perfusion and maybe lead to metabolic acidosis. However, generally, crystalloids are superior to colloids for volume resuscitation, except for intravenous IV albumin in patients with hepatorenal syndrome, primarily for those patients who need large-volume paracentesis. If we're going to give crystalloids, then balanced crystalloids are preferred over normal saline. So just to touch upon fluid overload again, fluid overload has systemic manifestations in every organ possible. What we always think about is pulmonary edema and respiratory failure, but it can have gut edema leading to poor absorption of medications. You can have myocardial edema, and also, interestingly, something that we fail to recognize even as a nephrologist is increased renal venous pressure, increased renal interstitial edema, further decreasing renal perfusion, and causing worsening AKI. And of course, from the surgery standpoint, poor wound healing is a major concern. So it is really a balancing act between volume resuscitation and avoiding fluid overload. So this is the PRISM investigators, which looked at early goal-directed therapy versus usual care in patients in the ICU. And essentially, there was no difference in renal replacement therapy or mortality in patients who received early goal-directed therapy versus usual care, so something to keep in mind. Looking at fluids, these studies that have compared head of starch to crystalloids have shown that patients are at higher risk for developing AKI and needing renal replacement therapy with head of starch. So what about balanced crystalloids? So the SMAR trial and the SALTED trials were both done at Vanderbilt as pragmatic cluster randomized crossover trials. So the SMART was done in all the ICUs, and it included 15,000 patients, and they compared normal saline to either lactate ringers or plasma-lite, and the primary endpoint was major adverse kidney events. And if you compare the balanced crystalloids to normal saline, there was a decrease in major adverse kidney event within 30 days in the balanced solutions group. And renal replacement therapy free days was also lower in the balanced crystalloid solutions. There was no difference in hospital death or receipt of neonatal replacement therapy, although it was trending towards benefiting balanced solutions. The SALTed trial, similar approach, but this time the patients were in the emergency room. And again, this time the primary endpoint was hospital free days, and secondary endpoint was make. And here also the major adverse kidney events within 30 days was better in the balanced crystalloid solution. So we know that hypervolemia and hypovolemia will increase the risk for acute kidney injury. So if the patient is hypovolemic, we have to volume resuscitate them, preferably with buffered crystalloids or maybe albumin in the appropriate patients. And if the patient is fluid overloaded, then we have to consider appropriate use of loop diuretics and maybe renal replacement therapy. Once the patient gets to a euvolemic status, then serial assessment for clinical signs of hypoperfusion is necessary. And you can give aliquots of crystalloid fluids if there's evidence of fluid responsiveness in the setting of hypotension. If there's no evidence of fluid responsiveness, it's important that we discontinue any further fluids as it can lead to hypervolemia. So to summarize, when you're assessing patient's hemodynamic stability, we have to go through the salvage phase where you give volume resuscitation if the patient is unstable. Then we optimize where the patients have improved organ perfusion and we're maintaining cardiac output and mean arterial pressures. Then we go to the stabilization and then de-escalation phase. And de-escalation phase is if the patient is very fluid overloaded and is not responding any more to volume. So the K-DIGO Bundle, which was put together in 2012, talks about an approach to patients with acute kidney injury who are deemed high risk and who progress through the various stages of acute kidney injury, stage one, two, and three. So we start out by ensuring appropriate volume status, hemodynamic, we make sure they're being monitored appropriately for blood pressure, and then we go to the stabilization phase for blood pressure and perfusion pressure. We monitor their renal function, we avoid hyperglycemia, and then we avoid radiocontrast procedures if possible. And as they go through, if those fails and the patient develops stage one, two, and three AKI, we go through additional workup, changes in drug dosing, and consideration for renal replacement therapy as patients get to stage three. So is there any way to know that we can prevent AKI by doing some of the measures that they discussed in the K-DIGO, what they call the K-DIGO Bundle? So the main thing is avoiding nephrotoxins, avoiding hyperglycemia, and optimizing volume status and hemodynamics. And this is a study, a single-center study from Germany by Melanie Marsh and Alexander Zerbach that looked at implementing these particular K-DIGO guidelines in patients with CT surgery. So they had about 1,000, they screened about 1,000 cardiac surgery patients, and they randomized 276 patients to control group and intervention group, and they used a biomarker, the nephrocheck biomarker, and if that was positive, the patients were randomized to these groups. And as you can see with the implementation of the biomarker, I'm sorry, but the implementation of these K-DIGO interventions in the biomarker-positive group, the incidence of acute kidney injury was significantly reduced compared to the control arm. So the take-home message is that implementation of these particular K-DIGO recommendations in high-risk patients who are biomarker-positive may reduce the frequency of acute kidney injury. Now, this was a single-center study, it's a relatively small study, and this will need to be validated. So let's talk about which particular nephrotoxins, and one of the commonest drugs used in the hospital is vancomycin, and in this meta-analysis of 15 trials, the incidence of vancomycin-associated acute kidney injury ranged anywhere from 5% to 43%. And as you can see by this figure, the higher the trough levels, the higher the risk of acute kidney injury. Definitely anything greater than 20 seems to increase the risk significantly. Most cases are reversible, but 3% of these patients required renal replacement therapy. And so the current recommendation is to follow the trough levels and keep it between 15 and 20. However, just last year, the pharmacology, pharmacy groups got together, infectious diseases pharmacy groups got together, and there is a recommendation to monitor AUC as opposed to the trough level, but that's very difficult to implement, has not been implemented in most institutions as yet. So right now, we're still following the vanc trough levels. What about intravenous contrast? So there's a lot of controversy on whether intravenous contrast is nephrotoxic or not. We know that the initial concern that we all had about it being extremely nephrotoxic is probably not real. However, in the high-risk patients, there's still potential for iodinated contrast given intravascularly to create an additional injury to the kidneys. And so it should be used judiciously. And this is the statement, a consensus statement from American College of Radiology and the National Kidney Foundation that for patients who have an EGFR less than 30, we recommend prophylaxis with intravenous normal saline. And in patients who are considered high-risk, so some of the ICU patients, you may wanna give intravenous fluids even with GFRs in the 30 to 44 range. So what we call stage three CKD. So the recommendation for normal saline is based on the PRESERVE trial. So PRESERVE trial compared normal saline to a bicarbonate and acetylcysteine to placebo. So it was a two by two factorial study. And basically the main thing that benefited was, or normal saline was just as good as giving sodium bicarbonate. And it's recommended for normal saline to be given few hours before and after procedure. So how about diagnosis of urine, I'm sorry, diagnosis of acute kidney injury. So it is recommended that urine microscopy be performed by an experienced physician. And most of the time, this will be a nephrologist. And why should we look at the urine microscopy? Well, we can look for cells, casts, crystals, et cetera. And urine microscopy is making a comeback because it's essentially what we would call a liquid biopsy, right? Most of these patients in the ICUs are not undergoing a kidney biopsy because of their high risk for complications. And so therefore, we should be considering at least a urine microscopy to see what could be happening in the kidney. So this is the typical finding in most of our ICU patients, muddy brown granular cast as the tubular epithelial cells are sloughed off and send down the tubules and into the urinary tract. However, if patients, if in some of the patients, you may find other things. So this is a courtesy of Dr. Seltzer who does a lot of urine microscopy. And as you can see, this patient has red blood cell cast, which immediately will make you suspicious for glomerulonephritis. And this is a patient we had a while ago who had acute kidney injury. And when we looked at the urine, had a lot of these crystals and it turned out to be indenovir crystals. So urine microscopy does give us a lot of answers and we should be performing it in every patient with acute kidney injury. Furosemide stress test. So in patients, this is a diagnostic test that was proposed by Dr. Jay Coyner at University of Chicago. And essentially patients with early stage one or two acute kidney injury is given one milligram per kilogram IV furosemide and the urine output is monitored over the next two hours. And urine output of less than 200 ml predicted progressive acute kidney injury with a sensitivity of 87%, a specificity of 84%. And so essentially this is a good test to determine if the patient might need renal replacement therapy shortly. And if the patient already has acute kidney injury and there's no acute indication for starting renal replacement therapy such as metabolic acidosis or severe hyperkalemia or uremic manifestations, then a loop diuretics are definitely okay to try. And the key point to remember is that if we are going to try a dose of loop diuretic, it has to be high enough to get to the site of action in the loop of Henle. And so you might need a dose ranging anywhere from 80 to 200 milligrams. So biomarkers have gotten a lot of interest in the last few years because creatinine by itself is not an injury marker per se. And also there's a significant delay with the timing of acute kidney injury and the rice and serum creatinine. And some of the proposed markers are noted below and the ones approved for use, Engal is approved for use in Europe and Canada, but not in the United States. And then you have the TIMP2 IGF binding protein product called Nefrocheck, which is approved in the United States and Europe. So how would you use this? So if you have a Nefrocheck that's positive between 0.3 and less than 2.0, you know the patient is probably at moderately high risk for AKI in the next 12 hours. And you can consider monitoring them more closely, avoiding nephrotoxic exposure, optimizing their hemodynamic status, maybe avoiding vancomycin or monitoring the levels closely, maybe being cautious with iodinated contrast procedures, and maybe considering furosemide stress tests in these patients if they're volume overloaded. If the patient has a level that's greater than 2.0, they're very high risk for acute kidney injury in the next 12 hours. And you can consider early nephrology consultation or other testing. So for the next talk, we'll talk about renal replacement therapy. We'll talk about different modalities, some of the technical considerations and fluid removal. So thank you for listening and I will see you soon for the talk on renal replacement therapy. Oh, to summarize acute kidney injury has multiple different etiologies. Most of the time it's multifactorial in the ICU, usually related to tubular injury, but we should be cautious and look out for other possibilities such as an acute glomerulonephritis or interstitial nephritis. Treatment is targeted towards the underlying illness, whether it's sepsis or hypotension, and we should be using IV fluids and diuretics very carefully in these patients. A biomarker is available to predict risk of AKI. And if that is available, we can consider early implementation of a KDGO bundle for biomarker positive patients. Thank you.

acute kidney injury

diagnosis

management

urine output

KDIGO criteria

biomarkers