Welcome, my name is Eric Tarula from the University of Wisconsin-Madison. Today we're going to discuss renal pathophysiology in the Neurocritical Care Review course. I have no conflicts of interest to report. Here are the learning objectives. In brief, we will be discussing acute kidney injury, the various neurologic manifestations of electrolyte disorders, review and approach to metabolic acid-based disorders, and then discuss renal replacement therapy. So let's start off with acute kidney injury. So what's the definition? Well, first off, I want to say that there's multiple definitions of acute kidney injury, and there are new definitions being worked on which incorporate biomarkers, so please be on the lookout for that. To the right here, you'll see some of these biomarkers, and what they're meant to do is to try to define function, areas of damage, and inflammation, and this is to ensure timely and accurate recognition of AKI and to prevent AKI. It's stated that at least 20-30% of AKI cases are considered preventable. So what are the definitions of acute kidney injury? Two of the main criteria that you'll run into is called the Rifle criteria and the Aitken criteria. To the left here, you have the Rifle criteria. You'll notice that the criteria is actually very similar. They're based off of increasing creatinine over baseline. The Rifle also includes GFR, and they're staged here by risk, injury, or failure, or stage 1, 2, 3. They both use urine output as a criteria, and those are the same. And for the Rifle criteria, they do have loss and end-stage renal disease as described by their failure of the decrease in GFR and increase in creatinine over weeks to months. So why do we care about AKI? Well, it's associated with increased mortality, cost, need for renal replacement therapy, I think that's obvious, and increased risk of progression of chronic kidney disease and into end-stage renal disease. And various studies have shown that AKI and neurological diseases to be between 8 and 38 percent. That's the incidence right there. And just for demonstration purposes here, in one study showing the mortality of, in this case, it was a cohort of traumatic brain injury, the mortality between those who had developed an acute kidney injury versus not. And you can see that's quite significant, almost a double in the mortality. So let's use a case example to look at acute kidney injury, define it, and to look at the etiology of it. Here we have a 61-year-old woman, or person, with a history of COPD, congestive heart failure, let's say they have an EF of 25 percent, and they're admitted for a subarachnoid hemorrhage. Their admissions labs are notable for slight hyponatremia, 134, elevated chloride at 105, BUN 25, and a baseline creatinine of 1.21. Four days later, they notice that their left MCA velocities on Doppler examination are up. She gets a CT angiogram and demonstrates mild spasm. This person is allowed to autoregulate, they have blood pressure of 150s, systolic, they're starting on 3 percent for hyponatremia. And on day six, it's noted that the blood pressures are still relatively high, on the higher side of 140s, systolic, the SP2 is 92 percent, we start to see crackles at the lung bases. There's anisarca, there's ascites, and there's a slightly firm abdomen. The labs on that day include a sodium of 144, chloride of 115, BUN of 55, creatinine 2.11, and a fluid balance of positive 4.2 liters. Their urine output has now been less than 0.5 mLs per kg per hour for the past 12 hours. A phena was obtained in less than 1 percent. So what's the etiology of acute kidney injury? Well, first off, let's start by defining it. She has AKI stage 2. They report that they have a phena of less than 1 percent, and now the phena, and here's a formula for that, looks at trying to decipher if the acute kidney injury is due to pre-renal or renal etiologies. So if you see a phena of less than 1 percent, does this mean that the patient needs more fluid or higher blood pressure? Well, some things you should know about the phena. It's only clinically validated in patients with oliguric acute kidney injury without diuretic use, chronic kidney disease, urinary obstruction, or acute glomerular disease. On top of that, I probably should have added, it's also not clinically relevant once you start using hypertonic sealing. In the chronic pre-renal stage, such as CHF and cirrhosis, you can have phena values less than 1 percent, and that's through the activation of the RAS system and sympathetic activation and vasopressin release as well. So you wouldn't be able to use a phena in this case. Phenas of phena less than 1 percent exclude acute tubular necrosis, that means pre-renal from renal. No. Acute tubular necrosis is an injury to tubular epithelial cells and is the most common cause of AKI. Etiologies include ischemic injury or direct toxic exposure. Now ischemic injury isn't just, we're talking about systolic blood pressure or MAPS, but also vascular congestion or medication-induced vasoactive changes, such as you get from NSAIDs, ACE inhibitors, or ARBs. It's defined by three phases. The initiation phase, which is the cell death, you get cast formation, and from that you get subsequent tubular obstruction and vasoconstriction. On the maintenance phase, you have a stabilized GFR, but this is where you get complications such as uremia. And this maintenance phase can last anywhere in the weeks. And this is clinically important for our neuro-ICU. In the recovery phase, you sometimes get complications of salt and water loss. So you may think of someone developing cerebral salt wasting, for instance, in these cases, but it's a post-ATN diuresis. Venous congestion. Well, in this case, the patient's fluid overloaded with a cytosine-affirmed abdomen. So this can possibly be due to an AKI due to venous congestion. Here we gave you fluid balance as well as some physical exam markers, but you should know that you can also use bedside ultrasound techniques to look at venous and arterial blood flow. In this case, you can use portal, hepatic, and intrarenal venous dopplers to assess venous blood flow. And here's an example of such a case. I don't have time to go over the particulars of this at the moment. This patient also had a firm abdomen. So one thing not to miss is that intra-abdominal hypertension, which I think is frequently missed, contributes to poor renal blood flow. And the physical exam has a poor correlation to actual pressure reading. So if you think intra-abdominal hypertension is present or can potentially be present and is clinically meaningful, then I think it's your job or duty to get a bladder pressure measurement and assess for that. This patient also got contrast. So did this patient have a contrast-induced CA-AKI? Well this is usually present within 48 hours of exposure, rarely in oliguric forms, and the risk factors include chronic kidney disease, diabetes, congestive heart failure, volume depletion, and contrast in volume. This patient did have chronic kidney disease and CHF, so this had some risk factors. What about the mechanisms? It's not fully understood. They think that there's some direct mechanism, which is just toxic, the actual ionated contrast agents or hyperosmolar or isosmolar contrast agents. And they also thought that there's indirect mechanisms through vasoconstriction that is nitric oxide-mediated. Now, I will say that contrast-induced AKI, some people consider it a myth, or there's a term coined pseudohypocreatinemia. If you look at some of the retrospective observational meta-analysis in isosmolar contrast, there have been reports of no clear association with the contrast administration and development of an AKI. There's no clear evidence of clinical meaningful harm. In this case, they're defining clinically meaningful harm as the need for renal replacement therapy. But there is some cagas to this. If a severely decreased GFR, some of the highest risk of developing this contrast-induced AKI were underrepresented in the propensity score studies. So there's a caveat to that. So the bottom line, it probably is real, but probably overstated in this contrast-induced AKI. In general, I say if IV contrast can provide information leading to better care, then you should consider ordering it. Obviously, there are some preventative measures, such as administration of IV sodium chloride or sodium bicarb, plus or minus oral and acetylcysteine. The evidence for this and decreasing the rates of contrast-induced AKI are soft, but this is a relatively benign and cheap drug. So revisiting this case one, what about the chloride load? That's been a hot topic. We've kind of introduced an iatrogenic hyperchloroemia. Well, something that you should know about chloride, hypokalemic acidosis from saline can cause systemic hypotension, can cause renal vasoconstriction. It's been demonstrated that chloride-restricted or balanced fluids decrease the incidence of AKI and the need for renal retacement therapy in critically ill patients. And just as a kind of a fun fact that I ran across, 0.9% saline is associated with longer times to first maturation compared to renal lactate infusion. And I don't know about you, but for whatever reason, my neuro-ICU patients all have issues with maturation. Let's move on. Case two, a 50-year-old is admitted for non-functioning pituitary adenoma resection. Post-update two, it's noted that the sodium is 130, was normal beforehand, and the patient is placed on 0.9% saline. And so that's those glucocorticoids. The following day, the sodium is lower, 126. Urine osmolality is 626, the urine sodium is relatively high at 262, and the patient is now placed on additional 3%. Later that day, sodium is now 124, and the patient looks a little bit more somnolent. What do you do next? Well, I think you should have a differential for this. It's either SIADH, or synapse of inappropriate antideriotic hormone, or cerebral salt wasting, which you sometimes see more commonly with subarachnoid hemorrhages. Well, in this case, the 0.9% was discontinued, 3% was increased, and you got some modest increase in the sodium. But then we did a fluid restriction, and by day five, the sodium was normal. Now speaking of subarachnoid hemorrhages, the prevalence of cerebral salt wasting in subarachnoid hemorrhages is estimated to be around 11 to 23%. Now the one thing, the difficulty lies in differentiating these two entities, is that there's a lot of overlap in lab markers, including AMP, the urine sodium is typically elevated in both, the urine osmolality as well, and usually have an elevated and exaggerated naturesis, usually from R-saline infusion to prevent hypovolemia. So kind of the end result is that cerebral salt wasting should include evidence of volume depletion in the presence of ongoing naturesis, which is sometimes difficult to, again, test clinically because we're really trying to prevent hypovolemia specifically in our subarachnoid hemorrhage patients. Moving on to case three, here we have a 26-year-old woman with HIV, alcoholism, brought to the emergency department in comatose, lying in bed for greater than 24 hours. I have a set of labs here, sodium 134, potassium 3, chloride 100, bicarbs low at 16, BUNs high at 62, creatinine is now 3.6, I'd say she had a baseline much lower than this, glucose of 80, calcium is low at 6.9, and phosphate is low at 2.1, also a lower albumin. The urine looks reddish-brown, the pH is 5.2, she has blood, some protein, including a sediment consisting of a muddy brown cast, she has a phenylase of 1%, she has cocaine in her system as well as some alcohol. So what other labs should you be looking for? In this case, you should be looking at the CK, right, we're looking at evidence of skeletal muscle injury, or myonecrosis, let's say her CK was 15,000, so she's developed rhabdomyolysis. So what's mostly the cause of her rhabdo? Well, in this case, she had immobility, but she also had the hypophosphatemia, which leads to depletion of ATP. The hypokalemia decreases the skeletal muscle blood flow, the cocaine, also through unknown or multiple mechanisms, can cause rhabdomyolysis on its own, and HIV is also a risk factor, through possibly an immune-mediated response, or process. So let's say that patient is aggressively fluid-resuscitated with some sodium bicarb, now nursing staff is calling your attention to some pre-op movements that she noticed in lab 10, she's concerned basically for seizures, it's pictured below, this is what she's seeing, or he's seeing. This is occurring about once every hour. The EKG shows QT prolongation at 510, so what lab should you get next? Here's a hint, the left arm also has the blood pressure cuff going off every hour. This is carpal-pedal spasm, in this case, hypokalemia is common in rhabdomyolysis through multiple mechanisms, so you should always get a calcium, or watch out for your calcium in these cases. There's a lot of neurologic manifestations of electrolyte disorders, here's a short list of hyponatremia, hypokalemia, magnesium, and hypophosphatemia. In general, most of them will cause some sort of weakness, some of them can cause coma, hypomagnesemia, hyper- or hyponatremia, most commonly hyponatremia, as well as seizures. Seizures particularly with hyponatremia and hypomagnesemia. Hyponatremia is common in your alcoholic patients, so if you have someone who's alcoholic now seizing, you should also consider or look at their magnesium as well, and not just simply attribute it to alcohol withdrawal or alcohol intoxication. Let's move on to acid-base disorders. We're going to review the Henderson-Hasselbalch approach, a simplified approach, which just states that metabolic disorders involve a primary change in certain bicarb and or the anion gap. The rest of the disorders involve primary changes in the pCO2 due to changes in CO2 removal by the lungs. As a reminder, your anion gap is just sodium minus your bicarb plus chloride. There are other approaches, I taught this last year, the Stewart approach, which may be more useful in quantifying disorders and for generating hypotheses regarding mechanisms, but this is a much longer approach and much more complicated, but just so you know that there is another approach and sometimes it can be useful. So let's look at metabolic pH abnormalities, a simplified approach. First you're going to look at your anion gap. If your anion gap is greater than 12, you've got to compare your delta gap and your delta bicarb. So your anion gap minus 10, 24 minus your measured bicarb. If your anion gap, delta anion gap is less than your delta bicarb, less than 1, then you're probably looking at a mixed anion gap and non-anion gap metabolic acidosis. If this ratio is 1 or 2, you're probably looking at just a pure anion gap metabolic acidosis. And if your ratio is greater than 2, you're probably looking at an anion gap metabolic acidosis along with the metabolic alkalosis or compensating risk-free acidosis. If your anion gap is less than 12, then you just got to look simply at your bicarb, bicarb less than 22, you're probably dealing with a non-anion gap metabolic acidosis. 22 to 23 is normal and greater than 28, a metabolic alkalosis. So what are the causes? For anion gap metabolic acidosis, some of the causes include ketoacidosis, like diabetic ketoacidosis, uremic acidosis, obviously in your chronic or end-stage renal diseases, lactic acidosis, either from shock or even a thymine deficiency, as we sometimes may see in a neuro ICU. Medication, such as metformin or excessive beta agonists, such as epinephrine infusion can cause this. And poisoning, such as ethanol, and I also include propofol because I think propofol can lead to poison. Some non-anion gap metabolic acidosis, probably the most common we'll see in a neuro ICU due to hypertonic saline is hyperglycemic from hyperglycemic fluids, so the chloride. We also got to think of GI bicarb loss from diarrhea or high output fistulas, renal insufficiency, and then renal tubular acidosis. I think this is under-recognized, and there's different types of renal tubular acidosis, which I won't review here, but that some of these actually can be unmasked or brought about by various medications. Some of these medications of which we use commonly in our ICU, such as butyric acid, you may see lithium, serozolomide, heparin or heparin infusion, hyperthyroidism, and autoimmune diseases, such as lupus. So renal replacement therapy, first, what are the indications? Indications include diuretic resistant clinically significant edema, medically refractory hyperkalemia, metabolic acidosis that's also medically refractory and clinically significant, uremic complications, such as encephalopathy and bleeding, and when you have dialyzable drug intoxication, such as lithium or salicylates. What about the timing? Well, this is a big topic, but in general, you should take a personalized approach. Also, in general, excluding life-threatening emergencies, renal replacement therapy can be safely delayed up to 72 hours in stage 3 AKIs, although there's little evidence in neurologically ill patients, and I'll put that as a caveat. So renal replacement therapy, let's say I have a patient that I know needs renal replacement therapy, but the intracranial compliance is low. So what are my RRT options? Well, the options for this include peritoneal dialysis, intermittent hemodialysis, and various forms of continuous renal replacement therapy, such as prolonged intermittent renal replacement therapy or sustained low-efficiency dialysis or SLED. So why is this even an option? Why is CRRT even an option? Well, the slower rates result in less hemodynamic changes, first of all, but in neurocritical care, we worry about the osmotic shifts. So here we got case 4, he's a 45-year-old man who suffered a severe TBI. He has a right-sided craniotomy with persistent midline shift and ICPs of 26 and a CPP of 55. He has new onset oligodenal failure that's multifactorial in nature, but will now require renal replacement therapy. He has no current hemodynamic issues with the following labs. He has a sodium 155, potassium 5.5, chloride 118, bu105, and creatinine 3.3. So renal replacement therapy and osmotic shift. So which mode of dialysis is safest, IHD or CRRT in this case? Well, RRT leads to removal of osmotically active substrates like urea from the blood. And there are several case reports of intermittent hemodialysis leading to acute deterioration in patients with various neurologic pathologies with poor intracranial compliance. Basically people have herniated and died from this. So CRRT has been proposed as the safest means of RRT given the slower rate of urea clearance. So I'll pose this question, does CRRT ensure that osmotic shifts will be clinically negligible? Well, you can tell by the way I said it, probably not. Here's one study that came out of Denmark showing intracranial pressures in various patients and the difference in their intracranial pressures, each line here represents a patient, the time of start of dialysis and their ICP changes. As you can tell here, IHD patients had, some of them had increasing clinically significant increases in the pressures, but also in CRRT, as you can see here, not just because they had CRRT, that doesn't mean that it was negligible, I mean it was negligible in some people, but some of them also had very significant changes in their pressures. Now they try to plot the changes in the intracranial pressure during dialysis compared to the plasma urea before dialysis. And there was some, there was a trend so that the higher the urea before dialysis, the higher the risk for, or the bigger the change in the ICPs, but not a perfect one. So is CRRT safe? Well, probably safer in terms of it may give you more time to act, but you still got to need to monitor this closely. Obviously there's less hemodynamic fluctuation, which we'd love in the neuro ICU. There's several novel membranes that have been introduced to remove pro-inflammatory cytokines. So this may augment in the future timing indications. So please be on the lookout for these future studies. One caveat to this, again, CRRT is associated with thrombocytopenia, and if they have thrombocytopenia, it's associated with an increase in mortality in ICU. Here's the citation for that. So I mentioned this because not only that it's associated with increased mortality, probably due to a variety of mechanisms such as immune system mechanisms, but also we worry about the platelets in a neuro ICU patient just from a bleeding risk stratification. What about renal failure and drug metabolism? Well, a lot of drugs are metabolized through the renal system. Here's just when we look at what's their GFR and what's the safe and appropriate drug dosing. One thing you should know that GFR does not reliably measure, first of all, AKI, and so you should properly drug doses and monitor them over time. And on top of that, if you have someone who is on dialysis, you need to consider drugs that are dialysable and thus will need redosing. Now this list is very large, but here I have a list of some of the more common medication that you may see in a neuro ICU, such as barbiturates, lacosamide, levatorastam, or Keppra, aminoclerotic acid, acyclovir, aspirin, steroids, Depakote, and Dabigatran. So let's start with the key points. Acute kidney injury is associated with a worse outcome. So I think it's imperative that we identify this quickly and try to prevent it or address it as soon as possible. Neurologic ill patients usually have a variety of potential ideologies for their acute kidney injury as we demonstrated with that first case. Simplified abnormalities have various neurologic manifestations. So we must be vigilant that not everything that's neurologic is purely from a simple CNS abnormality, but we have to think of systemic diseases and systems. Acid-based disorders are common, and a simplified approach can be used in most cases. We went through one such approach. And then care should always be taken in patient replacement therapy, such as IHC or CRRT, and those with potentially low intracranial compliance. Got to be vigilant. All right, thank you for your time.

renal pathophysiology

neurocritical care

acute kidney injury

electrolyte disorders

metabolic acid-base disorders

renal replacement therapy