Good afternoon, everyone. Thank you for joining me again. For the next half an hour or so, I'd like to discuss a few cases that I've put together. Some of these cases are pretty straightforward, but I have also added a couple of unusual cases that you might come across in your ICUs. My name is Anita Vijayan. I'm a professor of medicine in the Division of Nephrology at Washington University in St. Louis. For my first case, I will be discussing a 68-year-old woman who has a history of hypertension who presents to the ER with a fever, nausea, vomiting, and confusion. She is very febrile at 39.3. Heart rate is 98. Blood pressure is 130 or 59. Respiration 26. O2 sat is 92% on room air. She has right-sided costovertebral angle tenderness, and she's disoriented. Her WBC is 22,000. Serum creatinine is elevated from her baseline at 0.7 to 2.3. And she has greater than 50 white blood cells per high-power field. Imaging of her abdomen shows right perinephric stranding, but no stones or hydronephrosis bilaterally. Blood and urine cultures are obtained, and ceftriaxone is started, and she's admitted to the ICU. And so the question is, which of the following statements about this patient's acute kidney injury is most correct? A, the patient's acute kidney injury is likely due to ischemic ATN as a result of decreased blood flow. Her AKI is unlikely to be attributable to sepsis because she does not meet the current consensus definition of sepsis. Her AKI is unlikely to be attributable to sepsis given her normal blood pressure. The patient's AKI puts her at increased risk for secondary infections during hospitalization. Given her stage 3 AKI in the setting of sepsis, she would likely benefit from preemptive RRT before the development of an urgent indication. And I'll give you a few seconds to think about that. So the next case is a patient that we had at our hospital, a 45-year-old woman with Marfan syndrome and hypertension who presented with 3 weeks of sharp chest pain radiating to the back, left arm, and abdomen. She was on hydralazine, which had been increased over the last 4 years to 100 mg TID. Her arriving blood pressure was 210 over 107 with a heart rate of 47. Initial serum creatinine was 1.9 from a baseline of 0.9. Urinalysis showed a specific gravity, which was high, but this was after she had gotten IV contrast. She had 4 plus protein, 3 plus blood. Microscopy revealed 20 to 50 white blood cells per high-power field and greater than 50 red blood cells per high-power field. She received some prochloroperazine for nausea, which led to oropharyngeal angioedema complicated by cardiac arrest with successful resuscitation. A CT scan revealed type B dissection spanning the entire descending aorta, and she immediately was taken to the OR and underwent percutaneous endovascular repair of the aorta with stent placement. Post-op, she became completely aneuric, and serum creatinine rose to 4.3. Urine microscopy, which I personally performed, revealed several dysmorphic red blood cells and no granular casts, and she was started on emergent hemodialysis for hyperkalemia. Interestingly, A and N anko were sent, and these became positive. Anko in particular was particularly high at 1 to 112.80. And so what did I see in the urinalysis? So if you see this urine microscopy, which is not from her case, but this is courtesy of Dr. Seltzer from Missouri Baptist Hospital, who puts out a lot of urine microscopy images on Twitter, you can see that red blood cells in the urine, which normally should look like a peripheral smear if it's coming from lower urinary tract, it shows evidence of basically misshapen. They have Mickey Mouse ears like here, this one particular here, and these misshapen RBCs indicate that your glomerular barrier, basement membrane barrier, is disrupted and it's leaking RBCs through, and this usually suggests glomerulonephritis. So when I saw this patient's urine, which had a lot of dysmorphic RBCs and no granular cast, which I would have expected for somebody with severe ATN after major hemodynamic injury like that, I immediately became suspicious for glomerular disease. So the question is, what is the cause of her AKI? And the question is, is it acute tubular injury? Is it acute interstitial nephritis? Is it renal ischemia from aortic dissection with cortical necrosis? Or is it vasculitis with necrotizing glomerulonephritis? And I think I already gave you the answer here. So the correct answer is that, you know, patient probably doesn't have the first three based on her urine microscopy, and so most likely she has a necrotizing glomerulonephritis and anka-associated vasculitis. We did do a biopsy, and this is from her biopsy specimen itself, and you can see this glomerulus is very inflamed and has area of necrosis here, which is highlighted, and there's also the presence of crescent, which means there's extensive inflammation. And so the diagnosis on this patient was hydralazine-induced anka-associated vasculitis, and this is a diagnosis we should not miss because even though it's extremely rare, the mortality from vasculitis is so high that it should be treated immediately. So when we suspected this, we started giving her pulse steroids. We ended up giving her rituximab, plasmapheresis, and this woman became very sick. She was on ECMO for two weeks. She required several weeks of CRRT, and eventually she survived. So she, you know, spent a month in the ICU, a month in the step-down floor, and then she was discharged from the hospital after two to three months, and a year later, creatinine was 1.2. So if we had missed this diagnosis, this patient would not have survived this hospitalization. So very important to consider what other things could have contributed to AKI in ICU patients. So case three is from this critical care nephrology core curriculum, 64-year-old patient, 64-year-old man is admitted with cough, fever, and hypoxemia. Soon after, he develops respiratory distress and requires endotracheal intubation. Test X-ray shows diffused bilateral pulmonary opacities. Blood gas reveals PaO2 of 130 on 100% of FiO2, and bedside echo shows normal LV systolic function. Which of the following statements is correct about the management of this patient? So the next case is a 70-year-old man who is admitted with respiratory failure due to community-acquired pneumonia. Blood cultures are positive for pneumococcus. He started an appropriate antibiotic therapy and requires intubation. Shortly afterwards, he is initiated on norepinephrine and vasopressin therapy due to worsening hypotension despite intravenous fluids. Over the next 72 hours, the patient's creatinine level increases from 1.0 to 4.4, and urine output decreases to about 0 to 5 mL per hour. At a family meeting, it's planned to discuss the initiation of CRRT. So which of the following is most correct? Palliative care consultation would be inappropriate because he has AKI rather than chronic kidney disease. B, goals of care discussion is not indicated because most patients with AKI and respiratory failure survive. C, if the patient has started on RRT and survives the hospitalization, his quality of life at 60 days is likely to return to his pre-hospitalization baseline. D, CRRT should not be offered even if desired by the family because it would constitute futile care in this case. Or E, a time-limited trial of CRRT should be pursued in this case if CRRT is desired by the family. So palliative care in ICU, I don't think we discussed it enough, and there's no reason why a palliative care consultation would be inappropriate just because he has acute kidney injury. A lot of the patients who have acute kidney injury in the ICU have a lot of comorbid conditions. They might be nursing home residents, they might have severe heart failure, they might have end-stage liver disease, and if they're not candidates for liver transplant, then they may not necessarily benefit from RRT. So we should always consider a palliative care approach if we don't think the quality of life or even mortality is going to be significantly delayed by starting renal replacement therapy. So quality of life is a huge problem in nursing home residents who end up with critical illness and acute kidney injury. And in this case, I would say that considering palliative care would be important since the quality of life is unlikely to return. So a time-limited trial of CRRT is probably the most reasonable approach for a patient who might not survive the hospitalization or who may not have good quality of life left. So question five, a 54-year-old man hospitalized in the ICU for the last four days for septic shock following stem cell transplant is seen for AKI. He's hypotensive on vasopressors, oliguric, for the past two days. He has a positive fluid balance of 7 liters. He has bibasilar crackles, 3 plus edema, and patient is intubated for progressive respiratory distress. He develops AKI with a creatinine of 6.5, BUN of 86, and four days later, he is still on CRRT. He remains on ventilator, and even though he's awake, he fails waning trials. Two days later, he develops profound muscle weakness. His MAGK, calcium, are normal. His CK is 11,000. What is the most likely cause for his muscle weakness? Is it ICU-acquired myopathy? Is it Guillain-Barre syndrome? Is it hypoadrenalism, or is it severe hypophosphatemia? Just show you the composition of CRRT fluids. So all of the fluids have bicarbonate. In fact, all the fluids that are used in the United States in the ICU for CRRT are bicarbonate-based. It contains some potassium. There's one that has no potassium. All of them contain sodium, calcium in most of them, except the ones that we use for citrate. It has magnesium, chloride, and glucose, at least in these solutions. But it does not contain phosphorus. So hypophosphatemia is the most common electrolyte abnormality seen in the setting of CRRT. And the more intensive the dosing of CRRT, the more likely you are to get hypophosphatemia. Hypophosphatemia can develop very rapidly within even 24 to 72 hours of starting CRRT, depending on how high the phosphorus was when you started CRRT. So this is a post-hoc study from the ATN trial, which looked at intensive renal replacement therapy versus less intensive, and looked at extubation from mechanical ventilation. As you can see, the patients in the more intensive group were less likely to be extubated compared to the less intensive group. And if you look at their serum phosphorus, there was a correlation between the tertile of serum phosphate and patients who had events such as inability to wean from the ventilator. The lower the phosphorus, the less likely you're able to get off the ventilator. And this has been shown in other studies as well. So hypophosphatemia is an extremely common complication of CRRT. We have to check phosphorus daily while patients are on CRRT. Recommendation is to probably keep the phosphorus in the normal range, somewhere from 3 to 4.5. You can use phosphorus-containing solutions for CRRT. We don't do that in our institution. We start oral phosphorus supplementation as soon as phosphorus is less than 3.5 or 3.0, and the plan is to continue on CRRT. Of course, we can only give oral phosphorus if they're able to have a tube and have enteral nutrition or supplementation because if some of the patients have had extensive bowel surgery, this is not an option and we need to give IV phosphorus or add it to the TPN. So hypophosphatemia is something to pick, and I'm sorry for the spelling, hypophosphatemia is extremely common complication of CRRT. So case 6, 56-year-old patient with diabetes, developed diverticulitis, had bowel perforation, was taken to the OR for bowel resection post-op. He was in the ICU, remained on the ventilator for pneumonia, and had respiratory failure. He was hypotensive on two vasopressors. He developed acute kidney injury and was started on CVHDF. Due to concern for hypophosphatemia, foxylium, which is a phosphorus-containing CRRT solution, was used as a replacement fluid and dialysate. Now, we don't have foxylium at our institution. I based this case from a recent case report of what can happen when you use this solution if you're not careful. The patient subsequently developed bacteremia and sepsis, and two feeds were not started because of his recent bowel resection and ileus. And because of the bacteremia, the TPN was held until after the bacteremia could be controlled and a new catheter could be placed. So over the course of four days on CRRT, patient initially improved with decreasing vasopressor requirements. However, patient became progressively acidotic with CO2 trending from 24 to 17. PCO2 was 32. pH decreased from 7.38 to 7.36. Anion gap was 18, and blood glucose was between 120 to 160. Lactate was normal. So what is the most likely cause of his anion gap metabolic acidosis? Is it diabetic ketoacidosis? Is it CRRT-associated ketoacidosis? Is it D-lactate acidosis or propofol? So this is an interesting phenomenon that has recently been reported, and essentially what it is is the patient is developing ketoacidosis partly related to and remember I showed you a picture of the CRRT solutions where it all had 100 milligram per deciliter of glucose. Well, it turns out if you use foxylium solution where it has phosphorus, it does not have dextrose. So foxylium solutions which contain phosphorus have zero dextrose. And so what happens with these patients is that you're removing glucose and insulin via CRRT. You have impaired kidney gluconeogenesis because they do not have functioning kidneys. You have low glucose intake, especially like this patient who was not getting any nutrition. This leads to low glucose, low insulin, high glucagon with breakdown of lipase, sorry, breakdown of fatty acids, beta-oxidation of fatty acids, and eventually forming ketoacidosis. So you have to be very careful monitoring your patient's electrolyte and acid-base balance as they are on CRRT for many reasons. But this is a new phenomenon that has been recently described in the setting of phosphorus containing solutions. And it's really something to look out for because if you're not aware of this phenomenon, you would never think that these patients might be developing acidosis. The other scenario where an ICU patient might develop anion gap acidosis is not on CRRT per se, but patients who are receiving excessive doses of or high doses of acetaminophen as we're going away from using narcotics. We're giving a lot of Tylenol in the ICU. And in the patients who are malnourished, high dose of Tylenol can also result in anion gap metabolic acidosis by a pathway where it gets converted to phytoxoproline. So just to keep in mind these other causes of metabolic acidosis in the ICU. So this is from that paper that showed ketoacidosis with phosphorus containing solutions. And as soon as the patients received glucose and insulin infusion, you can see their pH corrected, their bicarbonate started to improve, and their ketones also resolved. So thank you so much for listening. And I'll be happy to answer any questions you might have by email if you need to reach me. I'm at avrijan at wustl.edu, W-U-S-T-L dot edu. Thank you so much.

acute kidney injury

ICU setting

ANCA-associated vasculitis

respiratory failure

CRRT

palliative care