So the belle of the ball at last year’s Kidney Week was the PISCES trial, which reported a striking ~50% reduction in cardiovascular outcomes with fish oil in hemodialysis patients. That is an extraordinary signal. If you’re unaware, or skeptical, I suggest ’d strongly encourage you to check out NephJC’s coverage or listen to the Freely Filtered episode. The data look clean, internally consistent, and genuinely impressive.
One of the practical sticking points, however, has been how to translate PISCES into practice. Specifically: what fish oil actually matches the intervention used in the trial? The investigators are working with their supplier to bring the study formulation to market, but in the meantime clinicians are left asking what, if anything, can reasonably substitute.
Earlier this year, I was invited to the Medical University of South Carolina to give grand rounds, and during the discussion Ruth Campbell mentioned that she has been using the FDA-regulated, prescription omega-3 product Lovaza for her patients. As it turns out, that choice is quite defensible: Lovaza is fairly similar to the fish oil used in PISCES, both in formulation (omega-3 ethyl esters) and dose, differing mainly in the EPA:DHA ratio:
Because Lovaza is an FDA-approved prescription drug, it fits cleanly into most EHR workflows, and patients are far more likely to receive what the label says they’re getting than with unregulated over-the-counter fish oil supplements. That reliability matters when you’re trying to reproduce a trial signal as strong as PISCES.
A few weeks ago I had to check the dose of pregabalin (Lyrica) in kidney failure and found myself in the drug label (this is what happens when you fact check ChatGPT) and saw that they included the Cockcroft and Gault equation…
It was like a blast from my earliest memories of internal medicine. The orignial eGFR formula, MDRD, was already being adopted when I arrived in Chicago for fellowship, but during residency, if you wanted to know the GFR, you would whip out the old CG equation and calculate the creatinine clearance. And if you were really good, you would recognize that the constant, 72, in the denominator would cancel the weight for patients with roughly a normal size, so the formula was just (140 – age) divided by serum Cr, something that could be approximated in your mind. This hack was even pointed out in the original paper:
This ability to mental math the GFR disappeared with the MDRD formula:
This was the early days of handheld PDAs in medicine and the MDRD equation was a great reason to carry one with a medical calculator (MedCalc doesn’t arrive until 2005).
For the last year I have been working a with a team of academic internists from West Virginia on an editorial about the difficulties of assessing GFR in hospitalized patients. and an examination of the absurdity that we have a a lot of approved drugs where dosing guidelines were explored and established using the obsolete Cockcroft and Gault equation. The editorial has faced an uphill battle and is currently being revised for submission to our fourth journal. This is a new experience for me since in the past if I wanted to get my opinion out there I wouldn’t go through a journal but just post it here. Blogs have ruined my patience for the editorial process.
Though CG is antiquated and poorly validated it still performs well enough for quick assessments and areas where precise knowledge of the GFR is not needed. For example, in an 80 year old male with Cr of 2:
Cockcroft Gault provides a CrCl of 30 ml/min
MDRD gives you an eGFR of 32 ml/min/ 1.73m^2
CKD-EPI creatinine-no-race (2021), generates an eGFR of 33 ml/min/ 1.73m^2
It is hard for me to imagine a situation where there is a meaningful difference between 30, 32, and 33.
Flip the gender and it still performs well:
CG: 25 ml/min
MDRD: 24 ml/min/ 1.73m^2
2021 CKD-Epi: 25 ml/min/ 1.73m^2
One last thought on this topic: Cockcroft Gault, MDRD, and CKD-Epi all come from a time when we thought that GFR was the defining characteristic of kidney disease. It underlies the entire concept of CKD, where the specific etiology of kidney disease is less important than the specific GFR, with the unstated assumption that all patients with a similar GFR behave similarly and can be approached similarly. I do not think this theory has borne out and I think nephrology would benefit from moving beyond eGFR to a more nuanced vision of kidney disease.
Addendum, this post was inspired by this post on Bluesky
How long until drug labels stop including the Cockcroft-Gault Equation?Here is Zyrica's (pregabalin) label
Back in the day…after a wild night of partying nephrology fellows would stumble into rounds with this tattooed on their arm and no memory of how it got there.
Last week Apple launched their Hypertension Notification Feature (HTNF) on Apple Watch Series 9 or later and Apple Watch Ultra 2 or later (excluding Apple Watch SE). According to Apple the feature is intended for people over 22 years of age who have not been diagnosed with hypertension. It is not intended to be used during pregnancy. The watch uses photoplethysmography (PPG) which is a 14 letter word (44 Scrabble points) which means that the watch looks at blood volume changes through the skin and uses that information to predict who has hypertension. The watch records 60-second segments of PPG signals as inputs. These signals are collected roughly every two hours throughout discreet 30-day evaluation windows. The watch uses the accelerometer to assure that only PPG signals collecting during rest are used are used to assess for hypertension. A blood pressure score is created for each segment. Segments scored during sleep are ignored. While creating the model, the actual blood pressure was assessed by home blood pressure measurement.
Apple’s white paper about the system provides a classic Table 1 of demographic information on training, validation, and testing cohorts:
When determining who to alert for hypertension Apple decided to value specificity over sensitiviy to minimize false positives. From the white paper:
A receiver operating characteristic (ROC) curve was generated based on thedevelopment data, and an operating point was selected to prioritize high specificity — minimizing false positives — while also notifying a significant portion of people with hypertension
After the system was complete, Apple did an additional clinical validation study. They enrolled 2,229 people from two cohorts, all-comers without known hypertension risk factors and those at risk of hypertension based on historical blood pressure values and other risk factors. Each person was sent an Omron Evolv Wireless BP cuff (which scored very high in this review) and instructed to measure their blood pressure twice a day as well as wear the Apple Watch at least 12 hours a day.
Blood pressures were scored using the 2017 ACC/AHA hypetension guidelines:
The cohort for this part of the study was admirably diverse
The full set analysis is the original cohort. The Notification Analysis set only includes people who performed the home blood pressure assessments and wore the watch as instructed. They look reasonably similar except for the higher fraction of women in the Notification Analysis Set.
The investigators found that roughly a third of the Notification Set had hypertension and the watch notified about 40% of them. Only 99 of the 1,278 people without hypertension were given notices that they may have hypertension.
Here is how the test performed using a two-way table:
Screenshot
I feel a little bit out of my element since this is a test being being done in the background on people with no pre-test suspicion of disease. This is unlike any test I have ever ordered or really thought about. It is not a screening situation, where you want to maximize sensitivity at the expense of specificity in order to move a disease enriched population to more specific downstream studies. I think Apple’s description for their rational for tuning the ROC curves is appropriate. They want to capture people with hypertension while minimizing any unnecessary alarm and inconvenience for their customers with false positives.
The FDA application includes information from a 2-year study that is not further described. But it describes that specificity remains high month after month and that almost all of the false positives occur only in the first month.
I am curious, but the data is not provided, if the Hypertension Detection System increases sensitivity over the two years? Does it pick up additional cases of true positives? That would be cool if it did.
If you want to turn hypertension detection on for your Apple Watch, go into the Health App and scroll down to hypertension detection and answer the two questions and then follow through the four panels which describe how it works and to do home BP monitoring for more accurate diagnosis and then a generic warning that hypertension is not a heart attack. The last panel is not from the onboarding flow, but is another panel about hypertension that the health app provides. Nice.
I have hypertension, but I want to play with the warning system so I lied during set up regarding this diagnosis.
Apple received FDA approval for Hypertension Detection. You can read the regulatory filing and response here.
I love getting e-mails from interesting people. Here is a recent one:
Dear Dr. Topf:
First of all, congratulation to your recent promotion!
I have one question regarding the treatment of acute hyperkalemia. As an avid follower of your posts and blogs, I have implemented your “bag of saline with furosemide”-approach for several years in appropriate patients and it works like a charm. I particularly like the beauty of the physiology behind this supposedly simple regime, with flow-induced recruitment of BK (FIKS) supporting the job of ROMK.
However, in my experience, this treatment is woefully underused in clinical practice and little known – even among nephrologists.
Are you aware of a publication substantiating its use? Which reference should I quote?
Looking forward to your reply!
Best wishes,
XXXXXXXXXX
Hmm? Data to support the use of loop diuretics in the treatment of hyperkalemia. Let’s see what we can find.
The first reference I found was this article from 1984
These authors looked at renal potassium excretion in patients following 40 mg of furosemide daily for three days. They did this experiment in 6 health volunteers five separate protocols:
Furosemide with a high salt diet (270 mEq of Na per day)
Furosemide with a low salt diet (15-20 mEq of Na per day)
Furosemide with a high salt diet and captopril, intended to prevent increased aldosterone release with diuresis
high salt diet and captopril, this time without furosemide
Furosemide with a high salt diet and water load, this was intended to suppress ADH, which has a kaliuretic effect
The authors found that in the acute phase, the first 3 hours after IV furosemide, patients excreted 16 mEq of additional potassium over baseline potassium excretion with the high sodium diet, 19 mEq over baseline in the low salt group (presumably due to higher aldosterone) and 13.5 mEq in the high salt and captopril group.
Interestingly, the authors found that following this acute phase of increased potassium excretion there was a compensatory period decreased potassium excretion in the high sodium group that resulted in just about neutral potassium balance.
It makes me wonder if adding fludrocortisone would be helpful to the “bag of saline with furosemide”-approach to hyperkalemia.
I posed this question to my favorite potassium expert, Melanie Hoenig, and she suggested these references.
First was a circulation manuscript from when furosemide was the new kid and we were still trying to feel it out. These authors were playing around with weekly infusions of furosemide to treat hypertension. And it worked. (Link)
But, helpfully, they also provided information on potassium excretion, but unhelpfully they provided the data in micro Equivalents of potassium per minute.
So we have to do the math
Let’s say:
188 µEq/min from 0 to 22 minutes = 4.1 mEq 180 µEq/min from 22 to 45 minutes = 4.1 mEq 122 µEq/min from 45 to 90 minutes = 5.5 mEq 62 µEq/min from 90 to 120 minutes = 1.9 mEq
So a total of 15.6 mEq in the first two hours with an average of 63 mg of furosemide
195 µEq/min from 0 to 22 minutes = 4.3 mEq 201 µEq/min from 22 to 45 minutes = 4.6 mEq 166 µEq/min from 45 to 90 minutes = 7.5 mEq 91 µEq/min from 90 to 120 minutes = 2.7 mEq
And 19.1 mEq in the first two hours with an average of 210 mg of furosemide for you cowboys.
The next reference she sent was from the Canadian Medical Association Journal in 1968 where they took 115 male students and gave them one of three diuretics:
And then tracked them for 24 hours. The results are interesting. Here are the potassium results:
73 mEq of potassium. Pretty impressive. But look how hydrochlorothiazide does just as well, though it is backloaded with most of the kaliuresis coming later in the monitoring period. Hydrochlorothiazide actually resulted in more sodium excretion, over 24 hours than a single dose of furosemide.
The last reference she sent was a 1964 manuscript from Circulation where the authors were playing with the, then novel furosemide. They were using it in edematous patients resistant to available diuretics (acetazolamide, spironolactone, meralluride, and thiazides) as well as normal patients. Here are the daily electrolyte losses with various doses of oral furosemide, from 50 to 600(!) mg.
Good to know that doses beyond 100 mg don’t seem to add much kaliuresis.
So to answer the question, a slug of furosemide seems to be good for 15 mEq of potassium removal acutely, given that the extracellular volume of 70 kg man is about 17 liters, this should drop the serum potassium by a little less than 1 mEq/L.
None of these studies look at patients with hyperkalemia, so I would really like to see any experimental evidence with that, so if you know of any, hit me up on socials.
In 2012, something unusual happened. For the first time since 1868, the year the Detroit Medical College (the future Wayne State University School of Medicine) was founded, a new medical school was opening in the Detroit area. Detroit would no longer hold the distinction of being the largest American city with only a single medical school.
As Oakland University William Beaumont School of Medicine assembled its inaugural faculty, academic appointments materialized from thin air. I was handed the title of Assistant Clinical Professor of Medicine. It felt like a gift.
Over the years, though, that “assistant” began to weigh heavily on me. By 2024, I committed to upgrading that qualifier and pursue promotion. I had no idea just how arduous the process would be. My promotion packet eventually included:
A five-page Achievements in Service letter
A 100-page Achievements in Education dossier, anchored by an eight-page letter and 92 pages of artifacts
A 37-page Achievements in Scholarship document, with a 13-page letter and 24 pages of supporting materials
A two-page Achievements in Patient Care letter
A three-page Personal Statement
My CV
And a list of a dozen associate professors and professors across North America willing to review and score my work
In July of 2024, I submitted well over 150 pages of narrative, documentation, and supporting evidence to OUWB’s Promotion and Tenure Committee.
A few weeks ago, I received the news I had been hoping for: my promotion was granted. As of July 1, I am officially an Associate Professor of Medicine.
It feels good. More than that, it feels validating. I am grateful that Oakland University viewed my work in social media and medical education not as a novelty, but as serious, productive scholarly activity worthy of recognition.
I hope you are well. I have a clarification question regarding milk alkali syndrome. In this disease mechanism, since you have a loop diuretic like effect on the NKCC2 transport proteins, will you have both hypercalcemia AND hypercalciuria?
Best,
XXXXXXXX MS2
This question asks about the urine calcium level and yes in this conditionyou will have both hypercalcemia and hypercalciuria.
The elevated calcium binds the calcium sensing receptor on the basal lateral side of the thick ascendig loop of Henle tubule cell. This signals a decrease in activity of the ROM-K channel on the apical side of the tubular epithelial cell.
The Na-K-2 Cl channel (NKCC) of the thick ascending limb depends on ROM-K to allow potassium to be recycled. Sodium and chloride are found at way higher concentrations in the tubular fluid than potassium, so without recycling the potassium, the NKCC would grind to a halt for want of K. Since the K that is reabsorbed by the NKCC is able to leave the cell via ROM-K there is always plenty of K available to keep the NKCC turning. It does however make the seemingly electroneutral NKCC (2 cations and 2 anions) become electrogenic, because the potassium just leaks out down its concentration gradient so there is only 1 net cation reabsorbed compared to 2 anions. This makes the tubule electropositive and provides the energy to drive the paracellular reabsorption of magnesium and calcium (and probably some sodium as well).
Following the Ca sensing receptor shuting down ROM-K, the NKCC slows for want of K, and the tubule loses its positive charge. This prevents calcium and Mg reabsorption leading to increased urinary calcium as well as loss of Na in the urine.
Note that this is an appropriate change in calcium handling to help restore a normal calcium level. The high calcium itself shuts down calcium reabsorption. However this is inadequate to normalize calcium in milk-alkali syndrome since the acute kidney injury lowers the GFR so far that not enough calcium escapes to normalize the serum calcium.
I hope you are doing well! My name is XXXXXX and I am a second year medical student at OUWB. Thank you for your excellent lectures [ed: I added the bold because I love flattery] that you gave yesterday on sodium and water metabolism. During the second lecture, I did get a little confused on some of the core concepts regarding hyponatremia. I am trying to conceptually understand why your urine volume decreases when you have low solute and high amounts of water intake.
I can see why you are confused, let me try to reteach this.
This slide is supposed to demonstrate how the kidney handles water and solute.
In the absence of kidney failure, solute absorption = solute kidney excretion. Often this will be abbreviated solute in = solute out since our indescrimnate GI tracts pretty much absorb all the minerals and protein they are exposed to and, besides the kidney, no other organ system does meaningful solute excretion. In fact, a failure of solute in = solute out to hold true is a pretty good functional definition of kidney failure.
For people on a western diet (i.e. omnivorous) solute intake can be estimated at 10 mOsm per Kg body weight. So for the 70 kg adult, estimate solute load at 700 mOsm per day.
The kidney can get rid of that solute load in a variable amount of urine. If the person is drinking a lot of water, lowering body osmolality, the hypothalamus will detect this and decrease ADH resulting in dilute urine as indicated on the left side of the slide (absence of ADH, urine osm of 50) and get rid of that solute load with 14 liters of water. The loss of 14 lite3rs of water will increase the serum osmolality back toward normal.
If the patient has a low water intake (which will push the serum osm up) or high serum osmolality, the hypothalamus will release ADH and the urine osm will increase so the body will excrete that same osmolar load of 700 mOsm in only 0.6 liters, retaining any water intake in excess of this 0.6 liters. This retained water will dilute the serum osmolality back toward normal.
This is supposed to demonstrate that the kidney can get rid of the daily osmolar load in a wide range of urine volume to balance water intake and excretion in order to maintain homeostasis.
The next slide shows what happens when the patient is not on a normal western diet. In this case instead of eating 700 mOsm a day the patient is only taking in 100 mOsm a day
Now with ADH turned down to zero, and the urine osmolality bottoming out at 50 mOsm/Kg H2O the maximum amount of urine the body can produce is only 2 liters, an amount that people may exceed with normal, habitual fluid intake. The serum osmolality is low and the body wants to get rid of excess water, but turning down urine osmolality does not produce the expected copious amount of dilute urine needed to correct this situation, because the urine volume is limited by a lack of ingested solute. To increase the urine output to 3 liters would require 150 mOsm of solute (3 L x 50 mOsm/g H2O) but they are only eating 100 mOsm! So while in most cases urine volume is determined by ADH, with increasing urine volume with decreasing urine Osm, once the daily osmolar load is excreted no more urine can be produced.
If the patient has hyponatremia, wouldn’t the interstitial medullary gradient not form due to little sodium being filtered in the tubules at all and thus leading to low amounts of sodium leaving the NKCC2 proteins, thus leading to lower amounts of water being reabsorbed (this is how I’m currently thinking and why I am getting so confused)?
This is not proper thinking. An example of severe case of hyponatremia would be a sodium of 110 mEq/L. Given a GFR of 100 ml/min (0.1 L/min), this is still:
110 x 0.1 L/min x 1440 min/day = 15,840 mEq of Na filtered.
Plenty of Na to keep the medullary interstitium fully concentrated. Also remember half of the osmoles in the medullary interstitium are urea which is not really affected by the hyponatremia (not entirely true, but true enough for MS2s)
I am also aware that ADH is still going to be low here because you wouldn’t want to continue reabsorbing water with hyponatremia,
True
so this is also why I got confused and thought you’d be producing a higher volume of urine rather than a lower one. With the patient drinking lots of water AND having low sodium (hyponatremia), can you reiterate what happens?
When the osmolality is low the body will suppress ADH, unless there is some other stimuli of ADH: volume depletion (hypovolemic hyponatremia) decreased perfusion (hypervolemic hyponatremia from cirrhosis or heart failure) or SIADH. THe lack of ADH increases urine production until all of the daily solute is excreted. With a normal diet, the 700 mOsm will allow the production of 14 liters of dilute urine, enough to correct just about any hyponatremia. But if the patient is drinking 15 liters of water, even with maximally dilute urine there will be progressive hyponatremia, must have fluid intake below excretion to normalize serum Na.
We had a patient with an active infection and bilateral below the knee amputations. The Creatinine was obviously going to over estimate kidney function due to the low muscle mass and I wasn’t prepared to trust the cystatin C in the presence of active inflammation. What to do? Can we McGyver the GFR by looking at the fall in vanco levels over time? Yes, of course we can.
Here’s how it works. Vancomycin is primarily eliminated by glomerular filtration, so its clearance approximates the GFR.
1. Get two vanco levels
You need two vancomycin concentrations drawn after the distribution phase (ideally 1–2 hours post-dose and a trough) and without an additional dose in between.
2. Calculate the Elimination Rate Constant (ke)
This gives you the rate at which the drug is disappearing from the plasma.
3. Estimate vancomycin Clearance
Vancomycin’s volume of distribution (Vd) is about 0.7 L/kg.
4. Convert to GFR (mL/min)
Since vancomycin is almost entirely renally cleared, its clearance approximates GFR.
The vanco clearance calculated above is in liters per hour, so to get conventional GFR units, multiply by 1000 and divide by 60
5. Caveats
This only works if renal function is stable (no AKI or wild fluid shifts)
Must use post-distribution levels
Non-renal clearance of vancomycin is minimal but not zero
Vd can be wildly off in critical illness, obesity, or fluid overload
Bottom Line
You can use vancomycin estimate GFR. It’s not perfect, but in the right context it’s a clever way to triangulate kidney function when the usual suspects lie.
I find using LLMs is a bit like skipping stones on a lake – in the right domain with the correctly-worded prompt, you can sometimes get impressive output.And sometimes you get an utterly unceremonious "kerplunk".
I had a patient’s potassium pop from 3.1 to 5.1. It struck a half-remembered dream of a mentor teaching me that the most common cause of hyperkalemia was the treatment of hypokalemia. It is one of those internship myths that are recited mostly just to scare caution into over eager interns. So I asked Bluesky if anyone knew about this.
I once was told that the most common cause of hyperkalemia was mistreatment of hypokalemia. Anyone know of a reference to support that?
I then immediately, I went to Chat GPT4o and asked the same question. I hit the right domain with a correctly-worded prompt and the rock skipped clear across the lake.
Looking at my choices, I went with the middle one, because it had a DOI and I don’t have a subscription to UpToDate.
And being from 1998 made it contemporary with the pearl of wisdom in question.
The NEJM review of Hypokalemia by G. John Gennari was part of a review series on Fluids and Electrolytes the NEJM ran around the turn of the century. They were foundational articles for me as a fellow from 2001-2003. And on page 6 of an 8 page review of hypokalemia he drops this bomb.
Principles of Potassium Replacement Potassium replacement is the cornerstone of therapy for hypokalemia. Unfortunately, supplemental potassium administration is also the most common cause of severe hyperkalemia in patients who are hospitalized,53 and this risk must be kept in mind when one is initiating treatment. The risk is greatest with the administration of intravenous potassium, which should be avoided if possible. When potassium is given intravenously, the rate should be no more than 20 mmol per hour, and the patient’s cardiac rhythm should be monitored. Oral potassium is safer, because potassium enters the circulation more slowly.
In that study he combed the lab computer to find every case of hyperkalemia over 5.9 in a year. He found 300 and examined every case. Forty-eight of them were associated with new medications. Here was his list of meds that could cause hyperkalemia
Notice he doesn’t include ACEi as a group. He uses group names for beta-blockers and potassium sparing diuretics. But he lists captopril as a single agent because there were no other ACEi. This is an old study.
Of those 300, 172 had sustained hyperkalemia. He found 43 to have a single etiology of hyperkalemia (Group 1). And of those 43, potassium chloride was the sole etiology in 29 patients. Renal insufficiency was right behind at 24, followed by digoxin at 21. Did I mention this was an old study?
The whole study sounds pretty flimsy and polluted with biases. It stinks of post hoc choices and arbitrary rationalizations. Research has become much more rigorous in the last 40 years.
I don’t find this data to be cempelling but I’m also sure this was the source of the myth in question. And Chat GPT4o was instrumental in finding it and finding it fast.
That said the discussion on Bluesky was absolutely delightful. Follow the original post to see the rich conversation that bubbled up.
Post script: I do not believe that the most common cause of hyperkalemia is the treatment of hypokalemia. I think we live in a world with so much angiotensin and aldosterone inhibition that mistreatment of hypokalemia is a tiny blip on the radar.
That seems uncommon these days (suspect MRAs, RASi, etc). If anything, most hypoK cases are tail-chasing quests that never fully correct. In that ‘87 paper, pt must have an underlying renal issue 2 switch 2 hyperK, did they? Like we discussed in the podcast, it’s never just the bananas!
Post-Post Script: the first reference, the review from 2017 in the NEJM by Kamel and Haperin, It doesn’t exist. The computer is pulling their book from 2016.
This came up on rounds today. I thought I covered this on the blog before but if I did, Google failed me. The answer is 7.7 mEq of Na and 7.7 mEq of HCO3 in every 650 mg tablet of baking soda.
This is just 650 mg divided by the molecular weight of NaHCO3, 84 mg/mmol.
