Tag: hyponatremia

Posted on

I helped create a significant manuscript on hyponatremia

I have been an avid reader of the hyponatremia literature for my entire medical career and last week I contributed to a significant new manuscript to the dysnatremia canon.

This link is supposed to work to give people access to the full text of the manuscript.

This feels like a major career achievement. For once I’m not just talking and teaching about other people’s work but sharing my own work, and in a topic I care a lot about. I am so grateful to Michael Fralick who approached me about helping out on the study way back in 2019.

ADH stood for Association with Demyelination in Hyponatremia

Amazing to see it mature from idea to data to manuscript.

The article looks at the rate of osmotic demyelination syndrome in hyponatremia and it is the largest examination of this question that I am aware of. I posted a Tweetorial on the topic here:

And what follows is the draft of the tweetorial (minus the gifs)

Never in the history of medicine has so much been done, by so many, so incompetently, with so little consequence as in the treatment of severe hyponatremia. #Tweetorial  1/10

We all know that we cannot correct hyponatremia too fast and that the speed limit is 8 mmol/L per day. We know this. We still do a terrible job at it. In the landmark George study, 41% of 1,490 patients had their sodium corrected faster than 8 mEq/L. Look at the poor slobs at the left of the graph whose Na actually went down in the first 24 hours 🤪 2/10

Thankfully this incompetence is rarely punished. Of the 611 (41% of 1490) patients who over-corrected in the George trial, only 7 developed osmotic demyelination syndrome (ODS). Screw the sodium correction and you can get away with it 99% of the time. 3/10

I had the good fortune to be invited to help with a research study looking at the incidence of ODS in hyponatremia using the GEMINI database which tracks internal medicine admissions in Toronto. It is amazing that someone you look up to and respect from the medical literature DMs you to join them on an important study. Thanks @FralickMike 4/10

So the gemini database was used to look up every case of hyponatremia (Na < 130) admitted to one of 5 academic hospitals in Toronto from 2010 through 2020. 

22,858 cases of hyponatremia. This is a massive study. By far the biggest ever. If you restrict the cohort to just people with a sodium < 120, it is nearly twice the size of George. 5/10

And the Canadians did a better job of correcting the sodium, but still went too fast in 18% of cases, 3632 patients. But hold on, a lot of these people had relatively mild hyponatremia (relative to George). When you break it down by starting Na, Canadians look just as bad the US at fixing the sodium slowly. 6/10

And what was the consequence of all that hyponatremia? And all that rapid correction? Twelve cases of osmotic demyelinating syndrome.
12 out of 22,858 cases of hyponatremia. 0.05%

If you divide by starting Na they found an incidence of:  

0.3% with a Na < 120

0.015% with a Na > 120

2.5% with a Na < 110 7/10

The part of the manuscript you are looking for that is not there is rates of ODS by rapid versus slow correction. We could not publish this because of the ethical guardrails on this trial. And the statisticians wouldn’t even whisper it in my ear because they know me. Kind of a bummer.  8/10

So what can we take from this paper? ODS is rare, Canadians are just at bad correcting the sodium slowly as the guys in Pittsburgh and the rate of ODS errally goes up as the initial sodium level goes does down, from a trial 0.015% at levels > 120 to 2.5% with sodiums < 110. 9/10

Take a look at the paper and at the George paper.

https://evidence.nejm.org/doi/10.1056/EVIDoa2200215

https://pubmed.ncbi.nlm.nih.gov/29871886/

10/10

Addendum:

Brian Locke asked how I made the gifs in the tweetorial

Did you make the animations? If so, they are awesome and I’d be interested to know how. #MedEd

Also, great study 👏

Originally tweeted by Brian Locke (@doc_BLocke) on March 30, 2023.

They are simple animations in Keynote. Here is the file so you can see how they work.

Hyponatremia-Gemini-TweetorialDownload

If this is interesting to you, don’t miss the editorial by Ayus. He’s ready to burn everything to the ground and he called his own number a few times in the refs.

Posted on

More OUWB M2 questions and Answers

The question: I am going through the real well made sodium and water guide you made and there is a concept I am having trouble with regarding ADH stimulation by decreased BP. In the text, one of the reasons that ADH release is triggered is:

Decrease in blood pressure: ADH is a potent vasoconstrictor. 10% drops in blood pressure are required to release ADH. This drop in pressure is usually due to a drop in the blood volume from volume loss but can also be due to heart failure or cirrhosis. ADH is relatively insensitive to changes in pressure, this is why it takes a full 10% drop in pressure to start to stimulate ADH.

I thought blood pressure goes up with heart failure as hypertension is a risk factor for heart failure. How do we get decreased blood pressure from heart failure? 

The Answer: Hypertension can cause heart failure. Hypertension is a risk factor for hypertension. Heart failure can present with hypotension or hypertension. Regardless of the blood pressure, heart failure can cause poor perfusion, and stimulate ADH release, with or without a low blood pressure. Sometimes the heart pumps so poorly that blood backs up in the venous circulation, this venous congestion slows perfusion since the back pressure blocks flow, even though blood pressure it good.

So the important point is that heart failure, cirrhosis and volume depletion all stimulate ADH release through decreased perfusion/blood pressure predisposing to hyponatremia.

Posted on

Another hyponatremia Tweetorial

This was a great case. The full the tweetorial is unrolled below.

🧂 Hyponatremia #CaseReport #Tweetorial

(and a gif for the people who want to rename this hyperhydronemia)

Chug Water GIF

Patient came to the hospital with abdominal pain, nausea, and vomiting. Patient has alcohol use disorder. Last drink was about a day prior to admission.

After arrival to the ER the patient has a seizure.

gross james van der beek GIF

Besides the weirdly elevated anion gap, and the hypokalemia, the initial labs just show some AKI. I don’t have an ABG but I suspect combined metabolic alkalosis and lactic acidosis.

This can be demonstrated by looking at the Delta Ratio which compares the change in bicarb to the change in anion gap. The ratio should be 1. If it is less than 1, there is an additional non-anion gap metabolic acidosis, > 2 additional metabolic alkalosis

A delta ratio of 6 is crazy high.

Kate Mckinnon Snl GIF by Saturday Night Live

A related calculation, called the “bicarb before” can tell you the serum bicarb without the anion gap acidosis, so if the patient has two disorders it allows you to look at the metabolic alkalosis (or non-anion gap metabolic acidosis) without the anion gap metabolic acidosis.

The “bicarb before” comes to a mostly unbelieable serum bicarb of 60.

But the reason I was intrigued by the case are the next two labs that come 10 and 20 hours after the initial labs…The sodium drops to 125 despite getting 150 an hour of 0.9% NS and the patient making 3600 ml of urine.

Additionally the full force of the metbolic alkalosis is revealed with the bicarb shooting from 29 to 41. I suspect this is due to vomiting. The urine chloride < 20 is consistent with this. This is Cl responsive metabolic alkalosis. It will (eventually) respond to the NS.

The urine also has a massive anion gap, around 90. What is the unmeasured anion?

(BTW the answer is bicarbonate)

But what is driving down the sodium? The patient appears volume depleted, and the steadily improving serum creatinine points to a patient with pre-renal AKI.

In volume depletion hyponatremia, giving fluid improves the serum sodium, it doesn’t make it worse. Also these patients do not typically make 3600 ml of urine

In it is highly unusual to make that much urine and have the sodium fall, usually that kind of urine output is associated with arising sodium. A hint to what is happenning can be found in the electrolyte free water clearance (Clefw).

The high urine sodium and really high urine potassium makes the urine essentially isotonic to plasma. Even though the patient is making 3.6 liters, it is like taking ladles of soup from a big pot, no matter how many ladles you take out it doesn’t change how salty the soup is.

pot wo GIF

Because the electrolyte free water is close to zero, those 3.6 liters of urine are not afffecting the serum sodium at all. So why is the sodium falling? I suspect this is due to the patient drinking (unrecorded) water.

So what would you do if faced with a falling sodium in a volume depleted patient?

I chose Tolvaptan plus continued 0.9% NS at 150/hr. The following day, the labs look…better.

I think this patient had nausea induced ADH in addition to severe metabolic alkalosis and volume deficiency. I found it interesting.

Originally tweeted by Joel M. Topf, MD FACP (@kidney_boy) on July 30, 2021.

What I loved about it is that the full lab interpretation required six different equations:

  • Anion Gap
    • the rare case of a relevant anion gap despite an increased serum bicarb
  • Gap Gap analysis
    • First the Delta Ratio
    • Then the Bicarbonate Before
  • Urine chloride in metabolic alkalosis
    • <20 mEq/L is chloride responsive
  • Urina anion gap
    • People think it is just for RTAs…not true
  • Electrolyte free water clearance

And I got great comments from Twitter. Some highlights:

Posted on

DDAVP Clamp

Bibliography:

Perianayagam, A. et al. (2008) ‘DDAVP is effective in preventing and reversing inadvertent overcorrection of hyponatremia’, Clinical journal of the American Society of Nephrology: CJASN, 3(2), pp. 331–336. https://cjasn.asnjournals.org/content/3/2/331

Gharaibeh, K. A. et al. (2015) ‘Risk factors, complication and measures to prevent or reverse catastrophic sodium overcorrection in chronic hyponatremia’, The American journal of the medical sciences, 349(2), pp. 170–175.

The initial infusion rate (mL/hr) of 3% saline can also be simply calculated as a product of patients’ weight (kg) and desired correction rate (mEq/L/hr)

Sood, L. et al. (2013) ‘Hypertonic saline and desmopressin: a simple strategy for safe correction of severe hyponatremia’, American journal of kidney diseases: the official journal of the National Kidney Foundation, 61(4), pp. 571–578. https://www.ajkd.org/article/S0272-6386(12)01471-0/fulltext

Rafat, C. et al. (2014) ‘Use of desmopressin acetate in severe hyponatremia in the intensive care unit’, Clinical journal of the American Society of Nephrology: CJASN, 9(2), pp. 229–237. https://cjasn.asnjournals.org/content/9/2/229

Adrogué, H. J. and Madias, N. E. (2000) ‘Hyponatremia’, The New England journal of medicine, 342(21), pp. 1581–1589. https://www.nejm.org/doi/pdf/10.1056/NEJM200005253422107

Mohmand, H. K. et al. (2007) ‘Hypertonic saline for hyponatremia: risk of inadvertent overcorrection’, Clinical journal of the American Society of Nephrology: CJASN, 2(6), pp. 1110–1117. https://cjasn.asnjournals.org/content/2/6/1110

Morris, J. H. et al. (2018) ‘Rapidity of Correction of Hyponatremia Due to Syndrome of Inappropriate Secretion of Antidiuretic Hormone Following Tolvaptan’, American journal of kidney diseases: the official journal of the National Kidney Foundation, 71(6), pp. 772–782. https://www.ajkd.org/article/S0272-6386(18)30004-0/fulltext

Posted on

Hyponatremia tweetorial

This is a basic approach to hyponatremia. (link if you would prefer to see the tweets on twitter)

1/ This is like the Bat Signal for me… #Tweetorial #Hyponatremia #LoudlyForThePeopleAtTheBackOfTheRoom

https://twitter.com/tonlajr/status/1304011343226437637
Batman Michael Keaton GIF

2/ The first question in hyponatremia is…

👏Is👏the👏patient👏seizing?👏

If you have a patient with cerebral edema from acute hyponatremia you need to 3% Saline first and ask questions later.

Matthew Smith Doctor Who GIF

3/ If patients have hyponatremia and have severe symptoms it is 150 ml of 3% then recheck the sodium and give another 150 ml of 3% (I’m using the European guidelines)

https://eje.bioscientifica.com/view/journals/eje/170/3/G1.xml

3b/ Be careful with vomiting there 👆🏻

This means if the hyponatremia is causing the vomiting, then you have severe symptoms and it is 3% time. But if the vomiting is causing the hyponatremia, that doesn’t automatically indicate severe symptoms and you will need to dig deeper.

4/ The goal of acute mgmt is to raise the Na by 5 or stop the symptoms, which ever is quicker. If you have raised the Na by 5 and they are still seizing, then it is probably not the low Na causing the seizure and you need to 👀 deeper. Algorithm from

https://eje.bioscientifica.com/view/journals/eje/170/3/G1.xml

5/ But when @tonlajr asked about the approach to sodium he wanted to know abot dx, not acute mgmt so…onward to diagnosis!

Glory Roomba GIF

6/ I am goin to be walking you through this algorithm:

https://twitter.com/hashamsarwar/status/1304476073627914240?s=20

Step one check the serum osmolarity (and get a repeat sodium, just to make sure it wasn’t a lab error)

7/ Since we are looking at a low sodium we expect a low osmolarity (don’t worry about the difference between osmolality and osmolarity, anyone who is a stickler about that is being a dick)

If we don’t find a low osmolality something weird is going on:

7b/ Low sodium with a normal osmolarity: Lab error from too much protein (IVIG, multiple myeloma) or too much fat (High lipids or triglys) in the blood. This throws off some lab machines.

(Specifically but not point of care iStats or ABG laboratories)

7c/ Low sodium with a high osmolality: this is usually due to hyperglycemia (but can be seen with mannitol, glycine and other edge cases). The hyponatremia is real, but due to another osmotically active particle (glucose in most cases) the are no consequences to the hyponatremia

7/d It is not as simple as that (it never is).
Read this manuscript journal📖: https://pubmed.ncbi.nlm.nih.gov/26002851/
or listen to this podcast🎧:

EMCrit 242 – ED Nephrology with Joel Topf aka KidneyBoy

8/ So that leaves the true hyponatremia. Low sodium and low osmolarity. The branch point here is:

What is the urine osmolarity?

The urine osmolarity tells us if the kidney is causing the hyponatremia or just unable to correct the hyponatremia (despite the best intentions)

9/ The urine osmolarity is less than 100 (maybe up to 150 or 200 if the patient has CKD). This indicates a lack of ADH and a kidney that is doing its best to correct the hyponatremia. The problem is not the character of the urine but the amount. Right urine, not enough.

9b/ The differential for low sodium, low serum and urine osmolarity is short:

• Oliguric kidney failure
• Tea and Toast
• Beer Drinkers potomania
• Psychogenic polydipsia

* Recovering from volume depletion hyponatremia

9c/ In low Na, low serum and urine osmolarity the urine is getting rid of excess water but the kidney cannot make enough urine because:

• Kidney failure (low GFR)
• T & T / Beer drinkers (lack of solute)
• Psychogen polydipsia (you are drinking faster than you are peeing)

10 Low sodium, low serum osm, high urine osm. This is ADH dependant hyponatremia. The kidney, stimulated by ADH, is causing the hyponatremia by generating free water. Making more urine here, just makes the Na fall further.

10b/ ADH can be 𝗽𝗵𝘆𝘀𝗶𝗼𝗹𝗼𝗴𝗶𝗰 due to low volume or a perceived low volume state:

Low volume states: GI losses like diarrhea or vomiting; renal losses like diuretics

High volume/low perceived volume: heart failure, liver failure, nephrotic syndrome

10c/ ADH can be 𝗻𝗼𝗻-𝗽𝗵𝘆𝘀𝗶𝗼𝗹𝗼𝗴𝗶𝗰 (euvolemic):

Adrenal insufficiency
Hypothyroidism (probably doesn’t exist, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4470237/)
SIADH (it’s usually this)

10d/ To differentiate these you can try to use a physical exam to determine volume status but doctors suck at it.

https://pubmed.ncbi.nlm.nih.gov/3674097/

10e/ Instead (or in addition) Check urine sodium and serum uric acid:

Urine sodium low, uric acid high in hypovolemic and hypervolemic hyponatremia

Urine sodium high and uric acid low in euvolemic hyponatremia

Originally tweeted by Joel M. Topf, MD FACP (@kidney_boy) on September 12, 2020.

Posted on

Lecture on hyperosmolar hypotonic hyponatremia

This is a formal lecture on the Tweetorial I posted about a patient with beer drinkers potomania but presented with increased serum osmolality due to ethanol intoxication.


Keynote • PowerPoint • PDF

Note: The presentations are designed to be displayed in KeyNote, the powerpoint version may be weird, especially the animations, and some of the icons maybe jaggy.

Following the lecture Scott and I recorded an episode of the EMCrit podcast, episode number 242! Scott Weingart is a medical education revolutionary. If you are interested in learning about him check out this episode of Explore the Space.

Posted on

A couple of new Tweetorials

The first was in response to Robert Centor’s excellent description of how he uses reciprocal creatinine. Honestly I had not thought about reciprocal creatinine in a long time. It was fun diving into some of the literature around it. Here is the tweetorial:

Today I did a second tweetorial on hyperosmotic hyponatremia

Here I had some technical problems. I wrote the entire tweetorial using chained tweets in Safari on MacOS. When I went to upload all tweets, Twitter hung and failed to upload more than the first 8. I had to then go through and re-post the remainder of the tweetorial. I was frustrated and failed to attach two of the animated gifs I made. I added them as additional tweets but they break the flow. Tweetorials are like writing email newsletters, once you publish the tweets (or hit send on the newsletter) there is no opportunity for editing.

Posted on

More questions from the minds of the M2s at OUWB

The minds of OUWB continue to provide thoughtful questions.
My roommate and I have encountered a question regarding the content on Sodium/Water Balance and also its application to SIADH. We have been using some outside resources to supplement the learning in class, and I feel that they have been somewhat contradictory in these 2 scenarios. The following are the scenarios that I am trying to think through
1) Patient eats a high salt meal, increasing total body Na+, resulting in an increase in ADH release (via increased plasma osmolarity) and eventually reaching baseline Na+ concentration and osmolarity at a higher ECV. Now, the increase in ECV would result in a down regulation of Sympathetic NS and RAAS; however, what I am hearing is that this down regulation would just return the kidney to Na+ in = Na+ out and would not actually return the individual to the original ECV. So, my question is how does this person get back to original ECV? What I am reading is that the person will continue to operate at this higher ECV until sodium restriction takes place. However, I am wondering how decreased RAAS (decrease aldosterone – decrease Na+ reabsorption – increase sodium excretion) wouldn’t do this, and also if pressure natriuresis wouldn’t do this also? Basically, why don’t these mechanisms do the work automatically, and why do you have to sodium restrict?
You have it right. That is the currently accepted understanding of sodium metabolism. It is not quite complete, because, though some subjects increase their blood pressure with increased sodium intake, not all patients increase their blood pressure. As to why the renin-angiotensin aldosterone system does not down regulate itself sufficiently to fully correct the volume overload situation, it is not well understood. The sodium regulating systems in the body strive to match sodium absorption with sodium excretion. With an increase in sodium intake there will be a modest expansion of the extracellular compartment until the sodium excretion is upregulated to match sodium intake. We can see evidence of the increase total body sodium with an increase in body weight associated with increased sodium intake.
2) In SIADH – high levels of ADH cause increased water reabsorption but euvolemic hyponatremia. Fitting in with my previous questions in the earlier scenario, how does the patient maintain euvolemic status? If increased water reabsorption occurs and the ECV is increased, the same down regulation of Sympathetic NS and RAAS would occur. Now, the outside resources in this case state that a decreased RAAS would actually cause increased sodium excretion that would allow for increased water excretion that would maintain euvolemic status. This makes sense because then the hyponatremia that results is not only an effect of the dilution from increased water reabsorption, but also from the increased excretion of Na+. But, this goes directly against the whole logic of needing to sodium restrict in the earlier case (i.e. RAAS can’t do the work to return the individual in scenario 1 back to a normal ECV).
So again you are well versed in what is happening in SIADH. SIADH is largely euvolemic and largely is a situation where patients are in sodium balance, i.e. sodium = sodium out. However if you do meticulous metabolic balance studies you will find that patients do gain weight during SIADH. There is excess water and this does serve to expand the patient’s extracellular volume. This also will suppress the renin-angiotensin-aldosterone-system so that patients will get a modest increase in urine sodium excretion. But I don’t quite understand how you think this is any different than the first scenario. There is a modest increase in sodium excretion but in the presence of continued unremitting ADH activity the patient continues to deal with the modest increase in volume. So like the first scenario, the modulation of the RAAS is unable to fully restore euvolemia.
For more on SIADH and volume status see this post.
Posted on

OUWB Question about pseudohyponatremia

First catch of the year.

I have a question regarding your OUWB lectures. I’m trying to grasp why hyperglycemia causes an increase in serum tonicity and decrease in serum sodium, but hyperlipidemia causes no change in serum tonicity and a decrease in serum sodium. For hyperglycemia, I understand that the glucose contributes to the serum osmolarity and can’t passively cross the membrane so causes water to move. However, I’m confused with the situation with lipids and was wondering if you could clarify. Thank you so much!

I may have over indexed on false hyponatremia stuff. This is something you need to be familiar with but a detailed understanding of the mechanism of pseudohyponatremia.

The student had perfect knowledge of the mechanism behind hyperglycemia induced hyponatremia associated with hyperglycemia.

The lipid situation is just a lab error. The lipids fool the lab machine into thinking the sodium is low. It is not low. That is why the osmolality is normal. The osmolality detector is not fooled by the high fats (or proteins) in the blood.

You will not need to know the mechanism for the lab error. I tried to explain it but that may be a situation where I causes more confusion than provided clarity.

The unexpectedly high protein or lipid fraction results in the sample being over diluted resulting in a false report of hyponatremia. The serum sodium is normal. Only about a third of clinical labs are susceptible to this error.