Category: Uncategorized

Posted on

Have a nephrology question? #AskRenal to the rescue.

A couple of days ago, this came across my notifications

Prolonged OTC ingestion of sodium bicarbonate causing extremely low K+ (<1.5)
Any ideas as to the mechanism? @kidney_boy #askrenal

— Morgan (@Morgansb) September 27, 2017

The answer came quick.

How does bicarbonate cause hypokalemia

The video is here (complete with misspellings) and the Keynote file is here (with misspellings corrected)


How metabolic alkalosis causes hypokalemia from joel topf on Vimeo.

Posted on

The beautiful #VisualAbstracts of the NEJM

This summer (I think) the NEJM began publishing visual abstracts on their twitter feed. Curiously, I was unable to find them on the page of the article that the visual abstracts references, or in the list of media types that you can search for.

The figure list on the right side does not include the striking visual abstract they created.

The “Browse Figures and Multimedia page has 19 different types of media, but visual abstract is not one of them.

The only way I could round up the visual abstracts was scrolling through the the NEJM Twitter feed. Here are the ones I found. Did I miss any?

Gorgeous work. Each one has a unique color palette and they have a pretty simple template, but three different ways of executing it. All of them look like they are from the same family except the tiotropium visual abstract. I really like that they give both the percentages and the raw numbers. No P-values or confidence intervals are found. These visual abstracts have as low an information density as I have seen. This is not a criticism, I think meh style has been increasing complexity to the detriment of my work. I need to turn up my inner NEJM.

Posted on

Another question from OUWB

Hi Dr. Topf 

First of all, apologies for sending this via email but I do not have a Twitter account (I know, its the 21st century, who doesn’t?). 

I had a quick question regarding a practice problem I was doing. Rather than summarize the question for you, I included a screenshot so that you have the primary source with the explanation provided. Below, I also included my explanation for my reasoning for choosing that option. Basically, I am confused as to why the bicarb would be decreased in this scenario.

So the stem describes acute trauma. Specifically crush injuries, so you should be thinking rhabdomyolysis where the body gets turned inside out. In my very first lecture we talked about the intracellular atmosphere versus the extracellular atmosphere:

So expect increased potassium and phosphorus.
The vital signs show a patient with circulatory insufficiency, i.e. shock. There are some initial labs drawn from the blood and urine right before resuscitation is initiated. The urine shows an osmolality of 800 mmol/kg H2O (highly concentrated, indicating a lot of ADH activity) and a urine sodium of 5 mEq/L (very low, indicating increased activity of the renin angiotensin aldosterone system).
The question then asks you to predict the serum labs. Cool question. The best testing strategy here is to cross off ones that make no sense. Here are the foils:
BUN. This should be elevated as the patient moves to a pre-renal physiology. This leads to an increased filtration fraction to maintain GFR in the face of decreased renal plasma flow. This causes an increase in the osmolality in the efferent arteriole and vasa recta. this increases osmotic reabsorption of fluid from the proximal tubule. BUN flows passively with the fluid, decreasing renal urea clearance and increasing serum BUN. So D is wrong. E is wrong.
Potassium (K+) ions. All the choices show that it rises as the rhabdomyolysis from the crush injury releases loads of intracellular potassium. No answers are eliminated here.
Sodium (Na+) ions. We have a mishmash of choices here. This is difficult to predict. The increased ADH released due to shock would tend to lower the sodium. Increased aldosterone and renin would tend to increase the sodium. Since both are happening together I would expect them to balance each other out resulting in no change in sodium concentration. This is especially true since the stem specifically says there has not been much urine output I would go with D, for no change in sodium concentration. NOTE: About activation of the RAAS as a cause of hypernatremia. Hypernatremia is commonly listed as a symptom of Cushing syndrome and hyperaldosteronism so it is possible to have hypernatremia from (at least the pathological) activation of the renin angiotensin aldosterone system) but this is very unusual as increased in sodium concentration stimulates thirst which dilutes the sodium back to normal.
THIS COLUMN IS MESSED UP. THIS IS AN ERROR BY THE QUESTION AUTHOR.
Hydrogen (H+) ions. All the choices show that it rises. In shock we expect an increase in hydrogen ions as patients move to anaerobic metabolism due to inadequate perfusion (the functional definition of shock). No answers are eliminated here.
 
Bicarbonate (HCO3-) ions. Bicarbonate is the primary buffer in the body. Increases in hydrogen ions will be buffered by bicarbonate and bicarbonate will be consumed.
So the right answer is bicarbonate will be decreased. This eliminates answer A.
This leaves us with B or C and the question hangs on what the sodium will do and the reality is the change in sodium is unknowable. Shit question. Sorry.
Here is what Kaplan claims to be true:
The forth bullet point is the one where the question fails. It is true that there is accumulation of plasma electrolytes in acute kidney injury. The problem is that these are electrolytes are measured as concentrations and there is also an accumulation of water (which is normally excreted by the kidney). This means that the effect on concentration is variable. Some, like hydrogen and potassium, reliably increase in AKI, but sodium is often decreased in AKI. Maybe the question writers were looking at an unconventional way measuring ions in the the plasma (as total amount rather than concentration).
The line in the answer that pre-renal azotemia is associated with hypernatremia is just wrong. You will encounter numerous patients with pre-renal disease that will have simultaneous hyponatremia. It is impossible to predict the serum sodium concentration from the volume status. This question reinforces the worst instincts of med students when it comes to predicting serum sodium. As I emphasized in the lecture, volume regulation and osmoregulation have two different regulatory systems and, though there is some cross talk, they are largely independent from each other.
The last paragraph tries to make the case that hyponatremia is more common in ATN while hypernatremia predominates in pre-azotemia. This is total fiction and does not exist. Though there can be more urinary sodium in ATN, if the patient is oliguric, it doesn’t matter how high the urine sodium concentration is, if the urine volume volume is close to zero there will not be much urinary sodium excretion.
Distinguishing between pre-renal azotemia and ATN is a constant problem in nephrology. Trust me you can’t make the determination by looking at the serum sodium. It aint that easy.
This question writer should never write another question about sodium.

 I posted this to twitter. The subsequent discussion was pretty interesting:

Kaplan Blows it on Hypernatremia and AKI

Posted on

Medical student questions about nephrology

I have the honor of teaching at Oakland University William Beaumont School of Medicine. I teach sodium and water and acid-base to the second year medical students. After the lectures there is a steady stream of questions that start to fill my in-box. I answer the e-mail but I also post the questions and answers on PBFluids. Here is a directory to this year’s crop of Q&A.

Sodium concentration versus sodium content. With a second question on pseudohyponatremia vs false hyponatremia vs factitious hyponatremia

Macula densa and TG feedback. With a second question on whether SIADH is really euvolemic or just mostly euvolemic.

Breaking down an acid-base question.

More on euvolemic hyponatremia and how does this affect uric acid.

Urine chloride in non-anion gap metabolic acidosis, where does it come from?

Starling forces and GFR. With additional Q&A on edema, and metabolic acidosis and ammonia-genesis.

Posted on

The Fluid Electrolyte and Acid Base Companion

After Medical School Sarah Faubel and I set out to explain sodium, potassium and acid-base in a programmed text. Four years later, in late 1999 we finished this journey. The result was The Fluid, Electrolyte and Acid Base Companion.

Get it right here, for free (PDF, 29 mb):

If you would prefer a zip file with each chapter in individual PDF files, click here.

5 stars on Amazon

Highly reviewed by Beaver Medic

And by Mark Yoffe.

Posted on

Dropbox problems

A while ago Dropbox changed how they handled public links. Then they announced that old links with the previous public folder system would be unsupported. I have no idea how many or where these links are littered through out PBFluids, but I suspect as of September 1, there will be a lot of them.

Today I received this tweet.

@kidney_boy Hi Dr Topf, I wanted to send med students your book, link says ‘File Not Found’; any advice? – https://t.co/hQ3nOGVRo6

— Paul Adams (@MSUPaul) September 7, 2017

So this bug struck the prized link on the whole damn blog. Annoying. Blogger is the only platform that doesn’t host files and forces you to store them elsewhere. This is the root cause of the problem. WordPress and SquareSpace allow you to host your files on their servers so you wouldn’t run into this problem reconnecting two services.

That’s like strike seven against Blogger. I’ve got to get out of this burning pile.

Posted on

OUWB Starling forces question


I was hoping I could ask you a few questions. I’m finding there is a lottttt of contradictory information.

  1. According to starling forces, decreased plasma oncotic pressure should increase GFR, but according to nephrotic syndrome, decreased albumin will cause edema and overall decrease GFR. Which one should I believe? 
  2. In general, it’s said that AT2 at low levels dilates the afferent arteriole to increase GFR, but at high level it constricts both efferent and afferent to decrease GFR. However, the SNS, which stimulates renin, constricts all arterioles in the body as well as activates the RAAS system. How does that work? Is the SNS more immediate until the aldosterone system is ready to say okay go ahead and dilate the afferent I’m ready to take up the water anyway? 
  3. This is a very basic question but sometimes I have moments of self doubt and this is one of them: So we always say edema is fluid buildup in ISF due to increased hydrostatic or decreased oncotic pressure (like nephritic syndrome hypoalbumineia) right? So why does fluid build up in ISF as opposed to go inside the cell where I guess technically there is more stuff to pull it in? 
  4. How does K suppress ammonia genesis? 

Thank you very much!

Let’s take these one by one,

NUMBER ONE
Nephrotic syndrome and GFR. Don’t connect those neurons. Proteinuria does not cause an immediate and hemodynamic change in GFR that is clinically meaningful. Yes, you are right that lower oncotic pressure should increase GFR, but those increases in GFR will be trimmed by tubuloglomerular feedback so that in the end there is not a meaningful change in GFR. Likewise the nephrotic syndrome will cause fluid to leak from the blood vessels decreasing effective circulating volume lowering renal plasma flow. However, once again these changes in volume are small enough that the kidney easily compensates with changes in AT2, PGE, filtration fraction, etc so that GFR remains stable.

Over a long time, proteinuria causes chronic kidney disease and decreases renal function, but not by the mechanisms you described.

Of note the model you are talking about with nephrotic syndrome causing fluid to leave the blood vessels and that resulting in decreased perfusion of the kidney is a model called underfill hypothesis of edema in nephrotic syndrome. Most nephrologists now ascribe by the overfill hypothesis which states that the primary abnormality is not loss of fluid from the capillaries from the decreased albumin, but increased sodium absorption by the diseased kidney. This results in volume overload and that causes the edema.
NUMBER TWO
As I understand it, angiotensin is only a vasoconstrictor. The proximal tubule is dilated by prostaglandin E. In volume depletion there is release of renin which activates angiotensin 2 (with help of angiotensin converting enzyme). Angiotensin 2 vasoconstricts both the afferent arteriole and efferent arteriole. But since the afferent arteriole is so much bigger to begin with, after the angiotensin 2 induced vasoconstriction the resistance in the afferent arteriole is less than the resistance in the efferent arteriole, this serves to increase the intraglomerular pressure, forcing more plasma through the glomerular slit membranes and increasing the filtration fraction and maintaining GFR in the face of volume depletion.
And yes the SNS is more immediate and the renin angiotensin aldosterone system is a bit slower.
NUMBER THREE
Where fluid builds up depends on what is being altered. In nephrotic syndrome, the (underfill) theory states (I’m an overfill believer) that decreased plasma albumin lowers the oncotic pressure drawing fluid from the interstium back into the capillaries at the venous end. This means more of the fluid remains in the interstium leading to edema. The oncotic agent of note here is albumin which determines the flux of fluid between the interstitial and plasma compartment.
In order to shift fluid between the intracellular and extracellular compartments you would need to change sodium and potassium which are the chief osmotically active particles of interest between those two compartments.
NUMBER FOUR
Hyperkalemia causes potassium to shift into the cells. To maintain electroneutrality hydrogen leaves the cell. One cation in, one cation out. The loss of hydrogen ions makes the cell alkalotic. This rise in pH tricks the proximal tubule cell into believing the entire body is suffering from metabolic alkalosis and since ammonia generation is used to increase hydrogen excretion, and correct metabolic acidosis, metabolic alkalosis shuts down ammonia generation.
Posted on

OUWB Non-anion gap question

Where does the chloride come from.

XXXXX and I had a question following the Acid-Base workshop. What is the origin of the increase in chloride ions in patients with NAGMA due to GI or renal causes?  

Thanks, 

 

So the key here is not to think of the body as static. As patients lose bicarb in the stool or in the urine, this will result in volume depletion which will be compensated for by renal retention of sodium and yes, chloride.

Great review of non-anion gap metabolic acidosis here:

Posted on

OUWB Euvolemic Hyponatremia question

Hello Dr. Topf,

I hope you are enjoying your weekend. I had a question in regards to one of your lectures. I was wondering why there is a low level of Uric acid in euvolemic hyponatremia but not in hypervolemic or hypovolemic hyponatremia. Also, how is it that Na taken in equals Na excreted in euvolemic hyponatremia?

All the best,

 
So why is there a low level of uric acid with euvolemic hyponatremia? Let’s first look at what happens to uric acid in the other causes of hyponatremia, namely hypovolemic and hypervolemic. In both of these situations the kidney is experiencing decreased perfusion, either from absolute volume depletion (diuretics, diarrhea) or perceived volume depletion from pump failure (CHF) or fluid maldistribution (cirrhosis and nephrotic syndrome).
In these volume depleted states there is an increase in the filtration fraction, i.e. more of the plasma that enters the glomerulus is actually filtered. This is how the kidney compensates for a decrease in renal plasma flow while maintaining GFR, it increases the fraction of fluid that is filtered.
A consequence of this, is that the oncotic pressure in the blood leaving the glomerulus is higher because more of the fluid (but none of the protein) has gone down the glomerular drain leaving the plasma in the efferent arterioles with a higher oncotic pressure.
This plasma then enters the vasa recta where it surrounds on the proximal tubule. Here the increased oncotic pressure pulls more fluid back.
This is an ideal situation. The increased filtration fraction maintains GFR in the face of decreased renal plasma flow, and the increased filtration fraction results in enhanced reabsorption of fluid in the proximal tubule limiting fluid loss in situations where patients have decreased perfusion.
Uric acid handling is complex and not fully worked out.
It appears that there is both uric acid secretion and reabsorption in the proximal tubule.
Functionally, uric acid clearance tracks with renal perfusion:
  • Decreased uric acid clearance with decreased renal perfusion
  • Increased uric acid clearance with increased perfusion of the kidney

This is similar to what we see with urea. The following description of urea handling gives a model that will work for uric acid, though the truth of uric handling is much more complex.

The key with urea is that it’s handling in the proximal tubule tracks with total fluid reabsorption in the proximal tubule.
With volume depletion, increased filtration fraction causes increased oncotic pressure in the vasa-recta increasing urea reabsorption in the proximal tubule.
In volume overload, decreased angiotensin 2 decreases sodium reabsorption resulting in less fluid reabsorption and less passive reabsorption of urea  so increased urea loss in the urine and lower serum urea.
Now what happens in euvolemic hyponatremia.
Sodium in equals sodium out. This means that these patients do not have a primary volume abnormality as we see in the hypovolemic and hypervolemic patients. Because of this their sodium regulation volume regulation system is not stressed, they are at homeostasis with regards to body sodium. When you are in homeostasis, in order to stay in homeostasis you need to excrete all the sodium that comes in. In other words sodium in equals sodium out.
However these patients are not in water balance. they have a disease that forces their ADH to 11. They have a fixed ADH secretion and it is set at full blast. This minimizes urinary water excretion, but they are able to stay in sodium balance. So the net of this is they make only a little bit of urine but that small amount of urine carries all of their ingested sodium (sodium in = sodium out) so the sodium is excreted in a small volume at a high concentration.
Now the obvious problem here is that they are holding on to an excess of water. And that will increase their total body volume. This is subtle and doesn’t cause edema, or heart failure, or fluid overload in the lungs, but it is there. This fluid overload suppresses angiotensin 2 and decrease sodium resorption in the proximal tubule and hence decreases urea (and uric acid in our model) reabsorption.
And yes this does mean it is not exactly sodium in = sodium out, there will be a slight excess of sodium excretion.
Posted on

OUWB Question: Acid-Base

Hi Dr. Topf,
(I don’t have Twitter) I wanted to ask you about question 6 on the week 2 quiz:
“An unresponsive woman is brought to the emergency room. She has a history of a suicide attempt a few years earlier. The lab tests are: Serum Na 140 mmol/L Serum K 4.0 mmol/L Serum Cl 100 mmol/L Serum HCO3 14 mmol/L, BUN 17 mg/dl, creatinine 0.7 mg/dL, serum osmolality 323 mOsm/Kg, Blood glucose 72 mg/dl, Blood gases: pH 7.28 pCO2 27 mmHg. What would you expect the urine pH to be in this patient?”
Why is it that we would expect the urine pH to be acidic? Since blood pH is 7.28, I would imagine that urinating out HCO3- (explaining the low serum HCO3) would have caused the acidic blood pH, thus making urine pH basic?

 

Thanks for your help,

When answering multiple choice board-style question try to figure out what they are looking for. Let’s break this down.
“An unresponsive woman is brought to the emergency room. She has a history of a suicide attempt a few years earlier.

This is the “tell” of the stem. Acid base + suicide = ethylene glycol toxicity

The lab tests are: Serum Na 140 mmol/L Serum K 4.0 mmol/L Serum Cl 100 mmol/L Serum HCO3 14 mmol/L, BUN 17 mg/dl, creatinine 0.7 mg/dL , Blood glucose 72 mg/dl, 

They don’t tell you the anion gap. Calculate it.

Anion gap = Na – (Cl + HCO3)
Anion gap = 140 – (100+14)
Anion gap = 26 (normal 6-12)

High anion gap.

serum osmolality 323 mOsm/Kg

More of the tell. They won’t tell you the osmolality unless they want you to calculate the osmolar gap (or it is a hyponatremia question)

Osmolar gap= Measured osmolality – (Na x2 + glucose/18 + BUN/2.8 + ethanol/3.6)
Osmolar gap = 323 – (280 + 4 + 6 + 0)

Osmolar gap = 323 – 290
Osmolar gap is a massive 33 (Upper limit of normal is 10, over 20 starts to gain a lot specificity for toxic alcohol)
This confirms our earlier suspicions of ethylene glycol toxicity

Blood gases: pH 7.28 pCO2 27 mmHg. What would you expect the urine pH to be in this patient?”

The ABG confirms the metabolic acidosis.

Let’s do Winters formula (not really needed for this question, but you know…practice)
1.5 x 14 =21 + 8 =29, measured CO2 is within ±2 of predicted so an appropriately compensated metabolic acidosis.

Why is it that we would expect the urine pH to be acidic? Since blood pH is 7.28, I would imagine that urinating out HCO3- (explaining the low serum HCO3) would have caused the acidic blood pH, thus making urine pH basic?

So the bicarbonaturia you are talking about would happen if the cause of the metabolic acidosis is renal loss of bicarbonate (what we call renal tubular acidosis).

RTA should only be considered if you are dealing with an normal (or non-anion gap) metabolic acidosis. Since we have an anion gap metabolic acidosis and functioning kidneys the kidneys will be working as hard as possible to clear the exogenous acid. This means the urine is acidic.

The urine would also be acidic if the patient had a non-anion gap metabolic acidosis from diarrhea.

Hope this helps