The Pre-brief
This interesting (and not so straightforward) case of hypernatremia is a great review of both physiology and clinical care.
This was a patient with severe COVID-19. Pt had developed acute tubular necrosis and required temporary renal replacement therapy but was now producing urine and both Cr and BUN were getting better. ARDS was also getting better and extubation felt close.
Then, patient developed fever and severe diarrhea 2-3 L/day that showed no improvement when enteral nutrition was discontinued. Patient developed hypokalemia and severe non-anion gap metabolic acidosis. Primary team replaced potassium loses and replaced intestinal fluid loses with isotonic bicarbonate solution. CT Scan revealed diffuse bowel inflammation. C. Diff toxin and fecal cultures were negative.

Laboratory tests revealed a serum Na+ of 160. For the next 3 days the patient was treated with oral free water (about 80 ml/hr for a calculated free water deficit of 2 L to improve Na+ at a rate of 10 meq/day). There was no improvement of serum Na+ (it remained at 156 meq/L).
There was concern about gut water malabsorption.
Question 1: Why aim for a correction rate 10 meq/day?
This was acute hypernatremia, the concern about rapid osmotic changes leading to brain edema do not apply (may even be exaggerated in the case of chronic hypernatremia). The correct aim should have been to normalize serum sodium with aggressive water repletion.
Question 2: What was the cause of hypernatremia?
Hypernatremia is very rarely the result of gigantic loads of exogenous Na+ administration. More frequently it is due to water losses:
[Na+] = (total body Na+ + total body K+)/total body water
The only way to increase serum sodium is to lose more water than salt (isotonic fluid losses will lead to hypovolemia, not to hypernatremia), so the cause of hypernatremia is almost always loss of electrolyte-free water (without Na+ or K+). This may come from renal, intestinal or skin (insensible) loss of water.
Question 3: Was diarrhea contributing to loss of electrolyte-free water?
This was very unlikely. While diarrhea can definitely cause free-water losses, this is always in the context of osmotic diarrhea. For example, the use of lactulose can create intestinal fluid losses that are low in sodium + potassium and thus, more water is lost than electrolytes.
To definitely rule this out, I measured electrolytes in stool:
Stool Na+ = 144 meq/L and Stool K+ = 14 meq/L
Electrolyte concentration of stool was exactly the same as serum sodium, this means fluid loss from diarrhea is not causing dehydration. However, fluid loss from diarrhea was causing VOLUME depletion and metabolic acidosis. Thus, this fluid loss has to be replaced with isotonic bicarbonate (according to the principle of pH guided fluid resuscitation
Question 4: If not the bowel, where was electrolyte-free water loss coming from?
In this case, all was needed was to look at urine output. Patient was producing urine at a rate of 96 ml/hr. For someone that’s dehydrated, this definitely feels abnormal.
Was this Diabetes Insipidus? Urine Osmolarity was 480 mosm/L. Definitely not DI. However, urine should have been maximally concentrated in the setting of hypernatremia. This was most likely related to still incomplete recovery of tubular function from ATN.
To determine if this urine output is generating loss of electrolyte-free water a sample of urine was sent to the lab:
U Na+ = 16 meq/L, U K+ = 24 meq/L
Such low values suggest most urine output is actually electrolyte-free water.
The formula for electrolyte-free water excretion is:
Urine Volume x {1 – (U Na+ + U K+ )/[Serum Na+]}
So, using this formula with a urine output of 96 ml/hr (2300 ml/day) we get:
2300 x {1 – (16 + 24)/[156]} = 1710
This means that this patient is losing 1710 ml of electrolyte-free water through the kidneys every day, or 71 ml/hr.
The treatment with oral free water that patient had been receiving the last 3 days was barely enough to replace urine losses!
Question 5: Replacing free-water accounting for ongoing losses
In order to normalize serum Na+ it is necessary to replace ongoing losses + water deficit. Urinary loss of water was:
1710 ml/day + insensible losses (arbitrarily calculated as 600 ml/day) = 2310 ml/day (96 ml/hr).
In addition, one needs to give additional water to correct the existent free-water deficit. Given that 3 days had passed and this was no longer “acute hypernatremia”, a goal of Na+ correction of 10 meq/day was set (aiming for a serum Na+ of 145 meq/L). Using med-calc, the free water deficit was estimated at 2000 ml.
So, in addition to ongoing water losses (2310 ml/day), we had to give an additional 2000 ml to account for the water deficit. This meant giving 4310 ml of water in one day (180 ml/hr).
Of course these calculations do not obviate the need for frequent lab checks to see if you are on the right track!

Question 6: Why was the kidney loosing free-water?
This was a combination of a concentration defect (from ATN) and increased urea excretion resulting from recovery from kidney failure. 1) ATN lead to increase in BUN, 2) after ATN resolved, a negative nitrogen balance is expected in order for BUN to return to normal, 3) This causes increased urea excretion
Just like giving urea tablets improves hyponatremia (because urea increases urinary excretion of free water), increased urea excretion in this case this lead to free water loss and hypernatremia!

Question 7: What caused the diarrhea?
Colonoscopy revealed aphtous ulcerations along the entire colon and short bowel. Biopsy reveled cytopathic inclusions. No organism was found. PCR for CMV and Herpes virus was negative. It resolved spontaneously after a week. So, I do not know…
The Debrief
- Hypernatremia should be treated aggressively to provide patient comfort (thirst)
- Water replacement should include free water deficit + ongoing water losses
- Secretory diarrhea is isotonic and does not cause hypernatremia; Only osmotic diarrhea will
- Increased urea excretion in the setting of recovery from kidney failure is a common cause of free water losses
- Ongoing urine losses of water can be calculated with the formula for electrolyte-free water clearance
References
Rose BD, Post TW. Clinical Physiology of Acid-Base and Electrolyte Disorders, 5th ed, McGraw-Hill, New York 2001. p.775