Tuesday, March 22, 2016

A novel look at hyponatremia in the alcoholics

Tavare and Murray in a recent NEJM image had an interesting case of hyponatremia correction. The case highlights development of central pontine myelinolysis(CPM) despite slow correction of hyponatremia.  CPM is known to occur in alcoholism, liver disease and malnourishment in the absence of hyponatremia, hypokalemia or hypophosphatemia. 

We wanted to suggest an algorithm that can be used in settings where alcoholics present with moderate to severe hyponatremia with similar symptoms as presented in this case and are at risk of CPM.  The figure below is a novel algorithm that uses brain imaging to help us guide the therapy for moderate to severe hyponatremia in alcoholics.  

If the patient is symptomatic with seizures, the correction of hyponatremia should be promptly started.  If the patient is asymptomatic  or with milder symptoms and is encephalopathic ( with several  confounding  etiologies : hyponatremia, alcoholism, liver disease), a MRI of the brain should be performed. If the MRI confirms cerebral edema, hyponatremia should be treated with the usual slow rate of correction of 6-9mmol/L per 24 hours.  If the MRI confirms CPM, the correction of hyponatremia should be put on hold.  

We hypothesize that often the hyponatremia  in alcoholics is  chronic  and correction, regardless of the rate, might cause harm in these patients.

We welcome comments from experts on this concept. 

Kenar D. Jhaveri, MD
Rimda Wanchoo, MD
Alessandro Bellucci, MD


  1. I appreciate the effort of developing this algorithm but there are several issues here:
    1. Even though over correction of hyponatremia is a major risk factor for ODS there are other risk factors. ODS have been associated with hyperglycemia and its correction, hypernatremia and its correction, and alcoholism without hyponatremia. So it is very difficult to determine from this single case if over correction of hyponatremia was involved in the development of ODS
    2. The goal of correction for a alcoholic hyponatremic patient is probably more close to 4 mEq/L and the limit of correction is no more than 8 mEq/L in 24h
    3. Instead of having an MRI brain showing CPM which in some cases does not appear abnormal after weeks of symptom onset, it would be better to say MRI brain without edema and then correct slowly.
    4. Finally, the team who took care of this patient did everything by the book. With the information we have of this entity, there is nothing different it could have been done to prevent this bad outcome. New literature actually suggest that most ODS cases are asymptomatic and the ones who present with symptoms have a favorable outcome most of the time

  2. Thanks for sharing this cute extra step, but I think rates as high as 6 to 9 mmol/L/day is on the high side for the 2nd step. As recommended by experts, we tend to adhere to the upper limit for the PNa to 4 mmol/hr if the first 6 to 9 mmol/L left the patient with cerebral edema would be my preference but you might want another MRI after the second cycle and this iteration may help minimize the risk of developing ODS and depending on the result decide to treat other causes.

    Mitchell Halperin, MD

  3. This comment has been removed by a blog administrator.

  4. This comment has been removed by a blog administrator.


All Posts

Search This Blog