Treating mild hyperkalemia

Somewhere along the road to becoming a doctor med students develop the hyperkalemia insulin reflex. See an elevated potassium give 10 units of IV insulin and 25 grams of D50. Yesterday I saw this done for a potassium of 5.6.

No. One. Dies. From. A. Potassium. Of. 5.6.

But there is morbidity from the insulin glucose antidote.

These guys from Rush University Medical Center looked at the risk of hypoglycemia from the standard 10 units of insulin followed by 25 grams of glucose. They found 13% of ESRD patients developed a glucose south of 60 mg/dl.

Remember the kidney both metabolizes insulin and is a source gluconeogenesis, so dialysis patients are naturally more prone to hypoglycemia.

 Love me some table one.

The patients that developed hypoglycemia were less likely to be diabetic, and had a lower glucose. In response to this study the authors suggested an alternative regimen of increased glucose monitoring and an additional 25 grams of dextrose:

Similar data was found in a study by Schafers. Schafers followed that study with this one, which suggests that weight based insulin dosing 0.1 units/kg up to 10 units could reduce hypoglycemia by about half:

What is truly terrifying is if some people start trying to really drive the potassium down with Sterns’ recommendations from KI:

Sterns suggests 6u bolus f/b 20u/hour insulin + 60 g glucose/hour for acute hyperkalemia in KI review https://t.co/vUTgibCyxE

ā€” Swapnil Hiremath, MD (@hswapnil) April 14, 2016

Also good factoid, there is less hypoglycemia if you add albuterol to insulin and glucose.

Systemic effects of formoterol and salmeterol: a dose-response comparison in healthy subjects

Let’s be careful out there and stop treating inconsequential hyperkalemia, let’s save our toxic antidote for truly toxic potassiums.

At what potassium would you give insulin + glucose?

ā€” Joel Topf, MD FACP (@kidney_boy) December 6, 2016