The Acute Treatment of Hyperkalaemia: A Systematic Review

Hyperkalaemia is relatively common. Severe hyperkalaemia is generally defined as K >6.5 mmol/L, however in neonates this increases to >7mmol/L.

This systematic review and meta-analysis assessed the effects of acute pharmacological intervention for the treatment of patients with hyperkalaemia, with and without cardiac arrest.

Our accepted approach to treating hyperkalaemia is:

1. Calcium to protect against arrhythmias
2. Insulin and beta2 agonist for intracellular shift of potassium and
3. Diuretics, potassium binders or dialysis for the excretion of potassium.

For patients in cardiac arrest the current approach is calcium, bicarbonate and insulin and glucose.

The Study
Jessen K.K et al. Pharmacological interventions for the acute treatment of hyperkalaemia: A systematic review and meta-analysis.Resuscitation 208 (2015) 110489.

What They Did

This was a review of acute pharmacological interventions, which included calcium chloride or calcium gluconate, bicarbonate, insulin/glucose frusemide and beta2 agonists.

101 articles were included for review.

What They Found

Cardiac Arrest Studies

Two observational studies( one adult and one paediatric) were included and both used Calcium.

• In adults the use of calcium, bicarbonate, or both resulted in a lower rate of spontaneous circulation
• In neonatal and paediatric patients calcium was frequently used in cardiac arrest and associated with worse outcomes.

Non-Cardiac Arrest Studies

Insulin plus Glucose/Dextrose/Fructose

Some of the findings included:

• 7 observational studies used 5U of insulin with a potassium change of -0.6 to -1.2 mmol/L within 360 minutes.
• 30 studies used 10U insulin with a change in potassium of -0.7 to -1.4 mmol/L within 420 min.
• 7 studies used 8-12U insulin (66 patients in total) in patients with hyperkalaemia and renal failure and found a mean change in potassium of -0.7 after 30-60 minutes.

Inhaled Beta2 Agonists

• 7 adult studies with a total of 87 patients were included and found that following 10-20mg inhaled salbutamol, there was a -0.9 change in potassium within 120 minutes.
• 3 neonatal studies (51 patients total) used 400mcg of inhaled salbutamol for a potassium change of -0.9 mmol/l within 240 minutes.

Intravenous Beta2 Agonists

• In 6 adult studies (100 patients total) 0.5mg IV salbutamol dissolved in glucose/dextrose decreased potassium by -1.0 mmol/l within 120 min.
• 4 studies (53 patients total) in neonates/paediatric patients used 4-5 mcg/kg IV salbutamol found a change in potassium of -1.0 mmol/l within 60 min.

Bicarbonate

5 adult studies were included with a total of 44 patients.

• use of bicarbonate doses of 50-390 mmol resulted in a -0.1 mmol/l change in potassium within 60 minutes.

Combination Treatments

• 3 studies (25 patients) using combinations of 0.5mg Salbutamol, 10U insulin and 25-40g glucose found a change in potassium of -1.2 mmol/l

Takeaway

This is an interesting study as it challenges our current accepted efficacy of hyperkalaemia treatment. Although the evidence was of low to very certainty, in some cases, it found, that in general, in patients with hyperkalaemia:

• The use of  insulin combined with glucose, beta2 agonists inhaled or intravenous (dissolved in glucose) resulted in an average reduction of potassium of 0.7 to 1.2 mmol/l.
• Bicarbonate did not really cause a change in potassium levels
• No evidence was found for a clinical benefit of calcium. This review found no evidence for its ability to prevent arrhythmias.
• Intravenous salbutamol is effective in reducing potassium.

We need to beware of interpreting the results in this meta-analysis. It included human and animal studies and made assumptions on the comparability of studies. The studies were small and they challenge what we sometimes clinically see. The effect clinically was not discussed, only the drop in the potassium levels.

I’m not sure what the significance of the time frames in the emergency department. I need what i’m giving to work quickly, not in 300 minutes etc.

I will take it on board that insulin and glucose and beta2 agonists work, however I cannot, not use calcium and bicarbonate. In terms of calcium in cardiac arrest, we know that it is associated with decreased survival, however in cardiac arrest secondary to hyperkalaemia, the evidence is so sparce that I cannot not use it.