An 82 year old man with a history of end stage renal disease, hypertension, coronary artery disease, and prior admissions for hyperkalemia is brought in by EMS with a history a 3 days of nausea, diarrhea, and fatigue, culminating in a witnessed collapse and a brief bout of bystander CPR. When the paramedics arrived on scene, he had a weak pulse and was only responding to painful stimuli. The ECG en route revealed a very strange appearing wide complex bradycardia, almost like a sine wave, and a VBG reveals a K+ of 9 mmol/L…
My approach to hyperkalemia
ABCs: Rapidly assess and manage the airway as necessary. Apply 100% oxygen. One nurse should apply the cardiac monitors and defibrillation/pacing pads while another is working on vascular access. I will start an IO if an IV cannot be placed within 2 minutes or after 2 attempts.
Call for help: It makes sense to notify whoever can arrange dialysis early, anticipating medical therapy may not work.
Stabilize the cardiac membrane
Key question: Is this patient on digoxin? Although the dreaded stone heart is probably a myth and I would not worry if I gave calcium and later discovered that the patient was on digoxin, digoxin toxicity has a specific antidote and is better treated with Fab than calcium.
Everyone else gets calcium: 3 amps or 30ml of calcium gluconate
- 3 amps (30ml) of calcium gluconate is equivalent to 1 amp (10ml) of calcium chloride
- Monitor the ECG continuously. You should see ECG improvement within 3 minutes. Dose can be repeated if there is no improvement
- Calcium chloride has a higher risk of tissue necrosis if extravasated
Alternate option: hypertonic saline
- Honestly, I am not sure why you would want to choose this over calcium. It has been shown to work, but only in patients with hyponatremia and concurrent hyperkalemia
- 50 ml of 3% saline pushed
Shift the potassium intracellularly
Insulin regular 10 units IV
- Make sure to flush after injecting the insulin, because it is a tiny volume that can easily remain in the port or IV tubing
- With insulin, give glucose. Dosing:
- If starting with a normal serum glucose: EITHER 2 amps (100ml) of D50W OR 1 amp (50ml) of D50W plus D10W run at 200ml/hr for 2 hours
- If starting with a markedly elevated serum glucose (>16mmol/L or 290mg/dL): No initial glucose
- Check glucose hourly x3. If less than 4mmol/L (72mg/dL) give 1 amp (50ml) of D50W
Albuterol 20mg nebulized, may repeat. (Unlikely to work in patients on beta-blockers)
Get the potassium out of the body
Key question: Can you use the kidneys? In patients with end stage renal disease who cannot excrete potassium renally, dialysis is the only option.
If hypovolemic, give fluids:
- Metabolic acidosis: isotonic sodium bicarbonate (3 amps of bicarb in 1L of D5W)
- No acidosis: Ringers lactate
- AVOID NORMAL SALINE. Although saline has traditionally been used to volume resuscitate and promote diuresis, it has actually been shown to induce hypercholermic metabolic acidosis and worsen hyperkalemia. (Of everything discussed, this probably has the strongest evidence. See this PulmCrit post for further discussion.)
- 80mg IV is my usual starting point, though there is no evidence to guide us
- Acetazolamide and mannitol are also occasionally used, and can be added for synergistic effect in the patient with life threatening hyperkalemia
If elimination of potassium through the urine is not successful, the patient requires dialysis.
The classic ECG progression in hyperkalemia is:
- Peaked T waves
- Prolonged PR interval and flattening of the P wave
- Widening of the QRS
- Sine wave
However. you cannot rely on the ECG to rule out significant hyperkalemia. There are a significant number of patients with hyperkalemia who do not exhibit any ECG changes (see Montague 2008 and Szerlip 1986 in the references.) Also, patients can progress directly from sinus rhythm to ventricular fibrillation (see Dodge 1953).
You will notice that sodium polystyrene sulfonate (kayexalate) is not included in this treatment plan. The evidence for SPS is extremely poor, but if it has any effect in acute hyperkalemia, that effect is not for many hours and comes with extra risk (colonic necrosis) that is not clearly outweighed by benefit. I never use SPS in the emergency department.
Does sodium bicarbonate work? The bicarb bottom line: We really don’t know. The best guess currently is it may be marginally effective as an infusion in the setting on metabolic acidosis. It does not appear to work in dialysis dependent patients, meaning that any effect seen is likely to be via renal elimination rather than intracellular shift. Boluses of hypertonic sodium bicarbonate (pushing 1 amp from the crash cart) have never been shown to be effective. In fact, hypertonic fluids increase serum potassium levels by shifting potassium out of cells.
Other FOAMed Resources
Weisberg LS. Management of severe hyperkalemia. Critical care medicine. 36(12):3246-51. 2008. [pubmed]
Pfennig CL and Slovis CM. Chapter 125. Electrolyte Disorders. In: Marx JA et al. eds. Rosen’s Emergency Medicine, 8e. Philadelphia: Elsevier Saunders; 2014.
Montague BT, Ouellette JR, Buller GK. Retrospective review of the frequency of ECG changes in hyperkalemia. Clinical journal of the American Society of Nephrology : CJASN. 3(2):324-30. 2008. [pubmed]
Szerlip HM, Weiss J, Singer I. Profound hyperkalemia without electrocardiographic manifestations. American journal of kidney diseases : the official journal of the National Kidney Foundation. 7(6):461-5. 1986. [pubmed]
DODGE HT, GRANT RP, SEAVEY PW. The effect of induced hyperkalemia on the normal and abnormal electrocardiogram. American heart journal. 45(5):725-40. 1953. [pubmed]
Morgenstern, J. Management of life-threatening hyperkalemia, First10EM, January 21, 2016. Available at: