Sunday, January 8, 2006

Treating Hyperkalemia (High Blood Potassium) According to the New 2005 CPR Guidelines

Hyperkalemia is a common problem that can range in severity from inconsequential to life-threatening. The treatments for hyperkalemia also vary widely and can include simply restricting dietary potassium; administering oral, intravenous or inhaled medications; and providing emergent dialysis for more extreme elevations.

Given a lack of standardization, it's not surprising that different doctors treat hyperkalemia in different ways. The new 2005 CPR guidelines from the American Heart Association provide recommendations for the treatment of hyperkalemia. Unfortunately, while these new guidelines are easy to follow, there are many potential problems, and I offer some criticisms.

For mild elevations (5 - 6 mEq/L), in addition to dietary and medication changes, the guidelines recommend removal of potassium from the body with
  1. Furosemide 40 - 80 mg IV. In my opinion, especially for slight elevations, in most cases intravenous diuretics are unnecessary, and oral furosemide could be just as easily substituted.
  2. Kayexalate 15 to 30 g orally in sorbitol (or by enema). While kayexalate is an important treatment for hyperkalemia, in my opinion, giving kayexalate routinely for any potassium elevation over 5 is a bad practice. It is often unnecessary and physicians frequently overlook the cramping, diarrhea, and discomfort it causes patients. Rarely, kayexalate in powdered form (which doesn't cause diarrhea) or occasionally florinef can be given to outpatients.
For moderate elevations (6 to 7 mEq/L), the guidelines recommend shifting potassium intracellularly. Previously, many algorithms suggested first obtaining an ECG to look for changes due to hyperkalemia -- "peaked" t-waves and new QRS widening -- and if either of these were present, the old algorithms recommended particularly aggressive treatment. The problem is, there is no good definition for what constitutes a peaked t-wave. (Various definitions have included "greater than 10 mm in the anterior leads" and "anything that looks like it would hurt you if you sat on it.") Another problem is that potentially, arrythmias may develop in patients without ECG changes. The new CPR guidelines do not recommend obtaining an ECG before treatment for moderate elevation. They recommend treatment with
  1. Glucose plus insulin (10 units of regular insulin and an amp of D50).
  2. Sodium bicarbonate 50 mEq IV over 5 minutes. In my opinion, for patients without metabolic acidosis, this is unlikely to be effective.
  3. Nebulized albuterol (10 to 20 mg) over 15 minutes. In my opinion, while this is an effective and rapid treatment for hyperkalemia, there are many patients with tachycardia or coronary artery disease who should not be given albuterol.
  4. The guidelines do not explicitly suggest treating moderate potassium elevations with kayexalate or furosemide, but these should also be routine treatments in my opinion.
For severe elevations (> 7 mEq/L with "toxic ECG changes"), the guidelines recommend shifting potassium into the cells, eliminating potassium from the body, and stabilizing the myocardial cell membrane. (Presumably, although not stated explicitly, severe potassium elevations without ECG changes should also be treated aggressively.)
  1. Calcium chloride 10%, 500 to 1000 mg (5 to 10 ml) IV over 2 to 5 minutes. Previously, many guidelines recommended 1 amp of calcium gluconate, which has about 1/3 the elemental calcium of calcium chloride. The common wisdom was that giving calcium chloride this rapidly had the potential to provoke arrythmias which outweighed the benefit of preventing arrythmias. Therefore, I found the recommendation to use calcium chloride (rather than gluconate) puzzling.
  2. Sodium bicarbonate 50 mEq IV over 5 minutes. See the above caveat.
  3. Glucose plus insulin (10 units of regular insulin and an amp of D50). In my opinion, this should be administered before sodium bicarbonate.
  4. Nebulized albuterol (10 to 20 mg over 15 minutes). See the caution about albuterol above.
  5. Furosemide 40 to 80 mg IV.
  6. Kayexalate 15 to 50 g in sorbitol orally or by enema. Kayexalate generally takes hours to work, so in my opinion, kayexalate should never be given to patients with severe hyperkalemia unless suggested by a nephrologist. Many patients with severe hyperkalemia require dialysis, which usually works faster than kayexalate. Being on dialysis with severe diarrhea from a medication you never needed is an unhappy experience for a patient.
  7. Dialysis is the ultimate treatment for hyperkalemia. Patients with end stage renal disease are routinely emergently dialyzed with severe hyperkalemia. For other patients with severe hyperkalemia due to acute renal failure (or other causes) the decision to dialyze is sometimes complicated and depends largely on the reversibility of the condition. Sometimes, patients with mild and moderate elevations of potassium are also treated with dialysis, depending on the situation.
For new posts, visit

(This post is for informational purposes only. Please see the disclaimer.)

Technorati Tags: , , ,


Richard V. Paul, M.D. said...

I agree, and would also like to point out:
1) Kayexalate would never be approved today - the original description of its effects, leading to FDA approval and marketing, included less than 20 patients. It is ineffectual (Kapral et al, Journal of the American Society of Nephrology, Vol 9, 1924-1930, 1998), unpleasant,and in combination with sorbitol can produce colonic necrosis, as documented in a NEJM CPC a few years ago. I can't believe anyone would recommend it for potassiums of between 5 and 6 - in fact, the normal range in most labs is up to 5.3 or 5.5, and mild elevation above this level is nearly always inconsequential.

2) I have seen 2 cases of full thickness skin necrosis of the forearm - 1 leading to a lawsuit that was settled for a substantial sum - from calcium chloride pushed into a presumably infiltrated peripheral IV. Calcium chloride has a very low pH and is painful to receive rapidly in undiluted form in a peripheral vein under the best of circumstances. IMHO,calcium gluconate should be the preferred agent, unless reliable central venous access has been established or the patient is near cardiac standstill.

Tyson Lewis said...

How does the Albuterol work in reversing hyperkalemia? I was taught to push Bicarb and Calcium if you've got significant EKG changes with alterations in mental status and/or cardiac arrest. And is it worthwhile to push the Lasix if the patient is in arrest?

To be honest, i definetly feel that metabolic imbalances is a weak point in my clinical knowledge. So one last question: I am a paramedic in New York city. If I have a patient with end stage liver failure in cardiac arrest that is unresponsive to standard therapies plus Bicarb administration, is it worthwhile to transport the patient still in arrest, or should we just pronounce?

Anonymous said...

You mention the patient discomfort associated with Kayexalate (sodium polystyrene sulfonate). This is from the drug itself, or just the sorbitol base? Would another liquid base, such as water alone, be better?

KidneyNotes said...

Answers to comments.

* Albuterol drives potassium into the cells.

* I don't think lasix is worthwhile if the person is in cardiac arrest.

* Re the person with liver failure: probably best to follow whatever the paramedic protocol is for pronouncements, but it sounds like they won't make it.

* Re Kayexalate: As an inpatient, kayexalate is usually mixed with sorbitol, which causes the discomfort. Kayexalate powder, which can be purchased as an outpatient, is mixed with water and actually is constipating.

Michael said...

Well, my particular grievance is that almost no physicians, and certainly no nurses, know that Kayexalate is intended to produce the cationic exchange of sodium and potassium in the large intestine. If Kayexalate is mixed with a laxative, like sorbitol, patients may certainly develop diarrhea--and that is a good thing if you are trying to unload potassium. But we do not give Kayexalate to produce diarrhea--in fact, Kayexalate is constipating, if anyhing. Still, it not rare to receive a call from a worried nurse who reports that a pt has received 30g of Kayexalate but has still failed to fill his sheets with diarrheal stool as intended.