Tuesday, January 12, 2010

HyperKalemia with Cardiac Arrest. Peaked T waves: Hyperacute (STEMI) vs. Early Repolarizaton vs. Hyperkalemia

For more on hyperkalemia, click here (includes a great case of incessant ventricular tachycardia).
For more on early repolarization, see these cases:
For differentiation of early repol from LAD occlusion, click here.

Hyperkalemia resulting in cardiac arrest

Although there is little high quality data on differentiating these entities, some general insights are useful, and illustrated with the following cases:

Case 1

The first ECG is one of a 27 y.o. patient who presented with ventricular fibrillation.

ECG #1, hyperkalemia

Because of the large T-waves, this ECG was interpreted as "hyperacute T-waves". However, these T waves are pathognomonic of hyperkalemia because they are peaked, "tented", come to point, have a very flat ST segment, and there is a long QRS (114 ms).

Conventional wisdom (with no hard data, to my knowledge) says that when hyperkalemia has a normal QRS that the QTc should be short. This idea conforms with the complex electrophysiology of hyperkalemia, but may not always be true in real life.

Case 2

ECG #2, LAD occlusion

In this ECG#2 above, the T-waves are slightly more blunt at the peak, there is a normal QRS duration with a long QTc at 450 ms, the ST segment is straightened (less upward concavity, steeper ST segment) which results in an area under the curve (integral) that is larger than in either hyperkalemia or early repol (i.e., the T-wave is "fat")

Case 3

This patient presented feeling moderately ill:

EKG #3, hyperkalemia, QTc 497 ms, QRS 102 ms (normal is less than 110ms)
See V4 especially.  The ST segment is horizontal until it abruptly rises to a very peaked T-wave.  The T-wave is "tented" to a point.

Due to the peaked T-waves, the residents were immediately concerned for hyperkalemia and sought an old ECG, which they immediately found (shown below, ECG #4):

EKG #4, early repolarization, QTc 455 ms, QRS 82 ms

Notice the ST segments here are not nearly as flat, and the rise to a peaked T-wave, especially in lead V4, not nearly as steep

After viewing this previous ECG, and knowing that the K was 4.5 at the time it was recorded, the residents believed that the peaked T waves in ECG #3 were this patient's baseline. It is true that early repolarization has tall and relatively peaked T waves, but not to the extent seen in ECG #3. Without seeing them side by side, it is hard to appreciate the difference, but the ST segment in V4 in EKG #3 is flat, making the base of the T wave much more narrow.

This patient, then did not get immediate treatment for hyperK. 

I saw these ECGs at a slightly later time than the resident, recognized the difference and, worried about the patient, started toward his bedside.  As I was approaching the patient, he had a v fib arrest.  He was immediately resuscitated, then his K returned at 7.0 mEq/L.   This was a presumed hyperkalemic arrest.  

Some say you don't need to treat hyperK unless there is QRS widening.  They claim that peaked T-waves are not enough.  This is only one case, and anecdotal, but we found no other etiology of arrest in this patient.  I always treat immediately if the ECG is affected by hyperK.

In addition, the QRS duration difference is important; the difference in QTc seems to defy conventional wisdom.

There is a definite difference, with EKG#3 pathognomonic for hyperkalemia.


  1. please i cant understand difference between early repolarization and hyperkalaemia

  2. HyperK has a tented T-wave, as if pulled up by a string to a sharp point. The T-waves are generally symmetric. In Early Repol, by constrast, there is a blunted peak to the T-wave, and the T-wave is assymmetric: slower upstroke than downstroke. HyperK may also, of course, manifest PR lengthening and QRS widening.

  3. For the LAD occlusion, would it be beneficial to the myocardium to give the patient calcium chloride or gluconate to decrease the stress on the heart?

    Also if I remember right if the patient goes into cardiac arrest and hyper k+ is suspected then high doses of CaCl (2-3g) might be therapeutical especially for refractory v-fib. At that time, during hyper k+, does amiodarone have any benefit for the patient?

  4. I know of no evidence that calcium would help in MI, nor of a mechanism that would support it. There were studies of GIK (glucose-insulin-potassium) that showed promise, but were not confirmed.

    Amio may be useful for hyperK, but why not just give calcium, as in this case in which I gave 15 amps of calcium gluconate:


  5. Can you kindly elaborate on differentiating LAD occlusion from the other two entities?

  6. The best way to differentiate hyper K is to measure the K! But hyperK has very tented T-waves, whereas LAD occlusion has bulky and wide T-waves whose peak is more blunt.

    Look here for how to differentiate LAD occlusion from Early repolarization:

  7. When treating hyperkalemia with peaked T waves but normal QRS width, do you just use shifting techniques only (b2 agonist, insulin/dextrose, bicarb)? And reserve calcium for QRS widening? There are some reports of tachyphylaxis to calcium when you really need it when calcium is used too early.

  8. I have not heard of tachyphylaxis to calcium. Can you provide some references?

  9. I think that comparing ekg 3 to 4 shows no real difference, especially with the qtc not following the rules. That V4 finding is so subtle, I can't believe anyone would treat based on that and a qrs of 102, which is still normal (without knowing the K level, of course.)
    Is this a case of the old retrospectogram?

  10. I personally took care of the patient and saw the difference just minutes before the arrest, so it is not ."the retro spectrogram". You need to look more closely if you cannot see the differenc. It is clear.

  11. The guys over at Emrap tv have a video on big twaves where they state hyperacute Ts are assymetric where as your blog and multiple other texts list hyperacute Ts as symmetric. Perhaps they were discussing the asymmetry associated with ste + the twave instead of isolated big t waves. Is symmetry useful at all. Tenting is easy to pick up on, I'm more concerned about getting burned on lvh with strain given the t waves should be assymetric

  12. I am not sure what EMRAP is saying, or who their expert is.

    I would say that symmetry is specific, but not sensitive, for pathology (whether AMI or hyperK).

    Perhaps it is better to say that as myocardial ischemia progresses, the T-waves become more symmetric. Early repolarization (normal variant ST elevation, especially in anterior leads) has very assymetric T-waves (slow upstroke and fast downstroke). This is true with normal T-waves too.

    Early after occlusion, T-waves may not yet be symmetric, but they are more symmetric than normal. T-waves which are symmetric are NOT normal and should greatly raise suspicion for MI (or, if peaked, of hyperK).

  13. Hi Dr Smith,

    Even there is no widened QRS, you will still give Calcium?

    1. Yes, it is safe and peaked T-waves are associated with instability.

  14. I tryed the formula for early repolarisation in the previous ECG, and I obtained 23,586 > 23,4 (ST60J = 1,5 ; RA = 15,5mm; QTC= 455ms)
    Maybe hypoCa+ was present in both previous and admission ECG, explaining the wide QTC?
    What was the etiology of hyperK in this patient?

    1. The formula can only be used when the differential diagnosis is early repol vs. anterior STEMI. And thus there must be ST elevation present.

  15. HI Dr Smith

    This is ACEP's posiiton on Ca in Hyperkalemia: "Because intravenous calcium can cause tachycardia, hypertension, and arrhythmias, as well as hypercalcemia, we urge that calcium not be given routinely for hyperkalemia in an otherwise stable patient with a normal QRS or just isolated, peaked T waves. " https://www.acep.org/Education/Continuing-Medical-Education-(CME)/Focus-On/Focus-On----Critical-Decisions--Hyperkalemia/

    Just would like to hear your thoughts on that. Thank you!

    1. This comment has been removed by the author.

    2. I agree with that in patients with no EKG findings, but do not agree if there are peaked T-waves. There is little literature that I am aware of. If you know of something that supports one way or the other, I would be interested. Notice that the statement in the ACEP article has no reference at all. I base my opinion on bad cases like the one above, and on literature that states that patients can die of hyperkalemia without EKG changes (I do not believe this, but rather that the treating physicians did not notice the changes, as in this case above).

  16. Hello sorry to revive such an old post, but is there a consensual electrocardiographicndefinition to peaked T waves in hyperkalemia.... Many thanks

    1. Well my question is: how tall should peaked T waves be (i.e in mm or small squares) to speak of hyperkalemia and in what leads shoud we look for them? Thanks for your efforts

    2. Peaked T-waves of HyperK are not necessarily tall, and no taller than hyperacute T-waves or early repol. What distinguishes them is the narrow base (which results in a very horizontal ST segment and a very sharp peak). I don't think anyone has ever quantified any of this, and so like much of ECG interpretation, it relies on an experienced interpreter recognizing morphology. I am working with a neural networking company to train its computers to recognize hyperk.

  17. Thank you for all these pearls useful in interpretation of ecg
    Can you kibkinshsre a link where you have explained about early repolarization in detail


DEAR READER: I have loved receiving your comments, but I am no longer able to moderate them. Since the vast majority are SPAM, I need to moderate them all. Therefore, comments will rarely be published any more. So Sorry.

Recommended Resources