ECG Guru - Instructor Resources

A gathering place for instructors of ECG and cardiac topics.


Subscribe to me on YouTube

Hyperkalemia in a DKA Patient

Sun, 05/04/2014 - 14:14 -- Dawn

For your collection, we present another interesting set of ECGs from Paramedic Erik Testerman.  They are from a 48 year old man who presented responsive only to painful stimuli, with deep, rapid (Kussmaul's) respirations.  His blood glucose in the field read as "HIGH" - too high for the glucometer to register a number.  He was treated with 3 large-bore IVs, 2 liters of NSS IV, O2.  At the hospital, his blood glucose again registered as "HIGH" on the glucometer, arterial O2 was 90%, CO2 15 (low), pH 6.8 (acidotic), HCO3 -2 (depleted).  His serum potassium was 7.0 ( greater than 5.5 is high ).  We do not have the rest of his chemistry panel.

The first ECG, at 5:59 am, shows some signs of early hyperkalemia.  One of these signs is wide QRS, at .188 sec (normal is less than .12).  This ECG even meets the criteria for LBBB, as noted in the machine's interpretation, but the widening is more likely due to the high potassium.  There is a right axis deviation.  Left axis deviation is more likely in LBBB. LBBB pattern with right axis deviation can be a sign of biventricular enlargement, but, again, this may be an intraventricular conduction delay that is NOT LBBB.  Another sign of hyperkalemia is that P waves are not evident.  They can either be flattened until they disappear, or the PR interval can become so long the P wave is lost in the preceding T wave.  The T waves are unusually tall and peaked in the chest leads - disproportionate to the wide QRS complexes.  There are ST depressions in the inferior leads.

For a good, systematic approach to the ECG changes associated with hyperkalemia, we recommend Life In The Fast Lane, by Ed Burns.

ECG number 2 was taken 13 minutes later, still in the pre-hospital phase.  The QRS is now .13 seconds, and the tall, narrow, peaked T waves are very evident in Leads V1 through V3.  There is  T wave inversion and ST depression in the inferior leads.  These are all possible signs of hyperkalemia, but also of other conditions.  Unfortunately, hyperkalemia is a "mimic" of many conditions on the ECG.  For a very interesting discussion of this topic, please go to Dr. Amal Mattu's ECG Discussion of the Week, October 14, 2013.   

Diabetic ketoacidosis and hyperglycemia are frequently associated with hyperkalemia, and it is a potentially deadly consequence of DKA.  Unless there are other complicating conditions, the total body potassium is usually not high.  The K has been forced out of the cells into the serum.  Treatment is aimed at fascilitating K entry into the cells.  Once this is accomplished, potassium may need to be replaced.  Metabolic acidosis is corrected with Na Bicarb and optimal respiration.  Potassium is "pumped" back into the cells by IV calcium followed by insulin and glucose.

The ECG changes of uncorrected hyperkalemia can progress very rapidly from tall, peaked T waves and slightly widened QRS complexes to very wide QRS complexes with no P waves, to sine waves and loss of pulses.  It is very important to learn to recognize the ECG signs because they are available immediately, and lab results take time.

Also worth mentioning, especially in cases of hyperkalemia with no ECG changes, pseudo hyperkalelmia has several causes - the most common being traumatic venipuncture.  Too much fist-squeezing, tight tourniquets, and rapid, forceful drawing back of the blood from the vein can damage red blood cells, causing release of their intracellular K into the serum in the tube.

Rate this content: 
Average: 5 (1 vote)

Comments's picture
     Interesting case by Dawn with appreciation to Eric Testerman for these 2 serial tracings taken only 13 minutes apart. One’s initial impression from ECG #1 (@05:59) would be a regular wide tachycardia at ~100/minute. Noticeably absent are P waves. V1 provides a hint of atrial activity - but this is not at all consistent - such that we need to assume no P waves (unless P waves are hidden with very long PR interval within the preceding T waves in lead II). Given the patient’s minimally responsive state - one’s initial differential diagnosis would include: i) preexisting LBBB with very wide complexes (QRS morphology in leads I,V1,V6 being consistent with this form of BBB); ii) QRS widening from some drug toxicity; or iii) Simply VT (Ventricular Tachycardia) - or in this case given the no-more-than-moderately increased rate (~100/minute) - AIVR (Accelerated IdioVenricular Rhythm)
  • The CLUE that none of these theories are operative is in the brief History we are given =  deep, rapid Kussmaul respirations + high Glucose reading in the field. The PEARL being DKA is a medical emergency in which the ECG is sometimes the strongest initial clue to the diagnosis, suggesting marked acidosis from transient hyperkalemia long before serum potassium values return.
  • Given the above clinical setting - ECG #1 shows 3 findings that are highly suggestive of marked hyperkalemia: i) QRS widening (over 0.16 second in this case); ii) Lack of atrial activity; and iii) Markedly peaked T waves that are easy to miss because of the QRS widening. But you can actually follow the peaked T wave in lead V2 up into the T wave in lead V1 above it. Note how pointed this T wave is. This is NOT simple LBBB - but rather strong clues suggestive of marked hyperkalemia. 
  • Dawn mentions details of this patient’s laboratory results are missing. The ECG picture in this 1st ECG to me suggests a serum K+ value significantly greater than 7.0 mEq/L - but rather above 8.0 mEq/L given tremendously wide QRS - no P waves - and persistent and incredibly peaked T waves despite this QRS widening.
  • Primary treatment of DKA is a little bit of insulin and a LOT of IV fluid. We are not told if any of a series of emergency measures were initiated specifically for hyperkalemia in this case (ie, Bicarb; Calcium; Albuterol nebulizer). Regardless - ECG #2 (@06:12) shows MARKED improved in terms of predicted serum K+ level. Sinus P waves look to have returned (albeit still of very small amplitude) - the QRS has narrowed considerably - and amazingly, the still enormously peaked T waves in ECG #2 are far smaller than they were in ECG #1. I would guess that the serum K+ value of 7.0 mEq/L that was reported in this case correlates MUCH better with ECG #2 than with ECG #1.
  • A handy bedside trick I was taught back during my residency days served me well over my clinical years. As an approximation - for every 0.1 pH unit - there is a change in serum K+ of ~ 0.6 mEq/L in the opposite direction. This approximate equation is based on presumptive ion flow during in and out of cells during acidosis. Since the pH in this case = 6.8 - that is 0.6 units below the norm of 7.4. Using my equation - this suggests 0.6 X 0.6 mEq/L = 3.6 mEq/L change (increase) in serum K+ as a result of this patient’s marked acidosis. This also suggests that when this patient’s acidosis is corrected (back to a norm of 7.4) - that his serum K+ value will drop from 7.0 minus 3.6 or to ~ 3.4 mEq/L. BOTTOM LINE: This patient will become HYPOkalemic once he is fully rehydrated and no longer acidotic - so the clinician needs to be aware that once the kidneys are working and as metabolic correction occurs - that this markedly hyperkalemic patient will actually need supplemental K+ as his course evolves.
  • Perhaps the most impressive aspect of this case is that the time between ECG #1 and ECG #2 was ONLY 13 minutes! Things can truly change quickly in the world of DKA with associated hyperkalemia.
THANK YOU Eric Testerman for contributing this case and these tracings.
  • For those wanting a Review of ECG signs of various Electrolyte Disorders - Click HERE to download this PDF excerpted from my ECG-2014-ePub.


Ken Grauer, MD   [email protected] 

Submitted by TestyLP on

I again can not thank you all enough for the excellent knowledge and wisdom that you all give with interpretation of these ECG's!!!! You all are teaching me so much, and making me love cardiology more and more each time!!

Erik Testerman, AAS, LP, NREMT-P

Submitted by TestyLP on

Just wanted to give this excellent group of clinicians, whom I plan to learn a lot from, a follow up on this patient. Last night while working, I was able to speak to the doctor who took care of this patient in the ER. I showed this to her and she told me that she thought he was not going to make it, but he actually walked out of the hospital 4 days later.

Erik Testerman, AAS, LP, NREMT-P

All our content is FREE & COPYRIGHT FREE for non-commercial use

Please be courteous and leave any watermark or author attribution on content you reproduce.