Thursday, March 8, 2018

Alcohol withdrawal and sinus tachycardia

This patient with alcohol withdrawal and tachycardia had an ECG recorded:
What do you think?

There is sinus tachycardia.

Look at the QT interval: it is far more than half the QT.  The computer reads the QT at 386 ms, and QTc of 462 (proprietary correction formula unknown!)

Read this article: 

QT Correction Formulas Compared to The Rule of Thumb ("Half the QT")

I measure the QT at 440 ms.  RR interval is 470 ms.  It is possible that what we are seeing is a fusion of the T- and U-wave, but the consequences are similar.

Here is the QTc:
Very long QT !!
(Fridericia is best; at heart rates over 60, Bazett results in a QT that is too long)

The magnesium returned at 1.2 mEq/L  (low).

There was also a profound metabolic alkalosis, with K = 3.0 mEq/L.

The patient is at high risk of Torsades.

Learning Point:

Do not trust the computer measurement of the QT interval when it looks long!


  1. Great case in which the ECG tells so much more than just about cardiac problems. The patient is in alcohol withdrawal — therefore subject to a series of nutritional and metabolic disorders. The presenting ECG shows a regular SVT at ~130/minute. This is a nice example of the uncommon instance in which lead II is not the best lead for detecting P waves — since it is not initially clear in lead II what is P vs T vs U vs some combination. Fortunately, lead V1 (generally the 2nd-best lead for detecting sinus activity) shows typical biphasic P waves with normal PR interval that suggest sinus rhythm. Looking down at simultaneously-recorded lead V2 confirms sinus activity — which tells us that the overly wide upright deflection toward the end of the R-R interval in lead II is at least partially made up of a sinus P wave. So the rhythm is sinus tachycardia with a VERY long QT interval. This should immediately suggest the common causes of a long QTc = i) Drugs; ii) Serum Electrolyte Disorders (including low K+, low Mg++ and low Ca++); and CNS catastrophes (bleeds, trauma, tumor, coma, etc.). We don’t know from the history about the patient’s mental status nor about medications — but given alcohol withdrawal, severe electrolyte disturbance is an excellent bet. Note: The “Rule of Thumb” that QTc should not measure more than half the R-R interval is meant as an estimate for rapid assessment as to whether or not QT prolongation is likely. It is not a precise measurement But it should ONLY be used when the heart rate is not overly fast — as this “Rule of Thumb” does not work when heart rate exceeds 90-100/minute. In cases like this when there is marked tachycardia — I’ve found the “Gestalt Method” works best (ie, what I instantly recognize having clinically correlated thousands of tracings over years) — and that to me instantly suggested that despite the rapid heart rate — there is marked QTc prolongation — which in a patient with alcohol withdrawal means probable low K+/low Mg++/possibly low Ca++ — plus making sure there is no CNS catastrophe being masked by symptoms of alcohol withdrawal. THANKS to Dr. Smith for presenting this insightful case!

    1. The delayed transition, is it the consequence of LAFB?

    2. The LAH (left anterior hemifascicle) lies to the left, superiorly and slightly “anterior” to the LPH (left posterior hemifascicle), which is more inferior, rightward AND slightly “posterior” to the LAH (hence derivation of the names left “anterior” or “posterior” hemifascicles …). Therefore, when there is LAHB — initial LV forces may be directed posteriorly (because there is now lack of opposition to the relatively posterior forces of the unblocked LPH) — and the result may be delayed transition (ie, “poor R wave progression) in the chest leads (as we see here). I always look for this whenever we have LAHB — and in my experience, sometimes you see delayed transition with LAHB and sometimes you don’t … but in cases like this in which we DO see delayed transition it certainly may be as a result of the LAHB. P.S. QRS morphology in leads V1-thru-V4 looks strikingly similar — so another possibility is that electrode lead placement might not be optimal …


Recommended Resources