A New Seizure in a Healthy 20-something

A 20-something year old who is the picture of good health presented with a new onset seizure.  A witness described what sounded like a 3 minute tonic-clonic seizure.  

Her seizure workup was negative and she was scheduled for an outpatient MRI and EEG.

Because she was persistently tachycardic, an ECG was recorded.  At the time her K was 3.2 mEq/L:

Here is the interpretation by the computer, confirmed by the over-reading physician:

JUNCTIONAL TACHYCARDIA

INTRAVENTRICULAR CONDUCTION DELAY  [130+ ms QRS DURATION]

ABNORMAL ECG

P-R Interval 116 ms

QRS Interval 158 ms

QT Interval 422 ms

QTC Interval 485 ms

P Axis 259

QRS Axis 88

T Wave Axis 36

What do you think?

When I saw it, I was immediately alarmed:

First, I think there are P-waves underneath all that artifact, so it is not a junctional rhythm.  

The abnormality is in the QRS and QT intervals.  I measure QT at 500 ms, with Hodges correction (what our computer uses) = 582 ms, Bazett = 668 ms. Part of this long QT is the wide QRS, which the computer measured at 158 ms. If you deduct that extra QRS duration, you get a QT of 440 ms.  I think the QRS duration was also erroneously measured and it is really 130 ms.  Thus, the QT would be about 470 ms without this extra QRS duration, with the corrections at 552 ms (Hodges) and 628 (Bazett).  

What constitutes a long QT in the setting of prolonged QRS such as LBBB, RBBB, and Paced rhythm is complicated and beyond the scope of this post, but suffice it to say that the JT and JTc intervals are very useful, as is the T-peak to T-end interval, which measures the part of the QT which is most prolonged in patients at risk for Torsades: the last part of the T-wave.

The findings are NOT due to a K of 3.2.

A QRS of this duration in an otherwise healthy patient is also a red flag — extremely abnormal — always requiring further investigation.   

NEVER trust the computer QT measurement.  NEVER.  

Below I put in links to 7 of the many cases I have in which the computer missed a dangerously long QT.

Here we write about this (full text): 

Screening for QT Prolongation in the Emergency Department: Is There a Better “Rule of Thumb?”   https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7081881/pdf/wjem-21-226.pdf

Summary: if the heart rate is over 60, then use the rule of thumb that if the QT is more than half the RR interval, it is probably long and must be measured by hand.  If the heart rate is under 60, measure the QT and if the raw QT is over 485 ms, then it is too long.

Case Continued

I saw the patient in the ED and syncope never occurred to me.  It seemed so much like seizure by history that I did not even order an ECG.  Fortunately, after I left and before the patient was discharged, she did get an ECG recorded.

Unfortunately, the findings were not appreciated.

So when I was signing my charts the next day, and found that an ECG was recorded, I naturally looked at it (it is the ECG at the top of the post).

I texted the ECG and the story to our electrophysiologist, and he texted back "WOW!"

Outcome.

Next day she returned after I called her back.  This was her ECG with a K of 4.0:

Now there are prominent U waves in almost all leads, creating a long QU interval.  The QRS is again long at 140 ms.

Here is another:

And this one is STANDING (which is useful in congenital long QT syndrome)!!

There is more to the story which makes it even more fascinating, but that will have to wait for a formal published case report.

More cases of long QT not measured correctly by computer (these are all fascinating ECGs/cases):

Bupropion Overdose Followed by Cardiac Arrest and, Later, ST Elevation. Is it STEMI?

Chest pain in high risk patient. Are these Hyperacute T-waves? What is going on here?

Syncope and Bradycardia

Syncope in a 20-something woman

Long QT: Do not trust the computerized QT interval when the QT is long

An Alcoholic Patient with Syncope

Cardiac Arrest. What does the ECG show? Also see the bizarre Bigeminy.