Sunday, May 3, 2020

A young woman with vomiting

A 20-something presented with vomiting.

An ECG was recorded:

HR is 101
The prior ECG was normal.
The computerized QT measurement was 353 ms, and Hodges QTc was 411 ms
What is going on?

There are down-up T-waves, in which the QT interval (measure wrong by the computer) is too long.

I measure it as 460 ms in lead V3, with a Hodges QTc of 532 ms.

There is also some diffuse non-specific ST depression.

Down-up T-waves should make you think one of two things:
1) reciprocal to up-down T-waves
  a) up-down waves in leads V7-V9 of reperfused posterior MI often manifest with down-up T-waves in V2 and V3
  b) in lead III, reciprocal to aVL
   c) in lead aVL, reciprocal to III

2) Hypokalemia, in which the up component is really a U-wave.  In these cases the apparent QT is very long, and is really a QU-wave.

DDx of ST depression in the presence of a normal QRS (i.e., not LVH, LBBB, RVH, RBBB, WPW, etc.) is:
1) non-pathologic
2) ischemia
3) digoxin
4) hypokalemia

Here are leads III and V3 magnified:

Hypokalemia is the etiology here.  K was 2.3 mEq/L.

MY Comment by KEN GRAUER, MD (5/3/2020):
Interesting case presented by Dr. Smith — especially from the viewpoint of clinical synthesis. The patient is a 20-something woman, who presented with vomiting and the ECG that I have labeled and reproduced in Figure-1.
  • I arrived at the same conclusion as Dr. Smith — although my path for getting there was slightly different.

QUESTIONS Regarding ECG #1Dr. Smith provides the conclusion to this case — which is that this patient has marked hypokalemia (ie, K+ = 2.3 mEq/L). But before arriving at this conclusion:
  • What was your Systematic Interpretation of the 12-lead ECG shown in Figure-1? The reason it is important to be systematic in your interpretation — is that there are several important ECG findings that would otherwise be all-too-easy to overlook. HOW MANY of these can YOU identify?
  • HINT: Whenever you see colored lines, arrows and numbers — these almost always contribute to the answer. WHY did I label ECG #1 as you see it in Figure-1?

Figure-1: The initial ECG in this case — which I have labeled (See text).

ANSWERS Regarding ECG #1I’ve emphasized the importance of routine use of a Systematic Approach to ECG interpretation on numerous occasions (See My Comment in the June 28, 2019 post of Dr. Smith’s ECG Blog). Failure to do so invariably leads to overlooking potentially important ECG findings. I interpreted the ECG shown in Figure-1 as follows:

Descriptive Analysis of ECG #1:
  • There is significant baseline artifact. Keep in mind that your interpretation of the ECGs of your patients is part of the medical record. I therefore feel it relevant to document (in your dictation/on the medical chartwhen significant artifact potentially impairs the accuracy of your interpretation. Given the importance of accurate measurement of the QT interval in this case — noting there is artifact in ECG #1 is relevant!
  • The rhythm in ECG #1 is sinus at ~100/minute (technically, sinus “tachycardia”).
  • Intervals: The PR interval is normal (ie, not more than 0.20 second). The QRS complex is normal (ie, not more than 0.10 second). The QT interval is prolonged (SEE below for details).
  • The frontal plane axis is normal (about +45 degrees).
  • There is no chamber enlargement.

Regarding Q-R-S-T Changes:
  • There is a small septal q wave in lead aVL. Although a small initial r wave is present in lead V1 — this initial r wave is lost in lead V2 — which manifests a large Q wave.
  • R wave progression looks bizarre. It does not make physiologic sense for there to be an initial r wave in lead V1 — that is then lost in lead V2 — to then show abrupt predominant positivity by lead V3 (SEE below for details).
  • Assessment of ST-T Waves — reveals ST-T wave abnormalities in virtually all leads! The most remarkable ST-T wave changes are in lead V3 — which manifests a deeply inverted T wave, with (as per Dr. Smith) a terminally positive T wave component. Multiple other leads show nonspecific ST-T wave flattening with slight ST depression.

Regarding the QT Interval:
In the last Dr. Smith ECG blog post (from April 29, 2020) — I emphasized in My Comment that the QT interval should be measured in THAT lead in which you most clearly see the end of the QT interval, and, in which lead the QT is longest.
  • KEY — Be SURE to look at all 12 leads when measuring the QTc. This is relevant in this case. First — baseline artifact makes it difficult to precisely measure intervals in ECG #1.
  • I suspect the computer selected lead aVL for its measurement of the QT interval. Based on this lead — I would have thought the QT interval was normal (RED double arrows indicate my measurement of 350 msec for the QT interval in this lead). This is why the computer indicated a normal corrected QT value (QTc = 411 msec).
  • Looking carefully at other leads in this 12-lead tracing — it should be apparent (as per Dr. Smith) that the QT interval is clearly prolonged. BLUE double arrows in several leads indicate my measurement of ~450 msec for the QT interval — which at a heart rate of ~100/minute, yields a corrected QTc = of at least 530 msec.
  • PEARL #1: As I emphasized in the April 29, 2020 post — IF it is difficult for YOU to determine the end of the QT interval (either because of baseline artifact and/or indistinct boundaries) — then it will also be difficult for the computer! (which is why the computer “thought” the QTc was normal in this case). In such instances — Be AWARE that you can not trust the computer evaluation of the QTc, and YOU need to assess the QTc yourself!

WHY does R Wave Progression look Bizarre?
As I mentioned above when assessing R wave progression — It does not make physiologic sense for there to be an initial r wave in lead V1 — that is then lost in lead V2 — and which then shows abrupt predominant positivity by lead V3. Nor does it make sense for the T wave to be clearly negative in lead V1 — then isoelectric in V2 — and then, not only negative again in V3 but markedly negative in V3!
  • PEARL #2: It is almost certain that Leads V1 and V2 are misplaced on the chest! Placing these first 2 precordial leads 1 or 2 interspaces too high is surprisingly common in clinical practice — even by experienced providers!
  • As I reviewed in My Comment in the November 4, 2018 post from Dr. Smith’s ECG Blog — there are 3 CLUES which allow rapid detection of probable V1,V2 lead misplacement — and all 3 of these clues are present in ECG #1: i) IF there is an r’ in leads V1 and/or V2, especially if the other finding for incomplete RBBB (ie, terminal s waves in lateral leads I and V6) is absent: ii) IF there is a significant negative component to the P wave in lead V1 and/or V2; andiii) IF the appearance of the QRS complex and the ST-T wave in leads V1 and/or V2 looks very much like the QRS and ST-T wave in lead aVR.
  • The relevance of recognizing probable V1,V2 lead misplacement is that: i) The large Q wave in lead V2 would probably disappear (and R wave progression will probably normalize) with repeat ECG after verifying lead placement: andii) We would be able to better appreciate WHY the T wave in lead V3 looks so very abnormal and, so very different from the ST-T waves in all the other leads (ie, I suspect a correctly placed lead V2 may show similar changes to V3).

Clinical IMPRESSION of ECG #1: Given that the patient in this case is a 20-something woman whose presenting complaint was vomiting (but presumably no chest pain or dyspnea) — the clinical probabilities would seem fairly limited even before looking at the ECG. I was thinking electrolyte disturbance, possibly with drug or substance ingestion ...
  • Putting Together the ECG findings we noted above — there is a lot of baseline artifact + sinus tachycardia at ~100/minute marked QTc prolongation + diffuse nonspecific ST-T wave abnormalities, with prominent T wave inversion in lead V3 (and potentially also in lead V2 if the ECG would have been repeated after verifying lead placement).
  • IF this patient had presented with new-onset shortness of breath — then it truly would have been essential to repeat ECG #1 after verifying lead placement — because if other anterior leads also showed prominent T wave inversion — then acute PE would have become a key diagnostic consideration.
  • IF instead of a 20-something woman with vomiting as her only symptom — the patient would have been older, and presenting with cardiac-sounding chest pain — then the finding of diffuse ST-T wave flattening with slight depression (and also with some ST elevation in lead aVR) would have raised the possibility of coronary disease that might be acute.
  • PEARL #3: In view of the above (in context of the patient being 20-something with vomiting) — the principal finding on this ECG is QT prolongation, here with diffuse nonspecific ST-T wave abnormalities. Over the years, I have found recall of a short LIST of Causes of QT Prolongation to be of invaluable assistance. As per My Comment in the March 19, 2019 post of Dr. Smith’s ECG Blog — Assuming there is no bundle branch block, ischemia or infarction (as these entities can prolong the QT) — THINK OF: i) DRUGS (many drugs prolong the QT interval — and combinations of drugs may result in marked prolongation)iiLYTES (ie, Think of low K+ and/or low Mg++ and/or low Ca++)and/oriii) a CNS Catastrophe (ie, stroke, bleed, coma, seizure, trauma, brain tumor). Clearly in this patient — recognition of the long QTc should immediately promptly check of electrolytes in this young woman with vomiting.

Our THANKS to Dr. Smith for presenting this case.


  1. Does high polarity fit the profile of a normal QRS? The patient seems to have this on her ECG. Do correct me if I'm wrong. As an aside, I struggle to differentiate such an entity from LVH, for example, in the absence of obvious strain pattern.

    1. @ Anonymous — I am sorry, but I do not understand what you mean by “high polarity”? If you could explain — then I might be able to answer your question. Otherwise (if I understand correctly) — you indicate having trouble diagnosing LVH when there is no “strain” pattern. First — this patient is in her 20s — and QRS amplitude is often increased in younger adults. As a result, different (larger) voltage criteria need to be used when assessing whether a young adult’s ECG satisfies “voltage for LVH”. As to the importance of identifying an LV “strain” pattern (or a “strain equivalent” pattern) in the ECG diagnosis of LVH — I discuss this topic in detail at this link — — Let me know after reading this if you still have questions — :)

  2. Great tracings and post. We can see "buoys on the bay" sign in lead II. Prominent U waves is seen in V3.I always have learn lot here.Thanks a lot my teachers!!!

    1. Thanks for the positive words. Glad this has been helpful — :)


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