Monday, October 1, 2018

A 60 year old patient with large T-wave inversions

Written by Andrus Alian and Pendell Meyers, with edits by Steve Smith

A female in her 60s with history of stage IV lung cancer presented to the ED with 3/10 chest pain and dyspnea waxing and waning for the last 24 hours. She had no personal or family history of coronary artery disease, drug use, HTN, or dyslipidemia. She did have a history of smoking. She recently had a 2 hour flight. She denied diaphoresis, nausea, or back pain. Vital signs were stable and she was afebrile.

Here is her initial ECG (during persistent 3/10 active chest pain):
Large T-wave inversions. What is their distribution? What is distinctive about them? What do they mean?

Here is her baseline ECG for comparison:
Basically normal.

Interpretation of presentation ECG:

This is likely a junctional rhythm or (less likely) a very low ectopic atrial rhythm, as the P-wave is awkwardly close to the QRS complex and has unusual morphology unlike normal P-waves. This is confirmed to be different from her baseline sinus rhythm P-waves.

There are very large, symmetric, whole T-wave inversions in II, III, aVF, and V3-V6. The computerized QT and QTc were 446 and 530 ms.

The differential in general for new T-wave inversions in both the inferior and anterior leads simultaneously (not the differential for this ECG in particular) is important and unique:

 Baseline abnormality
 - the T-waves in this case are too abnormal to be possibly present at baseline, but smaller less dramatic T-wave inversions could be baseline

Reperfusion of OMI that supplies both anterior and inferior distributions
 - usually type III "wraparound" LAD
 - should be present in the pain-free (or at least markedly reduced) state

Pulmonary embolism 
 - PE can cause simultaneous inferior and right precordial TWI in V1-V4, however this case has TWI predominantly in V3-V6

Takotsubo (or "stress") cardiomyopathy 
 - which can mimic many pathologies, especially acute MI

Primary prolonged QT interval 
 - with STD, T-wave inversion, and U-waves; however the TWI in this case are far too large to be caused by primary prolonged QT interval

This case, however, has one feature that seems to be fairly specific for one etiology:

The inverted T-waves are too big, too broad, too bizarre to be reperfusion T-waves or pulmonary embolism pattern alone. These T-waves are diagnostic of Takotsubo cardiomyopathy. After seeing hundreds of examples of reperfusion T-waves, I can't say I've seen more than a handful with reperfusion T-waves this large/tall/broad compared to their QRS complexes. The same goes for the T-wave inversions of pulmonary embolism. When the T-waves are bizarrely inverted with prolonged QTc like this, Takotsubo is almost certainly diagnosed in the eyes of an experienced electrocardiographer.

The fact that the patient is still symptomatic during T-wave inversion is a simple clue against reperfusion. But remember that Takotsubo can be a response to any stressor, including ACS, PE, and anything else.

Back to the Case Events:

Based on the triage ECG, the cath lab was activated.

One of the clinicians pointed out that there is a "new tall T-wave in V1" which is purported to be indicative of LAD occlusion. However, the upright T-wave in this case is simply because all leads which have any "upward" or superior/cephalad component (aVR, V1, aVL) are forced to register reciprocal large upright T-waves because of the proportionally massive T-wave inversions inferiorly and laterally. This is the T-wave equivalent of having STE in aVR (along with V1 and aVL) in the setting of diffuse severe supply/demand mismatch (diffuse subendocardial ischemia).

Even if there were no other explanation for the new tall T-wave in V1, this finding is only a soft sign of occlusion. Smith noted this feature in 14% of cases of early repol and 34% of LAD occlusions in the derivation study of the anterior OMI formula. The only times when NTTV1 makes sense as a sign of OMI is when there is anterior STE with upright T-waves, or when there is inferior OMI with RVMI.

After the EM and cardiology teams reviewed the ECGs, they correctly concluded that there was no STEMI criteria met. Because a point of care troponin was elevated, they offered the patient emergent cath, however the patient elected to avoid cath given her 6-9 month cancer prognosis and more palliative goals.

The first troponin T returned at 0.24 mg/dL.

At this time Dr. Alian (who completed an ultrasound fellowship and has extensive cardiac ultrasound training) performed a bedside US. Here is one relevant clip:

Dr. Alian explains his findings based on this and several other views:

"There is hyperkinesis of the basal posterior (aka inferior lateral wall) and anterior septal walls of the left ventricle (LV). The mid anterior and mid posterior walls are dyskinetic. The septal and inferior walls of the LV apex are also dyskinetic. Would a coronary lesion account for these wall motion abnormalities? 

The basal and mid ventricular segments of the anterior septum are usually supplied by the LAD. The LV posterior wall is supplied by the left circumflex or the right coronary artery (RCA). If this were a wraparound LAD lesion, the inferior wall and anterior septum could be affected. However, I have only seen case reports where a wraparound LAD supplies the posterior wall and it is an exceedingly rare occurrence. Perhaps, there is a left main coronary occlusion with left dominant circulation and the circumflex is supplying the posterior wall? If this patient’s symptoms were due to a left main occlusion with left dominant circulation, we would not expect to see LV hyperkinesis in the basal anterior septum and the basal  posterior wall with simultaneous dyskinesis of the LV apex, mid anterior septum, and mid posterior wall. Another finding that is absent is hyperkinesis of the wall segment opposite the infarct segment. When a patient has a wall motion abnormality as a result of the ACS, the wall opposite the ischemic segment becomes hyperkinetic. However, on this patient’s echo, there is no compensatory hyperkinesis present and the wall motion abnormalities are  symmetric. The wall motion abnormalities do not correspond to a coronary occlusion that would produce this patient’s wall motion abnormalities, so what’s is going on? 

The echocardiogram reveals apical and mid LV dyskinesia while there is simultaneous hyperkinesia of the LV base. 

The patient’s clinical presentation, the EKG, labs, and the bedside echocardiogram suggests that the patient has Takotsubo cardiomyopathy with ballooning of the mid and apical LV.  Despite the description of apical ballooning, studies show that 76% of patients with Takotsubo Cardiomyopathy have ballooning of the mid and apical LV.

In summary, the patient is 68 yo female with stage 4  lung cancer, with dyspnea and chest pain. The EKG is significant for new giant/deep t-waves leads V3-V6 and 2, 3, and avF, new  upright tall t-wave In V1, new QTc prolongation, and new shortening of the PR interval. The patient’s  echocardiogram and EKG are classic for Ventricular Takotsubo cardiomyopathy." 

Serial troponin values decreased slightly. No cardiac cath was performed. Formal echo confirmed the bedside findings. Pulmonary embolism was not seen on CT angiogram.

Here is her ECG the next day:
Similar findings, except likely return to sinus rhythm.

No obvious cause of Takotsubo was found during hospitalization. Symptoms resolved, and she was discharged.

Learning Points:

Takotsubo cardiomyopathy can mimic anything. It can be idiopathic, or it can be a consequence of just about any other significant illness.

Takotsubo cardiomyopathy may sometimes display characteristic bizarre large volume symmetric T-wave inversion with prolonged  QTc.

On ultrasound, Takotsubo may show the characteristic apical ballooning as seen above. There are also less common variants such as "reverse Takotsubo."

T-wave inversion simultaneously in inferior and anterior leads may be due to pulmonary embolism, reperfusion from OMI, Takotsubo cardiomyopathy, or other less important causes. However, bizarrely deep and wide T-wave inversions as in this case are due to Takotsubo CM.

Comment by KEN GRAUER, MD (10/1/2018):
Instructive case by Drs. Alian, Meyers and Smith — which serves to remind us of the multifaceted presentation of Takotsubo Cardiomyopathy (TC). As emphasized in the Learning Points — TC can mimic anything. And on ECG, it may sometimes display the bizarre, large volume symmetric T wave inversion with prolonged QTc, as seen in this case (Figure-1). I limit my comments here to 2 additional facets of the above presentation.
Figure-1 Initial ECG at the time of presentation (See text).
The Initial Rhythm — The rhythm in Figure-1 is interesting. As has been noted — P wave morphology is highly unusual, with a biphasic (negative-positive) appearance in lead II with short PR interval.
  • P waves are conducting — because each QRS complex is preceded by a P wave with fixed PR interval. However, the initial negative component of the P wave in lead II tells us this is definitely not a sinus rhythm. The QRS complex is narrow. The rhythm is regular at a rate of ~85/minute.
  • Negative conducting P waves with a short PR interval in lead II may be seen with either a junctional (AV nodal) rhythm — or with a low atrial rhythm. In either case, orientation of the path of atrial depolarization will be perceived by lead II (which is located at +60 degrees) — as moving away from lead II (thereby writing a negative P wave in this lead).
  • Rather than distance — it is the speed of conduction that a junctional impulse must travel retrograde through the atria relative to the speed of forward conduction through the ventricles that determines how short the PR interval of the negative P wave in lead II will be with a junctional rhythm. As a result, the specificity of a negative P wave in lead II with a short PR interval for confirming junctional (vs atrial) etiology is far from perfect.
  • In general, junctional rhythms are less likely than atrial rhythms to manifest a biphasic P wave in lead II. Junctional P waves are typically narrow and all negative in lead II.
  • Although I have no way of confirming my suspicion, given P wave morphology — I suspect the rhythm in Figure-1 is arising from the atria. And if the rhythm was arising from the AV node ... — then it would have to be an “accelerated” junctional rhythm, since the usual junctional escape rate in adults is between 40-60/minute.
  • Realizing that this discussion is primarily of academic interest for this clinical case of Takotsubo Cardiomyopathy (in which sinus rhythm was reestablished the next day) — I’ll introduce the algorithm in Figure-2 put forth by Das & Zipes for predicting the site of ectopic atrial rhythms (data in the Figure based on EP correlation for atrial tachycardia). Using the algorithm — the P wave in lead V1 is positive — it is bifid in lead II — and negative in all inferior leads — therefore likely to originate from the CS (Coronary Sinus) body.
Figure-2: Localization of atrial tachycardia based on P wave morphology (See text).
Giant T Waves — The diagnosis of Takotsubo Cardiomyopathy was firmly established in this case by the clinical presentation, the lead distribution of T wave inversion on ECG (Figure-1) — and, skilled use of bedside Echo by Dr. Alian. But it is well to consider other diagnostic entities that should be considered when you encounter an ECG with “Giant” T waves (ie, >5-10 mm amplitudein multiple leads (with or without a prolonged QTc).
  • In addition to etiologies mentioned in this case — CLICK HERE for further review of Giant T Waves on ECG.


  1. Why wouldn't takotsubo (drop in SV) be accompanied by a rise in HR? Also classically associated with some type of catecholamine surge where rise in HR might be expected.

    1. It should be. But of course some patients are on beta blockers (not sure here). And patients who do not compensate for the low SV and likely to go into shock.

  2. Dr.Smith, how long will this feature (giant T wave) last in Takotsubo and how will it change with time?

  3. How different is formal echo from bedside echo and who performs the formal echo? sorry about the basic question.

    1. Formal echo is done by more highly trained individuals (dedicated cardiac ultrasonographer and cardiologist) and is with IV bubble contrast. There are certainly some point of care cardiac sonographers who are just as accurate, but they rarely have bubble contrast.

    2. Bedside Echocardiography is usually done to Answer simple Questions. For example, is the heart moving in cardiac arrest? Is there a pericardial effusion? Is the ejection fraction normal?

      As an ED physician and RDCS (Registered Diagnostic Cardiac Sonographer), It is my believe that emergency echocardiography is still in its infancy as general ED ultrasound once was.

      In critical patients, advanced echo skills be life saving. I can recall a case Where my patient had a questionable ST elevation MI in the circumflex distribution. Cardiology did not believe it was an ST elevation MI. When I showed the cardiologist the bedside echo, he saw regional hypokinesis in the anterior lateral mid and basal segments. As a result, patient was taken for cardiac catheter cath And had 100% occlusion of the circumflex.

      Given the debatable ST segment elevation, the patient could have had an OMI with regional wall motion abnormalities. I believe that demonstrating regional wall motion abnormalities to the cardiologist Pushed the cardiologist to take the patient to the Cath Lab.

      Regardless, if this was a true STEMI or not, The patient had an OMI and Went to the Cath Lab as a result of in the ED physician performed bedside echo cardiogram. He got where he needed to go.

      Looking for regional wall motion abnormalities in Chest pain can be extremely useful in the right hands. However, basing Decisions on a bedside echo without proper training Is dangerous.

      As bedside echocardiography applications by ED Physicians expands, We will need to develop guidelines on proficiency in advanced applications of bedside echocardiography

    3. Agree. some findings are obvious and do not require extensive training. other findings are very subtle and those who are not trained well may not know what they are missing.


Recommended Resources