T-wave inversions and dynamic ST elevation

Written by Pendell Meyers, with edits by Steve Smith

I received a text message with no clinical information other than the following ECG, with the question “Is this Wellens? No prior ECG available.”

What do you think?

I responded that this ECG represented benign T-wave inversion (BTWI), not Wellens. I asked for more history.

It turns out this was a 25 year old male with no past medical history presenting after he was found “passed out” or laying down on the floor of the nursing home facility where he works. He was reported to be intermittently answering questions and seeming “not himself.” Family reported that he had 2 similar episodes in the past 3 months, which the patient describes as similar to “bad trips,” including increased fatigue, confusion, and vague hallucinations. He had been evaluated last month at an outside hospital by a psychiatrist. He denied any substance use on the day of presentation. There was no history of exertional syncope or sudden death in the family.

Apparently a single troponin was ordered on the basis of the perceived ECG findings, which was negative. Three serial ECGs were performed which were all identical (though BTWI is not necessarily always perfectly constant).

He was discharged.

Two days later he presented again with a similar story. Here is his ECG on arrival from the second visit:

This ECG shows sinus rhythm around 65 bpm. The QRS complex has moderately high voltage but otherwise normal morphology. There is STE in V1-V3 of 1.0, 1.5, and 1.5 millimeters, which is completely normal. There are prominent J-waves in leads V4-V6, as well as leads II, III, and aVF. Lead V3 shows the first and third complexes with terminal T-wave inversions, but the second complex does not appear to have this terminal T-wave inversion – whether this is due to a brief episode of baseline wandering / lead manipulation, or beat to beat variation is unknown, but I believe it is more likely brief baseline movement. The terminal T-wave inversion is no longer present in V4 as it was in the first presentation. Overall the second ECG is better characterized as “benign early repolarization” or simply “benign ST elevation” than by BTWI.

Another single troponin was normal. He was discharged with outpatient psychiatry and primary care.

How do I recognize this as BTWI? 

Most importantly because it matches the examples of BTWI I have seen on this blog, and does not match Wellens syndrome. The T-wave inversion in lead V3 of the initial ECG just does not look like reperfusion. Lead V4 of the initial ECG has the complete morphologic package of BTWI, including high voltage (large R wave), smaller S-wave, prominent J-wave followed by minimal STE and then characteristic terminal T-wave inversion. The history obviously helps as well, as this was a young African-American patient with symptoms of near-zero pretest probability of ACS. Some cases will not be so easy clinically.

Also important, “Wellens’ syndrome” requires clinical factors in addition to ECG findings, including chest pain which resolved prior to recording of the ECG. Wellens’ syndrome is not diagnosed during ongoing pain, as this would not be consistent with reperfusion (which should produce resolution of pain). On the ECG Wellens’ syndrome also requires that there are preserved R-waves in the precordial leads.

Below I have reproduced a list of findings of BTWI from a series of other blog posts on this topic on this site, and we will go through each one with respect to the first presentation ECG:

1. There is a relatively short QT interval (QTc less than 425)
YES. Computerized QTc in this case was 424 msec.

2. The leads with T-wave inversion often have very distinct J-waves
YES. J-waves are present in V4-6, as well as II, III, aVF.

3. The T-wave inversion is usually in leads V3-V6 (in contrast to Wellens’ syndrome, in which they are V2-V4)
Here the TWI is in leads V3-4 only. So this doesn’t really distinguish. Also, Wellens’ syndrome is simply one particular example of reperfusion (in this case, of the anterior wall). Reperfusion can obviously happen in any coronary distribution, and the same pattern of findings will happen in any affected wall.

4. The T-wave inversion does not evolve and is generally stable over time (in contrast to Wellens’, which always evolves)
There is certainly no evolution consistent with continued reperfusion from OMI (occlusion MI), as there would be in the case of Wellens’ syndrome. The changes from presentation 1 to 2 are clear, but these are potentially explainable by lead placement or simply the normal variation of ECG findings in normal healthy patients from day to day or hour to hour (we don’t really know the answer to this, except to say that we have a large amount of personal experience showing many cases of benign ECG patterns which show changes with repeat ECGs). We have shown many times on this blog that essentially no normal ECG is 100% guaranteed to be completely stable over time. See this case. And this case. And this case. And this very recent case with supporting references:

Huge Precordial ST Elevation in an ED Patient

The opposite, however, is fairly reliable: if the patient has OMI or reperfusion from OMI, the ECG abnormalities will evolve along the expected progression (depending upon continued occlusion and infarction or reperfusion). Therefore, lack of serial ECG changes in the appropriate time frame can be interpreted as evidence against ischemia, but presence of changes is not necessarily indicative of ischemia (it might be, but it might not because there is a significant amount of variation based on lead placement, day to day, etc). Those who have studied cases of OMI versus benign ECG patterns from this blog can usually tell the difference.

5. The leads with T-wave inversion (left precordial) usually have some ST elevation
In the first presentation ECG there is minimal STE. In the second presentation there is slightly more.

6. Right precordial leads often have ST elevation typical of classic early repolarization
In the first presentation, no. In the second presentation, yes.

7. The T-wave inversion in leads V4-V6 is preceded by minimal S-waves
YES, seen in V4 of the first presentation ECG.

8. The T-wave inversion in leads V4-V6 is preceded by high R-wave amplitude
YES, again seen in V4 of the first presentation ECG.

9. II, III, and aVF also frequently have T-wave inversion
YES, the first presentation ECG does have flat/minimally inverted T-waves in the inferior leads.

Learning Points:

Benign T-wave Inversion is recognizable and (with experience) reliably differentiated from anterior reperfusion (Wellens).

Almost all the patterns of normal variant ECGs may show some changes upon serial ECGs. This may be due to lead placement and/or actual primary morphologic change of unknown etiology and significance. These changes can be distinguished from ischemic progression changes.

If you record serial ECGs and there is no evolution over an appropriate time frame, then the lack of evolution is a fairly strong piece of evidence against ischemic ECG changes. The presence of dynamic changes on repeat ECGs, however, may be either progression of ischemia or meaningless, normal variation in a patient’s baseline – with the experience provided by this blog these are usually not difficult to differentiate.


1. Kambara et al. Early repol is not stable over time.  https://www.sciencedirect.com/science/article/pii/0002914976901429

2. Mahaveer C. Mehta, MD.  Abnash C. Jain, MD.  Early Repolarization on Scalar Electrocardiogram. The American Journal of the Medical Sciences. June 1995; 309(6):305–311.


Sixty thousand electrocardiograms were analyzed for 5 years. Six hundred (1%) revealed early repolarization (ER). Features of ER were compared with race-, age-, and sex-matched controls (93.5% were Caucasians, 77% were males, 78.3% were younger than 50 years, and only 3.5% were older than 70). Those with ER had elevated, concave, ST segments in all electrocardiograms (1—5 mv), which were located most commonly in precordial leads (73%), with reciprocal ST depression (50%) in aVR, and notch and slur on R wave (56%). Other results included sinus bradycardia in 22%, shorter and depressed PR interval in 38%, slightly asymmetrical T waves in 96.7%, and U waves in 50%. Sixty patients exercised normalized ST segment and shortened QT interval (83%). In another 60 patients, serial studies for 10 years showed disappearance of ER in 18%, and was seen intermittently in the rest of the patients. The authors conclude that in these patients with ER: 1) male preponderance was found; 2) incidence in Caucasians was as common as in blacks; 3) patients often were younger than 50 years; 4) sinus bradycardia was the most common arrhythmia; 5) the PR interval was short and depressed; 6) the T wave was slightly asymmetrical; 7) exercise normalized ST segment; 8) incidence and degree of ST elevation reduced as age advanced; 9) possible mechanisms of ER are vagotonia, sympathetic stimulation, early repolarization of sub-epicardium, and difference in monophasic action potential observed on the endocardium and epicardium.

3. Here is a good review of early repolarization by Mehta:

Powered by WPeMatico