Chest pressure and T-wave inversions. Not what you might think.

A 40-something male had been in the ED for many hours for altered mental status due to alcohol, when he became more alert and complained of chest pain.

They recorded an ECG:

What do you think?

The clinician called me over due to worry about Wellens' pattern, and showed this to me.

Here was my response: "I do not think this is Wellens.  These are not ischemic T-wave inversions. This is Pseudo-wellens, probably due to LVH.  It is maximal in V3-V6 and there is a lot of QRS voltage.  It is possible that it is Wellens, and so you should order 2 serial high sensitivity troponins, but I think he will rule out."

I would rule him out with our new high sensitivity troponins, or rule him in. But the ECG by itself does not worry me. [This is because it is not a truly ischemic ECG.  One should not apply troponin protocols to patients with truly ischemic ECGs, and our algorithm excludes them.  But our algorithm also says that T-wave inversions have many causes and only patients with truly ischemic ones should be excluded.

She said: "I know these are like faces to you and you can recognize them, but to me it looks like Wellens."

I said, "I know, I know.  Does the patient have active chest pain, or was the pain resolved at the time of the ECG?"

"It is active."

"Then it is very unlikely to be Wellens.  Wellens is a post-repefusion state, in which the T-waves can be called "reperfusion T-waves," and the patient is pain free."

Are T-waves ever inverted in ACTIVE ischemic (with active chest pain)?  Yes, occasionally, but only rarely. In such cases, the T-wave is inverted during ischemia and then becomes upright when the ischemia is gone.  These are different than Wellens (reperfusion) waves and, again, are quite rare.

She said: "But it is changed from the old one."

Let me see the old one:

"All I can say is that there are many causes of T-wave inversions and most are not ischemic.  Most T-wave inversions are nonspecific, even if changed from old."

Troponins in Wellens syndrome: There was a time when Wellens syndrome could have undetectable troponins, but I believe that day is passed.  It still may have all troponins below the 99th percentile upper reference value.  Although unstable angina still exists in the age of high sensitivity troponin, a patient with prolonged chest pain and T-wave inversions who has undetectable serial hs troponins does not have ACS.

So they measured serial troponins:

It is the Abbott Architect high sensitivity troponin I.  Limit of detection is 3.5 ng/L.  The 99% upper reference limit for men is 34 ng/L and 16 ng/L for women.

Any value less than 5 ng/L in a patient with chest pain for at least 3 hours from onset to blood draw has a negative predictive value of 99.7%.  (HIGH-STEACS data).  The limit of detection is 3.5 ng/L by FDA, though in Europe they are able to report less than 2.0 ng/L.

The first value was less than 4 ng/L (below LoD).  This is enough to rule out MI in someone with a non-ischemic ECG (the ECG may be abnormal, but not ischemic -- there may be LVH, non-ischemic T-wave inversions, LBBB, etc.)  But to repeat: the protocol should not be used for truly ischemic ECGs - this is not a truly ischemic ECG.

The 2-hour troponin returned at less than 4 ng/L again (any delta less than 3 ng/L rules the patient out again -- here the delta is zero)

The 4-hour troponin was less than 4 ng/L, again no delta.  The probability that these T-waves are ischemic is extremely low.

However, the clinician was just too uncertain to send the patient home, even with my reassurance.

He did get a formal bubble contrast echo the next morning, which was entirely normal.


Here is a case of classic Wellens, recorded while the patient is pain free. Later, the prehospital ECG was found and showed what the ECG looked like at the time the patient had anginal pain:

===================================

MY Comment by KEN GRAUER, MD (8/19/2020):

===================================

The question of what does and what does not constitute “Wellens’ Syndrome” is one that poses problems for many emergency providers. This case is illustrative in highlighting potential problems that arise when trying to make this diagnosis.

• For clarity — I’ve put the first 2 ECGs shown in today’s case together in Figure-1.

Figure-1: The first 2 ECGs shown above in today’s case (See text).

NOTE: The beauty of recognizing a true Wellens’ Syndrome — is that it is highly predictive of a tight proximal LAD narrowing. As per Dr. Smith — reasons why ECG #1 is unlikely to represent Wellens’ Syndrome in this case are:

• The History is “wrong”. Most often the history with Wellens’ Syndrome indicates that the patient did have chest pain — but that at the time the ECG changes of concern are noted, that chest pain has resolved. As per Dr. Smith — the reason the patient is most often pain-free at the time the ECG of concern is obtained — is that the ECG picture of ST-T wave changes characteristic of Wellens’ Syndrome is the result of reperfusion of the LAD that remains with a persistent, critical narrowing.
• The location of ST-T wave changes is “wrong”. As noted above — the clinical significance of recognizing a true Wellens’ Syndrome is that it localizes an area of severe coronary narrowing to the LAD, most often in a proximal part of this vessel. Rather than seeing maximal T wave inversion in the more anterior leads (ie, leads V2, V3) — T wave inversion in ECG #1 is maximal in leads V3-thru-V6 — as well as being present in leads I, II, III and aVF in the limb leads. True Wellens’ Syndrome is less likely to manifest such generalized T wave inversion in as many as 9 of 12 leads, yet showing no more than minimal (at most) change in lead V2.
• LVH may mimic the ST-T wave changes of Wellens’ Syndrome. I reviewed a series of Pearls & Pitfalls regarding the ECG diagnosis of LVH in My Comment at the bottom of the June 20, 2020 post in Dr. Smith’s Blog. I’ve excerpted Figure-2 (below) from that June 20 post — in which I list the criteria I favor for the ECG diagnosis of LVH. In the smaller print in Figure-2 — I find this list most helpful for recalling Cornell and Peguero criteria. While Cornell criteria are not satisfied in ECG #1 — Peguero Criteria are satisfied (ie, 22 mm [deepest S wave in lead V3] + 15 mm [S wave depth in V4] = 37mm, which is well above the required ≥28mm LVH threshold for men). This is not to say that Wellens’ Syndrome cannot occur when there is LVH — but rather to emphasize a much higher prevalence of false positive results in this circumstance.

Putting It All Together: I interpreted ECG #1 as showing sinus rhythm — normal intervals (including the QTc) and normal axis — LVH — a q wave of uncertain significance in the tiny, fragmented QRS complex in lead aVL — normal transition (with the R wave becoming taller than the S wave is deep between leads V3-to-V4) — and diffuse, symmetric T wave inversion in 9 of 12 leads.

• I thought the shape of the ST-T wave in lead V3 did resemble a shape that can be seen in Wellens’ Syndrome (ie, straightening of the ST segment takeoff, with no more than slight ST elevation — followed by steep downsloping into terminal T wave negativity).
• Details of the history are lacking (All we are told is that this 40-something man had been in the ED for many hours with altered mental status due to alcohol — and apparently only complained of “chest pain” when he was becoming more alert). While nothing is ruled out by this history — it does not sound as if new-onset chest pain was the complaint that brought this patient to the ED.

Regarding ECG #2:

• We are told that ECG #2 is an “old tracing” on this patient — but we have no idea of when, or under what circumstances this “old” ECG was obtained. As we’ve seen on many previous cases in Dr. Smith’s ECG Blog — this is extremely important information to know.
• As to ECG #2 — I interpreted it as showing sinus rhythm — normal intervals and axis — voltage for LVH (again satisfying Peguero criteria) — a tiny (insignificant) q wave in lead aVL — an rSr’ in V1 — normal transition (between V3-to-V4) — and essentially normal ST-T wave changes.

My Impression of ECGs #1 & 2:

• Despite the history of alcohol overuse to the point of producing altered mental status for hours — I found it difficult to ignore the diffuse T wave inversion I saw in ECG #1 (that was clearly different from the normal ST-T waves seen in this patient’s “old” ECG) — especially since this patient complained of chest pain at the time ECG #1 was obtained.
• As noted above — I did not think the patient had Wellens’ Syndrome. That said, given the changes between ECG #1 and ECG #2 — I would not feel comfortable ruling out ACS (Acute Coronary Syndrome) on the basis of only looking at ECGs #1 and #2.
• I don’t see mention of repeat ECGs in this case while the patient was in the ED.
• Three normal high-sensitivity troponin values (without any delta at all between them) came back over the next 4 hours. These negative results essentially do rule out ACS.
• A formal bubble contrast Echo was done the next day and reported as normal. But I don’t see mention of any form of stress test.

COMMENT: I think it’s important to distinguish between acute vs chronic coronary disease. Although 3 normal high-sensitivity troponin assays with a less-than-convincing history did rule out ACS in this case — this negative result does not rule out the possibility that this patient could have underlying coronary disease without ACS.

• Among the causes of new T wave inversion that may be seen in association with chest pain is underlying coronary disease. That said (as per Dr. Smith) — there are many potential causes of diffuse T wave inversion, so this finding of T wave inversion is not specific.
• We know nothing about this patient beyond the brief history of present illness that we were given. That is, we don’t know about coronary risk factors or about the presence or nature of chest discomfort outside of the several hours he was in the ED. We have no idea if there is any history of angina ... And we have no idea of when this patient's "old ECG" changed from normal ST-T waves to diffuse T wave inversion (ie, Was this months or longer ago? Days ago? or on the day, or just hours before he was admitted?).
• Therefore, while I’ve satisfied my ED mandate of ruling out ACS — I’d want to determine IF there is a reason for this patient’s new diffuse T wave inversion (when you compare ECG #1 to his prior tracing). Clearly more information is needed — including clinical follow-up and consideration for doing some form of stress test.

Figure-2: The criteria I favor for ECG diagnosis of LVH (See text).