Tuesday, September 13, 2022

Chest pain, peak troponin 100,000 ng/L and this ECG on discharge: what’s the diagnosis?

Written by Jesse McLaren, with edits by Smith


Usually cases are presented in chronological order the way they appear in real life. But this case will be represented in reverse chronological order. At each step we’ll ask the question, “what’s the diagnosis?” using the STEMI paradigm (was this STEMI or NSTEMI?) and OMI paradigm (was this Occlusion MI or Non-Occlusion MI?).





A 60 year old without prior cardiac history was admitted with chest pain, had a peak troponin of 100,000 ng/L (normal <16 in female and <26 in males), and was discharged with the following ECG (#4). What’s the diagnosis? 



There’s normal sinus rhythm, normal conduction, normal axis and normal voltages. There’s loss of R waves V2-3 and reperfusion T wave inversion prominently in V2-4 and more subtle in I/aVL/V5, indicating proximal LAD reperfusion after anterior infarct (i.e. not Wellens).


Was this STEMI or NSTEMI? We can’t say, because we don’t know if there was ST elevation on the ECG before angiography. So regardless of the angiographic findings, peak troponin, or evidence of reperfusion, the STEMI paradigm can’t answer, even in hindsight, the basic question “what’s the diagnosis” without the ECG that preceded the angiogram. But with the OMI paradigm, we can diagnose an LAD occlusion that led to loss of R waves (with massive peak troponin) before reperfusion.





Let’s go back in time to the angiogram: 100% proximal LAD occlusion. Was this STEMI or NSTEMI? We still can’t say, because the STEMI paradigm is not based on the angiogram, and 25% of “NSTEMI” have a totally occluded artery. On the other hand, we can obviously call this Occlusion MI based on the angiogram. But even if the artery fully opened (TIMI-3 flow) by the time of the angiogram (like 20% of “STEMI”), we could still say there was OMI based on the massive peak troponin (vs Non-Occlusion MI, with open artery and small peak troponin), and the post-angio ECG showing reperfusion T wave inversion.


Smith note: over 30% of arteries are open with TIMI-1, _2, or -3 flow, 20% with TIMI-3





Let’s go further back to the ECG (#3) prior to cath, to the ED physician assessment and repeat ECG: the patient had 7 hours of chest pain refractory to nitro, an initial troponin elevated at 300 ng/L, a run of VT, and an anterior regional wall motion abnormality on point of care ultrasound: was this STEMI or NSTEMI?

There’s only minimal concave ST elevation in V2-3, which doesn’t meet STEMI criteria. So this ECG, with an elevated troponin, would be “NSTEMI.” Yet this same situation is diagnostic of OMI with many factors not limited to the ECG: the patient has refractory chest pain, an elevated troponin, electrical instability with a run of VT and an anterior regional wall motion abnormality. These are indications for emergent reperfusion under the current paradigm, but are rarely followed.





Let’s go further back in time to the triage ED ECG (#2), when the patient had 5 hours of chest pain. Is this STEMI or NSTEMI?



Here the R waves are normal. There is only mild anterior ST elevation which does not meet STEMI criteria, but there are also hyperacute T waves, and mild reciprocal STD in lateral precordial and inferior limb leads. So this is still "STEMI negative" but diagnostic of OMI, i.e. STEMI(-)OMI.





Now let’s go back to the start, to the EMS ECG (#1), when the patient had 4.5 hours of chest pain: is this STEMI or NSTEMI?

There’s greater hyperacute T waves anteriorly, and greater ST elevation in V2, and there is ALMOST terminal QRS distortion in V3. Some might call this STEMI and some might not (there's poor inter-rater reliability even for basic STEMI criteria), but it is diagnostic of LAD occlusion.  

(Formula value = 18.9, above 18.2 cutoff for LAD occlusion; STE at J-point appears to be 2mm in both V2 and V3, which would meet STEMI criteria for both men and women)



Chronological order and identifying OMI in real time


Now let’s see how the case unfolded in chronological order.

-Patient brought as Code STEMI by paramedics based on their ECG (#1)

-Seen in ED by cardiology with repeat ECG (#2), both attributed to early repolarization and cancelled cath lab

-Seen by ED physician 2 hours later after troponin resulted. Patient had refractory pain, anterior regional wall motion abnormality on POCUS, and repeat ECG showing (#3) electrical instability: stat cardiology consult to reconsider cath lab

-Reassessed and cath lab activated: 100% proximal LAD occlusion, 5 hour door-to-balloon time with peak troponin of 100,000 ng/L (this is a massive MI)

-Discharge ECG (#4) showing loss of R waves and reperfusion T wave inversion


In hindsight, based on angiogram and peak troponin the patient had a LAD occlusion leading to a massive MI, and would have benefited from earlier reperfusion. They should therefore by classified as OMI, so that we can work backwards to find all the clinical, ECG and POCUS signs that can identify OMI and lead to early reperfusion. If we learn these, we can identify OMI prospectively: I sent the first triage ECG (#2) to Dr. Smith without any information (and without the prior or repeat ECG), and he immediately replied, “LAD occlusion”. This was followed by other signs of OMI—refractory pain, electrical instability and regional wall motion abnormalities—which led to cath lab activation even though the ECG didn’t meet STEMI criteria.


But because no ECG was interpreted to meet STEMI criteria, the discharge diagnosis was “NSTEMI”. Classifying acute coronary occlusion by whether the pre-cath ECG meets STEMI criteria or not is like classifying appendicitis by whether or not the WBC is initially elevated or not, regardless of CT findings. We would never delay CT scans for days because someone with migratory RLQ pain and fever/vomiting had a normal WBC, because we know it is a poor surrogate marker, neither sensitive nor specific, and relying on this dichotomy would lead to bad outcomes. Yet STEMI criteria are relied on to determine reperfusion, when they are even less sensitive for OMI than WBC are for appendicitis, and when acute coronary occlusion is more fatal.


The STEMI/NSTEMI dichotomy also undermines data and quality improvement: the EMS activation will be classified as inappropriate activation instead of inappropriate cancellation, and the 5 hour door-to-balloon time will not be flagged as an opportunity for improvement but will be buried among all the Non-OMI cases in the NSTEMI databases.


Take home

1.     MI should be classified by the actual pathology (Occlusion MI vs non-Occlusion MI) instead of a simple interpretation of the initial ECG (STEMI/NSTEMI), so that we can work backwards and learn signs of Occlusion

2.     STEMI(-)OMI can be identified prospectively—including hyperacute T waves, loss of R waves, and reciprocal change—and retrospectively with reperfusion T wave inversion

3.    Diagnosing OMI doesn’t depend on the ECG, and can include refractory ischemia, electrical or hemodynamic instability, and regional wall motion abnormalities


MY Comment, by KEN GRAUER, MD (9/13/2022):


Masterful presentation by Dr. McLaren that illustrates why the "STEMI paradigm" seemingly has to be relegated to 2nd position for our diagnostic decision-making. 

  • It of course remains valid to apply the term, "acute STEMI" — when a patient with acute infarction does manifest a "sufficient amount" of ST elevation on ECG to satisfy millimeter criteria. But to insist on fulfilling STEMI criteria before intervening on patients like the one in today's case, who so obviously demonstrates acute OMI ( = Occlusion-based MI) — is both an erroneous practice, and a major cause of suboptimal outcome.

  • Failure even after the fact (ie, after cardiac cath confirmed 100% proximal LAD occlusion) — to acknowledge that the cause of this patient's infarction was an acutely occluded coronary artery (ie, an "OMI") — is further demonstration of ignoring the evidence provided by the patient's history and the diagnostic tests that were performed (ie, refractory chest pain for hours — serial ECG findings as described in detail by Dr. McLaren above — localized anterior wall motion abnormality on Echo — markedly elevated troponin — and complete occlusion of the "culprit" artery on cath).

Dr. McLaren has covered in excellent fashion the details of the 4 serial ECGs that were obtained in today's case — in association with clinical events and other diagnostic testing as this case evolved.

  • I focus My Comment on some important additional findings on the 3rd ECG in today's case — which I felt to be fascinating! For clarity — I've reproduced this 3rd ECG (that was obtained prior to cardiac catheterization) in Figure-1

  • NOTE: Since there was no simultaneously-recorded long lead rhythm strip — I have numbered the 15 beats that appear on this tracing.


  • Take Another LOOK at ECG #3 in Figure-1. This tracing was interpreted as still not satisfying STEMI criteria — although there is some ST elevation in leads V2,V3 — and there is reciprocal ST depression in each of the inferior leads.
  • Fortunately, the composite of "other factors" in today's case (ie, refractory chest pain for hours, elevated troponin, regional wall motion abnormality on Echo) finally were enough to convince the consulting cardiologist to perform cardiac catheterization.

  • What ELSE do you see in ECG #3?

Figure-1: The 3rd ECG obtained in today's case

Additional Findings in ECG #3:
  • In addition to the modest ST elevation in leads V2 and V3 — there is also ST elevation in lead V1. The shape of this elevated ST segment in lead V1 is coved — which is definitely an abnormal finding in this lead (especially in view of how tiny the QRS is in lead V1).
  • By itself — the QS in lead V1 is not necessarily abnormal. But it is abnormal in association with the Q wave that we see in lead V2 (which should not be there when the QRS is predominantly positive in lead V2, as it is in ECG #3).

That said — The 2 parts of ECG #3 that I found fascinating relate to the cardiac rhythm. The lack of a simultaneously-recorded long lead rhythm strip made these findings more difficult to detect.
  • HINT: Follow the P waves in the chest leads ...
  • NOTE: The fact that there are ony 4 ECGs in today's case (ie, I see no ECG was performed in between the 3rd and 4th tracings shown above) — to me suggests that the treating clinicians did not recognize what was occurring with the last 5 beats on this tracing (Figure-2).

Figure-2: I've labeled the P waves in the chest leads of ECG #3.

Rhythm Abnormalities in ECG #3:
  • Beat #9 is a PVC. In addition to being wide and looking very different than the sinus-conducted beats before and after it (ie, beats #8 and 10) — there is transient AV dissociation. That is — the 2nd P wave (2nd RED arrow) in leads V1,V2,V3 occurs right on time — but the PR interval preceding beat #9 is too short to conduct all the way to the ventricles. This can only happen if the QRS complex that occurs after this 2nd P wave arises "from below" (ie, from the ventricles).
  • Actually — I strongly suspect that beat #9 is a Fusion Beat — because all P waves (RED arrows) are on time — and the 2nd P wave does occur before the QRS of beat #9, and does have at least some time to penetrate a portion through the ventricles. This is relevant — because a fusion beat manifests an intermediate QRS and ST-T wave morphology (between a sinus-conducted beat and a PVC) — which means that a "pure" PVC in leads V1 and V2 would manifest more ST elevation than what we see in beat #9 (For more on Fusion Beats — CLICK HERE)

  • As we have shown in other posts on Dr. Smith's ECG Blog — on occasion, acute ST-T wave changes of acute MI may only be seen in PVCs, and not in the rest of the tracing (See My Comment in the October 8, 2018 post in Dr. Smith's ECG Blog). If there was any more ST elevation than what we see for fusion beat #9 in leads V1,V2,V3 (especially for a PVC in lead V1!) — then we most probably could diagnose a "STEMI" based on this PVC ST elevation.

  • Did YOU Recognize that beats #11-thru-15 in Figure-2 represent a 5-beat run of VT? This run is possibly much longer — since the 12-lead tracing stops after beat #15, which means we have NO idea as to what happened after beat #15. And since no ECG was apparently done after ECG #3 — I suspect the health care team did not recognize this run of VT.

  • PEARL: The reason we know that beats #11-thru-15 represent a run of VT — is that there once again is transient AV dissociation — because there is an on-time P wave that occurs with a very short PR interval right before the QRS complex of beat #11 (double RED arrow in lead V6 of Figure-2) — which proves that beat #11 (and the 4 similar-looking QRS complexes that follow) have to be coming "from below" = from the ventricles.

  • NOTE: The rate of this 5-beat run of VT is not fast. It is ~115/minute — which is at the limit between AIVR (Accelerated IdioVentricular Rhythm) which is often benign and associated with reperfusion — and a "faster" VT rhythm. But since I see no ECG evidence of coronary reperfusion on ECG #3 — one has to be concerned about potential deterioration of this rhythm to VFib (ie, yet another reason for prompt cath).
  • For more on AIVR — See My Comment at the bottom of the page in the April 8, 2022 post of Dr. Smith's ECG Blog.

No comments:

Post a Comment

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