Monday, April 20, 2020

Why doesn't the name "OMI" say anything about the ECG, in contrast to STEMI? Because sometimes the ECG isn't always enough, no matter who interprets it.

Written by Pendell Meyers

A man in his 60s with history of prior CAD with PCI, HTN, HLD, and aortic insufficiency presented with acute chest pain radiating to the left arm. He had been walking on a treadmill for approximately 5 minutes when the symptoms began abruptly. He had associated shortness of breath, diaphoresis, and dizziness.

He arrived by private vehicle with relatively normal vitals signs and this initial ED ECG:

What do you think about the patients clinical presentation? About the ECG?







The patient: the treating team was completely convinced of ACS (or less likely another catastrophe such as dissection) by his symptoms and clinical appearance before the ECG was recorded.

The ECG read prospectively: Both I and Dr. Smith read this ECG (without any clinical information) as having no convincing evidence of OMI. We both noted the slight STE in V1 along with minimal STD in V6, I, and aVL, stating that these findings were possibly concerning and that we would perform serial ECGs as well as clinical info, echo, etc, to help the decisions further. We have described the pattern of STE in V1-V2 with lateral STD in cases of LAD occlusion, but we both agreed that something about this ECG just doesn't fit that pattern. But overall with this ECG in isolation (no clinical context, no baseline available to us at that time), we did not see evidence of OMI.

Smith comment: There is a touch of STE in V2 and V3, so one might wonder if this is subtle LAD occlusion. The formula value (using QTc of 400) is 16.88 however, which is very low and makes it very unlikely that this STE is due to LAD occlusion (which is not saying the same thing as there is unlikely to be LAD occlusion - sometimes it just does not show on the ECG).


There was a baseline ECG on file that they found later:
Baseline ECG which is basically normal.



Given ongoing ischemic pain and clinical appearance and pretest probability, the treating team called a "heart alert" (which summons the cardiology fellow to bedside immediately and coordinates a phone call between ED and Cardiology interventional attendings, to discuss potential activation of the cath lab).

Another management option might include recording both right sided and posterior leads. Sometimes when the standard 12-lead shows no evidence of OMI, the RV and/or posterior wall will manifest some ST elevation. It is worth a try.


Luckily, the cardiology attending was already in the building and came immediately to evaluate this patient whom he knew well.

They reviewed the case at bedside and agreed that his clinical appearance and history warranted emergent cath lab activation regardless of the lack of clear ECG findings, and before the first troponin would return.

The team performed a bedside echo which is not available, but they could not see any clear wall motion abnormality.

These two ECGs were recorded before he left the ED to the lab:






In the lab he was found to have complete occlusion of the proximal RCA (see images below). This was stented with excellent angiographic result.

Before intervention. Total proximal RCA occlusion.


After intervention, showing the course of the RCA.



Post cath ECGs:

This shows small terminal T-wave inversions which are indicative of inferior reperfusion in leads III ad aVF.


 The patient did well and was discharged home.


In retrospect, the only findings of proximal RCA occlusion that I can clearly see on the pre-cath ECGs is STD in aVL and lead I (which are minimal and obscured by baseline wander).



Learning Points:

The ECG is quite accurate overall when interpreted by those with significant experience. However, it will never be enough in isolation to detect all acute coronary occlusions. This is one of the reasons why the name of the paradigm designed to find acute coronary occlusion must be centered on the pathology we are trying to find, and must not contain specific ECG findings (such as "ST Elevation MI").

When you instead understand the better name "Occlusion MI," then you give your mind permission to keep considering the possibility of acute coronary occlusion despite the lack of traditional ECG findings, and your mind is free to consider other clinical features that may help with this important diagnosis such as clinical presentation, persistent pain, echo findings, etc.

Take it from someone who cares a lot about ECGs - it's not all about the ECG!

This case is an excellent example of the "no false negative paradox." This paradox is one of the major problems with the current STEMI paradigm, and it describes the situation that there is no such thing as a "false negative" in the STEMI paradigm. For every other paradigm, there is an actual outcome that defines whether the patient had the underlying disease or not, whether the case was a true negative or a false negative (for example, if you decide that a patient does not have appendicitis based on your history and physical exam findings, and the patient later turns out to have appendicitis, then this case would be classified as a false negative). But in the STEMI paradigm, the definition of the paradigm is a millimeter finding on a piece of paper, and has nothing to do with the patient's actual outcome(!), and there can be no false negative because NSTEMI by itself does not warrant emergent management by definition of the paradigm. Thus, even a patient who dies of acute coronary occlusion is not considered a "false negative" in the STEMI vs. NSTEMI paradigm, so long as their ECG doesn't manifest the specific criteria (which have pitiful accuracy). This case is an excellent example of a false negative in the OMI vs. NOMI paradigm, which cannot exist in the STEMI vs. NSTEMI paradigm. For more discussion of the no false negative paradox, hear the discussion in this lecture: https://emcrit.org/emcrit/emcrit-podcast-250-the-omi-manifesto-lecture-by-pendell-meyers/



===================================
MY Comment by KEN GRAUER, MD (4/20/2020):
===================================
Highly insightful case by Dr. Meyers — because (as Dr. Meyers emphasizes) — the ECG shows very little despite cath confirmation of acute proximal RCA occlusion.
  • Cases like this are humbling. I think they teach us a lot!
For illustrative purposes — I have put in Figure-1 the initial ECG that was done in the ED ( = ECG #1) — together with a baseline tracing ( = ECG #2) that was previously done on this patient.
  • Given the “high prevalence” situation in this case (ie, the patient is a man in his 60s with known coronary disease and new-onset worrisome chest pain!) — the onus falls on us to rule out an acute event, rather than the other way around. I therefore scrutinized ECG #1 for the smallest of details.

Figure-1: The initial ECG in this case — compared to a prior tracing (See text).



MY THOUGHTS on ECG #1: I agree completely with Drs. Meyers and Smith that ECG #1 does not manifest convincing evidence of OMI.
  • There is slight sinus arrhythmia and bradycardia (rate just under 60/minute). All intervals and the axis are normal. There is no chamber enlargement. There are small inferolateral q waves. Perhaps the inferior Q waves reflected a prior event in this patient with known coronary disease (?) — but there certainly were no acute inferior lead changes. ST-T waves in leads III and aVF, as well as in the lateral chest leads were a bit flattened — and the ST segment in lead aVL was straight — but these were clearly nonspecific changes.
  • Transition occurs early — in the form of a surprisingly tall R wave in lead V2 (R=S in this lead). Perhaps this reflected posterior infarction at some point in time — but again, nothing that looked acute.
  • There is some subtle straightening of the ST segment in lead V3 — and the J-point in leads V2 and V3 appears to be slightly elevated. That said — I was not convinced this was acute.
  • What did catch my eye — was the clearly abnormal ST elevation in lead V1. I did not know if this was ischemic or a technical mishap. The fact that there was actually an rSr’ complex in lead V1 of ECG #1, with a similar QRST appearance in lead aVR raised the possibility of too-high placement of the lead V1 electrode on the chest.
  • BOTTOM LINE: In a “high prevalence” situation in this patient who presented with worrisome new-onset chest pain — I thought the ST-T wave appearance in lead V1 of ECG #1 was clearly out of the ordinary. More information (serial tracings, stat Echo, troponin — and in this case, prompt cardiac cath) was in order.

ECG #2 (the Baseline Tracing) — Although there is slight difference in the frontal plane axis, and in QRS morphology of some chest leads in ECG #2 compared to ECG #1 — none of these slight differences looked acute.
  • The one lead that I thought did show a potentially significant change since the baseline ECG was done was lead V1. The ST-T wave in lead V1 of the baseline ECG ( = ECG #2) looked normal. In contrast — there was unexpected ST elevation in lead V1 on the initial ED tracing ( = ECG #1). That said, the terminal r’ was not present on the baseline tracing — which suggested the reason for ST elevation in lead V1 of ECG #1 might simply be the result of lead misplacement of the V1 electrode. (For more on recognizing lead misplacement of leads V1,V2 — Please see My Comment in the November 4, 2018 post in Dr. Smith’s ECG Blog).
  • Alas — nothing evolved on serial ECGs done prior to cath. None of these follow-up tracings showed abnormal ST elevation or a terminal r’ in lead V1.
Last Lesson — I learned from this case by going back through the serial ECGs that were done.
  • I thought it important to recognize in this patient with worrisome new-onset chest pain — that the ST elevation we see in lead V1 of ECG #1 is not a usual finding. Whether the reason for this ST elevation was lead misplacement of the V1 electrode and/or transient ST elevation from acute proximal RCA occlusion (that was not seen on subsequent pre-cath ECGs) — I think is uncertain.
  • Finally — I found it insightful in this case that serial ECG tracings showed very little despite cath confirmation of acute proximal RCA occlusion.
  • P.S. — I would have been curious to see what right-sided leads done at the time ECG #1 was recorded would have looked like.

Our THANKS to Dr. Meyers for presenting this case.


13 comments:

  1. I think the second EKG ,before the patient left the ED, started to show some changes, from baseline, in the inferior leads especially lead 3.

    ReplyDelete
    Replies
    1. THANKS for your comment. Looking at the 2 ECGs that were recorded before this patient left the ED (on the way to the cath lab) — while I agree that there might be slight change (a little more ST flattening in lead I — and ever-so-light-but probably real ST coving without elevation in lead III) — I still think these changes are non-diagnostic (and less impressive because there is really NO difference [evolution] occurring between these 2 pre-cath tracings). Anyway you look at it — I think it fair to say that serial ECGs were not diagnostic of the complete proximal RCA occlusion that was shown at cath. But the main points of this case being: i) the humbling reality that serial ECGs don’t always show diagnostic changes (even when interpreted by experts); and ii) Clinical decisions must sometimes be made in the absence of indicative ECG findings — on the basis of history and clinical symptoms + appearance of the patient in front of you. THANKS again for your comment! — :)

      Delete
  2. Thanks to Pendell and Ken for this very interesting ECG!

    I can't determine from the frames from the cath if the RCA was non-dominant. I suspect it was since there was no suggestion of inferior epicardial ischemia. Generally, an STE of this sort in V1 but not the adjacent precordial leads or inferior leads, I've seen only in 1) proximal occlusion of non-dominant RCA and 2) proximal occlusion of the LAD in which D1 was proximal to S1. (For what it's worth, I still did not anticipate the outcome until I read it.)

    I think when you look back at the pre-cath ECGs, there certainly is no suggestion of an LAD occlusion. I assume right-sided leads were not recorded. It was always my protocol that if the patient appeared to be having pain compatible with ACS and the 12-lead ECG was non-contributory, I always did posterior and right-sided leads. So we are in agreement there.

    I think that along with adoption of the OMI concept, we are going to have to hone our observation and deduction skills to a much higher degree.

    ReplyDelete
    Replies
    1. THANKS as always Jerry for your insightful comments! In my experience — posterior leads are rarely (if ever) indicative of acute changes not deducible from the standard 12-lead tracing. I am hard-pressed to remember even a single case in which posterior leads told me something I didn’t know from looking at the 12-lead. I accept that others differ with my opinion — and that some providers prefer to regularly assess posterior leads in cases like this one. That said — as you mention (and as I alluded to in the “P.S.” of my comment above) — right-sided leads CAN BE invaluable in certain cases — and right-sided leads probably would have been insightful in this case. Considering that 2 more ECGs were done after consulting with cardiology in the ED — I do not understand why the cardiology team did not order a right-sided tracing. As a result, we’ll never know if there in fact WAS clear evidence on ECG of acute RCA occlusion. THANKS again for your comment Jerry!

      Delete
  3. Ken...

    Thanks for your reply! While I totally agree that if you can see ST depression in Leads V1 - V3 in the context of an ACS, there is certainly no point in recording posterior leads. But if - as an expert electrocardiographer - one has been unable to decide on any diagnosis, then I feel posterior leads are well worth recording (in addition to right-sided leads). That was done in this case. The fact that those leads may only RARELY reveal anything is of absolutely no concern to me. Of, course, I don't record those leads when the diagnosis is already evident.

    I understand that you were addressing this issue in the context of ACUTE ischemic injury. I just want to caution those just learning about ECGs that recording posterior leads is NOT a waste of time under the appropriate circumstances. Searching for something and getting a negative result is NEVER a waste of time; however, searching INAPPROPRIATELY for something and getting a negative result is definitely a waste of time. In that sense, obtaining posterior leads when there is already significant ST depression in V1 - V3 would indeed be inappropriate and a waste of time. Obviously, even if STE were present in those leads - you ALREADY HAD your diagnosis!

    ReplyDelete
    Replies
    1. Hi Jerry. If I read the case correctly — neither posterior nor right-sided leads were done in this case (Pendell just mentioned that this could be “another management option”). As usual — I believe you and I ARE on the same line of the same page. What WOULD be a loss of time (that could be precious if in an out-of-hospital setting) is recording of posterior leads when careful assessment of the original 12-lead is already diagnostic of acute posterior MI (ie, positive “mirror test”). On the other hand, it IS perfectly appropriate to get posterior leads when you suspect posterior involvement, but can’t be certain about this from the standard leads. Many clinicians do this — and that is fine. I’m just offering my personal experience (admittedly anecdotal) that I’ve yet to see a case in which regular leads were not diagnostic, yet posterior leads were … As always — GREAT to discuss with you! — :)

      Delete
  4. Is it possible we can see a S1Q3T3 pattern in his ecg at the ER?

    ReplyDelete
    Replies
    1. @ Dexter — YES — by the “definition” — there is an S1Q3T3 in the ED ECGs, albeit that T wave in lead III is TINY! (ever-so-shallowly inverted). That said, in my experience — the isolated occurrence of an S1Q3T3 in the absence of a history suggestive/consistent with acute PE, and in the absence of other ECG findings suggestive of acute right heart strain has extremely poor sensitivity and extremely poor specificity for acute PE. I see nothing else on this ECG consistent with acute right heart strain — and the history of this 60s man with known documented coronary disease sounds cardiac — so I did not think the S1Q3T3 added to this case. I hope that makes sense — :)

      Delete
  5. Very intersting case as usual.
    This case proves that even without clear signs of ECG abnormalities indicative of myocardial ischemia it is recommended cath lab activation when the clinical scenario and symptoms are higly suggestive of acute coronary artery disease.
    As to the initial ECG and just for discussion, while admitting that this ECG did not show clear abnormalities, I must confess that I was immediately attarcted by leads V2-V3 that show minimal ST elevation but especially a relatively larger based T wave (much more evident in led V3). In retrospect, the fact that that above ECG “suggestions” were real is corroborated by the last post cath ECG (the second one) that shows biphasic (positive and negative) T waves in V3-V4.

    Mario Parrinello

    ReplyDelete
    Replies
    1. Thanks for your comment Mario! Yes, there were some subtle-but-present abnormalities in leads V2 and V3 of the initial ECG ( = ECG #1 in my Figure-1) — but I didn’t think these were definitive for acute OMI. There is some straightening in lead V3 compared to the baseline tracing (ECG #2) — but this difference was modest, and I didn’t think it evolved in the subsequent 2 ECGs that were done in the ED prior to the patient arriving in the cath lab. So clearly subtle (I thought non-definitive) ECG changes while the patient was in the ED — for which the clinical scenario played a KEY role in the decision for prompt cath. THANKS again for you comment! — :)

      Delete
  6. Catching up on posts, so I'm late to the game. Great case discussion! Like the authors, something about that first ECG didn't sit right with me, but I couldn't pin it down to anything, and 100% agree with their thinking and management.

    I wonder what the patient's left-sided circulation looked like on cath. In HINDSIGHT, the first tracing fits very well with an isolated RV OMI (aka non-dominant RCA culprit), with subtle injury in V1-V3 and a frontal ST-vector somewhere around 180 degrees. At cath though, it looks like this RCA supplies at least some of the inferior wall & probably some of the inferior septum, but it's hard to tell for sure. I wonder if the patient might have had a co-dominant LCx, which would explain the lack of inferior injury on the ECG.

    It's always interesting to see reperfusion T-waves in V3 & V4 with an RCA culprit. It just goes to show that the RV is really an "anterior" wall.

    ReplyDelete

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.