Anterior MI in paced rhythm, dismissed by cardiologist, patient died.

This was shown to me by a very astute Hennepin paramedic.

Although this comes from a Hennepin paramedic, the patient was not brought to Hennepin County Medical Center.

Another note:

One person wrote on Twitter that he was dismayed by the “tribalism.” of this post.  This is my reponse.

“This is not tribalism. It is essential to point out that cardiologists can be wrong also, and to always advocate for patients. It is important for cardiologists to realize that a paramedic may see something they do not. I am hoping to moderate the arrogance that so many who write to me and talk to me frequently encounter.”


An elderly patient had acute chest pain and 911 was called.  The medics recorded the following initial ECG at time 0:

The computer read (see below) gives no further comment beyond ventricular pacemaker.
What do you think?
It is difficult to appreciate the pacing spikes, but the computer detected them.
So just know that this is a ventricular paced rhythm.

Since you cannot see the pacing spikes, how would you know it is paced without the computer read?

Answer: There is a regular wide complex that looks like LBBB but there are no P-waves (i.e., underlying atrial fibrillation or atrial inactivity).  The only possibilities are accelerated idioventricular rhythm and paced rhythm.
The medic was very worried about the concordant STE and T-wave in aVR, and the downsloping concordant ST depression in II, III, aVF, as well as V5 and V6.
When he showed it to me, I said “Ouch, looks like an acute coronary occlusion.  What happened?”
The medic recorded another 12-lead en route 13 minutes later:
What do you think?
Now there is concordant STE in aVR, aVL and V4, with reciprocal concordant ST depression in II, III, and aVF, and extreme proportionally discordant STE in V2 and V3.
This is diagnostic of a proximal LAD occlusion, or possibly left main.
About 10 minutes later, and shortly before arrival at hospital (NOT Hennepin!), he recorded another:
time 20 minutes
It actually looks a bit better, but still diagnostic of LAD occlusion.
On arrival, the patient went into ventricular fibrillation.
For some reason unknown to me, the interventionalist was in the ED.  He refused to listen to the medic and was convinced that the ECG was due to hyperkalemia and that acute MI cannot be diagnosed in the presence of paced rhythm.    
Details cannot be shared here, but suffice it to say that inability to recognize acute occlusive myocardial infarction in the presence of ventricular paced rhythm contributed to a poor outcome.
This appears to be a common misconception that MI cannot be diagnosed in paced rhythm.  After showing an ECG with paced rhythm that clearly shows acute MI, K. Wang asked readers for an interpretation.

Here is the survey:

I’m not sure why it does not add up to 100%.  Until recently, it was 50-50.

While it is true that old MI, as diagnosed by Q-waves, is greatly obscured by LBBB and paced rhythm, acute occlusive MI appears to be fairly readily diagnosable, almost as much as in normal conduction.  In normal conduction, (i.e., no LBBB or paced rhythm), acute coronary occlusion MI (OMI) is diagnosed in approximately 75% of cases.  The other 25% are not diagnosed until rule-in by troponin and next day angiography.(1)
1. Khan AR, Golwala H, Tripathi A, et al. Impact of total occlusion of culprit artery in acute non-ST elevation myocardial infarction: a systematic review and meta-analysis.  European Heart Journal 38(41):3082-3089; November 1, 2017.  In this study, of 40,000 patients with NonSTEMI, 10,000 had OMI that was not diagnosed until the next day.
We have shown the ACO (OMI) can indeed be diagnosed in paced rhythm with comparable sensitivity and specificity.   It is important to realize however that many OMI may be difficult to diagnose in any ECG rhythm, and that NOMI (non-occlusive MI) is often impossible to diagnose on the ECG.
Here is our abstract from SAEM 2018.  We are writing the manuscript right now, with more patients enrolled.
The Smith-Modified Sgarbossa Criteria Accurately Diagnose Acute Coronary Occlusion in Emergency Department Patients With Ventricular Paced Rhythm.  
Kenneth W. Dodd1 , Deborah L. Zvosec2 , Michael A. Hart1 , Laura Bannister3 , Gary Mitchell4 , George Glass5 , Brooks Walsh6 , Harvey P. Meyers7 , David Miranda8 , James Corbett-Detig9 , Vaishal M. Tolia10, Louise Cullen11, Stuart Zarich6 , Sally Aldous12, William Brady5 , Timothy Henry13, Adam J. Singer14, Stephen W. Smith15, and PERFECT Study Author Group 1 Hennepin County Medical Center, 2 Minneapolis Medical Research Foundation, 3
Background: The Smith-modified Sgarbossa criteria (MSC) are frequently recommended for diagnosing acute coronary occlusion (ACO; STEMI-equivalent) in the setting of ventricular paced rhythm (VPR). The MSC are positive if one of the following criteria are met in at least 1 lead: concordant ST Elevation (STE) of at least 1 mm, concordant ST depression of at least 1 mm in V1-V3, or ST/S ratio less than -0.25 in leads with at least 1 mm STE. We hypothesized that the MSC will have higher sensitivity for diagnosis of ACO in VPR when compared to the original Sgarbossa criteria. Methods: The PERFECT study (#NCT02765477) is a retrospective, 16 center, international investigation of ED patients from 1/2008 – 12/2016 with VPR on the ECG and symptoms of acute coronary syndrome (ACS). Data from ten sites are presented here. Acute myocardial infarction (AMI) was defined by the Third Universal Definition of AMI. For this analysis, ACO was defined as angiographic evidence of coronary thrombosis with peak cardiac troponin-I (cTn-I) at least 10 ng/mL or cTn-T ≥ 1 ng/ mL. Blinded physicians adjudicated angiogram reports for coronary lesions and thrombolysis in myocardial infarction (TIMI) flow score. Separate, blinded physicians performed ECG measurements. Trained abstractors recorded data on standardized forms. Statistics were by Mann Whitney U, Chi-square, and McNemar’s test. Results: There were 46 encounters in the ACO group (median age 76 [IQR 65-82], 36 (76%) male) and 79 in the No-AMI group (median age 70 [61-75], 48 (61%) male). For ACO, median peak cTn-I was 65 ug/L [IQR 35-239] and cTn-T 3.3 ng/mL [IQR 2.2-8.3]. For No-AMI, median peak cTn-I was 0.015 ug/L [IQR 0-0.09]. For ACO, the sensitivity and specificity of the MSC and the original Sgarbossa criteria were 83% (95%CI 68-91) versus 63% (48-76; p less than 0.005) and 99% (92-100) versus 99% (92-100; p = 0.5). In pre-defined subgroup analysis of patients with TIMI 0-1 flow and peak cTnI at least 10 ug/L or peak cTnT at least 1 ng/mL (n = 29), the sensitivity was 87% (69-96) for the MSC versus 58% (39-75) for original Sgarbossa criteria (p less than 0.05). Conclusion: This represents the largest study of patients with VPR and angiographically-proven ACO. The MSC were highly sensitive and specific for the diagnosis of ACO in patients presenting to the ED with VPR and symptoms of acute coronary syndrome.

Comment by KEN GRAUER, MD (10/3/2018):
How good are paced tracings for detecting ACO (Acute Coronary Occlusion)? The answer is, much better than has previously been appreciated by still all-too-many clinicians. The goal of this blog post (and of the ongoing study by Dodd, Smith, et al — referenced above) is to increase appreciation of the need to pay special attention to paced ECGs in patients with new symptoms — since surprisingly often, ACO can be suspected on ECG despite cardiac pacing. And, in cases like the elderly patient with new-onset chest pain presented here — definitive diagnosis of acute STEMI is sometimes deceptively easy.
  • The abstract by Dodd, Smith et al (referenced above) focuses on application of quantitative criteria in the Smith-modified Sgarbossa (MSC) score.
  • My comments are aimed at qualitative assessment of ST-T wave changes in paced tracings. I selected the first 2 ECGs in this case to illustrate these points (Figure-1).
Figure-1: Comparison of the first 2 ECGs in this case (See text).

The easiest way to improve one’s ability to recognize suspicious findings in the ECG of a completely paced rhythm is to: iRealize this is a high prevalence population for coronary disease (ie, after all, these patients have a permanent pacer); and, iiListen to the patient. If a patient with a permanent pacer complains of worrisome new-onset symptoms — before you even look at the ECG — know that you are dealing with a very high prevalence population for possible acute occlusive coronary disease.
  • NOTE: As obvious as the above sounds — I have been amazed at how many clinicians on the numerous international ECG blogs I regularly participate in are still “stuck” in thinking it impossible to see acute ECG changes in a paced tracing — thereby ignoring the patient in front of them who is describing typical symptoms of acute coronary syndrome. It’s much easier to pick up abnormal ECG findings when you specifically look for them!
  • My qualitative approach to ECG assessment of paced tracings in patients with new-onset chest pain is as follows — Be suspicious when you see ST-T wave findings that shouldn’t be there
ECG Findings that Shouldn’t Be There:
To facilitate identification of ST segment deviations (elevation or depression) — I have added dotted BLUE lines in Figure-1 that I have drawn parallel to the heavy grid lines. These dotted blue lines correspond to the end of the QRS complex, and the beginning of the ST segment. Note the following in ECG#1 (TOP tracing in Figure-1):
  • The shape of the ST-T wave in the inferior leads in ECG #1 looks strange. This is best seen in lead II, but is also clearly present in lead aVF. The slope of the ST segment in these leads is clearly downward, and the T wave looks biphasic. These are changes that shouldn’t be there in these leads with an all negative QRS complex. The fact that ST-T wave appearance looks far less abnormal in lead III probably accounts for the lack of ST elevation in lead aVL in ECG #1.
  • Admittedly, artifact and beat-to-beat variation in morphology makes assessment of ST-T wave changes more difficult in Figure-1. But the ST-T in lead V5 resembles that seen in leads II and aVF — and in a lead in which the QRS complex is all negative, this negative ST segment and biphasic T wave that we see in V5 just shouldn’t be there. By comparison — Note that the T wave in lead V4 of ECG #1 is upright despite the negative QRS complex. There is no way the ST-T wave should change from lead V4-to-V5 as we see here given the similar all negative QRS appearance in these 2 leads, unless there is acute ischemia.
  • There are other more subtle ST-T wave changes in ECG #1 that by themselves would seem too nonspecific to be diagnostic — but in the context of clearly abnormal changes in leads II, aVF and V5 — serve to increase my suspicion of acute ischemia. These include disproportionately large upright and negative T waves in leads V2 and V6, respectively.
  • BOTTOM LINE: I didn’t think the ECG picture for the paced tracing in ECG #1 was definitive for acute occlusion. That said, the above noted changes in a paced tracing of a patient with new-onset chest pain was clearly suspicious to me for acute ischemia — and merited ongoing close follow-up.
13 minutes later — ECG #was obtained (BOTTOM tracing in Figure-1):
  • Now there is frank ST elevation in multiple leads that shouldn’t be there. These leads include lead aVL; and V2, V3, V4. Comparison of the ST-T wave appearance in these leads 13 minutes earlier in ECG #1 verifies beyond doubt that these are all acute changes.
  • The mirror-image opposite ST-T wave picture of ST elevation in lead aVL is seen in all 3 inferior leads (the dotted horizontal RED line on the ST segment baseline for leads II, III, aVF facilitates recognition of dramatic reciprocal  ST depression).
  • BOTTOM LINE: Despite cardiac pacing — ECG #2 is diagnostic of a very large acute anterior STEMI. Immediate cath and reperfusion are urgently indicated.

Powered by WPeMatico