Sunday, May 30, 2021

When the ECG is more revealing than the HPI


Written by Alex Bracey with edits from Pendell Meyers and Steve Smith

A woman in her 60s presented to the ED as a referral from an urgent care for weakness. When I interviewed her, she reported that she had experienced several months of shortness of breath and fatigue, which had worsened in the last several days. She had also experienced new dyspnea on exertion, along with a non-productive cough and fatigue. The only medical problem she was aware of was hypertension treated with hydrochlorothiazide. An ECG was performed as follows:

Sinus tachycardia
Convex, upward ST elevation in V1-V4 with subsequent T wave inversion
Right axis deviation

Based on this ECG, I was concerned about right heart strain, including pulmonary embolism (PE); however, this ECG could be seen in any condition that causes right heart strain, both acute and chronic. Her story did not neatly fit into any one diagnosis. Typically, my practice is to use a risk-stratification tool (e.g., Wells' Criteria for PE, modified Geneva) to help guide my diagnostic testing in the evaluation of PE. Importantly, ECG is not used in these commonly utilized tools. Despite an HPI and an exam that did not suggest PE as a clear diagnosis, I ordered a CT angiography of the pulmonary arteries while my resident performed a bedside cardiac ultrasound:

Imperfect parasternal short axis view 
Right ventricle is clearly greater than one-third the size of the left ventricle (nearly equal)
Septal flattening with probable septal bowing (D-sign)

Initial troponin I was undetectably low (negative), BNP was elevated to 582 pg/mL (reference values 0 - 100) and lactic acid was elevated to 2.2 mmol/L. A D-dimer ordered from triage resulted at 3.30 (Reference: <0.50 mg/L). Shortly afterwards, the patient underwent CTPA:

CT angiography of the pulmonary arteries in the axial view demonstrating bilateral, central pulmonary emboli

The patient was started on anticoagulant therapy and had an uncomplicated hospital course. 

Teaching points:

- Patients often have presentations that are not "classic" for a given disease. Paying attention to clues from ancillary testing may help to better diagnose, and ultimately to care for our patients. The subtle findings on this ECG helped to expediently come to the correct diagnosis in this case.
- ECG findings are not used in many commonly used risk-stratification tools for PE. In this case, had PE been considered as a diagnosis based on history and a risk-stratification tool subsequently been applied, testing with D-dimer would have been indicated and have had a positive result. Therefore, a CTPA would have been performed and the PE discovered. However, the ECG may heighten concern and expedite care, such as in this case. 
- The ECG findings in this case can be caused by process causing right heart strain, in this case pulmonary embolism was the culprit.

Other cases of right heart strain and pulmonary embolism on the blog:


A man in his 50s with schizophrenia, hypoglycemia, and vague chest pain and shortness of breath for a few days

Written by Pendell Meyers with edits by Steve Smith

Bystanders called EMS for a man in his 50s "not acting right." Medics found a man with altered mental status, immediately measured a blood glucose of 42 ng/mL, and administered glucose.  There was immediate improvement in his mentation, but it was "not back to normal." Now that the patient was able to give some history, he was able to complain of chest pain and shortness of breath off an on for "a few days." 

The impression that I get from the documentation is that the patient was still felt to be somewhat altered, and the history was felt to be somewhat unreliable and vague. The notes report that the patient has a history of schizophrenia, HTN, diabetes, and atrial fibrillation not on anticoagulation due to noncompliance. At triage, the patient was again able to independently report "shortness of breath" as the chief complaint. His vitals were within normal limits.

His triage ECG at 1741 is shown here:

What do you think?

A baseline ECG was available:

I see atrial fibrillation without any clear signs of ischemia. Borderline low voltage. Even with the baseline ECG available, I don't see a meaningful change.

A broad workup was ordered, including troponin and head CT. 

The head CT showed no acute intracranial abnormality. The patient was given food, and serial fingersticks did not show any further hypoglycemia. All reports continue to describe the patient as "encephalopathic" and a "poor historian," but all reports also agree that the patient would complain of vague chest pain and shortness of breath when asked.

The initial high sensitivity troponin I returned elevated at 62 ng/L (URL 14 ng/L). Due to the elevated troponin, another ECG was ordered at 2036 (unclear whether the patient had ongoing symptoms at this time):

This ECG is importantly and diagnostic different than the first! 
There is new STE in V3 through V5, and perhaps a hint inferiorly (but the baseline is not perfectly clear). More importantly the T waves in V3-V6, and also in inferior leads, are newly hyperacute compared to the first ECG.

These changes were not appreciated, and no more troponins were ordered in the ED. The patient was admitted to medicine.

The medicine team decided to trend troponins overnight, which showed a steady rise of 1260, 2749, and 6210 ng/L. 

At 0630 that morning, another ECG was ordered (unclear if due to change in pain, or because of the consistently rising troponin):

Evolution to obvious STEMI(+) OMI. Interestingly, the inferior hyperacute T waves are no longer hyperacute.

For unclear reasons, nothing seems to happen immediately based on this ECG.

At 0750, anther ECG is performed:

Still obvious STEMI(+) OMI.

Angiogram was finally done at around 0930 which showed a proximal LAD dissection with distal LAD thrombotic occlusion (TIMI 0 flow). They elected to treat the dissection medically. The report does not mention any intervention or aspiration of the distal LAD thrombus, and does not mention the state of the distal artery flow at the end of the case, but the post-cath note describes resolution of chest pain (and the post-cath ECGs below show reperfusion).

Here is the ECG after cath:

Beginning of reperfusion.

The next troponin resulted at greater than 25,000 ng/L (our lab does not report higher), and no further measurements were done.

Day 3 ECG 0900:

Day 3 1300:

Here is the whole sequence of ECGs, from baseline through reperfusion, showing just leads V3-6 side by side, with the relevant portions of the OMI progression for reference underneath:

Learning Points:

Serial ECGs with careful side by side inspection could have helped identify this OMI sooner.

Hyperacute T waves are not yet defined, but trained electrocardiographers know them when they see them. Serial ECGs can prove that otherwise reasonable T waves are actually hyperacute compared to prior. 

Ongoing ischemic symptoms, ongoing ECG ischemia, or ongoing troponin rise in ACS are reasons to pursue immediate reperfusion, and NOT to wait until clear STEMI criteria are manifest. These are indications for emergent angiogram in all guidelines around the world, but are simply not followed in many settings for unclear reasons. The STEMI paradigm makes most providers believe that they are not missing OMI, makes them believe there is no need to ever treat an NSTEMI emergently.

Remember, the myocytes do not "know" the etiology of their focal full thickness complete ischemia - the ECG findings are the same regardless of the etiology of OMI. Angiogram is the only way to tell whether it will be one of the majority of cases that benefit from an intervention. Lack of a discrete intervention does not mean that the patient and patients like him do not need to be taken to the cath lab.

Maintaining a high level of suspicion, particularly in those who may not be able to advocate for themselves (e.g., hx of psychiatric disorders) when faced with abnormal findings, such as a rising troponin, and learning to recognize electrocardiographic findings for OMI beyond STEMI can help to deliver the highest quality, most expedient care for our patients.

Thursday, May 27, 2021

Why is there ST Elevation in lead V2? Think Lead Placement.

 I was shown this ECG of a 40-something intoxicated male with altered mental status.  The provider was very worried about LAD occlusion.

What do you think?

There is ST Elevation in lead V2, but there is also an RSR' wave which creates a kind of saddle ("saddleback pattern").  This is rarely due to anterior MI.  

There is also tachycardia, which unless a patient has cardiogenic shock, is also uncommonly due to ACS.  

And, of course, without chest pain, the pretest probability is very low.

RSR' and saddleback can be due to leads being placed too high, but since the P-wave is upright in lead V2, I did not think that was the problem.

I said it was unlikely to be ischemic.  I recommended IV fluids and to measure a troponin.  A cardiac POCUS would be useful to be certain there is good LV function.  One would expect it to be hyperdynamic.

Then record another ECG after that.  And measure troponins.

3 hours later I went to see if another ECG was recorded.  As I walked into the patient's room, this ECG was being printed out from the ECG machine:

There is still an RSR' and STE in V2.
I looked at the lead placement.
Leads V1 and V2 were far too high on the chest.
I placed them correctly and instructed the tech to record again.

Here is what came out:
Correcting the lead placement resulted in a more normal ECG

All trops were negative.

Learning Points:

1. Pretest probability is crucial
2. Saddleback (which is almost always lead V2) is unusual as a manifestation of OMI
3. Lead placement is crucial.  V1 and V2 are often placed too high.  They should be in the 4th intercostal space.  V4-V6 in the 5th intercostal space.  4th and 5th spaces are not that far apart; thus, the leads should ALMOST be horizontal across the chest.  See images below.
4. Tachycardia should make you suspicious of another etiology than ACS
5. Placing leads V1 and V2 too high can result in:
   a. Q-waves mimicking septal MI
   b. RSR' mimicking Brugada
   c. ST Elevation mimicking Acute MI (OMI)
   d. Saddleback mimicking Type 2 Brugada
   e.  Intentional placement too high may uncover Brugada.
   f.  Usually if placed too high, P-wave in V2 is inverted (it should always be upright if leads are correctly placed)

Correct lead placement, from LITFL:

Incorrect lead placement, seen in similar images all over the internet:

Article 1: Great article by Brooks Walsh:

The precordial electrocardiogram (ECG) leads V1 and V2 are often misplaced. Such misplacement usually involves placing these leads too high on the chest. The resulting ECG may generate erroneous ECG patterns: e.g. incomplete right bundle branch block, anterior T wave inversion, septal Q waves, ST-segment elevation. These features may falsely suggest acute or old cardiac ischemiapulmonary embolism, or a type-2 Brugada pattern. On rare occasion, conversely, high placement of V1 and V2 may reveal a true type-1 Brugada pattern. The emergency clinician needs to be aware of the possibility of lead misplacement, and should know how to suspect it based on unusual P wave morphology in V1 and V2.

Article 2:  Walsh B, Watford C, DeGiulio V, Oto B, Sifford D, Grauer K, Smith SW.  google Image is a bad way to learn ECG lead placement: we all agree.  SMACC-Dub. June 2016.

Tuesday, May 25, 2021

Atrial Fibrillation w Rapid Ventricular Response and ST Depression Maximal V1-V4: Not always subendocardial.

A patient presented a few years ago with chest pain, but also cough, low grade fever, and malaise.  She had one prehospital saturation at 88%, but otherwise all vital signs and labs were normal, without tachycardia, elevated BP or anemia.  She was never in any distress.

She had this prehospital ECG:

What do you think?

There is diffuse ST depression, in I, II, III, aVF and V3-V6.  It is diagnostic of ischemia.  It is maximal in V5 and II, with reciprocal STE in aVR, and is thus most consistent with subnendocardial ischemia (Posterior OMI would have maximal STD in V1-V4, and Posterolateral OMI would have STD in lead III but without STD in lead I).  So this is not typical of OMI.  Most likely, this patient does not have OMI.

By the time of arrival in the ED, the ECG had normalized.  It is not clear that any physician ever saw that prehospital ECG.

She was admitted, not given aspirin in the ED, and here are the serial troponins:  the first 6 are from a previous visit, showing that she does not have chronic myocardial injury (chronically elevated troponins).  The last 5 are from a subsequent visit showing she does not have chronic injury afterwards either. 

This is a rise and fall of troponin diagnostic of acute myocardial injury.  When there is proven ischemia, as there is by the ECG, then acute injury is acute MI. Then you must decide if it is type I or type II.  

She was diagnosed with "Type II MI" (due to supply demand mismatch).  However, there were no recorded supply/demand mismatch (no persistent hypoxia, tachycardia, anemia, hypertension, or hypotension) that could have cause type II MI, so I think this was a case of "Nah, couldn't be," and probably was a type 1 MI, though not OMI (type I Non-OMI, or NOMI).  

When a patient has not only chest pain, but also some other symptoms (cough, low grade fever, malaise), it is easy to explain away ischemia as NOT being due to ACS.  

She had an echo the next day: Moderate concentric left ventricular hypertrophy.  Normal left ventricular size and systolic function with an estimated EF of 66%.  But diffuse subendocardial ischemia often has no acute wall motion abnormality.  Same for NOMI.  And this is also sometimes true for OMI that has rapid reperfusion.

She did not have an ACS workup (no stress test, CT coronary angiogram (CTCA), nor angiogram)

I believe this was her first episode of ACS and a CTCA or angiogram would have significant coronary disease.  A stress test is much less sensitive, but might have revealed it.


A few years later, she presented with CP and tachycardia, and was found to be in new Atrial Fibrillation with RVR.  The previous visit was apparently not noticed.

Here is her ED ECG:

There is profound ST depression, maximal in V3
Normally, STD maximal in V1-V4 is all but diagnostic of posterior OMI (we have submitted a paper showing this).  
However, when a patient has tachycardia, especially due to atrial fibrillation with RVR, it is not uncommon for subendocardial ischemia to present with maximal STD in V1-V4. 
Before concluding that it is posterior OMI, one should slow the heart rate and re-assess.  

Case Progression

She was electrically cardioverted, and this was her ECG 1 minute after cardioversion:
The heart rate is normal in sinus.  The ischemia is resolved.
This suggests that all that ischemia on the first ECG was really demand ischemia.  It makes it likely that she has coronary disease, with stenoses, but much less likely that there is ACS.

ACS is due to ruptured plaque, as opposed to stable plaque.
Stable plaque with significant stenosis limits coronary flow when more flow is required to supply the tachycardic myocardium. and such plaque is contributory to a type II MI)

First hs troponin I = 15 ng/L (URL = 16)

Peak hs trop = 749 ng/L (this is typical of a type II MI, but suggests that she at least has coronary stenoses that make her vulnerable to tachycardia)

The patient was admitted.  Unfortunately, she had no cardiology consult.  She was discharged soon thereafter.

5 days later she had sudden severe crushing chest pain and called 911.

Medics again found her in atrial fib with RVR.  They attempted cardioversion twice without success.

Here is her ED ECG:

Again, atrial fibrillation with RVR and ST depression maximal in V3.

Cardioversion was attempted twice in the ED without success.

She was given 1 mg of ibutilide over 10 minutes and cardioverted again.  After this "facilitation" with ibutilide, cardioversion was successful.

See this for more information on ibutilide: What to do when Atrial Fib with RVR will not Electrically Cardiovert. And how do you measure the QT in Atrial Fib?

Here is her ECG 30 minutes after cardioversion:

There is persistent ST depression maximal in V3.

The patient still had chest pain.

This should be considered diagnostic of posterior OMI until proven otherwise.

Cardiology was consulted, but because it was "not a STEMI" and was due to "subendocardial ischemia," the patient was not taken to the cath lab.

The patient continued to have chest discomfort, but it was explained away by the admitting physician as "chest wall pain."

Troponins overnight:

First hs trop = 76 ng/L

Next is 4 hours later = 27,600 ng/L

Next is 2 hours after (t = 6 hours) = 43,300 ng/L

Next is 3 hours after that (t = 9 hours) = over 50,000 ng/L (too high to measure)

After many hours, the ECG normalized.

Angiogram the next day showed BOTH 3 vessel disease and a culprit in the obtuse marginal (off the circumflex) with thrombus and 95% open.

Echo showed a new lateral (posterior) wall motion abnormality.

This is not a STEMI, but is an OMI of the obtuse marginal that led to a lot of myocardial loss.

Learning Points:

1. Do not diagnose Type II MI unless there are definite causes of increase oxygen demand or decreased supply.

2. ST depression in leads I, II, III, aVF and V4-V6 is usually subendocardial ischemia.  The ST vector is towards leads II and V5, with reciprocal ST elevation in aVR

3. ST depression maximal in V1-V4 in usually due to posterior OMI, but can be due to atrial fibrillation with RVR or other stresses; in the latter situations, it should resolve after resolution of tachycardia.

4. Patients with 3 vessel disease who have sudden chest pain and EKG abnormalities must have some culprit for their acute presentation, even if it isn't seen (in this case it was seen).  Just because there is 3 vessel disease does not mean there is not also an OMI.

5. Many, or even most, OMI will be open by the time of angiogram, but not before they have done a lot of damage.  This artery was 95% open, but it had been closed long enough that the peak troponin was over 50,000, with a wall motion abnormality.

6. Whether it is OMI or not, and whether you believe the ECG or not, if the patient has chest discomfort uncontrolled by aspirin, heparin (or LMWH), and Nitroglycerine, then such refractory chest discomfort is an indication for emergent angiogram by ACC/AHA guidelines. 

Do not call it "chest wall pain."  "Control" of pain does NOT include using opiates or analgesics; you should commit the patient to the cath lab if such medications are required to control pain.

Friday, May 14, 2021

What are these ST elevations, ST depressions, and tall T waves diagnostic of?

 Written by Pendell Meyers

Let's see this presentation ECG without any context first (no baseline ECG was available)

What do you think? What will you do?

The ECG is diagnostic of severe hyperkalemia. There is sinus tachycardia, the beginning of QRS widening, tall pointy peaked T waves with little area compared to their height. There is STE in V1-V3, aVR, and aVL, with STD in II, III, aVF, V4-V6. The appearance in V1 is similar to the Brugada morphology (as is often the case in hyperkalemia and Na channel blocker effects causing STE). Together these ST elevations in the right precordial leads and aVL are classic pseudoSTEMI (or pseudoOMI) patterns seen in hyperkalemia (see the end of the post for links to many similar cases). This ECG shows no evidence of OMI despite the fact that it meets STEMI criteria - it is all hyperK. Of course there may also be an additional component of acidosis, peri-ROSC, etc, given the context below, but hyperkalemia is the most contributory factor.

This was a man in his 30s who was found down and altered. He was seen normal the day before, then found in his bed by a friend, unresponsive. CPR was administered. EMS arrived and found him in PEA arrest. He was given epinephrine, calcium, and bicarbonate, and after several rounds of CPR he was found to have a pulse. He was intubated in the field. He was tachycardic and hypotensive. EMS recorded this ECG above just before arriving to the ED.

When the ED provider saw this ECG, they activated the cath lab thinking it was "STEMI" (presumably, this means that they thought the ECG meant that the patient was suffering from an acute coronary occlusion). Although it is irrelevant, this ECG does have acute STE that meets STEMI criteria in V1 and V2.

Cardiology reviewed the ECG remotely and heard the initial history, and decided that they did not want to take the patient to the cath lab.

The initial potassium returned at 7.9 mEq/L. 

At this point, further history and physical exam was performed, and it was realized that the patient was suffering from severe rhabdomyolysis and lower extremity compartment syndrome from prolonged down time, presumably due to drug ingestion.

Calcium was given IV, as well as insulin. I am not sure what dose or type of calcium.

Within minutes of calcium administration, a repeat ECG was recorded:

Findings almost completely resolved.

The patient unfortunately suffered a long and rocky ICU course. 

Learning Points:

Hyperkalemia is one of the most important diagnoses that can be made using the ECG. We must memorize these hyperkalemia ECG patterns to recognize them immediately in practice. Failure to do so may lead to early diagnostic closure (like in his case), which may result in delayed care, unnecessary procedures, and patient harm!

When hyperkalemia causes STEMI mimics, in my experience the most common areas for the STE are in the right precordial leads (V1-V2), lead aVL, and the inferior leads.

ECGs are almost always better when interpreted with the clinical context, but some ECGs are diagnostic by themselves in confusing scenarios.

Here are more examples of hyperkalemia OMI and STEMI mimics:

Monday, May 10, 2021

Typical Chest Pain: Would you activate the cath lab? Would you advocate if the interventionalist was not interested?

A 50-something woman complained of acute chest pain radiating to the left arm, onset while driving.  It would briefly improve with NTG.

Here is the first ED ECG:

What do you think?

When I saw this, I immediately said: "This is Aslanger's Pattern." 

1. Inferior OMI, with STE in lead III only, and reciprocal STD in aVL.

2. Diffuse subendocardial ischemia (ST depression, STD, in I, II, V3-V6) with reciprocal STE in aVR.

Aslanger's pattern is a combination of inferior OMI and diffuse subendocardial ischemia.  The subendocardial ischemia produces an ST depression vector toward leads II and V5 (with reciprocal STE in aVR) and the simultaneous inferior OMI results in STE in lead III only, but not in the other inferior leads.  

If there is inferior OMI, why is there no STE in II and aVF?  Because the ST depression (STD) vector towards lead II cancels part of the STE vector that would otherwise manifest in II and aVF.  That STD vector cannot, however, cancel the STE in III, which is too far to the right.  In fact, the STE vector must be directly rightward (180 degrees, or minus 180 degrees) in order not to register in lead aVF.  But of course, that rightward STE vector also registers as STE in aVR.  

When there is subendocardial ischemia alone, the STE vector is towards aVR, or actually a bit higher than aVR; thus, there is no STE in III or any other inferior lead.

So: lead III registers STE of subepicardial ischemia due to inferior MI and lead aVR registers reciprocal STE, reciprocal to the STD vector towards II. 

Since the ECG shows BOTH inferior OMI AND subendocardial ischemia, this pattern is associated with BOTH:

1. RCA/Circumflex occlusion (depending on dominance) AND 

2. At least one other ischemic vessel, and often 3 vessel disease, simultaneously.

See below a description of the article by Aslanger.  

Clinical Progression

The team activated the cath lab, but the interventionalist was skeptical.

An initial troponin I returned slightly elevated (0.379 ng/mL), and the patient's pain worsened.

Another ECG was recorded:

The findings are slightly worse.

Does this patient need immediate cath?  

There is ECG evidence of OMI.  There is refractory chest pain and clear ACS.  If the patient could be made pain free, then cath could be delayed.  But that was not the case here.

The patient was taken to the cath lab and was found to have 3 vessel disease:

LAD: 70% stenosis

Circ: 90% proximal, 90% first obtuse marginal.  It seems that these were not recognized as possible culprits for inferior OMI.

RCA: Prox: 50%, Mid: 50%, Distal: 90%

He underwent CABG to the posterior descending artery (PDA) of the RCA and internal mammary to LAD.  

The circumflex lesions were not intervened upon, either by PCI or CABG.

Here was the immediate post CABG EKG:

NEW Inferior and anterior STEMI (+) OMI
This looks like a wraparound LAD, but could also be 2 STEMIs at once.

The patient was immediately taken back to the cath lab, where they found:

--100% occlusion of the internal mammary graft to the LAD, AND 

--3 TIMI-2 occlusions in the circumflex:  

90% ostial circumflex culprit with TIMI-2 flow

90% 1st obtuse marginal with TIMI-2 flow

90% ostial circumflex culprit with TIMI-2 flow

These 4 occlusions of the 2 arteries resulted in inferior and anterior OMI:

The mid LAD was successfully stented (not the graft) and all 3 circ lesions were successfully stented.

Aslanger's Pattern:

I helped him a little bit and so I am a co-author. 

A new electrocardiographic pattern indicating inferior myocardial infarction


We define a new ECG pattern that is not uncommon among patients classified as non-STEMI (6.3%).

This new pattern has only 1 lead with any ST elevation.  Guidelines require 2 consecutive leads to make the diagnosis of STEMI.  Thus, those with only 1 lead, especially if less than 1 mm, are labelled Non-STEMI.

13.3% of inferior STEMIs presents with this pattern and may be deprived of primary PCI because of the incorrect label of non-STEMI.

It indicates an acute atherothrombotic event that is frequently due to inferior MI despite ECG not showing contiguous STE.

The patients with this pattern have higher short- and long-term risk for mortality.

Since there is more than one critical lesion, it may cause confusion about which lesion should be emergently opened.

2 Groups below:

Group IA is NonSTEMI with Aslanger's pattern 

Group IB is NonSTEMI without Aslanger's pattern

Group IA was compared with Group IB in terms of clinical outcomes, the patients in Group IA had a higher troponin rise in the first 24- hours, higher infarct size as evidenced by 24-hour troponin levels, higher frequency of angiographic culprit lesion and higher frequency of composite ACO endpoint (Table 2). They also had a higher frequency of circumflex artery involvement as the infarct-related artery, higher frequency of multivessel disease, higher frequency of the presence of concurrent chronic total occlusion, and a higher in-hospital and one year mortality compared to Group IB.

Although the patients with this pattern are classified as non-STEMI, they have an acute atherothrombotic event frequently resulting in inferior OMI (more often occlusion or near occlusion of the circumflex artery than the right coronary artery) with at least one accompanying stable but critical stenosis in one of the non-infarct-related arteries. They tend to have multiple vessel disease, multiple comorbidities and higher baseline risk, and show an increased short- and long-term mortality.

Recommended Resources