Killer ECG Patterns: Part 2

• ST elevation in lead I, aVL and V1-2
• ST depression and T-wave inversion in lead III
• Hyperacute T-waves and pathological Q waves in V1-2 — note the way that T waves “tower” over the preceding QRS complex in V1
• Trace ST elevation in V5-6

This ECG pattern, also known as the South Africa Flag Sign, is typically seen in “high lateral” Occlusion Myocardial Infarction (OMI), often due to complete occlusion of the first diagonal branch of the LAD (LAD-D1). Trace ST elevation seen here in leads V5-6 may be due to lateral wall involvement or early repolarisation.

There is no lead that directly reflects the “high lateral” region usually supplied by the LAD-D1 vessel. Infarction of this territory therefore often causes subtle ST segment changes that fail to meet STEMI criteria. Lead III is directly reciprocal to this region, and thus ST depression and T-wave inversion in lead III is often more pronounced than any ST elevation seen elsewhere (note above example).

NB: Although the “high lateral” term is so commonly used, there is no true high lateral wall in the left ventricle, and the term mid-anterior or basal MI is more anatomically correct as these are the corresponding segments.

This patient underwent emergent angiography demonstrating a 100% occlusion of the first diagonal branch of the LAD, as well as severe triple vessel disease. Successful PCI was performed to his LAD-D1 with resultant TIMI-3 flow.

See ECG Case 121 for a more in-depth discussion of this case.

• ST depression in leads V1-5 and I, maximal in V2-4 with horizontal ST segment morphology
• Trace ST elevation in lead III
• ST depression not extending to V6, and a lack of reciprocal ST elevation in aVR, suggests that these changes are not due to diffuse subendocardial ischaemia

The ECG pattern of ST depression maximal in leads V2-4, without progression to V5-6, is diagnostic of posterior OMI until proven otherwise.

Whilst posterior leads V7-9 should be performed, they may be falsely negative and should not dissuade against emergent reperfusion in the presence of the above changes. Electricity conducts poorly through aerated lung and thus ST elevations can be significantly reduced or absent in posterior leads, making standard ST elevation measurements inappropriate for detecting occlusion.

Angiography in this patient revealed a 100% proximal circumflex occlusion with TIMI 0 flow.

Isolated posterior OMI is most commonly due to acute proximal occlusion of the left circumflex artery. There will often be associated changes suggestive of early inferior wall injury (note lead III in above example) — such subtle ST elevation will likely not meet traditional STEMI criteria and may be overlooked on initial presentation.

Let’s take a look at another example:

• ST depression in precordial leads V2-4, without extension to V5-6
• Pathological Q waves and trace ST elevation in inferior leads II, III, aVF

This ECG is diagnostic for acute inferoposterior OMI. The patient went on to have delayed angiography revealing a complete proximal left circumflex artery occlusion.

• Up-sloping ST depression (> 1mm at J point) in precordial leads V2-6, plus leads I and II
• De Winter’s T waves — peaked anterior T waves, with the ascending limb of the T wave commencing below the isoelectric baseline
• ST elevation in aVR is simply a reciprocal change to widespread ST depression

The De Winter T wave pattern is seen in approximately 2% of acute LAD occlusions and is often under-recognised by clinicians, leading to delays in reperfusion therapy. Most guidelines now consider this pattern a STEMI-equivalent and indication for immediate reperfusion therapy.

Typical STEMI morphology may precede or follow the De Winter pattern. Take a look at this example of acute proximal LAD occlusion with features of both De Winter’s T waves and classical anterior STEMI:

• There is ST elevation in septal (V1-2) and high lateral (I and aVL) leads, with inferior reciprocal change, consistent with LAD occlusion proximal to the first diagonal branch
• Up-sloping ST depression and peaked T waves (De Winter’s T waves) in anterolateral leads V3-6

De Winter’s T waves — ECG Diagnostic Criteria

• T waves in leads III and aVF appear to tower over the preceding QRS complex, and have a wide, “bulky” appearance. Such changes are consistent with hyperacute T waves (HATW) representative of hyperacute occlusion and evolving ST elevation
• There is reciprocal ST depression in V1-3 and I

Note that we expect to see a “normal” degree of ST elevation in leads V2-3, hence an isoelectric ST segment in these leads (as seen in V3 here) should be considered relative ST depression and concerning for reciprocal change.

These changes are suggestive of hyperacute inferior OMI, likely due to occlusion of a dominant RCA given the lack of ST elevation seen in lateral leads. This ECG was faxed from the pre-hospital environment — by the time of arrival to an emergency department it is likely that we would see classical ST elevation evolving.

There is no formal, universal definition of what represents a HATW, however it is recognised that the ratio of T wave amplitude to the preceding QRS complex is of more significance than overall T wave size. HATWs are wider and generally more symmetric than normal T-waves. As is the case in bundle branch block, abnormal depolarisation should be followed by abnormal repolarisation. This extends to the context of a low amplitude QRS complex, which should be followed by a relatively low voltage T wave.

Serial ECGs should be performed as these changes generally precede classic STE findings or resolve if there is spontaneous reperfusion.

• Right ventricular paced rhythm with left bundle branch block (LBBB) morphology
• > 1mm concordant ST elevation in leads V4-5 and aVL — this is positive for both original and Smith-modified Sgarbossa criteria and indicates acute coronary occlusion
• ST elevation in leads V2-3 is excessively discordant and also positive for Smith-modified Sgarbossa criteria
• Concordant ST depression in inferior leads III and aVF is suggestive of reciprocal change

These changes are diagnostic of anterolateral OMI, and angiography revealed a 100% proximal LAD occlusion.

Although this is a relatively clear example, diagnosing acute coronary occlusion in patients with LBBB or a ventricular paced rhythm can be difficult. Abnormal depolarisation means we expect to see abnormal repolarisation and associated discordant ST segment changes. As such, a degree of ST elevation is considered acceptable and normal.

The Sgarbossa criteria allows us to determine when the nature of ST segment changes in LBBB is abnormal and suggestive of occlusion. The Smith-modified Sgarbossa criteria is binary, and takes into account size of the preceding QRS complex and proportionality of changes. Acute occlusion is diagnosed if any of the three criteria are met:

Whilst these criteria have not been validated in patients with right bundle branch block (RBBB), similar concordant changes may indicate occlusion and should prompt closer examination for other features suggestive of evolving infarction.

• Right ventricular paced rhythm with LBBB morphology
• Concordant ST depression in V2 and V3 is positive for Smith-modified Sgarbossa criteria
• T waves in inferior leads II and III appear hyperacute in nature

Note that whilst there is concordant ST depression extending to leads V4-6, these changes must be seen in V1-3 to be considered Sgarbossa-positive.

The morphology in V2-5 is reminiscent of posterior OMI, with horizontal ST depression and prominent upright T waves. This patient had a confirmed posterior infarction, requiring PCI to a completely occluded posterolateral branch of the RCA.

• Right ventricular paced rhythm with LBBB morphology
• Degree of ST elevation in leads V2-5 is > 25% of depth of the preceding S wave — this is positive for Smith-modified Sgarbossa Criteria
• Note concordant ST elevation in leads I and V6 is also positive for Sgarbossa criteria

This ECG is again diagnostic of anterolateral OMI — angiography revealed a 100% mid-LAD occlusion.

This patient actually had an ECG 60 minutes earlier on arrival to the emergency department:

• Note concordant ST elevation in lead I with a positive QRS — this isolated finding is Sgarbossa-positive but subtle and easy to miss, as it was in this case

• ST elevation in V1-6 plus I and aVL (most marked in V2-4)
• Reciprocal ST depression in III and aVF
• Pathological Q waves in V1-2, with reduced R wave height (a Q wave equivalent) in V3-4

This patient is having an extensive anterolateral STEMI. ST elevation in V1 suggests proximal LAD occlusion. There is a premature ventricular complex (PVC) with “R on T” phenomenon at the end of the ECG — this puts our patient at risk for malignant ventricular arrhythmias.

Whilst definitions exist for the degree of ST elevation which constitutes a “STEMI”, these authors do not rely on this to diagnose occlusion on the ECG. Firstly, there are many (particularly young) patients in which ST elevation seen will meet STEMI criteria but is simply normal variant due to early repolarisation — the experienced clinician will recognise and blissfully ignore such changes in the context of chest pain. Secondly, it is the constitution of changes which diagnoses acute occlusion. Even in our above clear-cut example, we see hyperacute T waves, pathological Q waves, and reciprocal ST depression. These may be the first features seen prior to ST segment elevation.

My advice would be to flip the way that we look for infarction on the ECG. We are all taught to examine closely for ST elevation, but a large proportion of the time such changes are normal variant. ST depression on the other hand is always an abnormal finding.

Look for ST depression first, then examine reciprocal leads closely for subtle ST elevation, hyperacute T waves, pathological Q waves and/or reduced R wave height.

• Pseudonormalisation refers to the appearance of a normal T wave morphology on the background of Wellens ECG changes – such changes actually represent HATW due to acute occlusion and may occur in patients experiencing chest pain

See this excellent diagram from @smithECGblog to better understand the evolution of these changes: