Post-revascularisation de Winter Pattern on Electrocardiography: Not Always the Left Anterior Descending Artery!

The de Winter pattern on ECG is conventionally associated with occlusive MI located in the proximal left anterior descending (LAD) artery. This pattern is often described as 1–3-mm upsloping ST-segment depression at the J point in the precordial ECG leads (V1–V6), with associated tall, positive symmetrical T waves in the same leads.^1,2 We report a unique case of a patient presenting with an acute posterior ST-segment elevation MI that had successfully undergone percutaneous coronary intervention (PCI), with the de Winter ECG pattern seen after revascularisation.

Case Presentation

A 71-year-old man presented with acute chest pain radiating to the neck, along with shortness of breath and palpitations. Aside from a significant smoking history, he had no known medical illness. He had first presented to a neighbouring hospital 2 hours prior but requested to be discharged at his own risk due to financial constraints. He later presented to our centre at 6 hours after onset of original symptom. His initial vital signs included a blood pressure of 105/70 mmHg following inotropic support in the form of IV infusion of dobutamine, pulse rate of 58 BPM of regular rhythm, respiratory rate of 20 breaths per minute and oxygen saturation of 93% requiring oxygen supplementation of 2 l/min. Initial ECG showed ST-segment depression in leads V2–V3, T wave inversions in leads aVL and V1–V4, and mild ST-segment elevation in leads V6, II, III and aVF (Figure 1 ). Following stabilisation, coronary angiography was performed (Figure 2). There was evidence of severe stenosis in both the proximal and distal left circumflex artery (LCx). The right coronary artery was both small and non-dominant, which suggests that the coronary lesions in the LCx were the likely culprit lesions. PCI was performed on both lesions, with a 2.5 × 22 mm drug-eluted balloon used in the distal LCx artery and 2.5 × 18 mm drug-eluted stent deployed in the proximal LCx artery, with subsequent thrombolysis in MI 3 flow within the artery.

Following transfer of the patient to the recovery bay, a post-reperfusion ECG was performed, revealing an upsloping ST-segment depression in leads V2–V4, with tall, symmetrical T waves seen in those leads signifying the de Winter pattern (Figure 3). Furthermore, the mild ST-segment elevation in leads V6, II, III and aVF visible from his initial ECG on presentation was no longer seen, hinting at transmural ischaemia in the territories overlying these leads rather than simply being a normal variant ST-segment for this patient. His peak high-sensitivity troponin T was 35,040 ng/l. A transthoracic echocardiogram (TTE) performed within 24 hours of PCI revealed a left ventricular ejection fraction of 40–45% with akinesia seen in the mid to apical inferolateral and mid to apical inferior territories. There was subsequent resolution of ECG pattern by day four of his inpatient stay, and he was subsequently discharged well with plans for secondary prevention and close follow-up. A repeat TTE performed at 6 weeks revealed a left ventricular ejection fraction of 50–55% with no regional wall motion abnormalities. The patient was at New York Heart Association functional class 1 and asymptomatic.

Discussion

There are various non-ST-segment changes on ECG that can be significant for MI, such as the de Winter ECG pattern representing regional subendocardial ischaemia and the Wellens’ sign representing post-ischaemia transient state reflecting myocardial oedema, as opposed to left ventricular dysfunction.^3,4 The de Winter ECG pattern is often associated with proximal LAD artery occlusions conventionally, although there have been several reports of de Winter pattern seen in other circumstances.^5–8 In fact, determining the exact location of coronary occlusion based on lead location of the de Winter pattern has not been shown to be completely accurate.⁷ The underlying pathophysiological process leading to de Winter pattern is largely unknown, but has been postulated to be due to regional subendocardial ischaemia with myocardial protection through either collateral circulation, ischaemic preconditioning or existing forward flow.⁷ The upsloping ST-segment depression and peaked T waves follow hypoxia-driven alterations in the adenosine triphosphate-dependent potassium channels within the myocardium. This results in sub-endocardial delayed repolarisation.⁷

In our patient, the de Winter pattern seen following successful revascularisation may have indicated ongoing, albeit less severe subendocardial ischaemia as a sequela of the initial insult from a subtotal coronary occlusion. ‘Recovery’ of the patient’s ECG may have been delayed due to the severity of the event (i.e., the patient was of Killip class 4 on arrival at our emergency department) with possible substantial amounts of myocardial oedema already developing, although this, we believe, had subsequently improved over time as evident from his TTE findings. Although cardiac MRI would have been helpful to confirm the presence or absence of oedema, this was not performed in our case. Alternatively, the pattern may also be reflective of reciprocal changes from a posterior infarct that has progressed with time and initial T wave inversions within the anterior ECG leads progressed into peaked T waves for the duration.

This present case highlights a unique manifestation of post-reperfusion of acute coronary syndrome and ischaemia recovery in the form of the de Winter pattern on ECG. Understanding the dynamics of myocardial oedema, myocardial dysfunction and repolarisation changes on ECG is important, and the clinical presentation of patients remains paramount so as to not mistakenly diagnose them as having a re-occlusion of their coronary arteries.