What Does a Heart Attack Look Like on an ECG?

A heart attack typically announces itself on an ECG through ST-segment elevation, a distinctive upward shift in the portion of the tracing that falls between the heartbeat’s main spike and the recovery wave that follows it. But that single hallmark barely scratches the surface. The full ECG picture of a heart attack includes changes that arrive before, during, and after the ST segment moves, and an entire category of heart attacks produces no ST elevation at all. What a clinician actually sees on the paper depends on which artery is blocked, how much muscle is at risk, how long the blockage has been there, and whether the patient’s baseline ECG was normal to begin with.

The Earliest Clue Most People Have Never Heard Of

Before the ST segment rises, the very first ECG change is often a subtle reshaping of the T wave. These so-called hyperacute T waves become unusually tall, broad, and symmetric compared with the patient’s normal tracing. They are the earliest described ECG sign of acute ischemia, appearing before ST elevation develops.¹ The window in which hyperacute T waves appear alone, without accompanying ST changes, can be brief, sometimes just minutes. That narrow window is one reason why a single normal-looking ECG in someone with chest pain does not rule out a heart attack. It also explains why emergency departments often obtain serial tracings: the ECG you capture at minute five may look different from the one at minute twenty.

Recognizing hyperacute T waves is tricky because tall T waves can also appear in other situations, most commonly high potassium levels. The morphology differs: potassium-driven T waves tend to be very narrow and pointy, while the hyperacute T waves of early ischemia are broader and more dome-shaped. Still, the overlap is real enough that clinicians consider lab values alongside the tracing.

ST-Segment Elevation and the Classic STEMI Pattern

When a coronary artery becomes completely blocked, the muscle it feeds loses its normal electrical behavior. The injured cells leak current in a way that pushes the ST segment upward on the ECG leads facing the damaged zone. The diagnostic threshold is modest: ST elevation of at least 1 mm (0.1 millivolt) in two neighboring leads is the standard criterion for calling a STEMI, often accompanied by T-wave inversion or the development of abnormal Q waves in the same leads.² That pattern triggers the fastest treatment pathway available: immediate reperfusion, usually by catheterization and stent placement.

At the cellular level, the ST elevation is driven by the opening of specific potassium channels on the heart muscle cell surface. When oxygen supply drops and ATP runs out, these channels activate and shorten the electrical signal in the injured area. A study using genetically engineered mice that lacked these channels showed markedly suppressed ST elevation after coronary artery ligation, and a drug that blocks the same channels in normal mice had the same suppressive effect.³ In other words, the ST shift you see on the tracing is a direct electrical consequence of starving heart muscle trying to cope with energy failure.

Reciprocal Changes and Why They Matter

A detail that separates experienced ECG readers from beginners is the recognition of reciprocal ST depression: leads on the opposite side of the heart from the injury zone often show the ST segment dipping below baseline. These mirror-image changes occur because the same electrical injury current that pushes the segment up in one direction pulls it down when viewed from the opposite angle.⁴

Reciprocal depression is not just a curiosity. In cases where the ST elevation itself is borderline or subtle, spotting the corresponding dip on the opposite wall can clinch the diagnosis. One study found reciprocal changes present in about 58% of heart attacks with only minimal ST elevation, making them an extremely useful clue when the primary changes are easy to miss.⁵ If someone with chest pain has ST depression in leads II, III, and aVF alongside even slight elevation in the high lateral leads, the combination is much more convincing than either finding alone.

When There Is No ST Elevation at All

Not every heart attack produces the classic upward shift. Non-ST-elevation myocardial infarctions (NSTEMIs) account for a large share of heart attacks and look quite different on the tracing. Instead of ST elevation, the ECG may show horizontal or downward-sloping ST depression, T-wave inversions, or sometimes both. The working definition used in clinical research is new ST depression of more than 0.05 millivolts in two neighboring leads, or new T-wave inversion deeper than 0.1 millivolts in leads with a prominent upward deflection.⁶

These findings carry real prognostic weight. A prehospital study of patients transported by ambulance for suspected heart problems found that those with T-wave inversion on their initial ECG had roughly 60% higher odds of developing new heart failure, and those with ST depression had higher odds of heart failure and death within 30 days.⁷ The absence of dramatic ST elevation does not mean the situation is benign; it just means the blockage is likely incomplete, or the damage involves the inner layer of the heart wall rather than its full thickness.

Which Leads Point to Which Artery

An ECG does not just confirm that a heart attack is happening. It also narrows down where. Each of the standard twelve leads looks at the heart from a different angle, so the pattern of abnormal leads acts like a rough map of the blocked artery.

Changes concentrated in leads I, aVL, and the chest leads V1 through V4 are strongly associated with a blockage in the left anterior descending artery, which supplies the front wall of the heart. Changes in leads II, III, and aVF point toward either the right coronary artery or the circumflex artery, both of which feed the bottom wall.⁸ This localization matters because the location of the blockage influences the risk of complications and sometimes alters the treatment strategy.

The Heart’s Blind Spots on a Standard ECG

A standard 12-lead ECG does not see every part of the heart equally well. Two areas are particularly easy to miss: the back wall (posterior) and the right ventricle.

A posterior heart attack can present as ST depression in the front chest leads V1 through V3 rather than elevation, because those leads view the back wall as a mirror image. The true ST elevation only becomes visible when additional leads (V7, V8, and V9) are placed on the patient’s back.⁹ One study of confirmed posterior heart attacks found that adding leads V5 or V6 elevation to posterior lead criteria achieved around 71% sensitivity and 89% specificity for identifying a circumflex artery blockage, and including right-sided leads helped pinpoint right coronary artery involvement.¹⁰ Without those extra leads, these patients risk having their STEMI mistaken for an NSTEMI, which can delay the emergency catheterization they need.

Right ventricular infarctions, which frequently accompany inferior wall heart attacks, are similarly invisible on a standard tracing. Right-sided chest leads, especially V4R, pick them up with high reliability. In patients presenting within 10 hours of symptom onset, ST elevation of at least 1 mm in V4R showed 93% sensitivity and 93% predictive accuracy for right ventricular involvement.¹¹ Identifying right ventricular infarction matters clinically because these patients are particularly sensitive to drops in blood volume, and treatments that are helpful for left-sided heart attacks, like nitroglycerin and diuretics, can cause dangerous drops in blood pressure.

Patterns That Act Like a STEMI but Do Not Look Like One

A handful of ECG patterns signal critical coronary artery blockages even though they do not meet the textbook definition of ST elevation. These are sometimes called STEMI equivalents, and missing them can cost time.

The de Winter pattern features tall, upsloping ST depression in the chest leads with very prominent T waves but no actual ST elevation. It indicates severe blockage of the left anterior descending artery, and patients with this pattern need emergency catheterization just as urgently as those with classic ST elevation.¹² Wellens syndrome shows deeply inverted or biphasic T waves in leads V2 and V3 during pain-free intervals, signaling a critical narrowing of the same artery that may progress to a full occlusion. Unlike the de Winter pattern, which appears during active chest pain, Wellens changes characteristically show up when the pain has temporarily resolved.

Other STEMI equivalents include isolated ST elevation in lead aVR (suggesting severe multi-vessel or left main disease) and new left bundle branch block patterns that meet specific criteria. All of these are considered high-risk presentations that warrant the same urgency as a traditional STEMI.¹³

The Left Bundle Branch Block Problem

Left bundle branch block (LBBB) is a pre-existing electrical conduction abnormality that dramatically changes the shape of the ECG. Because LBBB naturally produces ST-segment shifts that can look like ischemia, diagnosing a heart attack on top of it is one of the hardest calls in emergency medicine. For years, a new LBBB in a patient with chest pain was treated as presumptive STEMI and sent for emergency catheterization, but this led to many false activations.

The Sgarbossa criteria were developed to identify genuine heart attack patterns within LBBB tracings. A score of 3 or above has excellent specificity, around 98%, and strong positive predictive value for a true acute coronary blockage.¹⁴ A modified version of these criteria, which uses the ratio of ST elevation to the depth of the preceding S wave, improved sensitivity to about 91% while keeping specificity near 90%.¹⁵ A subsequent validation of a related algorithm confirmed sensitivity above 93% for catching a culprit lesion.¹⁶ These tools have significantly reduced both missed heart attacks and unnecessary catheterizations in LBBB patients, though the interpretation still requires clinical judgment.

What Else Can Look Like a Heart Attack on an ECG

ST elevation is not exclusive to heart attacks. Several other conditions produce similar-looking changes, and telling them apart in real time can be genuinely difficult.

Pericarditis, inflammation of the sac surrounding the heart, is the most common mimic. It produces widespread ST elevation across many leads, often with a concave-up shape rather than the convex or flat shape seen in STEMI. One of the most helpful distinguishing features is PR-segment depression, a subtle downward dip of the tracing just before the main QRS complex. PR depression in any lead had about 88% sensitivity for pericarditis, and when present in both chest and limb leads simultaneously, it had a positive predictive value of nearly 97% for pericarditis over STEMI.¹⁷ Another clue is Spodick’s sign, a downward slope of the TP segment. One study found it in 29% of pericarditis patients but only 5% of STEMI patients. However, the strongest differentiators were the presence of ST depression (overwhelmingly favoring STEMI), ST elevation greater in lead III than lead II (also favoring STEMI), and the absence of PR depression.¹⁸ Additional measures, including differences in QRS duration and QT interval spread across leads, can further improve the diagnostic accuracy when added to the classic criteria.¹⁹

Benign early repolarization is another frequent source of confusion. Some degree of ST elevation is a normal finding in the chest leads, particularly in young men. This pattern is common enough that it is considered a normal variant rather than a disease, but in a young person presenting with chest pain, it can be alarmingly difficult to distinguish from a true STEMI.²⁰ Having a prior baseline ECG on file is enormously helpful in these situations, since comparing the two tracings can reveal whether the elevation is new.

How the ECG Changes Over Hours and Days

A heart attack is not a static event, and neither is its ECG signature. The tracing follows a recognizable progression: hyperacute T waves appear first, followed by ST-segment elevation, then the development of pathologic Q waves (deep, wide initial downward deflections that indicate dead tissue), and eventually T-wave inversion as the ST segment returns toward baseline.²¹ This sequence can unfold over minutes to hours during the acute phase, with the Q waves sometimes appearing within the first hour and sometimes developing over several days.

Q waves were once considered permanent markers of completed heart attacks, but research has shown that they can regress, especially after successful early treatment. Patients whose Q waves disappeared over time showed significantly greater improvement in heart function over 24 months compared with those whose Q waves persisted and even compared with those who never developed Q waves at all.²² Q-wave regression is now understood as a sign that the treated tissue recovered rather than scarred permanently.

Continuous monitoring of the ST segment after treatment provides a real-time window into whether reperfusion is succeeding. Rapid resolution of ST elevation after catheterization is a favorable sign, while persistent elevation suggests ongoing damage. This dynamic monitoring supports clinical decisions about whether additional intervention is needed.²³

Age, Sex, and the Moving Target of “Normal”

One of the underappreciated complications in reading heart attack ECGs is that the definition of normal varies by who you are. Normal ST-segment height decreases with age and tends to be lower in women than in men, particularly in the chest leads.²⁴ This means that a 1 mm ST elevation threshold, the standard cutoff, may catch genuine heart attacks less reliably in older adults and women, whose baseline ST levels are already closer to zero. A modest rise that would be well within normal limits for a 25-year-old man could represent a significant change for a 70-year-old woman. Current guidelines have partially addressed this by using sex-specific thresholds in some leads, but the adjustment is not yet universally adopted.

Can Machines Read These Tracings Reliably?

Modern ECG machines print automated interpretations at the top of every tracing. These computerized readings are very good at ruling in STEMI when they flag it, with specificities above 95%. The problem is sensitivity: in one study of prehospital ECGs, the automated algorithm caught only about 63% of confirmed STEMIs, and its sensitivity for detecting any actionable coronary blockage dropped to roughly 37%.²⁵ In practical terms, when the machine says “acute MI,” it is almost always right. But when the machine says nothing alarming, a clinician still needs to look at the tracing themselves. Newer deep learning models trained on raw ECG signals have shown classification accuracies above 98% in research settings, but these remain largely experimental and are not yet standard in ambulances or emergency departments.²⁶

When the Problem Is the Technician, Not the Heart

Before attributing an alarming ECG pattern to a heart attack, it is worth considering a surprisingly mundane possibility: the leads were put on wrong. Swapping the left arm and left leg electrodes, for example, can make a genuine inferior wall heart attack look like a high lateral wall heart attack, potentially misdirecting treatment decisions.²⁷ Misplacements across multiple leads can generate bizarre tracings that mimic infarctions in locations that do not match the clinical picture at all.²⁸ Experienced clinicians develop a habit of checking for lead reversal whenever an ECG pattern seems inconsistent with the patient’s symptoms. A tracing that “looks wrong” sometimes is wrong, not because the heart has done something unusual, but because a sticker ended up on the wrong limb during a chaotic resuscitation.