The Widowmaker: An Interactive 3D Guide to the Heart

The left anterior descending artery is about the width of a drinking straw. It leaves the left main coronary artery near the top of the heart and runs down the front of it towards the apex, and along the way it feeds the front wall of the left ventricle, the septum between the two ventricles and usually the tip of the heart itself — somewhere around half of the muscle that does the actual pumping.

Block that vessel close to its origin and you lose that entire territory at once. That is what the slang term widowmaker refers to. It is not a medical diagnosis — clinicians call it a proximal LAD occlusion, or an anterior STEMI — but the nickname captures something real about the size of the loss and the speed of it.

What makes it genuinely worth understanding is not the anatomy, though. It is the silence. The plaque that eventually causes this event has usually been growing for three or four decades, and for almost all of that time the person carrying it feels entirely well — often right up to the morning it kills them. The model below is built to make that silence visible.

How to read the model

Drag to orbit the heart, scroll to zoom, and click any label to highlight the structure it names. The controls fall into three groups:

• •Rhythm — switch between sinus rhythm and atrial fibrillation, and set the heart rate. The gold glow is the electrical wave: it starts at the SA node, sweeps the atria, pauses at the AV node while the ventricles fill, then fires both ventricles at once. The strip below draws a live Lead II ECG from the same clock.
• •The plaque — the stenosis slider narrows the proximal LAD, and the inset shows the artery in cross-section. The age slider does the same thing on a biography: it walks a typical man from fifteen to sixty-six, growing the plaque the way autopsy and imaging studies say it actually grows.
• •Load and rupture — the exertion toggle raises demand to roughly two and a half times resting, which is what unmasks a narrowing. The rupture button does what a stress test cannot predict: it tears the cap and thromboses the vessel.

Two gauges track the consequences: anterior wall perfusion, the blood supply reaching the LAD's territory as a fraction of what it needs, and left atrial appendage stasis, which only matters once you switch to atrial fibrillation. The x-ray toggle makes the walls translucent, and the cutaway slider slices the heart open along whichever axis you are viewing it from — useful for seeing the thickness of the left ventricular wall against the right.

Why this artery, and not the others

Three vessels do nearly all the work of feeding the heart. The right coronary artery runs in the groove on the right side and typically supplies the inferior wall plus, in most people, both conduction nodes. The circumflex curves around the left side to the lateral wall. The LAD takes the front.

The asymmetry matters twice over. First, the LAD's territory is the largest — the anterior wall, the septum and the apex. Second, the further upstream the blockage, the more of that territory goes dark. A blockage in a small diagonal branch costs a patch of muscle; a blockage in the first centimetres of the LAD costs everything downstream of it.

Turn on x-ray walls and look at the wall thickness. The left ventricle is three to four times thicker than the right, because it pumps against systemic pressure rather than into the lungs. That thickness is also a vulnerability: thick muscle has a high oxygen demand and is compressed by its own contraction, so the coronary arteries only really fill during diastole, between beats. A fast heart rate shortens diastole — which is one reason ischaemia so often shows up during exertion.

Four decades in one slider

Drag the age slider slowly from fifteen to sixty-six and read the notes as they change. The sequence is the part most people have never seen:

• •Teens. Fatty streaks — cholesterol-laden foam cells sitting under the artery lining. Autopsy work on young accident and combat casualties, most systematically the PDAY study, found these in the great majority of young men. Atherosclerosis does not begin in middle age; it begins before anyone thinks about it.
• •Twenties. Thickening concentrates at branch points, especially where the left main splits into the LAD and circumflex. Blood flow there is disturbed and slow, and low-shear zones are where LDL particles accumulate. The widowmaker's location is not random — it is fluid dynamics.
• •Thirties. A true fibroatheroma forms: a soft lipid core with a fibrous cap over it. This is now a structure that can fail mechanically, and it is completely silent.
• •Forties. Glagov remodelling — described in 1987 and still one of the most important findings in the field. As plaque accumulates, the artery initially expands outward, so the channel through the middle stays open even as the wall thickens substantially. The disease progresses while the thing we measure, the lumen, barely changes.
• •Fifties and sixties. Outward expansion runs out and the lumen finally starts to narrow. Around 40 to 60 percent the first symptoms become possible on exertion — but many people still have none. The median age of a first heart attack in men is in the mid-sixties.

Watch the perfusion gauge through all of this. It stays at 100 percent almost the whole way.

Coronary flow reserve: the mechanism of the silence

The reason nothing happens for forty years is a safety margin called coronary flow reserve. The small vessels downstream of a coronary artery can dilate, increasing flow to something like three to four times resting levels when the heart needs it. As a plaque narrows the artery upstream, those vessels dilate to compensate, and resting flow is held completely normal — the reserve is being spent, invisibly.

You can see both thresholds in the model. Leave exertion off and drag the stenosis slider up: perfusion holds at 100 percent until you are past roughly 85 percent narrowing. Now turn exertion on and sweep it again — the gauge starts falling around 60 percent, and the ischaemia banner appears. Same artery, same plaque; the only difference is how much the muscle is asking for.

The counterintuitive part: the dangerous plaque is usually not the tightest one

If heart attacks were caused by arteries gradually closing, they would be preceded by months of worsening angina, and the tightest lesion would be the one to fear. Angiographic studies that compared images taken before an infarction with the culprit vessel afterwards found the opposite: most infarctions arose from lesions that had been less than 70 percent narrowed, and a large share from lesions under 50 percent.

The reason is that a heart attack is not a plumbing failure, it is a structural one. A plaque with a large soft lipid core, a thin inflamed fibrous cap and few smooth muscle cells holding it together can tear open regardless of how much room it leaves in the artery. When it does, the contents — intensely thrombogenic — meet flowing blood, and a clot forms across the vessel in minutes. A stable, calcified, heavily narrowed plaque may cause predictable angina for years and never do this.

So the widowmaker is rarely the story of a blockage nobody caught in time. It is more often the story of a moderate plaque that nobody would have treated with a stent, failing suddenly. Set the model to around 50 or 60 percent, confirm there are no symptoms at rest, and then press ⚡ Rupture plaque.

What rupture actually does

Four things change at once in the model, and all four have clinical counterparts.

• 1.The lumen goes to zero. The cross-section fills with thrombus. Perfusion to the anterior wall drops to nothing — not reduced, absent.
• 2.The muscle changes colour and stops moving properly. The wedge downstream turns grey and its contraction fails. Ischaemic muscle stops contributing to the pump within seconds — long before any cell actually dies.
• 3.The ECG changes. The ST segment lifts off the baseline. That elevation is the definition of a STEMI and the reason it is a "call the cath lab now" diagnosis rather than a "admit and observe" one. Compare it with what demand ischaemia looks like earlier in the model, where the ST segment sags below baseline instead.
• 4.The rhythm becomes unstable. The model flags this rather than simulating it: ischaemic muscle is electrically irritable, and ventricular fibrillation in the first minutes is the main reason a large anterior infarct kills people before they reach hospital.

Everything about emergency treatment follows from the arithmetic of point one. Muscle starved of blood is progressively lost over a few hours, so guidelines target opening the artery — usually by primary angioplasty — within ninety minutes of arrival at a capable hospital. Time is muscle, and muscle does not come back.

The other story in the model: atrial fibrillation

Press the atrial fibrillation button and watch the ECG. The P wave — the small bump before each spike that represents the atria contracting — disappears, replaced by a fine irregular wobble, and the beats stop being evenly spaced. That is the whole diagnosis: no organised atrial activity, irregularly irregular ventricles.

Mechanically, the atria are quivering at around 400 impulses a minute instead of contracting. The AV node rations those impulses through to the ventricles, which is why the pulse is chaotic but not lethally fast. The lost "atrial kick" — the final top-up of ventricular filling — costs something like 10 to 20 percent of cardiac output, which is why AFib often feels like breathlessness and fatigue rather than palpitations.

But the danger is the gauge climbing on the left. Blood that is not being squeezed through the atrium stagnates in the left atrial appendage, a narrow blind pouch off the left atrium — over 90 percent of clots in non-valvular AFib form there. Leave the model in AFib for a few seconds and a clot appears in the appendage. That is the mechanism behind AFib's roughly fivefold increase in stroke risk, and the reason anticoagulation matters more than whether the rhythm itself is restored.

What actually changes the odds

If the disease builds silently for forty years and the event is a structural failure rather than a gradual closure, then the useful interventions are the ones that act on the decades, not on the moment. Three follow directly from the model.

Lower the cumulative particle exposure. Plaque grows because ApoB-containing lipoproteins — LDL and its relatives — enter the artery wall and are retained there. The driver is exposure over time: concentration multiplied by years. This is why lipid-lowering started earlier does far more than the same treatment started late, and why ApoB — which counts particles rather than measuring the cholesterol they carry — is a more direct read on the process than LDL-C alone.

Find out whether you actually have plaque. Because the lumen stays open for decades, the question "is there disease" is answered by imaging the wall, not by testing flow. A coronary calcium score or CT angiography can show plaque that no stress test would detect. Whether these are appropriate for a particular person is a conversation with a clinician, not a decision to make from an article.

Stabilise what is already there. Blood pressure, smoking, glycaemic control and inflammation all influence whether a plaque stays quiet or develops the thin, inflamed cap that fails. hs-CRP is one window onto that residual inflammatory risk. Not smoking remains the single largest modifiable factor for plaque rupture specifically.

Frequently asked questions

The honest limits of this model

The anatomy is real — the heart geometry comes from BodyParts3D, published by the Database Center for Life Science and used here under CC-BY-SA 2.1 Japan — but everything layered on top of it is simplified deliberately. Perfusion is a smooth curve chosen to reproduce the two thresholds that matter (roughly 85 percent at rest, roughly 60 percent on exertion), not a computational fluid dynamics simulation. The ECG is synthesised from waveform components rather than derived from a cardiac electrical model. Collateral circulation, which genuinely protects some people with severe disease, is absent entirely.

The age timeline describes a population average for men and is not a prediction about any individual — plenty of sixty-six-year-olds have minimal plaque, and some forty-year-olds have a great deal. Use the model to understand why coronary disease behaves the way it does. Do not use it to estimate your own risk.