The Thunderbolt Is Not the Storm

Aug 28

12 min read

Updated: Aug 31

A cardiovascular endpoint trial named after the god of thunder dampens the concept of 'inflammation' as a principal cause of heart disease. Homer, of all people, explains why.

Ask the Evidence — explore this article with your AI assistant:

https://mhaat.vercel.app/ai/n-the-thunderbolt-is-not-the-storm-ai.md

For busy people, or to tune in when on the move, a Google NotebookLM audio podcast is available as a story beneath.

WHAT THE TRIAL DID

Somewhere near the start of Christopher Nolan's Odyssey — the film that has taken well over a billion dollars since July and is still filling IMAX screens into the autumn — a voice remarks that the old laws of Zeus have been smashed to pieces. It is a good line for a trailer. It is also, for anyone who spent the summer reading cardiology trial results, an unexpectedly precise description of where we find ourselves.

Because a few days after the film opened, a trial called ZEUS reported. And the laws it was built on turned out to be in rather worse repair than we had assumed.

ZEUS gave a monthly injection of ziltivekimab, an antibody that blocks interleukin-6, to 6,376 people with established atherosclerotic cardiovascular disease, chronic kidney disease and measurably raised inflammation. These were not marginal candidates. They were exactly the patients in whom the inflammatory hypothesis of atherosclerosis predicts the largest benefit — high risk, high residual inflammation, already on everything else we know how to give.

The drug did what it was designed to do. Free IL-6 fell. High-sensitivity CRP fell substantially. Target engagement was not in question at any point.

Cardiovascular death, non-fatal myocardial infarction, non-fatal stroke. Hazard ratio 0.99, confidence interval 0.88 to 1.11.

Nothing. Not a small benefit, not a trend, not a subgroup worth arguing about. The inflammatory signal was suppressed and the patients were no better off.

It is worth pausing on how strange that is. We had a coherent mechanism, a validated biomarker, a drug that hit its target, and a population selected for exactly the abnormality being treated. If the causal story were right, this should have worked.

And ZEUS was not alone. The year before, CLEAR-SYNERGY gave low-dose colchicine to 7,062 patients after myocardial infarction. CRP fell. Events occurred in 9.1 per cent of the treated group and 9.3 per cent of the placebo group — a hazard ratio, once again, of 0.99.

Two large, careful, well-conducted trials. Two convincing reductions in inflammatory signalling. Two flat lines.

WHAT ZEUS ACTUALLY DOES

Here the coincidence of the name becomes useful, though I should say plainly that it is a coincidence. The trial was named by its sponsor and not, so far as I know, by a classicist making a point. But the accident is instructive anyway, because most of us carry a cartoon of Zeus that is precisely wrong in the way our thinking about inflammation has been wrong.

The cartoon is Zeus as prime mover: the god who decides, who causes, who sets events in motion from the top of the mountain. Read the poem and he is nothing of the sort. Zeus does not author fate. He ensures it is carried out. He is the enforcer of the laws of hospitality, the arbiter between other gods' competing claims, the one who arrives after the transgression to see that consequence follows cause.

He is not the origin of the story. He is the mechanism by which the story reaches its conclusion.

Which is, I have come to think, a fair description of what interleukin-6 is doing in a coronary artery.

THE CATTLE OF THE SUN

There is one episode that makes the point better than any amount of argument.

Odysseus's crew, becalmed on the island of Thrinacia and slowly starving, are warned not to touch the cattle of Helios. Their captain falls asleep. They slaughter the cattle. And when they finally put to sea, Zeus destroys the ship with a thunderbolt and every man aboard drowns.

Now ask the question a trialist would ask. Suppose you could have disarmed Zeus. Suppose you had an antibody against thunderbolts, and it worked beautifully, and no lightning struck the ship at all. Would the crew have lived?

They were still starving. Still becalmed. Still a thousand miles from Ithaca with a captain who had already lost most of them. They had still eaten the cattle. The thunderbolt was the consequence of a situation that the thunderbolt did not create, and removing it changes the manner of the ending rather than the ending itself.

That is what I think we did in ZEUS, and in CLEAR-SYNERGY. We silenced the thunderbolt. The apolipoprotein B particles carried on entering the arterial wall. The blood pressure carried on. The insulin resistance, the visceral fat, the lipoprotein(a) that nobody was treating — all of it carried on, entirely undisturbed by the fact that we had made the sky quieter.

THE PART THAT SHOULD STOP US SHORT

There is a detail in the ZEUS report that deserves more attention than it will get, because it does not fit the simple story either.

Patients given ziltivekimab had more serious infections than those given placebo. This is not a surprise — it is a known consequence of blocking the IL-6 axis, and it appeared in CANTOS too. But it is worth saying what it means.

Zeus in the poem is not merely a punisher. He is the guarantor of order. His laws are what make the world habitable: the obligation to feed a stranger, to protect a guest, to honour a suppliant. When those laws lapse, Ithaca does not become peaceful. It fills with suitors eating another man's house down to the walls.

Interleukin-6 is likewise not a rogue agent. It is a load-bearing part of the immune system's response to injury and infection, and when you take it out of the wall, something else in the building starts to sag. The signal we were so keen to abolish was doing work.

WHERE THE METAPHOR BREAKS — AND IT DOES

I want to be careful here, because there is a version of this essay that is too tidy, and it would be wrong.

CANTOS worked. Canakinumab, blocking interleukin-1β in patients with previous myocardial infarction and raised CRP, reduced cardiovascular events without touching LDL cholesterol. That is a real result from a large randomised trial and it does not go away because two later trials were negative. Colchicine reduced events in COLCOT and in LoDoCo2. Patients with rheumatoid arthritis and psoriasis carry excess cardiovascular risk that no purely downstream account of inflammation explains comfortably.

Homer, to his credit, is less tidy than my argument too. Zeus does intervene decisively, and when he does it settles things. He sends Hermes to release Odysseus from Calypso's island, and seven years of stasis end. And in the poem's last book, with Ithaca on the edge of civil war between Odysseus and the families of the men he has slaughtered, a thunderbolt falls and Athena imposes a truce. The poem's final act is a peace enforced from above. Divine intervention at the top of the causal chain is not always ornamental.

Nolan cuts that book altogether. There is no orchard, no truce, no thunderbolt at the end of the film — the god who has been invoked throughout never arrives to settle the account. Odysseus takes back his hall, hands the kingdom to his son and sails west into exile as penance for what he did to get it. The consequences are not resolved. They are simply carried.

Which is, if you have spent this summer reading trial reports, an oddly familiar shape. We keep waiting for the intervention at the top of the chain that will settle the whole thing at once, and it keeps not coming.

So the position that survives contact with the evidence is narrower than the one I would prefer. It is not that inflammation does not matter. It is that in ordinary atherosclerosis, in the ordinary patient with a statin in the cupboard and a waist measurement nobody has taken, inflammation is largely the response to an injury that is still happening — and treating the response while the injury continues produces exactly the results we have just seen.

WHAT THE ARTERY IS ACTUALLY BEING ASKED TO ENDURE

If inflammation is the enforcement, what is the transgression? Our terminology can easily become confused here, so it is worth being precise before going further.

ApoB, LDL-C and the particle itself

Go back, for a moment, to the cattle of Helios. What made the crew's transgression a transgression was not the total weight of meat they ended up carrying. It was the number of sacred animals they had slaughtered — a herd Helios could count, one beast at a time, regardless of how large or small each one happened to be. That same distinction, oddly enough, is exactly where the ApoB argument lives.

The evidence here is not ambiguous, which is refreshing. Atherosclerosis is caused by the entry and retention of ApoB-containing lipoprotein particles — principally LDL — within the artery wall. The cholesterol and other lipids carried by those retained particles contribute to plaque formation.

LDL-C measures how much cholesterol the LDL particles happen to be carrying. ApoB tells us, more directly, roughly how many atherogenic particles are in circulation. Because particle number can vary substantially for a given LDL-C, ApoB can expose a burden that cholesterol mass alone conceals.

ApoB is not simply an inert counting label. It is the structural protein carried on these particles, and it takes part in their retention within the arterial wall. But LDL-C is not itself a particle, or a circulating toxin — it is a laboratory measurement of how much cholesterol is being transported within LDL. ApoB, in effect, counts the herd. LDL-C weighs the meat.

Once retained, these particles and their cargo are modified: lipids oxidise, macrophages are recruited and ingest what has been modified, and cholesterol accumulates in foam cells and eventually within the plaque and its necrotic core. The inflammatory response is not a separate event bolted on afterwards. It is inseparable from that process.

Not either / or — It is not that ApoB causes atherosclerosis while cholesterol is innocent. Nor is it that cholesterol is the toxin and ApoB merely counts it.ApoB-containing particles deliver and retain atherogenic lipid within the arterial wall; ApoB concentration gives us an unusually useful way of estimating how many of those particles are circulating.

Particle number matters more than particle size

This also corrects a persistent misunderstanding. Insulin resistance and metabolic dysfunction are frequently associated with small, dense LDL, and it is tempting to conclude that these particles are dangerous simply because they are small. The more important issue appears to be how many particles are present.

Every LDL particle carries exactly one ApoB molecule, whatever its size. Smaller, more cholesterol-depleted particles carry less cholesterol each, so a patient may need more of them in circulation to transport the same total amount of cholesterol. Two patients can therefore share an identical LDL-C while carrying substantially different numbers of LDL particles — the discordance ApoB is built to expose. A herd of large cattle and a herd of many small ones can yield the same tonnage of meat. Helios never cared about tonnage. He counted the herd.

Small dense LDL remains useful, but chiefly as a fingerprint of disturbed lipoprotein metabolism — particularly insulin resistance and raised triglycerides — rather than as a hazard in its own right. Particle number and cumulative years of exposure appear to matter more than particle size. That distinction matters considerably once visceral fat enters the discussion.

Blood pressure injures the vessel mechanically. Lipoprotein(a) — another population of ApoB-containing particles, with additional biology of its own — adds a burden that most patients have never had measured once. Insulin resistance and visceral adiposity supply a metabolic environment that can amplify several of these pathways at once: more hepatic VLDL and remnant particles, more dysglycaemia, more hypertension.

It would be too simple to say that visceral fat directly causes coronary plaque. A more defensible model runs through several steps:

1. Excess visceral adipose tissue

2. Insulin resistance and altered metabolic flux

3. Increased VLDL / remnant / ApoB particle burden, dysglycaemia, and hypertension

4. Greater arterial injury and ApoB-particle exposure

5. Retention of atherogenic particles and lipid within the arterial wall

6. Immune and inflammatory response

7. Atherosclerotic plaque

Which is why I think of visceral fat not as an alternative explanation to LDL or ApoB, but as a potentially important upstream amplifier of several already-established cardiovascular risk pathways.

And then there is tobacco, which is not a fifth item on that list but something more corrosive: the one exposure that degrades all four at once. It lowers HDL and shifts LDL towards a more readily oxidised, more readily retained form. It raises blood pressure and stiffens the arterial wall. It induces insulin resistance and raises the risk of type 2 diabetes. And it redistributes fat inward, so that smokers are frequently lighter than non-smokers while carrying proportionally more of the visceral depot that matters — which is one of the crueller tricks in cardiovascular medicine, because the scales appear to reward it. On top of all that, it injures the endothelium directly, without needing any of the four pillars as an intermediary.

In Britain and the United States this feels like a settled argument, which is a luxury of geography. Globally it is nothing of the kind. Around 1.2 billion adults still use tobacco — roughly one in five — and just under a billion of them are men. Prevalence among men in South-East Asia has almost halved since 2000 and still sits near 37 per cent. In Africa, prevalence is the lowest of any region and the absolute number of users is nonetheless climbing, because the population is growing faster than the habit is falling. Europe now has the highest prevalence in the world.

These are the cattle of the sun. They are being slaughtered daily, in clinics, by people who have been told their cholesterol is fine because a number on a page was under five — and, in much of the world, by people for whom the single most effective cardiovascular intervention available remains giving up cigarettes.

The artery is not inflamed because inflammation happened to it. It is inflamed because something is still being done to it.

There is a further wrinkle, and it is the reason the story is not finished. Coronary CT can now show us inflammation in the fat immediately surrounding a coronary artery, and that measurement carries real prognostic weight — in the ORFAN cohort of just over forty thousand patients, those without obstructive disease made up eighty-one per cent of the group and accounted for nearly two-thirds of the cardiac deaths. Something is going on in those arteries that a stenosis report does not capture and a blood test does not either.

What that measurement does when you remove the upstream injury, rather than when you suppress the inflammation directly, is a question I am in the middle of trying to answer with our own patients. I will write about it when I know, and not before.

THE INSTRUCTION

The therapeutic conclusion is unglamorous and completely unsurprising, which is probably why it keeps being passed over in favour of a new injectable.

Find what is injuring the artery. Measure it properly — particle number rather than cholesterol mass, blood pressure over days rather than in a corridor, lipoprotein(a) once in a lifetime, and a real look at the metabolic environment rather than a BMI. Then remove as much of it as can be removed, and see whether the inflammation follows.

Not: silence the thunderbolt and hope the sea calms.

Put like that it sounds vague, so let me be concrete. Four things determine most of what the coronary artery is being asked to endure, and all four can be measured and moved.

1. ApoB — how many atherogenic particles are arriving, not how much cholesterol they happen to be carrying.

2. Blood pressure — measured at home, over days, not once in a corridor with your coat on.

3. Glucose and insulin — fasting insulin and HbA1c, because insulin resistance arrives years before the glucose does.

4. Visceral fat — the depot that drives the other three, and the one a BMI cannot see.

And tobacco, which is not a fifth pillar but a solvent for all four. It worsens the particle profile, the pressure, the insulin resistance and the visceral depot simultaneously, and injures the artery directly besides. If it applies to you, nothing else on this list comes close.

These are the scaffold of the VAT Trap Blueprint — six panels covering what to measure, what insulin is doing, movement, the plate, protein, and what the options actually are. It is free, it is not selling anything, and it is where I would start.

Odysseus does get home. Not because Zeus is disarmed, and not because the gods are placated with a clever intervention, but because a man with no ship and no crew keeps moving in roughly the right direction for twenty years. And in Nolan's telling, getting home does not settle it either. He wins his hall back, gives the kingdom away and puts to sea again.

No thunderbolt arrives to make it right. What there is instead is the long, unglamorous business of continuing to row. There is a moral in there about cardiovascular prevention, and I suspect we already know what it is.

Summary

SUMMARY

A major cardiology trial called ZEUS tested a monthly injection that successfully reduced inflammation markers in high-risk heart patients. The drug worked exactly as designed. Cardiovascular deaths, heart attacks, and strokes were completely unchanged. The inflammatory hypothesis just got much harder to defend.

VAT Trap