At ESC, a new paper on coronary artery calcium scanning came out. Almost immediately, I saw cardiologists on social media saying some version of this:
“Why get a CAC? It won’t change management.”
Think about what that means. A test that directly measures atherosclerotic plaque in the coronary arteries is dismissed because “it won’t change management.” In other words, the treatment decision has already been made.
I think the reality is simpler: CAC can create uncomfortable situations.
- A 45-year-old with high LDL but a CAC score of 0.
- A 60-year-old with “normal labs” but a CAC score of 400.
Now the conversation becomes harder. Guidelines become less convenient. And the doctor has to explain risk.
The same risk category. Very different risks.
Ironically, the new paper demonstrates exactly why CAC can matter.
Among people considered borderline risk by PREVENT, the observed 10-year cardiovascular event rates were:
| CAC score | 10-year event rate |
|---|---|
| 0 | 1.9% |
| 1–99 | 3.9% |
| 100–299 | 7.4% |
| ≥300 | 14.3% |
These people were in the same general risk category. Yet depending on how much coronary plaque was actually present, observed event rates ranged from 1.9% to 14.3%.
That seems pretty relevant to me.
CAC doesn’t always fit neatly into a predetermined pharmaceutical pathway. But it answers a very important question: Do you actually have plaque?
That question becomes even more interesting when you look at two other major studies presented around ESC: STAREE and SAGA/SITE.
Statins work. Who benefits enough?
Statins work. They reduce cardiovascular events.
They also seem to harden and stabilize plaque, increase plaque calcification and, in some studies, modestly shrink plaque volume.
But they also have real side effects. They cause insulin resistance and diabetes, particularly in women. They deplete vitamin K2 and CoQ10 and are very likely to cause muscle aches.
The two studies from ESC make me question how much the known positive effects translate into the outcomes patients ultimately care about, particularly in older adults being treated for primary prevention.
Who will benefit, and who will be harmed?
In my own career, I have seen severe statin reactions requiring hospitalization, including rhabdomyolysis with kidney failure requiring acute dialysis and severe thrombocytopenia, a low platelet count. These experiences have included near-fatal cases.
STAREE: events and disability-free survival
STAREE randomized 9,971 adults aged 70 or older without cardiovascular disease, diabetes or dementia to atorvastatin 40 mg or placebo.
After approximately six years, atorvastatin reduced major cardiovascular events by about 30% in relative terms.
So again, statins work.
But in absolute terms, we’re talking about a couple fewer events per 100 people treated. And there was no significant improvement in disability-free survival, an outcome encompassing death, dementia or persistent physical disability.
SAGA/SITE: what happened when older adults stopped statins?
SAGA/SITE looked at essentially the opposite question. It randomized 1,160 adults aged 75 or older who were already taking statins for primary prevention to stop or continue treatment.
Stopping statins caused LDL to jump dramatically. Yet after three years, the results were:
| Outcome | Stop | Continue |
|---|---|---|
| Death | 7.2% | 7.9% |
| Major CV events | 5.0% | 4.4% |
Stopping statins was noninferior for all-cause mortality.
This doesn’t mean statins don’t work. They clearly reduce cardiovascular events. But it does make me question the magnitude of the real-world benefit in some patients.
Cognition: a question worth studying
STAREE found no significant reduction in dementia despite substantially lowering LDL and reducing cardiovascular events.
We also have a small prospective withdrawal/rechallenge study in patients with Alzheimer’s disease. MMSE scores improved by 1.9 points after statins were stopped, then worsened by 1.9 points when statins were restarted.
Small study. Open label. Definitely not proof.
But withdrawal, improvement, rechallenge and worsening is a signal worth studying.
Patient selection matters
So no, I’m not saying statins don’t work. They work.
I’m asking a different question: Who actually benefits enough to justify treatment?
If we’re treating 100 people for years to prevent a couple of cardiovascular events, without clearly improving mortality, dementia or disability in these older primary-prevention populations, then patient selection matters. A lot.
Which brings me back to CAC.
Why wouldn’t I want to know whether my patient actually has coronary atherosclerosis? And why wouldn’t I want more options once I know?
Why every patient may not need the same pathway
This is also why our recent ezetimibe paper is important.
We looked at patients who developed marked LDL elevations while following carbohydrate-restricted ketogenic diets.
Among the 12 patients treated with ezetimibe without a statin, median LDL fell 50.4%, an absolute median reduction of 137 mg/dL.
Ezetimibe is generic. It’s inexpensive. It works through intestinal cholesterol absorption rather than inhibiting cholesterol synthesis.
Our study was small and retrospective. It does not prove cardiovascular benefit in this population.
But it raises another question worth studying: Why are we acting as though every patient with an elevated LDL needs to travel down exactly the same therapeutic pathway?
Start with the disease. Understand the patient.
- Measure the disease.
- Understand the patient’s absolute risk.
- Understand why their LDL may be elevated.
- Discuss the magnitude of benefit.
- Discuss potential harms.
- Choose the intervention that makes sense for that patient.
That seems like medicine to me.
This article is for education and discussion. Do not stop or change a prescribed medication without discussing it with your clinician.
