Drivers Beneath the Numbers — What's Actually Causing Subclinical Atherosclerosis

I want you to picture two men walking into my office on the same morning. Same age. Same height. Same weight. Same BMI. If I lined them up on the bathroom scale, the number would read identical. If I dropped them both onto a standard lab panel, their LDL would come back about the same. But these two men are not the same.

One of them is muscular, fit, with most of his body fat sitting under his skin — on his arms, his legs, his backside. The other is what people now call the skinny-fat phenotype. Less muscle. More fat. And critically, that fat is sitting in a different place. It's wrapped around his liver. Around his pancreas. Pressed up against his intestines. Visceral fat.

If you put both of these men on a DEXA scan and looked at the android-to-gynoid ratio — essentially a measure of how much fat sits in the abdominal compartment versus the hips and thighs — the numbers would be dramatically different. And those numbers would tell you something the bathroom scale couldn't. Over the next thirty years, those two men are on completely different disease trajectories. Same BMI. Same weight. Different futures. That's a problem with the bathroom scale. The scale was built to tell us weight. Weight is not the metric.

The backdrop for this is the PESA study — the Spanish cohort of more than four thousand asymptomatic middle-aged adults — which showed that subclinical atherosclerosis is the rule, rather than the exception, in middle age. Sixty-three percent of those healthy-looking adults already had the disease. Sixty percent of the people who scored low risk on traditional risk scores had it too. So if risk factors are downstream — if you can have all the right numbers and still have the disease — what's actually driving the disease in the first place? Where, mechanistically, does atherosclerosis come from? And what are the levers we should be pulling that the standard lab panel and the standard risk score don't even put in front of us?

BMI is the wrong number

The body mass index is a simple ratio: your weight, divided by the square of your height. It's been the dominant measure of obesity in medicine for about half a century, mostly because it's cheap, reproducible, and computable from a tape measure and a scale. Here's the problem. BMI tells you essentially nothing about where your body fat is sitting. And in metabolic medicine, where your fat sits is more important than how much of it you have.

There are two types of fat in the body. Subcutaneous fat sits under the skin, on the arms, legs, and backside — the pattern heavy in subcutaneous fat is called gynoid, because it's more common in women. Visceral fat sits inside the abdominal wall, wrapped around the organs — the pattern heavy in visceral fat is called android, because it's more common in men. These two fats look different under a microscope, behave differently, and have very different effects on the body. Subcutaneous fat is mostly a quiet storage depot. Visceral fat is a hormonally active, pro-inflammatory endocrine organ. It secretes cytokines — interleukin-6, TNF-alpha, resistin — into the portal circulation that drains directly to the liver, and it actively contributes to insulin resistance.

The practical implication is that two people with identical BMIs can have radically different metabolic risk profiles. A man with a BMI of twenty-four and very low visceral fat is on a fundamentally different trajectory than a man with the same BMI of twenty-four who is skinny-fat. The bathroom scale treats them as identical. Their arteries are not.

So what do we measure instead? Waist circumference is the cheap one — a tape measure. The cutoffs cited in the literature are roughly forty inches in men and thirty-five inches in women; above those, risk for metabolic syndrome and cardiovascular disease begins to climb. It's not perfect, but it's an order of magnitude better than BMI. The other tool is the DEXA scan, which produces an android-to-gynoid ratio and an estimate of visceral adipose tissue volume directly. If you take one thing from this section: the metric is not your weight. It's your body composition, and specifically the distribution of your body fat. Visceral adiposity is the variable that does the metabolic damage. BMI doesn't see it.

Oxidized LDL — upstream of the obvious

So we have visceral adiposity, insulin resistance, and all the downstream consequences — high triglycerides, low HDL, high blood pressure, high glucose — which together make up metabolic syndrome. The question PESA asked is this: are visceral adiposity and insulin resistance the start of this cascade, or is there something upstream of them?

That requires a minute on a lab value most people haven't heard of: oxidized LDL, sometimes abbreviated oxLDL. Regular LDL is the cholesterol-carrying particle we've all heard about. It moves cholesterol around the body, does its job, and gets cleared. But LDL is susceptible to oxidative damage from free radicals. When that happens, the lipoprotein gets chemically modified — and oxidized LDL is a fundamentally more dangerous version of the particle. It gets recognized by immune cells in the artery wall as a damage signal; macrophages engulf it and turn into foam cells, the cellular building block of the atherosclerotic plaque. So oxidized LDL is, in many ways, the form of LDL that actually does the atherogenic work.

Holvoet and colleagues published a paper in the journal Diabetes in 2017 using nearly four thousand PESA participants without diabetes. Oxidized LDL was associated with metabolic syndrome — not surprising on its own. But when they put oxidized LDL into a multivariable model adjusted for waist circumference, BMI, and HOMA-insulin-resistance — accounting for central obesity and insulin resistance directly — oxidized LDL was still associated with metabolic syndrome. The odds ratio for being in the highest quartile of oxidized LDL versus the lowest was 2.57 after that adjustment.

Then they did a formal mediation analysis, asking how much of the relationship between waist circumference and the components of metabolic syndrome is mediated through oxidized LDL. The answer was about fourteen percent for triglycerides, and only one to three percent for HDL, blood pressure, and insulin. In other words, oxidized LDL is doing its own thing. It is not just a passive marker downstream of central obesity. The interpretation the PESA investigators offered — and I think this is right — is that oxidized LDL may be operating upstream of central obesity and insulin resistance, not downstream of them. It may be one of the early mechanisms by which atherosclerosis gets going, before the patient ever develops a paunch and before their fasting insulin starts to climb.

The practical takeaway is not "go get your oxidized LDL measured" — for most people that's not yet a routinely useful test, and there isn't a drug that specifically lowers it. The things that increase oxidative load are knowable, and they are the same things that drive atherosclerosis generally: smoking, ultra-processed food, sedentary lifestyle, poor sleep, and a high background level of inflammation. The takeaway is conceptual. The disease process starts earlier, and operates at more layers, than the standard story suggests. Waiting until your weight, your blood sugar, or your blood pressure are visibly abnormal is, by definition, late.

What you eat — and why most of us are eating wrong

PESA tracked diet. Every participant filled out a detailed food-frequency questionnaire, and the investigators used factor analysis to identify the eating patterns that actually showed up in the cohort. They found three patterns. About forty percent followed a Mediterranean pattern — high in vegetables, fruits, legumes, fish, olive oil, and whole grains, with moderate amounts of mostly poultry. About forty-one percent followed a Western pattern — high in refined carbohydrates, red and processed meats, and total fat.

The third is the one that struck me. About nineteen percent followed what the investigators named — and this is the actual term in the paper — a social-business eating pattern, characterized by high consumption of red and processed meat, pre-prepared meals, appetizers, snacks, alcoholic beverages, sugar-sweetened beverages, and frequent eating out. If you've known someone who's at a lot of business dinners and after-work drinks and doesn't cook for themselves much, that's the social-business pattern.

Here's what their arteries looked like. Compared with Mediterranean eaters, after adjusting for cardiovascular risk factors, the social-business group had a 1.3-fold higher prevalence of subclinical atherosclerosis — about thirty percent more disease. But the most striking part is how the disease was distributed. Among the social-business eaters, thirty-one percent had intermediate disease (atherosclerosis in two or three vascular territories) and thirty-seven percent had generalized disease (four or more territories). So almost seventy percent of the social-business group had multi-territorial atherosclerosis, in midlife, before symptoms.

Now, these people probably have other habits that travel with the eating pattern — and no statistical adjustment is perfect. So this isn't proof that swapping someone to a Mediterranean diet would drop their plaque burden. But it does tell us that the constellation of behaviors that wraps around how we eat in modern life is doing serious vascular work on people in their forties with no symptoms. The Mediterranean pattern, by contrast, has prospective evidence going back to Ancel Keys' Seven Countries Study in the 1950s, and it has held up through randomized trials — PREDIMED showed about a thirty-percent reduction in major cardiovascular events with a Mediterranean diet supplemented with olive oil or nuts. The diet that matters is not exotic. It is largely vegetables, fruits, legumes, fish, olive oil, and whole grains — a way of eating, not a three-week meal plan.

Sleep is doing vascular work

There is one more PESA finding I want to mention, and it's the one that gets the least attention: sleep. In 2019, Fernando Domínguez and the PESA team published a paper in the Journal of the American College of Cardiology looking at sleep duration, sleep quality, and subclinical atherosclerosis. Nearly four thousand participants wore accelerometers on their wrists for seven consecutive nights, which let the investigators measure not just self-reported sleep but objectively measured duration and fragmentation.

Compared to participants sleeping seven to eight hours a night, those sleeping less than six hours had a 1.27-fold higher odds of being in the highest tertile of atherosclerotic plaque burden, adjusted for the usual risk factors — about twenty-seven percent higher odds of having the most plaque, holding everything else constant. And sleep quality mattered separately: participants in the highest quintile of sleep fragmentation had a 1.34-fold higher odds of atherosclerosis in multiple vascular territories. So independent of how long you sleep, the quality of that sleep is doing vascular work. Coronary calcium did not differ between the sleep groups, but plaque burden on ultrasound did — consistent with the pattern that vascular ultrasound picks up disease earlier than coronary calcium.

Why does this matter for the drivers of atherosclerosis? Because sleep is not on the standard cardiovascular risk panel. Nobody is asking you about your sleep at your annual physical. It's not in any of the risk calculators. And yet, in a large prospective cohort, short sleep was independently associated with thirty percent more plaque burden, and disrupted sleep with thirty-four percent more multi-territorial disease. Sleep is when blood pressure dips, when inflammation gets cleaned up, when the brain's glymphatic system clears metabolic waste — and, evidently, when the arteries get a break from the day's metabolic load. If you sleep less than six hours, or your sleep is disrupted — by stress, by undiagnosed sleep apnea, by a phone in your bedroom — you are paying a vascular cost that does not show up on any standard lab.

Key takeaways

• BMI is the wrong number. The metric you should care about is body composition, and specifically visceral adiposity. Waist circumference is cheap; a DEXA scan is better. The bathroom scale doesn't tell you what you need to know.
• The drivers are earlier and more diffuse than the standard panel suggests. Oxidized LDL, in the PESA data, is doing damage independently of central obesity and insulin resistance. Waiting until weight, glucose, or blood pressure are visibly abnormal is waiting too long.
• Lifestyle is doing vascular work that no single lab shows. The diet-pattern and sleep findings are large-cohort, hard-endpoint results — a ~30% difference in subclinical disease in your forties. Lifestyle is the dose, delivered every day whether you track it or not.