Visceral fat vs BMI: which better predicts heart disease?

Visceral fat vs BMI is not a close contest in these studies: the measures that capture where fat sits predicted heart problems more consistently than BMI did. Visceral fat is the fat packed deep in the abdomen around the organs, rather than the softer subcutaneous fat that sits just under the skin, and body mass index, or BMI, is only your weight compared with your height - it carries no information about where that weight is.

The largest comparison pooled five international trials in adults with heart failure, the condition where the heart cannot pump or fill well enough to keep up with the body. BMI was recorded for 21,479 participants and waist measurements for 7,827 [13]. Waist to height ratio, your waist measurement divided by your height, rose in a straight line with heart failure events and deaths: the higher it went, the higher the risk. BMI did not. Its relationship with outcomes bent back on itself, so both low and high values looked bad [13].

A normal BMI turned out to be weak reassurance. Among participants in those trials whose BMI was under 30, the usual threshold for obesity, 89% still carried excess abdominal fat, most often the older ones and the women [13].

In UK Biobank, 17,232 adults who had prediabetes or diabetes and a normal BMI were followed for 12 years. Those whose waist was large relative to their hips were about 42% more likely to die of any cause, 54% more likely to die of cardiovascular causes and 43% more likely to have a heart attack than those without that pattern [14].

A second large study measured arterial stiffness, how rigid the big arteries have become, in 29,048 people with type 2 diabetes. Carrying less fat around the organs went with being about 15% less likely to have stiff arteries, while a lower BMI showed no clear effect in either direction [6]. Habits shifted that picture. Measured against the people with the least organ fat, those carrying more were about 89% more likely to have stiff arteries if they met one or none of seven heart health targets, and about 61% less likely if they met five or more [6].

An older, cheaper approach tried to infer how organ fat was behaving from a blood test and a tape measure. The visceral adiposity index combines waist size, BMI and two blood fats, triglycerides and HDL cholesterol. Tested in 1,498 primary care patients, a higher index went with being roughly two and a half times as likely to have had a cardiovascular event and about 63% more likely to have had a stroke or similar event in the brain's blood vessels [4]. Waist size and BMI on their own showed no relationship with insulin sensitivity, how well the body responds to the hormone that clears sugar from the blood, while the combined index did [4].

That mismatch is part of what feeds the obesity paradox, the odd finding that heavier patients sometimes do better than thinner ones: a BMI band gathers together people with very different biology and blood chemistry [5], [6]. Organ fat is also movable. In a 40 week trial of 185 adults with overweight or obesity at high cardiovascular risk, the group given daily liraglutide alongside a reduced calorie diet lost 12.5% of their visceral fat on scans, against 1.6% in the placebo group [2]. Liraglutide is a prescription medicine, and that trial measured fat, not heart attacks.

Fat around the organs is active tissue. As it expands under a long run of excess calories, it recruits macrophages, immune cells that normally clear debris, which switch into an inflammatory state and release cytokines, signalling proteins that interfere directly with insulin's message inside cells [8]. The result is insulin resistance: the hormone is there, but tissues respond to it poorly.

That resistance then shows up in several organs at once. Fat tissue releases more free fatty acids, which makes the inflammation worse, because those fatty acids can switch on the same immune receptors [8]. The liver produces more glucose, so blood sugar climbs. In the blood vessels and the kidneys, vessels narrow and sodium is retained, and blood pressure goes up [8].

Location matters because fat depots in different parts of the body contain different cell populations and release different signalling molecules [10]. Abdominal fat drains towards the liver and sits alongside ectopic fat, fat deposited inside organs such as the liver and heart where it does not belong. That combination is central to metabolic syndrome, the cluster of abdominal obesity, high blood sugar, abnormal blood fats and high blood pressure that raises the risk of heart disease and type 2 diabetes [7]. A BMI calculation cannot see any of it, because weight and height are all it uses.

Some of the damage appears to run through ageing itself. Across international cohorts, obesity measures were linked to cardiovascular disease partly through faster biological ageing, estimated from blood markers that describe how old the body looks next to its calendar age, and that pathway accounted for between 10% and 25% of the link [3]. In men it was mainly the body shape measures that ran through it; in women it was the visceral fat and metabolic measures [3].

Nearly all of this evidence comes from people who were already unwell, not from healthy adults:

• the five pooled trials enrolled adults with heart failure whose pumping fraction was mildly reduced or preserved, and who were willing to join a drug trial [13]
• the arterial stiffness study covered only people with type 2 diabetes, and it measured everyone at a single moment, so it cannot show that organ fat caused the stiffening [6]
• the 12 year cohort was restricted to adults who already had prediabetes or diabetes [14]
• the liraglutide trial was 92% female, ran at one US medical centre, and was funded by the drug's manufacturer [2]

The visceral adiposity index was built backwards, on records already collected, and its formula was modelled in one Italian population [4]. Separate versions exist for Chinese cohorts [1], [3], so a cut off from one population does not transfer to another.

Waist to height ratio is not neutral across groups either. At the same BMI, women in the pooled trials had higher waist to height ratios than men, and Asian and Black participants had higher ratios at lower BMI [13].

Organ fat and metabolic health do not simply add up. In a 9 year Chinese cohort of 8,358 adults, the highest cardiovascular risk sat in the group with high visceral adiposity but a low insulin resistance score, rather than in the group high on both [1].

And none of these studies tested whether measuring waists instead of BMI, or shrinking organ fat, prevents heart attacks. The liraglutide trial changed the fat and stopped there [2].

Two adults with the same BMI of 27 can sit in very different places in this research, and the difference shows up at the waist rather than on the scale. In the pooled heart failure trials, 89% of participants with a BMI below 30 still had excess abdominal fat, and higher waist to height ratios went with more heart events and deaths [13]. The measurements those studies used were ordinary: a tape around the waist, and that number divided by height. Seeing the organ fat itself takes a scan, which is how the visceral fat area and the liraglutide trials measured it [2], [6]. What any of those numbers means for one person, and whether anything should change because of them, is a question for a doctor rather than a calculator.