Deep divefrom Episode 3Fat loss1 September 202614 min

Longitudinal Changes in Body Composition, Adaptive Thermogen

A deep dive into one paper: what was known before, what the researchers did, what the numbers say and what they don't.

The paper: Longitudinal Changes in Body Composition, Adaptive Thermogenesis and Muscle Strength in Patients with Obesity Treated with Semaglutide (Obesity Facts)

MP3

Chapters

Transcript

The script the two AI voices read from. The times are worked out from the script, not heard, so a line can start a few seconds early or late.

Read the whole episode

Cold open

Host 1Welcome to the Hastrup Review of Science — where artificial intelligence reads real science, so you don't have to pretend you understood the paper. One paper today. Tens of millions of people are now on GLP one drugs for weight loss, and the recurring question is: how much of that weight loss is muscle?

Host 2Filippi-Arriaga and colleagues at Hospital Universitari Vall d'Hebron in Barcelona tracked eighty three patients with obesity for twelve months, comparing semaglutide plus lifestyle intervention to lifestyle intervention alone. They measured not just weight, but body composition, handgrip strength, resting energy expenditure, and adaptive thermogenesis. The paper is published in Obesity Facts.

What is semaglutide

Host 1Before we get into the study, let us set the stage. Semaglutide is the active ingredient in Ozempic and Wegovy. What does it actually do?

Host 2Semaglutide is a GLP one receptor agonist. GLP one, glucagon-like peptide one, is a hormone your gut releases after eating. It signals to the pancreas to release insulin, slows gastric emptying, and acts on the brain to reduce appetite. Semaglutide is a synthetic version engineered to last about a week in the body, compared to minutes for the natural hormone.

Host 1And the weight loss in clinical trials has been substantial.

Host 2In the landmark STEP trials, semaglutide at the two point four milligram weekly dose produced average weight loss of roughly fifteen percent of body weight over sixty eight weeks. That magnitude was previously only achievable through bariatric surgery.

Why body composition matters

Host 1So why does anyone care what kind of weight is lost? A kilogram is a kilogram.

Host 2Because a kilogram of fat and a kilogram of muscle are not equivalent. Skeletal muscle is your primary glucose disposal organ, your largest amino acid reservoir, and it is essential for mobility, balance, and independence as you age. Losing too much muscle while dieting can leave you weaker, more insulin resistant, and at higher risk of falls and frailty. That combination of low muscle mass with high body fat is called sarcopenic obesity.

Host 1And the concern with GLP one drugs specifically?

Host 2When people lose weight by any method, some of the loss is always lean tissue. The established range for caloric restriction is roughly twenty to forty percent of total weight lost. The concern has been that GLP one drugs, because they suppress appetite so effectively, might cause disproportionate muscle loss through reduced protein intake, reduced anabolic signalling, or some direct effect on muscle tissue.

Study design

Host 1That is what this paper set out to examine. Tell me about the study design.

Host 2This was a single-centre retrospective observational study at the obesity treatment unit in Vall d'Hebron, covering January 2024 through December 2025. They identified eighty three adults with obesity, defined as BMI of thirty or above, and without diabetes.

Host 1Why exclude diabetes?

Host 2Because diabetes itself affects body composition, muscle metabolism, and energy expenditure. Including diabetic patients would introduce confounders. The tradeoff is that the results may not generalise to the large population of people with both obesity and type two diabetes.

Host 1How were the groups structured?

Host 2Thirty nine patients received a lifestyle educational programme alone. Forty four received the same lifestyle programme plus semaglutide, with at least seventy percent treatment exposure during follow-up required for inclusion. Both groups were assessed at baseline and twelve months using anthropometry, dominant handgrip strength, body composition by bioelectrical impedance analysis, and resting energy expenditure by indirect calorimetry.

Host 1Indirect calorimetry. That is the gold standard for measuring metabolic rate at rest?

Host 2It is. The patient breathes into a sealed canopy while the machine measures oxygen consumption and carbon dioxide production. From those gas exchange rates you can calculate exactly how many calories the body burns at rest. Far more accurate than any predictive equation.

Weight loss results

Host 1What did the weight loss look like?

Host 2The semaglutide group lost significantly more weight at twelve months. The between-group difference was eleven kilograms for body weight, four kilograms per square metre for BMI, and seven centimetres for waist circumference. All highly significant after adjusting for baseline age, sex, and body weight.

Host 1And the waist-to-height ratio?

Host 2The lifestyle group showed a small, non-significant increase of point zero one. The semaglutide group showed a decrease of point zero four. The between-group difference was significant. That ratio is a useful marker of central adiposity, the visceral fat that drives cardiometabolic risk.

Body composition

Host 1Now the main question. What happened to muscle?

Host 2In absolute terms, the semaglutide group did lose lean tissue. Skeletal muscle mass dropped from 29.9 to 28.7 kilograms. Fat-free mass dropped from 53.8 to 52.7 kilograms. Both statistically significant.

Host 1So the muscle loss is real.

Host 2In absolute terms, about one point two kilograms of skeletal muscle and one point one kilograms of fat-free mass over twelve months. But they also lost substantially more fat mass, including reductions in visceral fat area. The overall weight loss was dominated by fat, not muscle.

The SMI-SDS metric

Host 1But how do you tell whether the muscle loss is proportionate or pathological? One kilogram sounds modest, but in someone who did not have much muscle to begin with, it could matter.

Host 2This is the study's most important methodological contribution. They used a metric called SMI-SDS, Skeletal Muscle Index Standardised Deviation Scores. This is a continuous reference score that tells you how much muscle mass a person has relative to what is expected for someone of their age and body size. It was validated against gold-standard MRI data.

Host 1So instead of asking did you lose muscle, it asks do you have more or less muscle than expected for your new body size.

Host 2Exactly. And here is the headline finding. The SMI-SDS improved by plus 0.52 standard deviations in the semaglutide group, compared to plus 0.09 in the lifestyle group. The adjusted p-value was below 0.001.

Host 1They lost muscle in absolute terms but gained muscle relative to their new body size.

Host 2The semaglutide group lost roughly eleven kilograms total. About one kilogram of that was skeletal muscle. The rest was predominantly fat. After that recomposition, their muscle mass relative to what is expected for someone of their new lower BMI was significantly better. The authors call this qualitative muscle mass optimisation despite absolute lean tissue loss.

Handgrip strength

Host 1Did they also measure strength directly? Because mass and function are not the same thing.

Host 2They measured dominant handgrip strength, the most widely validated clinical measure of overall muscle function. Absolute handgrip strength was preserved in both groups with no significant decline at twelve months.

Host 1And relative strength?

Host 2Relative handgrip strength, grip force divided by body weight, improved significantly in the semaglutide group with an adjusted p-value of 0.003. When you lose ten kilograms but maintain the same grip force, your strength-to-weight ratio goes up. That is functionally meaningful for daily activities and fall prevention.

A word from our sponsor

Host 1Are you tired of carrying around more body hair than you need?

Host 2The Hastrup Review of Science is brought to you by epilators. A device that removes hair from the root, one follicle at a time, with the patience of a retrospective observational study.

Host 1I used one once. The pain was significant and the between-group difference versus waxing was not statistically significant. But the results were preserved at twelve months.

Host 2Unlike adaptive thermogenesis, the effect does not attenuate over time. Your legs remain smooth while your metabolic rate remains unpredictable.

Host 1Epilators. Optimising your surface-to-hair ratio since the invention of electricity. Our referral link is in the show notes.

Adaptive thermogenesis

Host 1Back to the science. The paper also measured adaptive thermogenesis. What is that, and why does it matter for weight loss?

Host 2When you lose weight, your resting energy expenditure drops. Part of that drop is predictable, because you have a smaller body with less metabolically active tissue. But in many people the drop is larger than the math predicts. That extra reduction, the gap between what your metabolism should be and what it actually is, is called adaptive thermogenesis.

Host 1Your body is fighting to regain the weight.

Host 2The mechanisms include reduced thyroid hormone signalling, decreased sympathetic nervous system activity, and increased mitochondrial efficiency. Your body becomes more fuel-efficient in a way that makes maintaining weight loss harder. This was dramatically illustrated in the Biggest Loser study, where contestants showed persistent metabolic adaptation years after the competition ended.

Host 1What did this study find?

Host 2They derived a study-specific predictive equation for resting energy expenditure and calculated adaptive thermogenesis as the difference between measured and predicted values at baseline and twelve months, with a sensitivity analysis using the independently validated Ostendorf equation.

Host 1And the numbers?

Host 2Both groups showed substantial adaptive thermogenesis. The semaglutide group had a delta of minus 210 kilocalories per day. The lifestyle group had a delta of minus 373 kilocalories per day. The semaglutide group's metabolic slowdown was numerically smaller, but the between-group difference did not reach statistical significance.

Host 1So both groups experienced their metabolism slowing down more than predicted, and semaglutide did not make it worse.

Host 2That is a fair summary. The clinical implication is that adaptive thermogenesis is a real phenomenon during weight loss regardless of method, and monitoring resting energy expenditure with indirect calorimetry may help clinicians adjust dietary targets during treatment.

Caveats

Host 1Let us talk about what could be wrong with this study.

Host 2First and most important: this is a retrospective observational study, not a randomised controlled trial. Patients were not randomly assigned to semaglutide or lifestyle-only. Clinicians selected who received the drug, which introduces selection bias. The authors adjusted for baseline age, sex, and body weight, but that cannot fully compensate for unmeasured confounders.

Host 1Second?

Host 2Sample size. Eighty three patients total, split thirty nine and forty four. That is small for detecting between-group differences in outcomes like adaptive thermogenesis, where individual variability is large. The non-significant difference in adaptive thermogenesis could be a true null or simply underpowered.

Host 1Third?

Host 2Body composition was measured by bioelectrical impedance analysis, not by dual-energy X-ray absorptiometry or MRI. Bioelectrical impedance estimates lean and fat mass by passing an electrical current through the body and measuring resistance. It is clinically practical but less accurate than DEXA or MRI, particularly during rapid weight change when hydration status shifts. The SMI-SDS metric was validated against MRI, which adds confidence, but the individual measurements still carry more noise.

Host 1Fourth?

Host 2Single centre. All patients came from one obesity clinic in Barcelona. Demographics, dietary habits, clinical protocols, and dosing patterns at one Spanish hospital may not reflect practice elsewhere.

Host 1And fifth?

Host 2Handgrip is well validated as a clinical marker of overall muscle function, but it tests one movement pattern. Lower-body strength, which is arguably more relevant for mobility and fall prevention in people with obesity, was not measured. A comprehensive assessment would include leg press, chair-stand tests, or gait speed.

Bottom line

Host 1What is the bottom line?

Host 2In eighty three adults with obesity and without diabetes, twelve months of semaglutide plus lifestyle intervention produced roughly eleven kilograms more weight loss than lifestyle alone. Absolute muscle mass decreased by about one kilogram, but relative muscle mass, adjusted for age and BMI, actually improved. Handgrip strength was preserved, relative strength improved, and adaptive thermogenesis was present in both groups without significant difference.

Host 1So for clinicians?

Host 2The muscle loss observed during semaglutide treatment appears proportionate to overall weight loss in this cohort, not pathological. But this needs confirmation in larger, randomised studies with imaging-based body composition and comprehensive strength testing. The authors recommend that follow-up should include body composition assessment, handgrip strength, and indirect calorimetry. Not just a scale.

Sign-off

Host 1That is the deep dive. Longitudinal Changes in Body Composition, Adaptive Thermogenesis and Muscle Strength in Patients with Obesity Treated with Semaglutide, from Filippi-Arriaga and colleagues at Vall d'Hebron in Barcelona. Published in Obesity Facts. Link in the show notes.

Host 2This has been the Hastrup Review of Science. See you next week.