The phrase spread online as semaglutide and tirzepatide became common. People use it for a seat that looks flatter, smaller or looser after a large weight loss. A 2026 plastic surgery review lists it with “Ozempic face” as “informal terms used to describe volume loss and reduced skin elasticity” after large weight loss on these drugs [8].
The honest question behind it is narrow: when these drugs take fat off, how much comes from the hips and buttocks, and does muscle there shrink too?
The short answer is that no trial has measured it directly. What follows is what the trials did measure, region by region, and where each one stops.
What the labels say
Neither label uses the term. Both carry the same sentence in their pharmacology sections: the drug “lowers body weight with greater fat mass loss than lean mass loss” [1] [2].
That sentence is about the whole body. The labels mention the thigh and upper arm only as places to inject, which is covered in peptide injection sites, and they say nothing about where on the body the fat comes from.
What the tirzepatide scan substudy measured
SURMOUNT-1 scanned 160 of its participants by DXA, a low-dose X-ray that separates fat, lean tissue and bone, at baseline and week 72 [3]. They were 73% female, with a mean weight of 102.5 kg.
On pooled tirzepatide doses, body weight fell 21.3% and fat mass 33.9%, against 5.3% and 8.2% on placebo. Waist circumference fell 18.1 cm against 3.4 cm.
The only region the paper reports is visceral fat, the fat packed around the organs inside the abdomen. The scanner software estimated it in the lower 20% of the trunk region, and the authors note it was estimated rather than measured directly.
Estimated visceral fat fell 40.1% on tirzepatide and 7.3% on placebo. So fat deep in the belly fell by a larger share than fat across the whole body.
The paper does not split the rest of the fat loss between hips, thighs, arms and buttocks. Any figure for the seat would be a guess, and none is given here.
What an MRI study found in the thigh
The nearest muscle anyone has scanned is the thigh. The STEP UP trial randomized adults with a body mass index of 30 or more to semaglutide 7.2 mg, semaglutide 2.4 mg or placebo weekly for 72 weeks [4].
Its MRI analysis included 55 people with a valid baseline scan: 49 on semaglutide and only 6 on placebo. Against placebo, total fat tissue fell by 11.1 L and visceral fat by 1.5 L.
Thigh muscle volume fell by 1.1 L more than placebo, but the confidence interval ran from −2.3 to 0.1 L (p = 0.07), so that difference was not statistically significant. Fat marbled inside the thigh muscle fell by 0.92 percentage points.
On a sit-to-stand test, the pooled groups averaged 13 repetitions in 30 seconds at baseline and 15 at week 72. The gluteal muscles themselves were not reported, and six placebo scans is a very small comparison group.
| Study | Drug and length | Regions reported | Buttocks measured? |
|---|---|---|---|
| SURMOUNT-1 DXA substudy | Tirzepatide, 72 weeks, 160 people | Whole body; estimated visceral fat | No |
| STEP UP MRI analysis | Semaglutide, 72 weeks, 55 people | Total and visceral fat; thigh muscle | No |
| Kadouh 2020 | Liraglutide 3 mg, 16 weeks, 35 completers | Trunk, upper-body and lower-body fat | No, lower body only |
| SURPASS-3 MRI | Tirzepatide, 52 weeks, 296 people | Visceral, belly skin-level fat and liver fat | No |
The one trial that reported lower-body fat
A 16-week trial of liraglutide, an older daily GLP-1 drug, measured regional fat by DXA in 40 adults with obesity, of whom 35 completed [5]. The dose was escalated to 3 mg a day, alongside behavioral support.
Compared with placebo, the authors report “significant reductions in total body, trunk, and upper and lower body fat without reduction in lean body mass.” The abstract does not give the size of each regional change.
Lower-body fat on a DXA scan includes the hips and legs together, so it is the nearest published measure to the seat. It still does not isolate the buttocks, and the drug and duration differ from semaglutide and tirzepatide at full dose.
Does the fat come off the belly first?
A 2024 analysis of the SURPASS-3 MRI substudy compared 296 people with type 2 diabetes against matched groups of the same sex and similar body mass index, drawn from 40,172 adults in a British population imaging study [6]. That comparison shows where fat sits for a given size.
After 52 weeks on tirzepatide, visceral fat and liver fat moved lower than expected for a person of the same size. Fat under the skin of the belly moved slightly higher, from a starting point slightly below average.
The authors read this as “a possible treatment-related shift towards a more balanced fat distribution pattern with prominent VAT and LF loss.” VAT is visceral fat and LF is liver fat. Hips and buttocks were outside the scan.
A 2026 review of fat depots reaches a similar summary: GLP-1 drugs “mainly reduce visceral and ectopic fat alongside overall weight loss” [7]. The same review says lower-body fat under the skin “may provide a relatively safe lipid-buffering reservoir,” a place to store fat with less metabolic harm.
What this means for someone on these drugs
Two things are measured. Fat falls across the body, with visceral fat falling by a larger share. Lean tissue falls too, about a quarter of the weight lost in SURMOUNT-1 [3].
How much of that lean loss is muscle, and what limits it, is covered in GLP-1 muscle loss. The total amount lost, which drives every visible change, is in how much weight you can lose on tirzepatide.
Related reading
The same question asked of the face, with the one imaging study that measured it, is in Ozempic face. Shedding after fast weight loss is in semaglutide hair loss.