Your Stomach Is Probably Not Paralyzed: What the Gastric-Emptying Trials Found About Bloating

The explanation everybody gives you

Search for why a GLP-1 makes you bloated and every page returns the same sentence: these drugs slow your stomach down, so food sits there, so you feel full and swollen.

It is a tidy mechanism. It is also, as a description of what happens over months of treatment, not what the trials that actually measured it found.

This matters practically, not just pedantically. If you believe your stomach has stopped working, the sensible response is fear — and a lot of women arrive at their prescriber convinced they are developing gastroparesis when what they have is gas.

What was measured, twice

Gastric emptying can be measured. It has been, in semaglutide, in two placebo-controlled trials that reached compatible conclusions from different directions.

The 2018 trial. Thirty adults with obesity, a randomized double-blind two-period crossover, semaglutide escalated to 1.0 mg weekly against placebo, twelve weeks per period, emptying assessed by paracetamol absorption. First-hour emptying after a meal was delayed — an estimated treatment ratio of 0.73 (95% CI, 0.61 to 0.87). But overall gastric emptying across five hours was not statistically different between treatments¹.

The 2021 trial. Seventy-two adults with obesity, semaglutide escalated to 2.4 mg — the dose actually used for weight management — against placebo for twenty weeks. Five-hour paracetamol AUC rose 8% (P = 0.005), but the difference became non-significant once corrected for week-20 body weight (P = 0.12). There was no effect on first-hour emptying at all. The authors' stated conclusion: there was no evidence of delayed gastric emptying at week 20².

Read together: a first-hour delay at the lower dose and twelve weeks, no detectable first-hour delay at the higher dose and twenty weeks, and in neither study a meaningful slowing of overall emptying.

The 2021 trial was sponsored by Novo Nordisk, with three authors employed by and holding shares in the company and a contract research organization paid to help run it². Worth knowing — though note this result runs against the convenient story. It attributes the appetite effect to appetite suppression rather than to a stomach that has stopped emptying, which is the less dramatic claim.

So why do you feel like a balloon?

Because bloating and delayed emptying are not the same thing, and the label's own numbers separate them cleanly.

Here is what the Wegovy label reports for the gas-and-bloating cluster, injection 2.4 mg against placebo:

Two things jump out of that table.

Burping is the signature. Going from under 1% to 7% is the largest fold increase anywhere in the label's gastrointestinal list — a roughly sevenfold rise, and far sharper than nausea's or constipation's. If you have noticed one new thing, it is statistically most likely to be that.

Visible swelling is not. Abdominal distension moves from 5% to 7%. That is a two-point difference, and it deserves saying out loud because the internet describes GLP-1 bloating as though it were near-universal. On the label, most of the bloating sensation is not measurable distension.

The much larger signal in that table is the one at the bottom. Constipation more than doubles, 11% to 24% — and a slowed bowel produces exactly the pressure, fullness and gas that get described as bloating. That is a far better explanation for most women's experience than a paralyzed stomach, and unlike gastroparesis it is straightforwardly treatable. What actually works for GLP-1 constipation is the more useful page if that is your pattern.

The bacterial-overgrowth question, sized properly

A 2025 cohort study is circulating in GLP-1 forums as evidence that these drugs cause small intestinal bacterial overgrowth. It is a real study and the effect is real. The size is what gets lost.

Using a global database, researchers matched 216,173 adults with type 2 diabetes starting a GLP-1 or dual GLP-1/GIP agonist against an equal number starting other second-line diabetes agents. Short-term, diagnostically confirmed SIBO was more common on the GLP-1 drugs — hazard ratio 2.14 (95% CI, 1.13 to 4.07; P = 0.049)³.

A doubled risk sounds alarming until you read the rates it doubles:

That is fewer than two diagnoses per ten thousand patient-years, against roughly one. The long-term analysis did not reach significance (HR 2.02; 95% CI, 0.98 to 4.12)³.

Three further limits belong with it. The population was people with type 2 diabetes on second-line therapy, not women taking a GLP-1 for weight. "Diagnostically confirmed" means somebody ordered a breath test, so the figure partly measures which patients get investigated. And the authors' own recommendation is modest: symptom-driven breath testing may be warranted³.

Bloating on a GLP-1 is common. SIBO is rare. Both statements are true, and the second is not the explanation for the first in most people.

When it is not ordinary

There is a line, and it is worth knowing where it sits rather than worrying at every twinge.

Gastroparesis is a recognized and serious event with these drugs, and a 2026 systematic review has now cataloged its clinical features, diagnosis and management⁵. The distinction that matters is not intensity but pattern. Ordinary GLP-1 bloating fluctuates, is worse after eating, and settles between meals and across weeks as you adapt to a dose. What warrants a call is vomiting food eaten many hours or a day earlier, inability to keep fluids down, severe pain that does not ease, weight loss faster than intended, or symptoms that keep escalating rather than plateauing after a dose step.

One structural point deserves saying: a broad 2024 review in The Lancet Gastroenterology & Hepatology set out the mechanisms and management of these gastrointestinal effects and noted how much of the management guidance rests on clinical experience rather than trial evidence⁴. Advice about GLP-1 bloating, including much of what follows, is reasoning from mechanism rather than reading from a randomized comparison.

What actually tends to help

Ordered by how well supported each one is, which is not the order they usually appear in.

Treat the constipation first, and properly. It is the largest effect in the label's table by a wide margin and the most likely driver of what you are feeling. It also responds to ordinary measures.

Do not push through a dose step. The label's instruction for both the injection and the tablet is the same when a dose is not tolerated: consider delaying the escalation. Bloating that arrives with a dose increase and is still there three weeks later is information about the pace, not something to be endured on principle.

Slow the meal, shrink the meal. Air swallowed while eating quickly is a genuine contributor to eructation, and eating quickly is what happens when you eat standing at a counter between other people's needs. This is the least medical advice on the page and it is not the least effective.

Notice the carbonated drinks. Fizzy water is many women's substitute when appetite falls away and food stops appealing. It is also a direct source of the gas you are then attributing to the drug.

What does not have evidence behind it, despite being sold hard: digestive-enzyme supplements, "GLP-1 support" probiotic blends, and detox protocols. There are no trials of any of these against GLP-1 bloating.

The honest summary

Something real is happening and the popular explanation for it is mostly wrong.

The controlled measurements say overall gastric emptying is close to normal on the obesity dose at twenty weeks. The label says the drug's clearest gas-related effect is burping, that measurable distension barely moves, and that the biggest gastrointestinal change by far is constipation. The overgrowth risk exists and is very small in absolute terms.

Which is, on balance, reassuring — the mechanism most likely responsible for your bloating is the most treatable one on the list.