Gallbladder: what the data says

Why this guide is careful

The gallbladder question is where the internet is at its worst on this subject, in both directions. One half of it insists these drugs destroy your gallbladder. The other half insists the whole thing is a myth invented by people who dislike the medicines. Neither is what the evidence shows, and the actual position is more useful than either.

Sunil’s reading group went through this properly, and the honest summary is three sentences. Gallbladder and biliary events are reported more often on these drugs than on placebo in the trial safety data. Rapid weight loss by any means — surgery, very low intake, anything — is a long-established cause of gallstones, so there is a plausible route that has nothing to do with the drug directly. And nobody has cleanly separated those two explanations in humans.

That is the whole of what is known. Everything else in this guide is either mechanism, or the practical business of recognising gallbladder pain, which is the part that might actually matter to you one evening.

How gallstones form, in plain terms

The gallbladder is a small reservoir that stores bile made by the liver and squeezes it into the intestine when fat arrives. Stones form when the bile becomes supersaturated with cholesterol and crystals precipitate out — and that is made more likely by two things in particular.

Concentrated bile. During rapid weight loss the liver mobilises a large amount of cholesterol, which ends up in the bile.

A gallbladder that is not emptying often. The trigger for emptying is fat arriving in the small intestine. If you are eating much less, and much less fat, the gallbladder sits fuller for longer, and stasis lets crystals grow.

Both of those apply to anybody losing weight quickly for any reason, which is why bariatric surgery patients are routinely warned about gallstones and sometimes given preventive medication. And there is a third possible route specific to this drug class: GLP-1 signalling itself may reduce gallbladder contraction, which would add stasis on top of stasis. That mechanism is plausible and has some experimental support; it is not established as the explanation for what is seen in the trials.

What the trials reported

In STEP 1 (Wilding, NEJM, 2021), gallstones — cholelithiasis — were reported as an adverse event in a small percentage of participants on semaglutide 2.4 mg and in a smaller percentage on placebo. The numbers were low in both arms and the imbalance was real. Gallbladder-related events were also among the reasons a handful of participants had serious adverse events.

The tirzepatide programme (SURMOUNT-1, Jastreboff, NEJM, 2022) similarly lists gallbladder-related disorders among adverse events, again at low rates.

Two things Sunil insists on adding whenever this comes up. First, these trials were not designed to detect gallbladder disease, so the numbers are a by-product rather than a measurement — nobody was scanned. Second, the arms differ in the amount of weight lost, which is precisely the confounder you would need to remove to answer the causal question, and no trial has been built to do it.

So: real, uncommon, and of uncertain cause. That is a genuinely awkward answer and we would rather give it than a tidy one.

What gallbladder pain is actually like

This is the practically useful section, because gallbladder pain is distinctive enough to recognise and members routinely mistake it for "bad nausea" for weeks.

Biliary colic — a stone temporarily blocking the outlet — typically presents as:

• Pain in the upper right abdomen under the ribs, or in the centre just below the breastbone.
• Often radiating to the right shoulder blade or between the shoulders.
• Coming on after eating, classically after a fattier meal, often in the evening or waking you at night.
• Building over minutes to a steady, severe, gripping pain rather than a wave that comes and goes second by second.
• Lasting from around half an hour to several hours, then settling completely.
• Frequently with nausea and vomiting.

Members describe it as unmistakably different from GLP-1 nausea, which is a queasy fullness. This is pain. Rowan described a first episode as being convinced she had cracked a rib in her sleep.

The versions that need urgent care

Uncomplicated colic settles. Three complications do not, and they need same-day assessment:

• Cholecystitis — inflammation of the gallbladder. Pain that does not settle after several hours, with fever, feeling generally unwell, and tenderness that makes you catch your breath when pressed.
• A blocked bile duct — pain with yellowing of the eyes or skin, dark urine, pale stools, or itching.
• Cholangitis — infection in the bile ducts. Pain, fever with shaking chills, and jaundice together. This is a medical emergency.

And the one that overlaps and matters most: a gallstone can trigger pancreatitis. Severe pain that bores through to your back, is worse lying flat and better leaning forward, with vomiting, needs assessment today. The pancreatitis guide is the companion to this one and the red-flag list has all of it in one place.

None of this is something we can assess. If you are reading a symptom list at eleven at night and comparing it against your own abdomen, the answer is to ring somebody, not to keep reading.

What members have done about it

A mixture of practical measures and things worth asking about. Not advice.

• Mentioning gallbladder history before starting. Members with previous stones, or a family history, raised it with their prescriber and several were watched more closely. This is a genuinely useful conversation to have early.
• Not eliminating fat entirely. Counter-intuitive but mechanistically sensible: a gallbladder that never contracts is a gallbladder that stagnates. Members who had cut fat right down were sometimes advised to include some regularly. Take that from a clinician rather than from us.
• Asking about preventive medication. Ursodeoxycholic acid is used in some rapid weight loss settings to reduce stone formation. Whether it is appropriate for anyone on a GLP-1 is a clinical decision, and the evidence base for that specific use is thin. Ask; do not assume.
• Not ignoring the first episode. The commonest regret in circle. Members who reported an evening of severe right-sided pain and then said nothing generally had a second, worse episode.
• Life after removal. Several members have had their gallbladder out while on treatment and continued afterwards. Most report needing to adjust how they eat for a while, some report looser bowels for months, and nearly all report that the medicine question was a non-issue compared with the pain they had been living with.

Keeping the risk in proportion

Because it is easy to read a guide like this and become frightened of your own dinner.

Gallbladder events in the trials were uncommon. Most people on these medicines will never have one. The absolute numbers are low, and the same risk attends losing weight quickly by any method — which is worth remembering when you are weighing this against the outcomes seen in SELECT (Lincoff, NEJM, 2023) and FLOW (Perkovic, NEJM, 2024) for the people those trials studied. Risk goes on both sides of the ledger and a guide about one organ can distort the picture.

What we would take away: know the pain pattern, mention your history to your prescriber, do not soldier through an episode, and be sceptical of anyone online who tells you either that this never happens or that it happens to everybody. Sunil’s rule, as always: read the paper, not the post about the paper.