GLP-1 Drugs and Kidney Stones: What the Labels and Studies Show

No trial has been designed to find out whether Ozempic, Wegovy, Mounjaro or Zepbound cause kidney stones. The closest thing to an answer is one line on the Wegovy label: in a three-year heart trial, 1.2% of people on semaglutide reported a urinary stone against 0.8% on placebo.^[1] The Ozempic, Mounjaro and Zepbound labels say nothing about stones. One large veterans’ cohort linked GLP‑1 drugs to more stones than usual care,^[5] while urine tests in people who already form stones showed no shift toward stone formation during GLP‑1 weight loss.^[15] The verdict: a small possible signal, not an established side effect, and the most plausible route runs through fluid, not the drug acting on the kidney.

The evidence at a glance

What the labels say

Wegovy is the only GLP‑1 label that mentions stones. The figure comes from its cardiovascular outcomes trial, which randomized 17,604 adults with heart disease and followed them for a median of a little over three years.^[1]

In a CV outcomes trial, 1.2% of WEGOVY-treated patients and 0.8% of patients receiving placebo reported urolithiasis, including serious reactions that were reported more frequently among patients receiving WEGOVY (0.6%) than placebo (0.4%).

— Wegovy (semaglutide) prescribing information, section 6.1

Urolithiasis means a stone anywhere in the urinary tract, kidney included. The absolute gap is 0.4 percentage points, or about four extra people per thousand over three years. Two caveats apply. Stones were not a planned endpoint, and the label notes that safety reporting in this trial was limited to serious events, events that led people to stop, and events of special interest.^[1] The label gives no test of whether the difference could be chance.

The Ozempic, Mounjaro and Zepbound labels contain no stone terms at all.^[2][3] What every one of them carries, Zepbound included,^[4] is the acute kidney injury warning: injury reported after approval, mostly in people dehydrated by nausea, vomiting or diarrhea, with patients told to take precautions against fluid loss. That warning concerns sudden loss of kidney function, not stones, and is covered in GLP-1 drugs and your kidneys. It matters here because low urine volume is also the oldest known driver of stones.

The cohort studies: one signal

The largest direct look comes from US veterans with diabetes. Researchers compared 215,970 people who started a GLP‑1 drug with several control groups across 175 health outcomes. Against usual care, GLP‑1 use was linked to a higher risk of nephrolithiasis, alongside stomach disorders, low blood pressure, fainting, arthritic disorders and pancreatitis.^[5] Studies that scan 175 outcomes at once are built to generate leads, and veterans with diabetes are not the typical person taking these drugs for weight.

The randomized data on tirzepatide point the other way, with wide uncertainty. A meta-analysis of 15 trials (14,471 people) found tirzepatide did not raise the risk of nephrolithiasis compared with placebo, insulin or other GLP‑1 drugs.^[6] A second, of 20 trials, grouped kidney stones with other kidney adverse events and found a risk ratio of 1.22 with a confidence interval from 0.78 to 1.89, which is compatible with no effect.^[7] Trials of 26 to 72 weeks are short for a condition that can take years to show up.

Why the SGLT2 studies are not evidence against GLP-1 drugs

Most of the large studies that mention GLP‑1 drugs and stones were written about a different class. SGLT2 inhibitors such as Jardiance and Farxiga make the kidneys pass sugar into the urine, and water follows. Researchers testing whether that lowers stone risk needed a comparison group of people with diabetes starting another modern drug, and they chose GLP‑1 users.

Every study found fewer stones on the SGLT2 inhibitor.^[8][9] The pattern held in people who had already had stones^[10] and in two global databases.^[11][12] A 2026 European Association of Urology review of six such studies called the association consistent and judged all six at moderate risk of bias.^[13] The comparisons rank two drugs against each other. A lower rate on SGLT2 inhibitors is what you would see if those drugs protect, and the GLP‑1 rate could sit at the ordinary level for people with diabetes. Paik’s study also compared SGLT2 inhibitors with an older, weight-neutral class, DPP‑4 inhibitors, and found almost the same gap.^[9]

The table also shows how much the baseline moves. Diagnosed stones ran at 4 per 1,000 person-years among Danish GLP‑1 users and 21 per 1,000 among insured Americans, and recurrence in people with a stone history was 156 per 1,000.^[8][9] The rate climbs most steeply not with the drug but with a stone history.^[10] For how the two classes are combined, see taking an SGLT2 inhibitor and a GLP-1 together.

What happens to urine on a GLP-1

Stones form when urine carries more calcium, oxalate or uric acid than it can keep dissolved, so a 24-hour urine collection is the standard way to measure risk. Two studies looked at stone formers on GLP‑1 drugs.

• Dallas weight clinic. Among 44 stone formers with obesity who lost an average of 6.6 kg on GLP‑1 or tirzepatide treatment, daily urine oxalate fell from 40 to 32 mg. Calcium, citrate, uric acid, pH, total volume and the measures of how saturated the urine was did not change significantly. The authors concluded the weight loss was not linked to changes that favor stones.^[15]
• Medicare stone formers. Comparing 349 GLP‑1 users with matched controls, and 154 people tested before and after starting, researchers found no significant differences in any urine measure. SGLT2 users in the same study had higher urine volume (2.4 against 2.0 liters) and more citrate, which inhibits stones.^[14]

Both are small and observational. Neither is the reassurance a randomized trial would be, but neither found the shift in urine chemistry that a stone-causing drug would be expected to produce. The contrast with surgery is sharp: in a Scandinavian registry of 58,366 bariatric patients, every operation type was followed by roughly six times the kidney stone risk of matched controls.^[16]

Adverse-event reports: a signal, not a rate

Two studies of the FDA’s adverse-event database (FAERS) found nephrolithiasis reported with GLP‑1 drugs more often than the database average would predict. One listed it among unexpected signals for GLP‑1 monotherapy;^[17] the other found men on a GLP‑1 with metformin reported renal calculi more than women did.^[18] Disproportionality analyses of spontaneous reports show what gets reported, not how often stones happen or whether the drug caused them. Reports are shaped by publicity, by how many people take a drug, and by who files them.

Where a real risk would come from: fluid and food

What follows is inference from stone research, not from GLP‑1 trials. Three things that change on these drugs are known stone factors.

Less fluid. Urine volume is the best-tested lever in stone prevention. In a five-year randomized trial in people after a first calcium stone, 12 of 99 told to drink more water had a recurrence, against 27 of 100 given no treatment.^[20] On a GLP‑1, eating much less can mean drinking less too, and vomiting or diarrhea in the dose-escalation weeks removes more water. The labels already ask patients to avoid fluid depletion for the sake of kidney function.^[1] The same advice serves stone prevention. A related page covers urination and thirst on a GLP-1.

A different diet. High-protein, low-carbohydrate eating is common on these drugs, partly to protect muscle. In a six-week metabolic study of ten healthy adults, a low-carbohydrate high-protein diet dropped urine pH from 6.09 to 5.56, cut citrate from 763 to 449 mg a day, raised urine calcium from 160 to 258 mg, and more than doubled the urine’s saturation with uric acid.^[21] Those are pro-stone changes. A meat-heavy plan on a GLP‑1 is a different exposure from the drug itself. For eating patterns more broadly, see eating too little on a GLP-1.

Weight loss itself. Higher body weight raises stone risk: across three large US cohorts, a BMI of 30 or more carried 1.3 to 2.1 times the risk of a BMI of 21 to 22.9.^[19] Whether losing the weight lowers the risk is less clear. A 12-week low-calorie diet in 39 stone formers with obesity did not improve their urine chemistry.^[22] Losing weight is not a proven stone treatment, but nothing in the drug data suggests the loss itself raises risk.

Uric acid deserves a note because these drugs lower it in the blood. The trials behind that finding are set out in GLP-1 drugs and uric acid. A lower blood level has not been shown to mean fewer uric acid stones. In the Dallas study, urine uric acid did not change.^[15] In the diet study, by contrast, a falling urine pH came with a doubling of uric acid saturation, without any drug involved.^[21]

When to seek care

The NIH’s kidney institute, NIDDK, lists the symptoms of a stone and advises seeing a clinician right away if any appear:^[23]

• Sharp pain in the back, side, lower abdomen or groin, which may come in waves.
• Pink, red or brown urine.
• A constant need to urinate, pain while urinating, or passing only a little urine or none.
• Cloudy or bad-smelling urine.
• Nausea and vomiting with the pain, or fever and chills.

What to ask a prescriber

1. I have had a kidney stone. Does that change which drug suits me? Prior stones are the strongest risk factor in these studies, and in people with diabetes the cohort data favor SGLT2 inhibitors for stone risk.
2. Should I have a 24-hour urine test? For a repeat stone former, it shows which factor is driving the stones before weight loss changes the picture.
3. How much should I be drinking? Ask for a daily target, and a plan for the weeks when nausea makes drinking hard.
4. Is my diet raising my risk? Bring up protein shakes, a low-carbohydrate plan and any supplements.
5. What should make me call? Agree in advance what counts as vomiting or diarrhea severe enough to stop and phone.