Injected medications that treat diabetes and obesity increase the risk of a rare but serious side effect: stomach paralysis, according to new data on the real-world use of the drugs.

At least three new studies based on large collections of patient records show that the risk of being diagnosed with stomach paralysis, or gastroparesis, is higher for people who take GLP-1 agonists than for those who don’t.

The studies have not been scrutinized by outside experts or published in medical journals, so the data is considered preliminary. Two were presented Saturday at the medical conference Digestive Disease Week 2024 in Washington; the third is slated to be presented Monday.

Injected medications called GLP-1 agonists are in high demand because they have proved to be so effective for weight loss. In clinical trials, some of the stronger medications like Wegovy and Zepbound have been found to help people lose at least 10% of their starting weight. Studies have also concluded that they have benefits for the heart as well as the waistline. Drugmaker Novo Nordisk said 25,000 people are starting Wegovy every week in the US alone.

The drugs curb hunger by slowing passage of food through the stomach. They also help the body release more insulin and help send signals to the brain that turn down cravings.

In some people, however, these medications can also cause unpleasant-to-severe bouts of vomiting, which may require medical attention. They can also slow the stomach so much that medical tests show a condition called gastroparesis.

Most of the time, doctors say, gastroparesis will improve after stopping the medication. But some people say that their condition did not get better even months after coming off the drug, with life-altering consequences.

In the new studies, the risk of gastroparesis appears to be rare, but it is consistent. Compared with similar people who didn’t take GLP-1 medications, those who did had about a 50% higher risk of being diagnosed with the condition.

One study led by researchers at University Hospitals in Cleveland used records collected by the TriNetX database, which includes millions of patient records from 80 contributing heath care organizations. The analysis focused on adults who were obese, with a body mass index higher than 30, but who did not have a diagnosis of diabetes and had not been diagnosed with gastroparesis or pancreatitis at least six months before starting a GLP-1 medication. Records from more than 286,000 patients were included in the study.

Diabetes by itself can also increase the risk of gastroparesis, especially if a person’s blood sugar hasn’t been well-controlled for a long period of time.

Among people who were prescribed a GLP-1 medication for weight loss – such as semaglutide (branded as Ozempic and Wegovy), exenatide (Byetta) and liraglutide (Victoza) – 10 out of every 10,000, or 0.1%, were diagnosed with gastroparesis at least six months later. By comparison, 4 out of 10,000 people, or 0.04%, who were matched in the database based on their age, sex, ethnicity and other factors, but who were not taking a GLP-1 medication, developed the condition.

The difference, which was statistically significant, amounted to a 52% increased risk of being diagnosed with stomach paralysis while on a GLP-1 medication.

A second study, led by researchers at the University of Kansas, also used records from the TriNetX research network database. It included patients who were prescribed GLP-1 medications for diabetes or obesity between December 2021 and November 2022, and it compared them with people who had diabetes or obesity and had been seen by a doctor during the same time frame but had not been prescribed a GLP-1 medication. Records from nearly 300,000 patients were included in the study.

Compared with those who were not taking a GLP-1 medication, those who did were about 66% more likely to be diagnosed with gastroparesis. This study found that 0.53% of patients on GLP-1 medications were diagnosed with stomach paralysis, or about 1 case of gastroparesis for every 200 people taking the drugs.

People taking GLP-1 medications were also more likely to have nausea and vomiting or gastroesophageal reflux disease (GERD) and to be prescribed a proton pump inhibitor. They were more likely to have their gallbladders removed and experience drug-induced pancreatitis.

“Although these drugs do work and should be used for the right reason, we just want to caution everyone that if you do decide to start this, be prepared that you have a 30 percent chance that you may have GI side effects, and then the drug may have to be discontinued,” said study author Dr. Prateek Sharma, a professor of medicine at the University of Kansas School of Medicine.

Some side effects with the medications may also diminish over time as people get used to their doses. This is one reason doctors start with a low dose of the drug and work up to higher amounts over time.

Sharma noted that the study included people who had diabetes in both the group taking the GLP-1 medications and in the comparison group, and they still found a higher incidence of stomach paralysis in those taking the medications, suggesting that diabetes alone wasn’t driving the increased risk.

“The drug was the only thing which was different between these two groups,” he said.

“And we do show that all GI side effects or symptoms, nausea, vomiting and gastroparesis, were significantly higher in the GLP-1 takers as compared to the controls,” said Sharma, who is also president-elect of the American Society of Gastrointestinal Endoscopy.

Even though these drugs have been extensively studied, Sharma thinks it is possible that gastroparesis is rare enough that it didn’t show up in the drugs’ clinical trials because they didn’t include enough patients.

“You need hundreds of thousands of patients to come up with these conclusions, but that’s why I think these database studies are much more important there,” Sharma said.

Another reason it may have been missed in clinical trials was the way researchers often test for it, according to Dr. Michael Camilleri, a gastroenterologist and researcher at the Mayo Clinic who has studied gastroparesis with the GLP-1 drug liraglutide.

“It’s very important, if you’re going to study the problem with gastric emptying, you have to look at the gastric emptying of solids, not of liquids,” from the stomach, Camilleri said. Liquids pass through the stomach more rapidly than solids.

“When the pharmaceutical companies did the appraisal of the effects of this class of medications on gastric emptying, they usually use a method that assesses the emptying of liquids from the stomach,” he said.

It’s called the acetaminophen absorption test, and it’s often used because it’s faster and less expensive than a gastric emptying study with scintigraphy, which uses a radioactive tracer to see how much solid food is left in the stomach hours after a meal.

Acetaminophen is absorbed through the stomach and carried into the bloodstream by liquids. Measuring how quickly acetaminophen shows up in the blood can give an idea how fast liquids are passing through the stomach, but not solids. Camilleri and other experts say acetaminophen absorption is not an adequate test for gastroparesis on GLP-1 medications.

Camilleri was a co-author on a third study being presented Monday at Digestive Disease Week that looked at gastroparesis with GLP-1 medications.

That study combed through records of nearly 80,000 patients prescribed a GLP-1 medication by doctors in the Mayo Clinic’s health system. The researchers focused on a subset of 839 people who’d had symptoms of gastroparesis and who had gotten a gold-standard test for the condition: a procedure called gastric emptying scintigraphy.

About one-third of that group, 241 people, had food in their stomach four hours after eating a test meal, which means they qualified as having gastroparesis.

However, the study didn’t calculate the difference in the risk of gastroparesis between people using the drugs and those who were not.

Camilleri said it’s likely that the gastroparesis risk is be underestimated in these studies because not everyone who had symptoms would have ultimately gotten the test needed to diagnose it.

In the Mayo Clinic study, women and people who also reported having constipation on the GLP-1 medications were more likely to receive a diagnosis of gastroparesis.

Camilleri said that constipation may be one clue that people will have trouble with gastroparesis on a GLP-1 medication but that there are still many questions to answer.

“For the people who get this complication, it is extremely serious,” he said.