In the US study, significantly more patients taking semaglutide developed gastroparesis than those using other weight loss strategies (bupropion/naltrexone or gastric sleeve surgery).
A possible link between semaglutide and gastroparesis has already been described in diabetics – presumably in connection with the slowed gastric emptying that is one of the known effects of GLP-1 agonists. Whether this risk also exists in people without a history of diabetes was previously unclear – i.e. in patients without the known risk factor of diabetes, which itself is associated with an increased prevalence of gastroparesis.
This retrospective cohort study therefore aimed to investigate the risk of gastroparesis under semaglutide in obese patients without type 2 diabetes – in comparison to two other weight reduction strategies established in the USA: bupropion/naltrexone, a pharmacologically different preparation with no effect on gastric emptying (no longer available in Germany), and sleeve gastrectomy as a surgical approach.
The data was based on the US insurance database Merative MarketScan (2018–2022). It included 55,460 obese adults (BMI ≥ 30 kg/m², aged 18–64) without pre-existing gastroparesis or type 2 diabetes. The three treatment groups comprised 36,990 individuals receiving semaglutide, 7,369 receiving bupropion/naltrexone and 11,101 who had undergone sleeve gastrectomy. The mean follow-up period ranged from 1.0 to 2.1 years, depending on the group.
The study examined how often gastroparesis was diagnosed for the first time after the start of the respective intervention. After evaluating the insurance data, a clear difference in the frequency of gastric emptying disorders was found between the groups studied:
Even when applying a stricter definition of delayed gastric emptying – namely, only when a prescription for a prokinetic agent was present in addition to the diagnosis of gastroparesis – the risk remained significantly increased under semaglutide (vs. bupropion-naltrexone: HR 2.46; vs. SG: HR 4.36).
Some groups of people in the semaglutide arm were particularly susceptible to gastroparesis: in this group, there was a higher prevalence of metabolically associated fatty liver disease (MAFLD; HR 2.11), gastroesophageal reflux disease (GERD; HR 1.80), a body mass index between 30 and 34 (vs. ≥ 40; HR 1.93) and female gender (HR 1.56) were significantly associated with an increased risk of gastroparesis.
Despite an overall low incidence of gastroparesis in the study population, the study shows a significantly increased risk with semaglutide in individuals who use the drug for weight loss. In comparison, the surgical option of sleeve gastrectomy performed best.
The results could be important for practice both in the choice of therapy for weight normalisation and in perioperative management, especially in patients who already have a history of gastrointestinal problems. The authors point to a potential risk of aspiration due to delayed gastric emptying during sedation, even though the data on this is inconsistent so far.
However, the results must be viewed in the context of methodological limitations: the diagnosis of gastroparesis was based solely on ICD codes and was not confirmed by gastric emptying tests. Furthermore, it remained unclear whether and how regularly the prescribed medications were actually taken. The follow-up period was also limited. Despite these limitations, the analysis highlights that potential gastrointestinal side effects of GLP-1 agonists should be given greater consideration in future – both in studies and in practical care.