A phase 2 liver trial where the add-on drug didn't help — and the base drug quietly did
Read the Sunday Brief →

Study wrapper · #177

Ketoacidosis Risk in Non-diabetic Patients Using Semaglutide Versus Tirzepatide for Obesity: A Disproportionality Analysis of the FDA Adverse Event Reporting System.

Makhmutov A, Qureshi F Cureus. 2026.

Editor's note

A pharmacovigilance disproportionality study of the FDA Adverse Event Reporting System (FAERS), comparing ketoacidosis reports for semaglutide versus tirzepatide in non-diabetic people using these drugs for weight. The design is important: FAERS is a spontaneous-reporting database, so a disproportionality signal (reporting odds ratio) flags an association worth watching, not an incidence rate or a causal link. Reporting is voluntary, subject to media-driven and prescribing-volume biases, and denominators are unknown. With those caveats, the findings are notable. Researchers reported a significant ketoacidosis signal for both agents, stronger for semaglutide (ROR 3.15) than tirzepatide (ROR 1.22), with roughly three-quarters of cases requiring hospitalisation and a sharp rise in tirzepatide reports through 2025. Euglycaemic ketoacidosis in non-diabetic users is biologically plausible and clinically serious. This is a hypothesis-generating safety signal that argues for clinical vigilance; it cannot establish how often the event actually occurs, and confirming it requires cohort or registry data with proper denominators.

Plain-language abstract

Semaglutide and tirzepatide are injected weight-loss medicines increasingly used by people who do not have diabetes. Ketoacidosis, a dangerous build-up of blood acids, has emerged as a worry in this group. Researchers searched the FDA's voluntary adverse-event database (FAERS) from January 2021 to December 2025, keeping reports where these drugs were the main suspected cause and where the person was not diabetic, and counted ketoacidosis cases. Because reporting is voluntary and the total number of users is unknown, this method can show whether a problem is reported more often than expected, but not how common it truly is or whether the drug caused it. Among non-diabetic reports, ketoacidosis appeared in 261 semaglutide and 209 tirzepatide cases. Both showed a statistically raised reporting signal, and it was higher for semaglutide than for tirzepatide. About 74% of semaglutide and 71% of tirzepatide cases needed hospital admission, and tirzepatide reports climbed steeply, from 1-2 per quarter in 2022 to 28-34 per quarter by 2025. The authors describe this as an emerging safety concern warranting clinical vigilance.