Pharmacovigilance study flags Wernicke encephalopathy with semaglutide and tirzepatide

Researchers in Israel who reviewed 195,979 adverse event reports identified 15 cases of Wernicke encephalopathy — a rare brain disorder caused by severe vitamin B1 (thiamine) deficiency — and reported that the condition was disproportionately associated with GLP-1 drugs compared with other medications ^[1]. Most of those cases involved semaglutide (sold as Ozempic, Wegovy and Rybelsus) or tirzepatide (sold as Mounjaro and Zepbound) ^[1].

The study is a disproportionality analysis, meaning it compares how often a side effect shows up in reports for one drug versus all other drugs in a safety database. That kind of analysis can flag a possible signal, but it cannot prove the drug caused the problem and cannot tell you how often it happens per patient. The findings were published in a peer-reviewed journal and covered this week by several national outlets ^[1][2].

What the researchers found

Of the 15 Wernicke encephalopathy cases, most patients had gastrointestinal problems — including weight loss, vomiting, loss of appetite or malnutrition — all of which can lead to poor nutrition and thiamine deficiency ^[1]. GLP-1 medications slow digestion and reduce appetite, which is how they help people eat less and lose weight, but that same effect can cut nutrient intake ^[1].

One detail stands out for anyone trying to spot the problem early: only two of the 15 patients had the classic "triad" of Wernicke symptoms — confusion or altered mental state, trouble walking, and loss of control over eye movements ^[1]. That is consistent with the condition in general, where only about 10% to 16% of people show all three signs, which makes it easy to confuse with dementia or intoxication ^[1]. As many as 80% of cases go undiagnosed, and some are found only at autopsy ^[1].

The stakes are high. Untreated, Wernicke encephalopathy can be fatal in up to 20% of cases, and as many as 85% of survivors develop permanent brain damage ^[1]. It is treatable if caught early, usually by replacing the missing vitamin B1, often through an IV ^[1].

Several things are not answered by the reporting available here: how many people were taking GLP-1 drugs in the database, what time period the reports covered, how the 15 patients fared, and whether other factors such as alcohol use or prior surgery played a role in individual cases. The researchers themselves described Wernicke encephalopathy as a "potentially rare" adverse event while urging clinicians to watch for it, especially in patients on semaglutide or tirzepatide ^[1].

Why it matters for patients

More than one in 10 American adults now use GLP-1 medications for weight loss, so even a rare complication affects a large denominator ^[1]. Fifteen cases out of 195,979 reports is a very small number, and adverse event databases capture only what someone chooses to report — so this is a flag for further study, not a measure of personal risk.

The practical takeaway from the study's own logic is about recognition. The people who developed Wernicke encephalopathy were not random: nearly all had ongoing vomiting, appetite loss, rapid weight loss or malnutrition ^[1]. Those are also the most common reasons people struggle on GLP-1 drugs. And because most patients did not have the textbook three symptoms, confusion or unsteadiness in someone who has been vomiting or eating very little may not immediately look like a vitamin deficiency to a clinician or a family member ^[1].

Wernicke encephalopathy is not unique to these drugs. It is most often linked to chronic alcohol use, and has also been tied to bariatric surgery, anorexia, and repeated vomiting from cannabinoid hyperemesis syndrome ^[1] — all situations where nutrients either are not absorbed or never get in.

What happens next

No regulator has announced a labeling change or new warning in response to this analysis, based on the available reporting. Signals from disproportionality studies typically need follow-up in larger datasets that include how many people actually took the drug, so the size of any real risk remains unknown. Nothing in these sources should be read as guidance about starting, continuing or changing a medication; questions about symptoms belong with a clinician.