The explosion in GLP-1 receptor agonist use represents one of modern medicine’s most rapid adoptions. Semaglutide and tirzepatide have transformed weight management and diabetes care for tens of millions globally in less than a decade. This speed of uptake is a triumph of innovation and unmet clinical need. It is also, necessarily, an experiment in real-world pharmacovigilance conducted at unprecedented scale. A new study published in Clinical Nutrition this September offers a sobering reminder of why that vigilance cannot lag behind adoption.
Researchers analyzing FDA FAERS adverse event reports and published literature identified fifteen cases of Wernicke encephalopathy disproportionately associated with GLP-1 drug use. Wernicke encephalopathy—a severe neurological emergency caused by acute thiamine (vitamin B1) deficiency—carries mortality rates reaching 20 percent if untreated, yet remains undiagnosed in up to 80 percent of cases. The reported patients shared a common clinical thread: gastrointestinal symptoms including vomiting, profound appetite suppression, and malnutrition, precisely the mechanisms by which these medications work. The signal is real. The question is not whether we noticed it, but whether we were looking carefully enough.
This is where intellectual honesty matters. The study itself, properly, makes no causal claim. Fifteen cases from adverse-event surveillance systems cannot prove that GLP-1 drugs cause Wernicke encephalopathy. As clinician Karthik Achari observed in response, the absolute number remains vanishingly small against a user base now measured in the tens of millions. But this is precisely where thinking becomes muddled in our era of rapid pharmaceutical scale. The absence of proof of causation in an adverse-event signal is not proof of absence of risk. Rare complications—especially those that mimic or present atypically—require decades to surface in post-marketing surveillance. Bariatric surgery, which shares the malabsorptive mechanism with GLP-1-induced appetite suppression, took years before Wernicke encephalopathy was widely recognized as a complication and before thiamine supplementation became standard prophylaxis.
The relevant comparison is instructive. When a new drug reaches one million users, pre-approval trials have already screened approximately 5,000 to 10,000 patients over years. Once fifty million people are taking it, the mathematics invert: even genuinely rare events—those occurring in 1 per 100,000 users—become visible only if someone is systematically watching. The FDA FAERS system and published case reports did exactly this work. The signal emerged not because GLP-1 drugs are uniquely dangerous, but because post-marketing surveillance, though imperfect, functioned as it should.
What follows from this should be neither panic nor dismissal. Clinicians prescribing GLP-1 drugs to patients with significant gastrointestinal side effects—particularly vomiting, sustained anorexia, or nutritional compromise—should counsel on thiamine intake and monitor for the neurological red flags of Wernicke disease: acute confusion, ophthalmoplegia, ataxia. Regulatory bodies should ensure that product labeling reflects this signal and that healthcare systems routinely capture vitamin micronutrient status in patients on these agents. And the pharmaceutical industry, which has invested billions in demonstrating the cardiovascular and metabolic benefits of GLP-1 agonists, must invest equally in the unglamorous work of pharmacovigilance networks that catch the needle of rare complications in the haystack of millions.
The Wernicke signal is not an indictment of GLP-1 drugs. It is, instead, evidence that our post-marketing surveillance systems are working as designed—imperfectly, but working. The test now is whether we act on what we have learned.
Reference
Lev D, Leibowitz A, Lang A, Shlomai G, Twig G, Eden-Friedman Y, Engel T, Cukierman-Yaffe T, Dankner R, Gerstein HC, Goldman A, et al. Glucagon-like peptide-1 receptor agonists and Wernicke encephalopathy: A pharmacovigilance study and literature review. Clinical Nutrition. 2026 Feb;57:106571. doi: 10.1016/j.clnu.2025.106571. PMID: 41534460.
Was this article helpful?
100% of readers found this helpful (1 votes)
Disclaimer. This article is health journalism intended for general information and education. It is not medical advice and is not a substitute for professional diagnosis or treatment. Always consult a qualified healthcare provider about your individual circumstances. Full disclaimer →
Editorial standards. This article was produced under the GMJ News editorial process, with oversight by the GMJ Editorial Board. Our editorial process. Spotted an error? Contact the editorial team.