Semaglutide and tirzepatide, two GLP 1 based obesity and diabetes drugs targeting the same pathway, show markedly different NAION (non arteritic anterior ischemic optic neuropathy) risk profiles, with EMA mandating label updates for semaglutide (~1 additional case/10,000 person…
Two drugs. Same target. Very different headlines. Semaglutide, sold as Ozempic and Wegovy, has been linked to a rare form of blindness that cost it its EU label and has US regulators debating. Tirzepatide, sold as Mounjaro and Zepbound, was supposed to be the safer option—and in one major analysis, it showed no elevated optic nerve risk. Now a pharmacovigilance analysis has found an emerging signal with tirzepatide too, a complication that arrives just as patients and doctors are trying to weigh real risks against real benefits.
The optic nerve condition itself is non-arteritic anterior ischemic optic neuropathy—sudden, painless, usually permanent vision loss from interrupted blood flow to the optic nerve. It occurs in a small number of people without any drug at all. For semaglutide, the European Medicines Agency found approximately one additional case per 10,000 person-years of treatment and required a label update. The FDA has not followed suit.
The proposed explanation for why tirzepatide looked safer hinges on receptor biology. In a head-to-head analysis reported by MedPage Today, tirzepatide showed no elevated NAION signal while semaglutide did, with a dose-dependent pattern—higher-dose Wegovy riskier than lower-dose Ozempic. A large JAMA Ophthalmology study found patients with type 2 diabetes on semaglutide were roughly four times more likely to develop NAION than patients on other diabetes drugs; patients who were overweight or obese faced roughly seven times the risk. The drug activates not just GLP-1, the gut hormone targeted by semaglutide, but also a second receptor called GIP. Researchers have hypothesized that dual receptor activation may stabilize blood flow and buffer the rapid metabolic and fluid shifts that could stress the optic nerve’s blood supply. That hypothesis now has a complicating data point attached to it: a separate Hospital Pharmacy pharmacovigilance analysis from early 2026 identified 28 reported cases of ischemic optic neuropathy in tirzepatide patients, with a ROR of 2.60—which the authors noted may reflect lower reporting rates or a shorter observation window rather than a genuine elevated risk.
What is not complicated is the discontinuation arithmetic. Washington University researchers reported in March 2026 that patients who stopped GLP-1 therapy had substantially elevated cardiovascular risk compared with those who stayed on treatment—risk rising as much as 22% after two years off the drugs. Clinical trial data from Lancet EClinicalMedicine00614-5/fulltext) show most patients who stop regain more than half the weight they lost within a year, and the heart protection fades with it. Roughly half of patients prescribed these drugs stop within the first year anyway, often because of gastrointestinal side effects or insurance coverage gaps, according to a medRxiv preprint.
On a recent Moonshots podcast hosted by Peter Diamandis, Harvard longevity researcher David Sinclair said “people waking up blind because of it” and put the figure at “about 20 to 30,000 people in the US a year.” That appears to conflate total US NAION cases with drug-attributable excess. Background incidence in the general population and in patients with diabetes is already substantial; the drug-linked excess estimated by European regulators is far smaller.
For patients holding these drugs in their hands, the choice is between documented rare harm and documented common harm. The blindness risk is real but affects roughly one in 10,000 patients per year. Stopping carries consequences that are more common and have been measured at population scale in hundreds of thousands of patients. Neither is a neutral option. Patients on Ozempic, Wegovy, Mounjaro, or Zepbound should talk to a clinician before stopping—and should not ignore new visual symptoms. The FDA has not added a NAION warning to US semaglutide labels. In March 2026, it sent Novo Nordisk a warning letter over deficiencies in adverse event reporting—a separate concern, but not a reassuring backdrop for a safety debate that still lacks settled science.