Recently Andrew Huberman stated the risk of going blind from GLP-1 was roughly the same risk as going blind from masturbation. That is a reassuring soundbite. But I don’t want a soundbite.
The current headlines of people suing GLP-1 manufacturers because they went blind. From this are two sub-headlines: those who say it isn’t a big deal and barely rises above the noise from those who have diabetes and go blind. The second group are the GLP-1 skeptics, who will magnify any negative side effects and say if you buy their coaching plan, you won’t go blind.
There is a real medical condition behind the headlines. Rare as it might be, that isn’t a zero risk. There is also an anatomical predisposition to this effect.
I have taken tirzepatide since October of 2024. Going from over 225 pounds, losing 50 and maintaining, I am happy. It has had a positive impact on my health, from all my labs, it has changed food noise. But would I make that trade if I knew I would go blind? What risk would I accept?
What is this blindness everyone is talking about?
The condition is called nonarteritic anterior ischemic optic neuropathy, mercifully abbreviated NAION. It is an injury to the optic nerve caused by inadequate blood supply to the portion of the nerve where it enters the back of the eye. Don’t worry - let me explain.
The optic nerve is the cable that carries visual information from the eye to the brain. The fiber for the internet that sends information. But that fiber is a living tissue and needs a blood supply to keep it alive. If the blood supply is lost, you go blind, that fiber is frayed and cannot transmit information.
If the blood flow to the nerve is decreased, you may damage the nerve to lose some sight. Meaning, you don’t have perfect vision, like having a bad cable, you don’t have great internet. You may lose some vision on the side of the eye, or in the center, or things can get fuzzy. The loss of sight is usually described as sudden and painless loss.
Most often, this happens in one eye, and the amount of vision lost varies. This condition does not mean a person becomes completely blind. Often that is not the case.
This is not the same as diabetic retinopathy, where diabetes damages the blood vessels of the retina. The retina is at the back of the eye. It is what converts light into the electrical signals. Think of it as the keyboard. Diabetic retinopathy is like having a bad keyboard, where you can lose some keys or all of them, or they don’t work as well. So retinopathy is a different condition.
NAION has been known about for a long time, long before GLP-1 medications. NAION is like a heart attack to the arteries that feed the optic nerve. Anything that can cause damage to a blood supply can cause this condition. Mostly diabetes, but also high blood pressure, sleep apnea, high cholesterol, etc.
This is part of what makes the current controversy difficult to sort out. Many people taking GLP-1 medications already have some of the conditions associated with NAION. If someone taking Ozempic develops this problem, did the medication cause it, contribute to it, or was it something that would have happened anyway? This is not a question to dismiss based on someone with pre-existing conditions. The question is more, what is the additional risk of partial or complete visual loss added by GLP-1 medications?
The anatomy of the problem
Andrew Huberman’s says there is a specific anatomy (a crowded optic disc) that leads to NAION. While Huberman is not a physician, he happens to be a specialist in the nerves in the eye. So here is what he is talking about.
Where the optic nerve goes from the eye into the brain is called the optic disc. Think of it as all the cables coming together, forming that nerve and bringing that information to the brain.
If blood flow to the optic nerve becomes decreases, swelling can develop within that confined space. Anytime you get a loss of blood flow swelling takes place. Ever put a rubber band around a finger, and you see it swell. That is what happens if there is blood loss in this area. When the area swells, the nerve gets damaged. Just as if you left the rubber band long enough on the finger, the finger would become damaged. The difference is with the finger, you have nerves that tell you it is painful. In the eye, you don’t have those pain fiber nerves, which is why this is painless and you don’t see it coming (sorry).
Some people have a smaller area, and they are far more susceptible to NAION. But the real question is what caused the original interruption in blood supply.
A person can have a crowded optic disc for decades without developing NAION. In fact, most people with a crowded disc never experience the condition.
We see this throughout medicine. Having a predisposition to a problem does not mean you will get the problem. You can smoke for years and never get lung cancer or heart disease. A person may have coronary artery disease for years, and never have a heart attack. You can have a narrow spinal canal (spinal stenosis) and never have back pain (go yoga).
What did the regulators find?
In June 2025, the European Medicines Agency reviewed the evidence linking semaglutide to NAION. They concluded that NAION should be recognized as “a very rare adverse effect of semaglutide.” They state this includes Ozempic, Wegovy, and Rybelsus.
The agency estimated approximately one additional case of NAION for every 10,000 people treated with GLP-1 for one year. One in ten thousand. But it is not zero.
The agency also reported that people with type 2 diabetes had approximately twice the risk of NAION in people taking semaglutide compared with those who were not. That makes sense when you realize that diabetes damages arteries far more than any other underlying condition.
A doubling of a rare event is still a rare event. But rare is not zero. And if you suffer from partial or complete blindness, it is hardly comfort. Just like if a shark bites you, a rare event, you still have an injury (this is why I scuba), I have a pathologic fear of sharks, so I wanted to overcome it).
The European regulators recommended that semaglutide’s prescribing information include NAION as an adverse effect. I would agree, informed consent requires this.
What about Zepbound?
There is one more distinction before we get into the studies.
Semaglutide and tirzepatide (Zepbound) are not the same drug.
The European regulatory determination specifically concerns semaglutide. We cannot automatically apply the same risk to every medication that acts on the GLP-1 receptor. It may not be the GLP-1 receptor, but we simply don’t know. Yes, it seems logical, but as we have seen in medicine before, what seems logical does not work when applied to medicine. We have to see. Not that human beings aren’t logical, we are not, but lots of great theories and models don’t automatically apply.
So what do we do? That is one of those deeply personal reflections and questions you have to ask yourself. I have my vision checked yearly by an optometrist for my glasses. This year, an ophthalmologist will check it for a more in-depth look at my retina and optic disc. But to be clear - having a non-crowded optic disc does not guarantee you won’t have an issue -here is where Huberman was wrong.
I will go more into detail in the paid section below.
