Brian Reed: Hey — you caught the Fox News piece this morning? August 23rd, Cialis, 22% higher glaucoma risk.
Eliza Ward: Yeah. And my first read was — wait, is that the whole story? Because 22% sounds like a lot until you ask, compared to what.
Brian Reed: Right — and the framing was pretty much immediate alarm. Current users, safety concern, the whole thing.
Eliza Ward: Okay but glaucoma aside, the condition that actually stopped me — they mention NAION. Non-arteritic anterior ischemic optic neuropathy. Which is — it cuts off blood flow to the optic nerve, and the vision loss can be permanent.
Brian Reed: Permanent. Not blurry — gone.
Eliza Ward: That's the one that's been in the literature longest for this drug class — PDE5 inhibitors, Cialis, Viagra, that whole group.
Brian Reed: So the Fox report is really the entry point here — but the NAION finding is older and, honestly, potentially scarier than the glaucoma number.
Eliza Ward: That's the part I want to pull on — yeah.
Brian Reed: But here's where the framing falls apart — the 22% glaucoma figure and the 85% combined risk figure aren't the same study. They're not even measuring the same thing.
Eliza Ward: Right. And that's — okay, think of it this way. Two different studies, same neighborhood, different houses. You can't add them together and call it one number.
Brian Reed: Like adding apples and fire trucks.
Eliza Ward: Exactly. The University of British Columbia study — Dr. Etminan's work — that's the 85% figure. But it's covering three bundled conditions: serous retinal detachment, ischemic optic neuropathy, retinal vein occlusion. Combined. Adjusted IRR of 1.85, confidence interval 1.41 to 2.42. That's actually a fairly tight interval, which — wait, that's the part that's harder to wave away.
Brian Reed: So the UBC number is more robust — but it's three different conditions bundled. The glaucoma figure is separate, and we're not even sure the study Fox cited is actually new research.
Eliza Ward: That's the honest answer — we don't know the publication date. It might be new, it might be a re-amplification of something older. I want to be clear: that's a genuine uncertainty, not a rhetorical hedge.
Brian Reed: The American Academy of Ophthalmology has published guidance on exactly this question — ED medications and vision risk. So this isn't fringe territory. It's just... the headline merged two separate signals into one alarm.
Eliza Ward: Which covers sildenafil, tadalafil, vardenafil, avanafil — the whole PDE5 class. The UBC study looked at all of them. The glaucoma finding is tadalafil-specific. Those are different scopes.
Brian Reed: So the actual signal — what's genuinely supported — is that regular PDE5 inhibitor use is associated with a higher risk of serious eye events. The specific number depends entirely on which study, which drug, which condition you're talking about.
Eliza Ward: But that association — that's the part that's doing a lot of work in the circulating take. 'Cialis causes blindness' is essentially what's moving on social. And the confounding problem is... I mean, it's not a footnote. It's potentially the whole story.
Brian Reed: That's the take I want to name directly. Men prescribed ED medications — tadalafil, sildenafil — are disproportionately the men who already have hypertension, type 2 diabetes, cardiovascular disease. The same conditions that independently elevate your risk for NAION and glaucoma.
Eliza Ward: Right. So you can't — wait, you can't just look at who got the drug and who got eye disease. You have to ask: would those men have been at higher risk anyway?
Brian Reed: And the FDA actually said this out loud in 2005. Their own statement — when they required the label updates on Viagra, Cialis, and Levitra — was that they could not determine whether the drugs caused the vision loss, or whether underlying vascular risk factors were responsible. That's the agency's own hedge, on the record.
Eliza Ward: Hold on — that's the Bayer pilot study territory, right? Where the overall odds ratio was 1.81 but the confidence interval ran from 0.51 to 6.37. That interval straddles 1.0. Statistically that's basically — you haven't cleared the bar.
Brian Reed: Mostly. Except — and that's uncomfortable — in men with a prior myocardial infarction, that same study puts the odds ratio at 10.7. Confidence interval 1.3 to 95.8. So the overall picture is murky, but the heart-attack-history subgroup is a very different number.
Eliza Ward: 10.7. That's — okay, that's not noise.
Brian Reed: No. But here's where I want to be honest: that subgroup is small, the interval is enormous — 1.3 to 95.8 — which means very wide uncertainty. It might be real, it might be a small-sample artifact. We don't know. And that's actually the problem. Picture a man, late 60s, type 2 diabetes, high blood pressure, picks up his tadalafil refill, reads the warning label — that label was written in 2005, before the UBC study existed. It tells him nothing about whether his specific risk profile changes the math.
Eliza Ward: And that's what we haven't touched yet — the label hasn't moved in twenty years despite everything that's accumulated since. That's the part we need to get into.
Brian Reed: And that's the thing about the label — it's not that the FDA missed the signal. They caught it. May 18, 2005, they had 43 adverse event reports. Thirty-eight on sildenafil alone. They required updates on Viagra, Cialis, Levitra that same year. That's an action. And then — nothing on ocular risk since.
Eliza Ward: Nothing. Well — one move. 2007, they updated the labels again. But that was for sudden sensorineural hearing loss. Not eyes. So the ocular warning is still sitting at 2005 language.
Brian Reed: While the UBC study, the glaucoma research, all of it — accumulates after that.
Eliza Ward: Right — and Health Canada noticed. UK regulators noticed. But there's no coordinated international update. Everyone's clocked the same signal, nobody's moved together.
Brian Reed: So what does a man with — let me think about this concretely — a man with a history of high blood pressure, refilling his tadalafil prescription, actually read on that label right now?
Eliza Ward: He reads a 2005 warning that says, essentially — we noticed some cases of vision loss, we can't confirm the drug caused it. That's it. The FDA said that explicitly when they required the update: they could not determine whether the drug caused the harm or whether underlying factors did. That epistemic humility is still baked into the label. Which makes the inaction harder to read — it's not that they decided the risk was zero. They decided they didn't know, and then... stopped deciding.
Brian Reed: Stopped deciding. That's — yeah, I don't know what to call that except administrative drift.
Eliza Ward: So — watch the FDA. Does the Fox report, the renewed attention, actually produce a formal review request? Because the question isn't whether the agency updates the label to say 'danger.' It's whether updated guidance gives a prescribing doctor something actionable — like, for patients with prior cardiovascular events, flag this specifically. That's the version of a label update that actually changes a clinical conversation.
Brian Reed: And the version that doesn't — is just alarming language that prompts people to stop a medication without any framework for what to actually do instead. Which, given that the American Academy of Ophthalmology already has guidance on this, isn't a hypothetical failure mode. It's the one that happens when alarm outruns instruction.
Eliza Ward: And that's — I mean, that's actually the live question now. Not whether the FDA acts. Whether the form of action matters. Because there's a version of updated guidance that says 'men with prior myocardial infarction, that 10.7 odds ratio, flag them, monitor them specifically' — and there's a version that just says 'danger, vision loss possible' and every current tadalafil user panics and stops their prescription cold.
Brian Reed: And stopping abruptly — that's not a neutral act. There are risks to discontinuing a prescribed PDE5 inhibitor without physician guidance. That's not hypothetical, that's in the evidence. So the alarm-everyone label doesn't just fail to help the MI subgroup — it potentially harms the people who don't need to stop.
Eliza Ward: Right. So the diabetics, the hypertensives — they're the ones who actually need the conversation with their doctor. Not a label that treats a 45-year-old with no cardiovascular history the same as someone with a prior heart attack.
Brian Reed: And we genuinely don't know if the current research is strong enough to force that distinction. The UBC finding is real — adjusted IRR of 1.85, reasonably tight interval. But is that enough for the FDA to move past 2005 language and actually stratify guidance by risk group? I don't have an answer to that.
Eliza Ward: Neither do I. That's — wait, that's actually where I want to leave it. Because the honest current state is: the signal exists, the high-risk subgroup is identifiable, and the regulatory machinery hasn't demonstrated it can tell those two versions of a response apart.
