Beyond the basics: John Berdahl, MD, on AI-driven IOL planning, PCO prevention, and lessons from the mission field

In the second installment of a 2-part conversation, Ophthalmology Times continues its discussion with John Berdahl, MD, a board-certified ophthalmologist at Vance Thompson Vision in Sioux Falls, South Dakota, and a professor at the University of South Dakota. Where part 1 examined the global cataract surgical gap and emerging dementia data, part 2 turns to the decisions surgeons face in the OR—how machine learning–based IOL formulas compare with vergence-based approaches in challenging eyes, which material properties and edge designs most reliably reduce posterior capsule opacification rates, and how surgical mission work in low-resource environments has shaped Dr. Berdahl's technique and philosophy at home.

This is part 2 of a 2-part conversation with Berdahl, read part 1 here.

This interview has been edited for length and clarity.

How are machine learning–based IOL performing relative to established vergence-based approaches—particularly in challenging eyes such as those with prior refractive surgery, high axial myopia, or irregular corneas? And from your perspective, what are the outstanding scientific, methodologic, or standardization gaps that must be addressed before AI-assisted IOL planning can be considered the unambiguous standard of care?

Berdahl: In normal eyes the AI formulas have basically caught the whole field up to a ceiling it seems we are all bumping into. In the hard eyes, post-refractive, extreme myopia, irregular corneas, they pull modestly ahead.

The truth about normal eyes is that leading formulas are all quite good now. In an elderly cataract population, predictability within ±0.5 D was about 72% for both Kane and Barrett Universal II, with no significant difference between them.¹ The modern AI formulas (Kane, Hill-RBF) and the best analytic vergence formula (Barrett) are similar in routine eyes. The pioneers of formulas have done us all an incredible service.Predicting how the eye is going to heal, especially astigmatically, remains a major limiting factor for normal eyes.

Where AI separates itself is in the tails. In extreme axial myopia, in eyes 32 mm and longer, Kane had the lowest mean and median absolute error, 0.44 D and 0.40 D, and both Kane and Hill-RBF had far fewer big misses than SRK/T—refractive errors beyond ±1.0 D in 7.5% of eyes versus 42.5% for SRK/T.² Think about that—a 42% large-miss rate with an old vergence formula in a long eye is basically a refractive-surprise factory. That's the real case for AI. Not that it wins the average eye, but that it shrinks the disasters.

On the post-refractive question specifically, which is the one that matters most for my spectacle-independence patients, the field still leans on dedicated approaches like Barrett True-K and the ASCRS calculator, with the AI methods earning a seat at the table but not yet owning the room.

My read: AI formulas are already first-line reasonable and especially help to avoid outlier outcomes.

With posterior capsule opacification (PCO) remaining the most common long-term complication of cataract surgery, what IOL material properties, optic edge designs, or pharmacologic adjuncts show the most promise for meaningfully reducing Nd:YAG capsulotomy rates?

Berdahl: The single biggest lever is one we've had for 20 years, a square optic edge followed by hydrophobic material. The pharmacology is interesting but not yet practice-changing.

The square edge is the heavyweight, and it's not close. The mechanism is mechanical: a sharp posterior edge creates a discontinuous bend in the capsule that physically blocks lens epithelial cells from migrating across the posterior capsule. The long-term data are emphatic. In a 9-year randomized fellow-eye study, square-edge IOLs had a 2% YAG rate versus 37% for round-edge lenses of the same PMMA material. Same material, same surgeon, same patient's two eyes. The only variable was the edge, and it cut YAG dramatically.³

Material is the second lever. Pooled across trials, hydrophilic acrylic lenses had roughly a 7-fold higher YAG capsulotomy rate than hydrophobic acrylic at 2 years, RR 6.96.⁴ To put a number on it, one clinic reported a 17.7% two-year capsulotomy rate for hydrophilic IOLs.⁵

The newer wrinkle is that not all hydrophobic, square-edge lenses are created equal. A network meta-analysis found AcrySof had a lower YAG risk than other hydrophobic acrylics, with sharp-edged hydrophilic acrylic carrying about a 9-fold higher hazard, and a modeled 3-year YAG rate of 4.19% for AcrySof versus 1.82% for the newer Clareon material.⁶

The other real contributors: a well-centered, appropriately sized capsulorhexis with 360-degree overlap on the optic, thorough cortical cleanup, good capsule polishing, and a posterior square edge that runs the full optic, not just the center.

On drugs, nothing has cleared the bar of being safe to the rest of the eye while reliably ablating epithelial cells. The risk-benefit of YAG is low, because PCO is already treatable with a 60-second laser, therefor any preventive drug must be extraordinarily safe to earn its place.

What drew you to surgical mission work, and how has operating in low-resource environments shaped your approach to cataract surgery back in your own practice?

Berdahl: Mission work changes my heart, and I complain less for about 6 months after I return. So, I should probably go about every six months. At my core, I am a teacher. Teaching residents overseas, and helping those with the least among us fills my cup in an unselfish way that other daily efforts don’t.

There's almost nothing else in medicine where you take someone from blind to seeing in 15 minutes, for a few hundred dollars, and hand them back their independence, their livelihood, often their place in their own family and a caretaker. Cataract surgery is one of the most cost-effective interventions in all of healthcare.⁷ Standing in a recovery line the morning after, watching people see their grandkids for the first time in years, you feel that statistic in your chest, not your head. My faith is lived through service, and this is service with an unusually clear benefit.

How it changed my practice at home, concretely:

1. It made me question what actually matters. When you don't have the phaco machine, the premium IOL, sterilize a whole OR between each patient, or even reliable power, you find out fast which 10% of your tools deliver 90% of the outcome. You come home and realize how much of what we do is refinement at the margins. That sharpens your judgment about when a fancier technology is truly worth it and when it's just expensive.
2. It made me a better, faster, more deliberate and creative surgeon.
3. It recalibrated gratitude. It's genuinely hard to be annoyed about an EHR click after a week operating by headlamp.
4. The biggest thing it changes is the heart. Yes, it helps the individuals who got surgery, and that matters and teaching residents pays dividends in those communities for years and is the sustainable answer. At the end of the day, we are humans, and serving each other lightens the heart.

References
1. Reitblat O, Gali HE, Chou L, et al. Intraocular lens power calculation in the elderly population using the Kane formula in comparison with existing methods. J Cataract Refract Surg. 2020;46(11):1501-1507. doi:10.1097/j.jcrs.0000000000000308
2. Suzuki Y, Kamoi K, Uramoto K, Ohno-Matsui K. Artificial intelligence driven intraocular lens power calculation in extreme axial myopia. Sci Rep. 2025;15:36921. doi:10.1038/s41598-025-20899-6
3. Haripriya A, Chang DF, Vijayakumar B, et al. Long-term posterior capsule opacification reduction with square-edge polymethylmethacrylate intraocular lens: randomized controlled study. Ophthalmology. 2017;124(3):295-302. doi:10.1016/j.ophtha.2016.10.022
4. Li Y, Wang J, Chen Z, Tang X. Effect of hydrophobic acrylic versus hydrophilic acrylic intraocular lens on posterior capsule opacification: meta-analysis. PLoS One. 2013;8(11):e77864. doi:10.1371/journal.pone.0077864
5. Von Tress M, Marotta JS, Lane SS, Sarangapani R. A meta-analysis of Nd:YAG capsulotomy rates for two hydrophobic intraocular lens materials. Clin Ophthalmol. 2018;12:1125-1135. doi:10.2147/OPTH.S159840
6. Thom H, Ender F, Samavedam S, et al. Effect of AcrySof versus other intraocular lens properties on the risk of Nd:YAG capsulotomy after cataract surgery: a systematic literature review and network meta-analysis. PLoS One. 2019;14(8):e0220498. doi:10.1371/journal.pone.0220498
7. GBD Vision Loss Expert Group. "Global estimates on the number of people blind or visually impaired by cataract: a meta-analysis from 2000 to 2020." *Eye (London)*, 2024;38(11):2156-2172. (2020 blindness/MSVI figures; rising absolute burden despite falling age-standardized prevalence) https://www.nature.com/articles/s41433-024-02961-1