This Week in Ophthalmology — Aug 29, 2026

Welcome to This Week in Ophthalmology. This week we're covering 10 notable papers spanning macular disease and the value of treating early, the surgical side of cataract and glaucoma care including the hardware we hold in our hands, and paediatric eye care from myopia control to retinopathy of prematurity. Let's dive in.

We start with the macula, where two papers converge on the theme of timing and risk. In the American Journal of Ophthalmology, Dhoot and colleagues report the pooled efficacy and safety of early versus delayed pegcetacoplan for geographic atrophy, using up to 48 months of continuous treatment across the OAKS and DERBY trials and the GALE open-label extension. Over four years, continuous treatment slowed the growth of atrophy by up to about a quarter relative to projected sham, which the authors translate into just under two square millimetres of retinal tissue preserved. The more striking signal is in the eyes that started earlier: among eyes with non-subfoveal atrophy, those on monthly treatment from the outset preserved roughly three times more retinal tissue than the eyes that crossed over from sham at 24 months. Risk of progression to absolute scotoma across the central loci fell by around a third to just over 40 percent, and safety over four years was consistent with the parent trials. The caveat is important — the comparator is a projected sham curve rather than a concurrent one, because the sham arms crossed over — but the functional and anatomic direction of travel supports the argument that delay costs tissue you cannot get back. Alongside that, in Retina, report 10 from the PRECISE study by Kazantzis and colleagues quantifies a complication we all counsel about but rarely have good numbers for. Among nearly 1,930 treatment-naive eyes with neovascular age-related macular degeneration, retinal pigment epithelial tears occurred in about 3.5 percent after the three-injection aflibercept loading phase. Male sex, poorer baseline acuity, thicker central subfield, haemorrhagic and taller pigment epithelial detachments, subretinal hyperreflective material and polypoidal choroidal vasculopathy all raised the odds, while the presence of drusen was protective. Practically, the number to remember is a baseline pigment epithelial detachment height of about 244 microns, which discriminated tear risk with a sensitivity and specificity both around three quarters. That is a concrete threshold for a franker consent conversation before loading a tall vascularised detachment.

Staying with imaging but shifting to inflammation, Ophthalmology Retina publishes the eighteenth report of the Multimodal Imaging in Uveitis taskforce, in which Bansal and colleagues used systematic review plus a formal nominal group consensus process to define what imaging actually earns its place in Behçet disease uveitis. The conclusion is refreshingly deflationary: colour fundus photography, fluorescein angiography and optical coherence tomography, with particular emphasis on ultra-widefield capture, are the modalities that matter. Angiography remains the arbiter of activity through vascular, macular and disc leakage, while tomography characterises cystoid macular oedema, the partial-thickness inner retinitis smudge sign, and subretinal fluid. Autofluorescence, indocyanine green angiography and optical coherence tomography angiography were judged to add little. If you manage Behçet uveitis, this is a framework for standardising what you order and what you document.

The surgical theme this week is uncomfortable in a useful way, because it asks whether our equipment and our service models are silently shaping outcomes. In the British Journal of Ophthalmology, Ahmed and colleagues interrogated the 2024 National Ophthalmology Database cataract audit, covering more than 166,000 operations across 24 National Health Service centres with a mean unadjusted posterior capsule rupture rate of around 1.5 percent. Comparing platforms across some 44,000 procedures, one manufacturer's systems carried roughly 60 percent greater odds of capsule rupture than another's, and flared phaco tips were associated with a rupture rate of about 2.1 percent versus 1.3 percent for straight tips. In a separate eye-level analysis at two tertiary centres, flared tips remained associated with more than double the odds of rupture even after adjusting for surgeon experience. These are unadjusted centre-level comparisons with no correction for case complexity, technique, training or fluidics settings, so this is a hypothesis rather than a verdict — but it is a hypothesis about consumables that deserves a prospective, risk-adjusted answer. The same journal pairs this with a global look at where surgery happens. Arazi and colleagues pooled almost 42,000 cataract-operated eyes from 168 population-based Rapid Assessment of Avoidable Blindness surveys across 61 countries. Outreach camps, which accounted for about one in ten surgeries, delivered presenting acuity of 6/18 or better roughly six percentage points less often than fixed facilities, a difference that translated to about 40 percent lower odds. Within fixed facilities, non-governmental hospitals performed similarly to government hospitals, while private hospitals did modestly better. Outcomes improved everywhere over two decades, but the gap between settings did not close. Where outreach carries substantial volume, that persistent gap is a real constraint on effective cataract surgical coverage.

Glaucoma this week gives us one risk-stratification paper and one device comparison. Also in the American Journal of Ophthalmology, Bak and colleagues followed more than 17,000 Korean adults with normal baseline discs through a health screening programme for a mean of 13 years. Structural glaucomatous conversion occurred in about 2.3 percent of participants, roughly 180 cases per 100,000 person-years. Beyond the expected contributions of older age and higher intraocular pressure — where each additional millimetre of mercury raised the hazard by about 16 percent — retinal arteriosclerosis visible on a routine fundus photograph was independently associated with close to 50 percent higher hazard of conversion, alongside modest signals from serum uric acid and total cholesterol. That is an argument for reading the vasculature, not just the disc, on screening photographs. On the interventional side, the British Journal of Ophthalmology reports a multicentre retrospective comparison from Jacobs and colleagues of excimer laser trabeculostomy versus trabecular micro-bypass stenting, each combined with phacoemulsification, in 343 eyes across three European centres. At one year, pressure fell by about 21 percent in both arms with no meaningful difference, and overall medication reduction and adverse event rates were comparable. The one point of separation was that a higher proportion of eyes achieved any medication reduction after laser trabeculostomy, roughly 46 percent versus 31 percent. With only 70 percent of eyes reaching one-year follow-up and a retrospective design, this supports laser trabeculostomy as a reasonable alternative rather than a clear winner.

Finally, three paediatric papers. In JAMA Ophthalmology, Zhong and colleagues randomised 120 children aged 6 to 12 in Guangzhou to virtual reality-based digital defocus vision training for 15 minutes a day at home plus single-vision spectacles, or spectacles alone. Over six months, axial elongation was about 0.1 millimetres less in the training group, with a corresponding difference in spherical equivalent of just over 0.2 dioptres, and no training-related adverse events. The authors are appropriately restrained: these are small absolute differences over a short horizon, and whether the effect persists or accumulates is unknown. Treat it as promising and unproven rather than ready for routine recommendation. In the same journal, Wu and colleagues compared HOTV letter-matching against the tumbling E chart in 995 children aged 3 to 15, each tested with both charts in randomised order. Tumbling E systematically underestimated acuity, by about a line in the youngest children and narrowing to under one letter by age 8. Among 183 preschoolers, HOTV cut the referral rate from roughly 31 percent to under 19 percent while simultaneously improving the true positive rate for amblyopia and refractive error and raising specificity from about 0.79 to 0.93. In other words, the easier chart was both more accurate and less wasteful in children under 8. That is an immediately implementable change in any screening programme still using tumbling E. Rounding out the paediatric section, the British Journal of Ophthalmology publishes work from Liu and colleagues on aggressive retinopathy of prematurity across 23 sites in and outside the United States. Of 1,600 eyes with retinopathy, 283 had the aggressive subtype, and these infants had lower birth weight and gestational age. Each additional 100 grams of birth weight cut re-injection risk by about a quarter, and initial treatment with ranibizumab rather than bevacizumab was associated with nearly five times the risk of needing repeat anti-vascular endothelial growth factor treatment. Zone II eyes appeared to need more retreatment than zone I, but that signal did not survive correction for bilaterality. For the smallest infants treated with the shorter-acting agent, plan surveillance accordingly.

If you only have time for one paper this week, make it the pegcetacoplan early-versus-delayed analysis in the American Journal of Ophthalmology [1]. It reframes the geographic atrophy conversation from whether complement inhibition works to what a two-year delay actually costs a patient in preserved retina and preserved central function.

Here are the key takeaways from this week in Ophthalmology. First, in geographic atrophy, timing matters — eyes treated from the outset preserved roughly three times more retinal tissue than those starting two years later, though the comparator was a projected rather than concurrent sham. Second, before loading a treatment-naive neovascular eye, measure the pigment epithelial detachment; above roughly 244 microns, retinal pigment epithelial tear risk is materially higher and consent should reflect it. Third, phaco platform and tip architecture were associated with real differences in posterior capsule rupture in a national audit — provocative, unadjusted, and worth a prospective answer before you change practice. Fourth, cataract outreach camps continue to deliver poorer visual outcomes than fixed facilities globally, and that gap has not narrowed in twenty years. And fifth, for vision screening in children under 8, HOTV beats tumbling E on both accuracy and unnecessary referrals — a low-cost change with immediate benefit.

That's your roundup for This Week in Ophthalmology. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And think of one colleague — in any specialty — who never has time to keep up with the literature. Tell them about AudioScholar: a free ten-minute weekly for every specialty, to listen to in any podcast app, or to read at audioscholar dot C C.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.