This week in Ophthalmology

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

Welcome to This Week in Ophthalmology. This week we're covering 10 notable papers spanning glaucoma from the nursery to the screening clinic, the safety and prognosis of retinal injections and surgery, and drug-related and HLA-B27 ocular inflammation. Let's dive in.

We start with glaucoma, where this week's papers range from surgical choice in infants to who might be flagged years before a field defect appears. The most rigorous of the set is a randomized noninferiority trial in Ophthalmology from Fang and colleagues, comparing trabeculotomy alone with combined trabeculotomy-trabeculectomy in children up to three years old with primary congenital glaucoma and a horizontal corneal diameter of 12 to 14 millimetres [1]. Among 248 children, each contributing one eye, qualified success at three years, meaning a pressure between 5 and 21 with or without drops and no vision-threatening complication or further surgery, was reached in about 82 percent of the trabeculotomy eyes and about 85 percent of the combined-procedure eyes. That small gap sat comfortably within the prespecified margin, so trabeculotomy was judged noninferior, and the per-protocol analysis told the same story. Pressure reduction was similar, at roughly 12 to 13 millimetres of mercury in both arms, and adverse events, mostly self-resolving hyphema, were numerically less frequent with trabeculotomy alone, though not significantly so. The authors frame trabeculotomy as a reasonable alternative in this specific corneal size band; the trial does not speak to larger corneas or previously operated eyes.

Two papers address risk stratification in adults. In the American Journal of Ophthalmology, Gobira and colleagues followed 330 eyes with ocular hypertension for a median of just over four years, during which 14 percent of eyes developed repeatable visual field damage [2]. The established OHTS-EGPS risk calculator discriminated only modestly on its own. Thinner baseline macular ganglion cell-inner plexiform layer roughly doubled the hazard of conversion per standard deviation of thinning, and adding it to the calculator meaningfully improved discrimination, whereas global peripapillary nerve fibre layer thickness added essentially nothing. It is retrospective and relatively small, but it strengthens the case for macular imaging as part of baseline risk assessment in ocular hypertension. Pressure characterization itself is the subject of a paper in Ophthalmology from Jensen and colleagues, who modelled home rebound tonometry data from about 400 academic patients and validated it in nearly 1,600 more from a national database [3]. Roughly six to seven days of multiple daily readings captured most of the achievable accuracy for mean, peak, range and variability, but eyes with primary open-angle glaucoma needed nine to ten days, and limiting patients to three readings a day stretched the requirement to around ten days or more. Notably, about three quarters of peak pressures occurred outside typical clinic hours. The authors conclude that seven to ten consecutive days is the window needed to represent a patient's true pattern, which gives structure to home monitoring protocols, though outcome benefit was not tested.

Finally in this theme, van Meyel and colleagues, also in Ophthalmology, externally validated the winning algorithm from the AIROGS glaucoma screening challenge in the population-based Rotterdam Study, nearly 10,000 participants aged 50 to 80 [4]. At a fixed 95 percent specificity, it caught about 70 percent of perimetric glaucoma eyes, with an area under the curve of about 0.91, but because glaucoma was rare in this population, around five referrals were generated for each true case found. Missed cases tended to have milder field loss. Among detected cases with earlier images, just over half of those patients had been flagged before diagnosis, typically four to seven years ahead. Performance held on disc-centred photos despite training on macula-centred images, and poor optic nerve head visibility drove many false positives. The authors see a scalable triage tool, with image acquisition standards as the practical hinge.

Our second theme is retina, specifically the safety of what we inject and what follows vitreoretinal surgery, plus a new candidate prognostic marker. In Survey of Ophthalmology, Leung and colleagues pooled 14 comparative studies covering 6.2 million anti-VEGF injections and found that prefilled syringes were associated with roughly a third lower odds of endophthalmitis than vial-drawn syringes, consistently for ranibizumab and aflibercept, and with culture-positive cases cut even further [5]. In absolute terms, that is about one fewer case per 10,000 injections. The result survived adjustment for calendar era and small-study effects, though the data remain observational and residual confounding cannot be excluded.

Safety of newer combinations is examined in Ophthalmology Retina by Esmaeilkhanian and colleagues, using a United States retina-practice record database of about 55,000 eyes and over a million injections [6]. Eyes receiving both anti-VEGF and anti-complement therapy did not show higher overall rates of serious adverse events than monotherapy, and endophthalmitis ran at about one per 10,000 injections overall. Same-day dual injections showed higher per-injection rates, but these rested on a single endophthalmitis and a single retinitis case, too few to interpret. The authors call for larger prospective data, and the dual-therapy cohort was small, so this is reassurance rather than settled safety.

From the operating room, Talaparthy and colleagues in Retina used the TriNetX network to compare about 26,000 matched vitrectomy patients on systemic immunosuppression with controls [7]. No drug class raised endophthalmitis risk in the first six weeks, but between six weeks and a year, risk was roughly 75 percent higher with any immunosuppressant, with corticosteroids and biologics standing out. The authors note this timing is atypical for surgically acquired infection and that some events may be endogenous or inflammatory, so the finding argues for a longer window of suspicion rather than a change in perioperative technique, with the caveats of claims-based data.

Rounding out the retina theme, Ferreira and colleagues, again in Survey of Ophthalmology, meta-analysed 17 studies of nearly 10,000 eyes treated for neovascular macular degeneration [8]. Eyes with the greatest fluctuation in fluid or retinal thickness gained roughly five to ten fewer letters at two years than the most stable eyes, across every compartment examined, and had about twice the odds of fibrosis and atrophy. Certainty of evidence was low to very low, largely because fluctuation lacks a standard definition, and the authors are explicit that prospective testing with a common metric is required before stability becomes a treatment target.

Our final theme is ocular inflammation, with one paper on a drug trigger and one on treatment intensity. In Ophthalmology, a nationwide Taiwanese case-case-time-control study using over half a million incident cases of uveitis, scleritis and cystoid macular edema examined short-term bisphosphonate exposure [9]. Any bisphosphonate use was associated with a modest rise in uveitis risk, driven by anterior uveitis, but the signal concentrated in intravenous agents, and zoledronic acid in particular was associated with nearly tripled odds. Oral bisphosphonates showed no association, and no clear link emerged for scleritis or macular edema. Negative-control drugs were null, which supports the specificity of the signal, and the authors suggest it may inform monitoring of ocular symptoms after intravenous dosing.

In the American Journal of Ophthalmology, Ruiz-Lozano and colleagues reviewed 462 patients with HLA-B27-associated acute anterior uveitis at Bascom Palmer [10]. Adding oral prednisone to topical steroids shortened the median time to a quiet anterior chamber from about 41 days to 28, roughly doubling the rate of achieving control after adjustment, yet total time on medication was essentially unchanged at around two months. Prednisone courses beyond 14 days offered no added benefit, and more intensive initial topical dosing, eight or more drops a day, was linked to earlier discontinuation. The authors acknowledge that oral steroids were probably reserved for more severe presentations, so this single-centre retrospective evidence suggests faster control rather than proving a better overall course.

If you only have time for one paper this week, make it the congenital glaucoma trial from Fang and colleagues in Ophthalmology [1]. It is a rare randomized trial in paediatric glaucoma surgery, and it provides three-year evidence that the simpler procedure holds its own in children with moderately enlarged corneas, a question previously answered largely by surgeon preference.

That's your roundup for This Week in Ophthalmology. The full transcript and references are available on the episode page. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if you have a paper you have been meaning to read, upload the PDF, or paste any link, at audioscholar dot C C. We will turn it into audio like this one, in any of thirty-one languages.

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