This Week in Ophthalmology — Oct 4, 2026

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 monitoring and surgery, the retina from macular degeneration to diabetic vitrectomy, and the inflammatory and ischaemic disease that sits between ophthalmology and the rest of medicine. Let's dive in.

Glaucoma leads the week, with three papers asking how we measure risk and how much surgery is enough. In Ophthalmology, Pizarro Peñaranda and colleagues looked back at 94 treated patients, 150 eyes, who had been prescribed home tonometry for at least a week. Mean pressure sat around 13 to 14 millimetres of mercury whether measured at home or in clinic, and none of the standard metrics, including mean, peak, range or fluctuation, tracked with visual field decline. What did track with faster progression were new threshold-based measures from home readings, namely how often and by how much pressure exceeded the target the glaucoma specialist had set for that patient. The same measures taken from office readings showed no association [1]. This is a retrospective cohort of modest size, so it points toward an idea rather than proving it: that pressure burden above an individual target, captured outside the clinic, may carry information that the office snapshot misses. The authors call for prospective validation before it shapes monitoring. Complementing that, the British Journal of Ophthalmology carries a ten-year Japanese multicentre study from Nishijima and colleagues, following 127 eyes with early central field defects on 10-2 testing for close to thirteen years. Just over four in ten eyes progressed. Eyes with inferior defects, or with defects spanning both hemifields at baseline, had a bit more than three times the hazard of progression, and specific archetypal patterns, an inferonasal inferior pattern and a nasal both-hemifield pattern, also predicted decline [2]. Again retrospective, but the long follow-up lends weight to the idea that the baseline location of a central defect may help identify eyes that warrant closer central surveillance. On the surgical side, also in the British Journal of Ophthalmology, Wang and colleagues ran a prospective multicentre comparison across 19 Chinese centres of goniotomy arc size in juvenile open-angle glaucoma, 126 eyes with a mean age around 23. At one year, complete success without medication was similar for 240 and 360 degree incisions, at roughly 55 and 65 percent, and both clearly outperformed a 120 degree arc, where only about one in eight eyes succeeded without drops. With medications allowed, the three groups were statistically indistinguishable. Hyphema was common and notably more frequent with the full 360 degree incision, affecting about two thirds of those eyes versus about a third in the smaller-arc groups [3]. Arc size was not randomised, so the authors frame 240 degrees as a promising balance of efficacy and simplicity that needs a randomised trial to confirm.

The retina section is the busiest this week, beginning with a potential drug safety signal. In Ophthalmology, Xue and colleagues used drug-target Mendelian randomization in more than 450,000 UK Biobank participants and about 7,000 people in the Singapore Epidemiology of Eye Diseases study to ask whether inhibiting cholesteryl ester transfer protein, the target of a cardiovascular drug class, might raise macular degeneration risk. Greater genetically proxied inhibition was associated with roughly a 45 percent higher odds of advanced AMD in Europeans, and in East Asians with roughly tripled odds of exudative disease, including both typical neovascular AMD and polypoidal choroidal vasculopathy. The observational genetic score in the UK Biobank showed only a small increase in incident AMD, and the Singapore data traced a U-shaped curve rather than a straight line [4]. Genetic proxies are not drug exposure, and the authors are careful to call this a signal requiring validation in trials and clinical cohorts, but it is consistent across two ancestries and worth watching as these agents return to cardiology. Turning to how future AMD therapies will be tested, Progress in Retinal and Eye Research publishes a roadmap from Berni, Rosenfeld and colleagues on intermediate AMD. The core problem is that intermediate disease can sit stable for five years or more, while a trial needs an answer in about two. Their proposal uses a single OCT acquisition both to enrich for high-risk eyes, drawing on drusen volume, hyperreflective foci, calcified drusen, reticular pseudodrusen and photoreceptor band integrity, and to supply the endpoint, using the onset and growth of large hypertransmission defects and outer retinal band loss, paired with microperimetry [5]. Also in Progress in Retinal and Eye Research, Jin and colleagues review artificial intelligence across retinal decision-making and land on a sober conclusion: the evidence is strongest for diabetic retinopathy screening, which has prospective and real-world validation, while tools for treatment decisions in diabetic macular oedema and neovascular AMD, and for inherited retinal disease, remain less consistently validated. They highlight a persistent gap between model performance and clinical utility, emphasising external validation, calibration and prospective evaluation [6].

Two more retina papers address surgery and structure in the diabetic and myopic eye. In Ophthalmology Retina, Butt and colleagues pooled 33 studies, about 3,400 eyes and 12 randomised trials, on preoperative anti-VEGF before vitrectomy for proliferative diabetic retinopathy. Anti-VEGF was associated with less than half the odds of early postoperative vitreous haemorrhage, a fivefold reduction in the odds of intraoperative bleeding, fewer reoperations and less need for endodiathermy, with no difference in late haemorrhage, retinal detachment, silicone oil use or tractional detachment [7]. The timing question is less settled. Haemorrhage rates were lowest when injection preceded surgery by one to seven days, but when the interval was modelled continuously it showed no association with bleeding, and the signal of worse outcomes beyond two weeks comes from a single cohort. Certainty was moderate for early postoperative haemorrhage and low or very low elsewhere, so the benefit of pretreatment is reasonably supported while the optimal interval remains hypothesis-generating. Finally, in Investigative Ophthalmology and Visual Science, Jonas and colleagues used the population-based Beijing Eye Study, more than 3,300 eyes, to describe extrafoveal paravascular retinoschisis. It was rare overall, about one percent, but rose steeply with axial length beyond about 25.5 millimetres, reaching six in ten eyes longer than 28 millimetres. It was strongly linked to dragging of the central retinal vessel trunk and to a longer fovea-to-disc distance, leading the authors to hypothesise that it reflects stretching of the large temporal vessels, a mechanism distinct from myopic foveoschisis [8].

Our last theme crosses into neuro-ophthalmology and orbital disease, where systemic conditions show up in the eye. In the British Journal of Ophthalmology, Hanna and colleagues reviewed OCT from two tertiary practices, comparing 80 patients with biopsy-proven giant cell arteritis against 179 with non-arteritic anterior ischaemic optic neuropathy. Paracentral acute middle maculopathy appeared in 28 percent of the arteritis patients and in none of the non-arteritic patients. It was found in a similar proportion of arteritis patients with and without visual symptoms, persisted after corticosteroids were started, and did not correlate with presenting acuity or field loss [9]. That suggests the sign can reveal subclinical retinal ischaemia even in patients with normal vision, and could help separate arteritic from non-arteritic disease. The data are retrospective and from specialist centres, so how it performs in a general emergency setting is still unknown. In the American Journal of Ophthalmology, Sun and colleagues present a meta-analysis of teprotumumab for thyroid eye disease, combining 6 randomised trials and 30 real-world studies. In trials, proptosis response was about four and a half times more likely with teprotumumab, and in routine practice just over four in five patients achieved it. Diplopia response roughly doubled in trials, with about six in ten real-world patients improving. The safety side is equally consistent: hyperglycaemia was more frequent in trials and occurred in about one in six real-world patients, and hearing impairment was roughly quadrupled in trials and affected nearly a fifth of real-world patients [10]. The authors conclude the efficacy holds across study designs, while the real-world rates of hyperglycaemia and hearing impairment support careful patient selection and monitoring.

If you only have time for one paper this week, make it the British Journal of Ophthalmology study by Hanna and colleagues on paracentral acute middle maculopathy in giant cell arteritis [9]. It reopens the question of what a routine macular OCT can contribute to one of the most time-critical diagnoses in ophthalmology, including in patients who have no visual complaints at all.

Here is what this week's evidence adds up to in Ophthalmology. First, glaucoma risk may be better captured by individualised measures, how far and how often pressure exceeds a patient's own target at home, and where a central defect sits at baseline, though both findings are retrospective and await prospective confirmation. Second, preoperative anti-VEGF before diabetic vitrectomy is now supported by moderate-certainty evidence for less early bleeding, while the ideal injection interval remains unresolved. Third, genetic data raise a cross-ancestry signal that cholesteryl ester transfer protein inhibition could increase macular degeneration risk, a signal that has not yet been tested in drug-exposed patients. Fourth, paracentral acute middle maculopathy appears specific to arteritic rather than non-arteritic ischaemic optic neuropathy in specialist cohorts, with broader validation still needed. And fifth, teprotumumab's benefits for proptosis and diplopia hold up in real-world practice, but so do meaningful rates of hyperglycaemia and hearing impairment.

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 if it saves you time, a quick rating in your podcast app helps other physicians find the show.

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