Audio link: 

https://open.spotify.com/episode/5oAdDy8eWldXuLzetm88tL?si=bb926d634f604242

In this week’s issue

• In patients with ROP, neurodevelopmental outcomes do not differ between anti-VEGF and laser photocoagulation treatments.
• AS-OCT identifies plateau iris, which carries an 11-fold higher risk of progression to angle closure, making it a potentially useful screening tool.
• Patients presenting with ocular complaints, and those presenting to eye doctors, receive more timely care in GCA than those presenting to their PCP or with headache or systemic symptoms. 
• Depression is subjectively strongly associated with dry eye disease severity, but not associated with objective clinical signs.

Survey for Trainees and Faculty

You are invited to participate in a research study from the University of Michigan Kellogg Eye Center examining how clinical training/experience level influences interpretation and communication of ophthalmic examination findings. Eligible participants include fourth-year medical students interested in ophthalmology, ophthalmology residents, fellows, and attending ophthalmologists. Participation involves completing an anonymous, web-based survey consisting of five questions. The survey is expected to take approximately 5-10 minutes to complete.

If you would like to participate, please access the survey here

Ophthalmology

Neurodevelopmental outcomes of anti-VEGF versus laser photocoagulation surgery for treatment of retinopathy of prematurity

When treating the smallest patients, even a tiny injection can raise big questions. Intravitreal anti-VEGF therapy is effective for retinopathy of prematurity (ROP), but systemic absorption and VEGF suppression have raised concerns about neurodevelopment. This study reviewed 26 comparative studies comparing primary anti-VEGF therapy to laser photocoagulation. Twenty of 26 studies (77%) found no statistically significant difference in neurodevelopmental outcomes between treatments. Four studies reported worse outcomes after bevacizumab, while two reported worse outcomes after laser; however, testing methods, follow-up periods, and adjustment for prematurity-related confounders varied considerably. Among five studies assessing intelligence quotient after age four, none found a significant difference between treatments. Overall, current evidence does not show worse neurodevelopment following anti-VEGF therapy compared with laser, although no study was powered to detect small differences, and longer-term data remain necessary.

JAMA Ophthalmology 

Can imaging predict who will develop angle-closure glaucoma? 

Sometimes the narrowest angle isn't the whole story. Primary angle-closure glaucoma is a leading cause of irreversible blindness worldwide, but not everyone with anatomically narrow angles will progress to disease. Identifying which patients are at the highest risk could help clinicians better target preventive treatment. In this post hoc analysis of the randomized ANA-LIS trial, investigators analyzed 322 eyes from patients with primary angle-closure suspect (PACS) using anterior segment optical coherence tomography and ultrasound biomicroscopy. While only 16 eyes (5%) progressed to primary angle closure, eyes with plateau iris configuration, a structural variation in which the iris remains crowded despite a relatively deep central anterior chamber, had an approximately 11-fold higher risk of progression. Narrower anterior chamber angles and higher baseline intraocular pressure were also independently associated with progression. These findings suggest that baseline anterior segment imaging, particularly assessment for plateau iris, may help identify high-risk patients earlier, leading to preventative treatment.

American Journal of Ophthalmology 

Different symptoms, different timelines: how presenting symptoms shape giant cell arteritis diagnosis

When giant cell arteritis shows up with blurry vision, everyone pays attention. When it shows up with vague symptoms, not so much. Giant cell arteritis (GCA) is an ophthalmic emergency that can cause irreversible vision loss if diagnosis and treatment are delayed. However, diagnosis can be challenging given its wide range of cranial, ocular, and systemic manifestations. This retrospective, multicenter cohort study of 300 biopsy-proven GCA cases identified from a nationwide U.S. Veterans Health Administration database examined whether presenting symptoms and the specialty of the first provider seen influenced diagnostic delays. Patients presenting with ocular complaints were treated significantly sooner (8.3 days after presentation) compared to those presenting with headache (12.9 days) or systemic symptoms (20.2 days). Patients first evaluated by an eye care provider or the emergency department also received treatment more than twice as quickly as those initially seen in primary care. This study highlights the importance of recognizing nonocular presentations of GCA, particularly since primary care was the first point of contact for 40% of patients.

British Journal of Ophthalmology

How much does the interpretation of post-op IOP measurements influence reported glaucoma surgical success?

Success might just depend on how you measure it. In glaucoma surgery studies, intraocular pressure (IOP) thresholds are commonly used to define post-op surgical success or failure, but patients often have multiple IOP measurements within the same recommended follow-up window. In this retrospective study, investigators examined how six different approaches to handling these sequential measurements affected reported outcomes in 934 eyes undergoing trabeculectomy and 1760 eyes undergoing deep sclerectomy, with median follow-up of 41 and 45 months, respectively. Using an IOP threshold of ≤21 mm Hg with ≥20% reduction from baseline, 5-year success rates varied substantially depending on the method used: in trabeculectomy eyes, from 38.8% using all visit measurements to 54.8% using the minimum IOP; in deep sclerectomy eyes, from 60.4% to 74.5%, respectively. Strategies incorporating peak IOP or all available measurements produced the highest failure rates, potentially reflecting transient IOP fluctuations or outlier measurements, whereas using the minimum IOP may overestimate success. Although eyes classified as surgical failures showed faster progression in mean visual field deviation, substantial overlap remained between success and failure groups regardless of the strategy used. These findings highlight that reported glaucoma surgical success can vary depending on both the IOP threshold and how multiple measurements are handled, underscoring the need for standardized reporting and greater emphasis on clinically meaningful outcomes, such as visual field progression.

Nature

A cross-sectional analysis of the associations between dry eye disease and depression

Sometimes the eye tells a story beyond what's on the surface. Dry eye disease (DED) has increasingly been linked to mental health, with the DREAM study demonstrating that patients with depression often report more severe DED. However, despite Hong Kong having one of the highest reported prevalences of DED worldwide (7–59%), the relationship between depression and DED has not been well characterized in this population. Researchers evaluated 614 participants using the Ocular Surface Disease Index (OSDI) questionnaire alongside objective measures of DED, including tear osmolarity, ocular surface staining, lid wiper epitheliopathy, tear break-up time, Schirmer testing, and meibography. Depression was defined by self-reported depression with concurrent antidepressant use. Participants with depression reported significantly more severe dry eye symptoms, although objective ocular surface measurements were not significantly different between groups. A key limitation was the small number of participants with depression (18, of which 88.9% female), limiting statistical power. Overall, these findings suggest that depression is associated more strongly with patients' perceived severity of DED than with objective clinical signs. This study highlights t

Nature Medicine

AI improves the accuracy of inherited retinal disease diagnoses

Can AI models improve the diagnosis of inherited retinal diseases? Inherited retinal diseases (IRDs) are among the leading causes of blindness in working-age adults. Having a molecular diagnosis can be the gateway to gene therapy and clinical trials, but the pathway to getting there involves seeing multiple specialists and consultations. Jia et al. built Retina4IRD, a model that predicts 17 genotype categories from color fundus photographs, OCT, and basic clinical metadata. The model was trained on 1,843 patients (3,376 eyes) with IRD diagnoses confirmed genetically. Researchers tested the model in a randomized controlled trial with 300 patients suspected of having IRD. Patients were randomized to specialists who worked with Retina4IRD or to specialists who did not receive additional AI assistance. Next-generation sequencing was used as the gold standard for diagnosis. The AI model helped the clinician include the correct diagnosis anywhere in their top 5 predictions 88.5% of the time; compared to only 67.3% of the time among clinicians who did not use the model. Similarly, the AI model helped the clinician diagnose IRDs correctly the first time 37.8% of the time; an improvement from only 22.4% without the AI model. These findings show that Retina4IRD can help narrow the field of potential diagnoses for IRD, but sequencing is still likely to be required for diagnosis.

Sometimes the best intervention is none at all. The DRCR.net Protocol V study sought to determine the optimal management of center-involving diabetic macular edema. In this randomized controlled trial, 702 individuals (n=702 study eyes) with type 1 or type 2 diabetic with baseline visual acuity of 20/25 or better were enrolled. Participants were placed into 3 study groups: initial observation (n=236), initial laser photocoagulation (n=240), and initial aflibercept (n=226). Aflibercept was initiated in the observation and laser groups as needed based on reduced vision and worsened macular edema.

Key Points: 

• At 2 years, there was no significant difference in the rate of vision loss (5-letter reduction) between the observation (19%), laser photocoagulation (17%), and aflibercept group (16%)
• There was a small increase in the proportion of patients that were 20/20 at year 2 for the Aflibercept group (77%) compared to initial observation (66%) (P = 0.03); this was not true for Aflibercept versus laser

Overall, the DRCR V is a landmark study because it was the first large RCT to evaluate management outcomes in diabetic macular edema after the development of anti-VEGF treatments. Anti-VEGF treatments provided ophthalmologists with a new option for diabetic macular edema; however, there was no evidence prior to this study to demonstrate that early treatment initiation improved visual acuity outcomes for these patients. This study demonstrated that individuals with good baseline visual acuity did not require initiation of treatment with anti-VEGF intravitreal injections or laser photocoagulation.