Four articles. Four minutes. | Issue #21
Diagnostics: Frequent home perimetry yielded more precise mean-deviation slope estimates in an exploratory analysis
Surgery: Repeat trabeculectomy and Ahmed implantation showed similar visual field progression rates in a retrospective cohort
Genetics: Polygenic risk scores showed moderate POAG discrimination in European-ancestry cohorts
Artificial Intelligence: RNFL estimates from fundus photographs yielded continuous glaucoma likelihood ratios
Weekly Home Monitoring of Glaucoma with Virtual Reality Perimetry: A 2-Year Prospective Longitudinal Study
Methods
This prospective cohort enrolled 53 patients (106 eyes) with open-angle glaucoma at Wills Eye Hospital for unsupervised weekly VisuALL home perimetry and supervised Humphrey 24-2 testing every 6 months. The primary outcome was agreement in mean deviation (MD) and pattern standard deviation (PSD); 51 participants contributed follow-up data over a mean of 16.6 months.
Results
Home and office MD and PSD correlated strongly through 12 months (r = 0.81-0.94), with home MD averaging 1.4 dB higher. Participants completed 4,817 home tests, with no significant difference in baseline test-retest variability between devices. Exploratory analysis found more precise home MD slope estimates (standard error 0.40 vs 0.53 dB/year; unadjusted P = 0.049), while PSD slope precision did not differ significantly. During months 18 to 24, 26 of 30 eligible participants completed at least one session, but only 9 averaged at least three sessions monthly.
Conclusion
Unsupervised home perimetry was feasible in this selected cohort and yielded more precise MD slope estimates, although testing frequency declined over time. The precision finding remains exploratory.
Our Angle
Home perimetry is most useful when a patient's rate of field loss remains uncertain between clinic visits, particularly in suspected fast progressors or patients who cannot return frequently. This single-center cohort involved motivated adults with open-angle glaucoma who could use the headset after training, with mild or moderate damage in most eyes. The tighter MD slope estimates appear to come from repeated sampling rather than better individual tests; single-test variability was similar to office testing. Because HomeVF MD averaged about 1.4 dB higher than Humphrey, I would keep separate baselines and follow each platform's own trend. Adherence is also an important practical limitation.
These findings support home fields as an adjunct for selected patients, with established office surveillance continuing. The practical question is whether more frequent testing changes management, not simply whether it produces a tighter slope estimate. The analysis was exploratory and unadjusted for multiple comparisons, and improved treatment decisions or visual outcomes were not established. Weekly testing was the goal, but by 18 to 24 months only 30% of eligible participants averaged at least 3 sessions per month; we still need validated action thresholds, a manageable review workflow, and evidence that earlier detection preserves vision.
Surgery
Long-Term Visual Field Outcomes after failed trabeculectomy: Same site Trabeculectomy versus Ahmed Glaucoma Valve Implantation
Design: Retrospective clinical cohort study
Journal: American Journal of Ophthalmology, September 2026
Authors: Khaliliyeh et al.
In 186 eyes with a failed trabeculectomy, repeat same-site trabeculectomy and Ahmed valve implantation had similar observed visual field outcomes over median field follow-up of 6.4 to 6.6 years. Progression confirmed by at least two methods occurred in 31% and 35%, respectively (P = 0.6), although repeat trabeculectomy eyes had lower postoperative IOP and used fewer medications. Among 108 eyes with preoperative fields, progression fell after repeat trabeculectomy (45% to 27%) but showed no clear reduction after Ahmed implantation (31% to 29%). Higher postoperative peak IOP was the only measured factor associated with progression after adjustment. Surgeons chose the procedure, Ahmed eyes began with higher IOP and more prior ocular procedures, and the trabeculectomy subset had faster preoperative loss, so these findings cannot establish equivalence or a preferred operation.
Genetics
Predictive Performance of Polygenic Risk Scores for Primary Open-Angle Glaucoma: A Systematic Review and Meta-Analysis
Design: Systematic review and meta-analysis
Journal: Ophthalmology Science, September 2026
Authors: Zidan et al.
Polygenic risk scores showed moderate discrimination for primary open-angle glaucoma in a 14-study meta-analysis of approximately 692,000 participants after accounting for confirmed cohort overlap. The pooled AUC was 0.75 across 10 studies, and nine studies yielded an odds ratio of 1.70 per standard-deviation increase in genetic risk after harmonizing different reporting scales. Higher scores were also associated with earlier diagnosis, higher IOP, faster visual field deterioration, and more glaucoma surgery, although most of these outcomes came from only one or two cohorts. Between-study heterogeneity was extreme (I² approximately 99%), and all included studies were from European-ancestry populations. A PubMed-only search, residual participant overlap, and conversion of six extreme-group comparisons to a common scale limit how confidently these pooled estimates can be applied to an individual patient.
Artificial Intelligence
Diagnostic Likelihood Ratios for Glaucoma Using Deep Learning-Predicted Retinal Nerve Fiber Layer Thickness from Fundus Photographs
Design: Cross-sectional diagnostic study
Journal: Ophthalmology Science, October 2026
Authors: Nilforoushan et al.
A deep learning model converted RNFL thickness estimates from fundus photographs into continuous likelihood ratios for glaucoma in 359 participants. In this cross-sectional comparison of 232 participants with glaucomatous optic neuropathy and 127 healthy controls, predicted thickness and measured OCT thickness each had an adjusted AUC of 0.986. A predicted thickness of 70 µm corresponded to a likelihood ratio of 35.7, versus 0.17 at 90 µm; around 83.4 µm, the ratio was 1 and added little diagnostic information. This approach updates pretest probability across the measurement range without labeling each result simply normal or abnormal. The likelihood-ratio curve was derived in this same clinic cohort, which used younger, highly screened healthy controls and a glaucoma reference standard incorporating OCT and visual-field abnormalities, so calibration and screening performance require external validation.
A first-in-human phase 1/2 gene therapy trial will test retinal neuroprotection in adults with open-angle glaucoma.
Four-year follow-up of femtosecond laser trabeculotomy found sustained IOP reduction in a small first-in-human cohort, with glaucoma medications continued.
FDA launches an expedited IND pilot that lets selected sponsors submit application components for rolling review before complete filings.
An oral NLRP3 inhibitor enters a 180-participant phase 2 trial for diabetic macular edema, with week-12 visual acuity as the primary endpoint.
What was the most common ancient Greek and Roman explanation for how vision worked?
A. Objects released tiny images that entered the eye
B. The eye sent light or fire toward the object being viewed
C. The brain projected pictures forward through the optic nerve
D. Ambient air carried complete images into the pupil
See answer at bottom of newsletter.
Get the next issue in your inbox
The Open Angle is a weekly glaucoma research digest for busy eye care professionals.
Four articles worth knowing, in under four minutes.
Edited by Jella An, MD, MBA and Jason Dossantos, MD.
The Open Angle is an educational editorial product. It is not medical advice. Readers should review original sources before changing practice.
TRIVIA ANSWER EXPLANATION
What was the most common ancient Greek and Roman explanation for how vision worked?
B. The eye sent light or fire toward the object being viewed
Many Greek and Roman thinkers believed the eye sent light, fire, or rays toward the object being viewed. The outward stream was thought to connect the observer with the object, allowing vision to occur without physical contact.
Other ancient thinkers argued that objects released small images or particles that entered the eye. Both accounts predated modern optics. The debate focused on the direction of the process, centuries before the retinal image and the role of incoming light could be studied.