Week in review: AI-assisted retinal imaging, sleep and glaucoma, uveitis risk factors in JIA

A weekly roundup of ophthalmic news from around the web.

Retinal vasculometry using AI may be the future of stroke and MI detection. Retinal images from approximately 72,000 participants in 2 large-scale cohort studies were assessed with an automated, artificial intelligence (AI)–enabled system, and the retinal vasculometry findings on vessel width were included in a risk score to predict stroke, myocardial infarction (MI), and circulatory mortality. Risk prediction with this score—which also took into account gender, age, medical history, and smoking status—was similar to that of an algorithm based on Framingham risk scores that used data from blood samples and blood pressure measurement. As retinal vasculometry is less invasive than drawing blood, it may be a screening option that can be more easily used in a wide variety of community settings. British Journal of Ophthalmology

Have sleep issues? You could be at greater risk for glaucoma. The link between sleep behavior and glaucoma risk was assessed in a cohort study of 409,000 individuals enrolled in the UK Biobank. Participants completed a questionnaire about their sleep and were followed for a mean period of 10.7 years. During that time, 2.1% of participants developed glaucoma, with insomnia, snoring, frequent daytime sleepiness, and overly short or long sleep durations each linked to increased risk. Because sleep behavior is modifiable, the authors note that “these findings underscore the necessity of sleep intervention for individuals at high risk of glaucoma and potential ophthalmologic screening among individuals with chronic sleep problems to help prevent glaucoma.” BMJ Open

Certain factors increase risk of uveitis development in children with JIA. Researchers at the University Medical Center Utrecht in the Netherlands have developed a model for predicting 2-, 4-, and 7-year risk of new-onset chronic uveitis in juvenile idiopathic arthritis (JIA). The model used data from 5393 children with JIA under treatment with nonsteroidal anti-inflammatory drugs or disease-modifying antirheumatic drugs, and data from 2 additional JIA cohorts were used to validate the model. Younger age at JIA diagnosis, treatment with methotrexate or adalimumab, presence of rheumatoid factor–positive polyarthritis, and positivity for antinuclear antibodies were linked to a greater risk of developing uveitis. Arthritis & Rheumatology

What leads to better postsurgical visual outcomes in patients with diabetes? To answer that question, researchers reviewed results from the ACCORD and ACCORD-Eye studies of cardiovascular outcomes in type 2 diabetes, focusing on 1136 eyes (784 patients) that underwent cataract surgery from 2001 to 2009. Sixty-seven percent of the eyes achieved visual acuity (VA) ≥20/40 after surgery. The factors associated with improved visual outcomes included better preoperative VA, access to the Veterans Affairs clinical network, and higher education levels. Clinicians should take these factors into account when providing postoperative care to their patients with diabetes. British Journal of Ophthalmology