BMJ Open Ophthalmology最新文献

Objective: Biosimilars are helping to reduce the cost burden of treatment and widen patient access to therapies. This multicentre trial compared the efficacy, safety and immunogenicity of the biosimilar aflibercept FYB203 with reference aflibercept in patients with neovascular age-related macular degeneration (nAMD).

Methods and analysis: Patients aged ≥50 years with newly diagnosed nAMD and a best-corrected visual acuity (BCVA) between 20/40 and 20/200 Snellen equivalent were randomised (1:1) to double-masked treatment with 2 mg FYB203 or EU-approved reference aflibercept by intravitreal injection every 4 weeks for three doses (baseline, weeks 4 and 8) then every 8 weeks up to week 48. The primary efficacy endpoint was the change from baseline in BCVA by Early Treatment Diabetic Retinopathy Study (ETDRS) letters at week 8 in the study eye. Therapeutic equivalence of FYB203 and reference aflibercept was demonstrated if, depending on the regulatory requirement with respect to the significance level, the two-sided 90.4% and 95.2% CIs were within the predefined equivalence interval of (-3.5 to 3.5) ETDRS letters.

Results: A total of 433 patients received treatment with FYB203 (n=215) or reference aflibercept (n=218). Mean improvement in BCVA from baseline to week 8 was 6.6 ETDRS letters with FYB203 and 5.6 ETDRS letters with reference aflibercept, with an estimated mean treatment difference of 1.0 and the two-sided 90.4% CI (-0.3 to 2.2) and 95.2% CI (-0.6 to 2.5) fully contained within the pre-defined equivalence margins, confirming therapeutic equivalence between FYB203 and reference aflibercept. Safety and immunogenicity profiles were similar between groups.

Conclusion: Although conducted during the COVID-19 pandemic in a potentially vulnerable elderly population and affected by geopolitical disruption in Ukraine, mitigation measures minimised the overall impact of these events. FYB203 demonstrated therapeutic equivalence to reference aflibercept in patients with nAMD, supporting similar clinical performance across all approved indications.

Trial registration number: Clinicaltrials.gov: NCT04522167; EudraCT: 2019-003923-39.

Objective: Sweden has a long-standing tradition of actively managing infants born extremely preterm at 22-23 weeks' gestational age. This study analyses screening and treatment pathways for retinopathy of prematurity (ROP) in these infants, assessing costs and health outcomes to develop a model of direct healthcare costs.

Method and analysis: The cohort included all 399 infants born at 22-23 weeks in Sweden (2007-2018) who underwent ROP screening, recorded in the national ROP registry SWEDROP. A health economic model estimated costs based on three primary pathways: (1) no sight-saving treatment, (2) laser as initial treatment and (3) anti-vascular endothelial growth factor (anti-VEGF) as initial treatment. Pathways 2 and 3 were further divided into single and multiple treatments. Costs were calculated using screening frequency, treatment and neonatal care expenses. Register data were verified against medical records. An expanded model incorporated gestational age and comorbidities.

Results: In the basic model, 36% received laser (16 screenings on average; 32% required retreatment), while 7% received anti-VEGF injections (25 screenings; 69% required retreatment). The cost of screening and treating an infant with laser was Int$ 18 590, compared with Int$ 20 792 for anti-VEGF. The expanded models showed similar screening and treatment frequencies.

Conclusion: Despite similar overall costs, the higher screening burden in the anti-VEGF group (25 vs 16 screenings) raises concerns regarding cost-effectiveness and potential health impacts. Main limitations include the use of cost data from a single hospital, potential selection bias between treatment groups and limited precision in small subgroups. These findings lay the groundwork for future research on long-term health and cost outcomes in this vulnerable population.

Introduction: Pars plana vitrectomy with intravitreal gas tamponade has been the standard treatment for full-thickness macular holes (FTMHs) for the last 30 years. Gas tamponade has multiple drawbacks: it results in very poor vision for several weeks, limiting patients' ability to drive and work; many undertake intensive face-down posturing for about a week after surgery, to float the gas bubble onto the macula; gas is cataractogenic and prevents adequate fundus examination; it can elevate intraocular pressure, mandating regular hospital review; patients cannot fly and the fluorocarbon tamponades are extremely potent and long-lived greenhouse gases.Six uncontrolled retrospective case series have reported promising results utilising novel FTMH surgery without gas tamponade.The study aims to establish if it is feasible to recruit, retain and evaluate patients with FTMHs into a pivotal randomised control trial of vitrectomy without gas tamponade. The main clinical aim is to collect preliminary safety and efficacy data, comparing gasless with standard vitrectomy.

Methods and analysis: This randomised, assessor-masked, surgical, feasibility study aims to recruit 60 participants with FTMH from six English National Health Service sites. Participants will be randomised 1:1 to either standard vitrectomy, gas tamponade and postoperative posturing, or to gasless vitrectomy without posturing. Gasless surgery involves folding a hinged flap of internal limiting membrane over the hole, secured by gravity and a viscoelastic gel. Participants attend postoperatively for examination and imaging. The main feasibility outcomes are recruitment and retention rates, cross-over and participant treatment acceptability. The main clinical outcomes are primary FTMH closure (3 months), best-corrected visual acuity (6 months) and adverse events.

Ethics and dissemination: South Central-Hampshire A research ethics committee gave a favourable opinion of the study (Reference: 24/SC/0019, 16 February 2024). The study results will be published in a peer-reviewed journal.

Trial registration number: NCT06079593 .

Aims: To compare the changes in ocular parameters after implantation of two types of intraocular lenses (IOLs) following cataract surgery. One IOL transmits violet light (VL), and the other does not.

Methods: A total of 402 patients were randomly assigned to receive either a VL-non-transmitting IOL or a VL-transmitting IOL. Ocular parameters were measured preoperatively and postoperatively, and participants completed a lifestyle questionnaire. The choroidal thickness (CT) was measured using the PLEX Elite 9000 (Carl Zeiss Meditec). The associations between changes in CT and age, sex, type of IOL, preoperative CT, time spent outdoors, time spent using a smartphone or tablet and time spent reading were examined using stepwise multiple regression analysis (SPSS V.28.0 for Windows, IBM-SPSS).

Results: The baseline mean spherical equivalents were -4.00±5.97 D and -2.19±4.73 D, the axial lengths were 24.84±1.82 mm and 24.19±1.65 mm, and the CTs were 213.78±102.93 µm and 239.20±104.98 µm for the VL-non-transmitting and VL-transmitting IOL groups, respectively. The mean changes in the CT from 3 to 12 months postoperatively were -0.30±18.32 µm for the VL-non-transmitting IOLs and 3.48±16.46 µm for the VL-transmitting IOLs (p=0.012). Multiple regression analysis identified significant choroidal thickening associated with female sex (p=0.004), VL-transmitting IOL implantation (p=0.049) and increased outdoor exposure (p=0.008).

Conclusion: The choroidal thickening after cataract surgery was associated with the VL-transmitting IOL and longer time spent outdoors with exposure to abundant VL. Despite these findings, this study has several limitations, including a relatively short follow-up period, and it did not assess postoperative lifestyle, VL exposure or actual peripheral defocus in the patients.

Trial registration number: UMIN000038961.

Purpose: This study aims to evaluate and compare the diagnostic reasoning of five large language models (LLMs) in complex neuro-ophthalmological cases. We assessed the performance of GPT-o1 Pro, GPT-4o, Google Gemini, Grok 2 and DeepSeek in handling clinical scenarios related to neuro-ophthalmology.

Method: 18 clinical scenarios, derived from six complex neuro-ophthalmological cases, were presented to five LLMs: GPT-o1 Pro, GPT-4o, Google Gemini, Grok 2 and DeepSeek. The responses generated by these models were evaluated using the Revised-IDEA (R-IDEA) assessment tool. R-IDEA scores for high-quality responses ranged from 6 to 10, with 'Excellent' responses defined as those scoring between 8 and 10. In addition, the simplicity of each response was evaluated based on word count using a readability tool.

Result: GPT-o1 Pro (8.80) significantly outperformed GPT-4o (6.80) and Grok 2 (6.94) in the R-IDEA scores (p=0.001). It achieved 100% high-quality responses, compared with 72.2% for GPT-4o, 77.8% for Grok 2 and 83.3% for both Gemini and DeepSeek (p=0.175). Regarding 'Excellent' responses, GPT-o1 Pro achieved 88.9% of its responses rated as Excellent, significantly outperforming the other models: 27.8% for GPT-4o, 38.9% for Grok 2 and 55.6% for both Gemini and DeepSeek (p=0.003). GPT-o1 Pro used the fewest words, showing significant differences compared with GPT-4o (p<0.001) and Gemini (p=0.032).

Conclusion: The study underscores the superior clinical reasoning capabilities of ChatGPT-o1 Pro in neuro-ophthalmology compared with other LLMs, highlighting its potential for enhancing diagnostic processes in this complex field.

The myopia control field stands at an ethical crossroads. Withholding proven soft contact lens interventions in control arms of myopia management trials is increasingly viewed as ethically questionable and operationally challenging, particularly given that MiSight 1-day contact lens has demonstrated efficacy in at least four randomised controlled trials. The recent methodological evolution towards non-inferiority trials against an established control rather than a placebo (single vision) is a welcome change. However, this shift raises a critical challenge: setting non-inferiority margins that are both clinically meaningful and statistically robust.

Objective: To analyse ocular complications, their surgical treatments and risk factors in anterior uveitis (AU).

Methods and analysis: From a single-centre registry of all adult patients with AU in 2009 (n=413), ocular complications of 641 uveitic eyes, their operative treatments and risk factors were studied comparing patients with or without complications.

Results: A total of 269 (42%) uveitic eyes and 189 (46%) patients had ocular complications. The most common complication was cataract in 217 (34%) eyes. Glaucoma was noted in 69 (11%) eyes, cystoid macular oedema in 39 (6%), posterior synechiae in 30 (5%), band keratopathy and optic disc oedema in 11 (2%), and iris atrophy and cystic corneal oedema in one eye (0.2%). Complications were typically observed in the chronic course of AU, without notable sex difference and with non-infectious aetiology. A steroid response, the course of uveitis, eye involvement, oral, subconjunctival and prophylactic dosage of steroids, methotrexate and immunosuppressants management formed significant risk factors for complication development in AU. Respectively, significant risk was also observed with the steroid response in the development of complications other than glaucoma.

Conclusions: A remarkable portion of eyes affected by AU-two out of five-develop ocular complications during long follow-up. Our results highlight the significance of uveitic inflammation and its corticosteroid treatment leading to earlier onset of cataract, glaucoma and their surgical treatments. For the first time, a significant risk was observed between the number of corticosteroid responders and the development of complications, extending beyond glaucoma. The retrospective design of the study can be considered as the main limitation of the study.

Objective: This study investigated the relationship between estimated glucose disposal rate (eGDR) and the risk of age-related ocular diseases, including macular degeneration, glaucoma, cataracts, diabetic retinopathy (DR) and retinal detachment (RD), in individuals with stages 0-3 of cardiovascular-kidney-metabolic (CKM) syndrome.

Methods and analysis: This prospective cohort study included 223 120 participants from the UK Biobank. The CKM stages were defined based on adiposity, metabolic risk factors and subclinical cardiovascular disease. Lower eGDR values indicate greater insulin resistance. Outcomes were incidences of macular degeneration, glaucoma, cataract, DR and RD. HRs and 95% CIs were estimated using Cox proportional hazards models. Non-linear relationships were explored using restricted cubic splines.

Results: The study showed that macular degeneration (HR 0.92, 95% CI 0.87 to 0.97, p=0.001) and glaucoma (HR 0.91, 95% CI 0.87 to 0.95, p<0.001) were linearly associated with eGDR. Cataracts exhibited a U-shaped association with eGDR (P non-linear=0.001) and DR exhibited an L-shaped association (P non-linear=0.018). Quartile stratification of eGDR significantly differentiated risk in DR (Q4 vs Q1: HR 0.15, 95% CI 0.04 to 0.52, p=0.003) and RD (Q4 vs Q1: HR 0.65, 95% CI 0.47 to 0.89, p=0.007). Stratified effects analysis revealed that these associations were more significant in advanced CKM syndrome stages.

Conclusions: eGDR is associated with ocular diseases risk in CKM syndrome, especially in advanced stages. This finding suggests the potential use of eGDR for guiding ophthalmic screening in CKM management.

Purpose: To analyse drug-associated immune-mediated uveitis (IMU) risk using large-scale pharmacovigilance data.

Methods: We conducted a retrospective pharmacovigilance study using data from the FDA Adverse Event Reporting System (FAERS) and the Japanese Adverse Drug Event Report (JADER) databases, encompassing reports from Q1 2004 to Q4 2024. Disproportionality analysis identified drug-IMU associations using reporting odds ratios (ROR) and information component (IC). Time-to-onset (TTO) was also assessed.

Results: Disproportionality analysis identified 17 176 IMU cases in FAERS and 2580 in JADER. IMU prevalence was higher in females, with the 30-80 age group most affected. Antineoplastic and immunomodulating agents, sensory organ drugs and anti-infectives were the most frequently associated drug classes with IMU. Among 117 drugs with positive signals for IMU in FAERS, brolucizumab (ROR₀₂₅=145.89, IC₀₂₅=7.08), faricimab (ROR₀₂₅=134.06, IC₀₂₅=6.97) and rifabutin (ROR₀₂₅=82.23, IC₀₂₅=6.32) exhibited the strongest associations. The three drugs also ranked as the top three drugs with the strongest association with IMU in JADER. The study also highlights significant differences in TTO among various drug classes, with musculoskeletal and sensory organ drugs showing longer TTO compared with other categories.

Conclusions: Our study highlights the diverse range of drugs implicated in the occurrence or progression of IMU. These findings emphasise the need for early risk assessment, careful monitoring and positive intervention in managing treatments involving high-risk drugs. Future research should focus on elucidating underlying mechanisms and risk factors to develop safer therapeutic strategies.

Objective: To evaluate the effectiveness of a deep learning-based style adaptation strategy in improving the diagnostic accuracy and cross-camera generalisability of artificial intelligence (AI) for detecting diabetic retinopathy (DR).

Methods and analysis: This diagnostic study involved prospective recruitment of patients aged 50 years and older attending the outpatient clinic at a tertiary eye hospital in Southern India, between 14 June and 5 August 2022. Paired macula-centred retinal images were captured using two fundus cameras: Optain Resolve (portable, automated) and Topcon NW400 (static, manual). A style adaptation model, the Style-Consistent Retinal Image Transformation Network (SCR-Net), was applied to align image styles across cameras. The AI-based DR detection model, developed using the InceptionNeXt-T architecture, was trained on images from the EyePACS data set and evaluated under three scenarios: (1) training and testing on original images (2) training and testing on SCR-Net-adapted images; and (3) training on a combined (original+adapted) data set and testing on adapted images. Diagnostic accuracy and preservation of image quality were evaluated.

Results: The mixed training/testing approach (scenario 3) achieved the highest diagnostic accuracy for Optain images at 79.2% (95% CI 75.9% to 82.6%) with a Cohen's kappa of 0.893 (95% CI 0.867 to 0.917). Adapted images preserved critical diagnostic features (peak signal-to-noise ratio, 29.35; structural similarity index measure, 0.847). Style adaptation reduced false positives in Optain images while maintaining robust diagnostic performance for Topcon images, effectively addressing cross-camera variability.

Conclusion: Style adaptation using SCR-Net enhances the consistency and generalisability of AI-based DR detection systems by reducing false positives and maintaining robust performance across camera systems. This approach has the potential to democratise access to early DR diagnosis in underserved regions. This study was conducted at a single centre using a limited set of fundus cameras, which may affect the generalisability. Nonetheless, further validation across diverse imaging systems and clinical settings is warranted to support broader applicability.