Sriwijaya Journal of Ophthalmology

Continuous 24-Hour Intraocular Pressure Monitoring with a Smart Contact Lens Predicts Visual Field Progression in Normal-Tension Glaucoma

2026-06-17T04:00:03+00:00

Introduction: Normal-tension glaucoma (NTG) may progress despite normal office intraocular pressure (IOP), suggesting that 24-hour IOP dynamics captured by a smart contact lens (SCL) sensor may improve risk stratification. This study evaluated whether SCL-derived nocturnal IOP parameters independently predict Humphrey 24-2 visual field (VF) progression in NTG.

Methods: This prospective observational cohort enrolled 62 NTG eyes (62 patients; one eye per patient) who underwent 24-hour SCL monitoring and serial Humphrey 24-2 SITA-Standard VF testing over 24 months at a private hospital in Palembang, Indonesia. The primary outcome was VF mean deviation (MD) slope (dB/year). Multivariable linear regression and receiver operating characteristic (ROC) analysis were performed.

Results: Eyes with a nocturnal IOP acrophase (n = 36) had a significantly faster MD decline (−1.24 ± 0.31 dB/year) than eyes without acrophase (n = 26; −0.42 ± 0.19 dB/year; mean difference −0.82 dB/year, 95% CI −0.97 to −0.67, p < 0.001; Cohen d = 3.20). Independent predictors of MD slope were nocturnal acrophase (β = −0.71, p < 0.001), number of long peaks (β = −0.18, p < 0.001), and baseline RNFL thickness (β = +0.034, p = 0.002); adjusted R² = 0.64. Nocturnal amplitude yielded an AUC of 0.83 (95% CI 0.74–0.91).

Conclusion: SCL-derived nocturnal IOP parameters independently predict VF progression in NTG. Integration of 24-hour SCL monitoring may enhance risk stratification beyond office IOP measurement.

Diagnostic Accuracy of a Federated Learning Algorithm for Proliferative Diabetic Retinopathy Detection: A Multicenter Indonesian Study

2026-06-17T06:39:52+00:00

Introduction: Proliferative diabetic retinopathy (PDR) remains a leading cause of preventable blindness in resource-limited settings. Federated learning (FL) enables collaborative artificial intelligence (AI) model training without sharing patient data. This study evaluated the diagnostic accuracy of an FL-based algorithm for PDR detection across three Indonesian ophthalmology centers.

Methods: This multicenter, prospective, diagnostic accuracy study enrolled 512 eyes from 289 patients with type 2 diabetes at three private hospital ophthalmology clinics in Palembang, Jakarta, and Surabaya, Indonesia (January 2023–December 2024). All eyes underwent standardized fundus photography, spectral-domain optical coherence tomography, and comprehensive ophthalmic examination. The FL-based deep learning algorithm was evaluated against two independent retinal specialists using the International Clinical Diabetic Retinopathy classification. Primary outcomes were sensitivity, specificity, and area under the receiver operating characteristic curve (AUC).

Results: The FL-based AI achieved a sensitivity of 94.6% (95% CI 81.8–99.3), specificity of 91.2% (95% CI 88.2–93.6), and AUC of 0.962 (95% CI 0.943–0.981) for PDR detection. Agreement with retinal specialists was substantial (κ = 0.87). Performance was consistent across centers (AUC 0.955–0.968; p = 0.841). Media opacity was the strongest predictor of misclassification (OR 3.42; 95% CI 1.87–6.25; p < 0.001).

Conclusion: The FL-based AI demonstrated high diagnostic accuracy for PDR detection comparable to retinal specialists across multiple Indonesian centers. This privacy-preserving approach may facilitate scalable diabetic retinopathy screening in resource-limited ophthalmology settings.

Safety and Efficacy of Subretinal AAV-CRISPR/Cas9 Gene Editing for Rhodopsin-Mutant Retinitis Pigmentosa: A Phase I/IIa Dose-Escalation Triall

2026-06-19T06:33:19+00:00

Introduction: Autosomal dominant retinitis pigmentosa (adRP) from rhodopsin (RHO) mutations lacks an approved gene therapy. This Phase I/IIa trial evaluated subretinal AAV5-CRISPR/Cas9 for RHO-adRP.

Methods: An open-label, 3+3 dose-escalation design with expansion at the recommended Phase II dose (RP2D) was used. Eighteen patients (18 study eyes; one eye per patient) received subretinal AAV5-CRISPR/Cas9 at low (1.5×10¹⁰ vg, n = 3), mid (5.0×10¹⁰ vg, n = 3), or high (1.5×10¹¹ vg, n = 6) doses, plus an expansion cohort (n = 6) at RP2D, at a private hospital in Palembang, Indonesia. The primary endpoint was dose-limiting toxicities (DLTs); secondary endpoints were BCVA (LogMAR), SD-OCT ellipsoid zone (EZ) width, microperimetry, and ffERG.

Results: No DLTs occurred. The high-dose cohort showed a BCVA improvement of −0.14 LogMAR (95% CI −0.22 to −0.06, p = 0.003; Cohen d = 1.82), an EZ width increase of +312 µm (95% CI +187 to +437, p < 0.001; d = 2.45), and a microperimetry gain of +3.8 dB (95% CI +2.1 to +5.5, p < 0.001; d = 1.94). Dose-response trends were significant (Jonckheere–Terpstra p-trend: BCVA 0.008, EZ 0.002).

Conclusion: Subretinal AAV5-CRISPR/Cas9 demonstrated acceptable safety with dose-dependent structural and functional improvements at 12 months. Phase II/III trials are warranted.

Sustained Anti-VEGF Delivery via Hydrogel Implant Achieves Non-Inferior Visual and Anatomical Outcomes in Neovascular AMD: A 24-Month Randomized Controlled Trial

2026-06-19T07:43:08+00:00

Introduction: Neovascular age-related macular degeneration (nAMD) requires frequent anti-VEGF intravitreal injections, creating substantial treatment burden. This study compared the 24-month efficacy and safety of a sustained-release anti-VEGF hydrogel implant versus monthly ranibizumab in treatment-naive nAMD eyes.

Methods: This prospective randomized controlled trial enrolled 126 eyes (96 patients; 30 bilateral) with treatment-naive nAMD at a private hospital in Palembang, Indonesia (January 2022–December 2024). Eyes were randomized 1:1 to a biodegradable anti-VEGF hydrogel implant (n=64) or monthly ranibizumab 0.5 mg (n=62). The primary outcome was non-inferiority of the mean best-corrected visual acuity (BCVA) change (LogMAR) at 24 months (margin: 0.10 LogMAR). Analyses used linear mixed-effects models with generalized estimating equations to account for inter-eye correlation.

Results: Mean BCVA improved by −0.24±0.18 LogMAR (≈12 letters) in the hydrogel group and −0.22±0.20 LogMAR (≈11 letters) in the ranibizumab group (difference: −0.02; 95% CI: −0.09 to 0.05; p=0.578; non-inferiority confirmed). The hydrogel group required 77.8% fewer injections (4.8±1.2 vs 21.6±3.4; p<0.001). Safety profiles were comparable.

Conclusion: The sustained-release anti-VEGF hydrogel implant achieved non-inferior visual and anatomical outcomes versus monthly ranibizumab while reducing injection burden by 77.8% over 24 months. Multicenter validation is warranted.

Machine Learning Prediction of Binocular Vision Recovery in Accommodative Esotropia: A Prospective Multicenter Study

2026-06-22T02:52:26+00:00

Introduction: Accommodative esotropia is the most common childhood convergent strabismus, yet predicting binocular vision recovery after treatment remains challenging. This study developed and validated machine learning (ML) models to predict binocular vision recovery using baseline clinical parameters.

Methods: This prospective multicenter study enrolled 156 patients (aged 2–12 years) with accommodative esotropia across three private hospitals in Indonesia. The unit of analysis was patients. Baseline binocular vision parameters were used to train four ML models (gradient boosting, random forest, neural network, logistic regression) with 5-fold stratified cross-validation. Treatment success was defined as stereoacuity ≤100 arc seconds at 12 months. Model performance was evaluated using AUC-ROC and SHAP feature importance.

Results: Treatment success was achieved in 104 patients (66.7%). Gradient boosting achieved the highest AUC of 0.903 (95% CI: 0.854–0.952; sensitivity 0.875; specificity 0.827). The strongest predictors were baseline stereoacuity ≤400 arc seconds (OR = 4.15; p < 0.001), deviation angle ≤20 PD (OR = 3.42; p < 0.001), and Worth 4-dot fusion (OR = 3.21; p = 0.001).

Conclusion: ML models accurately predicted binocular vision recovery in accommodative esotropia, identifying clinically interpretable predictors that may optimize treatment selection in pediatric strabismus management.