A review of deep learning-based information fusion techniques for multimodal medical image classification.

Multimodal medical imaging plays a pivotal role in clinical diagnosis and research, as it combines information from various imaging modalities to provide a more comprehensive understanding of the underlying pathology. Recently, deep learning-based multimodal fusion techniques have emerged as powerful tools for improving medical image classification. This review offers a thorough analysis of the developments in deep learning-based multimodal fusion for medical classification tasks. We explore the complementary relationships among prevalent clinical modalities and outline three main fusion schemes for multimodal classification networks: input fusion, intermediate fusion (encompassing single-level fusion, hierarchical fusion, and attention-based fusion), and output fusion. By evaluating the performance of these fusion techniques, we provide insight into the suitability of different network architectures for various multimodal fusion scenarios and application domains. Furthermore, we delve into challenges related to network architecture selection, handling incomplete multimodal data management, and the potential limitations of multimodal fusion. Finally, we spotlight the promising future of Transformer-based multimodal fusion techniques and give recommendations for future research in this rapidly evolving field.

Artificial intelligence-based prediction of diabetic retinopathy evolution (EviRed): protocol for a prospective cohort.

INTRODUCTION: An important obstacle in the fight against diabetic retinopathy (DR) is the use of a classification system based on old imaging techniques and insufficient data to accurately predict its evolution. New imaging techniques generate new valuable data, but we lack an adapted classification based on these data. The main objective of the Evaluation Intelligente de la Rétinopathie Diabétique, Intelligent evaluation of DR (EviRed) project is to develop and validate a system assisting the ophthalmologist in decision-making during DR follow-up by improving the prediction of its evolution. METHODS AND ANALYSIS: A cohort of up to 5000 patients with diabetes will be recruited from 18 diabetology departments and 14 ophthalmology departments, in public or private hospitals in France and followed for an average of 2 years. Each year, systemic health data as well as ophthalmological data will be collected. Both eyes will be imaged by using different imaging modalities including widefield photography, optical coherence tomography (OCT) and OCT-angiography. The EviRed cohort will be divided into two groups: one group will be randomly selected in each stratum during the inclusion period to be representative of the general diabetic population. Their data will be used for validating the algorithms (validation cohort). The data for the remaining patients (training cohort) will be used to train the algorithms. ETHICS AND DISSEMINATION: The study protocol was approved by the French South-West and Overseas Ethics Committee 4 on 28 August 2020 (CPP2020-07-060b/2020-A01725-34/20.06.16.41433). Prior to the start of the study, each patient will provide a written informed consent documenting his or her agreement to participate in the clinical trial. Results of this research will be disseminated in peer-reviewed publications and conference presentations. The database will also be available for further study or development that could benefit patients. TRIAL REGISTRATION NUMBER: NCT04624737.

Switching to subtenon triamcinolone acetonide does not jeopardize the functional and anatomic outcomes of dexamethasone implant treated eyes with diabetic macular edema.

BACKGROUND: Intraocular dexamethasone implant (DEXi) is an efficient treatment for diabetic macular edema (DME). However, it may be unavailable or contraindicated. Triamcinolone acetonide is another corticosteroid that has proved to be safe and effective in treating macular edema complicating various diseases including diabetes. The purpose of this study is to evaluate the outcomes of a switch from DEXi to subtenon triamcinolone acetonide (STTA) and back, in eyes with DME. METHODS: Retrospective study. DME eyes that had been treated with DEXi and switched to STTA between October 2018 and February 2019 (stock shortage of DEXi) were included. The functional and anatomical outcomes of the switch and switch-back were studied. RESULTS: 26 eyes of 17 patients (mean age 67.1 ± 8.2 years) were considered. The mean baseline visual acuity (VA) was 0.35 ± 0.17 decimals remaining stable after DEXi, STTA and switch-back to DEXi. The mean central macular thickness (CMT) was 492.7 ± 32.8 µm initially, decreasing to 294.3 ± 133.4 µm after DEXi, 369.9 ± 182.3 µm after STTA and 297.6 ± 72.0 µm after switching back to DEXi (all p < 0.05 versus baseline). Compared to baseline, the CMT reduction was numerically better after DEXi and switching back to DEXi than after STTA (mean reduction: -200.4 µm, -167.7 µm, and -95.08 µm respectively, p = 0.13). Intraocular pressure was comparable after DEXi and STTA. CONCLUSION: DEXi is the steroid of choice in DME. However, STTA can be a cost-effective alternative when DEXi is unavailable or contraindicated. This study suggests that STTA may be used in the context of a step therapy in DME.

DISCOVER: 2-D multiview summarization of Optical Coherence Tomography Angiography for automatic diabetic retinopathy diagnosis.

Diabetic Retinopathy (DR), an ocular complication of diabetes, is a leading cause of blindness worldwide. Traditionally, DR is monitored using Color Fundus Photography (CFP), a widespread 2-D imaging modality. However, DR classifications based on CFP have poor predictive power, resulting in suboptimal DR management. Optical Coherence Tomography Angiography (OCTA) is a recent 3-D imaging modality offering enhanced structural and functional information (blood flow) with a wider field of view. This paper investigates automatic DR severity assessment using 3-D OCTA. A straightforward solution to this task is a 3-D neural network classifier. However, 3-D architectures have numerous parameters and typically require many training samples. A lighter solution consists in using 2-D neural network classifiers processing 2-D en-face (or frontal) projections and/or 2-D cross-sectional slices. Such an approach mimics the way ophthalmologists analyze OCTA acquisitions: (1) en-face flow maps are often used to detect avascular zones and neovascularization, and (2) cross-sectional slices are commonly analyzed to detect macular edemas, for instance. However, arbitrary data reduction or selection might result in information loss. Two complementary strategies are thus proposed to optimally summarize OCTA volumes with 2-D images: (1) a parametric en-face projection optimized through deep learning and (2) a cross-sectional slice selection process controlled through gradient-based attribution. The full summarization and DR classification pipeline is trained from end to end. The automatic 2-D summary can be displayed in a viewer or printed in a report to support the decision. We show that the proposed 2-D summarization and classification pipeline outperforms direct 3-D classification with the advantage of improved interpretability.

A novel RLBP1 gene mutation associated to a flecked retina syndrome.

PURPOSE: To present an unusual fleck retina condition associated to a novel RLBP1 gene mutation.Methods/Results: A 25-year old male presented flecks on fundoscopic examination. Clinical presentation, multimodal imaging and ERG were compatible with the diagnosis of benign familial fleck retina. Genetic analysis detected an RLBP1 gene, a gene commonly associated with more severe retinal disease. CONCLUSION: Flecked retina syndromes and other genetic retinal diseases have a complex genotype-phenotype relation and need further research for their pathophysiology to be fully understood.

Incidence of macular displacement after rhegmatogenous retinal detachment surgery using ultra-widefield fundus autofluorescence.

PURPOSE: The incidence of retinal displacement after rhegmatogenous retinal detachment (RRD) surgery is variable and its clinical consequences are unclear. The aim of this study was to assess the incidence and clinical features of retinal displacement after RRD surgery by using ultra-widefield (UWF) imaging. METHODS: Retrospective observational study including all consecutive patients who underwent RRD surgery at the Rothschild Foundation Hospital. Postoperative data included the visual acuity and symptoms of visual impairment. Macular retinal displacement occurrence and its features were assessed and measured by using the autofluorescence images. RESULTS: A total of 123 eyes were included. UWF fundus autofluorescence revealed the presence of macular retinal displacement in 14 (11%) eyes. All displacements were inferior, with a mean angle of 3.8°. Patients with and without macular displacement did not differ in postoperative visual acuity. The retinal detachment extent and preoperative macular involvement were not significantly associated with the occurrence of retinal displacement. CONCLUSION: In this representative cohort of eyes that underwent RRD surgery with systematic screening for postoperative retinal displacement by UWF fundus autofluorescence, 11% of eyes experienced an inferior retinal shift. As in other cohorts, the presence of metamorphopsia was not associated with the occurrence of retinal shift.

Central serous chorioretinopathy: An evidence-based treatment guideline.

Central serous chorioretinopathy (CSC) is a relatively common disease that causes vision loss due to macular subretinal fluid leakage and it is often associated with reduced vision-related quality of life. In CSC, the leakage of subretinal fluid through defects in the retinal pigment epithelial layer's outer blood-retina barrier appears to occur secondary to choroidal abnormalities and dysfunction. The treatment of CSC is currently the subject of controversy, although recent data obtained from several large randomized controlled trials provide a wealth of new information that can be used to establish a treatment algorithm. Here, we provide a comprehensive overview of our current understanding regarding the pathogenesis of CSC, current therapeutic strategies, and an evidence-based treatment guideline for CSC. In acute CSC, treatment can often be deferred for up to 3-4 months after diagnosis; however, early treatment with either half-dose or half-fluence photodynamic therapy (PDT) with the photosensitive dye verteporfin may be beneficial in selected cases. In chronic CSC, half-dose or half-fluence PDT, which targets the abnormal choroid, should be considered the preferred treatment. If PDT is unavailable, chronic CSC with focal, non-central leakage on angiography may be treated using conventional laser photocoagulation. CSC with concurrent macular neovascularization should be treated with half-dose/half-fluence PDT and/or intravitreal injections of an anti-vascular endothelial growth factor compound. Given the current shortage of verteporfin and the paucity of evidence supporting the efficacy of other treatment options, future studies-ideally, well-designed randomized controlled trials-are needed in order to evaluate new treatment options for CSC.

Imaging Modalities for Assessing the Vascular Component of Diabetic Retinal Disease: Review and Consensus for an Updated Staging System.

Purpose: To review the evidence for imaging modalities in assessing the vascular component of diabetic retinal disease (DRD), to inform updates to the DRD staging system. Design: Standardized narrative review of the literature by an international expert workgroup, as part of the DRD Staging System Update Effort, a project of the Mary Tyler Moore Vision Initiative. Overall, there were 6 workgroups: Vascular Retina, Neural Retina, Systemic Health, Basic and Cellular Mechanisms, Visual Function, and Quality of Life. Participants: The Vascular Retina workgroup, including 16 participants from 4 countries. Methods: Literature review was conducted using standardized evidence grids for 5 modalities: standard color fundus photography (CFP), widefield color photography (WFCP), standard fluorescein angiography (FA), widefield FA (WFFA), and OCT angiography (OCTA). Summary levels of evidence were determined on a validated scale from I (highest) to V (lowest). Five virtual workshops were held for discussion and consensus. Main Outcome Measures: Level of evidence for each modality. Results: Levels of evidence for standard CFP, WFCP, standard FA, WFFA, and OCTA were I, II, I, I, and II respectively. Traditional vascular lesions on standard CFP should continue to be included in an updated staging system, but more studies are required before they can be used in posttreatment eyes. Widefield color photographs can be used for severity grading within the area covered by standard CFPs, although these gradings may not be directly interchangeable with each other. Evaluation of the peripheral retina on WFCP can be considered, but the method of grading needs to be clarified and validated. Standard FA and WFFA provide independent prognostic value, but the need for dye administration should be considered. OCT angiography has significant potential for inclusion in the DRD staging system, but various barriers need to be addressed first. Conclusions: This study provides evidence-based recommendations on the utility of various imaging modalities for assessment of the vascular component of DRD, which can inform future updates to the DRD staging system. Although new imaging modalities offer a wealth of information, there are still major gaps and unmet research needs that need to be addressed before this potential can be realized. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

TENAYA and LUCERNE: Two-Year Results from the Phase 3 Neovascular Age-Related Macular Degeneration Trials of Faricimab with Treat-and-Extend Dosing in Year 2.

PURPOSE: To evaluate 2-year efficacy, durability, and safety of the bispecific antibody faricimab, which inhibits both angiopoietin-2 and VEGF-A. DESIGN: TENAYA (ClinicalTrials.gov identifier, NCT03823287) and LUCERNE (ClinicalTrials.gov identifier, NCT03823300) were identically designed, randomized, double-masked, active comparator-controlled phase 3 noninferiority trials. PARTICIPANTS: Treatment-naive patients with neovascular age-related macular degeneration (nAMD) 50 years of age or older. METHODS: Patients were randomized (1:1) to intravitreal faricimab 6.0 mg up to every 16 weeks (Q16W) or aflibercept 2.0 mg every 8 weeks (Q8W). Faricimab fixed dosing based on protocol-defined disease activity at weeks 20 and 24 up to week 60, followed up to week 108 by a treat-and-extend personalized treatment interval regimen. MAIN OUTCOME MEASURES: Efficacy analyses included change in best-corrected visual acuity (BCVA) from baseline at 2 years (averaged over weeks 104, 108, and 112) and proportion of patients receiving Q16W, every 12 weeks (Q12W), and Q8W dosing at week 112 in the intention-to-treat population. Safety analyses included ocular adverse events (AEs) in the study eye through study end at week 112. RESULTS: Of 1326 patients treated across TENAYA/LUCERNE, 1113 (83.9%) completed treatment (n = 555 faricimab; n = 558 aflibercept). The BCVA change from baseline at 2 years was comparable between faricimab and aflibercept groups in TENAYA (adjusted mean change, +3.7 letters [95% confidence interval (CI), +2.1 to +5.4] and +3.3 letters [95% CI, +1.7 to +4.9], respectively; mean difference, +0.4 letters [95% CI, -1.9 to +2.8]) and LUCERNE (adjusted mean change, +5.0 letters [95% CI, +3.4 to +6.6] and +5.2 letters [95% CI, +3.6 to +6.8], respectively; mean difference, -0.2 letters [95% CI, -2.4 to +2.1]). At week 112 in TENAYA and LUCERNE, 59.0% and 66.9%, respectively, achieved Q16W faricimab dosing, increasing from year 1, and 74.1% and 81.2%, achieved Q12W or longer dosing. Ocular AEs in the study eye were comparable between faricimab and aflibercept groups in TENAYA (55.0% and 56.5% of patients, respectively) and LUCERNE (52.9% and 47.5% of patients, respectively) through week 112. CONCLUSIONS: Treat-and-extend faricimab treatment based on nAMD disease activity maintained vision gains through year 2, with most patients achieving extended dosing intervals. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Early remodeling and loss of light-induced dilation of retinal small arteries in CADASIL.

A major hurdle to therapeutic development in cerebral small vessel diseases is the lack of in-vivo method that can be used repeatedly for evaluating directly cerebral microvessels. We hypothesised that Adaptive Optics (AO), which allows resolution images up to 1-2 µm/pixel at retinal level, could provide a biomarker for monitoring vascular changes in CADASIL, a genetic form of such condition. In 98 patients and 35 healthy individuals, the wall to lumen ratio (WLR), outer and inner diameter, wall thickness and wall cross-sectional area were measured in a parapapillary and/or paramacular retinal artery. The ratio of vessel diameters before and after light flicker stimulations was also calculated to measure vasoreactivity (VR). Multivariate mixed-model analysis showed that WLR was increased and associated with a larger wall thickness and smaller internal diameter of retinal arteries in patients. The difference was maximal at the youngest age and gradually reduced with aging. Average VR in patients was less than half of that of controls since the youngest age. Any robust association was found with clinical or imaging manifestations of the disease. Thus, AO enables the detection of early functional or structural vascular alterations in CADASIL but with no obvious link to the clinical or imaging severity.

Efficacy and Safety of Faricimab for Macular Edema due to Retinal Vein Occlusion: 24-Week Results from the BALATON and COMINO Trials.

PURPOSE: To evaluate the 24-week efficacy and safety of the dual angiopoietin-2 (Ang-2) and vascular endothelial growth factor (VEGF)-A inhibitor faricimab versus aflibercept in patients with vein occlusion. DESIGN: Phase 3, global, randomized, double-masked, active comparator-controlled trials: BALATON/COMINO (ClincalTrials.gov identifiers: NCT04740905/NCT04740931; sites: 149/192). PARTICIPANTS: Patients with treatment-naïve foveal center-involved macular edema resulting from branch (BALATON) or central or hemiretinal (COMINO) RVO. METHODS: Patients were randomized 1:1 to faricimab 6.0 mg or aflibercept 2.0 mg every 4 weeks for 24 weeks. MAIN OUTCOME MEASURES: Primary end point: change in best-corrected visual acuity (BCVA) from baseline to week 24. Efficacy analyses included patients in the intention-to-treat population. Safety analyses included patients who received ≥ 1 doses of study drug. RESULTS: Enrollment: BALATON, n = 553; COMINO, n = 729. The BCVA gains from the baseline to week 24 with faricimab were noninferior versus aflibercept in BALATON (adjusted mean change, +16.9 letters [95.03% confidence interval (CI), 15.7-18.1 letters] vs. +17.5 letters [95.03% CI, 16.3-18.6 letters]) and COMINO (+16.9 letters [95.03% CI, 15.4-18.3 letters] vs. +17.3 letters [95.03% CI, 15.9-18.8 letters]). Adjusted mean central subfield thickness reductions from the baseline were comparable for faricimab and aflibercept at week 24 in BALATON (-311.4 µm [95.03% CI, -316.4 to -306.4 µm] and -304.4 µm [95.03% CI, -309.3 to -299.4 µm]) and COMINO (-461.6 µm [95.03% CI, -471.4 to -451.9 µm] and -448.8 µm [95.03% CI, -458.6 to -439.0 µm]). A greater proportion of patients in the faricimab versus aflibercept arm achieved absence of fluorescein angiography-based macular leakage at week 24 in BALATON (33.6% vs. 21.0%; nominal P = 0.0023) and COMINO (44.4% vs. 30.0%; nominal P = 0.0002). Faricimab was well tolerated, with an acceptable safety profile comparable with aflibercept. The incidence of ocular adverse events was similar between patients receiving faricimab (16.3% [n = 45] and 23.0% [n = 84] in BALATON and COMINO, respectively) and aflibercept (20.4% [n = 56] and 27.7% [n = 100], respectively). CONCLUSIONS: These findings demonstrate the efficacy and safety of faricimab, a dual Ang-2/VEGF-A inhibitor, in patients with macular edema secondary to retinal vein occlusion. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

XTEND: Two-Year Results from a Global Observational Study Investigating Proactive Dosing of Intravitreal Aflibercept in Neovascular Age-Related Macular Degeneration.

INTRODUCTION: XTEND (NCT03939767) is a multicenter, observational, prospective study of patients with treatment-naïve neovascular age-related macular degeneration (nAMD) in routine clinical practice. The study aims to examine treatment outcomes of proactive intravitreal aflibercept (IVT-AFL) treatment regimens (fixed dosing or treat-and-extend) according to local marketing labels. METHODS: Study eyes received IVT-AFL injections as per the local label. The mean changes in best-corrected visual acuity (BCVA) and central subfield thickness (CST) from baseline to month (M) 12 and M24 were measured and stratified by baseline factors. Treatment exposure and safety data were evaluated. Statistical analysis was descriptive. RESULTS: Overall, 1466 patients from 17 countries were treated. For the overall population, the mean ± standard deviation (SD) age was 78.7 ± 8.5 (range 50-100) years, and 891 patients (60.8%) were female. The mean ± SD baseline BCVA was 54.3 ± 20.3 letters and CST was 374 ± 126 µm. At M12 and M24, mean (95% confidence interval [CI]) BCVA change was + 4.3 (3.4, 5.3) and + 2.3 (1.3, 3.3) letters, respectively. Mean (95% CI) CST was - 106 (- 114, - 99) µm and - 109 (- 117, - 102) µm at M12 and M24, respectively. At M24, 41.5% of patients had a BCVA ≥ 70 letters. Patients received a mean ± SD of 7.7 ± 2.7 injections by M12 and 10.8 ± 5.0 injections by M24 (3.1 injections between M12 and M24). Adverse events were consistent with the known safety profile of IVT-AFL. CONCLUSION: The 24-month results indicate that, in routine clinical practice, a proactive IVT-AFL regimen achieves functional improvements in patients with treatment-naïve nAMD. The proportion of patients achieving ≥ 70 letters at M24 increased, and patients with baseline BCVA ≥ 70 letters maintained vision regardless of the followed IVT-AFL label. TRIAL REGISTRATION: ClinicalTrials.gov identifier: NCT03939767. A video abstract is available for this article. Supplementary file2 (MP4 364624 KB).

Cerebro-Ocular Toxocariasis Revealed by Bilateral Cystoid Macular Edema.

DESIGN: Case Report Case description: This report describes the case of a female patient diagnosed with oculo-cerebral toxocariasis manifesting initially in the form of isolated bilateral cystoid macular edema. Diagnosis was made by means of positive anterior chamber and lumbar puncture western blots. The unusual presentation, ancillary findings and treatment are discussed. The control of intraocular inflammation that was only partially responsive to steroids was eventually achieved with pegylated interferon alfa-2a. CONCLUSION: Isolated macular edema is a rare presentation of ocular toxocariasis. Interferon alfa-2a may prove useful in case of insufficient control of inflammation.

Natural History and Surgical Outcomes of Lamellar Macular Holes.

PURPOSE: To assess the natural history and surgical outcomes of lamellar macular holes (LMHs). DESIGN: Retrospective and consecutive case series. SUBJECTS: Patients with LMHs from multiple tertiary care centers. METHODS: Clinical charts and OCT scans were reviewed. MAIN OUTCOME MEASURES: The visual acuity (VA) changes and the occurrence rate of full-thickness macular hole (FTMH) were studied in both groups. Within the operated group, factors associated with 6-month VA and development of FTMH were explored. RESULTS: One hundred seventy-eight eyes were included, of which 89 were monitored and 89 underwent surgery. In the observation group, the mean VA decreased from 0.25 ± 0.18 to 0.28 ± 0.18 logarithm of the minimum angle of resolution (logMAR; P = 0.13), with 14 eyes (15.7%) that lost ≥ 0.2 logMAR VA, after 45.7 ± 33.3 months. Nine eyes (10.1%) spontaneously developed an FTMH. In the operated group, the mean VA increased from 0.47 ± 0.23 to 0.35 ± 0.25 logMAR at 6 months (P < 0.001) and 0.36 ± 0.28 logMAR (P = 0.001) after 24.1 ± 30.1 months. By multivariate analysis, better baseline VA (P < 0.001), the presence of an epiretinal membrane (P = 0.03), and the peeling of the internal limiting membrane (ILM; P = 0.02), with a greater effect of ILM perihole sparing, were associated with a greater 6-month VA. Perihole epiretinal proliferation sparing was associated with a better postoperative VA by univariate analysis (P = 0.03), but this was not significant by multivariate analysis. Eight eyes (9.0%) developed a postoperative FTMH. Using Cox proportional hazard ratios [HRs], pseudophakia at baseline (HR, 0.06; 95% confidence interval [CI], 0.00-0.75; P = 0.03) and peeling of the ILM (HR, 0.05; 95% CI, 0.01-0.39; P = 0.004) were protective factors, while ellipsoid zone disruption (HR, 10.5; 95% CI, 1.04-105; P = 0.05) was associated with an increased risk of FTMH. CONCLUSION: Observed eyes with LMH experienced, on average, progressive VA loss. Patients with LMH and altered vision may benefit from surgery. Internal limiting membrane peeling, with perihole ILM sparing, represents a crucial step of the surgery associated with a greater VA and a lower risk of postoperative FTMH. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

PIGMENT EPITHELIAL DETACHMENT THICKNESS AND VARIABILITY AFFECTS VISUAL OUTCOMES IN PATIENTS WITH NEOVASCULAR AGE-RELATED MACULAR DEGENERATION.

PURPOSE: To evaluate the impact of pigment epithelial detachment (PED) thickness (i.e., height) and thickness variability on best-corrected visual acuity outcomes in patients with neovascular age-related macular degeneration in the Phase 3 HAWK and HARRIER trials. METHODS: Optical coherence tomography images from the pooled brolucizumab 6 mg and aflibercept 2 mg arms were analyzed for the maximum PED thickness across the macula at baseline through to week 96. Best-corrected visual acuity outcomes were compared in patients with different PED thickness and variability cut-off thresholds. RESULTS: Greater PED thickness at baseline or at week 12 was associated with lower mean best-corrected visual acuity gain from baseline to week 96 (baseline PED ≥200 µ m: +4.6 letters; <200 µ m: +7.0 letters; week 12 PED ≥100 µ m: +5.6 letters; <100 µ m: +6.6 letters). Eyes with the largest PED thickness variability from week 12 through week 96 gained fewer letters from baseline at week 96 (≥33 µ m: +3.3 letters; <9 µ m: +6.2 letters). Furthermore, increased PED thickness at week 48 was associated with higher prevalence of intraretinal and subretinal fluid. CONCLUSION: In this treatment-agnostic analysis, greater PED thickness and PED thickness variability were associated with poorer visual outcomes in patients with neovascular age-related macular degeneration and greater neovascular activity.

Macular oedema secondary to rhegmatogenous retinal detachment repair: risk factors for resistance to first-line therapy and long-term response to dexamethasone intravitreal implant.

OBJECTIVES: To assess the risk factors for resistance to first-line therapy and long-term response to dexamethasone intravitreal implant (Ozurdex®) of patients with macular oedema (MO) secondary to rhegmatogenous retinal detachment repair (RRDR). METHODS: This was a retrospective, consecutive cohort study conducted in patients who underwent RRDR between January 2014 and December 2020 in the Rothschild Foundation Hospital and experienced postoperative MO (POMO) with a follow-up of at least 18 months. RESULTS: Of the 1152 patients screened, 36 eyes (3.1%) experienced POMO. The mean follow-up duration was 45.2 months (18.0-80.5 months). Twenty-five eyes (69.4%) were resistant to first-line therapy and received at least one Ozurdex® injection (mean number: 2.7 [1-12]). The multivariate analysis showed an increased risk of resistance in patients who underwent perfluorocarbon liquid (PFCL)-assisted drainage (adjusted odds ratio: 8.65; 95% confidence interval: 1.97-15.33; p = 0.01). Significant differences in best-corrected visual acuity and central macular thickness were found between before Ozurdex® injection and the last follow-up visit: from 0.57 ± 0.47 LogMAR to 0.34 ± 0.32 LogMAR (p = 0.02) and from 483.0 ± 124.0 µm to 354.6 ± 96.5 µm (p = 0.001), respectively. The absence of serous retinal detachment and the presence of hyperreflective foci at baseline were associated with a higher resistance and a poorer response to Ozurdex®. Two patients (8%) experienced hypertony, that was well controlled with hypotonic drops. CONCLUSION: MO secondary to RRDR is challenging. Ozurdex® could be reasonably proposed as first-line treatment, at least when MO occurs following PFCL-assisted drainage, given the favourable long-term benefit/risk ratio.

PARACENTRAL ACUTE MIDDLE MACULOPATHY ASSOCIATED WITH SEVERE PLASMODIUM FALCIPARUM MALARIA.

PURPOSE: To report a case of bilateral paracentral acute middle maculopathy lesions on spectral domain-optical coherence tomography(OCT) secondary to severe Plasmodium falciparum malaria. METHODS: Retrospective case report. Spectral domain-OCT, ultra-widefield fluorescein angiography, and OCT angiography were performed and analyzed. RESULTS: A 54-year-old healthy man presented with acute vision loss in both eyes few days after being diagnosed with severe Plasmodium falciparum malaria. Ophthalmoscopic examination was unremarkable, but near-infrared reflectance imaging showed patchy hyporeflective areas located at the terminal tips of the venous branches. Corresponding spectral-domain OCT demonstrated alternating bands of hyperreflectivity involving the inner nuclear layer, consistent with skip paracentral acute middle maculopathy lesions. Optical coherance tomography angiography illustrated corresponding flow signal loss at the level of the deep capillary plexus. Ultra-widefield fluorescein angiography showed peripheral retinal vein staining and capillary nonperfusion. CONCLUSION: Paracentral acute middle maculopathy may be an OCT manifestation of malarial retinopathy associated with severe Plasmodium falciparum infection.

Large language models and their impact in ophthalmology.

The advent of generative artificial intelligence and large language models has ushered in transformative applications within medicine. Specifically in ophthalmology, large language models offer unique opportunities to revolutionise digital eye care, address clinical workflow inefficiencies, and enhance patient experiences across diverse global eye care landscapes. Yet alongside these prospects lie tangible and ethical challenges, encompassing data privacy, security, and the intricacies of embedding large language models into clinical routines. This Viewpoint highlights the promising applications of large language models in ophthalmology, while weighing up the practical and ethical barriers towards their real-world implementation. This Viewpoint seeks to stimulate broader discourse on the potential of large language models in ophthalmology and to galvanise both clinicians and researchers into tackling the prevailing challenges and optimising the benefits of large language models while curtailing the associated risks.