Multimodal deep transfer learning to predict retinal vein occlusion macular edema recurrence after anti-VEGF therapy.

Purpose: To develop a multimodal deep transfer learning (DTL) fusion model using optical coherence tomography angiography (OCTA) images to predict the recurrence of retinal vein occlusion (RVO) and macular edema (ME) after three consecutive anti-VEGF therapies. Methods: This retrospective cross-sectional study consisted of 2800 B-scan OCTA macular images collected from 140 patients with RVO-ME. The central macular thickness (CMT) > 250 µm was used as a criterion for recurrence in the three-month follow-up after three injections of anti-VEGF therapy. The qualified OCTA image preprocessing and the lesion area segmentation were performed by senior ophthalmologists. We developed and validated the clinical, DTL, and multimodal fusion models based on clinical and extracted OCTA imaging features. The performance of the models and experts predictions were evaluated using several performance metrics, including the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity. Results: The DTL models exhibited higher prediction efficacy than the clinical models and experts' predictions. Among the DTL models, the Vgg19 performed better than that of the other models, with an AUC of 0.968 (95 % CI, 0.943-0.994), accuracy of 0.913, sensitivity of 0.922, and specificity of 0.902 in the validation cohort. Moreover, the fusion Vgg19 model showed the highest prediction efficacy among all the models, with an AUC of 0.972 (95 % CI, 0.946-0.997), accuracy of 0.935, sensitivity of 0.935, and specificity of 0.934 in the validation cohort. Conclusions: Multimodal fusion DTL models showed robust performance in predicting RVO-ME recurrence and may be applied to assist clinicians in determining patients' follow-up time after anti-VEGF therapy.

PDGFD switches on stem cell endothelial commitment.

The critical factors regulating stem cell endothelial commitment and renewal remain not well understood. Here, using loss- and gain-of-function assays together with bioinformatic analysis and multiple model systems, we show that PDGFD is an essential factor that switches on endothelial commitment of embryonic stem cells (ESCs). PDGFD genetic deletion or knockdown inhibits ESC differentiation into EC lineage and increases ESC self-renewal, and PDGFD overexpression activates ESC differentiation towards ECs. RNA sequencing reveals a critical requirement of PDGFD for the expression of vascular-differentiation related genes in ESCs. Importantly, PDGFD genetic deletion or knockdown increases ESC self-renewal and decreases blood vessel densities in both embryonic and neonatal mice and in teratomas. Mechanistically, we reveal that PDGFD fulfills this function via the MAPK/ERK pathway. Our findings provide new insight of PDGFD as a novel regulator of ESC fate determination, and suggest therapeutic implications of modulating PDGFD activity in stem cell therapy.

Analysis of Microcirculation Changes in the Macular Area and Para-Optic Disk Region After Implantable Collamer Lens Implantation in Patients With High Myopia.

Myopia has become an important public health problem to be solved urgently. Posterior chamber phakic implantable Collamer lens (ICL) implantation is one of the latest and safest products for myopia correction worldwide. This prospective cross-sectional case series aimed to observe changes in the macular retinal thickness, retinal nerve fiber layer (RNFL) thickness of para-optic disk region, and blood flow density after posterior ICL implantation in patients with high myopia using optical coherence tomography angiography (OCTA). A total of 67 eyes of 67 patients with high myopia, who underwent ICL implantation at The Affiliated Eye Hospital of Nanjing Medical University from January 2020 and December 2020, were included. The spherical equivalent (SE) of the operative eyes was >-6.00 D. The changes in vision, intraocular pressure (IOP), SE, and vault were observed pre-operatively, and follow-up were performed 1 week, 1 month, and 3 months. OCTA was used to observe the changes in the CRT, retinal thickness of paracentral fovea, FAZ, superficial and deep retinal blood flow density in the macular area, RNFL thickness of para-optic disk region, and blood flow density before and after ICL implantation. The uncorrected distance visual acuity (UDVA) and best corrected distance visual acuity (CDVA) of the patients post-operation were significantly improved (P < 0.001). The IOP increased in comparison with other time points at 1 week post-operation (P < 0.05). There were no significant changes in CRT post-operation. The retinal thickness in the upper, lower, nasal, and temporal quadrants of the paracentral fovea increased significantly at 1 month and 3 months post-operation (P < 0.05). The FAZ area at all postoperative time points were decreased (P < 0.001). At 3 months post-operation, the blood flow density of the superficial and deep retinal layers in the upper, lower, and nasal macular area were significantly reduced (P < 0.05). At 1 month post-operation, the RNFL thickness in the temporal para-optic disk region and blood flow density were significantly reduced (P = 0.001 and P < 0.05, respectively). ICL implantation for highly myopic eyes led to an increase of the retinal thickness in the upper, lower, nasal, and temporal regions of the paracentral fovea; reduction of RNFL thickness in the temporal area of para-optic disk; decrease in FAZ area; and decrease in the blood flow density of some deep and superficial retinal layers as well as that of the temporal para-optic disk region.

Modified internal limiting membrane flap technique for large chronic macular hole: Two case reports.

RATIONALE: Internal limiting membrane (ILM) peeling and gas tamponade are the standardized treatments for macular holes (MHs). However, the close rate is low, and postoperative vision is unsatisfactory in large, chronic MHs. Currently, various modifications of the ILM flap techniques are being gradually applied for large MHs in the hope of obtaining better postoperative effects. This study described 2 successful cases achieved by "Sandwich-type" modified ILM flap covering technique in patients with large, chronic MHs. PATIENT CONCERNS: A 62-year-old woman presented with decreased vision and visual distortion of the left eye for 18 months. Optical coherence tomography (OCT) showed the absence of full-thickness neuroepithelial tissue in the central fovea, with a minimum MH diameter of 742 µm and a base diameter of 1630 µm. A 57-year-old man experienced decreased visual acuity for 8 months. OCT showed the absence of full-thickness neuroepithelial tissue in the central fovea, with a minimum MH diameter of 713 µm and a basal diameter of 939 µm. DIAGNOSES: Two patients were diagnosed with large, chronic MH based on the OCT results and duration of the hole. INTERVENTIONS: The 2 patients were treated with the "sandwich-type" modified ILM flap covering technique. OUTCOMES: Large, chronic MH closure was observed using SD-OCT, and the BCVA improved. The patients were very satisfied with the postoperative results. LESSONS: "Sandwich-type" modified ILM flap covering technique may be a safe, effective way for large, chronic MH.

Outcomes of combined treatments in patients with retinal arterial macroaneurysm.

PURPOSE: To evaluate the outcome of the combined approach between intravitreal ranibizumab (IVR) and focal laser photocoagulation (FLP) in the treatment of symptomatic retinal arterial macroaneurysm (RAM). METHODS: A total of 10 patients were included in this clinical case series report. They were diagnosed with symptomatic RAM (one eye in each) and assessed by a comprehensive ophthalmologic examination, including fluorescein angiography (FA), optical coherence tomography angiography (OCT-A), and indocyanine green angiography (ICGA). All patients were treated with an IVR followed by an FLP 2 weeks later. If necessary, a second IVR was given 1 month after the first one (or 2 weeks after the first FLP), which was followed by a second FLP treatment 2 weeks later in the needed cases. All cases were followed up for 6 months after the last treatment. RESULTS: Both the retina hemorrhage and edema were resolved by the treatment. No ocular and/or systemic side effects were evident, and no recrudescence of RAM was seen within the 6 months of follow-up. CONCLUSION: The combined treatment of IVRs and FLPs was successful in the management of symptomatic RAM.

Outcomes of a Foldable Capsular Vitreous Body Implantation: A Retrospective Study.

BACKGROUND: The vitreous body is an important part of the ocular body fluid. A foldable capsular vitreous body (FCVB) is designed to treat chronic adverse complications in severe ocular trauma and silicone oil-dependent eyes. This study is aimed at investigating a method for implanting an FCVB, its postoperative efficacy, and clinical value. METHODS: A retrospective analysis was performed on data from 18 patients who underwent vitrectomy and FCVB implantation for severe ocular trauma and silicone oil-dependent eyes between March 2019 and May 2020. All treated eyes underwent clinical examinations involving the best-corrected visual acuity, intraocular pressure, FCVB position, anterior segment photography, and wide-angle fundus photography regularly after surgery. RESULTS: Eighteen eyes from 18 patients were enrolled in this study. A total of 2.00-4.20 (3.46 ± 0.78) ml of silicone oil were injected into the FCVB during surgery. The patients were followed up at 1, 2, and 4 weeks and 3, 6, and 12 months after surgery. Twelve months after surgery, visual acuity improved in 7 (38.89%) eyes. In contrast, 10 (55.56%) eyes showed no obvious improvement, and 1 (5.56%) eye had decreased vision. Intraocular pressure at 12 months was 10.13 ± 3.52 mmHg, which was comparable to that before the surgery (t = 0.38, P = 0.71). The anterior chamber depth examined by slit lamp was 2.00-3.00 cornea thickness (CT) in 7 eyes, 1.00-2.00 CT in 2 eyes, and <1.00 CT in one eye. The anterior chamber disappeared in eight eyes. There were eight eyes with clear cornea, four eyes with localized opacity, and two eyes with obvious gray-white opacity. There was no case of severe FCVB deflection, rupture, or exposure during the observation period. CONCLUSION: FCVB implantation is an effective and safe treatment for eyes with severe ocular trauma and silicone oil-dependent eyes. It may support retinal reattachment, slow down eyeball atrophy, reduce the risk of chronic adverse complications such as corneal endothelial decompensation, and maintain intraocular pressure and preoperative visual function.

Sequence effect in the treatment of proliferative diabetic retinopathy with intravitreal ranibizumab and panretinal photocoagulation.

PURPOSE: To compare the outcome of the sequence in the two treatments (intravitreal ranibizumab and panretinal photocoagulation) in high-risk proliferative diabetic retinopathy. METHODS: This retrospective study included 35 patients with newly diagnosed high-risk proliferative diabetic retinopathy in 43 eyes; 18 (22 eyes) received intravitreal ranibizumab before panretinal photocoagulation (intravitreal ranibizumab+ group), while the other 17 (21 eyes) received panretinal photocoagulation before intravitreal ranibizumab (panretinal photocoagulation+ group). Each subject received three intravitreal ranibizumabs that were interleaved with three panretinal photocoagulations. The first treatment (either intravitreal ranibizumab or panretinal photocoagulation) was done 1 week before the second one. The interval between intravitreal ranibizumabs was 4 weeks, panretinal photocoagulation was 2 weeks. The power and pulse duration were determined based upon the status of each retinal spot before each panretinal photocoagulation. The retinal non-perfusion region was measured with fundus fluorescein angiography before and 1 month after the final treatment. The central macular thickness was measured with optical coherence tomography within 1 week before the first treatment, before each panretinal photocoagulation, and 1 month after the final intravitreal ranibizumab. RESULTS: The panretinal photocoagulation energy required for effective treatment was lower in intravitreal ranibizumab+ group in the first and second sessions and in total energy (p < 0.05). Central macular thickness reduction before the second panretinal photocoagulation session was significant in the intravitreal ranibizumab+ group (p < 0.05). CONCLUSION: The sequence used in intravitreal ranibizumab+ group showed clear advantages over that in panretinal photocoagulation+ group in the treatment of proliferative diabetic retinopathy, not only in the use of lower energy for panretinal photocoagulation but also in the more rapid regression of neovascularization and less need of additional treatment.

Targeting Keap1 by miR-626 protects retinal pigment epithelium cells from oxidative injury by activating Nrf2 signaling.

Activation of the NF-E2-related factor 2 (Nrf2) cascade can offer significant protection against oxidative stress in retinal pigment epithelium (RPE) cells. Here, we identified a novel kelch-like ECH-associated protein 1 (Keap1)-targeting microRNA, microRNA-626 (miR-626) that activates Nrf2 signaling. In ARPE-19 cells and primary human RPE cells, ectopic overexpression of miR-626 targeting the 3'-UTR (3'-untranslated region) of Keap1 downregulated its expression, promoting Nrf2 protein stabilization and nuclear translocation, leading to expression of ARE-dependent genes (HO1, NOQ1 and GCLC). Functional studies showed that miR-626 protected RPE cells from hydrogen peroxide (H2O2)-induced oxidative injury. Conversely, miR-626 inhibition induced Keap1 upregulation and Nrf2 cascade inhibition, exacerbating oxidative injury in RPE cells. Further studies demonstrated that miR-626 was ineffective in Keap1-knockout or Nrf2-knockout RPE cells. Importantly, miR-626 also activated Keap1-Nrf2 signaling cascade in human lens epithelial cells (HLECs) and primary human retinal ganglion cells (RGCs), providing protection from H2O2. At last, we show that plasma miR-626 levels are significantly downregulated in age-related macular degeneration (AMD) patients than those in the healthy donors. We conclude that targeting Keap1 by miR-626 protects RPE cells and other ophthalmic cells from oxidative injury via activation of Nrf2 signaling cascade.