Volume 62 · Issue 07 · 2026

Neuro-ophthalmic disorders feature complex etiology. Certain ocular manifestations may hint at severe neurological diseases. Current clinical practice faces no shortage of diagnostic examinations. The core challenge consists in inadequate recognition, integration and interpretation of multifaceted information during initial consultation and non-specialist visits, potentially resulting in misdiagnosis, missed diagnosis, delayed referral and excessive examinations. Artificial intelligence has demonstrated promising performance in optic disc lesion detection, image interpretation of optic neuropathy, visual field analysis, as well as eye movement and pupillary function evaluation. Nevertheless, AI is not intended to replace specialists to deliver definitive diagnoses. Instead, it helps boost early detection of high-risk signs, facilitates analysis of complicated test results, and aids clinical decision-making, patient referral and follow-up management. This article elaborates on the application prospects, existing dilemmas and translational strategies of artificial intelligence in neuro-ophthalmology, aiming to offer references for relevant research and clinical practice.

Primary blepharospasm (PBS) is a form of focal dystonia characterized by involuntary, intermittent, or sustained contractions of the orbicularis oculi and periorbital muscles, resulting in involuntary eyelid closure. In clinical practice, PBS has received insufficient attention, and ophthalmologists often have limited experience in its diagnosis and treatment. Its early symptoms can be easily confused with other conditions, such as dry eye disease, leading to misdiagnosis and missed diagnosis, which delays treatment. As PBS is a progressive disorder, it may extend to adjacent regions, increasing treatment difficulty and negatively affecting patients′ quality of life. Currently, there is no curative treatment for PBS, and standardized approaches for symptomatic relief remain lacking. To improve the clinical diagnosis and treatment of PBS, the Ophthalmic and Visual Science Committee of the Chinese Research Hospital Association convened experts from relevant specialties. Based on a comprehensive review and assessment of relevant literature from both domestic and international sources, and following standardized methodologies for guideline and consensus development, the panel formulated consensus-based recommendations addressing 11 clinical questions related to the diagnosis and treatment of PBS. These recommendations are intended to serve as a reference for Chinese ophthalmologists and healthcare professionals involved in the PBS-related health management.

To investigate the role of BCL-2-interacting mediator of cell death (BIM), a pro-apoptotic molecule, in retinal ganglion cell (RGC) injury after optic nerve crush (ONC), and to analyze its association with changes in the expression of inflammation-related genes.

This was an experimental study. The study was conducted from January 2025 to December 2025. Healthy male mice aged 6-8 weeks were randomly divided into four groups: control, ONC, AAV2-scramble, and AAV2-shBim, with 6 mice in each group. The control group received no intervention, the ONC group underwent ONC only, the AAV2-scramble group received intravitreal injection of AAV2-mediated scrambled negative control sequence after ONC, and the AAV2-shBim group received intravitreal injection of AAV2-mediated short hairpin RNA targeting the BIM gene after ONC. Immunohistochemical staining was used to detect the protein expression of Bim, complement component 3 (C3), and lipocalin 2 (Lcn2). Hematoxylin-eosin (HE) staining was used to observe retinal structural changes. Retinal flat-mount immunofluorescence staining was used to assess RGC survival. Optical coherence tomography (OCT) was used to measure ganglion cell complex (GCC) thickness. Flash visual evoked potential (F-VEP) and flash electroretinography (F-ERG) were used to evaluate visual electrophysiological function. RNA sequencing was performed to analyze retinal transcriptomic changes after BIM knockdown. Quantitative real-time PCR (qPCR) was used to validate inflammation-related differentially expressed genes. Independent-sample t-test and one-way analysis of variance were used for statistical analysis.

The proportions of Bim-positive RGCs in the peripheral and central retina were 0.56±0.06 and 0.63±0.06 in the ONC group, respectively, both of which were higher than those in the control group (0.00±0.00) (t=21.60, 24.61; both P<0.001). The numbers of RNA-binding protein with multiple splicing(RBPMS)-positive RGCs in the peripheral and central retina were 74.2±4.4 and 118.5±8.0 in the ONC group, respectively, both of which were lower than those in the control group (222.7±6.0 and 325.0±6.5, respectively) (t=48.94, 48.88; both P<0.001). Significant differences were observed among the four groups in ganglion cell complex thickness, the number of TUJ1-positive RGCs, F-VEP N2-P2 amplitude, and F-ERG b-wave amplitude (F=57.42, 1 216.78, 467.88, 423.76; all P<0.001). In the AAV2-shBim group, ganglion cell complex thickness, the number of TUJ1-positive RGCs, F-VEP N2-P2 amplitude, and F-ERG b-wave amplitude were 54.64±2.61 μm, 242.8±13.1, 11.13±0.80 μV, and 318.00±25.14 μV, respectively, all of which were higher than those in the ONC group [(44.29±1.95) μm, 140.0±5.3, (3.43±0.48) μV, and (190.68±25.50) μV, respectively] (all P<0.001). RNA sequencing showed that the expression levels of the inflammation-related genes C3, CCL12, LCN2, S100A9, CCL6, and ANGPTL4 were lower in the AAV2-shBim group than in the ONC group (t=10.21, 12.02, 8.98, 12.19, 7.33, 9.41; all P<0.001), and the quantitative polymerase chain reaction results were consistent with the RNA sequencing results. The C3-positive cell rates in the central and peripheral retina were 0.11±0.04 and 0.08±0.02 in the AAV2-shBim group, respectively, both of which were lower than those in the ONC group (0.64±0.06 and 0.57±0.05, respectively) (t=18.63, 21.04; both P<0.001). The Lcn2-positive cell rates in the central and peripheral retina were 0.08±0.03 and 0.09±0.02 in the AAV2-shBim group, respectively, both of which were lower than those in the ONC group (0.55±0.06 and 0.48±0.07, respectively) (t=17.19, 12.38; both P<0.001).

BIM expression is upregulated after ONC. AAV2-mediated BIM knockdown alleviates RGC loss, retinal structural damage, and visual electrophysiological dysfunction, accompanied by downregulation of inflammation-related gene expression.

To construct an inducible carnitine acetyltransferase (Crat) gene expression system and provide a controllable experimental platform for research on neuroprotection in glaucoma.

This experimental study was conducted from January 2025 to November 2025 using the mouse hippocampal neuronal HT22 cell line. A high-efficiency tetracycline-inducible lentiviral gene expression system was established by optimizing the multiplicity of infection (MOI) and polybrene concentration, as well as comparing the activities of the cytomegalovirus (CMV) promoter and the elongation factor 1α (EF1A) promoter. According to promoter type, the cells were divided into two groups: the CMV promoter group, in which enhanced green fluorescent protein (EGFP) fluorescence was detected, and the EF1A promoter group, in which mCherry fluorescence was detected. Conditional Crat overexpression was induced by doxycycline (Dox). Unmodified HT22 cells were assigned to an uninfected control group and an uninfected Dox-treated group, whereas successfully constructed HT22-TRE-mCrat/rtTA cells were assigned to an infected non-Dox-treated group and an infected Dox-treated group. The neuroprotective effect of the Crat overexpression system was evaluated in an H₂O₂-induced oxidative injury model by measuring lactate dehydrogenase (LDH) release, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) staining, and the mRNA expression levels of apoptosis-related factors, including caspase-3, B-cell lymphoma 2-associated X protein (Bax), and B-cell lymphoma 2 (Bcl-2), using quantitative PCR (qPCR). Statistical analyses were performed using the independent-samples t-test or one-way analysis of variance, with Tukey′s test for pairwise comparisons.

A tetracyline-inducible gene expression system was successfully constructed and stably established in HT22 cells. The EF1A promoter showed higher driving efficiency than the CMV promoter, with a significantly greater number of positive cells in the EF1A promoter group than in the CMV promoter group [(303.0±28.0) vs. (88.7±11.4) cells; t=17.37, P<0.001]. In the H₂O₂-induced oxidative injury model, the LDH release rate was significantly lower in the infected Dox-treated group than in the uninfected Dox-treated group (27.1%±4.7% vs. 49.8%±8.2%; P<0.001). The apoptotic cell rate determined by TUNEL staining was significantly lower in the infected Dox-treated group than in the uninfected Dox-treated group (14.1%±3.4% vs. 41.3%±4.1%; t=16.23, P<0.001). The caspase-3 mRNA level was significantly lower in the infected Dox-treated group than in the uninfected Dox-treated group (140.3±17.1 vs. 192.0±24.3; t=4.26, P=0.002). Bax mRNA expression was significantly lower in the infected Dox-treated group than in the uninfected Dox-treated group (132.0±13.1 vs. 169.8±12.6; t=5.09, P<0.001). Bcl-2 mRNA expression was significantly higher in the infected Dox-treated group than in the uninfected Dox-treated group (83.2±6.9 vs. 58.3±8.1; t=5.74, P<0.001). All of these differences were statistically significant.

This study successfully established a controllable Crat overexpression platform using the mouse hippocampal neuronal HT22 cell line, enabling temporal regulation of Crat gene expression and cell-specific studies in this cell line.

To investigate the molecular mechanism of toll-like receptor adaptor protein SARM1 in glaucomatous optic neuropathy.

The experimental study was conducted from February 2024 to October 2025. A chronic ocular hypertension glaucoma model was established by injecting micro-magnetic beads into the anterior chamber of 8- to 10-week-old male Wistar rats. At 3 days, 1 week, and 2 weeks post-modeling, retinal and optic nerve tissues from 6 eyes of 6 rats were collected as the chronic ocular hypertension glaucoma model group, and 6 eyes from 6 wild-type rats that received an equal volume of saline via anterior chamber injection served as the control group. Intraocular pressure was measured using a TonoLab tonometer. Retinal whole-mounts were prepared and POU domain class 4 transcription factor 1 (POU4F1 or Brn3A) immunofluorescence staining was used to detect retinal ganglion cell loss. Western blotting was performed to detect the expression levels of SARM1 and SNPH in the rat retina and optic nerve. Immunofluorescence staining was used to examine their distribution in these tissues. Furthermore, CRISPR/Cas9 technology was used to knock down the expression of SARM1 and SNPH in mouse 661W retinal ganglion cells, respectively. Cells were collected 48 hours after transfection, and Western blotting was performed to detect the expression levels of SARM1 and SNPH. Normally distributed continuous data are presented as mean±SEM. Comparisons between two groups were performed using the Student′s t-test, while comparisons among multiple groups were assessed by the one-way analysis of variance followed by the Tukey′s multiple comparisons test.

Western blot analysis revealed that in the glaucoma model group, the relative expression level of SARM1 protein in the optic nerve at one week post-modeling (1.22±0.06) was significantly higher than that in the control group (1.03±0.01; P=0.027, q=4.38). In contrast, the expression level of SARM1 in the retina at three days post-modeling (0.79±0.02) was significantly lower than that in the control group (1.04±0.03; P<0.001, q=6.86). Concurrently, the expression level of SNPH at three days post-modeling (0.74±0.01) was lower than that in the control group (1.03±0.04; P=0.040, q=0.58), and its expression at one week post-modeling (1.19±0.10; P=0.002, q=4.36) was significantly higher than that at three days (0.74±0.01). Consistent with the Western blot results, immunofluorescence staining results showed that in the optic nerve of the glaucoma model group, the expression of SARM1 was significantly higher than that in the control group at one week post-modeling, while the expression of SNPH was lower than that in the control group at three days post-modeling. Both proteins partially co-localized with the neuronal marker β3-tubulin. Immunofluorescence staining also revealed co-localization of these two proteins within axons. Additionally, SARM1 co-localized with the mitochondrial marker protein TOM20. Western blot results from 661W cells showed that the knockdown of SARM1 expression (0.54±0.04) significantly reduced SNPH expression (0.54±0.05; P=0.003, q=7.98), whereas the knockdown of SNPH expression (0.39±0.06) did not markedly affect SARM1 levels (0.75±0.05; P=0.010, q=6.39).

The elevated expression of SARM1 protein in the axons of the rat glaucoma model can promote retinal ganglion cell axonal pathology by localizing to axonal mitochondria and regulating SNPH expression.

To evaluate the ocular surface changes and their correlation with filtering bleb morphology after phacoemulsification cataract extraction with intraocular lens implantation (PEI) combined with trabeculectomy for primary angle-closure glaucoma (PACG).

A retrospective case-series study. PACG patients who underwent PEI combined with trabeculectomy between October 2021 and March 2023 were included. Preoperatively and at 3 and 6 months postoperatively, best corrected visual acuity (BCVA), intraocular pressure (IOP), surgical complications, 5-fluorouracil subconjunctival injection, and ocular surface data, including the noninvasive first tear film breakup time (NIF-BUT) and average tear film breakup time, tear meniscus height, tear film dynamic scattering index, Schirmer Ⅰ test, corneal staining score, meibomian gland function score, and ocular surface disease index questionnaire (OSDI), were recorded. At 6 months postoperatively, filtering blebs were graded according to the Indiana Bleb Appearance Grading Scale to explore the correlation between bleb morphology and ocular surface function.

A total of 51 patients (51 eyes) were included, comprising 18 males (35.3%) and 33 females (64.7%), with 22 left eyes (43.1%) and 29 right eyes (56.9%). The mean age was 63.9±7.6 years, and the mean follow-up duration was 7.5±2.1 months. The BCVA values at baseline, 3 months, and 6 months postoperatively were 0.22 (0.1, 0.52), 0.15 (0.05, 0.4), and 0.15 (0.1, 0.22), respectively. At 6 months, the BCVA was significantly improved compared with baseline (P=0.004). The IOP at baseline, 3 months, and 6 months was 26.9 (18.6, 33.2) mmHg(1 mmHg=0.133kPa), 12.8 (11.0, 16.3) mmHg, and 12.9 (11.2, 14.5) mmHg, respectively, and was significantly lower at both 3 and 6 months than that at baseline (P<0.001). The complete success rates at 3 and 6 months were 98.0% and 96.1%, respectively, and the qualified success rates were both 100%. No statistically significant differences were found in the noninvasive average tear film breakup time, tear meniscus height, tear film dynamic scattering index, Schirmer Ⅰ test, or corneal staining score among preoperative and postoperative 3-month and 6-month data. The NIF-BUT values at baseline, 3 months, and 6 months were 7.46 (4.40, 11.47) s, 5.22 (3.50, 8.15) s, and 7.84 (5.93, 11.24) s, respectively; the NIF-BUT at 3 months was significantly lower than that at 6 months (P=0.004). The OSDI scores at baseline, 3 months, and 6 months were 30.56 (25.00, 38.89), 23.47 (17.56, 32.88), and 25.00 (17.62, 30.56), respectively; the score at 3 months was significantly lower than that at baseline (P=0.016). The meibomian gland scores at baseline, 3 months, and 6 months were 3 (2, 3), 3 (2, 4), and 3 (3, 4), respectively; the score at 6 months was significantly higher than that at baseline (P=0.003). At 6 months postoperatively, except for the positive correlations of the corneal staining score with bleb vascularity and the number of 5-fluorouracil injections (r=0.405, P=0.003; r=0.401, P=0.003), no significant correlations were found between the bleb grading or number of 5-fluorouracil injections and the postoperative visual acuity, IOP, other ocular surface parameters, or OSDI scores.

The PACG patients can achieve good visual improvement and high surgical success rates after PEI with trabeculectomy. A transient decline in the objective ocular surface parameter NIF-BUT and the subjective OSDI score can be observed at 3 months postoperatively, with recovery by 6 months. The meibomian gland function gradually deteriorates over time. The corneal staining can be positively correlated with the postoperative bleb vascularity and the number of 5-fluorouracil injections.

A 54-year-old male patient presented for medical attention with right eyelid swelling and visual loss for half a month following a wasp sting to his right eye. His chief complaints included visual loss in the right eye, accompanied by ocular redness, pain, and lacrimation. Ophthalmic examinations revealed severe visual impairment, eyelid edema, and conjunctival hyperemia in the right eye. The cornea, anterior chamber, iris, and lens were all affected, and the fundus was not visible. Auxiliary examinations showed ciliary body detachment, severe vitreous opacity, and suspected exudative retinal detachment in the right eye. After admission, the patient received conservative treatments including systemic and topical anti-inflammatory therapy and mydriasis, with no obvious improvement. Vitrectomy combined with lensectomy was then performed. Intraoperatively, the retina was attached, but optic nerve atrophy and retinal vascular abnormalities were detected. No visual improvement was observed in the patient postoperatively.

Increased intracranial pressure (ICP) is commonly induced by neurological diseases, and early diagnosis and treatment are crucial for improving patient prognosis. Although invasive intracranial pressure monitoring via external ventricular drainage (EVD) serves as the gold standard, it is not suitable for early screening. Multimodal imaging approaches for ICP evaluation include optic nerve sheath diameter (ONSD) measurement, optical coherence tomography (OCT), and transcranial Doppler ultrasound. Among these techniques, ONSD measurement is the most convenient modality, and OCT has also been validated for the early diagnosis of elevated ICP. This review elaborates on the clinical value, diagnostic efficacy, and applicable scenarios of three ONSD measurement techniques and OCT in predicting increased ICP. It clarifies that multimodal ocular imaging helps optimize the monitoring system for elevated ICP and provides evidence-based guidance for its early diagnosis.

Since the release of the Evidence-based guidelines for diagnosis and treatment of demyelinating optic neuritis in China (2021), significant progress has been achieved in the diagnostic classification, biomarker detection and targeted therapy of optic neuritis. A quantitative evaluation system combining multimodal imaging and serological antibody testing has been established for clinical application. The updated McDonald criteria have incorporated the optic nerve into the lesion localization of multiple sclerosis. Detection of aquaporin-4 (AQP4) and myelin oligodendrocyte glycoprotein (MOG) antibodies serves as a key tool for the definitive diagnosis of neuromyelitis optica spectrum disorder and MOG antibody-associated disease. Plasma exchange can improve the prognosis of severe cases, and biological agents represent the first-line therapeutic strategy for AQP4-positive patients. Considering the current domestic status of antibody detection capabilities and drug accessibility, this article proposes implementing routine antibody screening at initial diagnosis, optimizing multidisciplinary collaboration models, and constructing a real-world clinical data system.