Ferdinand K. Hui

BACKGROUND AND OBJECTIVES: Cerebral venous outflow disorders (CVD) are conditions characterized by impairment of cerebral venous drainage. Associated symptoms have been poorly defined, as no standardized diagnostic criteria exist for this class of conditions; most studies to date have attempted to report on symptoms solely based on suggestive noninvasive imaging findings, which is prone to selection bias. The aim of this study was to characterize symptoms that may be present in the setting of suspected internal jugular vein outflow stenosis. METHODS: A survey was conducted among patients with suspected CVD who had previously undergone diagnostic dynamic cerebral venography and subsequent internal jugular vein stenting or styloidectomy from 2019 to 2025. RESULTS: Following distribution to 133 patients, 60 (45.1%) responded. Fifty (83.3%) of the 60 respondents indicated an improvement in symptoms following intervention. Brain fog (49, 98.0%), headache (48, 96.0%), and fatigue/low energy (46, 92.0%) were the most frequently reported symptoms pre-procedure. Uncontrollable seizures or shaking spells, headache, and eye pain were the most responsive to intervention, with improvement rates of 87.5%, 81.3%, and 81.0%, respectively. Nineteen of 50 respondents endorsed a diagnosis of Postural Orthostatic Tachycardia Syndrome; 8 of 19 reported improvement in Postural Orthostatic Tachycardia Syndrome symptoms after their procedure. CONCLUSION: There is a wide range of symptoms that may be present in patients with CVD. This survey illustrates the most common complaints among suspected patients with CVD and elicits which symptoms are likely due to cerebral venous outflow impairment, based on responsiveness to intervention.

Post-stroke aphasia is a common and disabling consequence of ischemic stroke, substantially affecting communication and quality of life. Early identification of patients at risk is essential for prognosis and rehabilitation planning. Diffusion tractography has recently been employed to map white matter pathways implicated in language processing and to evaluate their predictive value for post-stroke aphasia. In this study we aimed to evaluate the prognostic value of diffusion tensor imaging characteristics of key white matter tracts. A systematic search was conducted in PubMed, Embase, and Scopus until April 2025. Eligible studies applied diffusion MRI tractography to assess structural language pathways in patients with stroke and investigated their association with post-stroke aphasia occurrence or severity. Extracted data included study characteristics, tractography methodology, language assessment tools, and key imaging findings, including tract-specific diffusion metrics and tractography-derived measures. Due to data insufficiency, qualitative synthesis was conducted. Twenty-one studies comprising 550 patients with post-stroke aphasia were included. Participants were predominantly older adults, with mean/median ages generally in the late 50s to early 70s. Baseline NIHSS values were commonly in the ∼5-14 range. The timing of diffusion/DTI acquisition varied substantially, spanning hyperacute/acute imaging, early subacute assessments, and later evaluations in the chronic phase. Studies identified left arcuate fasciculus (AF) integrity, fractional anisotropy (FA), axial diffusivity (AD), or reconstruction status, as the strongest single predictor of aphasia severity and recovery. Loss of AF reconstruction was associated with markedly lower follow-up AQ. Ventral-stream involvement added prognostic value in several studies. Right-hemisphere contributions were inconsistent, though higher callosal anisotropy predicted comprehension gains in one cohort. Overall, tract-specific microstructure improved prognostic models beyond lesion volume or NIHSS, accounting for an additional 12-29% of outcome variance in several multivariate analyses. Methodological variability limits cross-study comparability and precluded quantitative pooling and as a result, generalizability. Tractography provides valuable insights into the structural substrates of post-stroke aphasia, with the arcuate fasciculus emerging as the most robust predictor. Standardization of tractography protocols and multicenter validation are needed before routine clinical translation.

Timely treatment of ischemic stroke is critical, yet access to specialized neurointerventionalists remains limited, particularly in remote or underserved areas. This gap in care delivery motivates the development of technologies enabling remote neurointerventions. This feasibility study evaluated the performance of a magnetically steerable guidewire system, operated with manual catheter advancement, in comparison to standard manual navigation techniques. Experiments were conducted in an in-vitro silicone vascular model simulating both standard and complex anatomies. Nine neurointerventionalists participated after less than one hour of pre-procedural training on the robotic system. Procedures were performed via femoral and radial access routes. The magnetically steerable guidewire system demonstrated procedural success and navigation times comparable to standard manual navigation methods, with no statistically significant differences. Importantly, the magnetically steerable guidewire system approach significantly reduced the average number of devices required per procedure (three vs. five, p < 0.0001), suggesting a simplification of procedural workflow. The magnetically steerable guidewire system offers a promising step toward enabling remote neurointerventions by maintaining procedural efficacy while reducing device complexity. These findings support further development of robotic systems to assist in stroke treatment that are both effective and operationally efficient.

Prolonged venous transit (PVT) on CT perfusion is an established marker of poor collaterals and stroke severity; feasibility on MR perfusion (MRP) is uncertain. We evaluated feasibility and prognostic value of PVT on MRP in 59 patients with suspected acute ischemic stroke using a "straight-to-MRI" workflow. Board-certified neuroradiologists visually defined PVT as Tmax ≥10 s in the superior sagittal sinus and/or torcula. Pearson correlation tested associations between binary PVT and discharge NIHSS and 90-day modified Rankin Scale (mRS) in the full cohort and treated sub-cohort (IV thrombolysis and/or thrombectomy). PVT correlated with worse outcomes in the full cohort (NIHSS r=0.469, p=0.002; mRS r=0.387, p=0.02) and more strongly in treated patients (NIHSS r=0.678, p=0.005; mRS r=0.631, p=0.01). These findings demonstrate that PVT can be assessed on MRP and is a potentially promising post-revascularization prognostic biomarker aligning with CTP-based physiology; larger studies are needed.

Idiopathic intracranial hypertension (IIH) is a complex and increasingly prevalent disorder that results in significant morbidity despite a broad array of available treatments. While lifestyle modifications, pharmacologic agents, and surgical interventions can produce meaningful short-term symptomatic improvements, long-term outcomes are poor, characterized by high rates of symptom recurrence. Weight loss remains the only disease-modifying therapy, though sustained reductions in weight are rarely achieved outside of bariatric surgery. In addition, a subset of patients is not overweight, limiting applicability. Pharmacologic therapies such as acetazolamide, topiramate, and glucagon-like peptide-1 (GLP-1) receptor agonists offer benefits but are limited by the side effect profile and poor efficacy. Surgical approaches often address only a portion of IIH’s multifactorial pathophysiology, and recurrent symptoms may arise from persistent venous hypertension or new venous stenoses. This review evaluates the current evidence on medical and surgical treatment failures in IIH, emphasizing unresolved questions in disease pathogenesis and the need for personalized therapeutic approaches that advance the development of more durable treatment options.

Background: Dynamic internal jugular vein (IJV) stenosis is increasingly recognized in patients with cerebral venous outflow disorders (CVD). Although the hemodynamic effects of dynamic jugular stenosis have been characterized, its role in disease remains undefined. This study evaluated the safety and early outcomes of IJV stenting for symptomatic rotational stenosis. Methods: This retrospective, single-institution study included adult patients that underwent IJV stenting for dynamic, symptomatic IJV stenosis between 2023 and 2025. Inclusion required dynamic venography demonstrating >75% rotational IJV stenosis and ≥6 mmHg trans-stenotic pressure gradient. Demographic, procedural, and clinical data were analyzed. Symptom severity was assessed using the Cerebral Venous Disorder Symptom Severity (CVDSS) scale and the Headache Impact Test (HIT-6). Results: Sixteen patients (mean age 37.1 years, 75% female) were included. All underwent successful IJV stent placement without periprocedural complications. The median trans-stenotic pressure gradient improved from 7 mmHg (range 6–18) to 1 mmHg (range 0–9) after stenting. CVDSS scores improved from 13.5 (11.0–16.0) to 8.5 (4.75–12.0) at peak improvement (p<0.001), with sustained improvement (10.5 (8.0–12.3), p<0.001) at a median follow-up of 10.1 months. Median HIT-6 scores decreased from 65.5 (64.0–67.3) to 61.0 (57.0–64.0) (p<0.001). Thirteen patients reported meaningful symptom improvement at last follow-up, and 88% were satisfied with their decision to pursue stenting. Conclusion: Jugular vein stenting in refractory CVD patients with severe, symptomatic rotational IJV stenosis is technically feasible and associated with meaningful clinical improvement in patients though partial symptom recurrence is common. Optimal patient selection and procedural expertise are essential to maximize patient safety and efficacy.

Background: Dynamic internal jugular vein (IJV) stenosis has been demonstrated with 90 degree head rotation, often producing large trans-stenotic pressure gradients and elevations in intracranial venous pressures. However, the prevalence of IJV stenosis and occlusion with partial head rotation remains unknown. Methods: This retrospective, multi-institutional study included patients with suspected cerebral venous outflow disorders that underwent diagnostic cerebral venography with pressure manometry across progressive head rotations (neutral, 25 degrees, 45 degrees, and 90 degrees bilaterally). Superior sagittal sinus and IJV pressures were recorded, and the degree of IJV stenosis and associated pressure gradients were quantified. Results: Twenty patients underwent cerebral venography and pressure manometry with fractional head rotation. The mean age of the patients in this cohort was 43.4 years, and 70% of the patients were female. Severe or occlusive stenosis occurred in 15–20% of patients at 25 degrees of ipsilateral head rotation, 40% of patients at 45 degrees of ipsilateral head rotation, and 70–80% of patients at 90 degrees of ipsilateral head rotation. The highest median pressure gradient occurred with ipsilateral 90 degree head rotation in both the right and left IJV, measuring 6.0 mmHg and 5.0 mmHg respectively. Trans-stenosis pressure gradients and superior sagittal sinus pressures increased in a stepwise fashion with greater degree of head rotation. Conclusion: Fractional head rotation frequently produces IJV stenosis and significant pressure gradients. Even modest head turns can precipitate severe stenosis and increased intracranial venous pressures, suggesting that dynamic jugular stenosis and occlusion may play a greater role in daily symptom exacerbation than previously recognized.

Background: As large language models (LLMs) become increasingly accessible to the public, patients are turning to these tools for medical guidance — including in highly specialized fields like interventional neuroradiology. Despite their growing use, the safety, completeness, and reliability of LLM-generated information in subspecialty medicine remain unclear. Methods: Five publicly available LLMs — ChatGPT, Gemini, Claude, Perplexity, and DeepSeek — were prompted with four neurointerventional patient-facing clinical questions spanning ischemic stroke, hemorrhagic stroke, venous disorders, and procedural device use. Each model was queried three times per question to generate unique responses. Eight blinded raters scored each response on accuracy, completeness, safety, and actionability using Likert scales. Plagiarism analyses were also performed. Results: DeepSeek consistently outperformed other LLMs in accuracy, completeness, and actionability across four prompts, while Gemini frequently ranked worse, including in plagiarism levels. ChatGPT performed well in accuracy. Physicians were more critical than non-physicians across accuracy, completeness, and safety, whereas non-physicians rated actionability significantly lower. Overall, LLMs were rated relatively high (median of >4 on a 5-point scale) in medical safety, suggesting low risk of overtly harmful advice. Conclusion: Recent-generation LLMs offer medically safe, though often incomplete or imprecise, information in response to patient-oriented neurointerventional queries. Including non-physician raters revealed valuable differences in perception that are relevant to how patients may interpret LLM outputs. As benchmark frameworks like HealthBench improve LLM evaluation, inclusion of lay perspectives and subspecialty contexts remains essential. Responsible use by clinicians and ongoing patient education will be critical as LLM use in healthcare expands.