Interventional Neuroradiology最新文献

BackgroundQuantitative angiography (QA) can extract hemodynamic information during neurointerventional procedures by leveraging contrast flow biomarkers. However, clinical adoption of two-dimensional (2D) QA remains limited compared with three-dimensional (3D) methods such as CT perfusion due to view-dependent biases when projecting 3D flow onto 2D images. Variations in tube potential (kVp) further modulate iodine attenuation, introducing intensity inconsistencies that confound QA measurements. This study evaluates a path-length correction (PLC) method designed to reduce orientation- and acquisition-related biases in 2D digital subtraction angiography (DSA).Materials and methodsThree cerebrovascular cases containing rotational and 2D DSAs were retrospectively analyzed. 3D volumes were reconstructed using cone-beam algorithms, and synthesized projections were spatially aligned with 2D DSA images using affine and non-linear transformations. Frame intensities were normalized for kVp using an iodine mass attenuation lookup. Path-length maps derived from aligned 3D volumes were then used to normalize DSA intensities, generating PLC images. QA parameters-peak height and area under the curve-were computed from pixel-wise time-density curves (TDCs) at matched regions of interest within aneurysm and parent vessels. PLC performance was evaluated by comparing root mean squared errors (RMSEs) between frontal and lateral TDCs and discrepancies in QA parameters before and after correction.ResultsAcross three cases, PLC improved cross-view consistency by reducing TDC RMSE from 0.23 ± 0.04 to 0.14 ± 0.04, peak height RMSE from of 0.42 ± 0.16 to 0.15 ± 0.11, and area under the curve RMSE from 0.43 ± 0.13 to 0.14 ± 0.13.ConclusionsThe PLC method reduces foreshortening bias in 2D DSA and improves consistency of QA metrics, enhancing reliability in cerebrovascular assessment and treatment evaluation using clinical DSA.

BackgroundBalloon guide catheters (BGCs) have been associated with higher rates of first pass effect (FPE) in endovascular thrombectomy. Large-bore aspiration catheters (ACs) placed at the face of the embolus may mimic the flow arrest effect seen with BGCs; combined use and its effect on FPE have not been previously well-studied. This study aims to evaluate the impact of this dual-device strategy on achieving FPE.MethodsWe conducted a retrospective cohort study across three sites (2018-2024). A multicenter registry was queried for adult patients with anterior circulation acute ischemic stroke treated with thrombectomy. The primary outcome was rate of FPE (modified Treatment in Cerebral Infarction (mTICI) 2C/3 on the first pass). Multivariable logistic regressions for likelihood of FPE were performed.Results973 patients were included, with a median age of 69 years (IQR 59-79). BGCs were used in 26.3%. Large-bore ACs were used in 68.7%. In most cases, large-bore ACs were used without a BGC (BGC-: 77.5% vs. BGC+: 43.8%, P < .001). FPE was achieved in 3-0.6% of procedures. BGC usage was significantly associated with FPE (aOR: 1.94, 95% CI: 1.01-3.72, P = .04). When sub-stratified by BGC use, large-bore AC use was found to be significantly associated with FPE only when a BGC was not used (aOR: 3.34, 95% CI: 1.01-12.14, P = .04).ConclusionBGC use is associated with an increased likelihood of FPE; when BGCs were not used, large-bore ACs were significantly associated with a higher likelihood of FPE. These findings are important when considering which tools contribute to successfully achieving FPE.

BackgroundEpidural contrast extravasation (ECE) at the level of lumbar puncture site during CT myelography (CTM) is a noted phenomenon without determined clinical relevance. While prior randomized studies in spinal analgesia have shown benefits of pencil-type spinal needles compared to cutting-type for dural punctures, rates of ECE on imaging have not been closely studied, and no prior study has investigated the effect of spinal needle type technical factors on ECE rate.MethodsAll CTM cases over an 8-year period for any indication (e.g., spontaneous intracranial hypotension CSF leak evaluation) were retrospectively reviewed. A large series of cases (n = 276) were reviewed by two independent neuroradiologists for presence of puncture site ECE (n = 63), ECE extending at least one vertebral body level (n = 15), or no ECE (n = 198). Rates were compared by several technical factors including spinal needle type, gauge, and puncture site level.ResultsPencil-type spinal needles had significantly greater rates of any ECE (32%) versus cutting-type (25%, p = 0.030) and for ECE at puncture site only (25% vs. 20%, respectively) and ECE with extension greater than one vertebral body level (7% vs. 4%, respectively, p = 0.004). The rate of ECE did not differ by needle gauge, needle type/gauge combinations, or level accessed (p > 0.05). The need for post-dural puncture targeted epidural blood patch did not differ by presence of ECE (p = 0.190).ConclusionsThe rate of ECE in CTM was common (28% of cases) and occurred with slightly greater frequency with pencil-type spinal needles. As CTM and the use of pencil-type spinal needles become more prevalent in the investigation of spinal CSF leak, it is important to convey that asymptomatic ECE can be a common expected post-dural puncture finding with both pencil-type and cutting-type spinal needles and should not be confused for the site of leak when evaluating patients for spontaneous spinal CSF leak. Further, our results that immediate ECE do not correlate with symptomatic post-dural puncture headache differ from a recent retrospective series, and the imaging finding alone of ECE does not warrant further evaluation in an asymptomatic patient.

PurposeTo develop and implement a novel tool for evaluating neointimal healing over flow diverting aneurysm devices in vitro using a biocompatible, dye-based flow visualization method.MethodsBiocompatibility of Brilliant Blue FCF (BB FCF) dye was established using an alamarBlue metabolic viability assay. Next, a custom MATLAB image analysis script to quantify three intra-aneurysmal dye transport parameters (fill time, washout time, and max intensity) was evaluated through benchtop testing. Video recordings of BB FCF injections in four silicone aneurysm models with increasing levels of device-facilitated occlusion (0-100%) were used to quantify corresponding changes in dye transport parameters. Finally, the method was applied in a 14-day endothelialized silicone blood vessel mimic (BVM) study to assess the impact of repeated BB FCF injections on vessel construct morphology.ResultsBB FCF was noncytotoxic at concentrations ≤500 µM. Dye transport parameters differed as expected between occlusion models; fill and washout times were prolonged in partially occluded models compared to the patent control (p < .01), and maximum dye intensity decreased across all models as a function of occlusion (p < .0001). Repeated BB FCF injections in BVMs showed no morphological differences between endothelial linings in injection-treated vessels vs. no-injection controls.ConclusionThe methods and results documented in this study demonstrate that a non-angiographic dye-based flow visualization method has potential to provide a repeatable, non-destructive way to assess gradual healing-mediated occlusion performance of flow diverter devices in vitro, complementing traditional imaging techniques. This method lays the groundwork for correlating flow-based transport metrics with endothelial coverage to better understand healing-based aneurysm occlusion.

BackgroundAs 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.MethodsFive 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.ResultsDeepSeek 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.ConclusionRecent-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.

BackgroundMechanical thrombectomy (MT) has become the standard of care for ischemic stroke caused by large vessel occlusion (LVO). Direct aspiration techniques (ADAPT) have demonstrated efficacy comparable to stent retrievers (SRs).ObjectiveTo describe the safety and efficacy of the Raptor catheter as a first-line aspiration device for LVO MT in a national multicenter setting.MethodsWe prospectively, enrolled 46 patients with acute ischemic stroke due to LVO. Patients were treated with the Raptor catheter as a first-line ADAPT technique. Demographic, clinical, and procedural variables were analyzed. Efficacy and safety outcomes were recorded. Data was analyzed using descriptive statistics.ResultsMedian age was 66 years (IQR 59-77), with 39.1% women. The median National Institutes of Health Stroke Scale at admission was 18 (IQR 13.5-21), and the M1 segment accounted for 67.4% of the LVOs. First-pass effect (FPE), defined as mTICI ≥2c, was achieved in 54.3% (25/46), and final recanalization mTICI ≥2b in 93.5% (43/46). The median number of passes was 1 (IQR 1-2), and the mean puncture-to-recanalization time was 25.7 ± 28.4 min. Rescue SRs were used in 17.4% (8/46). Complications were documented in five patients (10.9%) with symptomatic intracranial hemorrhage in 3/46. Ninety-day mortality was 6.5%. At 90 days follow-up, 61.9% were functionally independent.ConclusionThe Raptor catheter proved to be safe and effective as a first-line strategy for LVO MT, with recanalization and FPE rates comparable to other large-bore aspiration devices with low rates of intracranial complications and mortality. Comparative and more extensive studies are needed to validate these findings.

BackgroundComputer-generated holograms (CGHs) are advanced, glassless, three-dimensional (3D) representations. This study reports the first surgical application of CGH for preoperative and intraoperative assistance.MethodsIn this single-center case series and observational survey study, three consecutive patients with intracranial aneurysms were scheduled for endovascular treatment; in one case, observation was selected instead of treatment. Twelve neurosurgical residents were recruited to inspect the CGHs and complete a questionnaire. CGHs were inspected pre- and intraoperatively in two patients with flow-diverter stents. For validation, neurosurgical residents inspected conventional 3D rotational angiograms, then CGHs, and completed a questionnaire. We compared correct answer rates on a questionnaire assessing 3D anatomical understanding at 10 locations using CGHs versus conventional methods. Simulator Sickness Questionnaire (SSQ) scores were used to assess safety. Outcomes were defined a priori.ResultsCorrect answer rates were higher with CGH than with two-dimensional working angle images (median, 75% vs. 60%; p < .01). Mean SSQ scores were nausea,0.80; oculomotor, 13.9; disorientation, 12.8; and total severity, 10.6. Intraoperative CGH reference particularly assisted the operating surgeon in visualizing and simulating the course of the guidewire across the neck in flow-diverter placement of a giant internal carotid artery aneurysm.ConclusionsThe 3D visualization of anatomical structures in CGH was quantitatively validated in the present study. Comprehension of CGH did not require particular training or prior experience, and its safety profile was also evaluated. CGH may serve as an effective tool for surgical assistance and education, particularly in anatomically complex cases.

The use of vertebrobasilar junction (VBJ) traversal provides an effective alternative for treating complex posterior circulation aneurysms when standard ipsilateral access is restricted, by enabling retrograde access to the posterior inferior cerebellar artery (PICA) branching from the targeted aneurysm. While experienced operators have employed this technique, a formal video-based demonstration of the approach is notably absent from the existing educational literature, underscoring the significance of this contribution as a resource for trainees and early-career neurointerventionists. We present two cases illustrating the utility of contralateral vertebral artery access via VBJ traversal. In the first case, a recurrent left PICA aneurysm was treated with stent-assisted coiling. In the second case, a ruptured left V4 dissecting aneurysm was managed with balloon-assisted coil sacrifice. In both cases, bilateral radial access facilitated a favorable microcatheter trajectory through the contralateral vertebral artery, enabling effective aneurysm treatment while preserving PICA patency. Bilateral radial access has been associated with reduced complication rates and improved procedural control in recent literature. Both patients had excellent clinical and angiographic outcomes, with no recurrence observed on follow-up imaging. Written informed consent was obtained from both patients for the procedures performed and for the inclusion of their anonymized clinical data and images in this report. These cases demonstrate the feasibility and potential benefits of VBJ traversal in treating posterior circulation aneurysms, particularly when conventional access strategies are technically challenging.

BackgroundChronic subdural hematoma (CSDH) is increasingly treated with middle meningeal artery embolization (MMAE). This study evaluated MMAE using a precipitating hydrophobic injectable liquid low viscosity (PHIL-LV) liquid embolic agent under general anaesthesia.MethodsConsecutive patients who underwent MMAE for CSDH with PHIL-LV at a single comprehensive neuroscience centre during a two-year period (2023-2025) were analysed. Patient clinical characteristics, imaging characteristics and procedural characteristics were recorded and described.ResultsTwenty-two patients (mean age 76 ± 13) were included in this study. Mean CSDH size was 18.4 ± 6.7 mm; 9/22 (40.9%) patients had a Charlson Comorbidity index >4. MMA embolization was successful in all 22/22 (100%) patients. Symptomatic improvement was observed in 14/17 (82.4%) of patients. During mean follow up (10.7 ± 9.2 (range 1-36)) weeks, mean reduction in CSDH thickness was 12.1 ± 9.9 mm; 13/16 (81.3%) patients had at least partial radiologic resolution of CSDH. No patients required retreatment.ConclusionPHIL-LV MMAE was safe and effective for treatment of CSDH in this case series.

BackgroundMechanical thrombectomy (MT) is the standard of care in acute ischemic stroke, but the efficacy of differently sized stent retrievers (SRs) for MT in M2 segment middle cerebral artery (MCA) occlusions has not been extensively evaluated. We conducted a retrospective study comparing outcomes of 3, 4, and 6 mm SRs in MT at the M2 segment.MethodsData collected from M2 MCA strokes in a comprehensive stroke center were retrospectively analyzed. Patient outcomes were quantified using National Institutes of Health Stroke Scale (NIHSS) 24 h post-procedure and modified Rankin Scale (mRS) scores at discharge and after 90 days. Successful outcomes were defined as an NIHSS score of <10 and an mRS score of <3. First pass success rate and thrombolysis in cerebral infarction (TICI) scores of 2b or greater were also compared. Analysis was performed using chi-square and Kruskal-Wallis tests.ResultsA total of 139 patients with similar baseline demographics underwent MT with SRs sized 3 mm (n = 23), 4 mm (n = 56), and 6 mm (n = 60). No significant difference between groups was noted in 24-h NIHSS (p = 0.83), mRS at discharge (p = 0.597), mRS score at 90 days (p = 0.251), time from groin to recanalization (p = 0.12), or first pass success (p = 0.058). All three sizes had similar complication rates. However, the rate of successful recanalization (final TICI ≥2b) differed significantly between groups (91.3% for 3 mm, 80.4% for 4 mm, and 95.0% for 6 mm; p = 0.043).ConclusionThis retrospective study showed successful recanalization of isolated M2 occlusions was greatest with 6 mm SR.