BackgroundArtificial intelligence can help to identify irregular shapes and sizes, crucial for managing unruptured intracranial aneurysms (UIAs). However, existing artificial intelligence tools lack reliable classification of UIA shape irregularity and validation against gold-standard three-dimensional rotational angiography (3DRA). This study aimed to develop and validate a deep-learning model using computed tomography angiography (CTA) for classifying irregular shapes and measuring UIA size.MethodsCTA and 3DRA of UIA patients from a referral hospital were included as a derivation set, with images from multiple medical centers as an external test set. Senior investigators manually measured irregular shape and aneurysm size on 3DRA as the ground truth. Convolutional neural network (CNN) models were employed to develop the CTA-based model for irregular shape classification and size measurement. Model performance for UIA size and irregular shape classification was evaluated by intraclass correlation coefficient (ICC) and area under the curve (AUC), respectively. Junior clinicians’ performance in irregular shape classification was compared before and after using the model.ResultsThe derivation set included CTA images from 307 patients with 365 UIAs. The test set included 305 patients with 350 UIAs. The AUC for irregular shape classification of this model in the test set was 0.87, and the ICC of aneurysm size measurement was 0.92, compared with 3DRA. With the model’s help, junior clinicians’ performance for irregular shape classification was significantly improved (AUC 0.86 before vs 0.97 after, P<0.001).ConclusionThis study provided a deep-learning model based on CTA for irregular shape classification and size measurement of UIAs with high accuracy and external validity. The model can be used to improve reader performance.

Background:Hypertension is a common comorbidity in patients with unruptured intracranial aneurysms, which is closely related to the instability of aneurysms. Though anti-hypertension therapy has been recommended by several guidelines, the optimal blood pressure range remains unknown. This trial aims to determine the optimal range of blood pressure by comparing standard blood pressure reduction and intensive blood pressure reduction to reduce the instability of unruptured intracranial aneurysms. Methods/design:This trial is a multicenter, prospective, open-label, randomized controlled trial with minimization to ensure allocation concealment. Five hundred seventy patients with unruptured intracranial aneurysms and hypertension will be recruited by nine centers in China. Patients will be allocated to the standard blood pressure lowering (SBPL) group (systolic blood pressure at 120-140 mmHg) or the enhanced blood pressure lowering (EBPL) group (systolic blood pressure at <120 mmHg). The primary outcome is aneurysm instability within 24 months after appropriate blood pressure lowering therapy, including aneurysm growth and aneurysm rupture. During follow-up, Blood pressure data will be collected monthly, while radiological examination for aneurysms will be performed at 6 ± 1 months, 12 ± 1 months and 24 ± 1 months. The trial has 85% power to reduce 60% of aneurysm instability. Discussion:The China Antihypertensive Trial for Intracranial Aneurysm (ChATIA) trial is a randomized trial about blood pressure lowering therapy for UIA patients. In this study, a new strategy and evidence will be provided for blood pressure management in the future. Clinical trial registration:http://www.clinicaltrials.gov, identifier NCT05941377.

Endovascular treatment (EVT) for basilar artery trunk aneurysm (BTA) is inadequately studied due to its rarity. The aim of this study was to report our experience regarding treatment modalities, complications and outcomes. A retrospective analysis of 2759 aneurysm patients who underwent EVT between January 2018 and December 2022 was performed. Patients with BTAs were involved, and their clinical characteristics, treatment modalities, complications, and clinical and angiographic outcomes were collected. Literatures from 2013 to 2024 were reviewed and studies included more than 5 BTA cases were summarized. Thirty-seven patients were involved, including 6 (16.2

Purpose To develop a deep learning model for the morphologic measurement of unruptured intracranial aneurysms (UIAs) based on CT angiography (CTA) data and validate its performance using a multicenter dataset. Materials and Methods In this retrospective study, patients with CTA examinations, including those with and without UIAs, in a tertiary referral hospital from February 2018 to February 2021 were included as the training dataset. Patients with UIAs who underwent CTA at multiple centers between April 2021 and December 2022 were included as the multicenter external testing set. An integrated deep learning (IDL) model was developed for UIA detection, segmentation, and morphologic measurement using an nnU-Net algorithm. Model performance was evaluated using the Dice similarity coefficient (DSC) and intraclass correlation coefficient (ICC), with measurements by senior radiologists serving as the reference standard. The ability of the IDL model to improve performance of junior radiologists in measuring morphologic UIA features was assessed. Results The study included 1182 patients with UIAs and 578 controls without UIAs as the training dataset (median age, 55 years [IQR, 47-62 years], 1012 [57.5%] female) and 535 patients with UIAs as the multicenter external testing set (median age, 57 years [IQR, 50-63 years], 353 [66.0%] female). The IDL model achieved 97% accuracy in detecting UIAs and achieved a DSC of 0.90 (95% CI: 0.88, 0.92) for UIA segmentation. Model-based morphologic measurements showed good agreement with reference standard measurements (all ICCs > 0.85). Within the multicenter external testing set, the IDL model also showed agreement with reference standard measurements (all ICCs > 0.80). Junior radiologists assisted by the IDL model showed significantly improved performance in measuring UIA size (ICC improved from 0.88 [95% CI: 0.80, 0.92] to 0.96 [95% CI: 0.92, 0.97], P < .001). Conclusion The developed integrated deep learning model using CTA data showed good performance in UIA detection, segmentation, and morphologic measurement and may be used to assist less experienced radiologists in morphologic analysis of UIAs. Keywords: Segmentation, CT Angiography, Head/Neck, Aneurysms, Comparative Studies Supplemental material is available for this article. © RSNA, 2024 See also the commentary by Wang in this issue.

Importance:Recent longitudinal studies in patients with unruptured intracranial aneurysms (UIAs) suggested that aneurysm wall enhancement (AWE) on magnetic resonance imaging (MRI) predicts growth and rupture. However, because these studies were limited by small sample size and short follow-up duration, it remains unclear whether this radiological biomarker has predictive value for UIA instability. Objective:To determine the 4-year risk of instability of UIAs with AWE and investigate whether AWE is an independent predictor of UIA instability. Design, Setting, and Participants:Individual patient data were obtained from 3 prospective multicenter cohort studies conducted in 83 Chinese centers between January 2017 and December 2024. Included were patients aged 18 to 75 years with at least 1 asymptomatic, saccular UIA greater than or equal to 3 mm. Exposures:All patients had 3-T MRI gadolinium-enhanced aneurysm wall imaging and computed tomography angiography (CTA) at baseline, and CTA at follow-up. Main Outcomes and Measures:The primary outcome measure was aneurysm growth or rupture (instability) during follow-up. The absolute risk of aneurysm instability in UIAs with circumferential, focal, and no AWE was determined with Kaplan-Meier estimates at 4 years after baseline aneurysm wall imaging. Cox proportional hazards regression was used to investigate AWE as a potential predictor of instability. Results:Of the 1453 patients who had baseline 3-T MRI aneurysm wall imaging, 41 patients were excluded because of loss to follow-up or no follow-up CTA, and 61 patients were excluded because of low-quality CTA. We included 1351 patients (median [IQR] age, 56 [48-63] years; 750 female [56%]) with 1416 UIAs and 4884 aneurysm-years of follow-up. Instability within 4 years occurred in 235 of 1416 UIAs (16.6%). The absolute cumulative risk of instability at 4 years was 36.8% (95% CI, 30.7%-43.0%) in UIAs with circumferential AWE, 17.2% (95% CI, 13.4%-21.1%) in UIAs with focal AWE, and 11.4% (95% CI, 11.9%-16.1%) in UIAs with no AWE. Circumferential AWE predicted 4-year instability (hazard ratio [HR], 3.80; 95% CI, 2.82-5.14) and after adjusting for size ratio, aneurysm location, aneurysm shape, and bifurcation configuration (adjusted HR, 2.21; 95% CI, 1.56-3.13). Conclusions and Relevance:Within 4 years after baseline wall imaging, instability occurred in one-third of UIAs with circumferential AWE. These results suggest that MRI aneurysm wall imaging may be used for predicting the risk of aneurysm instability.

Spontaneous intracranial artery dissection (sIAD) is the leading cause of stroke in young individuals. Identifying high-risk sIAD cases that exhibit symptoms and are likely to progress is crucial for treatment decision-making. This study aimed to develop a model relying on circulating biomarkers to discriminate symptomatic sIADs. The study prospectively collected sIAD tissues and corresponding serums from January 2020 to December 2022 as the discovery cohort. Symptomatic sIADs were defined as those with mass effect, hemorrhagic, or ischemic stroke. A stratification model was developed using the machine-learning algorithm within the derivation cohort (a cross-sectional cohort including from January 2018 to August 2022) and validated within the validation cohort (a longitudinal cohort including from January 2017 to April 2023). In the discovery cohort (n = 10, 5 symptomatic), analyses of tissues and serums revealed 15 proteins and 2 cytokines with significance between symptomatic and asymptomatic sIADs. Among these biomarkers, six proteins and one cytokine, participating in the immune response and inflammatory-related pathways, have a good consistency in expression level between sIAD tissues and serums. In the derivation cohort (n = 181, 77 symptomatic), a model incorporating these 7 biomarkers was highly discriminative of symptomatic sIADs (area under curve [AUC], 0.95). This model performed well in predicting the occurrence of sIAD-related symptoms in the validation cohort (n = 84, 26 symptomatic) with an AUC of 0.88. This study revealed seven circulating biomarkers of symptomatic sIAD and provided a high-accuracy model relying on these circulating biomarkers to identify symptomatic sIADs, which may aid in clinical decision-making for sIADs.

BACKGROUND:Post-translational modifications (PTMs) serve as crucial regulatory mechanisms in stroke-induced brain injury. Nevertheless, the specific roles of PTMs in subarachnoid hemorrhage (SAH) development and the functional properties of associated genes are not yet fully understood. METHODS:We performed differential expression analysis on SAH dataset and intersected results with PTM-related genes to identify PTM-associated genes of SAH differential expression (DE-PTMRGs), followed by functional enrichment analysis. Mendelian randomization (MR) was applied to determine causal PTM genes affecting SAH risk. Machine learning algorithms were used to screen feature genes from these causal genes. Candidate biomarkers derived from feature gene analysis were validated using an independent transcriptomic dataset. Subsequently, these biomarkers were employed for constructing clinical diagnostic models, performing single-gene enrichment and regulatory network analyses, conducting clinical experimental validation, and predicting potential drug repurposing opportunities. RESULTS:We identified 227 SAH-associated PTM genes, of which 12 demonstrated causal relationships through Mendelian randomization analysis. Machine learning algorithms screened ZRANB1 and ATG7 as key PTM biomarkers. The diagnostic model achieved excellent performance with AUC = 0.833 (0.701-0.966), calibration curve mean error of 0.088, and capability to predict 20%-90% SAH risk. Network analysis revealed distinct regulatory patterns: ZRANB1 showed extensive connectivity with 27 microRNAs and 6 transcription factors, while ATG7 exhibited focused regulation by 16 microRNAs and 1 transcription factor. Functionally, ZRANB1 primarily regulates metabolism and genetic maintenance, while ATG7 coordinates broader cellular processes including autophagy and signal transduction. Through clinical validation, we found that the levels of ATG7 and ZRANB1 were significantly altered between SAH patients and healthy control. Both biomarkers predicted Tretinoin as a potential therapeutic compound. Molecular docking showed strong binding affinities: ZRANB1-Tretinoin (-7.7 kcal/mol) and ATG7-Tretinoin (-6.1 kcal/mol). Molecular dynamics simulations confirmed stable dynamic binding between ATG7 and Tretinoin, while ZRANB1-Tretinoin complex showed instability. CONCLUSION:The work established a novel ATG7-ZRANB1 biomarker combination with superior diagnostic accuracy for SAH prediction and identified Tretinoin as a promising therapeutic candidate, demonstrating PTM-based precision medicine potential.

Background Aspirin is widely used for preventing ischaemic events. About 20%–40% of patients have aspirin resistance (ASR), which prevents them from benefiting from aspirin medication. This study aimed to develop and validate a model based on single-nucleotide polymorphism (SNP) to distinguish ASR patients. Methods We included patients with spontaneous intracerebral haemorrhage and continuing antiplatelet therapy from a multicentre, prospective cohort study as the derivation cohort. Thromboelastography (inhibition of arachidonic acid channel<50%) was used to identify ASR. Genotyping was performed to identify the ASR-related SNP. Based on the result of the logistic analysis, the aspirin resistance in the Chinese population score (ASR-CN score) was established, and its accuracy was evaluated using the area under the curve (AUC). Patients receiving dual antiplatelet therapy for unruptured intracranial aneurysm embolism were prospectively included in the validation cohort. After embolism, 30-day ischaemic events, including ischaemic stroke, new or more frequent transient ischaemic attack, stent thrombosis and cerebrovascular death, were recorded. Results The derivation cohort included 212 patients (155 male patients and the median age as 59). 87 (41.0%) individuals were identified with ASR. The multivariate logistic analysis demonstrated six SNPs of GP1BA, TBXA2R, PTGS2 and NOS3 as risk factors related to ASR. The ASR-CN score integrating these SNPs performed well to discriminate ASR patients from non-ASR patients (AUC as 0.77). Based on the validation cohort of 372 patients receiving antiplatelet therapy after embolism (including 130 ASR patients), the ASR-CN score continued to distinguish ASR patients with good accuracy (AUC as 0.80). Patients with high a ASR-CN score were more likely to suffer from 30-day ischaemic events after embolism (OR, 1.28; 95% CI, 1.10 to 1.50; p=0.002). Conclusion GP1BA, TBXA2R, PTGS2 and NOS3 were SNPs related to ASR. The ASR-CN score is an effective tool to discriminate ASR patients, which may guide antiplatelet therapy. Clinical trial registration Surgical Treatments of Antiplatelet Intracerebral Hemorrhage cohort (unique identifier: ChiCTR1900024406, http://www.chictr.org.cn/edit.aspx?pid=40640&htm=4 ).

Background Precise diagnosis and rapid treatment for acute complex intracranial hemorrhage (ICH) are crucial. The neurosurgical hybrid operating platform integrates traditional open neurosurgery operating room functionalities with endovascular therapy capabilities and is developing in the neurosurgical practice. However, its effect on the emergent complicated neurovascular cases needs pilot exploration.Methods In this prospective study, a total of 103 cases of both spontaneous and non-spontaneous ICH were consecutively recruited between June 2019 and June 2023. Demographic data, including age, gender distribution, and types of hemorrhage, were collected. Surgical interventions were tailored based on DSA, including spontaneous and non-spontaneous hemorrhages. Functional outcomes were assessed using the modified Rankin Scale (mRS) preoperatively and postoperatively.Results Over the study period from June 2019 to June 2023, a cohort of 103 ICH cases underwent emergency hybrid surgical treatment utilizing the E-HOPE platform. Among these cases, 88 were classified as spontaneous ICH, while 15 were non-spontaneous. The mean age at diagnosis for the entire cohort was 54.0 ± 3.7 years, with a slight predominance of male patients. Spontaneous ICH encompassed a diverse spectrum of etiologies, including arteriovenous malformations, aneurysms, arteriovenous fistulas, cavernous malformations, moyamoya disease, and cryptogenic hemorrhages. Surgical interventions were tailored to address the specific pathology. Notably, improvements in mRS scores were observed in a majority of cases, with some patients experiencing stabilization or deterioration postoperatively. Non-spontaneous cases (n = 15) were primarily iatrogenic (n = 13) due to tumors adjacent to the internal carotid artery, necessitating stent graft deployment. Surgical approaches, including stent graft deployment and middle meningeal artery embolization, were effective in managing these cases. Postoperative functional outcomes varied depending on the nature of the hemorrhage, with a subset of patients demonstrating improvement in mRS scores while others showed no significant change.Conclusions Emergency hybrid surgical treatment utilizing the E-HOPE platform offers promising outcomes for ICH patients. Tailored surgical approaches result in favorable postoperative functional outcomes, highlighting the importance of a multidisciplinary approach in managing these complex cases.

Importance:Previous randomized clinical trials did not demonstrate the superiority of endovascular stenting over aggressive medical management for patients with symptomatic intracranial atherosclerotic stenosis (sICAS). However, balloon angioplasty has not been investigated in a randomized clinical trial. Objective:To determine whether balloon angioplasty plus aggressive medical management is superior to aggressive medical management alone for patients with sICAS. Design, Setting, and Participants:A randomized, open-label, blinded end point clinical trial at 31 centers across China. Eligible patients aged 35 to 80 years with sICAS defined as recent transient ischemic attack (<90 days) or ischemic stroke (14-90 days) before enrollment attributed to a 70% to 99% atherosclerotic stenosis of a major intracranial artery receiving treatment with at least 1 antithrombotic drug and/or standard risk factor management were recruited between November 8, 2018, and April 2, 2022 (final follow-up: April 3, 2023). Interventions:Submaximal balloon angioplasty plus aggressive medical management (n = 249) or aggressive medical management alone (n = 252). Aggressive medical management included dual antiplatelet therapy for the first 90 days and risk factor control. Main Outcomes and Measures:The primary outcome was a composite of any stroke or death within 30 days after enrollment or after balloon angioplasty of the qualifying lesion or any ischemic stroke in the qualifying artery territory or revascularization of the qualifying artery after 30 days through 12 months after enrollment. Results:Among 512 randomized patients, 501 were confirmed eligible (mean age, 58.0 years; 158 [31.5%] women) and completed the trial. The incidence of the primary outcome was lower in the balloon angioplasty group than the medical management group (4.4% vs 13.5%; hazard ratio, 0.32 [95% CI, 0.16-0.63]; P < .001). The respective rates of any stroke or all-cause death within 30 days were 3.2% and 1.6%. Beyond 30 days through 1 year after enrollment, the rates of any ischemic stroke in the qualifying artery territory were 0.4% and 7.5%, respectively, and revascularization of the qualifying artery occurred in 1.2% and 8.3%, respectively. The rate of symptomatic intracranial hemorrhage in the balloon angioplasty and medical management groups was 1.2% and 0.4%, respectively. In the balloon angioplasty group, procedural complications occurred in 17.4% of patients and arterial dissection occurred in 14.5% of patients. Conclusions and Relevance:In patients with sICAS, balloon angioplasty plus aggressive medical management, compared with aggressive medical management alone, statistically significantly lowered the risk of a composite outcome of any stroke or death within 30 days or an ischemic stroke or revascularization of the qualifying artery after 30 days through 12 months. The findings suggest that balloon angioplasty plus aggressive medical management may be an effective treatment for sICAS, although the risk of stroke or death within 30 days of balloon angioplasty should be considered in clinical practice. Trial Registration:ClinicalTrials.gov Identifier: NCT03703635.

Chinese population have a high prevalence of unruptured intracranial aneurysm (UIA). Clinical and imaging risk factors predicting UIA growth or rupture are poorly understood in the Chinese population due to the lack of large-scale longitudinal studies, and the treatment decision for UIA patients was challenging. Develop a decision tree (DT) model for UIA instability, and validate its performance in multi-center studies. Single-UIA patients from two prospective, longitudinal multicenter cohort studies were analyzed, and set as the development cohort and validation cohort. The primary endpoint was UIA instability (rupture, growth, or morphological change). A DT was established within the development cohort and validated within the validation cohort. The performance of clinicians in identifying unstable UIAs before and after the help of the DT was compared using the area under curve (AUC). The development cohort included 1270 patients with 1270 UIAs and a follow-up duration of 47.2 ± 15.5 months. Aneurysm instability occurred in 187 (14.7

ImportancePrevious randomized clinical trials did not demonstrate the superiority of endovascular stenting over aggressive medical management for patients with symptomatic intracranial atherosclerotic stenosis (sICAS). However, balloon angioplasty has not been investigated in a randomized clinical trial.ObjectiveTo determine whether balloon angioplasty plus aggressive medical management is superior to aggressive medical management alone for patients with sICAS.Design, Setting, and ParticipantsA randomized, open-label, blinded end point clinical trial at 31 centers across China. Eligible patients aged 35 to 80 years with sICAS defined as recent transient ischemic attack (<90 days) or ischemic stroke (14-90 days) before enrollment attributed to a 70% to 99% atherosclerotic stenosis of a major intracranial artery receiving treatment with at least 1 antithrombotic drug and/or standard risk factor management were recruited between November 8, 2018, and April 2, 2022 (final follow-up: April 3, 2023).InterventionsSubmaximal balloon angioplasty plus aggressive medical management (n = 249) or aggressive medical management alone (n = 252). Aggressive medical management included dual antiplatelet therapy for the first 90 days and risk factor control.Main Outcomes and MeasuresThe primary outcome was a composite of any stroke or death within 30 days after enrollment or after balloon angioplasty of the qualifying lesion or any ischemic stroke in the qualifying artery territory or revascularization of the qualifying artery after 30 days through 12 months after enrollment.ResultsAmong 512 randomized patients, 501 were confirmed eligible (mean age, 58.0 years; 158 [31.5%] women) and completed the trial. The incidence of the primary outcome was lower in the balloon angioplasty group than the medical management group (4.4% vs 13.5%; hazard ratio, 0.32 [95% CI, 0.16-0.63]; P < .001). The respective rates of any stroke or all-cause death within 30 days were 3.2% and 1.6%. Beyond 30 days through 1 year after enrollment, the rates of any ischemic stroke in the qualifying artery territory were 0.4% and 7.5%, respectively, and revascularization of the qualifying artery occurred in 1.2% and 8.3%, respectively. The rate of symptomatic intracranial hemorrhage in the balloon angioplasty and medical management groups was 1.2% and 0.4%, respectively. In the balloon angioplasty group, procedural complications occurred in 17.4% of patients and arterial dissection occurred in 14.5% of patients.Conclusions and RelevanceIn patients with sICAS, balloon angioplasty plus aggressive medical management, compared with aggressive medical management alone, statistically significantly lowered the risk of a composite outcome of any stroke or death within 30 days or an ischemic stroke or revascularization of the qualifying artery after 30 days through 12 months. The findings suggest that balloon angioplasty plus aggressive medical management may be an effective treatment for sICAS, although the risk of stroke or death within 30 days of balloon angioplasty should be considered in clinical practice.Trial RegistrationClinicalTrials.gov Identifier: NCT03703635

Background and objectivesRegarding the anti-inflammatory effect, methylprednisolone is a candidate to prevent patients with unruptured intracranial aneurysms (UIAs) from postoperative bleeding (PB) after flow diverter (FD) treatment. This study aimed to investigate whether methylprednisolone is related to a lower incidence of PB after FD treatment for UIAs.MethodsThis study retrospectively reviewed UIA patients receiving FD treatment between October 2015 and July 2021. All patients were observed until 72 h after FD treatment. The patients receiving methylprednisolone (80 mg, bid, for at least 24 h) were considered as standard methylprednisolone treatment (SMT) users, otherwise as non-SMT users. The primary endpoint indicated the occurrence of PB, including subarachnoid hemorrhage, intracerebral hemorrhage, and ventricular bleeding, within 72 h after FD treatment. This study compared the incidence of PB between SMT users and non-SMT users and investigated the protective effect of SMT on PB after FD treatment using the Cox regression model. Finally, after controlling the potential factors related to PB, we performed subgroup analysis to further confirm the protective effect of SMT on PB.ResultsThis study finally included 262 UIA patients receiving FD treatment. PB occurred in 11 patients (4.2%), and 116 patients (44.3%) received SMT postoperatively. The median time from the end of surgery to PB was 12.3 h (range: 0.5–48.0 h). SMT users had a lower incidence of PB comparing with non-SMT users (1/116, 0.9% vs. 10/146, 6.8%, respectively; p = 0.017). The multivariate Cox analysis demonstrated that SMT users (HR, 0.12 [95%CI, 0.02–0.94], p = 0.044) had a lower risk of PB postoperatively. After controlling the potential factors related to PB (i.e., gender, irregular shape, surgical methods [FD and FD + coil] and UIA sizes), the patients receiving SMT still had a lower cumulative incidence of PB, comparing with patients receiving non-SMT (all p < 0.05).ConclusionSMT was correlated with the lower incidence of PB for patients receiving FD treatment and may be a potential method to prevent PB after the FD treatment.

Background To compare the safety and efficacy of pipeline embolization device (PED) and Tubridge flow diverter (TFD) for unruptured posterior circulation aneurysms. Methods Posterior aneurysm patients treated with PED or TFD between January, 2019, and December, 2021, were retrospectively reviewed. Patients’ demographics, aneurysm characteristics, treatment details, complications, and follow-up information were collected. The procedural-related complications and angiographic and clinical outcome were compared. Results A total of 107 patients were involved; PED was applied for 55 patients and TFD for 52 patients. A total of 9 (8.4%) procedural-related complications occurred, including 4 (7.3%) in PED group and 5 (9.6%) in TFD group. During a mean of 10.3-month angiographic follow-up for 81 patients, complete occlusion was achieved in 35 (85.4%) patients in PED group and 30 (75.0%) in TFD group. The occlusion rate of PED group is slightly higher than that of TFD group. A mean of 25.0-month clinical follow-up for 107 patients showed that favorable clinical outcome was achieved in 53 (96.4%) patients in PED group and 50 (96.2%) patients in TFD group, respectively. No statistical difference was found in terms of procedural-related complications ( p = 0.737), occlusion rate ( p = 0.241), and favorable clinical outcome (0.954) between groups. Conclusions The current study found no difference in complication, occlusion, and clinical outcome between PED and TFD for unruptured PCAs.

Abstract Background Coil embolization is a common method for treating unruptured intracranial aneurysms (UIAs). To effectively perform coil embolization for UIAs, clinicians must undergo extensive training with the assistance of senior physicians over an extended period. This study aimed to establish a deep-learning system for measuring the morphological features of UIAs and help the surgical planning of coil embolization for UIAs. Methods Preoperative computational tomography angiography (CTA) data and surgical data from UIA patients receiving coil embolization in our medical institution were retrospectively reviewed. A convolutional neural network (CNN) model was trained on the preoperative CTA data, and the morphological features of UIAs were measured automatically using this CNN model. The intraclass correlation coefficient (ICC) was utilized to examine the similarity between the morphologies measured by the CNN model and those determined by experienced clinicians. A deep neural network model to determine the diameter of first coil was further established based on the CNN model within the derivation set (75% of all patients) using neural factorization machines (NFM) model and was validated using a validation set (25% of all patients). The general match ratio (the difference was within ± 1 mm) between the predicted diameter of first coil by model and that used in practical scenario was calculated. Results One-hundred fifty-three UIA patients were enrolled in this study. The CNN model could diagnose UIAs with an accuracy of 0.97. The performance of this CNN model in measuring the morphological features of UIAs (i.e., size, height, neck diameter, dome diameter, and volume) was comparable to the accuracy of senior clinicians (all ICC > 0.85). The diameter of first coil predicted by the model established based on CNN model and the diameter of first coil used actually exhibited a high general match ratio (0.90) within the derivation set. Moreover, the model performed well in recommending the diameter of first coil within the validation set (general match ratio as 0.91). Conclusion This study presents a deep-learning system which can help to improve surgical planning of coil embolization for UIAs.

BACKGROUND The role of endovascular therapy for acute stroke with a large infarction has not been extensively studied in differing populations. METHODS We conducted a multicenter, prospective, open-label, randomized trial in China involving patients with acute large-vessel occlusion in the anterior circulation and an Alberta Stroke Program Early Computed Tomography Score of 3 to 5 (range, 0 to 10, with lower values indicating larger infarction) or an infarct-core volume of 70 to 100 ml. Patients were randomly assigned in a 1:1 ratio within 24 hours from the time they were last known to be well to undergo endovascular therapy and receive medical management or to receive medical management alone. The primary outcome was the score on the modified Rankin scale at 90 days (scores range from 0 to 6, with higher scores indicating greater disability), and the primary objective was to determine whether a shift in the distribution of the scores on the modified Rankin scale at 90 days had occurred between the two groups. Secondary outcomes included scores of 0 to 2 and 0 to 3 on the modified Rankin scale. The primary safety outcome was symptomatic intracranial hemorrhage within 48 hours after randomization. RESULTS A total of 456 patients were enrolled; 231 were assigned to the endovascular-therapy group and 225 to the medical-management group. Approximately 28% of the patients in both groups received intravenous thrombolysis. The trial was stopped early owing to the efficacy of endovascular therapy after the second interim analysis. At 90 days, a shift in the distribution of scores on the modified Rankin scale toward better outcomes was observed in favor of endovascular therapy over medical management alone (generalized odds ratio, 1.37; 95% confidence interval, 1.11 to 1.69; P = 0.004). Symptomatic intracranial hemorrhage occurred in 14 of 230 patients (6.1%) in the endovascular-therapy group and in 6 of 225 patients (2.7%) in the medical-management group; any intracranial hemorrhage occurred in 113 (49.1%) and 39 (17.3%), respectively. Results for the secondary outcomes generally supported those of the primary analysis. CONCLUSIONS In a trial conducted in China, patients with large cerebral infarctions had better outcomes with endovascular therapy administered within 24 hours than with medical management alone but had more intracranial hemorrhages. (Funded by Covidien Healthcare International Trading [Shanghai] and others; ANGEL-ASPECT ClinicalTrials.gov number, NCT04551664.).

Intracranial aneurysm is the leading cause of nontraumatic subarachnoid hemorrhage. Evaluating the unstable (rupture and growth) risk of aneurysms is helpful to guild decision-making for unruptured intracranial aneurysms (UIA). This study aimed to develop a model for risk stratification of UIA instability. The UIA patients from two prospective, longitudinal multicenter Chinese cohorts recruited from January 2017 to January 2022 were set as the derivation cohort and validation cohort. The primary endpoint was UIA instability, comprising aneurysm rupture, growth, or morphology change, during a 2-year follow-up. Intracranial aneurysm samples and corresponding serums from 20 patients were also collected. Metabolomics and cytokine profiling analysis were performed on the derivation cohort (758 single-UIA patients harboring 676 stable UIAs and 82 unstable UIAs). Oleic acid (OA), arachidonic acid (AA), interleukin 1β (IL-1β), and tumor necrosis factor-α (TNF-α) were significantly dysregulated between stable and unstable UIAs. OA and AA exhibited the same dysregulated trends in serums and aneurysm tissues. The feature selection process demonstrated size ratio, irregular shape, OA, AA, IL-1β, and TNF-α as features of UIA instability. A machine-learning stratification model (instability classifier) was constructed based on radiological features and biomarkers, with high accuracy to evaluate UIA instability risk (area under curve (AUC), 0.94). Within the validation cohort (492 single-UIA patients harboring 414 stable UIAs and 78 unstable UIAs), the instability classifier performed well to evaluate the risk of UIA instability (AUC, 0.89). Supplementation of OA and pharmacological inhibition of IL-1β and TNF-α could prevent intracranial aneurysms from rupturing in rat models. This study revealed the markers of UIA instability and provided a risk stratification model, which may guide treatment decision-making for UIAs.

OBJECTIVE: To propose a machine learning (ML) model predicting the favorable outcome of stereotactic radiosurgery (SRS) for residual brain arteriovenous malformation (hAVM) after partial embolization. METHODS: One hundred and thirty bAVM patients who underwent partial embolization followed by SRS were reviewed retrospectively. Patients were split at random split into training datasets (n = 100) and testing datasets (n = 30). Radiomics and dosimetric features were extracted from pre-SRS treatment images. Feature selection was performed to select appropriate radiomics and dosimetric features. Three ML algorithms were applied to construct models using selected features respectively. A total of 9 models were trained to predict favorable outcomes (obliteration without complication) of bAVMs. The efficacy of these models was evaluated on the testing dataset using mean accuracy (ACC) and area under the receiver operating characteristic curve (AUC). RESULTS: The obliteration rate of this cohort was 70.77% (92 of 130) with a mean follow-up of 43.8 months (range, 12-108 months). Favorable outcomes were achieved in 89 patients (68.46%). Four radiomics features and 7 dosimetric features were selected for ML model construction. The dosimetric support vector machines (SVM) model showed the best performance on the training dataset, with an ACC of 0.74 and AUC of 0.78. The dosimetric SVM model also showed the best performance on the testing dataset, with an ACC of 0.83 and AUC of 0.77. CONCLUSIONS: Dosimetric features are good predictors of prognosis for patients with partially embolized bAVM followed by SRS therapy. The use of ML models is an innovative method for predicting favorable outcomes of partially embolized bAVM followed by SRS therapy.

Background Rebleeding can cause a catastrophic outcome after aneurysmal subarachnoid hemorrhage. A clinical + morphology nomogram was promoted in our previous study to assist in discriminating the rupture intracranial aneurysms (RIAs) with a high risk of rebleeding. The aim of this study was to validate the predictive accuracy of this nomogram model. Method The patients with RIAs in two medical centers from December 2020 to September 2021 were retrospectively reviewed, whose clinical and morphological parameters were collected. The Cox regression model was employed to identify the risk factors related to rebleeding after their admission. The predicting accuracy of clinical + morphological nomogram, ELAPSS score and PHASES score was compared based on the area under the curves (AUCs). Results One hundred thirty-eight patients with RIAs were finally included in this study, 20 of whom suffering from rebleeding after admission. Hypertension (hazard ratio (HR), 2.54; a confidence interval of 95% (CI), 1.01–6.40; P = 0.047), bifurcation (HR, 3.88; 95% CI, 1.29–11.66; P = 0.016), and AR (HR, 2.68; 95% CI, 1.63–4.41; P < 0.001) were demonstrated through Cox regression analysis as the independent risk factors for rebleeding after admission. The clinical + morphological nomogram had the highest predicting accuracy (AUC, 0.939, P < 0.01), followed by the bifurcation (AUC, 0.735, P = 0.001), AR (AUC, 0.666, P = 0.018), and ELAPSS score (AUC, 0.682, P = 0.009). Hypertension (AUC, 0.693, P = 0.080) or PHASES score (AUC, 0.577, P = 0.244) could not be used to predict the risk of rebleeding after admission. The calibration curve for the probability of rebleeding showed a good agreement between the prediction through clinical + morphological nomogram and actual observation. Conclusion Hypertension, bifurcation site, and AR were independent risk factors related to the rebleeding of RIAs after admission. The clinical + morphological nomogram could help doctors to identify the high-risk RIAs with a high predictive accuracy.

Background This study aimed to investigate clinical and angiographic outcomes of Pipeline embolization device (PED) treatment of large or giant basilar artery (BA) aneurysms and examine associated factors. Methods Clinical and angiographic data of 29 patients (18 men, 11 women) with large or giant BA aneurysms were retrospectively examined. Mean age was 44.1 ± 21.2 years (range, 30–68). Mean aneurysm size was 22.2 ± 8.3 mm (range, 12.0–40.1). Results Mean angiographic follow-up was 18.3 ± 3.4 months (range, 4.5–60). The rate of adequate aneurysmal occlusion (O'Kelly–Marotta grade C–D) was 87%. The overall complication rate was 44.8%; most complications (84.6%) occurred in the periprocedural period. Univariable comparison of patients who did and did not develop complications showed significant differences in aneurysm size (p < 0.01), intra-aneurysmal thrombus (p = 0.03), and mean number of PEDs used (p = 0.02). Aneurysm size (odds ratio, 1.4; p = 0.04) was an independent risk factor for periprocedural complications in multivariable analysis. Mean clinical follow-up was 23.5 ± 3.2 months (range, 0.1–65). Nine patients (31%) had a poor clinical outcome (modified Rankin scale score ≥3) at last follow-up, including 7 patients who died. Univariable comparisons between patients with favorable and unfavorable clinical outcomes showed that aneurysm size (p = 0.009) and intra-aneurysmal thrombus (p = 0.04) significantly differed between the groups. Multivariable analysis showed that aneurysm size (odds ratio, 1.1; p = 0.04) was an independent risk factor for poor clinical outcome. Conclusion PED treatment of large or giant BA aneurysms is effective and can achieve a satisfactory long-term occlusion rate. However, the treatment complications are not negligible. Aneurysm size is the strongest predictor of perioperative complications and poor clinical outcome.