Citations

Abstract Background and aims Patient foramen ovale (PFO) is a stroke risk factor evaluated using transthoracic echocardiogram (TTE), transesophageal echocardiogram (TEE), transcranial Doppler ultrasound (TCD), and cardiac computed tomography (CCT). TCD is non-invasive, but its diagnostic performance relative to other modalities is unclear. We hypothesized that TCD has high sensitivity but low specificity for PFO detection. Methods We conducted an IRB-approved, single-center retrospective study of patients undergoing TCD for right-to-left shunt (RLS) detection between January 2024 and April 2025. Patients were identified from the Neurosonology Laboratory Quality Database. Records were reviewed for demographics, comorbidities, and TTE, TEE, and CCT findings. Sensitivity, specificity, and positive and negative predictive values were calculated for TCD, TTE, and CCT relative to TEE, the gold standard. The sensitivity and specificity of TCD versus TTE were compared using McNemar test. Results Ninety-two patients were included. Mean age was 52.66 ± 15.78 years, and 51% were female; qualifying events included ischemic stroke or transient ischemic attack. RLS was detected in 53 patients (57.6%) by TCD, 17 (30%) by TTE, 18 (28.5%) by TEE, and 3 (21%) by CCT. TCD demonstrated 94% sensitivity and 53% specificity, (PPV 66%, NPV 89%). TTE showed 50% sensitivity and 82% specificity (PPV: 75%, NPV 61%), while CCT showed 33% sensitivity and 100% specificity (PPV 100%, NPV 33.3%). TCD had significantly higher sensitivity than TTE for RLS detection (p=0.039) with a non-significant difference in specificity (p=0.062), using TEE as the reference. Conclusions TCD has high sensitivity for RLS detection in patients with ischemic stroke or transient ischemic attack. Conflict of interest Rachel Knopp: nothing to disclose; Kelly Flemming: nothing to disclose; Eugene Scharf: nothing to disclose; Zafer Keser: nothing to disclose; Jessica Prink: nothing to disclose; Shari Johnson: nothing to disclose; Guy Reeder: nothing to disclose; Robert Brown: nothing to disclose; William Harmsen: nothing to disclose; Muhib Khan: nothing to disclose Table 1 - belongs to Results

Abstract Background and aims Carotid calcification features are readily quantifiable by CTA, yet their relationship with plaque vulnerability and rupture remains controversial. We investigated whether calcium score, volume, density, and residual (volume-normalized) calcium density are associated with carotid plaque vulnerability, symptomatic status, and recurrent ipsilateral ischemic stroke and transient ischemic attack (TIA). Methods Patients with recent TIA or ischemic stroke and ≥2mm carotid plaque with <70% stenosis underwent baseline CTA and MRI. Carotid calcifications were segmented on CTA (≥130 Hounsfield units), yielding calcium score, volume, and density. Volume-normalized residual calcium density was calculated as the standardized residuals from a log-transformed calcium density-volume model (Figure 1). Vulnerable plaque features (intraplaque hemorrhage, thin-or-ruptured fibrous cap, and lipid-rich necrotic core, and ulceration) were linked to calcium metrics. Associations between calcium metrics and recurrent ipsilateral TIA or ischemic stroke were assessed using Cox regression and adjusted for age, sex, stenosis. Results Among 193 patients (mean age 67.8±8.5 years, 144 (75%) men), 26 experienced ipsilateral ischemic stroke or TIA over median 5-year follow-up. Lower calcium density and residual calcium density were independently associated with vulnerable plaques (adjusted ORs: 0.40-0.60; Figure 2). Residual calcium density remained independently associated with symptomatic side (adjusted OR: 0.64, 95%CI: 0.47-0.88) and with recurrent TIA/stroke (adjusted HR: 0.53, 95%CI: 0.29-0.95). Calcium score, volume, and density were not independently associated with symptomatic side or recurrent events (Figure 3). Conclusions Residual calcium density is a robust, independent marker of carotid plaque vulnerability and ipsilateral ischemic stroke and TIA risk in patients with symptomatic carotid artery disease. Conflict of interest Juul Bierens: nothing to disclose; Daniel Bos: nothing to disclose; Paul J. Nederkoorn: nothing to disclose; Pim A. de Jong: nothing to disclose; Robert J. van Oostenbrugge: nothing to disclose; Luc J.M. Smits: nothing to disclose; Jae W. Song: nothing to disclose; Giuseppe Lanzino: nothing to disclose; M. Eline Kooi: nothing to disclose; Luca Saba: nothing to disclose. Figure 1 - belongs to Methods Figure 2 - belongs to Results Figure 3 - belongs to Conclusions

Abstract Background and aims Intravenous thrombolysis (IVT) is the most common acute ischemic stroke (AIS) treatment. This study compared the efficacy and safety of different IVT agents to determine the optimal option for AIS management. Methods A comprehensive search was conducted in PubMed, Scopus, Web of Science, and Cochrane CENTRAL until 23 March 2025. Randomized controlled trials (RCTs) on alteplase (tPA), tenecteplase (TNK), reteplase, and recombinant human pro-urokinase (rhPro-UK) of any dose in AIS were included. A frequentist network meta-analysis was conducted using R software to pool risk ratios (RRs) with 95% confidence intervals (CIs). Results Twenty-one RCTs (10,852 patients) were included. Compared with tPA 0.9 mg/kg, tPA 0.6 mg/kg showed a lower risk of symptomatic intracranial hemorrhage (RR, 95% CI: 0.48 [0.27–0.86]), whereas TNK 0.4 mg/kg showed a higher risk (RR, 95% CI: 1.88 [1.06–3.35]). Reteplase 18 + 18 mg (RR, 95% CI: 1.13 [1.05–1.21]) and TNK 0.25 mg/kg (RR, 95% CI: 1.05 [1.01–1.10]) were associated with higher rates of excellent functional outcome at 90 days (mRS 0–1) compared with tPA 0.9 mg/kg. No significant differences were observed among interventions in 90-day mortality and favorable function outcomes (mRS 0–2), or serious adverse events. Conclusions Thrombolytic efficacy and safety in AIS appear to be dose dependent. A lower tPA dose (0.6 mg/kg) improved safety; reteplase 18 + 18 mg and TNK 0.25 mg/kg favored functional recovery, whereas a higher TNK dose increased sICH risk, underscoring the importance of dose optimization in IVT selection. Conflict of interest Haneen Sabet: nothing to disclose, Moaz Elsayed Abouelmagd: nothing to disclose, Abdelrahman Shata: nothing to disclose, Mohamed El-Moslemani: nothing to disclose, Diana Kadi: nothing to disclose, Dina Essam Abo-elnour: nothing to disclose, Abdallah Abbas: nothing to disclose, Ramanathan Kadirvel: nothing to disclose, Sherief Ghozy: nothing to disclose Figure 1 - belongs to Results

Abstract Background and aims A substantial proportion of large vessel strokes remain untreated worldwide due to limited access to specialized neuroendovascular care. Remote robotic neuroendovascular thrombectomy performed by expert interventionalists from a distant hub has been proposed as one solution to address these disparities. In this study, we test the feasibility of remote robotic thrombectomy in a live animal model. Methods Two live porcine experiments were conducted, involving Mayo Clinic Arizona (MCA), University Hospital Zurich (USZ), and Inselspital—Bern University Hospital (USB), Switzerland. Aspiration thrombectomy at USB was performed remotely from USZ (regional hub, 95 km; two experiments) and MCA (international hub, one experiment). Five aspiration thrombectomies were completed using a proprietary robotic magnetic navigation system consisting of a mobile magnetic field generator, a three-axis robotic advancer, controlling an 80 cm 6f sheath, a 6f aspiration catheter and a 0.035″ magnetically trackable guidewire, and a teleoperated console. The primary endpoint was the feasibility of successful clot aspiration using the robotic magnetic navigation system. Results Five out of seven thrombectomy procedures were successfully completed, achieving full clot retrieval. In the remaining two procedures, more distal clots were targeted, which could not be reached due to the limited catheter length. Magnetic navigation was successful in all cases. Image quality was excellent to acceptable. Conclusions Remote robotic thrombectomy using robotic magnetic navigation technology is technically feasible over both regional and transcontinental distances. This first-in-vivo regional and transatlantic study establishes a potential model of stroke intervention that may one day bridge access to endovascular stroke treatment. Conflict of interest Patrick Thurner and Tilman Schubert have consulting agreements with Nanoflex Robotics.

Abstract Background and aims Dementia represents a growing global health burden, with vascular dementia and Alzheimer's disease accounting for most cases through cumulative cerebrovascular damage. While periodontal disease (PD) is associated with cognitive decline, dental caries represents a distinct oral infection pathway characterized by bacterial invasion and infection, potentially contributing to vascular brain injury through microbial translocation, endothelial dysfunction, systemic inflammation, and cerebral microvascular damage. The relationship between dental caries and incident dementia within a vascular framework remains understudied. Methods We analyzed 6,542 participants from the Atherosclerosis Risk in Communities (ARIC) Study who received dental examinations at Visit 4 (1996–1998) and followed for incident dementia through 2023 using standardized surveillance (Figure-1). Cox proportional hazards models with competing risk accounted for death. Multivariable models adjusted for demographic factors, cardiovascular risk factors, BMI, physical activity, education, income, insurance status, APOE4 genotype, PD, and regular dental care. Results Mean age was 62.4 years; 54% female, 82% White, and 18% Black. Dental caries was present in 20%. Over median 25-year follow-up, individuals with dental caries demonstrated significantly higher cumulative dementia incidence (Gray’s test p<0.0001). After full adjustment, dental caries remained independently associated with increased dementia risk (adjusted HR 1.20, 95% CI 1.02–1.42), independent of PD and APOE4 status (Figure-2). Conclusions Dental caries was independently associated with increased incident dementia risk. These findings support a potential role for caries-related oral infection in vascular brain injury pathways relevant to vascular dementia and stroke-related cognitive impairment. Oral infection prevention may represent a modifiable target within cerebrovascular disease prevention strategies. Conflict of interest Alana E. Barton, Souvik Sen, Kevin Moss, James Selph, Ryan Demmer, Pamela L. Lutsey, Priya Palta, Kamakshi Lakshminarayan, Karly Pikel, Wayne Rosamond: nothing to disclose Figure 1 - belongs to Conclusions Figure 2 - belongs to Conclusions

Abstract Background and aims Systemic inflammation has been implicated in cognitive decline and dementia pathogenesis. Interleukin-6 (IL-6), a pro-inflammatory cytokine associated with cardiovascular disease, may play a role in cerebrovascular and neurodegenerative processes. We examined the relationship between IL-6 levels and incident dementia risk using the Atherosclerosis Risk In Communities (ARIC) study. Methods ARIC study participants at Visit 4 (1996-1998) had IL-6 measurements and were followed for incident dementia over 25 years. IL-6 levels were dichotomized at the median value of 1.5 pg/mL. Incident dementia cases were identified through standardized surveillance. Cox proportional hazards regression calculated hazard ratios (HR) with 95% confidence intervals (CI), adjusting for age, race, gender, hypertension, diabetes, obesity, physical activity, smoking, alcohol use, education, income, insurance status, and ApoE-ε4 allele status. Results Participants (n=5,662; mean age 62.4±5.6 years, 55% female, 83% White, 17% Black) were followed for a median of 25 years. Those with IL-6 >1.5 pg/mL demonstrated significantly increased dementia risk versus IL-6 ≤1.5 pg/mL. The crude HR was 1.23 (95% CI 1.11-1.37). After adjustment, the association remained significant with HR 1.13 (95% CI 1.01-1.27). Kaplan-Meier survival curves showed progressive divergence between groups, with lower dementia-free survival in the higher IL-6 group, log-rank p<0.001. Conclusions Elevated mid-life IL-6 levels (>1.5 pg/mL) independently predict increased long-term dementia risk over 25 years, even after adjusting for cardiovascular and sociodemographic confounders. These findings support systemic inflammation's role in dementia pathogenesis and highlight IL-6 as a potential biomarker for dementia risk. Conflict of interest Karly Pikel, Natasha Carmichael, Alana E. Barton, Kevin Moss, Christine Downey, Ryan T. Demmer, Pamela L. Lutsey, Souvik Sen: Nothing to disclose Figure 1 - belongs to Conclusions

Abstract Background and aims BotulinumtoxinA (BoNT-A) is a guideline-recommended treatment for spasticity following stroke. We evaluated BoNT-A treatment persistence drivers of discontinuation and predictors of non-persistence in patients with stroke. Methods Eligible adults had received ≥1 treatment cycle with BoNT-A at the Mayo Clinic (January 2010–June 2023) and had an International Classification of Diseases (ICD9/10) code for stroke. Patients with ICD9/10 codes for migraine, cerebral palsy, cervical dystonia, or bladder dysfunction were excluded. Index date (baseline) was the date of first BoNT-A treatment; follow-up was 2 years post-index. Patients were persistent if all treatment intervals post-index were ≤the 80th percentile of all patients. Reasons for discontinuation were identified by AI natural language processing analysis of physicians’ free-text notes. Baseline characteristics associated with non-persistence were identified using step-wise logistic regression analyses. Results Overall, 407 patients were included. The 80th percentile of the BoNT-A treatment interval was 154 days; 5.7% of patients were persistent. Mean (95%CI) number of BoNT-A treatments over the 2-year follow-up was 2.43 (2.24–2.63) cycles. Comorbidity burden was high (Table 1). The most common reasons for BoNT-A discontinuation were patient- and physician-reported lack of treatment effectiveness (Table 2). Older age, Hispanic/Latino/unknown/undisclosed ethnicity, presence of comorbidities, and use of medications of interest at baseline (skeletal muscle relaxants, antidepressants, anticonvulsants, opiates, sedatives/hypnotics, NSAIDs) were significantly associated with non-persistence (Table 3). Conclusions Over 2 years, BoNT-A treatment persistence was low. Lack of treatment effectiveness was the most common reason for discontinuation. Addressing drivers of non-persistence may help improve BoNT-A treatment outcomes. Conflict of interest Funding: Ipsen. Disclosure of Conflicts of Interest: Michael Wainberg: has received consulting fees from Ipsen. Billie Schultz: has received consulting fees from Ipsen. Jinming Zhang, Simon Page, Jonathan Bouchard, Amandeep Mann, Mathieu Beneteau are employees of Ipsen. Table 1 - belongs to Conclusions Table 2 - belongs to Conclusions Table 3 - belongs to Conclusions

Abstract Background and aims Cardiac CT is increasingly used in the acute diagnostic work-up of acute ischemic stroke (AIS), but commonly used scan protocols do not allow assessment of left ventricular ejection fraction (LVEF), a known predictor of functional outcome. We aimed to develop a prediction model for reduced LVEF using clinical and imaging data. Methods This post-hoc analysis of the Mind the Heart study included AIS patients who underwent acute cardiac CT (acquired during diastole) and routine diagnostic work-up, including transthoracic echocardiography (TTE), which served as the reference standard for LVEF. Multivariable logistic regression models incorporating baseline variables, arterial input function (AIF) derived from CT-perfusion, and left atrial and ventricular cardiac CT volumes were developed using the training set (80%) to predict reduced LVEF (<50%). Performance was evaluated on the testing set (20%) using receiver operating characteristic analysis, reporting the area under the curve (AUC), sensitivity, and specificity. Non-linear associations were modeled using restricted cubic splines, and Least Absolute Shrinkage and Selection Operator regression was used for variable selection. We evaluated whether cardiac CT information improved model fit. Results We included 298 AIS patients (61% male, median age 71 [IQR 61-79] years), of whom 87 (29%) had reduced LVEF. The final model showed moderate discrimination (AUC 0.71), sensitivity (0.67) and specificity (0.70). Including cardiac CT volumes in addition to baseline and AIF variables improved model fit (p=0.04). Conclusions Our model may aid risk stratification for reduced LVEF in AIS patients undergoing acute cardiac CT and could support identification of patients in whom additional TTE is warranted. Conflict of interest MMW received the Arthur Fonville Award for Stroke Research in 2025 for this work. SSN received travel grants from the Cultuurfondsbeurs, Remmert Adriaan Laan Fonds, CONTRAST consortium and Dr. Jan Meerwaldt Stichting outside the submitted work, JMC reports grants from Medtronic, Siemens, AstraZeneca and Bayer outside the submitted work (all paid to institution). JMC is shareholder and co-founder of TrianecT. CBM reports grants from Stryker and Boeringher-Ingelheim outside the submitted work (paid to institution) and is shareholder of Nicolab (minority interest). LAR was supported by a personal Dekker Junior Clinical Scientist Grant from the Dutch Heart Foundation.The other authors have no financial conflicts of interest.

Abstract Aims To compare the prevalence of autism spectrum disorder (ASD) in children with and without epilepsy using clinical diagnoses and chart review with research‐defined criteria and to examine sex ratios, age at ASD onset, and intellectual disability between groups. Method The Olmsted County birth cohort included 30 490 individuals, of whom 257 (0.84%) had epilepsy diagnosed before 19 years of age. The cohort was previously screened for ASD using three case definitions: research‐inclusive (ASD‐R I ), research‐narrow (ASD‐R N ), and clinically diagnosed ASD (ASD‐C). For those with ASD‐R I , we assessed sex, intellectual disability (IQ < 70), and age at ASD onset. Means and proportions are reported when available. Results ASD prevalence was significantly higher in participants with epilepsy across all definitions: ASD‐R I 21.4% versus 3.2%; ASD‐R N 14.0% versus 1.6%; and ASD‐C 7.9% versus 0.7%. Children with both epilepsy and ASD‐R I were more often female (38.2% vs 25.8%), had higher rates of intellectual disability (56.5% vs 15.4%), and met the ASD‐R I criteria at a younger age (mean 7 years 5 months vs 8 years 8 months). Interpretation Co‐occurring ASD and epilepsy are frequently associated with intellectual disability, earlier age at ASD diagnosis, and reduced male predominance.

Abstract Background and aims Interfacility transfers in rural areas are challenging due to limited ambulance and bed availability, image sharing issues, and access disparities. Regional transfers to strategically located primary stroke centers can reduce time to care, decrease helicopter use, support local economies, and conserve comprehensive stroke center (CSC) resources. This study established a primary stroke center as a regional transfer center (rSTC) within a telestroke network. Methods Over one year, stakeholders created regionally appropriate transfer workflows defined as patients without large vessel occlusion, low likelihood for neurosurgical decompression or critical illness within a rSTC and 4 surrounding centers. Continuous education and prospective case logs were maintained by the rSTC stroke coordinator and a retrospective chart review was conducted one year following the intervention period. Results Of 99 patients requiring transfer, 73 (73.7%) went to the rSTC, with an average stay of 2.1 days, no deaths, and no secondary CSC transfers. The remaining 26 (26.3%) were sent to the CSC, with longer stays (10.6 days) and higher mortality (23.1%). Transfers to the rSTC were faster (60 min vs. 188 min). Financial analysis showed significant EMS savings ($455,111) and local health system benefits ($730,000) when compared to the pre-intervention period. Conclusions Establishing an effective regional transfer model is possible with broad stakeholder engagement and regular network maintenance. rSTCs provide the right level of care for the right patient, improving efficient allocation of healthcare resources and avoiding futile transfers with costly admissions far from home while opening beds and reducing transfer times to CSCs. Conflict of interest Nothing to disclose