Citations

Background: CT perfusion (CTP) imaging is an important tool for identifying candidates for mechanical thrombectomy (MT). This modality is increasingly available in regional facilities, but whether pre-transfer perfusion imaging studies may be relied upon for treatment decisions is unclear. We analyzed a cohort of stroke transfers to quantify changes in CTP imaging profiles over time. Methods: A cohort of confirmed ischemic stroke patients who received CTP before and after transfer to a large mid-western comprehensive stroke center (CSC) were identified. We compared perfusion mismatch (MM), core infarct volume (CIV), and favorability of imaging for mechanical thrombectomy (MT) candidacy (defined as the presence of LVO and MM/CIV ratio of >1.8) between studies. Regression models were used to examine predictors of CIV growth including hypoperfusion intensity ratio (HIR) and time between studies. Results: Over a 24-month period, 61 patients met inclusion criteria. Median age was 71 (IQR 60-80), 52% were female, median NIHSS was 14 (IQR 4-19), and most patients had occlusion of the internal carotid and/or middle cerebral arteries (77%). The median time from last known well (LKW) to initial CT was 148 minutes (IQR 71-620) and median time between imaging studies was 149 minutes (IQR 126-194). Median CIV growth was 0 mL (IQR 0-6) and 8% (5/61) had growth ≥25 mL. In linear regression, higher HIR predicted CIV growth (β=24.3, p=0.04), but time from LKW to presentation and time between imaging studies did not. Imaging studies taken at the regional and CSC sites demonstrated excellent agreement in classifying patients as candidates for MT (kappa 0.778 [95% CI: 0.611-0.944]). A favorable regional CTP profile was 93% sensitive for MT delivery at the CSC with a positive predictive value of 63%. Conclusion: In our cohort, CIV growth between facilities was minimal despite relatively long interfacility transport times and was predicted by the regional HIR. A favorable regional imaging profile demonstrated high agreement with the imaging result at the CSC and was strongly associated with MT treatment.

Introduction: The safety and efficacy of intravenous thrombolysis (IVT) in cervical artery dissection-related acute ischemic stroke (CeAD-AIS) remains unclear. We performed a retrospective study of data from the National Inpatient Sample investigating outcomes in CeAD-AIS patients treated with and without IVT. Methods: We included adult patients with CeAD-AIS hospitalized between 2016 to 2019. CeAD-AIS was defined by concurrent CeAD and AIS diagnoses, identified with ICD-10 codes. The primary outcome was home discharge. The safety outcomes were inpatient death and intracerebral hemorrhage (ICH). We used survey-weighted multivariable adjusted logistic regressions comparing the groups (IVT versus no IVT), followed by interaction analyses to examine for effect modifications based on age, medical history, stroke severity, and endovascular treatment. Results: Between 2016-2019, 1,360 (12.1%) of 11,285 CeAD-AIS patients received IVT. Patients treated with IVT- had higher NIH Stroke Scale (NIHSS) scores (median 8 [4-17] versus 3 [1-11]; p < 0.001). In adjusted analyses, treatment with IVT was associated with a greater odds of home discharge (adjusted odds ratio [aOR] 1.49; 95% confidence interval [CI] 1.09-2.05, p = 0.014), but not ICH (aOR 0.90, 95% CI 0.43-1.86, p = 0.773) or inpatient death (aOR 1.22, 95% CI 0.72-2.06, p = 0.454). In interaction analyses, stroke severity significantly interacted with IVT ( p for interaction = 0.001), where IVT was significantly associated with increased likelihood of home discharge in patients with moderate to severe strokes (NIHSS > 4) (aOR 2.28, 95% CI 1.59-3.26, p < 0.001) but not in those with mild stroke (NIHSS 0-4) (aOR 1.03 95% CI 0.62-1.70, p = 0.922). Conclusion: IVT improved the likelihood of home discharge in patients with CeAD-AIS without increasing the risk of inpatient death or ICH.

Background: Emergency medical services (EMS) routing of stroke patients is challenging in rural settings. To address this, the American Stroke Association (ASA) has developed a template severity-based EMS triage algorithm to direct EMS to either the nearest stroke-ready versus the nearest Comprehensive Stroke Center (CSC) hospital, but its real-world impact is uncertain. Methods: This is an interim analysis of a before-and-after quality improvement study examining the impact of implementation of a severity-based EMS stroke destination protocol by three medical control authorities (MCAs) covering 4 rural counties in western Michigan. We defined optimal destination according to the ASA algorithm, which recommends bypass for patients within 24 of last known well if large vessel occlusion suspected and bypass will not delay thrombolysis. Chi square tests were used to compare optimal destination selection before and after implementation among EMS suspected stroke cases originating from the target counties. We also compared the time from EMS scene arrival to ischemic stroke (IS) treatment (thrombolysis or endovascular therapy [EVT]) before and after implementation using Wilcoxon Rank Sum tests. Results: From November 2021 to May 2024, EMS transported 616 suspected stroke cases, 272 (44.2%) of which ultimately received a diagnosis of ischemic stroke (IS) or transient ischemic attack. The protocol was implemented by each MCA in random order at months 21, 23, and 29, resulting in 73 MCA-months before and 17 after implementation, excluding a one-month washout for each MCA. Optimal hospital destinations were selected for 310/499 (62.1%) patients prior to the protocol and 86/117 (73.5%) after (p=0.021). Thrombolytics were administered to 57 patients (48 before and 9 after) and 46 (42 before and 4 after) received EVT. Median times from first EMS contact to needle decreased from 98 (interquartile range [IQR]: 79-107) to 77 (IQR: 75-85) minutes (p=0.012); median times from EMS contact to groin puncture increased nonsignificantly from 165 (IQR: 132-216) to 196 (IQR: 191-228) minutes (p=0.110). Conclusions: Interim analysis of implementation of a rural EMS severity-based triage algorithm for suspected stroke cases resulted in improved destination selection and faster thrombolysis but did not improve EVT treatment times. This analysis will be repeated once follow-up is completed in 7 months.

Background: Left ventricular (LV) injury is a cardioembolic etiology of ischemic stroke. We aimed to 1) evaluate recurrent stroke risk among patients with cryptogenic stroke and LV injury, and 2) determine if there is a treatment effect with anticoagulation in this population. Methods: We analyzed the multicenter, retrospective Cardiac Abnormalities in Stroke Prevention and Recurrence cohort of patients with cryptogenic ischemic stroke. The exposure was LV injury defined by 1) LV ejection fraction (EF) 20-40% and/or 2) LV wall motion abnormality (hypokinesis or akinesis). The primary composite outcome included recurrent stroke, major bleeding, or death. Cox proportional hazard models were built and sequentially adjusted for 1) age and sex (Model 1), and 2) age, sex, race, ethnicity, anti-thrombotic use, and vascular risk factors (Model 2). We then assessed if there was heterogeneity in treatment effect by LV injury status with an interaction term of LV injury and oral anticoagulation use in Model 2. Results: Of 2,641 patients in CASPR, the analytic cohort included 2,191 patients with complete exposure and outcome data (median age 65 years, 49.8% female, median follow-up 1.55 years). There were 113 (5.2%) patients with LV injury (median LVEF 30%, interquartile range 25-34%). There were 559 incident events (321 recurrent strokes [288 ischemic], 71 major hemorrhages, 250 deaths). Among patients with LV injury, 15.9% were prescribed an anticoagulant within 7 days of index stroke (vs 9.6% without LV injury, p=0.03). The unadjusted risk of the primary outcome was non-significantly higher among patients with LV injury vs without (HR 1.40; 95% confidence interval [CI], 0.98-2.00), but became significant in Model 1 (HR 1.50; 95% CI, 1.21-1.85) and in Model 2 (HR 1.35, 95% CI, 1.11-1.64). In the LV injury subgroup, anticoagulation vs antiplatelet therapy was associated with a significant reduction in the primary outcome (HR 0.63, 95% CI, 0.43-0.94). In patients without LV injury, anticoagulation was associated with a higher risk of the composite outcome (HR 4.35, 95% CI, 1.69-11.16, p[interaction]=0.002). Conclusion: In this observational study, there was an increased risk of the primary outcome which included recurrent stroke among patients with cryptogenic stroke and LV injury. Anticoagulation use in LV injury is associated with a significant reduction in the risk of the primary outcome. These findings warrant confirmation in a dedicated randomized controlled trial.

Introduction: Malignancy represents an important subgroup of embolic stroke of undetermined source (ESUS). We aim to compare infarct features, potential embolic sources, and long-term outcomes in ESUS patients with and without malignancy. Methods: In this pre-specified secondary analysis of the Cardiac Abnormalities in Stroke Prevention and Recurrence (CASPR) retrospective cohort study, active cancers were identified by site investigators and adjudicated as malignant if they were non-squamous or non-basal cell skin cancers. We compared the prevalence of various potential embolic sources (e.g., patent foramen ovale, left ventricular dysfunction, post-stroke atrial fibrillation) among patients with and without malignancy. The risk of recurrent stroke, major bleeding, or death was evaluated using adjusted Cox proportional hazards models, accounting for age, sex, vascular risk factors, competing potential embolic sources, and initial treatment with anticoagulation versus non-anticoagulant (e.g., antiplatelet) regimen. Results: Of the 2201 patients followed over a median of 564 days, 236 (10.7%) were diagnosed with cancer at the time of stroke. Patients with malignancy were older, more frequently hypertensive and did not differ with regards to prevalence of specific infarct patterns or potential cardiac embolic sources. No difference was observed in the composite outcome of recurrent stroke, major bleeding, or death (adjusted HR 1.02, 95% CI, 0.71-1.46). Age (adjusted HR per year 1.02, 95% CI, 1.01-1.03), baseline stroke severity (adjusted HR 1.03 per NIHSS point, 95% CI, 1.01-1.05), prior stroke (adjusted HR 1.68, 95% CI, 1.38-2.05), and multi-territorial infarct pattern (adjusted HR 1.67, 95% CI, 1.33-2.08) were significantly associated with this composite outcome, but anticoagulation was not. There was a nominally lower rate of major bleeding with malignancy (adjusted HR 0.45, 95% CI, 0.19-1.04), a significantly higher rate of death (adjusted HR 1.58, 95% CI, 1.09-2.30), but no significant difference in recurrent ischemic stroke (p=0.31). Conclusions: Patients with malignancy and ESUS had higher mortality but did not differ in recurrent stroke risk or major bleeding when compared to patients without malignancy. Anticoagulation was not independently associated with better outcomes. Further analysis of infarct topology and risk factors may better identify patients for whom anticoagulation may benefit.

Background: Tricuspid regurgitation (TR) is a common occurrence in patients with heart failure (HF), and its role in disease progression has gained attention in recent years. While TR can worsen the clinical outcomes in HF patients, its impact on gender, racial and socioeconomic factors remains largely unexplored. With growing evidence supporting the role of percutaneous interventions for the treatment of significant TR, understanding these disparities is more crucial than ever. Methods: Data was extracted from the National Inpatient and National Readmission 2016 - 2020 Databases. We used ICD-10 code I50 to identify the patients primarily admitted for HF and subdivided the cohort into two groups based on presence or absence of TR. We performed multivariable Logistic regression analysis to determine odds of in-hospital mortality and multivariable Cox regression analysis to assess 30- and 90-day hospital readmission in HF patients with and without TR. All the analyses were adjusted for age, gender, insurance status, Charlson co-morbidity index (CCI) and hospital characteristics. STATA 16 software was used for analysis. Results: There was no difference in in-hospital mortality among HF patients with and without TR (OR: 1.04, 95% CI 0.94 – 1.16, p = 0.442) except in certain subgroups of HF patients. HF patients with TR were 6% more likely to have HF-specific readmission in 30-days (HR: 1.06, 95% CI 1.00 – 1.13, p = 0.044) and 9% more likely to have HF-specific readmission in 90-days (HR: 1.09, 95% CI 1.03 – 1.15, p = 0.002). Subgroup analysis revealed significant gender, racial and socioeconomic disparities in in-hospital mortality and readmission outcomes of HF patients with TR compared to those without TR. Conclusion: In our population-based survey analysis, we observed significant gender, racial and socioeconomic disparities in clinical outcomes of HF patients with TR compared to those without TR.

Apart from ancestry, personal or environmental covariates may contribute to differences in polygenic score (PGS) performance. We analyzed the effects of covariate stratification and interaction on body mass index (BMI) PGS (PGS BMI ) across four cohorts of European (N = 491,111) and African (N = 21,612) ancestry. Stratifying on binary covariates and quintiles for continuous covariates, 18/62 covariates had significant and replicable R 2 differences among strata. Covariates with the largest differences included age, sex, blood lipids, physical activity, and alcohol consumption, with R 2 being nearly double between best- and worst-performing quintiles for certain covariates. Twenty-eight covariates had significant PGS BMI –covariate interaction effects, modifying PGS BMI effects by nearly 20% per standard deviation change. We observed overlap between covariates that had significant R 2 differences among strata and interaction effects – across all covariates, their main effects on BMI were correlated with their maximum R 2 differences and interaction effects (0.56 and 0.58, respectively), suggesting high-PGS BMI individuals have highest R 2 and increase in PGS effect. Using quantile regression, we show the effect of PGS BMI increases as BMI itself increases, and that these differences in effects are directly related to differences in R 2 when stratifying by different covariates. Given significant and replicable evidence for context-specific PGS BMI performance and effects, we investigated ways to increase model performance taking into account nonlinear effects. Machine learning models (neural networks) increased relative model R 2 (mean 23%) across datasets. Finally, creating PGS BMI directly from GxAge genome-wide association studies effects increased relative R 2 by 7.8%. These results demonstrate that certain covariates, especially those most associated with BMI, significantly affect both PGS BMI performance and effects across diverse cohorts and ancestries, and we provide avenues to improve model performance that consider these effects.