Search

Background and Purpose: The spectrum of brain infarction in patients with embolic stroke of undetermined source (ESUS) has not been well characterized. Our objective was to define the frequency and pattern of brain infarcts detected by magnetic resonance imaging (MRI) among patients with recent ESUS participating in a clinical trial. Methods: In the NAVIGATE ESUS trial (New Approach Rivaroxaban Inhibition of Factor Xa in a Global Trial Versus ASA to Prevent Embolism in Embolic Stroke of Undetermined Source), an MRI substudy was carried out at 87 sites in 15 countries. Participants underwent an MRI using a specified protocol near randomization. Images were interpreted centrally by those unaware of clinical characteristics. Results: Among the 918 substudy cohort participants, the mean age was 67 years and 60% were men with a median (interquartile range) of 64 (26–115) days between the qualifying ischemic stroke and MRI. On MRI, 855 (93%) had recent or chronic brain infarcts that were multiple in 646 (70%) and involved multiple arterial territories in 62% (401/646). Multiple brain infarcts were present in 68% (510/755) of those without a history of stroke or transient ischemic attack before the qualifying ESUS. Prior stroke/transient ischemic attack ( P 0 ( P P =0.01) were associated with multiple infarcts. Topographically, large and/or cortical infarcts were present in 89% (757/855) of patients with infarcts, while in 11% (98/855) infarcts were exclusively small and subcortical. Among those with multiple large and/or cortical infarcts, 57% (251/437) had one or more involving a different vascular territory from the qualifying ESUS. Conclusions: Most patients with ESUS, including those without prior clinical stroke or transient ischemic attack, had multiple large and/or cortical brain infarcts detected by MRI, reflecting a substantial burden of clinical stroke and covert brain infarction. Infarcts most frequently involved multiple vascular territories. Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT02313909.

Background: We sought to determine and validate a pre-reperfusion blood pressure threshold for MCA occlusions patients with better thrombectomy outcomes. Methods: We reviewed patients from 2013 to 2015 treated with thrombectomy for a MCA M1 occlusion at our center. Blood pressure measurements from available records at the time of EMS activation to the time of reperfusion were used to calculate an averaged pre-reperfusion mean arterial pressure (MAP). All patients in our analysis had successful reperfusion. Our endpoint was a mRS ≤2 at 90 days. The averaged MAP was grouped as follows: ≥85mmHg, ≥90mmHg, ≥95mmHg, ≥100mmHg, ≥110 mmHg and ≥115mmHg. We considered hourly epochs from symptom onset to reperfusion, pre-treatment ASPECTS, thrombolytics, and collateral status. Significant parameters from a univariate analysis were included into a multivariate logistic regression to determine independent predictors of outcome. The multivariate findings were then validated with a prospective set of similar patients evaluated with our mobile stroke treatment unit. Results: We reviewed 52 patients. Mean age was 70; NIHSS was 16±6; 40% (21) received t-PA; median ASPECTS score was 10; TICI 3 score was achieved in 63% (33), and 67% (35) had a good outcome. The number of patients having a mean arterial blood pressure threshold were as follows: ≥85mmHg (96%), ≥90mmHg (83%), ≥95mmHg (73%), ≥100mmHg (58%), ≥110mmHg (31%) and ≥115mmHg (15%). Reperfusion occurred as follows: ≤3hrs (4%), ≤4hrs (21%), ≤5hrs (39%), ≤6hrs (52%), ≤7hrs (67%), ≤8hrs (79%), ≤9hrs (87%), and ≤12hrs (90%). Our multivariate logistic regression model identified an average pre-reperfusion MAP ≥95mmHg as the sole factor statistically associated with a better outcome post successful thrombectomy [OR 15.1, CI 1.3-170.1,p=0.02]. Good clinical outcome in those patients with an averaged MAP ≥95mmHg was 76% and 43% in those below the threshold [OR 5.2, p=0.02]. Our validation cohort consisted of 32 patients, and a pre-reperfusion averaged MAP ≥ 95 mmHg had a sensitivity of of 93% for a good outcome. Conclusion: Our analysis suggests with validation that a pre-reperfusion averaged MAP ≥ 95mmHg results in better outcomes with successful thrombectomy and may be used as a therapeutic target in the pre-hospital setting.

Introduction: In the VNS-REHAB trial, vagus nerve stimulation (VNS) paired with task-specific arm and hand rehabilitation (Paired VNS) led to clinically meaningful improvements in both impairment and function of the upper extremity in people with chronic ischemic stroke. In this post hoc analysis of trial data, we assessed whether improvements were driven by proximal (shoulder and elbow) and/or distal (wrist and hand) components of the Upper Extremity Fugl-Meyer Assessment (FMA-UE). We also wanted to determine whether FMA-UE improvements correlated with improvements in functional outcome. Methods: Chronic stroke participants (n=108) with moderate-severe UE impairment (FMA-UE 35.1±8) were implanted with the VNS device and randomized to Paired VNS (n=53) or Controls (n=55). Participants underwent 6 weeks of in-clinic therapy followed by 90 days of a home exercise program. Outcomes, including the FMA-UE and Wolf Motor Function Test (WMFT), were collected after in-clinic therapy (Post-1) and 90 days of home therapy (Post-90). Results: Distal FMA-UE change was significantly greater after Paired VNS compared to Controls, both at Post-1 (2.43±2.52 vs. 0.64±2.33; 95% CI 0.87-2.72, p Conclusions: The data suggest that Paired VNS leads to greater improvement in the distal upper extremity compared to rehabilitation alone and that improvements in FMA-UE scores are correlated with improvements in function. These findings suggest that the improvements seen with Paired VNS are of functional importance. A greater understanding of region-specific arm and hand recovery after Paired VNS therapy will facilitate planning and implementation of a personalized neurorehabilitation approach based on patient-specific impairments.

Introduction: For patients diagnosed with central retinal artery occlusion (CRAO), timely screening for underlying risk factors to prevent secondary stroke and other cardiovascular events is required. Given the profound impact on vision, there is also great interest in the potential to treat CRAO. Thrombolytic therapy is the current standard of care in acute ischemic cerebral stroke, but the window for administration is short. As part of a larger pilot to investigate the prospect of intravenous thrombolysis for patients presenting with acute CRAO in our institution, we set out to evaluate the specific presentation patterns of this population within our system. Methods: Retrospective study performed on 804 patients correctly diagnosed with CRAO from 2011-2020. From this cohort of patients, we identified 484 for whom we could accurately determine the time of first symptoms, the time of first contact with the healthcare system, and the time of evaluation by an ophthalmologist. Results: While 18.0% (87/484) of patients waited over 24 hours from time of vision loss to contact the health care system, 50.2% of patients did connect in under 4.5 hours (243/484). However, only 16.7% (81/484) of patients saw an ophthalmologist within that window, and 35.5% (172/484) did not have any ophthalmologic evaluation within 24 hours of vision loss. Conclusion: Half of CRAO patients within our institution connect with the health care system within a potential window for intravenous thrombolytic treatment, but the majority do not receive ophthalmic diagnosis within that time frame. In conclusion, public health education efforts to speed up presentation times and system-based strategies to decrease the time to ophthalmic diagnosis are needed if thrombolytic treatment either via compassionate use or within a clinical trial setting is to be more widely offered.

Background and Purpose: The Tigertriever is a novel, radially adjustable, fully visible, stentriever that permits the operator to align radial expansion with target vessel diameters. This multicenter trial compared the Tigertriever’s effectiveness and safety compared with established stent retrievers. Methods: Single arm, prospective, multicenter trial comparing the Tigertriever to efficacy and safety performance goals derived from outcomes in 6 recent pivotal studies evaluating the Solitaire and Trevo stent-retriever devices with a lead-in and a main-study phase. Patients were enrolled if they had acute ischemic stroke with National Institutes of Health Stroke Scale score ≥8 due to large vessel occlusion within 8 hours of onset. The primary efficacy end point was successful reperfusion, defined as core laboratory-adjudicated modified Thrombolysis in Cerebral Ischemia score 2b-3 within 3 passes of the Tigertriever. The primary safety end point was a composite of 90-day all-cause mortality and symptomatic intracranial hemorrhage. Secondary efficacy end points included 3-month good clinical outcome (modified Rankin Scale score 0–2) and first-pass successful reperfusion. Results: Between May 2018 and March 2020, 160 patients (43 lead-in, 117 main phase) at 17 centers were enrolled and treated with the Tigertriever. The primary efficacy end point was achieved in 84.6% in the main-study phase group compared with the 63.4% performance goal and the 73.4% historical rate (noninferiority P P P =0.004; superiority P =0.57). Good clinical outcome was achieved in 58% at 90 days. Conclusions: The Tigertriever device was shown to be highly effective and safe compared with Trevo and Solitaire devices to remove thrombus in patients with large-vessel occlusive stroke eligible for mechanical thrombectomy. Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT03474549.

Stroke is the second leading cause of death and disability worldwide, with a disproportionate burden on low- and middle-income countries. Critical elements of guideline-based stroke care developed in high-income countries are not applicable to resource-limited settings, where lack of access to neuroimaging prevents clinicians from distinguishing between ischemic stroke and intracranial hemorrhage, requiring challenging clinical decision-making, particularly in the acute setting. We discuss strategies for acute inpatient management of stroke of unknown type with a focus on blood pressure management and antiplatelet therapy when neuroimaging is unavailable, and review some of the challenges and strategies for successfully implementing stroke unit care in resource-limited health care settings.

Introduction: Compared to White patients, non-White minority populations are less likely to receive acute treatment for ischemic stroke, and when they do, it can be delayed. Hypothesis: To evaluate the interaction between arrival method (transfer vs. emergency medical services [EMS] vs. private vehicle) and race/ethnicity (non-Hispanic Black [NHB] vs. non-Hispanic White [NHW] vs. Hispanic) on door-to-arterial puncture (DTAP) time in thrombectomy. Methods: Multicenter retrospective cohort study of consecutive patients (≥18 years) treated with endovascular thrombectomy at 14 US Comprehensive Stroke Centers (1/1/2019-07/31/2020). Door (arrival) to arterial puncture (DTAP) was evaluated using a linear mixed model to estimate the effect of race/ethnicity and arrival mode on treatment time. Results: Of the 1,908 included patients, 356 (19%) of whom were Hispanic, stroke severity was similar between race/ethnicity groups, however Hispanic and NHB patients experienced significant delays in DTAP compared to NHWs (adjusted β Hisp = 0.32, 95%CI 0.08-0.55; β NHB = 0.23, 95%CI 0.04-0.41). In the final linear mixed model, more severe neurologic deficits were associated with shorter DTAP (β NIHSS per point -0.03, 95%CI -0.04- -0.03), and arrival via EMS or private vehicle versus transfer were associated with longer delays (β EMS = 0.57, 95%CI 0.41-0.74; β PV = 1.27, 95%CI 0.85-1.70). There was a significant interaction between Hispanic ethnicity and arrival by EMS (p=0.03) or private vehicle (p=0.04) in a direction favoring shorter treatment delays. Conclusions: Endovascular therapy is delayed in Hispanic and NHBs when compared to NHWs. While arrival by EMS or private vehicle (vs. inter-hospital transfer) also contributed to treatment delays, Hispanic ethnicity modified this effect such that there was less delay in DTAP. This may be related to communication barriers between caregivers and healthcare providers when a patient is suddenly disabled from stroke.

Background: Stroke is the 5th leading cause of death in the US. National stroke mortality among Asian subgroups has never been reported. Our objective is to disaggregate national stroke mortality data among the largest Asian American groups by sex and characterize mortality trends across a 12-year period. Methods: We extracted National Vital Statistics System mortality data from 2006 to 2017 for 6 Asian subgroups and non-Hispanic Whites (NHWs). Stroke mortality was classified by ICD-10, including ischemic stroke (I630-699), intracerebral hemorrhage (I610-629), and subarachnoid hemorrhage (I601-609). 2006-2017 American Community Survey population data was extracted to calculate age-standardized mortality rates (AMRs) stratified by sex and race. Results: Participants included 2,593 Asian Indians, 7,585 Filipinos, 8,212 Chinese, 2,551 Koreans, 3,179 Vietnamese, 4,667 Japanese, and 781,966 NHWs. Mortality from hemorrhagic strokes exceeded ischemic stroke in groups except NHW (Figure 1A). From 2006-2017, deaths from ischemic stroke increased for all groups, with higher mortality in Filipino males compared to females (figure 1B). Deaths from subarachnoid hemorrhage increased only in the Vietnamese subgroup (figure 1D). A similar trend was observed for intracerebral hemorrhage in Vietnamese males (figure 1C). Conclusion: Disaggregation of US stroke mortality revealed important variations within the 6 largest Asian subgroups, including increased ischemic stroke mortality in all subgroups and increased hemorrhagic stroke mortality among Vietnamese. Considerations for this variation include differences in risk factor prevalence (cigarette smoking, obstructive sleep apnea) and social determinants of health. Future studies must collect disaggregated Asian subgroup data to provide tailored interventions to further reduce stroke mortality. Figures: See below

Background: The effect of anesthesia choice on endovascular thrombectomy (EVT) outcomes is unclear. Collateral status on perfusion imaging may help identify the optimal anesthesia choice. Methods: In a pooled patient level analysis of EXTEND-IA, EXTEND-IA TNK, EXTEND-IA TNK part II and SELECT, EVT Functional outcomes (mRS distribution) were compared between general anesthesia (GA) vs non-general anesthesia (non-GA). Further, we assessed the impact of collateral flow on perfusion imaging evaluated by hypoperfusion intensity ratio (HIR) - Tmax10 sec/Tmax6 sec) on the association between anesthesia type and EVT outcomes. Results: Of 731 treated with EVT, 305 (42%) received GA and 426 (58%) non-GA. The baseline characteristics were similar, except for presentation NIHSS (median [IQR] GA 18 [13-22], non-GA 16[11-20], p Conclusion: GA was associated with worse functional outcomes after EVT, particularly in patients with poor collaterals. These findings have implications for randomized trials of GA vs non-GA.

Background: Depression and anxiety impact recovery and quality of life in 1 in 3 stroke survivors. Caregivers also experience burnout. Creative art-based therapy boosts feelings of accomplishment, self-esteem, and neuroplasticity by stimulating diverse brain regions. Few outpatient art therapy programs exist for stroke survivors and caregivers, limited further during the COVID-19 pandemic. We aim to implement an outpatient, student-run virtual art therapy curriculum using a patient-carer team approach to foster a supportive peer community, reduce depressive symptoms in survivors, and increase relief for caregivers. Methods: A multidisciplinary team of stroke physician, nurse, occupational therapist, clinic managers, and undergraduate student volunteers created an evidenced-based art therapy curriculum feasible for stroke survivors. An art educator trained volunteers in empathetic communication, teaching techniques, and patient privacy. Participants pre-registered for weekly 1 hour classes held via video conference and received art materials by mail. We administered surveys at initial registration and quarterly. An institutional grant provided funding. Results: From September 2020 - July 2021, Healing Strokes hosted 30 classes using the curriculum (Table 1) for 71 survivors and caregivers from 9 states at a 1:5 volunteer-to-participant ratio. 20 participants (14 survivors, 6 caregivers) self-reported benefits of peer support, creative inspiration, increased sense of accomplishment, and improvement in speech and fine motor skills. Conclusion: Implementation of a virtual outpatient art-based therapy program for stroke survivors and caregivers is feasible with participants self-reporting benefits in functional recovery and mood. Post-stroke supportive care programs can expand to include art therapy during a pandemic. Future studies can validate the impact on peer support, function, and post-stroke depression and anxiety.

Background and Purpose— It remains unclear how experience influences outcomes after the advent of stent retriever technology. We studied the relationship between site experience and outcomes in the Trevo Acute Ischemic Stroke multicenter registry. Methods— The 24 sites that enrolled patients in the Trevo Acute Ischemic Stroke registry were trichotomized into low-volume (4 cases/month). Baseline features, imaging, and clinical outcomes were compared across the 3 volume strata. A multivariable analysis was performed to assess whether outcomes were influenced by site volumes. Results— A total of 624 patients were included and distributed as low- (n=188 patients, 30.1%), medium- (n=175, 28.1%), and high-volume (n=261, 41.8%) centers. There were no significant differences in terms of age (mean, 66±16 versus 67±14 versus 65±15; P =0.2), baseline National Institutes of Health Stroke Scale (mean, 17.6±6.5 versus 16.8±6.5 versus 17.6±6.9; P =0.43), or occlusion site across the 3 groups. Median (interquartile range) times from stroke onset to groin puncture were 266 (181.8–442.5), 239 (175–389), and 336.5 (221.3–466.5) minutes in low-, medium-, and high-volume centers, respectively ( P =0.004). Higher efficiency and better outcomes were seen in higher volume sites as demonstrated by shorter procedural times (median, 97 versus 67 versus 69 minutes; P P ≤0.0001), and higher rates of good outcome (90-day modified Rankin Scale [mRS], ≤2; 39% versus 50% versus 53.4%; P =0.02). There were no appreciable differences in symptomatic intracranial hemorrhage or 90-day mortality. After adjustments in the multivariable analysis, there were significantly higher chances of achieving a good outcome in high- versus low-volume (odds ratio, 1.67; 95% CI, 1.03–2.7; P =0.04) and medium- versus low-volume (odds ratio, 1.75; 95% CI, 1.1–2.9; P =0.03) centers, but there were no significant differences between high- and medium-volume centers ( P =0.86). Conclusions— Stroke center volumes significantly influence efficiency and outcomes in mechanical thrombectomy.

Background and Purpose— Mechanical thrombectomy has been shown to improve clinical outcomes in patients with acute ischemic stroke. However, the impact of balloon guide catheter (BGC) use is not well established. Methods— STRATIS (Systematic Evaluation of Patients Treated With Neurothrombectomy Devices for Acute Ischemic Stroke) was a prospective, multicenter study of patients with large vessel occlusion treated with the Solitaire stent retriever as first-line therapy. In this study, an independent core laboratory, blinded to the clinical outcomes, reviewed all procedures and angiographic data to classify procedural technique, target clot location, recanalization after each pass, and determine the number of stent retriever passes. The primary clinical end point was functional independence (modified Rankin Scale, 0–2) at 3 months as determined on-site, and the angiographic end point was first-pass effect (FPE) success rate from a single device attempt (modified Thrombolysis in Cerebral Infarction, ≥2c) as determined by a core laboratory. Achieving modified FPE (modified Thrombolysis in Cerebral Infarction, ≥2b) was also assessed. Comparisons of clinical outcomes were made between groups and adjusted for baseline and procedural characteristics. All participating centers received institutional review board approval from their respective institutions. Results— Adjunctive technique groups included BGC (n=445), distal access catheter (n=238), and conventional guide catheter (n=62). The BGC group had a higher rate of FPE following first pass (212/443 [48%]) versus conventional guide catheter (16/62 [26%]; P =0.001) and distal access catheter (83/235 [35%]; P =0.002). Similarly, the BGC group had a higher rate of modified FPE (294/443 [66%]) versus conventional guide catheter (26/62 [42%]; P P =0.003). The BGC group achieved the highest rate of functional independence (253/415 [61%]) versus conventional guide catheter (23/55 [42%]; P =0.007) and distal access catheter (113/218 [52%]; P =0.027). Final revascularization and mortality rates did not differ across the groups. Conclusions— BGC use was an independent predictor of FPE, modified FPE, and functional independence, suggesting that its routine use may improve the rates of early revascularization success and good clinical outcomes. Clinical Trial Registration— URL: https://www.clinicaltrials.gov . Unique identifier: NCT02239640.

Introduction: Classification of intracranial bleeding (ICB) by trauma status is important in research relating to anticoagulant therapy. We evaluated trauma status of ICB subtypes - intracerebral hemorrhage (ICH), subarachnoid hemorrhage (SAH) and subdural hematoma (SDH) - in a large population of patients with non-valvular atrial fibrillation (NVAF) receiving oral anticoagulant (OAC) therapy. Methods: Using the IQVIA Medical Research Data UK primary care database, we followed 45,164 patients with NVAF newly prescribed OAC therapy to identify incident ICB cases, and categorized them as ICH, SAH or SDH based on the coded description. We identified cases with a code suggesting the event was traumatic (e.g. head injury) or non-traumatic within 90 days either side of the ICB event, and manually reviewed the patient records for cases without such a code to identify other entries indicating trauma in the time period. ICH and SAH cases without a trauma/non-trauma code were classed as non-traumatic and SDH cases without a trauma/non-trauma code were classed as traumatic. Results: A total of 286 ICB cases were identified: 45.5% ICH, 41.9% SDH, 12.6% SAH. Classification of cases is shown in the Table. Most traumatic cases were SDH (77.3%, 85/110); most non-traumatic cases were ICH (66.5%, 117/176). Classification by the presence of a trauma/non-trauma code was 82.5% for SDH, 33.3% for SAH and 10.8% for ICH. Classification of all but one non-traumatic ICH case and all non-traumatic SAH cases were by the absence of a trauma/non-trauma code. For SDH, 75.3% of traumatic SDH cases were based on a trauma code, and 100% of non-traumatic SDH cases were based on a non-trauma code. Conclusion: Among patients with NVAF receiving OAC therapy, the majority of incident ICB cases were non-traumatic. Classification of SDH was predominantly made from trauma/non-trauma codes; cases of ICH and SAH were mostly based on the absence of trauma/non-trauma codes using non-trauma as the default category.

Introduction: Magnetic resonance imaging (MRI) has historically been contraindicated for patients with MRI non-conditional (i.e. legacy) cardiac implantable electronic devices (CIED). Recent trials have demonstrated safety of MRI in legacy CIED patients, with the Center for Medicare & Medicaid Services (CMS) revising MRI coverage to include these patients in 4/2018. We sought to determine the effect of this policy change on MRI utilization for legacy CIED patients with acute ischemic stroke or transient ischemic attack (AIS/TIA) and contemporary use of MRI for these patients. Methods: We performed an interrupted time series analysis of MRI utilization for AIS/TIA patients with the CMS MRI coverage revision for legacy CIED patients serving as the intervention. Using Optum claims data from 1/2012 to 7/2019, we identified AIS/TIA hospitalizations and CIED implantations and interrogations using ICD-9/10 and CPT codes, respectively. The intervention’s effect on MRI utilization for AIS/TIA was determined for patients with and without CIEDs separately. For patients treated after the CMS coverage revision, we used multivariable logistic regression to determine the association between lack of CIED and MRI utilization for AIS/TIA. Results: We identified 417,899 patients hospitalized for AIS/TIA, of which 30,425 patients (7%) had a CIED (CIED patients: age 78.0 ±10.2 years, 45% female; non-CIED patients: age 74.1 ±11.8 years, 55% female). From 2012 to 2019, MRI utilization for AIS/TIA increased from 3% to 20% for CIED patients and 58% to 66% for non-CIED patients. The CMS coverage revision was associated with a 4.2% absolute (25% relative) additional increase in MRI utilization for CIED patients with AIS/TIA. In multivariable regression, non-CIED patients treated after the CMS coverage revision, as compared to CIED patients, were substantially more likely to undergo MRI for AIS/TIA (adjusted OR 6.7, 95% CI: 6.3-7.1, p Conclusions: Despite an increase in MRI utilization for AIS/TIA patients with CIEDs attributable to the CMS coverage revision and trials demonstrating safety, a large disparity in use of MRI for AIS/TIA patients with CIEDs persists. Identification and resolution of barriers to appropriate MRI use in AIS/TIA patients with CIEDs are needed.

Introduction: Successful and fast reperfusion with endovascular thrombectomy (EVT) improves outcomes in acute ischemic strokes due to large vessel occlusion (LVO). While complete reperfusion (mTICI 3) is the ideal target, multiple passes could prolong EVT and increase complications likelihood without improving clinical outcomes. We hypothesized that pts with a single pass mTICI2b would have similar good outcomes to those with multiple passes mTICI3 with a better safety profile. Methods: From the prospective multicenter cohort study of imaging selection (SELECT), EVT pts who achieved mTICI2b and mTICI3 were stratified into those with single pass vs multiple passes. Functional independence rates at 90 day mRS and safety (sICH, neuro-worsening and mortality) were compared between pts with single pass mTICI 2b and multiple passes mTICI 3. Results: Of 361 pts enrolled, 285 received EVT, of those 70 (25%) achieved mTICI 2b and 159 (56%) mTICI 3. 89 (31%) achieved mTICI 3 with single pass, 70 (25%) with multiple passes, 33 (12%) had mTICI 2b with a single pass and 37 (13%) with multiple passes. Baseline characteristics were similar between the two groups except for larger perfusion deficit (Tmax >6s) volume in pts with multiple passes mTICI3 55 (25, 99) cc vs first pass mTICI 2b 43 (4, 79), p=0.047). Functional independence rates were higher with single pass mTICI 2b as compared to multiple passes mTICI 3 (70% vs 56%, aOR=1.51, 95% CI=0.48-4.76, p=0.78), fig 1 but did not reach statistical significance. Multiple passes mTICI 3 was associated with numerically higher mortality (13% vs 3%, p=0.16) and Neuro-worsening (13% vs 3%, p=0.16) with similar sICH: 4% vs 3%, p=1.00 rates. Conclusion: Complete reperfusion with multiple passes did not confer better outcome rates than single pass mTICI 2b and was associated with worse safety profiles. These results suggest that in the absence of complete reperfusion on first attempt, a single pass mTICI 2b may be sufficient for a successful thrombectomy.

Introduction: To evaluate overall ischemic stroke rates, specific subtypes, and clinical presentation during the COVID-19 pandemic in a multicenter observational study from eight states across US. Methods: We compared all ischemic strokes admitted between January 2019 and May 2020, grouped as; March-May 2020 (COVID-19 period), March-May 2019 (seasonal pre-COVID period) and November 2019-January 2020 (immediate pre-COVID-19 period). Primary outcome was stroke severity at admission measured by NIHSS stratified as mild (0-7), moderate (8-14), and severe (>14) symptoms. Secondary outcomes were number of large vessel occlusions (LVOs), stroke etiology, IV-tPA rates, and disposition. Results: Of the 7,969 patients diagnosed with acute stroke during the study period, 933 (12%) presented in the COVID-19 period, 1319 (17%), and 1254 (16%) presented in the seasonal pre-COVID-19 and immediate pre-COVID-19 periods, respectively. Significant decline was observed in the weekly mean volume of newly diagnosed strokes (98±7.3 vs 50±20, p Conclusions: We observed a decrease in newly diagnosed stroke cases and rates of IV-tPA administration, while the LVO frequency remained unchanged during the COVID-19 pandemic. Additionally, these stroke patients had more severe presentations.

Background and Purpose: Time elapsed from last-known well (LKW) and baseline imaging results are influential on endovascular thrombectomy (EVT) outcomes. Methods: In a prospective multicenter cohort study of imaging selection for endovascular thrombectomy (SELECT [Optimizing Patient’s Selection for Endovascular Treatment in Acute Ischemic Stroke], the early infarct growth rate (EIGR) was defined as ischemic core volume on perfusion imaging (relative cerebral blood flow2. The primary outcome was 90-day functional independence (modified Rankin Scale score =0–2). Results: Of 445 consented, 361 (285 EVT, 76 medical management only) patients met the study inclusion criteria. The optimal EIGR was P P P P =0.001). EIGR independently correlated with functional independence after EVT, adjusting for age, National Institutes of Health Stroke Scale, time LKW to groin puncture, reperfusion (modified Thrombolysis in Cerebral Infarction score of ≥2b), IV-tPA (intravenous tissue-type plasminogen activator), and transfer status (aOR: 0.78 [0.65–0.94], P =0.01). Slow progressors had higher functional independence rates (121 [61%] versus 30 [35%], P P =0.001) as compared to fast progressors, who had substantially worse clinical outcomes both in early and late time window. The odds of good outcome decreased by 14% for each 5 mL/h increase in EIGR (aOR, 0.87 [0.80–0.94], P Conclusions: The EIGR strongly correlates with both collateral status and clinical outcomes after EVT. Fast progressors demonstrated worse outcomes when receiving EVT beyond 6 hours of stroke onset as compared to those who received EVT within 6 hours. Registration: URL: https://clinicaltrials.gov . Unique identifier: NCT02446587.

Background and Purpose— Mechanical thrombectomy with stent retrievers has become standard of care for treatment of acute ischemic stroke patients because of large vessel occlusion. The STRATIS registry (Systematic Evaluation of Patients Treated With Neurothrombectomy Devices for Acute Ischemic Stroke) aimed to assess whether similar process timelines, technical, and functional outcomes could be achieved in a large real world cohort as in the randomized trials. Methods— STRATIS was designed to prospectively enroll patients treated in the United States with a Solitaire Revascularization Device and Mindframe Capture Low Profile Revascularization Device within 8 hours from symptom onset. The STRATIS cohort was compared with the interventional cohort of a previously published SEER patient-level meta-analysis. Results— A total of 984 patients treated at 55 sites were analyzed. The mean National Institutes of Health Stroke Scale score was 17.3. Intravenous tissue-type plasminogen activator was administered in 64.0%. The median time from onset to arrival in the enrolling hospital, door to puncture, and puncture to reperfusion were 138, 72, and 36 minutes, respectively. The Core lab–adjudicated modified Thrombolysis in Cerebral Infarction ≥2b was achieved in 87.9% of patients. At 90 days, 56.5% achieved a modified Rankin Scale score of 0 to 2, all-cause mortality was 14.4%, and 1.4% suffered a symptomatic intracranial hemorrhage. The median time from emergency medical services scene arrival to puncture was 152 minutes, and each hour delay in this interval was associated with a 5.5% absolute decline in the likelihood of achieving modified Rankin Scale score 0 to 2. Conclusions— This largest-to-date Solitaire registry documents that the results of the randomized trials can be reproduced in the community. The decrease of clinical benefit over time warrants optimization of the system of care. Clinical Trial Registration— URL: http://www.clinicaltrials.gov . Unique identifier: NCT02239640.

Background: The effect of anesthesia choice on endovascular thrombectomy (EVT) outcomes is unclear. Methods: In the prospective multicenter cohort study of imaging selection for EVT (SELECT), patients were stratified based on their anesthesia type into general anesthesia (GA) and conscious sedation (CS). EVT times and outcomes were compared. Further we assessed the impact of ischemic core size (rCBF Results: Of 361 enrolled, 285 received EVT. 129 (45%%) received GA and 156 (54%) CS. The baseline characteristics were similar, except for presentation NIHSS (GA 17(13-21), CS 15(11-20), p=0.027) and ischemic core volume (GA 14.1 cc (0-38) vs CS 6.3(0-26.1), p=0.034). GA was associated with numerically longer arrival to GP times 92 (68—115) vs. 85(60-117) mins, p=0.58. After adjustment for baseline imbalances, patients who received CS had a shift toward better outcome (adj cOR 1.72, 95% CI=1.08-2.75, p=0.022) with higher functional independence rates 56.8% vs 48.8%, p=0.75. Furthermore, GA was associated with higher mortality rates (19% vs 9%, p=0.017), figure 1A. In patients with core volume ≥ 50 cc, there was a trend for a shift towards better outcomes (adj cOR=5.84, 95%CI= 0.90-38.00, P=0.065), figure 1B while there was no difference in patients with core volume < 50 cc (adj cOR=1.01 (95%CI 0.53-1.94, P=0.96), figure 1C. There was an interaction between core volume size and anesthesia type on functional outcome (p=0.042). For every 10cc increase in the core volume, the odds of attaining better functional outcome decreased by 29% (adjusted cOR: 0.71, 95% CI=0.61-0.83, p Conclusion: Conscious sedation was associated with a shift towards better EVT outcomes. This effect was driven by patients with larger ischemic core volumes and has implications for randomized trials of conscious sedation vs general anesthesia.

Background: Whether IV tPA has adjunctive benefit to endovascular thrombectomy (EVT) is unclear. Methods: In a prospective multicenter cohort study of imaging selection for EVT (SELECT), patients who received IV tPA vs. no IV tPA were compared stratified by stroke severity and ischemic core size (rCBF Results: Of 361 enrolled, 285 received EVT; 226 presented within 4.5 hrs, 162 (72%) received IV tPA. IV tPA patients had lower median ASPECTS (8 vs 9, p=0.007) and larger ischemic core size (11.4 (1.5-37) vs 3.9 (0-32.15), p=0.042, otherwise similar at baseline. There were no delays in EVT delivery associated with tPA: median time (IQR) from arrival to groin puncture 95.0 min (66.0, 118.0) tPA vs 81.5 (63.5, 107.5) no tPA, p=0.21. IV tPA use was associated with higher mRS 0-2 rates (57% vs 44%), aOR 2.02 (95% CI 1.01-4.03, p=0.046) after adjustment for baseline differences with a shift towards better outcomes on all mRS levels (cOR 2.06, 95% CI 1.18-3.59, p=0.01) with lower mortality (11% vs 22%, p=0.026) and similar sICH rates (and 6% vs 6%, p=1.0). In patients with NIHSS Conclusion: IV tPA did not result in thrombectomy delivery delays and may result in better outcomes. Patients with less severe strokes and smaller infarct size had a stronger association between the use of IV tPA and favorable outcomes.

Background: The efficacy of endovascular thrombectomy (EVT) in M2 occlusions is uncertain. Methods: In a prospective multicenter cohort study of imaging selection (SELECT), EVT outcomes were compared to medical managment (MM) in M2 occlusions. Further, we assessed for potential treatment benefit in patients with higher stroke severity (NIHSS) and a larger perfusion deficit on CTP (Tmax > 6 sec - ischemic core volume)The primary outcome was excellent outcome (mRS 0-1). Results: of 361 patients enrolled in SELECT, 87 had isolated M2 occlusion (EVT 59, MM 28). Baseline NIHSS median (IQR) (EVT 14 (10-20), MM 15 (9.5-19.5), p=0.72) and infarct volume rCBF10 had better outcomes with EVT (46%) vs MM (10%), aOR=11.39, 95% CI=1.80-72.11, p=0.01 as shown in figure 1. As perfusion deficit lesion size increased, the odds of achieving excellent outcomes was reduced (for each 10cc by 11%, aOR: 0.89, 95% CI=0.79-1.00, p=0.05). Excellent outcomes declined in patients with MM as perfusion deficit lesion size increased, yet in the EVT they were maintained as shown in figure 2. Similar results were obtained for mRS 0-2. Conclusion: EVT may result in better rates of excellent outcomes in isolated M2 occlusions, especially those with more severe strokes and larger perfusion deficits who are more likely to have worse outcomes without emergent reperfusion.

Background: Whether Endovascular Thrombectomy (EVT) is cost-effective in large ischemic core infarcts is unknown. Methods: In the prospective multicenter cohort study of imaging selection study (SELECT), large core was defined as CT ASPECTS < 6 or CTP ischemic core volume (rCBF Results: Of 361 enrolled, 105 had large core on CT or CTP (EVT 62, MM 43). 19 (31%) EVT patients achieved mRS 0-2 vs 6 (14%) MM (aOR: 3.27, 95% CI: 1.11-9.62; P = .03) with a shift towards better mRS (adj cOR: 2.12, 95% CI: 1.05-4.31, P = .04). Over 20 years EVT was associated with $26589 (C.I. $8672- $43978) incremental costs and a gain of 1.18 QALYs (C.I. 0.091- 2.2) per patient. EVT could avert 75 deaths over a theoretical cohort of 1000 patients (MM 861 vs EVT 786) thus the ICER of EVT compared to MM was $22400 per QALY (CI. $10109 - $66140), which is EVT has a higher NMB compared to MM at the lower and upper WTP thresholds (EVT $86,3 and 271,4 million vs MM $53,6-$179,3 million), Tab 2. The PSA confirmed the results (fig 1). The CEAC showed 94% and 97% cost-effectiveness probability of EVT at the lower and upper values respectively of the maximum WTP, fig 2. EVT ICER in SELECT large core ($22400/QALY) was higher but still comparable to those in HERMES ($16882/QALY), DAWN ($7335/QALY) and DEFUSE3 ($14673/QALY), Tab 3. Conclusion: EVT may result in better outcomes and more lives saved in large core patients with higher QALYs, NMB and an acceptable ICER. The results were comparable to other EVT RCTs.

Background and Purpose- It remains unclear how experience influences outcomes after the advent of stent retriever technology. We studied the relationship between site experience and outcomes in the Trevo Acute Ischemic Stroke multicenter registry. Methods- The 24 sites that enrolled patients in the Trevo Acute Ischemic Stroke registry were trichotomized into low-volume (2 cases/month), medium-volume (2-4 cases/month), and high-volume centers (4 cases/month). Baseline features, imaging, and clinical outcomes were compared across the 3 volume strata. A multivariable analysis was performed to assess whether outcomes were influenced by site volumes. Results- A total of 624 patients were included and distributed as low- (n=188 patients, 30.1%), medium- (n=175, 28.1%), and high-volume (n=261, 41.8%) centers. There were no significant differences in terms of age (mean, 66±16 versus 67±14 versus 65±15; P=0.2), baseline National Institutes of Health Stroke Scale (mean, 17.6±6.5 versus 16.8±6.5 versus 17.6±6.9; P=0.43), or occlusion site across the 3 groups. Median (interquartile range) times from stroke onset to groin puncture were 266 (181.8-442.5), 239 (175-389), and 336.5 (221.3-466.5) minutes in low-, medium-, and high-volume centers, respectively (P=0.004). Higher efficiency and better outcomes were seen in higher volume sites as demonstrated by shorter procedural times (median, 97 versus 67 versus 69 minutes; P0.001), higher balloon guide catheter use (40% versus 36% versus 59%; P≤0.0001), and higher rates of good outcome (90-day modified Rankin Scale [mRS], ≤2; 39% versus 50% versus 53.4%; P=0.02). There were no appreciable differences in symptomatic intracranial hemorrhage or 90-day mortality. After adjustments in the multivariable analysis, there were significantly higher chances of achieving a good outcome in high- versus low-volume (odds ratio, 1.67; 95% CI, 1.03-2.7; P=0.04) and medium- versus low-volume (odds ratio, 1.75; 95% CI, 1.1-2.9; P=0.03) centers, but there were no significant differences between high- and medium-volume centers (P=0.86). Conclusions- Stroke center volumes significantly influence efficiency and outcomes in mechanical thrombectomy.