DWI Lesion Research Articles

Clopidogrel Plus Aspirin in Patients With Different Types of Single Small Subcortical Infarction.

Background: We aim to investigate the effects and safety of clopidogrel plus aspirin in patients with different types of single small subcortical infarction (SSSI) in the Clopidogrel in High-risk patients with Acute Non-disabling Cerebrovascular Events (CHANCE) trial.Methods: SSSI was defined as single DWI lesion of ≤2.0 cm. Patients with SSSI were divided into SSSI + PAD (parent artery disease) and SSSI – PAD, according to the stenosis of the parent artery. The efficacy outcome was stroke recurrence during 90-day follow-up. Cox proportional hazards models or logistic regression models were used to assess the interaction of the treatment effects of clopidogrel plus aspirin vs. aspirin alone among patients with and without PAD.Results: Among 338 patients with SSSI included in the subanalysis, 105 were with PAD and 233 without. The efficacy of clopidogrel plus aspirin compared with aspirin alone on any stroke was consistent between patients with [adjusted hazard ratio (HR) 0.84; 95% confidence interval (CI), 0.25–2.75] and without PAD (adjusted HR 1.03; 95% CI, 0.40–2.68, interaction P = 0.83). In patients with SSSI + PAD, the rate of stroke recurrence in those treated with dual antiplatelet therapy and mono antiplatelet therapy was not significantly different (10.9 vs. 13.6%, P = 0.77). The number of bleeding events was similar between the clopidogrel-aspirin group and aspirin group regardless of SSSI + PAD or SSSI – PAD.Conclusions: There was no significant difference in the efficacy of clopidogrel plus aspirin compared with aspirin alone between patients with SSSI + PAD and SSSI – PAD in the CHANCE trial. Studies in other populations and with adequate power are needed to further verify such findings.

Incidence of Silent Ischemic Infarct after Diagnostic Conventional Cerebral Angiogram.

This study was carried out to determine the risk of new ischemic events, demonstrated by diffusion weighted magnetic resonance imaging (DW-MRI) in patients undergoing diagnostic conventional cerebral angiogram. MEDLINE, EMBASE, Current Contents, Cochrane Central, Expert Files, and bibliographies of included articles were systematically searched to identify prospective studies that reported MRI findings after diagnostic angiogram. Eligibility criteria included the following elements: manuscripts published between 1999 and 2019; prospective design; only diagnostic angiogram imaging performed prior to the MRI; DW-MRI done before and after the angiogram procedure; and number of new diffusion weighted lesions documented after each procedure. Six studies met the eligibility criteria. The total number of diagnostic angiograms reported was 430. DWI lesions were associated with diagnostic angiograms in 106 (24.65%) of total procedures. Mean time fluoroscopy, procedure by resident operator, and vascular risk factors were the recognised independent risk factors for silent cerebral ischemia after diagnostic angiogram. Key Words: Stroke, Diffusion magnetic resonance imaging, Cerebral infarction, Angiography, Brain ischemia.

Abstract P335: Post-Treatment Blood-Brain Barrier Disruption is Associated With Cerebral Edema and Extends Into Penumbral Tissue in Anterior Circulation Large Vessel Occlusion

Background: Blood Brain Barrier (BBB) disruption is central to vasogenic edema development after ischemia in preclinical studies. We investigated the relationship between BBB disruption and cerebral edema in patients receiving reperfusion therapies for anterior circulation large vessel occlusion. Methods: In a post-hoc pooled analysis of the Tenecteplase versus Alteplase before Endovascular Therapy for Ischemic Stroke (EXTEND-IA TNK) part 1 and 2 multicenter RCTs, the DWI lesion on 24h post-treatment MRI and peri-infarct salvaged penumbra derived from pre-treatment CT-Perfusion (CTP) were segmented and registered to 24h MR Perfusion. A validated permeability measure (percentage of measured cerebral blood volume lost due to T1 effect from gadolinium leakage across the BBB) was calculated for each ROI. The association between post-treatment BBB disruption in the infarct and cerebral edema assessed on a trichotomized ordinal scale of negligible (<1mm midline shift), mild (≥1 to <5mm) or severe (≥5mm) was analyzed using ordinal logistic regression. Results: Of 220 patients analyzed, median (IQR) BBB disruption was 2.36(1.4-4.1)% in the infarct, 1.61(1.0-2.6)% in salvaged penumbra and 0.98(0.7-1.5)% in normal brain (p<0.001). There were 119 (54.1%) patients with negligible, 90 (40.8%) mild and 11 (5.0%) severe cerebral edema at 24h. In multivariable analysis, infarct BBB disruption was associated with increased cerebral edema (cOR=1.11 per%, 95%CI 1.02-1.21;p=0.012, adjusted for age, admission NIHSS and reperfusion (mTICI2b/3 or >50% reperfusion on early CTP 2h post-lysis). Infarct BBB disruption correlated with follow-up infarct volume (rho=0.37;p<0.001), and was associated with parenchymal hematoma (aOR=1.18, 95%CI 1.00-1.40;p=0.048) and worse outcome assessed on 90-day modified Rankin Scale (cOR=1.16 per%, 95%CI 1.08-1.26;p<0.001, adjusted for age, admission NIHSS and reperfusion). Conclusions: BBB disruption after reperfusion treatment extends beyond the infarct lesion, and is associated with cerebral edema development, hemorrhagic transformation and poor outcome. Further studies to evaluate BBB integrity as an imaging biomarker and potential therapeutic target in malignant cerebral edema after ischemic stroke are needed.

Abstract P331: Accuracy of CT Perfusion Core Estimates for Predicting Infarct Size in the SELECT Study

Background: The accuracy of CT perfusion imaging for estimating the ischemic core has been questioned. Methods: In SELECT, a prospective cohort study of imaging selection, pts who achieved complete reperfusion after EVT were stratified on time from LKW to imaging acquisition and time from imaging to reperfusion. The difference between baseline CTP core volume and f/up infarct volume (on DWI after EVT) was classified as over-estimation (core >10 cc larger than infarct), adequate, or under-estimation (≥ 25 cc smaller). F/up DWI lesion was outlined using a semiautomated algorithm and co-registered to CTP. Results: Of 361 enrolled, 117 achieved TICI 3. F/up MRIs were acquired at 21 (13-30) hrs from EVT with median infarct volume of 16.4 cc, median 8.1 cc larger than baseline core. Median (IQR) time from imaging acquisition to groin puncture (GP) was 70 (50-95) min. Reperfusion was achieved at 35 (25-54) min of GP. The frequency of overestimation decreased as time LKW to imaging increased: < 90 min 6 (14%), 90 – 270 min 3 (6%) and > 270 min 1 (4%), and adequate estimation increased (< 90 min 21 (50%), 90 – 270 min 32 (65%) and > 270min 19 (73%), p for trend 0.048) Fig 1. Overestimation primarily occurred in pts imaged within 90 min who had short imaging to reperfusion times Fig 2. Volumetric correlation between pre-procedure and f/up imaging improved as LKW time to imaging acquisition increased; Spearman’s ρ: <90 min: 0.41 (p=0.007), 90-270 min: 0.35 (p=0.01), >270 min: 0.79 (p<0.0001). Spatially, overestimation occurred predominantly in white matter juxtacortical areas. Adjusting rCBF threshold from < 30% to < 20% in the 6 pts with overestimation ≤ 90 min from LKW resulted in adequate core estimation in all 6, Fig 3. Conclusion: In patients who achieve reperfusion, the correlation between baseline CTP ischemic core volume and f/up DWI volume improved as time LKW to imaging increased. Core estimation accuracy improved by using the < 20% CBF threshold for patients imaged within 90 minutes of LKW.

Abstract P325: Validation of Deep Learning Based Critical Hypoperfusion and Ischemic Core Prediction in a Multicenter External Randomized Controlled Trial

Objective: We previously developed two separate deep learning (DL) models to segment the ischemic core and critically hypoperfused tissue on baseline imaging of acute ischemic stroke patients. We aimed to validate the models in an external, multi-center randomized clinical trial (DEFUSE3) and compare with the current clinical standard. Methods: The DL models were previously trained in a separate dataset in which follow-up MRI, obtained at 3-7 days, was used as the reference for critically hypoperfused tissue in patients who did not reperfuse and as the reference for the ischemic core in patients who did reperfuse. For validation, we included DEFUSE3 patients with adequate quality baseline MR perfusion and a 24-hour follow-up DWI scan. The 24-hour DWI lesion served as the reference for ischemic core in patients in the thrombectomy arm and for critically hypoperfused tissue for patients in the medical arm. RAPID was used to generate perfusion maps (Tmax, cerebral blood flow, cerebral blood volume, and mean transient time). The accuracy of segmenting the ischemic core and critically hypoperfused tissue on baseline imaging was compared between the DL approach and the traditional thresholding approach implemented in RAPID. Results: In the 46 patients included for analysis, 24 were in the medical arm and 22 in the thrombectomy arm. Compared to a traditional thresholding method, the DL model segmented the ischemic core more accurately (AUC of 0.92 vs 0.72, p=0.0001 and volume difference of -8ml vs -21ml, p=0.001). Similarly, the DL model segmented critically hypoperfused tissue more accurately (AUC of 0.93 vs 0.80, p<0.0001; volume difference 14ml vs. 55ml, p=0.0005). However, great heterogeneity in final infarct was noticed in medical arm. See tables and figures. Conclusions: The DL-based critical hypoperfusion and ischemic core prediction provides more accurate prediction on final infarct than a commonly used thresholding method in this external validation.

Abstract P370: Early Brain Volume Change After Stroke: Subgroup Analysis From the Axis-2 Trial

Objective: The timing of brain atrophy after stroke is not well understood. We investigated the associations between age and imaging features and brain volume change in the first month after stroke. Methods: We retrospectively studied patients with acute ischemic stroke enrolled in the AXIS-2 trial. Total brain volume change from hyperacute MRI data to the first month after stroke was assessed using unified segmentation in SPM12. We hypothesized that age, ischemic brain lesion size and white matter changes were associated with larger brain volume change. Enlarged perivascular spaces (EPVS) and white matter hyperintensities (WMH) were rated visually and the presence of lacunes was assessed. Results: We enrolled 173 patients with a mean age of 67±11 years, 44 % were female. There was a median 6 mL decrease in volume (25th percentile -1mL-75th percentile 21mL) over time, equivalent to a median 0.5% (IQR-0.07% - 1.4%), decrease in brain volume. Age was associated with larger brain volume loss (per 10 years of age, 5 mL 95% CI 2 mL-8 mL). Baseline DWI lesion volume was not associated with greater volume loss (per 10 mL of lesion volume, change by 0 mL (95% CI -0.1 mL-0.1mL). Increasing Fazekas scores of deep white matter hyperintensity were associated with greater tissue loss (5 mL, 95% CI 1 mL-10 mL). Brain volume change was associated with ordinal change on the modified Rankin scale at 90 days (per 10 ml, OR 1.085, 95% CI 1.031-1.141) but the association was not present when adjusting for age, baseline diffusion lesion volume and NIHSS score (OR 1.044, 95% CI 0.990-1.010).Interpretation-Global brain volume loss occurs over 1 month after stroke and is associated with age and deep white matter disease. We did not find evidence that more severe strokes lead to increased early tissue loss.

Collateral and permeability imaging derived from dynamic contrast material-enhanced MR angiography in prediction of PH 2 hemorrhagic transformation after acute ischemic stroke: a pilot study.

To evaluate the role of collateral and permeability imaging derived from dynamic contrast material-enhanced magnetic resonance angiography to predict PH 2 hemorrhagic transformation in acute ischemic stroke. The secondary analysis of a published data from participants with acute ischemic stroke. The multiphase collateral map and permeability imaging were generated by using dynamic signals from dynamic contrast material-enhanced magnetic resonance angiography obtained at admission. To identify independent predictors of PH 2 hemorrhagic transformation, age, sex, risk factors, baseline National Institutes of Health Stoke Scale (NIHSS) score, baseline DWI lesion volume, collateral-perfusion status, mode of treatment, and successful early reperfusion were evaluated with multiple logistic regression analyses and the significance of permeability imaging in prediction of PH 2 hemorrhagic transformation was evaluated by subgroup analysis. In 115 participants, including 70 males (mean (SD) age, 69 (12) years), PH 2 hemorrhagic transformation occurred in 6 participants with very poor collateral-perfusion status (MAC 0). MAC 0 (OR, 0.06; 95% CI, 0.01, 0.74; P = .03) was independently associated with PH 2 hemorrhagic transformation. In 22 participants with MAC 0, the permeable signal on Kep permeability imaging was the only significant characteristic associated with PH 2 hemorrhagic transformation (P = .009). The specificity of Kep permeability imaging was 93.8% (95% confidence interval: 69.8, 99.8) in predicting PH 2 hemorrhagic transformation. Individual-based prediction of PH 2 hemorrhagic transformation in patients with acute ischemic stroke may be possible with multiphase collateral map and permeability imaging derived from dynamic contrast material-enhanced magnetic resonance angiography.

Combined Perfusion and Permeability Imaging Reveals Different Pathophysiologic Tissue Responses After Successful Thrombectomy

Despite successful recanalization of large-vessel occlusions in acute ischemic stroke, individual patients profit to a varying degree. Dynamic susceptibility-weighted perfusion and dynamic T1-weighted contrast-enhanced blood-brain barrier permeability imaging may help to determine secondary stroke injury and predict clinical outcome. We prospectively performed perfusion and permeability imaging in 38 patients within 24 h after successful mechanical thrombectomy of an occlusion of the middle cerebral artery M1 segment. Perfusion alterations were evaluated on cerebral blood flow maps, blood-brain barrier disruption (BBBD) visually and quantitatively on ktrans maps and hemorrhagic transformation on susceptibility-weighted images. Visual BBBD within the DWI lesion corresponded to a median ktrans elevation (IQR) of 0.77 (0.41–1.4) min−1 and was found in all 7 cases of hypoperfusion (100%), in 10 of 16 cases of hyperperfusion (63%), and in only three of 13 cases with unaffected perfusion (23%). BBBD was significantly associated with hemorrhagic transformation (p < 0.001). While BBBD alone was not a predictor of clinical outcome at 3 months (positive predictive value (PPV) = 0.8 [0.56–0.94]), hypoperfusion occurred more often in patients with unfavorable clinical outcome (PPV = 0.43 [0.10–0.82]) compared to hyperperfusion (PPV = 0.93 [0.68–1.0]) or unaffected perfusion (PPV = 1.0 [0.75–1.0]). We show that combined perfusion and permeability imaging reveals distinct infarct signatures after recanalization, indicating the severity of prior ischemic damage. It assists in predicting clinical outcome and may identify patients at risk of stroke progression.

Abstract P492: The Hypoperfusion Intensity Ratio is a Poor Discriminator of Infarct Growth in Fully Reperfused Patients

Background and Purpose: The hypoperfusion intensity ratio (HIR) is a perfusion-weighted imaging parameter defined as the ratio of Tmax&gt;10 seconds : Tmax&gt;6 seconds volume and is believed to be reflective of collateral strength and consequently influence infarct growth. We sought to assess the utility of the HIR in predicting infarct growth in patients undergoing thrombectomy at a comprehensive stroke center (CSC). Methods: Consecutive acute ischemic stroke patients transferred to our CSC from 09/2010-11/2018 were identified and included if the following criteria were met: 1)computed tomography perfusion (CTP) imaging enabling assessment of baseline ischemic core volume and HIR 2) follow-up neuroimaging for assessment of final infarct volumes and 3)modified Thrombolysis In Cerebral Infarction scale (mTICI) 2c status or greater post-thrombectomy. Infarct growth rate (IGR) was calculated as the difference between infarct volume on follow-up imaging and the acute DWI lesion volume, divided by time from CTP to reperfusion in hours. Results: 461 patients (median age, 64 [55-75] years, median baseline NIHSS, 16 [12-21]) were eligible for this analysis. HIR poorly correlated with IGR (Spearman’s rho=0.001, p=0.89). An HIR cut-off of 0.5 was not able to discriminate ‘fast progressors’ (IGR&gt;5 mL/hr) (AUC 0.42, sensitivity 40%, specificity 51%), or IGR at thresholds of either 2.5 or 10 mL/hr (AUC 0.44 and 0.49 respectively, with 95% confidence intervals [0.35-0.52] and [0.41-0.57], respectively). Similarly, an HIR of 0.5 only weakly distinguished ‘fast progression’ in patients reperfused beyond 120 min from imaging and patients with early CTP (last known well to CTP&lt;6 hrs) (AUC 0.59, sensitivity 43%, specificity 68% and AUC 0.50, sensitivity 45%, specificity 55%). On multiple regression analysis, HIR was not predictive of infarct growth (regression equation=18.09+8.48x, F=2.46, p=0.11, R 2 =0.13) but was predictive of ‘fast progression’ (OR 0.22, 95% CI [0.09-0.60], p=0.003, pseudo-R 2 =0.16). Conclusions: Though predictive of fast progression, the HIR is a poor discriminator of infarct growth in successfully reperfused thrombectomy patients who undergo perfusion imaging at a CSC, and thus should not be factored into treatment decision-making.

TNF-R1 Correlates with Cerebral Perfusion and Acute Ischemia Following Subarachnoid Hemorrhage.

Early cerebral hypoperfusion and ischemia occur after subarachnoid hemorrhage (SAH) and influence clinical prognosis. Pathophysiological mechanisms possibly involve inflammatory mediators. TNF-α has been associated with complications and prognosis after SAH. We investigated the relation of perfusion parameters and ischemic lesions, with levels of TNF-α main receptor, TNF-R1, after SAH, and their association with prognosis. We included consecutive SAH patients admitted within the first 72h of SAH onset. Blood samples were simultaneously collected from a peripheral vein and from the parent artery of the aneurysm. Levels of TNF-R1 were measured using ELISA (R&D Systems Inc., USA). CT perfusion and MRI studies were performed in the first 72h. Correlation and logistic regression analysis were used to identify outcome predictors. We analyzed 41 patients. Increased levels of TNF-R1 correlated with increased Tmax (arterial: r = -0.37, p = 0.01) and prolonged MTT (arterial: r = 0.355, p = 0.012; venous: r = 0.306, p = 0.026). Increased levels of both arterial and venous TNF-R1 were associated with increased number of lesions on DWI (p = 0.006). In multivariate analysis, venous TNFR1 levels > 1742.2pg/mL (OR 1.78; 95%CI 1.18-2.67; p = 0.006) and DWI lesions (OR 14.01; 95%CI 1.19-165.3; p = 0.036) were both independent predictors of poor outcome (mRS ≥ 3) at 6months. Increased levels of TNF-R1 in arterial and venous blood correlate with worse cerebral perfusion and with increased burden of acute ischemic lesions in the first 72h after SAH. Venous levels of TNF-R1 and DWI lesions were associated with poor outcome at 6months. These results highlight the pathophysiological role of TNF-α pathways in SAH and suggest a possible role of combined imaging and laboratorial markers in determining prognosis in acute SAH.

Cerebral Microembolism in Intracerebral Hemorrhage: A Prospective Case–Control Study

Small and remote acute ischemic lesions may occur in up to one-third of patients with spontaneous intracerebral hemorrhage (ICH). Possible mechanisms include cerebral embolism, small vessel disease, blood pressure variability and others. The embolic mechanism has not been adequately studied. Using transcranial Doppler (TCD), we assessed the incidence of spontaneous microembolic signals (MESs) in patients with acute ICH. Twenty acute ICH patients were prospectively evaluated within 48h of hospital admission. Clinical and imaging data were collected. Continuous TCD monitoring was performed in both middle cerebral arteries for a one-hour period on days 1, 3 and 7 of hospital admission. Monitoring was performed in the emergency room, ICU or ward, according to patient location. We compared the frequency and risk factors for MES in patients with ICH and in 20 age- and gender-matched controls without history of ischemic or hemorrhagic stroke. The mean age was 57.5 ± 14.1years, and 60% were male. MESs were detected in 7 patients with ICH and in one control patient without ICH (35% vs 5%, p = 0.048). The frequency of MES on day 1 was 15% (3 of 20 patients), on day 3, 26% (5 of 19 patients) and on day 7, 37.5% (3 of 8 patients). Among patients with ICH, those with MES had a tendency to higher frequencies of dyslipidemia (83% vs 33%, p = 0.13) and lobar location of hemorrhages (71% vs 30%, p = 0.15). Two out of 6 patients with ICH who also underwent MRI had remote DWI lesions, of whom one showed MES on TCD. Micro-embolic signals occur in over one-third of patients with ICH. Further research is needed to identify the sources of cerebral microembolism and their relationship with small acute infarcts in ICH.

A Prospective Analysis of Lesion-Symptom Relationships in Acute Vestibular and Ocular Motor Stroke.

Background: Diagnosing stroke as a cause of acute vertigo, dizziness, or double vision remains a challenge, because symptom characteristics can be variable. The purpose of this study was to prospectively investigate lesion-symptom relationships in patients with acute vestibular or ocular motor stroke.Methods: Three hundred and fifty one patients with acute and isolated vestibular or ocular motor symptoms of unclear etiology were enrolled in the EMVERT lesion trial. Symptom quality was assessed by the chief complaint (vertigo, dizziness, double vision), symptom intensity by the visual analog scale, functional impairment by EQ-5D-5L, and symptom duration by daily rating. Acute vestibular and ocular motor signs were registered by videooculography. A standardized MRI (DWI-/FLAIR-/T2-/T2*-/3D-T1-weighted sequences) was recorded within 7 days of symptom onset. MRIs with DWI lesions were further processed for voxel-based lesion-symptom mapping (VLSM).Results: In 47 patients, MRI depicted an acute unilateral stroke (13.4%). The chief complaints were dizziness (42.5%), vertigo (40.4%) and double vision (17.0%). Lesions in patients with vertigo or dizziness showed a large overlap in the cerebellar hemisphere. VLSM indicated that strokes in the medial cerebellar layers 7b, 8, 9 were associated with vertigo, strokes in the lateral cerebellar layer 8, crus 1, 2 with dizziness, and pontomesencephalic strokes with double vision. Symptom intensity and duration varied largely between patients. Higher symptom intensity and longer duration were associated with medial cerebellar lesions. Hemispheric lesions of the cortex were rare and presented with milder symptoms of shorter duration.Conclusions: Prospective evaluation of patients with acute vestibular or ocular motor stroke revealed that symptom quality, intensity and duration were not suited to differentiating peripheral from central etiologies. Lesions in the lateral cerebellum, thalamus, or cortex presented with unspecific, mild and transient symptoms prone to being misdiagnosed.

Predictors of malignant middle cerebral artery infarction after mechanical thrombectomy

IntroductionSeveral predictors have been described to early diagnose malignant middle cerebral artery infarction (MMI) and select patient for hemicraniectomy. Nevertheless, few studies have assessed them among patients with acute ischemic stroke undergoing mechanical endovascular thrombectomy (MET). The overall objective in this study was to evaluate these predictors in patients undergoing MET in the purpose to guide the medical care in the acute phase. MethodsWe selected patients from a prospective local database which reference all patients eligible for treatment with Alteplase thrombolysis and/or mechanical endovascular thrombectomy in acute stroke. We investigated demographic, clinical, and radiological data. Multivariate regression analysis was used to identify clinical and imaging predictors of MMI. ResultsIn 32 months, 66 patients were included. Eighteen (27.3%) developed MMI. Malignant evolution was associated with: severity of neurological deficit and level of consciousness at admission, infarct size in DWI sequence and involvement of other vascular territories. Study groups didn’t differ in terms of successful reperfusion. Two variables were identified as independent predictors of MMI: DWI infarct volume (p<0.001) and time to thrombectomy (p=0.018). A decision tree based on these two factors was able to predict malignant evolution with high specificity (100%) and sensibility (73%). ConclusionOur study proposes a practical decision tree including DWI lesion volume and delay before thrombectomy to early and accurately predict MMI in a subgroup of patients with MCA infarction undergoing MET regardless to the status of reperfusion.

Osmotherapy for malignant cerebral edema in a phase 2 prospective, double blind, randomized, placebo-controlled study of IV glibenclamide

Background/ObjectiveMalignant edema can be a life-threatening complication of large hemispheric infarction (LHI), and is often treated with osmotherapy. In this exploratory analysis of data from the GAMES-RP study, we hypothesized that patients receiving osmotherapy had symptomatic cerebral edema, and that treatment with intravenous (IV) glibenclamide would modify osmotherapy use as compared with placebo. MethodsGAMES-RP was a phase 2 multi-center prospective, double blind, randomized, placebo-controlled study in LHI. Patients were randomized to IV glibenclamide (e.g. IV glyburide) or placebo. Cerebral edema therapies included osmotherapy and/or decompressive craniectomy at the discretion of the treating team. Total bolus osmotherapy dosing was quantified by “osmolar load”. Radiographic edema was defined by dichotomizing midline shift at 24 h. Clinical changes were defined as any increase in NIHSS1a. ResultsOsmotherapy was administered to 40 of the 77 patients at a median of 39 [27–55] h after stroke onset. The median baseline DWI lesion volume was significantly larger in the osmotherapy treated group (167 [146–211] mL v. 139 [112–170] mL; P=0.046). Adjudicated malignant edema (75% v. 16%; P<0.001) was more common in the osmotherapy treated group. There were no differences in the proportion of patients receiving osmotherapy or the median total osmolar load between treatment arms. Most patients (76%) had a decrease in consciousness (NIHSS item 1A ≥1) on the day they began receiving osmotherapy. ConclusionsIn the GAMES-RP trial, osmolar therapies were most often administered in response to clinical symptoms of decreased consciousness. However, the optimal timing of administration and impact on outcome after LHI have yet to be defined.

Reversible diffusion-weighted imaging lesions in acute ischemic stroke: A systematic review.

To systematically review the literature for reversible diffusion-weighted imaging (DWIR) lesions and to describe its prevalence, predictors, and clinical significance. Studies were included if the first DWI MRI was performed within 24 hours of stroke onset and follow-up DWI or fluid-attenuated inversion recovery (FLAIR)/T2 was performed within 7 or 90 days, respectively, to measure DWIR. We abstracted clinical, imaging, and outcomes data. Twenty-three studies met the study criteria. The prevalence of DWIR was 26.5% in DWI-based studies and 6% in FLAIR/T2-based studies. DWIR was associated with recanalization or reperfusion of the ischemic tissue with or without the use of tissue plasminogen activator (t-PA) or endovascular therapy, earlier treatment with t-PA, shorter time to endovascular therapy after MRI, and absent or less severe perfusion deficit within the DWI lesion. DWIR was associated with early neurologic improvement in 5 of 6 studies (defined as improvement in the NIH Stroke Scale (NIHSS) score by 4 or 8 points from baseline or NIHSS score 0 to 2 at 24 hours after treatment or at discharge or median NIHSS score at 7 days) and long-term outcome in 6 of 7 studies (defined as NIHSS score ≤1, improvement in the NIHSS score ≥8 points, or modified Rankin Scale score up to ≤2 at 30 or 90 days) likely due to reperfusion. DWIR is seen in up to a quarter of patients with acute ischemic stroke, and it is associated with good clinical outcome following reperfusion. Our findings highlight the pitfalls of DWI to define ischemic core in the early hours of stroke.

Abstract WP488: Imaging Correlates of Vascular Cognitive Impairment After Recent Ischemia in Intracranial Atherosclerosis: Evidence From SAMMPRIS

Background: More than 50% of individuals with recently symptomatic intracranial atherosclerotic disease (ICAD) manifest cognitive dysfunction. The imaging correlates of such cognitive impairment due to large vessel disease remain largely unknown. We examined the degree of cognitive dysfunction associated with detailed quantification of MRI DWI and FLAIR lesions at baseline in the SAMMPRIS trial. Methods: Central, blinded, adjudication of baseline MRI included measurement of DWI and FLAIR lesions and corresponding volumes. Cognitive function was assayed with the Montreal Cognitive Assessment (MoCA) score (0-30). Statistical analyses were used to describe MRI lesion characteristics and the correlation with baseline MoCA score. Results: At enrollment in SAMMPRIS, baseline DWI was available in 309/451 (69%) subjects, with FLAIR in 293/451 (65%). Baseline MoCA was median 25.0 (10-30). Chronic ischemic lesion burden on FLAIR was median 2.7 cc (0-87.0), with greater extent in those older than 60 years (p=0.03) and those on anti-thrombotics (p=0.01). DWI lesion volume (median 1.45 cc, range 0-71.84) was associated with NIHSS score (p&lt;0.01), antithrombotic use (p=0.01) and time from qualifying event to enrollment (p=0.05). The number of DWI lesions (median 9, range 1-69) correlated (r=0.63, p&lt;0.01) with total volume of acute infarction and was associated with more than 7 days from qualifying event (15.3 vs. 10.7, p&lt;0.01). Chronic FLAIR lesion burden was associated with worse cognitive function or lower MoCA (r=-0.18, p=0.01) at baseline. Importantly, the volume or number of acute DWI lesions was unrelated to MoCA. Conclusions: Cognitive impairment associated with recently symptomatic intracranial atherosclerosis reflects the underlying burden of chronic, not acute, ischemia. Routine MRI, including FLAIR may inform future studies of vascular cognitive impairment in large vessel disease.

Abstract 7: Cerebral Small Vessel Disease Burden and Association With Post-Surgical Rebleeding and Long-Term Outcome in Intracerebral Hemorrhage Patients From the Mistie III Trial

Objectives: To examine the effect of cerebral small vessel disease (CSVD) markers and CSVD burden on hematoma expansion and long-term functional outcomes in large spontaneous intracerebral hemorrhage (ICH) &gt;30 mL. Method: Retrospective analysis of 288 patients from the MISTIE III trial with qualified MRI sequences. MISTIE III evaluated minimally invasive surgery plus alteplase (vs medical management). We identified 6 CSVD markers including lacunes, cerebral microbleeds (CMB), enlarged perivascular space (EPVS), white matter hyperintensities (WMH), cortical superficial siderosis (cSS), and acute DWI positive lesions. First 5 components were reviewed on day 1 MRI and DWI lesions on day 7. The primary outcome was death or major disability at one year (modified Rankin score 4-6). Secondary outcome was post-surgical rebleeding, defined as symptomatic or asymptomatic hemorrhage expansion within 72 hours post last dose of alteplase. Result: Unfavorable outcome occurred in 65.9% at one year. Post-surgical ICH expansion occurred in 34 (22.9%) of 148 surgical patients. Mean time from symptom onset to first MRI was 0.94 (IQR 0.2-6.3) days. Most individual CSVD markers were more common in patients with unfavorable vs. favorable outcome: CMB≧5 (26.1% vs 15.0%, p=0.03), DWI lesions (65.8%vs 49.6%, p=0.006), cSS (16.8% vs 7.1%, p=0.02) severe WMH (67.7% vs 29.9%, p&lt;0.001) severe EPVS in basal ganglia (19.3% vs 10.2%, p=0.047), and lacunes (11.2% vs 6.3%, p=0.21), In the multivariable adjusted model, cumulative CSVD score (one point for each component: CMB≥5, severe EPVS in basal ganglia, lacunes, severe WMH and cSS) was independently associated with unfavorable outcome (OR 0.56, 95%CI 0.41-0.76, p&lt;0.001; AUC 0.855). We did not find a relationship between either independent CSVD markers or CSVD score with post-surgical ICH expansion. Cumulative CSVD score in multivariable analysis adjusted for predictors of hematoma expansion had OR 0.77 (95% CI 0.50-1.19, p=0.77; AUC 0.659). Conclusion: In large volume ICH patients with long-term follow-up, heavy burden of CSVD at ICH onset remains an independent predictor of unfavorable outcome, and particularly severe white matter hyperintensities. CSVD markers did not show association with post-surgical ICH expansion.

Abstract TP64: Ischemic Stroke Lesion Identification in Non-Contrast CT Using Deep Learning

Background: Automatic measurement of the acute stroke lesion volume on DWI and CT-CBF has been used in recent late window trials. Despite non-contrast CT (NCCT) being the most widely used imaging modality in the acute stroke setting, quantification of acute stroke volumes on NCCT has not been employed in trials because of the difficulty outlining territory with very mild Hounsfield unit depression. Deep learning algorithms have been effective at solving many image processing tasks and may outperform human readers given enough training data. The goal of this study was to train and test a deep learning model on NCCT scans with synthetic stroke lesions and to determine the optimal model design. Methods: Training: 20 NCCT scans without acute stroke were combined with 20 DWI lesions using co-registration producing 400 non-contrast scans with lesions. The region of the NCCT that coincided with the DWI lesion was depressed by 2 Hounsfield units to simulate an acute infarct. An independent validation dataset of 100 cases was created in the same way. Two models were used: a standard “Unet” model and a symmetry-aware Unet model. The models were compared in terms of segmentation accuracy in the independent validation dataset. Results: Both the symmetry aware U-net and the standard U-net detected some part of the true lesion in 100% of the cases. The symmetry aware U-net was more sensitive, median [iqr], (45% [27-68] vs 17% [6-54], p&lt;0.00001) but slightly less specific (98% [93-98] vs 99% [94-99], p&lt;0.0008) than the standard U-net. Conclusion: The symmetry aware U-net shows great promise in detection of acute strokes on NCCT; lesions with Hounsfield unit depressions that are barely visible to the eye can be automatically segmented by this model. Additional training data and architectural enhancements are likely to improve the current spatial sensitivity to above 45%.

Abstract TP351: Microembolic Signals Monitoring in Acute Intracerebral Hemorrhage

Background: The etiology of remote DWI lesions in acute intracerebral hemorrhage (ICH) is still unknown. Postulated mechanisms include intracranial or extracranial emboli, small vessel abnormalities and ischemia following acute intracranial hypertension. Our aim is to evaluate the presence of spontaneous microembolic signals (MES) using transcranial Doppler (TCD) in acute ICH patients. Methods: Twenty patients with acute ICH were prospectively enrolled and monitored with TCD for 1 hour on admission days 1, 3 and 7. TCD monitoring was performed using 2MHz probes. Results: Of the 20 patients evaluated, 40% were females and mean age was 55.6±14.1. Eight patients (40%) had dyslipidemia, 15 (75%) hypertension, 5 (20%) diabetes, 2 (10%) ethanol abuse, 6 (30%) smoking and 1 (10%) had prior ischemic stroke. Most frequent location was lobar (9 patients). The mean hematoma volume was 13,5±17,9 ml. Of six patients who underwent MRI, 2 (20%) had remote DWI lesions. Embolic sources were found in 3 patients (1 with atrial fibrillation and 2 with large artery atherosclerosis). Microembolic signals were detected in seven patients (35%). Conclusion: The high occurrence of microemboli in patients admitted with acute ICH indicates a possible embolic mechanism for DWI lesions in these patients.

Abstract WP56: The Hypoperfusion Intensity Ratio is a Poor Discriminator of Infarct Growth in Fully Reperfused Patients

Background and Purpose: The hypoperfusion intensity ratio (HIR) is a perfusion-weighted imaging parameter defined as the ratio of Tmax&gt;10 seconds : Tmax&gt;6 seconds volume and is believed to be reflective of collateral strength and consequently influence infarct growth. We sought to assess the utility of the HIR in predicting infarct growth in patients undergoing thrombectomy at a comprehensive stroke center (CSC). Methods: Consecutive acute ischemic stroke patients transferred to our CSC from 09/2010-11/2018 were identified and included if the following criteria were met: 1)computed tomography perfusion (CTP) imaging enabling assessment of baseline ischemic core volume and HIR 2) follow-up neuroimaging for assessment of final infarct volumes and 3)modified Thrombolysis In Cerebral Infarction scale (mTICI) 2c status or greater post-thrombectomy. Infarct growth rate (IGR) was calculated as the difference between infarct volume on follow-up imaging and the acute DWI lesion volume, divided by time from CTP to reperfusion in hours. Results: 461 patients (median age, 64 [55-75] years, median baseline NIHSS, 16 [12-21]) were eligible for this analysis. HIR poorly correlated with IGR (Spearman’s rho=0.001, p=0.89). An HIR cut-off of 0.5 was not able to discriminate ‘fast progressors’ (IGR&gt;5 mL/hr) (AUC 0.42, sensitivity 40%, specificity 51%), or IGR at thresholds of either 2.5 or 10 mL/hr (AUC 0.44 and 0.49 respectively, with 95% confidence intervals [0.35-0.52] and [0.41-0.57], respectively). Similarly, an HIR of 0.5 only weakly distinguished ‘fast progression’ in patients reperfused beyond 120 min from imaging and patients with early CTP (last known well to CTP&lt;6 hrs) (AUC 0.59, sensitivity 43%, specificity 68% and AUC 0.50, sensitivity 45%, specificity 55%). On multiple regression analysis, HIR was not predictive of infarct growth (regression equation=18.09+8.48x, F=2.46, p=0.11, R 2 =0.13) but was predictive of ‘fast progression’ (OR 0.22, 95% CI [0.09-0.60], p=0.003, pseudo-R 2 =0.16). Conclusions: Though predictive of fast progression, the HIR is a poor discriminator of infarct growth in successfully reperfused thrombectomy patients who undergo perfusion imaging at a CSC, and thus should not be factored into treatment decision-making.