Trans-radial spinal angiography: A single-center experience

INTRODUCTION

Spinal angiography (SpAn) is the gold standard for diagnosis of spinal dural fistulas and arteriovenous malformations. A complete spinal angiogram necessitates the interrogation of the segmental arteries arising from the aorta at every level as well, the internal iliac; and median sacral arteries at the caudal end; and the cervical vasculature at the cranial end. SpAn has traditionally been performed with transfemoral arterial access and could be challenging. Of late, transradial arterial access has emerged as a popular alternative for endovascular surgical Neuroradiology (ESN) procedures including SpAn. However, there is paucity of the literature regarding transradial access for spinal angiography.

METHODS

After IRB approval, records and imaging were reviewed in a series of patients who underwent SpAn at our institution.

RESULTS

A total of nine spinal angiograms were performed via transradial access in a consecutive series of eight patients between July 2019 and December 2020 at our institution. Eight of these were diagnostic SpAn's, and one patient underwent SpAn with transradial approach for the treatment of a type I spinal dural arteriovenous fistula. No complications occurred during or subsequent to the procedures.

CONCLUSION

SpAn can be successfully and safely accomplished via transradial access. This approach appears to provide a stable method for interventions, as well.

BOBBY balloon guide catheter thrombectomy in large-vessel occlusion stroke: Initial experience

BACKGROUND AND PURPOSE

Nonrandomized studies have found Balloon Guide Catheter (BGC) use to improve technical and functional outcomes in patients undergoing mechanical thrombectomy (MT).

MATERIALS AND METHODS

We performed a retrospective analysis on prospectively collected data of consecutive ischemic stroke patients undergoing MT at our institution (December 2020-October 2021). Interventions where BOBBY BGC (BBGC, MicroVentionTM, Aliso Viejo, CA) was used were identified. Baseline demographics and clinico-radiographic characteristics were retrospectively collected and analysed using descriptive statistics.

RESULTS

A total of 43 patients received BBGC-MT (male: female = 26:17, median age 72 years [IQR 62-82]). The most common occlusion site was the middle cerebral artery (MCA) (60.4%). Over half (51.2%) received intravenous thrombolytics. The BBGC tracked well over tortuous aortic arches (type II 34.8%, type III 16.3%), with median arteriotomy-to-perfusion time of 29 min (IQR 20-46). Thromboaspiration was used as first-line MT technique in 69.7% cases, with 1 (IQR 1-2) median MT pass achieving modified TICI (thrombolysis in cerebral ischemia) scores of 3 and 2b/3 in 74.4% and 95.3% respectively. Our overall first pass effect (FPE, defined as mTICI 3 after firs-pass) and modified FPE (defined as, mTICI 2b/3 after first-pass) rates were 51.1% and 79.1% respectively, with rates of 92.3% and 100% respectively when stentretriever and thromboaspiration were combined. The median reduction in National Institutes of Health Stroke Scale (NIHSS) was 9 (IQR 4-15, p < 0.0001), with a median 90-day modified Rankin Score (mRS) of 1.5 (IQR 0-2).

CONCLUSIONS

BOBBY BGC use resulted in a high first-pass effect rate and may contribute towards improved functional outcomes.

Percutaneous n-butyl cyanoacrylate embolization of cervical metastatic disease via an anterolateral approach

Symptomatic vascular spinal metastases will benefit from pre-operative tumor embolization - percutaneous with or without adjunct endovascular embolization. However, when a transpedicular approach is not feasible, an anterolateral approach may be a viable alternative. The authors report a 57-year-old woman with prior C3-T1 instrumentation who presented with acute cord compression from a pathologic C5 vertebral body fracture related to metastatic renal cell carcinoma. The patient underwent CT-guided direct tumor embolization with 33% n-butyl-2-cyanoacrylate via an anterolateral approach, followed by C5-corpectomy and anterior cervical C4-C6 fusion and plating with minimal blood loss (est. 20 cc) and a stable neurological exam post-operatively. In patients with highly vascular cervical metastatic disease who lack a viable transpedicular approach for preoperative tumor embolization, a CT-guided anterolateral approach is a viable alternative.

Anatomical predictors of need for decompressive craniectomy after stroke using voxel-based lesion symptom mapping

BACKGROUND AND PURPOSE

Malignant cerebral edema (MCE) secondary to ischemic stroke is a highly morbid condition. Decompressive craniectomy (DC) is the only treatment for MCE that has been shown to reduce mortality. We examined whether early infarction and/or hypoperfusion in specific topographic regions was predictive of the need for later DC.

METHODS

A retrospective database of patients evaluated for large vessel occlusion (LVO) stroke at Stanford between 2010 and 2019 was used. Thirty patients with LVO and baseline perfusion MRI who underwent DC were evaluated. Propensity matching based on age, lesion size, and recanalization status was performed on the remaining cohort. Baseline masks of apparent diffusion coefficient (ADC) + Tmax  >6 seconds lesions were generated using automated perfusion software. Voxel-based lesion symptom maping was used to perform logistic regression at each voxel to generate statistical maps of lesion location associated with DC. Hemispheres were combined to increase statistical power.

RESULTS

Sixty patients were analyzed. After adjusting for age, lesion size, and recanalization status as covariates, scattered cortical regions, predominately within the temporal and frontal lobe, were mildly to moderately predictive of the need for DC (z-scores: 2.4-6.74, p < .01).

CONCLUSIONS

Scattered temporal and frontal lobe regions on baseline diffusion and perfusion MRI were found to be mildly to moderately predictive of the need for subsequent DC in patients with LVO stroke.

Interventional compared with medical management of symptomatic carotid web: A systematic review

BACKGROUND

Carotid web (CaW) is non-atheromatous, shelf-like intraluminal projection, generally affecting the posterolateral wall of the proximal internal carotid artery, and associated with embolic stroke, particularly in younger patients without traditional stroke risk factors. Treatment options for symptomatic CaWs include interventional therapy with carotid endarterectomy or carotid stenting versus medical therapy with antiplatelet or anticoagulants. As safety and efficacy of these approaches have been incompletely delineated in small-to-moderate case series, we performed a systematic review of outcomes with interventional and medical management.

METHODS

Systematic literature search was conducted and data analyzed per PRISMA guidelines (Preferred Reporting Items for Systemic Reviews and Meta-Analyses) from January 2000 to October 2021 using the search strategy: "Carotid web" OR "Carotid shelf" OR "Web vessels" OR "Intraluminal web". Patient-level demographics, stroke risk factors, technical procedure details, medical and interventional management strategies were abstracted across 15 series. All data were analyzed using descriptive statistics.

RESULTS

Among a total of symptomatic 282 CaW patients across 14 series, age was 49.5 (44-55.7) years, 61.7% were women, and 76.6% were black. Traditional stroke risk factors were less frequent than the other stroke causes, including hypertension in 28.6%, hyperlipidemia 14.6%, DM 7.0%, and smoking 19.8%. Thrombus adherent to CaW was detected on initial imaging in 16.2%. Among 289 symptomatic CaWs across 15 series, interventional management was pursued in 151 (52.2%), carotid artery stenting in 87, and carotid endarterectomy in 64; medical management was pursued in 138 (47.8%), including antiplatelet therapy in 80.4% and anticoagulants in 11.6%. Interventional and medical patients were similar in baseline characteristics. The reported time from index stroke to carotid revascularization was median 14 days (IQR 9.5-44). In the interventional group, no periprocedural mortality was noted, major periprocedural complications occurred in 1/151 (0.5%), and no recurrent ischemic events were observed over follow-up range of 3-60 months. In the medical group, over a follow-up of 2-55 months, the recurrence cerebral ischemia rate was 26.8%.

CONCLUSION

Cumulative evidence from multiple series suggests that carotid revascularization is a safe and effective option for preventing recurrent ischemic events in patients with symptomatic carotid webs.

Outcomes Among Patients With Reversible Cerebral Vasoconstriction Syndrome: A Nationwide United States Analysis

BACKGROUND AND PURPOSE

Reversible cerebral vasoconstriction syndrome (RCVS) is a well-established cause of stroke, but its demographics and outcomes have not been well delineated.

METHODS

Analysis of the United States Nationwide Inpatient Sample database (2016-2017) to characterize the frequency of hospitalizations for RCVS, demographic features, inpatient mortality, and discharge outcomes.

RESULTS

During the 2-year study period, 2020 patients with RCVS were admitted to Nationwide Inpatient Sample hospitals, representing 0.02 cases per 100 000 national hospitalizations. The mean age at admission was 47.6 years, with 85% under 65 years of age, and 75.5% women. Concomitant neurological diagnoses during hospitalization included ischemic stroke (17.1%), intracerebral hemorrhage (11.0%), subarachnoid hemorrhage (32.7%), seizure disorders (6.7%), and reversible brain edema (13.6%). Overall, 70% of patients were discharged home, 29.7% discharged to a rehabilitation facility or nursing home and 0.3% died before discharge. Patient features independently associated with the poor outcome of discharge to another facility or death were advanced age (odds ratio [OR], 1.04 [95% CI, 1.03-1.04]), being a woman (OR, 2.45 [1.82-3.34]), intracerebral hemorrhage (OR, 2.91 [1.96-4.31]), ischemic stroke (OR, 5.72 [4.32-7.58]), seizure disorders (OR, 2.61 [1.70-4.00]), reversible brain edema (OR, 6.26 [4.41-8.89]), atrial fibrillation (OR, 2.97 [1.83-4.81]), and chronic kidney disease (OR, 3.43 [2.19-5.36]).

CONCLUSIONS

Projected to the entire US population, >1000 patients with RCVS are hospitalized each year, with the majority being middle-aged women, and about 300 required at least some rehabilitation or nursing home care after discharge. RCVS-related inpatient mortality is rare.

Abstract P431: White-Matter Disease Burden and Cerebral Microbleeds After Extracorporeal Membrane Oxygenation Support

Introduction: Cerebral microbleeds (CMBs) have been increasingly reported in patients receiving extracorporeal membrane oxygenation (ECMO) support. Both, CMBs and cerebral white-matter disease (WMD) are thought to be a result of microvascular lipo-hyalinosis; as commonly seen in an aging brain. The pathogenesis of microbleeds after ECMO, has not yet been definitively established. We sought to examine the relationship between cerebral WMD burden and microbleeds after ECMO-support at a single tertiary referral academic hospital with high volumes of ECMO patients.

Methods: All patients receiving venovenous (vv) and venoarterial (va) ECMO between January 2013 to January 2018 were retrospectively examined. The distribution of white matter hyperintensities was quantified using the Walhund age-related white matter changes (ARWMC) scale and correlated with the presence of cerebral microbleeds on brain MRI studies with Susceptibility Weighted Imaging (SWI) performed shortly after cessation of ECMO.

Results: A total of 307 ECMO patients were reviewed, among whom 44 patients (vv:va= 13:31; male:female= 29:15) received at-least one MRI-brain study with SWI sequences after ECMO decannulation. The median duration of ECMO support was 4 days (range 1-25 days), with median duration from decannulation-to-MRI being 11.5 days (range 3-724 days). Microbleeds were present in 77.3% (n=34) patients, with 38.2% (n=13), 14.7% (n=5), and 47.1% (n=16) having mild (<10), moderate (10-30) and severe (>30) CMBs respectively. The median Walhund ARWMC score was 1 (range 0-22). Age and sex adjusted Walhund scores were not found predictive of CMB presence (p=0.578 [0.90-1.19]). Of 44 patients with at-least one post-ECMO MRI, 10 patients had a follow-up MRI scan at a median duration of 4 months (range 0.22-55 months); all of whom had unchanged CMB burden. At 10-month median clinical follow-up (range 0-76 months), none of the 34 patients with CMBs on initial MRI-study had presented with an intracranial hemorrhage.

Conclusion: Development of CMBs after ECMO support is independent of microvascular lipo-hyalinosis, as estimated from cerebral white-matter disease burden. This requires further study in a larger sample-size.

Abstract P612: Iron Deficiency Anemia and Ischemic Stroke in Patients With Pulmonary Arteriovenous Fistulas

Introduction: Pulmonary-arteriovenous-fistulas (PAVFs) are pathologic right-to-left shunts resulting in paradoxical embolism causing acute-ischemic-stroke (AIS). Recent single-center studies have identified that in patients with AIS associated with PAVF (AIS-P), traditional stroke risk-factors are not prominent and instead stroke-risk is associated with low serum iron. Single-centre studies have the risk of introducing a selection bias, while multicentre trials are challenging since PAVF still remains a rare and under-recognised entity. We thus seek epidemiological validation of such stroke predictors in patients with PAVF.

Methods: We conducted a retrospective analysis of all AIS-admissions within the Nationwide-Inpatient-Sample (NIS) database (2005-2014). Baseline characteristics were compared across AIS populations [AIS-P (with PAVF) and R(routine)-AIS (without PAVF)]. We also compared morbidity, mortality and management trends of AIS in patients with and without PAVF.

Results: Of 4,271,910 patients admitted with AIS, 822 (0.02%) were diagnosed with a PAVF. Over this decade the prevalence of PAVF per million AIS-admissions, rose from 197 to 368 (P trend =0.026). Patients with PAVF were younger with a median age (IQR) of 57.5 (42.2 -70.4) years vs. 72.5 (60.8-82.1) years (p<0.001); but had comparable age-adjusted inpatient morbidity (χ 2 p=0.71) and all-cause mortality (χ 2 p=0.26). On multivariate analyses, the odds ratios (95% confidence-interval) favouring PAVF as the cause for AIS were 9.0 (6.79-11.94) for hypoxemia, 4.64 (3.84-5.60) for patent-foramen-ovale, 4.52 (3.42-5.97) for pulmonary hypertension, 4.07(2.23-7.44) for epistaxis, and 2.12 (1.60-2.82) for iron deficiency anaemia [all p-values <0.001].

Conclusion: Pulmonary-arteriovenous-fistula related AIS represents a significantly younger demographic, which suffers inpatient morbidity and mortality comparable to routine ischemic-stroke. They carry a unique set of stroke-risk markers, including treatable conditions such as iron deficiency anemia. Further studies are needed to examine a causal role for such markers on ischemic-stroke risk in this cohort.

Cauda Equina Syndrome in Neurosarcoidosis

Neurosarcoidosis (NS) is a mimicker of many infectious, neoplastic, and inflammatory diseases. It most commonly involves the cranial nerves followed by meninges, ventricles, hypothalamic-pituitary axis, spinal cord, and brainstem/cerebellum. While NS myelopathy has been increasingly recognized, pathophysiological/prognostic and management principles in NS-mediated cauda equina (CE) and conus medullaris (CM) syndromes, which constitute a small and rare minority of this subset, remain elusive. We present the case of a 49 -year-old Hispanic man who developed a peripheral facial palsy and primary hypogonadism within a span of 12 months and eventually got diagnosed with NS after he presented with CE syndrome. We also performed an extensive literature review, with a discussion on the underlying pathophysiology and current management recommendations for NS-mediated CE/CM syndrome. CE/CM syndromes in a middle-aged man should prompt the consideration of NS as a possible differential diagnosis. While steroid responsive, the majority of NS-CE/CM patients are left with residual neurodeficits with quick relapses when steroids are tapered, making the case for early institution of immunosuppressive therapies.

Identifying Gaps and Missed Opportunities for Intravenous Thrombolytic Treatment of Inpatient Stroke

Background: Inpatient stroke-codes (ISC) have traditionally seen low treatment rates with IV-thrombolytic (IVT). The purpose of this study was to identify the predictors of true stroke, prevalent IVT-treatment gap and study the factors associated with such missed treatment opportunities (MTO).

Methods: A retrospective chart review identified ISC from March 2017 to March 2018. Clinical, radiographic and demographic data were collected. Primary analysis was performed between stroke vs. non-stroke diagnoses. Dichotomous variables were analyzed using Chi-Square test of proportions and continuous variables with Wilcoxon-Ranked-Sum test. Significant factors were then tested in a multivariate logistic regression model for independence.

Results: From 211 ISC, 36% (n = 76) had an acute stroke. Hemorrhagic stroke (HS) was present in 5.7% (n = 12). Of the remaining 199, 44% (n = 87) were IVT-eligible but only 3.4% (n = 3) were treated. Of the remaining 84 IVT-eligible-but-untreated patients, 69(82.1%) were mimics, while 15 (17.9%) had an ischemic stroke (IS), constituting a MTO of 1 in 6 IVT-eligible patients, with National Institutes of Health Stroke Scale (NIHSS) ≤4 being the commonest deterrent. Independent predictors of stroke were ejection fraction (EF) <30% (p = 0.030, OR = 3.06), post-operative status (p = 0.001, OR = 3.71), visual field-cut (p = 0.008, OR = 3.70), and facial droop (p = 0.010, OR = 2.59).

Conclusion: In our study, one in three ISC were true strokes. IVT treatment rates were low with a MTO of 1 in 6 IVT-eligible patients. The most common reason for not treating was NIHSS ≤4. Knowing predictors of true stroke and the common barriers to IVT treatment can help narrow this treatment gap.

Abstract TP72: Should Hypoperfusion Intensity Ratio Influence Patient Selection for Mechanical Thrombectomy?

Background: Current data utilizes clinical-radiographic mismatch (core and mismatch ratio) as patient selection criteria for mechanical thrombectomy in LVO. High HIR (Hypoperfusion Intensity Ratio) is known to correlate with core size, infarct growth and outcome though influence on patient selection has not been yet determined.

Hypothesis: Patients with High HIR and malignant profile (Tmax >10s greater than 50% of penumbra) indicative of fast growing infarct may influence final clinical outcome irrespective of reperfusion.

Methods: We retrospectively identified all AIS patients with LVO who underwent CTP imaging between January and June2018 within 24 hours from symptom onset. Demographics, CTP imaging variables, reperfusion status and outcomes (discharge NIHSS and mRS) were analyzed. HIR was dichotomized by proportion of greater and less than 0.5 into malignant vs favorable profile. Association with core size, infarct growth velocity, reperfusion (defined as TICI 2b or 3) and impact on outcomes was analyzed using Wilcoxon Ranked Sum tests for the (skewed) continuous and ordinal variables; chi-square test of proportion were used for categorical variables. The independent contribution of HIR and reperfusion predicting the major outcomes was assessed with logistic regression.

Results: A total of 67 patients with LVO were identified with a median age of 78 (IQR 62-87), NIHSS of 16 (IQR 11-21) and time from last seen normal to CT 404 minutes (IQR 113-734). Five patients were excluded due inadequate CTP data. Patients with high HIR (n=23) had a higher core size (median 39 cc; IQR 16-73) compared to 0 cc (IQR 0-12) than patients with low HIR (n=39; median 0; IQR 0-12) (p=<0.001) and faster Infarct growth rate 14.8 cc/hr (IQR 3.6-29.7) vs. 0 cc/hr (IQR 0-1.12) (p=<0.001). After adjusting for reperfusion, median discharge NIHSS was not significantly different (p=0.22) in groups with low vs high HIR, however in-hospital mortality differed (p=0.02).

Conclusion: Higher HIR and malignant profile is associated with larger index core size and faster growth rate. However, the influence of this profile on clinical outcomes after recanalization is yet to be established. Ongoing studies evaluating the utility of HIR on patient selection for thrombectomy are needed.

Abstract WP374: 30 Days Unplanned Readmission (30-DR) in Cerebral Vein Thrombosis: US National Retrospective Analysis

Introduction: Thrombosis of the cerebral veins and/or dural sinuses (CVST) results in 0.5% to 1% of all strokes. CVST has a lower mortality rate compared to arterial stroke; however it predisposes to certain complications, which lead to frequent hospital readmissions.

Objective: To study the readmission rate, reasons for readmissions and predictors of intracranial hemorrhage (ICH).

Method: Using the Nationwide Readmission Database (NRD) from 2013 -2014 patients with a primary discharge diagnosis of CVST (ICD 9 CM code: 437.6, 325, and 671.5) were identified. Their post-hospitalization course was tracked using the variable “NRD_visitlink,” and the time between re-admissions was calculated by subtracting the variable “NRD_DaysToEvent”. SAS 9.4 was used for data analysis with categorical and continuous variables tested using the Rao Scott Chi-square test and Student’s t-test, respectively. Multivariate logistic regression models were used for categorical dependent variable.

Result: A total of 3,611 (weighted) Index discharges of CVST were identified. After excluding elective readmissions, 11.32%, 17.04% and 21.81% were the observed readmission rates at 30, 90 and 180 day-follow-up periods respectively. The common causes of 30-day readmission (30-DR) were complications of the puerperium (8.76%), ICH (6.13%), migraine (4.23%), ischemic stroke (2.86%), sepsis (2.30%) and hypertensive complications (1.83%). Epilepsy related readmissions were observed only in the 90-day (1.95%) and 180-day (2.51%) follow up periods. The mean 30-DR cost was lower vs. Index admission (Mean ± SE $ 12,439±1139 vs. $ 16314±652; P value <0.0001) while the length of stay (LOS) during readmission was also lower compared to Index LOS (Mean ± SE 5.0±0.4 vs. 6.4±0.2 ;< 0.0001). After adjusting age and gender, multivariate predictors of increase risk of ICH readmission were hypertension (OR 4.79 CL 1.38-16.59; P=0.0068), primary hypercoagulable state (OR 3.34, CL 1.12-9.99; P=0.0306), meningitis (OR 17.77 CL 2.51-125.96, P=0.0041) and chronic kidney disease (OR 3.80CL 1.04-13.85, P=0.0434).

Conclusion: Puerperal complications, ICH, epilepsy and sepsis are the most common complications prompting re-hospitalization in patients diagnosed with CVST.

Abstract WP424: Identifying Gaps and Missed Opportunities for IV-Thrombolytic Treatment of Inpatient Stroke

Background: Inpatient stroke-codes have traditionally seen low treatment rates with IV-thrombolytic due do an abundance of stroke mimics and contraindications for treatment. However, data regarding missed treatment opportunities are lacking.

Objective: To identify the treatment gap and factors associated with missed treatment opportunities for inpatient strokes.

Methods: A retrospective chart review was performed identifying all inpatient stroke codes from March 2017 to March 2018. Clinical, radiographic and demographic patient data were collected. Primary analysis was performed between stroke vs. non-stroke final diagnosis. Dichotomous variables were analyzed using Chi-Square test of proportions and continuous variables with Wilcoxon Ranked Sum test. Significant factors were then tested in a multivariate logistic regression model for independence.

Results: Out of 211 codes, 36% (n=76) of patients had an acute stroke. An intracranial hemorrhage was present in 5.7% (n=12). Of the remaining 199 codes, 44% (n=87) were IV tPA eligible but only 3.4% (n=3) were treated. All treated patients had a confirmed stroke. Of the remaining 84 tPA-eligible patients, 44% (n=37) had >1 reason to hold treatment. The most frequent reason cited was NIHSS ≤4 in 40% (n=62), suspected metabolic encephalopathy in 23% (n=47) and abnormal blood pressure or blood sugar in 6.3% (n=13). From the eligible-but-untreated cohort, 82% (n=69) were stroke mimics while 18% (n=15) had strokes, constituting a missed treatment opportunity of 1 in 6 patients. Independent predictors of stroke were ejection fraction <30% (p=0.030, OR 3.06), post-operative status (p=0.003, OR 3.00), visual field cut (p=0.048, OR 2.61) and facial droop (p=0.048, OR 2.07). Sedative use (p=0.013, OR 0.33) and seizure at onset (p=0.015, OR 0.07) were inversely predictive of stroke.

Conclusion: In the inpatient setting, 1 in 3 codes are true strokes and treatment rates with IV thrombolytic are low with a missed treatment opportunity of 1 in every 6 eligible patients. The most frequent reasons for not treating include NIHSS ≤4 and suspicion of metabolic encephalopathy. Identifying patients with NIHSS ≤4 and knowing predictors of true stroke can help narrow this treatment gap.

Abstract TP409: Prevalence and Predictors of Stroke in Patients With Short-Term Mechanical Circulatory Support Devices: A Single-Center Study

Introduction: Short-term mechanical circulatory support devices (ST-MCS) include intra-aortic balloon-pump (IABP), extracorporeal membrane oxygenation (ECMO) and the CentriMag® and Impella® ventricular assist systems (VAS). Despite an exponential increase in their use, data regarding stroke prevalence and predictors are lacking.

Objective: To identify the prevalence and predictors of stroke in ST-MCS.

Methods: Data was collected prospectively into a database from January 2016 to June 2018 and retrospectively extracted and analyzed. Primary analysis was performed between acute-stroke vs. non-stroke diagnoses. Dichotomous and continuous variables were analyzed using Chi-Square test of proportions and Wilcoxon Ranked Sum test respectively.

Results: Out of 203 ST-MCS-patients [IABP 31.5% (n=64), Impella 24.6% (n=50), ECMO 31% (n=63) and CentriMag®VAS 12.8% (n=26)], 8.4% (n=17) had an acute stroke. Among them 6.4% (n=13) had ischemic stroke and 1% (n=2) had hemorrhagic stroke, with 1% (n=2) having both. CentriMag®VAS had a higher stroke rate than the other devices (23.1% [n=6] vs. 6.2% [n=11]; p=0.011, OR 4.53). Predictors of stroke in all patients were central cannulation (p=0.044, OR 3.08), duration >4 days (p=0.025, OR 3.21) and use of another ST-MCS device before primary device (p=0.043, OR 1.45). Flow-rate (p=0.86) and catheter size (p=0.15) did not predict stroke. Only 1 patient was eligible for thrombolytic therapy and received IV tPA, with the most common reasons to hold treatment being unknown last-seen-normal (n=8), coagulopathy (n=2) and established infarct on head CT (n=2). A large vessel occlusion was present in 20% (n=3), but none underwent a mechanical thrombectomy due of established infarction. All hemorrhagic strokes and 47% (n=7) ischemic strokes led to withdrawal of care.

Conclusion: About 1 in 12 patients placed on a ST-MCS device may have an acute stroke, but this can be as high as 1 in 5 with the use of the CentriMag®VAS. Factors such as central cannulation, duration >4 days and use of another ST-MCS device before the primary device may be predictive of acute stroke in these patients. Further research in the identification of such predictors, in conjunction with early symptom recognition could help improve treatment rates.

Abstract WP262: The Migraine-PFO-Stroke Triangle: Observations From the National Inpatient Sample

Introduction: Patients with migraine are known to have a higher risk of stroke. Migrainous-infarction, and more recently, paradoxical embolism via a patent foramen ovale (PFO) are proposed to be the causative pathophysiology. The prevalence of PFO in migraineurs with cryptogenic stroke has been shown to be as high as 80%.

Objective: To determine from the Nationwide Inpatient Sample (NIS), the prevelance of acute ischemic stroke (AIS) in migraineurs with PFO who have had a PFO-closure procedure and compare them to those without a PFO-closure.

Methods: Using the NIS from 2005-2014, patients carrying a diagnosis of migraine with aura (ICD 9: 346.0, 346.3, 346.5, and 346.6), migraine without aura (ICD9: 346.1, 346.2, 346.4, 346.7, 346.8, and 346.9) and PFO (ICD 9: 745.61, 745.5) were identified. From them, those who had a PFO closure (ICD 9: 35.52, 35.51, 35.71, 35.61) were identified. SAS 9.4 was used for data analysis with categorical and continuous variables tested using the Chi-Square test of proportions and Student’s t-test, respectively. Cox proportional hazard regression was used to adjust for confounders.

Results: A total of 45,00,000 patients with migraine were identified, among which 3% (n=132,349) had migraine with aura. PFO detection rates in migraineurs with and without aura were 2% (n=2620) and 0.5% (n=23,225) respectively. In migraineurs with aura, PFO closure was performed in 10.5% (n=274) among which 12% (n=33) had an AIS as compared to 27.5% (n=647) AIS within the non-closure group (p<0.095, OR 0.48). However in migraineurs without aura, PFO closure was performed in 17% (n=3845) among which 7% (n=266) had an AIS as compared to 26.2% (n=5046) AIS within the non-closure group (p<0.0001, OR 0.29).

Conclusion: In a nationally representative sample, PFO closure was associated with a lower prevalence of AIS in migraineurs without aura. A trend was observed favoring PFO-closure in migraineurs with aura, but lacked statistical significance which maybe secondary to the low PFO-detection rate observed in our sample. These findings may support the hypothesis of a paradoxical embolic phenomenon as being the predominant underlining pathophysiology for acute ischemic stroke in patients with migraine.

Abstract WP291: Utilization of Novel Imaging Algorithms for Patient Triage at Primary Stroke Centers in a Stroke System of Care

Background: The 24 hour time window for endovascular therapy (EVT) in ischemic strokes due to large vessel occlusions (LVO) is a significant advance. Although many stroke systems of care have adopted a paradigm of transferring all potential EVT patients to the hub center for further evaluation, we developed an approach using advanced screening and imaging at spoke hospitals to improve transfer triage.

Hypothesis: Utilizing a novel algorithm combining clinical and imaging (CTA plus CT Perfusion “RAPID”) criteria at the transferring facilities (spokes), more appropriate patient transfer decision to the CSC (hub) can be made.

Methods: We developed and implemented a clinical and imaging screening algorithm for our 3 PSCs equipped with CTA and CTP-“RAPID capabilities”. Patients at these 3 hospitals with NIHSS> 6 or fulfilling Stroke VAN (Hemiparesis and Visual field cut, aphasia or neglect) criteria and presenting between 4.5 to 24 hours from last seen normal undergo CTA plus CTP, and a decision for transfer if LVO plus core ≤50cc is met. Data including diagnosis, clinical and radiographic features, transfer status and final diagnosis of 102 patients from January 2018 to June 2018 were analyzed.

Results: The mean age was 69 (SD 1.7) years and mean NIHSS was 9 (SD 0.73). NIHSS> 6 or Stroke VAN criteria was met in 78% (n=80). Of all screened patients, 26% (n=27) were found to have LVO. The mean CTP core size for patients transferred (n=20) to the hub was lower than those who remained at the spoke (n=7) but not significant due to small sample size (8.6 cc SD 2.5 vs. 38 cc SD 22 p =0.31). Five additional patients without LVO were transferred for higher level of care. Of all transferred LVO patients, 35% (n=7) were treated with EVT; 65% (n=13) did not meet treatment criteria. Overall, only 7 of 102 patients (7%) were treated with EVT. Of the 75% (n=77) non-transferred patients, 55% (n=41) were stroke mimics based on final diagnosis..

Conclusion: Using our current algorithm, 60% of screened patients were ischemic strokes and 26% had LVO’s, but only 7% qualified for EVT. Using advanced clinical and imaging paradigms, 75% of screened patients did not require transfer. While challenges with implementation and patient selection remain, triage at the spoke facilities is feasible and effective.

Abstract TP577: Prevalence and Associations of Ischemic Stroke in Pulmonary ArterioVenous Fistulas

Introduction: Pulmonary arteriovenous fistula (PAVF) is an abnormal intrapulmonary right-to-left shunt. About 1 in 4 patients with untreated PAVF may suffer a paradoxical embolic acute ischemic stroke (AIS) by 65 years of age. Current expert recommendations are to treat all CT scan visible PAVF, even if asymptomatic. The American Heart Association Stroke Guidelines recommend using antiplatelet agents for secondary prophylaxis of AIS related to PAVF.

Objective: To determine from the Nationwide Inpatient Sample (NIS), the prevelance and management trends of AIS in patients diagnosed with PAVF.

Methods: Using the NIS from 2001-2014, patients carrying a secondary diagnosis of PAVF (ICD-9 CM code: 417.0) and a subsequent hospitalization with AIS were identified. SAS 9.4 was used for data analysis with categorical and continuous variables tested using the Chi-Square test of proportions and Student’s t-test, respectively. Cox proportional hazard regression was used to adjust for confounders.

Results: A total of 4430 (weighted) patients (>18 years) were diagnosed with PAVF at a mean age of 55.6 years (SE±0.7). About 7% (n=312, p<0.0001) patients with PAVF were admitted with AIS at a mean age of 58.5 years (SE ±2.5). From 2001 to 2014, there was a two-fold increase in the prevalence of PAVF (204 vs. 455) as well as AIS related admissions (5% vs. 12%). On univariate analysis, epistaxis (p= 0.0184 OR 5.7), migraine (p= 0.0007, OR 5.7), patent foramen ovale (PFO) (p=0.0002, OR 3.2) and cannabis use (p= 0.009, OR 5.8) were significantly associated with AIS. Migraine (p=0.0062, OR 4.17) and PFO (p=0.0004, OR 3.68) retained significance on multivariate analysis. None of the ischemic strokes received reperfusion therapy. Only about 8% (n=24) were on long-term antiplatelet therapy. Approximately 3.2% (n= 141) of all patients with PAVF had a percutaneous embolization of their fistula.

Conclusion: In a nationally representative sample, the prevalence of AIS in patients with PAVF has roughly doubled from 2001 to 2014. Utilization of reperfusion therapy and secondary stroke prophylaxis with antiplatelet agents is low. Consensus guidelines and randomized control trials are need to establish the role of fistula embolization in these patients.

Wernicke's Encephalopathy

Wernicke’s encephalopathy (WE) is a neurologic emergency that requires immediate attention to prevent permanent neurological morbidity and mortality. It presents with confusion, ophthalmoplegia and gait ataxia which together comprise its classic triad. Thiamine deficiency related to alcohol abuse remains the primary culprit; non-alcoholic WE, however, can have an atypical clinical presentation and is often missed. Thus, although the diagnosis of WE remains primarily clinical, neuroimaging plays an important role, especially in the diagnosis of non-alcoholic WE. Here, we present a case of non-alcoholic WE with an atypical clinical presentation but typical magnetic resonance imaging (MRI) findings in a woman with a history of non-bariatric gastrointestinal surgery. Thiamine replacement therapy rapidly reversed her neurologic symptoms and MRI findings.