Search

OBJECTIVE Machine learning algorithms have shown groundbreaking results in neuroimaging. The authors herein evaluated the performance of a newly developed convolutional neural network (CNN) to detect and analyze intracranial aneurysms (IAs) on CTA. METHODS Consecutive patients with CTA studies between January 2015 and July 2021 at a single center were identified. The ground truth determination of cerebral aneurysm presence or absence was made from the neuroradiology report. The primary outcome was the performance of the CNN in detecting IAs in an external validation set, measured using area under the receiver operating characteristic curve statistics. Secondary outcomes included accuracy for location and size measurement. RESULTS The independent validation imaging data set consisted of 400 patients with CTA studies, median age 40 years (IQR 34 years) and 141 (35.3%) of whom were male; 193 patients (48.3%) had a diagnosis of IA on neuroradiologist evaluation. The median maximum IA diameter was 3.7 mm (IQR 2.5 mm). In the independent validation imaging data set, the CNN performed well with 93.8% sensitivity (95% CI 0.87–0.98), 94.2% specificity (95% CI 0.90–0.97), and a positive predictive value of 88.2% (95% CI 0.80–0.94) in the subgroup with an IA diameter ≥ 4 mm. CONCLUSIONS The described Viz.ai Aneurysm CNN performed well in identifying the presence or absence of IAs in an independent validation imaging set. Further studies are necessary to investigate the impact of the software on detection rates in a real-world setting.

In animal models, inflammation caused by experimental acute pancreatitis (AP) promotes pancreatic carcinogenesis that is preventable by suppressing inflammation. Recent studies noted higher long-term risk of pancreatic ductal adenocarcinoma (PDAC) after AP. In this study, we evaluated whether the long-term PDAC risk after AP was influenced by the etiology of AP, number of recurrences, and if it was because of progression to chronic pancreatitis (CP).This retrospective study used nationwide Veterans Administration database spanning 1999-2015. A 2-year washout period was applied to exclude patients with preexisting AP and PDAC. PDAC risk was estimated in patients with AP without (AP group) and with underlying CP (APCP group) and those with CP alone (CP group) and compared with PDAC risk in patients in a control group, respectively, using cause-specific hazards model.The final cohort comprised 7,147,859 subjects (AP-35,550 and PDAC-16,475). The cumulative PDAC risk 3-10 years after AP was higher than in controls (0.61% vs 0.18%), adjusted hazard ratio (1.7 [1.4-2.0], P0.001). Adjusted hazard ratio was 1.5 in AP group, 2.4 in the CP group, and 3.3 in APCP group. PDAC risk increased with the number of AP episodes. Elevated PDAC risk after AP was not influenced by the etiology of AP (gallstones, smoking, or alcohol).There is a higher PDAC risk 3-10 years after AP irrespective of the etiology of AP, increases with the number of episodes of AP and is additive to higher PDAC risk because of CP.

It is believed that acute pancreatitis (AP), recurrent AP (RAP) and chronic pancreatitis (CP) represent stages of the same disease spectrum. We aimed to identify risk factors, clinical presentation and outcomes in patients with prior RAP who develop CP.We retrospectively reviewed patients with CP who were seen at our Pancreas Center during 2016-2021. We divided them into two groups: with and without RAP (≥2 episodes of AP). We compared demographics, clinical presentation and resource utilization between the two groups.We identified 440 patients with CP, of which 283 (64%) patients had preceding RAP. These patients were younger (55.6 vs 63.1 years), active smokers (36% vs 20%) and had alcohol-related CP (49% vs 25%) compared to those without RAP and CP (p 0.05). More patients with RAP had chronic abdominal pain (89% vs 67.9%), nausea (43.3% vs 27.1%) and exocrine pancreatic insufficiency (65.8% vs 46.5%) (p 0.05). More patients with RAP used opioids (58.4% vs 32.3%) and gabapentinoids (56.6% vs 34.8%) (p 0.05). They also had more ED visits resulting in an opioid prescription (9.68% vs 2%) and more CP flares requiring hospitalization (3.09 vs 0.87) (p 0.05).Young age, smoking and alcohol use are seen in patients with RAP who progress to CP. These patients are highly symptomatic and use more healthcare resources, suggestive of an overall a more course compared to those patients who develop CP without preceding RAP. Early identification and counselling of these patients may slow down progression to CP.

Chronic subdural hematoma (CSDH) is an increasingly prevalent disease in the aging population. Patients with CSDH frequently suffer from concurrent vascular disease or develop secondary thrombotic complications requiring antithrombotic treatment.To determine the safety and impact of early reinitiation of antithrombotics after middle meningeal artery embolization for chronic subdural hematoma.This is a single-institution, retrospective study of patients who underwent middle meningeal artery (MMA) embolizations for CSDH. Patient with or without antithrombotic initiation within 5 days postembolization were compared. Primary outcome was the rate of recurrence within 60 days. Secondary outcomes included rate of reoperation, reduction in CSDH thickness, and midline shift.Fifty-seven patients met inclusion criteria. The median age was 66 years (IQR 58-76) with 21.1% females. Sixty-six embolizations were performed. The median length to follow-up was 20 days (IQR 14-44). Nineteen patients (33.3%) had rapid reinitiation of antithrombotics (5 antiplatelet, 11 anticoagulation, and 3 both). Baseline characteristics between the no antithrombotic (no-AT) and the AT groups were similar. The recurrence rate was higher in the AT group (no-AT vs AT, 9.3 vs 30.4%, P = .03). Mean absolute reduction in CSDH thickness and midline shift was similar between groups. Rate of reoperation did not differ (4.7 vs 8.7%, P = .61).Rapid reinitiation of AT after MMA embolization for CSDH leads to higher rates of recurrence with similar rates of reoperation. Care must be taken when initiating antithrombotics after treatment of CSDH with MMA embolization.

Background: Successful reperfusion is one of the strongest predictors of functional outcomes after mechanical thrombectomy (MT). Despite continuous advancements in MT technology and techniques, reperfusion failure still occurs in ≈15% to 30% of patients with large vessel occlusion strokes undergoing MT. We aim to evaluate the safety and efficacy of rescue intracranial stenting for large vessel occlusion stroke after failed MT. Methods: The SAINT (Stenting and Angioplasty in Neurothrombectomy) Study is a retrospective analysis of prospectively collected data from 14 comprehensive stroke centers through January 2015 to December 2020. Patients were included if they had anterior circulation large vessel occlusion stroke due to intracranial internal carotid artery and middle cerebral artery-M1/M2 segments and failed MT. The cohort was divided into 2 groups: rescue intracranial stenting and failed recanalization (modified Thrombolysis in Cerebral Ischemia score 0–1). Propensity score matching was used to balance the 2 groups. The primary outcome was the shift in the degree of disability as measured by the modified Rankin Scale at 90 days. Secondary outcomes included functional independence (90-day modified Rankin Scale score 0–2). Safety measures included symptomatic intracranial hemorrhage and 90-day mortality. Results: A total of 499 patients were included in the analysis. Compared with the failed reperfusion group, rescue intracranial stenting had a favorable shift in the overall modified Rankin Scale score distribution (acOR, 2.31 [95% CI, 1.61–3.32]; P P P =0.04) at 90 days. Rates of symptomatic intracerebral hemorrhage were comparable across both groups (7.1% versus 10.2%; aOR, 0.99 [95% CI, 0.42–2.34]; P =0.98). The matched cohort analysis demonstrated similar results. Specifically, rescue intracranial stenting (n=107) had a favorable shift in the overall modified Rankin Scale score distribution (acOR, 3.74 [95% CI, 2.16–6.57]; P P P =0.03) at 90 days with similar rates of symptomatic intracerebral hemorrhage (7.5% versus 11.2%; aOR, 0.87 [95% CI, 0.31–2.42]; P =0.79) compared with patients who failed to reperfuse (n=107). There was no heterogeneity of treatment effect across the prespecified subgroups for improvement in functional outcomes. Conclusions: Acute intracranial stenting appears to be a safe and effective rescue strategy in patients with large vessel occlusion stroke who failed MT. Randomized multicenter trials are warranted.

Medically refractory idiopathic intracranial hypertension (IIH) is frequently treated with venous sinus stenosis stenting with high success rates. Patient selection has been driven almost exclusively by identification of supraphysiological venous pressure gradients across stenotic regions based on theoretical assessment of likelihood of response.To explore the possibility of benefit in low venous pressure gradient patients.Using a single-center, prospectively maintained registry of patients with IIH undergoing venous stenting, we defined treatment groups by gradient pressures of ≤4, 5 to 8, and8 mmHg based on the most frequently previously published thresholds for stenting. Baseline demographics, clinical, and neuro-ophthalmological outcomes (including optical coherence tomography and Humphrey visual fields) were compared.Among 53 patients, the mean age was 32 years and 70% female with a mean body mass index was 36 kg/m 2 . Baseline characteristics were similar between groups. The mean change in lumbar puncture opening pressure at 6 months poststenting was similar between the 3 groups (≤4, 5-8, and8 mmHg; 13.4, 12.9, and 12.4 cmH 2 O, P = .47). Papilledema improvement was observed across groups at 6 months (100, 93, and 86, P = .7) as were all clinical symptoms. The mean changes in optical coherence tomography retinal nerve fiber layer (-30, -54, and -104, P = .5) and mean deviation in Humphrey visual fields (60, 64, and 67, P = .5) at 6 weeks were not significantly different.Patients with IH with low venous pressure gradient venous sinus stenosis seem to benefit equally from venous stenting compared with their higher gradient counterparts. Re-evaluation of our restrictive criteria for this potentially vision sparing intervention is warranted. Future prospective confirmatory studies are needed.

Aim: Stent-retriever (SR) thrombectomy has demonstrated superior outcomes in patients with acute ischemic stroke compared with medical management alone, but differences among SRs remain unexplored. We conducted a Systematic Review/Meta-Analysis to compare outcomes between three SRs: EmboTrap ® , Solitaire™, and Trevo ® . Methods: We conducted a PRISMA-compliant Systematic Review among English-language studies published after 2014 in PubMed/MEDLINE that reported SRs in ≥25 patients. Functional and safety outcomes included 90-day modified Rankin scale (mRS 0-2), mortality, symptomatic intracranial hemorrhage (sICH), and embolization to new territory (ENT). Recanalization outcomes included modified thrombolysis in cerebral infarction (mTICI) and first-pass recanalization (FPR). We used a random effects Meta-Analysis to compare outcomes; subgroup and outlier-influencer analysis were performed to explore heterogeneity. Results: Fifty-one articles comprising 9,804 patients were included. EmboTrap had statistically significantly higher rates of mRS 0-2 (57.4%) compared with Trevo (50.0%, p = 0.013) and Solitaire (45.3%, p Conclusion: The results of our Systematic Review/Meta-Analysis suggest that EmboTrap may be associated with significantly improved functional outcomes compared with Solitaire and Trevo. EmboTrap and Trevo may be associated with significantly lower rates of sICH and mortality compared with Solitaire. No significant differences in recanalization and ENT rates were found. These conclusions are tempered by limitations of the analysis including variations in thrombectomy techniques in the field, highlighting the need for multi-arm RCT studies comparing different SR devices to confirm our findings.

Background The relationship between pass number during endovascular stroke therapy (EST) and outcomes in anterior circulation large vessel occlusion (LVO) acute ischemic stroke (AIS) has been well studied. However, the association between EST attempts and outcomes for patients with posterior circulation LVO remains unclear. Methods From our prospectively maintained multicenter registry, we identified consecutive patients with LVO AIS who underwent EST. The number of thrombectomy passes and per‐pass reperfusion grades (thrombolysis in cerebral infarction) were recorded prospectively at the time of procedure. The primary outcome was the association of number of EST passes and functional independence (modified Rankin scale 0–2) at 90 days. Secondary outcomes included likelihood of attaining substantial reperfusion (thrombolysis in cerebral infarction 2b‐3). Results Among 894 patients with LVO AIS who underwent EST, median age was 68 [interquartile range 58–78], 49% were female, and median National Institutes of Health Stroke Scale was 16 [interquartile range 11–21]. Of this cohort, 86 (9.6%) had posterior circulation LVO, including 65 (7%) with basilar artery occlusion and 9 (1%) with vertebral artery occlusion. The diminishing benefit on successful reperfusion of additional passes was similar for anterior and posterior circulation AIS. Among patients with posterior circulation stroke, successful reperfusion with fewer passes was associated with greater likelihood of good functional outcome at 90 days (40.0% versus 4.8% modified Rankin scale 0–2 with 1–2 versus 3+ passes; P = 0.003). The likelihood of good outcome in patients with posterior circulation LVO treated with EST and successful reperfusion in 1–2 passes was comparable to patients with anterior circulation LVO (40.0% versus 40.2%, posterior versus anterior; P = 0.553). Conclusions The number of EST passes has a significant association with functional independence at 90 days for patients with posterior circulation LVO AIS. After 2 passes, there is significantly reduced benefit conferred by additional passes. Overall clinical outcomes in patients with successful reperfusion with 1–2 passes were similar in patients with posterior circulation compared with anterior circulation LVO.

Introduction There are two approaches for treating stroke patients with tandem occlusions: the anterograde approach (AA, extracranial lesions first) and the retrograde approach (RA, intracranial lesion first). Both techniques are associated with favorable functional outcomes. We aimed to compare both techniques for efficacy and safety outcomes in a multicenter study. Methods Patient data were pooled from 17 centers and divided into AA and RA groups. We performed multivariable logistic regressions to evaluate the association between each group with efficacy and safety outcomes. Results 552 patients were included in the study, 270 (48.4%) were treated with the AA, and 288 (51.6%) with the RA. There were no differences between groups for functional outcome (mRS 0–2) at 90 days (aOR = 0.93, 95%CI: 0.58‐1.48, p = 0.75), and successful reperfusion [mTICI >2b] (aOR = 0.83, 95%CI: 0.44‐1.56, p = 0.57). Similarly, we did not observe any differences for safety outcomes related to sICH (OR = 0.57, 95%CI: 0.25‐1.29, p = 0.18), parenchymal hematoma type 2 (OR = 0.61, 95%CI: 0.3‐1.22, p = 0.16), and all‐cause mortality at 90 days (OR = 1.22, 95%CI: 0.66‐2.26, p = 0.52). However, the median puncture‐to‐reperfusion time was higher in AA compared with RA [59 minutes (IQR: 38–92) vs 53.5 minuntes (IQR: 37–87)]. Conclusions The AA and RA approaches for the treatment of tandem occlusions seem to achieve similar efficacy and safety outcomes. These results are consistent with those of a previous multi‐center study. Additionally, and in line with previous research, there is a puncture‐to‐reperfusion time difference between both approaches, which suggests a potential benefit when using the RA. However, further prospective randomized studies are needed to elucidate its benefit in achieving better clinical outcomes.

Introduction The Area Deprivation Index (ADI) is a validated neighborhood‐level measure that utilizes variables such as income, education, and employment to quantify relative socioeconomic disadvantages. Here we explore the impact of disparities on EVT access. Methods From our prospectively maintained multi‐hospital registry, we identified patients with LVO AIS from January 2019‐ June 2020. Patient addresses and zip‐codes were validated using US Postal Service codes and matched to census‐tract level ADI scores that were obtained from Neighborhood Atlas. ADI were categorized into high and low using the median ADI as the cuto!. The primary outcome was utilization of EVT and IV tPA and was determined using multivariable logistic regression and expressed as OR [95% CI]. All p‐values are two‐sided with p < 0.05 defined as statistically significant. All analyses were conducted using RStudio (Version 1.2.5001). Results Among 637 patients with LVO AIS, median age was 68, 46% were female, 53% were white, 27% were black, and 78% identified as Hispanic. Median state ADI was 5 IQR [5]. NIHSS was similar between low/high ADI (mean(SD): 13.3(7.75) vs 13.6(8.62), p‐value 0.69) regions. ADI was significantly associated with race (6.41 vs 4, black vs. white, p‐value 0.03). In the univariable analysis, patients treated with EVT had lower mean ADIs (5.2 vs. 4.6, no EVT vs. EVT, p< 0.02). In multivariable analysis adjusted for age, sex, race, ethnicity and NIHSS, higher ADI was significantly associated with greater rates of IV tPA usage (OR 1.7 [1.01‐ 2.98]) but not EVT usage (OR 0.63 [0.04‐1.0]) Conclusions Patients residing in disadvantaged neighborhoods may have reduced rates of reperfusion therapy, despite comparable acute stroke presentation symptoms. These findings are consistent with prior studies demonstrating poorer health outcomes in these populations.

Introduction Chronic Subdural Hematoma (cSDH) is projected to be the most common neurosurgical disease in the US by the end of the decade. MMA embolization is a promising new treatment; however, its efficacy in patients with complex, septated cSDH remains uncertain. Methods From our prospectively maintained registry of patients with cSDH treated with MMA embolization (with or without concurrent surgical drainage), we identified patients with and without septations. Septations were defined as hyperdense septa between the inner and outer membranes on a lower‐density background. The primary outcome was recurrence of cSDH, which was defined as any radiographic evidence of increase in thickness and/or new acute hemorrhage. Secondary outcomes included reduction in cSDH thickness, midline shift and rate of reoperation. Results Among 84 patients with 100 cSDHs, median age was 70 [IQR 59‐77] with 26.2% females. 35 CSDHs (35%) had membranes identified on imaging. Evacuation with burr holes was performed in 45.2% and craniotomy in 16.7% of the total cohort. Baseline characteristics between the patients with no septations (no SEP) and those with septatations (SEP) were similar except for median age (no SEP vs SEP, 66 vs 74, p = 0.006). Recurrence rate was lower in the SEP group (no SEP vs SEP, 21.5 vs 2.9%, p = 0.017) even when adjusting for clinically relevant factors (OR 0.07, p = 0.017). Despite larger baseline thickness in the SEP groups, the mean absolute reduction in thickness was more pronounced (no SEP vs SEP, ‐4.6 vs ‐8.0 mm, p = 0.016) with similar midline shift change. Rate of reoperation did not differ (6.2 vs 2.9%, p = 0.65). Recurrence free survival was significantly improved in patients with septations even after adjustment for age and evacuation strategy (Figure 1, p = 0.04). Conclusions MMAE in traditionally higher risk septated cSDHs appears to be highly effective with an even larger reduction in volume and lower risk of recurrence than non‐septated hematomas. These findings support the mechanistic theory of MMAE as a devascularization procedure of membrane neovasculature and may aid in improved patient selection.

Introduction Venous sinus stenosis (VSS) stenting in medically refractory Idiopathic Intracranial Hypertension (IIH) patients has emerged as an effective treatment with low rates of failure and recurrence. However, the best treatment strategy following recurrence of symptoms in previously stented patients with initial response is unclear and frequently leads to shunt placement. We investigated the role of re‐stenting in patients with recurrence after prior successful stenting. Methods This is a retrospective review of a prospectively maintained IIH registry. Inclusion criteria included patients with confirmed IIH and angiographically demonstrable VSS who underwent interventions from 2012–2022. Patients were divided into those who underwent a single stenting procedure (Group S) and those who underwent placement of an additional stent due to recurrence of clinical symptoms and angiographic stenosis (Group R). All re‐stenting was performed in adjacent or remote regions of stenosis. Bivariate analyses were utilized to compare clinical outcomes. Results 86 patients were included with a median age of 33 [IQR 26‐39] and 94% were females. Group S (79/86) and Group R (7/86) had similar baseline characteristics. There was similar improvement in post‐stenting opening pressure (22.4±6.7 vs. 25.2±4.6 cm H2O, p = 0.35, Group S vs. Group R), change in opening pressure (11.6±10.4 vs. 5.3±12.2 cm H2O, p = 0.20, Group S vs. Group R) and resolution of tinnitus (91.8 vs. 80%, p = 0.28, Group S vs. Group R). Headache improvement was higher in Group S at 6 weeks, but similar at 6 months (92.8 vs. 60%, p = 0.02 and 88.2 vs. 80%, p = 0.59, Group S vs. Group R). Group S also had higher improvement in visual disturbances at 6 months (93.8 vs. 60%, p = 0.01, Group S vs. Group R), but similar improvement in papilledema (92.2 vs 100%, p = 0.68, Group S vs. Group R). None of the re‐stented patients required VPS rescue. Conclusions VSS re‐stenting in IIH patients with recurrence after initial stenting is feasible with similar efficacy in reducing intracranial pressure and symptoms.

Introduction The recently published CHOICE trial found that intra‐arterial infusion of alteplase improved clinical outcomes in patients undergoing EVT for LVO AIS, an effect that was possibly due to a reduction in reperfusion‐related injury. Here, we examine whether intravenous alteplase, administered after successful endovascular reperfusion, results in improved clinical outcomes. Methods From our prospectively collected multicenter institutional registry drawn from four comprehensive stroke centers in the greater Houston area, we identified cases who underwent EVT with substantial reperfusion (TICI 2b/3) prior to termination of the intravenous alteplase continuous drip. Control patients were then identified as patients who underwent EVT with TICI 2b/3 after completion of intravenous alteplase, matching on age, baseline Modified Rankin Scale (mRS), initial NIH Stroke Scale (NIHSS), and comorbidities. The primary outcome was 90 day disability and was determined by mRS shift analysis, comparing the case and control populations. Results Among 21 cases and 24 matched controls, there were no significant difference between the mean age, gender, race, initial NIHSS, baseline mRS, or ASPECT score. Time from last known well to recanalization was longer in controls (48 min vs 151 min, p< 0.01). Patients who achieved reperfusion prior to termination of the alteplase continuous drip had better 90 day clinical outcomes relative to their presentation mRS when compared to patients who achieved reperfusion after termination of the alteplase. A change in mRS of 1 or less was seen in 71% vs 33% in the continued infusion group vs controls respectively. Conclusions Patients undergoing EVT who continued to be treated with intravenous alteplase following reperfusion had better 90d clinical outcomes relative to those with reperfusion after infusion termination. These findings further support the possible efficacy of thrombolytics following endovascular reperfusion.

Introduction Recent randomized controlled trials (RCT) failed to demonstrate non‐inferiority of skipping IV tPA in patients with planned endovascular therapy (EVT). None of these studies included patients from the US due to regulatory challenges. Given practice patterns vary relative to Asia and Europe, we sought to address this topic using a validated alternative to RCTs, fuzzy regression discontinuity design (RDD). Methods From our prospectively maintained multi‐center registry we identified patients with LVO AIS treated with EVT with and without IV tPA treatment from 1/2018 ‐ 9/2021. We used the time cutoff for IV tPA as our discontinuity and assumed subjects on either side of the cutoff have markedly different probabilities of receiving the treatment but are similar in other relevant characteristics. The primary outcome was good functional outcome defined as 90‐day mRS 0–2 and it was compared between these two populations immediately adjacent to the cutoff using local linear regressions. Results Among 694 patients with LVO AIS who received EVT, median age was 69 [IQR 59‐79], 50%, were female, 44% White, 24% Black, and 14% Hispanic. 51% received IV tPA, with median onset to treatment time of 109 min [IQR 79‐160]. We observed a sharp drop (47%) in the probability of tPA around the cutoff time of 4 hours (allowing 30 minutes for in‐hospital evaluation), while there were no significant differences in other relevant features at the cutoff, validating the underlying RDD assumptions (Figure A). Overall, 33% of patients achieved good functional outcomes and there were no significant differences around the cutoff time (Figure B). In fuzzy RDD, there was no evidence of an association of receiving tPA with good functional outcome with regression discontinuity of only 1.0% (p = 0.98)There were no significant differences in rates of hemorrhage in patients treated with or without IV tPA (22% vs 21%, p = 0.68). Conclusions Our study provides the highest quality US‐based evidence supporting the findings of the outside US trials, demonstrating no benefit of skipping IV tPA in patients with planned EVT.

Introduction Unruptured cerebral aneurysms (UCAs) have a relatively low prevalence of approximately 3%, but detection can prevent devastating consequences of subarachnoid hemorrhage. Here, we assess the performance of a machine‐learning (ML) algorithm to identify UCAs and determine whether routine use of the algorithm would have improved detection rates and patient care. Methods From a prospectively maintained multi‐center registry across 8 certified stroke centers (1 comprehensive, 7 primary), we identified patients who underwent CT angiogram for evaluation of stroke from 3/14/21 – 11/31/21. An FDA‐cleared convolutional deep neural network (Viz ANEURYSM, Viz.ai, Inc.) trained to identify UCAs at least 4mm analyzed the images. Ground truth was provided by independent expert neuroradiology read. The primary outcome was rate of UCAs detected by the ML algorithm but not detected or addressed in the clinical radiology report or clinical notes, which was determined by two independent researchers. Results Among 1191 CT angiogram scans performed during the study period, 49 were flagged by the ML algorithm as possibly demonstrating an UCA, of which 26 cases were confirmed as true positive (PPV 53%).The most common locations included posterior communicating artery (22%), followed by MCA bifurcation (19%). Of these cases, 9 (35%) were not noted in the clinical radiology report or clinical notes, with a median size of 4.2 mm [IQR 3–7.5 mm], and 22 (85%) were not referred for follow up, with median size of 5 mm [IQR 3.7‐11.3 mm]. Of the 22 cases not referred for follow up, 13 (59%) had been noted in the radiology report. 46% (6/13) of the detected but not referred cases had a diameter greater than 10mm. Conclusions UCAs of sizes and intra‐dural locations that may warrant treatment are frequently missed or not followed up in routine clinical care. An ML algorithm that flags studies and notifies clinicians may minimize missed treatment opportunities.

Intro & Background: The Area Deprivation Index (ADI) is a validated neighborhood-level measure that utilizes variables such as income, education, and employment to quantify relative socioeconomic disadvantages. Here we explore the impact of disparities on EVT access. Methods & Materials: From our prospectively maintained multi-hospital registry, we identified patients with LVO AIS from January 2019- June 2020. Patient addresses and zip-codes were validated using US Postal Service codes and matched to census-tract level ADI scores that were obtained from Neighborhood Atlas. ADI were categorized into high and low using the median ADI as the cutoff. The primary outcome was utilization of EVT and IV tPA and was determined using multivariable logistic regression and expressed as OR [95% CI]. All p-values are two-sided with p Results: Among 637 patients with LVO AIS, median age was 68, 46% were female, 53% were white, 27% were black, and 78% identified as Hispanic. Median state ADI was 5 IQR [5]. NIHSS was similar between low/high ADI (mean(SD): 13.3(7.75) vs 13.6(8.62), p-value 0.69) regions. ADI was significantly associated with race (6.41 vs 4, black vs. white, p-value 0.03). In the univariable analysis, patients treated with EVT had lower mean ADIs (5.2 vs. 4.6, no EVT vs. EVT, p Conclusion: Patients residing in disadvantaged neighborhoods may have reduced rates of reperfusion therapy, despite comparable acute stroke presentation symptoms. These findings are consistent with prior studies demonstrating poorer health outcomes in these populations.

Background Current stroke guidelines recommend advanced imaging (computed tomography [CT] perfusion or magnetic resonance imaging) prior to endovascular therapy (EVT) in patients with late presentation of large vessel occlusion. Adherence to guidelines may be constrained by resources or timely access to imaging. We sought to understand the factors which influence late window imaging selection for EVT candidates with large vessel occlusion. Methods We conducted an international survey from January to May 2022. The questions aimed to identify advanced imaging and treatment decisions based on access to imaging, time delays, and simulated patient scenarios. Results There were 3000 invited participants and 1506 respondents, the majority (89.6%) from comprehensive stroke centers in high‐income countries. Neurointerventionalists comprised 31.8% and noninterventionalists 68.2% of respondents. Overall, 70.7% reported routine use of advanced imaging for late EVT selection, and 63.6% reported its usage in every case. There was greater availability of advanced imaging in comprehensive stroke centers versus primary stroke centers (67.0% versus 33.7%; P P Conclusion Current guidelines for imaging late window EVT candidates are inconsistent with imaging decisions by physicians. Most respondents consider an imaging delay of greater than 20 minutes unacceptable. Access to advanced imaging was greater in comprehensive stroke centers and high‐income countries. In the case of limited access most respondents would consider EVT based on CT only.