scholarly journals Machine Learning-Based Model for Predicting Incidence and Severity of Acute Ischemic Stroke in Anterior Circulation Large Vessel Occlusion

2021 ◽  
Vol 12 ◽  
Author(s):  
Junzhao Cui ◽  
Jingyi Yang ◽  
Kun Zhang ◽  
Guodong Xu ◽  
Ruijie Zhao ◽  
...  

Objectives: Patients with anterior circulation large vessel occlusion are at high risk of acute ischemic stroke, which could be disabling or fatal. In this study, we applied machine learning to develop and validate two prediction models for acute ischemic stroke (Model 1) and severity of neurological impairment (Model 2), both caused by anterior circulation large vessel occlusion (AC-LVO), based on medical history and neuroimaging data of patients on admission.Methods: A total of 1,100 patients with AC- LVO from the Second Hospital of Hebei Medical University in North China were enrolled, of which 713 patients presented with acute ischemic stroke (AIS) related to AC- LVO and 387 presented with the non-acute ischemic cerebrovascular event. Among patients with the non-acute ischemic cerebrovascular events, 173 with prior stroke or TIA were excluded. Finally, 927 patients with AC-LVO were entered into the derivation cohort. In the external validation cohort, 150 patients with AC-LVO from the Hebei Province People's Hospital, including 99 patients with AIS related to AC- LVO and 51 asymptomatic AC-LVO patients, were retrospectively reviewed. We developed four machine learning models [logistic regression (LR), regularized LR (RLR), support vector machine (SVM), and random forest (RF)], whose performance was internally validated using 5-fold cross-validation. The performance of each machine learning model for the area under the receiver operating characteristic curve (ROC-AUC) was compared and the variables of each algorithm were ranked.Results: In model 1, among the included patients with AC-LVO, 713 (76.9%) and 99 (66%) suffered an acute ischemic stroke in the derivation and external validation cohorts, respectively. The ROC-AUC of LR, RLR and SVM were significantly higher than that of the RF in the external validation cohorts [0.66 (95% CI 0.57–0.74) for LR, 0.66 (95% CI 0.57–0.74) for RLR, 0.55 (95% CI 0.45–0.64) for RF and 0.67 (95% CI 0.58–0.76) for SVM]. In model 2, 254 (53.9%) and 31 (37.8%) patients suffered disabling ischemic stroke in the derivation and external validation cohorts, respectively. There was no difference in AUC among the four machine learning algorithms in the external validation cohorts.Conclusions: Machine learning methods with multiple clinical variables have the ability to predict acute ischemic stroke and the severity of neurological impairment in patients with AC-LVO.

Stroke ◽  
2017 ◽  
Vol 48 (suppl_1) ◽  
Author(s):  
Ryan McTaggart ◽  
Shadi Yaghi ◽  
Daniel C Sacchetti ◽  
Richard Haas ◽  
Shawna Cutting ◽  
...  

Background: There is very limited data on the use of advanced neuroimaging to select patients with acute ischemic stroke and large vessel occlusion for intraarterial therapy beyond 6 hours from onset. Our aim is to report the outcome of patients with acute ischemic stroke and large artery occlusion who presented beyond 6 hours from onset, had favorable MRI imaging profile, and underwent mechanical embolectomy. Methods: This is a single institution retrospective study between December 1st, 2015, and July 30 th , 2016 with acute ischemic stroke and anterior circulation large vessel occlusion (LVO) with ASPECTS of 6 or more and beyond 6 hours from symptoms onset. Favorable imaging profile was defined as 1) DWI lesion volume (as defined as apparent diffusion coefficient < 620 X 10-6 mm2/s) of 70 mL or less AND 2) Penumbra volume (as defined by volume of tissue with Tmax >6 sec) of 15 mL or greater AND 3) A mismatch ratio of 1.8 or more AND 4) Volume of tissue with perfusion lesion with Tmax > 10 sec is less than 100 mL. Good outcome was defined as a 90 day mRS≤2. Results: In the study period, 41 patients met the inclusion criteria; 22 (53.6%) had favorable imaging profile and underwent mechanical embolectomy. The median age was 75 years (59-92), 68.2% were females; the median time from last known normal to groin puncture was 684.5 minutes (range 363-1628) and the median admission NIHSS score was 17.5 (range 4-28). The rate of good outcomes in this series was similar to that in a patient level pooled meta-analysis of the recent endovascular trials (68.2% vs. 46.0%, p=0.07). The rate of good outcome matches that of the EXTEND-IA trial that selected patients using perfusion imaging (68.2% vs. 71.0%, p = 1.00). None of the patients in our cohort had symptomatic intracereberal hemorrhage. Conclusion: Advanced MR imaging may help select patients with acute ischemic stroke and anterior circulation large vessel occlusion for embolectomy beyond the treatment window used in most endovascular trials.


Stroke ◽  
2019 ◽  
Vol 50 (10) ◽  
pp. 2842-2850 ◽  
Author(s):  
Wouter H. Hinsenveld ◽  
Inger R. de Ridder ◽  
Robert J. van Oostenbrugge ◽  
Jan A. Vos ◽  
Adrien E. Groot ◽  
...  

Background and Purpose— Endovascular treatment (EVT) of patients with acute ischemic stroke because of large vessel occlusion involves complicated logistics, which may cause a delay in treatment initiation during off-hours. This might lead to a worse functional outcome. We compared workflow intervals between endovascular treatment–treated patients presenting during off- and on-hours. Methods— We retrospectively analyzed data from the MR CLEAN Registry, a prospective, multicenter, observational study in the Netherlands and included patients with an anterior circulation large vessel occlusion who presented between March 2014 and June 2016. Off-hours were defined as presentation on Monday to Friday between 17:00 and 08:00 hours, weekends (Friday 17:00 to Monday 8:00) and national holidays. Primary end point was first door to groin time. Secondary end points were functional outcome at 90 days (modified Rankin Scale) and workflow time intervals. We stratified for transfer status, adjusted for prognostic factors, and used linear and ordinal regression models. Results— We included 1488 patients of which 936 (62.9%) presented during off-hours. Median first door to groin time was 140 minutes (95% CI, 110–182) during off-hours and 121 minutes (95% CI, 85–157) during on-hours. Adjusted first door to groin time was 14.6 minutes (95% CI, 9.3–20.0) longer during off-hours. Door to needle times for intravenous therapy were slightly longer (3.5 minutes, 95% CI, 0.7–6.3) during off-hours. Groin puncture to reperfusion times did not differ between groups. For transferred patients, the delay within the intervention center was 5.0 minutes (95% CI, 0.5–9.6) longer. There was no significant difference in functional outcome between patients presenting during off- and on-hours (adjusted odds ratio, 0.92; 95% CI, 0.74–1.14). Reperfusion rates and complication rates were similar. Conclusions— Presentation during off-hours is associated with a slight delay in start of endovascular treatment in patients with acute ischemic stroke. This treatment delay did not translate into worse functional outcome or increased complication rates.


Neurology ◽  
2021 ◽  
pp. 10.1212/WNL.0000000000012827
Author(s):  
Adam de Havenon ◽  
Alicia Castonguay ◽  
Raul Nogueira ◽  
Thanh N. Nguyen ◽  
Joey English ◽  
...  

ObjectiveTo determine the impact of endovascular therapy for large vessel occlusion stroke in patients with pre-morbid disability versus those without.MethodsWe performed a post-hoc analysis of the TREVO Stent-Retriever Acute Stroke (TRACK) Registry, which collected data on 634 consecutive stroke patients treated with the Trevo device as first-line EVT at 23 centers in the United States. We included patients with internal carotid or middle cerebral (M1/M2 segment) artery occlusions and the study exposure was patient- or caregiver-reported premorbid modified Rank Scale (mRS) ≥2 (premorbid disability, PD) versus premorbid mRS score 0-1 (no premorbid disability, NPD). The primary outcome was no accumulated disability, defined as no increase in 90-day mRS from the patient’s pre-morbid mRS.ResultsOf the 634 patients in TRACK, 407 patients were included in our cohort, of which 53/407 (13.0%) had PD. The primary outcome of no accumulated disability was achieved in 37.7% (20/53) of patients with PD and 16.7% (59/354) of patients with NPD (p<0.001), while death occurred in 39.6% (21/53) and 14.1% (50/354) (p<0.001), respectively. The adjusted odds ratio of no accumulated disability for PD patients was 5.2 (95% CI 2.4-11.4, p<0.001) compared to patients with NPD. However, the adjusted odds ratio for death in PD patients was 2.90 (95% CI 1.38-6.09, p=0.005).ConclusionsIn this study of anterior circulation acute ischemic stroke patients treated with EVT, we found that premorbid disability was associated with a higher probability of not accumulating further disability compared to patients with no premorbid disability, but also with higher probability of death.Classification of EvidenceThis study provides Class II evidence that in anterior circulation acute ischemic stroke treated with EVT, patients with premorbid disability compared to those without disability were more likely not to accumulate more disability but were more likely to die.



Stroke ◽  
2021 ◽  
Author(s):  
Imad Derraz ◽  
Mohamed Abdelrady ◽  
Nicolas Gaillard ◽  
Raed Ahmed ◽  
Federico Cagnazzo ◽  
...  

Background and Purpose: White matter hyperintensity (WMH), a marker of chronic cerebral small vessel disease, might impact the recruitment of leptomeningeal collaterals. We aimed to assess whether the WMH burden is associated with collateral circulation in patients treated by endovascular thrombectomy for anterior circulation acute ischemic stroke. Methods: Consecutive acute ischemic stroke due to anterior circulation large vessel occlusion and treated with endovascular thrombectomy from January 2015 to December 2017 were included. WMH volumes (periventricular, deep, and total) were assessed by a semiautomated volumetric analysis on fluid-attenuated inversion recovery–magnetic resonance imaging. Collateral status was graded on baseline catheter angiography using the American Society of Interventional and Therapeutic Neuroradiology/Society of Interventional Radiology grading system (good when ≥3). We investigated associations of WMH burden with collateral status. Results: A total of 302 patients were included (mean age, 69.1±19.4 years; women, 55.6%). Poor collaterals were observed in 49.3% of patients. Median total WMH volume was 3.76 cm 3 (interquartile range, 1.09–11.81 cm 3 ). The regression analyses showed no apparent relationship between WMH burden and the collateral status measured at baseline angiography (adjusted odds ratio, 0.987 [95% CI, 0.971–1.003]; P =0.12). Conclusions: WMH burden exhibits no overt association with collaterals in large vessel occlusive stroke.


Stroke ◽  
2017 ◽  
Vol 48 (suppl_1) ◽  
Author(s):  
Ilana M Ruff ◽  
Fan Z Caprio ◽  
Linda Jiang ◽  
Scott J Mendelson ◽  
Christopher T Richards ◽  
...  

Introduction: With randomized data proving the benefit of endovascular therapy for large vessel occlusion (LVO), there is increasing interest in developing a rapid screening tool to predict LVO in acute ischemic stroke (AIS) patients in the emergency department (ED) setting. Methods: We implemented a new LVO screening tool in our ED in March 2016. The LVO score ranged from 0-6 and included one point for each of the following individual components: level of consciousness, gaze deviation, aphasia, dysarthria, facial droop, and arm weakness. LVO was defined as an intracranial occlusion of the internal carotid artery, M1 and M2 segments of the middle cerebral artery, A1 and A2 segments of the anterior cerebral artery, P1 and P2 segments of the posterior cerebral artery, basilar artery, or vertebral arteries. We calculated c-statistics to determine the discrimination of the LVO score for presence of LVO on emergent vascular imaging and identified the optimal cut-off score with maximum sensitivity and specificity. Results: Among 65 consecutive confirmed AIS patients between March and July 2016, 63.1% had a positive LVO screen (score > 0); 97% of the patients underwent emergent CT angiography. Eleven (16.9%) patients had an LVO in the territory of ischemic infarct, 10 of which were in the anterior circulation. The LVO score had a c-statistic of 0.77 (95%CI 0.59-0.96, p< 0.005) for predicting LVO. An optimal cut-off score > 2 was present in 19 patients (29.2%) and was associated with 72.7% sensitivity and 79.6% specificity; positive and negative predictive values were 42.1% and 93.5%, respectively. Conclusions: In a preliminary analysis, a simple LVO screening tool had acceptable discrimination for predicting LVO in AIS patients in the ED. Further validation and refinement of the tool is necessary.


Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Gaurav Thakur ◽  
Ciaran J Powers ◽  
Shahid M Nimjee ◽  
Patrick Youssef ◽  
Sushil Lakhani ◽  
...  

Introduction: A quarter of ischemic stroke patients with initial mild deficits have a poor outcome. We sought to determine the rate of early neurological decline in acute ischemic stroke patients with large vessel occlusion (LVO) who presented with mild deficits. Methods: Among 1022 acute ischemic stroke patients who received intravenous tissue plasminogen activator (IVtPA) admitted to our institution from January 1, 2014 to March 31, 2019, we identified 313 (30.6%) with LVO. We defined anterior circulation LVO as M1, M2, or carotid artery terminus (ICAT). Mild deficits were defined as National Institute of Health Stroke Scale (NIHSS) ≤ 7. Data was abstracted on demographics, neuroimaging, last known well (LKW), time to IVtPA, intra-arterial therapy (IAT) revascularization, Thrombolysis in Cerebral Infarction score (TICI), clinical presentation, and outcome. Early neurologic decline was defined as NIHSS worsening of ≥ 4 points within 24 hours. Results: Among 313 patients with LVO, we identified 94 (30%) who presented with initial low NIHSS (≤ 7) due to anterior circulation LVO. We excluded 13 patients who did not have natural history data (underwent IAT with mild deficits), leaving 81 patient for analysis. The mean age was 65.8 years (range 25 to 93) and 41% were female. IVtPA time from LKW was a mean 2.5 hours (range, 0.8 to 7). LVO sites were as follows: 5 (6%) ICAT, 23 (28%) M1, and 53 (65%) M2 occlusions. Among the 81 patients, 27 (33.3%) had early neurological decline. Patients with decline were significantly older (71.2 vs 63.1 years, p=0.03). Among the 27 patients with decline, the mean change in NIHSS was 10.5 (range, 4 to 22) and 12 patients (44%) underwent rescue IAT resulting in TICI 2B (6) and TICI 3 (6) revascularization. On hospital discharge, patients with decline were less likely to be discharged home (26% vs 65%, p=0.006). Conclusions: Among LVO patients who received IVtPA, 30% present with initial mild deficits. Early neurological decline occurred in one-third of LVO patients with initial mild deficits despite receiving IVtPA, and patients with decline were less likely to be discharged home. Clinicians need to be aware of the natural history of LVO with initial mid deficits, as patients who decline would be eligible for rescue IAT in the expanded 24 hour window.


Sign in / Sign up

Export Citation Format

Share Document