Prediction of delayed cerebral ischemia after cerebral aneurysm rupture using explainable machine learning approach.

BACKGROUND: Aneurysmal subarachnoid hemorrhage results in significant mortality and disability, which is worsened by the development of delayed cerebral ischemia. Tests to identify patients with delayed cerebral ischemia prospectively are of high interest. OBJECTIVE: We created a machine learning system based on clinical variables to predict delayed cerebral ischemia in aneurysmal subarachnoid hemorrhage patients. We also determined which variables have the most impact on delayed cerebral ischemia prediction using SHapley Additive exPlanations method. METHODS: 500 aneurysmal subarachnoid hemorrhage patients were identified and 369 met inclusion criteria: 70 patients developed delayed cerebral ischemia (delayed cerebral ischemia+) and 299 did not (delayed cerebral ischemia-). The algorithm was trained based upon age, sex, hypertension (HTN), diabetes, hyperlipidemia, congestive heart failure, coronary artery disease, smoking history, family history of aneurysm, Fisher Grade, Hunt and Hess score, and external ventricular drain placement. Random Forest was selected for this project, and prediction outcome of the algorithm was delayed cerebral ischemia+. SHapley Additive exPlanations was used to visualize each feature's contribution to the model prediction. RESULTS: The Random Forest machine learning algorithm predicted delayed cerebral ischemia: accuracy 80.65% (95% CI: 72.62-88.68), area under the curve 0.780 (95% CI: 0.696-0.864), sensitivity 12.5% (95% CI: -3.7 to 28.7), specificity 94.81% (95% CI: 89.85-99.77), PPV 33.3% (95% CI: -4.39 to 71.05), and NPV 84.1% (95% CI: 76.38-91.82). SHapley Additive exPlanations value demonstrated Age, external ventricular drain placement, Fisher Grade, and Hunt and Hess score, and HTN had the highest predictive values for delayed cerebral ischemia. Lower age, absence of hypertension, higher Hunt and Hess score, higher Fisher Grade, and external ventricular drain placement increased risk of delayed cerebral ischemia. CONCLUSION: Machine learning models based upon clinical variables predict delayed cerebral ischemia with high specificity and good accuracy.

Benefit of Intravenous Alteplase before Thrombectomy Depends on ASPECTS.

PURPOSE: Baseline variables could be used to guide the administration of additional intravenous alteplase (IVT) before mechanical thrombectomy (MT). The aim of this study was to determine how baseline imaging and demographic parameters modify the effect of IVT on clinical outcomes in patients with ischemic stroke due to large vessel occlusion. METHODS: Multicenter retrospective cohort study of ischemic stroke patients triaged by multimodal-CT undergoing MT treatment after direct admission to an MT-eligible center. Inverse-probability weighting analysis (IPW) was used to assess the treatment effect of IVT adjusted for baseline variables. Multivariable logistic regression analysis with IPW-weighting and interaction terms for IVT was performed to predict functional independence (mRS 0-2 at 90-days). RESULTS: 720 patients were included, of which 366 (51%) received IVT. In IPW, the treatment effect of IVT on outcome (mRS 0-2) distinctively varied according to the ASPECTS subgroup (ASPECTS 9-10: +15%, ASPECTS 6-8: +7%, ASPECTS <6: -11%). In multivariable logistic regression analysis, IVT was independently associated with functional independence (aOR: 1.57, 95% CI: 1.16-2.14, p = 0.003) and the interaction term was significant for ASPECTS and IVT revealing that IVT was only significantly associated with better outcomes in patients with higher ASPECTS. No other significant baseline variable interaction terms were identified. INTERPRETATION: ASPECTS was the only baseline variable that showed a significant interaction with IVT for outcome prediction. Use of IVT prior to MT in patients with an ASPECTS of <6 was not associated with a treatment benefit and should be considered carefully. ANN NEUROL 2022;92:588-595.

Intravenous tPA (Tissue-Type Plasminogen Activator) Correlates With Favorable Venous Outflow Profiles in Acute Ischemic Stroke.

BACKGROUND: Intravenous tPA (tissue-type plasminogen activator) is often administered before endovascular thrombectomy (EVT). Recent studies have questioned whether tPA is necessary given the high rates of arterial recanalization achieved by EVT, but whether tPA impacts venous outflow (VO) is unknown. We investigated whether tPA improves VO profiles on baseline computed tomography (CT) angiography (CTA) images before EVT. METHODS: Retrospective multicenter cohort study of patients with acute ischemic stroke due to large vessel occlusion undergoing EVT triage. Included patients underwent CT, CTA, and CT perfusion before EVT. VO profiles were determined by opacification of the vein of Labbé, sphenoparietal sinus, and superficial middle cerebral vein on CTA as 0, not visible; 1, moderate opacification; and 2, full. Pial arterial collaterals were graded on CTA, and tissue-level collaterals were assessed on CT perfusion using the hypoperfusion intensity ratio. Clinical and demographic data were determined from the electronic medical record. Using multivariable regression analysis, we determined the correlation between tPA administration and favorable VO profiles. RESULTS: Seven hundred seventeen patients met inclusion criteria. Three hundred sixty-five patients received tPA (tPA+), while 352 patients were not treated with tPA (tPA-). Fewer tPA+ patients had atrial fibrillation (n=128 [35%] versus n=156 [44%]; P=0.012) and anticoagulants/antiplatelet treatment before acute ischemic stroke due to large vessel occlusion onset (n=130 [36%] versus n=178 [52%]; P<0.001) compared with tPA- patients. One hundred eighty-five patients (51%) in the tPA+ and 100 patients (28%) in the tPA- group exhibited favorable VO (P<0.001). Multivariable regression analysis showed that tPA administration was a strong independent predictor of favorable VO profiles (OR, 2.6 [95% CI, 1.7-4.0]; P<0.001) after control for favorable pial arterial CTA collaterals, favorable tissue-level collaterals on CT perfusion, age, presentation National Institutes of Health Stroke Scale, antiplatelet/anticoagulant treatment, history of atrial fibrillation and time from symptom onset to imaging. CONCLUSIONS: In patients with acute ischemic stroke due to large vessel occlusion undergoing thrombectomy triage, tPA administration was strongly associated with the presence of favorable VO profiles.

Favourable arterial, tissue-level and venous collaterals correlate with early neurological improvement after successful thrombectomy treatment of acute ischaemic stroke.

BACKGROUND AND PURPOSE: Early neurological improvement (ENI) after thrombectomy is associated with better long-term outcomes in patients with acute ischaemic stroke due to large vessel occlusion (AIS-LVO). Whether cerebral collaterals influence the likelihood of ENI is poorly described. We hypothesised that favourable collateral perfusion at the arterial, tissue-level and venous outflow (VO) levels is associated with ENI after thrombectomy. MATERIALS AND METHODS: Multicentre retrospective study of patients with AIS-LVO treated by thrombectomy. Tissue-level collaterals (TLC) were measured on cerebral perfusion studies by the hypoperfusion intensity ratio. VO and pial arterial collaterals (PAC) were determined by the Cortical Vein Opacification Score and the modified Tan scale on CT angiography, respectively. ENI was defined as improvement of ≥8 points or a National Institutes of Health Stroke Scale score of 0 hour or 1 24 hours after treatment. Multivariable regression analyses were used to determine the association of collateral biomarkers with ENI and good functional outcomes (modified Rankin Scale 0-2). RESULTS: 646 patients met inclusion criteria. Favourable PAC (OR: 1.9, CI 1.2 to 3.1; p=0.01), favourable VO (OR: 3.3, CI 2.1 to 5.1; p<0.001) and successful reperfusion (OR: 3.1, CI 1.7 to 5.8; p<0.001) were associated with ENI, but favourable TLC were not (p=0.431). Good functional outcomes at 90-days were associated with favourable TLC (OR: 2.2, CI 1.4 to 3.6; p=0.001), VO (OR: 5.7, CI 3.5 to 9.3; p<0.001) and ENI (OR: 5.7, CI 3.3 to 9.8; p<0.001), but not PAC status (p=0.647). CONCLUSION: Favourable PAC and VO were associated with ENI after thrombectomy. Favourable TLC predicted longer term functional recovery after thrombectomy, but the impact of TLC on ENI is strongly dependent on vessel reperfusion.

The Cerebral Collateral Cascade: Comprehensive Blood Flow in Ischemic Stroke.

BACKGROUND AND PURPOSE: Robust cerebral collaterals are associated with favorable outcomes in patients with acute ischemic stroke due to large vessel occlusion treated by thrombectomy. However, collateral status assessment mostly relies on single imaging biomarkers and a more comprehensive holistic approach may provide deeper insights into the biology of collateral perfusion on medical imaging. Comprehensive collateralization is defined as blood flow of cerebral arteries through the brain tissue and into draining veins. We hypothesized that a comprehensive analysis of the cerebral collateral cascade (CCC) on an arterial, tissue and venous level would predict clinical and radiological outcomes. MATERIALS AND METHODS: Multicenter retrospective cohort study of acute stroke patients undergoing thrombectomy triage. CCC was determined by quantifying pial arterial collaterals, tissue-level collaterals, and venous outflow. Pial arterial collaterals were determined by CT angiography, tissue-level collaterals were assessed on CT perfusion. Venous outflow was assessed on CT angiography using the cortical vein opacification score. 3 groups were defined: CCC+ (good pial collaterals, tissue-level collaterals, and venous outflow), CCC- (poor pial collaterals, tissue-level collaterals, and venous outflow) and CCCmixed (remainder of patients). Primary outcome was functional independence (modified Rankin Scale: 0-2) at 90-days. Secondary outcome was final infarct volume. RESULTS: 647 patients met inclusion criteria: 176 CCC+, 345 CCC mixed and 126 CCC-. Multivariable ordinal logistic regression showed that CCC+ predicted good functional outcomes (OR=18.9 [95% CI 8-44.5]; p<0.001) compared to CCC- and CCCmixed patients. CCCmixed patients likely had better functional outcomes compared to CCC- patients (OR=2.5 [95% CI 1.2-5.4]; p=0.014). Quantile regression analysis (50th percentile) showed that CCC+ (ß: -78.5, 95% CI -96.0- -61.1; p<0.001) and CCCmixed (ß: -64.0, 95% CI -82.4- -45.6; p<0.001) profiles were associated with considerably lower final infarct volumes compared to CCC- profiles. CONCLUSION: Comprehensive assessment of the collateral blood flow cascade in acute stroke patients is a strong predictor of clinical and radiological outcomes in patients treated by thrombectomy.

Distinct intra-arterial clot localization affects tissue-level collaterals and venous outflow profiles.

BACKGROUND AND PURPOSE: Arterial clot localization affects collateral flow to ischemic brain in patients with acute ischemic stroke due to large vessel occlusion (AIS-LVO). We determined the association between vessel occlusion locations, tissue-level collaterals (TLC), and venous outflow (VO) profiles and their impact on good functional outcomes. METHODS: We conducted a multicenter retrospective cohort study of consecutive AIS-LVO patients who underwent thrombectomy triage. Baseline computed tomographic angiography (CTA) was used to localize vessel occlusion, which was dichotomized into proximal vessel occlusion (PVO; internal carotid artery and proximal first segment of the middle cerebral artery [M1]) and distal vessel occlusion (DVO; distal M1 and M2), and to assess collateral scores. TLC were assessed on computed tomographic perfusion data using the hypoperfusion intensity ratio. VO was determined on baseline CTA by the cortical vein opacification score. Primary outcomes were favorable VO and TLC; secondary outcome was the modified Rankin Scale after 90 days. RESULTS: A total of 649 patients met inclusion criteria. Of these, 376 patients (58%) had a PVO and 273 patients (42%) had a DVO. Multivariate ordinal logistic regression showed that DVO predicted favorable TLC (odds ratio [OR] = 1.77, 95% confidence interval [CI] = 1.24-2.52, p = 0.002) and favorable VO (OR = 7.2, 95% CI = 5.2-11.9, p < 0.001). DVO (OR = 3.4, 95% CI = 2.1-5.6, p < 0.001), favorable VO (OR = 6.4, 95% CI = 3.8-10.6, p < 0.001), and favorable TLC (OR = 3.2, 95% CI = 2-5.3, p < 0.001), but not CTA collaterals (OR = 1.07, 95% CI = 0.60-1.91, p = 0.813), were predictors of good functional outcome. CONCLUSIONS: DVO in AIS-LVO patients correlates with favorable TLC and VO profiles, which are associated with good functional outcome.

Association of Venous Outflow Profiles and Successful Vessel Reperfusion After Thrombectomy.

OBJECTIVE: Robust arterial collaterals are associated with successful reperfusion after thrombectomy treatment of acute ischemic stroke due to large vessel occlusion (AIS-LVO). Excellent venous outflow (VO) reflects excellent tissue perfusion and collateral status in AIS-LVO patients. To determine whether favorable VO profiles assessed on pre-treatment CT angiography (CTA) images correlate with successful vessel reperfusion after thrombectomy in AIS-LVO patients. METHODS: Multicenter retrospective cohort study of consecutive AIS-LVO patients treated by thrombectomy. Baseline CTA was used to assess collateral status (Tan scale) and VO using the cortical vein opacification score (COVES). Favorable VO was defined as COVES ≥3. Primary outcome was excellent vessel reperfusion status (modified Thrombolysis In Cerebral Infarction [TICI] 2c-3). Secondary outcome was good functional outcome defined as 0-2 on the Modified Ranking Scale (mRS) after 90 days. RESULTS: 565 patients met inclusion criteria. Multivariable logistic regression analysis showed that favorable VO (OR= 2.10 [95% CI 1.39-3.16]; p<0.001) was associated with excellent vessel reperfusion during thrombectomy, regardless of good CTA collateral status (OR= 0.87 [95%CI 0.58-1.34]; p=0.48). A favorable VO profile (OR= 8.9 [95%CI 5.3-14.9]; p<0.001) and excellent vessel reperfusion status (OR = 2.7 [95%CI 1.7-4.4]; p<0.001) were independently associated with good functional outcome adjusted for age, sex, glucose, tPA administration, good CTA collateral status and presentation NIHSS. CONCLUSION: A favorable VO profile is associated with reperfusion success and good functional outcomes in patients with AIS-LVO treated by endovascular thrombectomy.

Venous Outflow Profiles Are Linked to Cerebral Edema Formation at Noncontrast Head CT after Treatment in Acute Ischemic Stroke Regardless of Collateral Vessel Status at CT Angiography.

Background Ischemic lesion net water uptake (NWU) at noncontrast head CT enables quantification of cerebral edema in patients with acute ischemic stroke (AIS) due to large vessel occlusion (LVO). Purpose To assess whether favorable venous outflow (VO) profiles at CT angiography are associated with reduced NWU and good functional outcomes in patients with AIS due to LVO. Materials and Methods This multicenter retrospective cohort study evaluated consecutive patients with AIS due to LVO who underwent thrombectomy triage between January 2013 and December 2019. Arterial collateral vessel status (Tan scale) and venous output were measured at CT angiography. Venous outflow was graded with use of the cortical vein opacification score, which quantifies opacification of the vein of Labbé, sphenoparietal sinus, and superficial middle cerebral vein. Favorable VO was regarded as a score of 3-6 and unfavorable VO as a score of 0-2. NWU was determined at follow-up noncontrast CT. Multivariable regression analyses were performed to determine the association between favorable VO profiles and NWU after treatment and good functional outcome (modified Rankin Scale, ≤2). Results A total of 580 patients were included. Of the 580 patients, 231 had favorable VO (104 women; median age, 73 years [interquartile range {IQR}, 62-81 years]) and 349 had unfavorable VO (190 women; median age, 77 years [IQR, 66-84 years]). Compared with patients with unfavorable VO, those with favorable VO exhibited lower baseline National Institutes of Health Stroke Scale score (median, 12.5 [IQR, 7-17] vs 17 [IQR, 13-21]), higher Alberta Stroke Program Early CT Score (median, 9 [IQR, 7-10] vs 7 [IQR, 6-8]), and less NWU after treatment (median, 7% [IQR, 4.6%-11.5%] vs 17.9% [IQR, 12.3%-22.2%]). In a multivariable regression analysis, NWU mean difference between patients with unfavorable VO and those with favorable VO was 6.1% (95% CI: 4.9, 7.3; P < .001) regardless of arterial CT angiography collateral vessel status (b coefficient, 0.72 [95% CI: -0.59, 2.03; P = .28]). Favorable VO (odds ratio [OR]: 4.1 [95% CI: 2.2, 7.7]; P < .001) and reduced NWU after treatment (OR: 0.77 [95% CI: 0.73, 0.83]; P < .001) were independently associated with good functional outcomes. Conclusion Favorable venous outflow (VO) correlated with reduced ischemic net water uptake (NWU) after treatment. Reduced NWU and favorable VO were associated with good functional outcomes regardless of CT angiography arterial collateral vessel status. © RSNA, 2021 Online supplemental material is available for this article.

Favorable Venous Outflow Profiles Correlate With Favorable Tissue-Level Collaterals and Clinical Outcome.

BACKGROUND AND PURPOSE: Patients with acute ischemic stroke due to large vessel occlusion and favorable tissue-level collaterals (TLCs) likely have robust cortical venous outflow (VO). We hypothesized that favorable VO predicts robust TLC and good clinical outcomes. METHODS: Multicenter retrospective cohort study of consecutive acute ischemic stroke due to large vessel occlusion patients who underwent thrombectomy triage. Included patients had interpretable prethrombectomy computed tomography, computed tomography angiography, and cerebral perfusion imaging. TLCs were measured on cerebral perfusion studies using the hypoperfusion intensity ratio (volume ratio of brain tissue with [Tmax >10 s/Tmax >6 s]). VO was determined by opacification of the vein of Labbé, sphenoparietal sinus, and superficial middle cerebral vein on computed tomography angiography as 0, not visible; 1, moderate opacification; and 2, full. Clinical and demographic data were determined from the electronic medical record. Using multivariable regression analyses, we determined the association between VO and (1) favorable TLC status (defined as hypoperfusion intensity ratio ≤0.4) and (2) good functional outcome (modified Rankin Scale score, 0-2). RESULTS: Six hundred forty-nine patients met inclusion criteria. Patients with favorable VO were younger (median age, 72 [interquartile range (IQR), 62-80] versus 77 [IQR, 66-84] years), had a lower baseline National Institutes of Health Stroke Scale (median, 12 [IQR, 7-17] versus 19 [IQR, 13-20]), and had a higher Alberta Stroke Program Early Computed Tomography Score (median, 9 [IQR, 7-10] versus 7 [IQR, 6-9]). Favorable VO strongly predicted favorable TLC (odds ratio, 4.5 [95% CI, 3.1-6.5]; P<0.001) in an adjusted regression analysis. Favorable VO also predicted good clinical outcome (odds ratio, 10 [95% CI, 6.2-16.0]; P<0.001), while controlling for favorable TLC, age, glucose, baseline National Institutes of Health Stroke Scale, and good vessel reperfusion status. CONCLUSIONS: In this selective retrospective cohort study of acute ischemic stroke due to large vessel occlusion patients undergoing thrombectomy triage, favorable VO profiles correlated with favorable TLC and were associated with good functional outcomes after treatment. Future prospective studies should independently validate our findings.

Impact of Clot Shape on Successful M1 Endovascular Reperfusion.

Objectives: The susceptibility-vessel-sign (SVS) allows thrombus visualization, length estimation and composition, and it may impact reperfusion during mechanical thrombectomy (MT). SVS can also describe thrombus shape in the occluded artery: in the straight M1-segment (S-shaped), or in an angulated/traversing a bifurcation segment (A-shaped). We determined whether SVS clot shape influenced reperfusion and outcomes after MT for proximal middle-cerebral-artery (M1) occlusions. Methods: Between May 2015 and March 2018, consecutive patients who underwent MT at one comprehensive stroke center and who had a baseline MRI with a T2* sequence were included. Clinical, procedural and radiographic data, including clot shape on SVS [angulated/bifurcation (A-SVS) vs. straight (S-SVS)] and length were assessed. Primary outcome was successful reperfusion (TICI 2b-3). Secondary outcome were MT complication rates, MT reperfusion time, and clinical outcome at 90-days. Predictors of outcome were assessed with univariate and multivariate analyses. Results: A total of 62 patients were included. 56% (35/62) had an A-SVS. Clots were significantly longer in the A-SVS group (19 mm vs. 8 mm p = 0.0002). Groups were otherwise well-matched with regard to baseline characteristics. There was a significantly lower rate of successful reperfusion in the A-SVS cohort (83%) compared to the S-SVS cohort (96%) in multivariable analysis [OR 0.04 (95% CI, 0.002-0.58), p = 0.02]. There was no significant difference in long term clinical outcome between groups. Conclusion: Clot shape as determined on T2* imaging, in patients presenting with M1 occlusion appears to be a predictor of successful reperfusion after MT. Angulated and bifurcating clots are associated with poorer rates of successful reperfusion.

Perfusion imaging-based tissue-level collaterals predict ischemic lesion net water uptake in patients with acute ischemic stroke and large vessel occlusion.

Ischemic lesion Net Water Uptake (NWU) quantifies cerebral edema formation and likely correlates with the microvascular perfusion status of patients with acute ischemic stroke due to large vessel occlusion (AIS-LVO). We hypothesized that favorable tissue-level collaterals (TLC) predict less NWU and good functional outcomes. We performed a retrospective multicenter analysis of AIS-LVO patients who underwent thrombectomy triage. TLC were measured on cerebral perfusion studies using the hypoperfusion intensity ratio (HIR; volume ratio of brain tissue with [Tmax > 10 sec/Tmax > 6 sec]); favorable TLC were regarded as HIR ≤ 0.4. NWU was determined using a quantitative densitometry approach on follow-up CT. Primary outcome was NWU. Secondary outcome was a good functional outcome (modified Rankin Scale [mRS] 0-2).580 patients met inclusion criteria. Favorable TLC (ß: 4.23, SE: 0.65; p < 0.001) predicted smaller NWU after treatment. Favorable TLC (OR: 2.35, [95% CI: 1.31-4.21]; p < 0.001), and decreased NWU (OR: 0.75, [95% CI: 0.70-0.79]; p < 0.001) predicted good functional outcome, while controlling for age, glucose, CTA collaterals, baseline NIHSS and good vessel reperfusion status.We conclude that favorable TLC predict less ischemic lesion NWU after treatment in AIS-LVO patients. Favorable TLC and decreased NWU were independent predictors of good functional outcome.