Infarct Evolution in Patients with Anterior Circulation Large-Vessel Occlusion Randomized to IV Alteplase and Endovascular Treatment versus Endovascular Treatment Alone

BACKGROUND AND PURPOSE

Infarct evolution after endovascular treatment varies widely among patients with stroke and may be affected by baseline characteristics and procedural outcomes. Moreover, IV alteplase and endovascular treatment may influence the relationship of these factors to infarct evolution. We aimed to assess whether the infarct evolution between baseline and follow-up imaging was different for patients who received IVT and EVT versus EVT alone.

MATERIALS AND METHODS

We included patients from the Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Netherlands (MR CLEAN)-NO IV trial with baseline CTP and follow-up imaging. Follow-up infarct volume was segmented on 24-hour or 1-week follow-up DWI or NCCT. Infarct evolution was defined as the follow-up lesion volume: CTP core volume. Substantial infarct growth was defined as an increase in follow-up infarct volume of >10 mL. We assessed whether infarct evolution was different for patients with IV alteplase and endovascular treatment versus endovascular treatment alone and evaluated the association of baseline characteristics and procedural outcomes with infarct evolution using multivariable regression.

RESULTS

From 228 patients with CTP results available, 145 patients had follow-up imaging and were included in our analysis. For patients with IV alteplase and endovascular treatment versus endovascular treatment alone, the baseline median CTP core volume was 17 (interquartile range = 4-35) mL versus 11 (interquartile range = 6-24) mL. The median follow-up infarct volume was 13 (interquartile range, 4-48) mL versus 17 (interquartile range = 4-50) mL. Collateral status and occlusion location were negatively associated with substantial infarct growth in patients with and without IV alteplase before endovascular treatment.

CONCLUSIONS

No statistically significant difference in infarct evolution was found in directly admitted patients who received IV alteplase and endovascular treatment within 4.5 hours of symptom onset versus patients who underwent endovascular treatment alone. Collateral status and occlusion location may be useful predictors of infarct evolution prognosis in patients eligible for IV alteplase who underwent endovascular treatment.

Association between Dural AVFs and Cerebral Venous Thrombosis

BACKGROUND AND PURPOSE

Recent reports suggest an association between dural AVFs and cerebral venous thrombosis. We aimed to investigate the specific temporal and anatomic association between the 2 conditions.

MATERIALS AND METHODS

A consecutive cohort of adult patients with dural AVFs was seen at Amsterdam University Medical Centers (2007-2020). An experienced neuroradiologist re-evaluated the presence and imaging characteristics of dural AVFs and cerebral venous thrombosis on all available imaging. The temporal (previous/concurrent/subsequent) and anatomic (same/adjacent/unrelated venous sinus or vein) association between dural AVFs and cerebral venous thrombosis was determined.

RESULTS

Among 178 patients with dural AVFs, the mean age was 58.3 (SD, 13.2) years and 85 (48%) were women. Of 55 patients (31%) with cerebral venous thrombosis, 34 (62%) were women. Four patients (7%) had cerebral venous thrombosis before the development of a dural AVF, 33 (60%) had cerebral venous thrombosis at the time of dural AVF diagnosis (concurrent), and 18 (33%) developed cerebral venous thrombosis during follow-up after conservative treatment. The incidence rate of cerebral venous thrombosis after a dural AVF was 79 per 1000 person-years (95% CI, 50-124). In 45 (82%) patients with dural AVFs and cerebral venous thrombosis, the thrombosis was located in the same venous sinus as the dural AVF, whereas in 8 (15%) patients, thrombosis occurred in a venous sinus adjacent to the dural AVF.

CONCLUSIONS

One-third of patients with a dural AVF in this study were diagnosed with cerebral venous thrombosis. In almost two-thirds of patients, cerebral venous thrombosis was diagnosed prior to or concurrent with the dural AVF. In 97% of patients, there was an anatomic association between the dural AVF and cerebral venous thrombosis. These data support the hypothesis of a bidirectional association between the 2 diseases.

Unsupervised Deep Learning for Stroke Lesion Segmentation on Follow-up CT Based on Generative Adversarial Networks

BACKGROUND AND PURPOSE

Supervised deep learning is the state-of-the-art method for stroke lesion segmentation on NCCT. Supervised methods require manual lesion annotations for model development, while unsupervised deep learning methods such as generative adversarial networks do not. The aim of this study was to develop and evaluate a generative adversarial network to segment infarct and hemorrhagic stroke lesions on follow-up NCCT scans.

MATERIALS AND METHODS

Training data consisted of 820 patients with baseline and follow-up NCCT from 3 Dutch acute ischemic stroke trials. A generative adversarial network was optimized to transform a follow-up scan with a lesion to a generated baseline scan without a lesion by generating a difference map that was subtracted from the follow-up scan. The generated difference map was used to automatically extract lesion segmentations. Segmentation of primary hemorrhagic lesions, hemorrhagic transformation of ischemic stroke, and 24-hour and 1-week follow-up infarct lesions were evaluated relative to expert annotations with the Dice similarity coefficient, Bland-Altman analysis, and intraclass correlation coefficient.

RESULTS

The median Dice similarity coefficient was 0.31 (interquartile range, 0.08-0.59) and 0.59 (interquartile range, 0.29-0.74) for the 24-hour and 1-week infarct lesions, respectively. A much lower Dice similarity coefficient was measured for hemorrhagic transformation (median, 0.02; interquartile range, 0-0.14) and primary hemorrhage lesions (median, 0.08; interquartile range, 0.01-0.35). Predicted lesion volume and the intraclass correlation coefficient were good for the 24-hour (bias, 3 mL; limits of agreement, -64-59 mL; intraclass correlation coefficient, 0.83; 95% CI, 0.78-0.88) and excellent for the 1-week (bias, -4 m; limits of agreement,-66-58 mL; intraclass correlation coefficient, 0.90; 95% CI, 0.83-0.93) follow-up infarct lesions.

CONCLUSIONS

An unsupervised generative adversarial network can be used to obtain automated infarct lesion segmentations with a moderate Dice similarity coefficient and good volumetric correspondence.

Clinical Outcome After Endovascular Treatment in Patients With Active Cancer and Ischemic Stroke: A MR CLEAN Registry Substudy

OBJECTIVE

To explore clinical and safety outcomes of patients with acute ischemic stroke (AIS) and active cancer after endovascular treatment (EVT).

METHODS

Using data from the MR CLEAN Registry, we compared patients with active cancer (defined as cancer diagnosed within 12 months prior to stroke, metastatic disease, or current cancer treatment) to patients without cancer. Outcomes were 90-day modified Rankin Scale (mRS) score, mortality, successful reperfusion (eTICI scores≥2b), symptomatic intracranial hemorrhage (sICH), and recurrent stroke. Subgroup analyses were performed in patients with a pre-stroke mRS score of 0 or 1 and according to treatment setting (curative or palliative). Analyses were adjusted for prognostic variables.

RESULTS

Of 2583 patients who underwent EVT, 124 (4.8%) had active cancer. They more often had pre-stroke disability (mRS≥2: 34.1% vs. 16.6%). The treatment setting was palliative in 25.3% of the patients. There was a shift towards worse functional outcome at 90 days in patients with active cancer (adjusted common OR 2.2, 95% CI 1.5-3.2). At 90 days, patients with active cancer were less often independent (mRS 0-2: 22.6% vs. 42.0%, aOR 0.5, 95% CI 0.3-0.8), and more often dead (52.2% vs. 26.5%, aOR 3.2, 95% CI 2.1-4.9). Successful reperfusion (67.8% vs. 60.5%, aOR 1.4, 95% CI 1.0-2.1) and sICH rates (6.5% vs. 5.9%, aOR 1.1, 95 %CI 0.5-2.3) did not differ. Recurrent stroke within 90 days was more common in patients with active cancer (4.0% vs. 1.3%, aOR 3.1, 95% CI 1.2-8.1). The sensitivity analysis of patients with a pre-stroke mRS of 0 or 1 showed that patients with active cancer still had a worse outcome at 90 days (acOR 1.9, 95% CI 1.2-3.0). Patients with active cancer in a palliative treatment setting regained functional independence less often compared to patients in a curative setting (18.2% vs. 32.1%) and mortality was also higher (81.8% vs. 39.3%).

CONCLUSIONS

Despite similar technical success, patients with active cancer had significantly worse outcomes after EVT for AIS. Moreover, they had an increased risk of recurrent stroke. Nevertheless, about a quarter of the patients regained functional independence and the risk of other complications, most notably sICH, was not increased.

CLASSIFICATION OF EVIDENCE

This study provides Class I evidence that patients with active cancer undergoing EVT for AIS have worse functional outcomes at 90 days compared to those without active cancer.

Presentation outside office hours does not negatively influence treatment times for reperfusion therapy for acute ischemic stroke

BACKGROUND

Treatment outside office hours has been associated with increased workflow times for intravenous thrombolysis (IVT) in acute ischemic stroke (AIS). Limited data suggest that this "off-hours effect" also exists for endovascular treatment (EVT). We investigated this phenomenon in a well-organized acute stroke care region in the Netherlands.

METHODS

Retrospective, observational cohort study of consecutive patients with AIS who received reperfusion therapy in the Greater Amsterdam Area, consisting of 14 primary stroke centers and 1 comprehensive stroke center (IVT: 2009-2015, EVT: 2014-2017). Office hours were defined as presentation during weekdays between 8 AM and 5 PM, excluding National Festive days. Primary outcome was door-to-treatment time (door-to-needle [DNT] for IVT, door-to-groin [DGT] for EVT). For DGT, we used the door time of the first hospital. Other outcomes were in-hospital mortality, modified Rankin Scale (mRS) score at 90 days and symptomatic intracranial hemorrhage (sICH). We performed multivariable linear and logistic regression analyses and used multiple imputation to account for missing values.

RESULTS

In total, 59% (2450/4161) and 61% (239/395) of patients treated with IVT and EVT, respectively, presented outside office hours. Median DNT was minimally longer outside office hours (32 vs. 30 min, p = 0.024, adjusted difference 2.5 min, 95% CI 0.7-4.2). Presentation outside office hours was not associated with a longer DGT (median 130 min for both groups, adjusted difference 7.0 min, 95% CI - 4.2 to 18.1). Clinical outcome and sICH rate also did not differ.

CONCLUSION

Presentation outside office hours did not lead to clinically relevant treatment delays for reperfusion therapy in patients with AIS.

Data-efficient deep learning of radiological image data for outcome prediction after endovascular treatment of patients with acute ischemic stroke

Treatment selection is becoming increasingly more important in acute ischemic stroke patient care. Clinical variables and radiological image biomarkers (old age, pre-stroke mRS, NIHSS, occlusion location, ASPECTS, among others) have an important role in treatment selection and prognosis. Radiological biomarkers require expert annotation and are subject to inter-observer variability. Recently, Deep Learning has been introduced to reproduce these radiological image biomarkers. Instead of reproducing these biomarkers, in this work, we investigated Deep Learning techniques for building models to directly predict good reperfusion after endovascular treatment (EVT) and good functional outcome using CT angiography images. These models do not require image annotation and are fast to compute. We compare the Deep Learning models to Machine Learning models using traditional radiological image biomarkers. We explored Residual Neural Network (ResNet) architectures, adapted them with Structured Receptive Fields (RFNN) and auto-encoders (AE) for network weight initialization. We further included model visualization techniques to provide insight into the network's decision-making process. We applied the methods on the MR CLEAN Registry dataset with 1301 patients. The Deep Learning models outperformed the models using traditional radiological image biomarkers in three out of four cross-validation folds for functional outcome (average AUC of 0.71) and for all folds for reperfusion (average AUC of 0.65). Model visualization showed that the arteries were relevant features for functional outcome prediction. The best results were obtained for the ResNet models with RFNN. Auto-encoder initialization often improved the results. We concluded that, in our dataset, automated image analysis with Deep Learning methods outperforms radiological image biomarkers for stroke outcome prediction and has the potential to improve treatment selection.

Flow Patterns in Carotid Webs: A Patient-Based Computational Fluid Dynamics Study

BACKGROUND AND PURPOSE

Carotid webs are increasingly recognized as an important cause of (recurrent) ischemic stroke in patients without other cardiovascular risk factors. Hemodynamic flow patterns induced by these lesions might be associated with thrombus formation. The aim of our study was to evaluate flow patterns of carotid webs using computational fluid dynamics.

MATERIALS AND METHODS

Patients with a carotid web in the Multicenter Randomized Clinical Trial of Endovascular Treatment of Acute Ischemic Stroke in the Netherlands (MR CLEAN) were selected for hemodynamic evaluation with computational fluid dynamics models based on lumen segmentations obtained from CT angiography scans. Hemodynamic parameters, including the area of recirculation zone, time-averaged wall shear stress, transverse wall shear stress, and the oscillatory shear index, were assessed and compared with the contralateral carotid bifurcation.

RESULTS

In our study, 9 patients were evaluated. Distal to the carotid webs, recirculation zones were significantly larger compared with the contralateral bifurcation (63 versus 43 mm², P = .02). In the recirculation zones of the carotid webs and the contralateral carotid bifurcation, time-averaged wall shear stress values were comparable (both: median, 0.27 Pa; P = .30), while transverse wall shear stress and oscillatory shear index values were significantly higher in the recirculation zone of carotid webs (median, 0.25 versus 0.21 Pa; P = .02 and 0.39 versus 0.30 Pa; P = .04). At the minimal lumen area, simulations showed a significantly higher time-averaged wall shear stress in the web compared with the contralateral bifurcation (median, 0.58 versus 0.45 Pa; P = .01).

CONCLUSIONS

Carotid webs are associated with increased recirculation zones and regional increased wall shear stress metrics that are associated with disturbed flow. These findings suggest that a carotid web might stimulate thrombus formation, which increases the risk of acute ischemic stroke.

Accuracy of CT Angiography for Differentiating Pseudo-Occlusion from True Occlusion or High-Grade Stenosis of the Extracranial ICA in Acute Ischemic Stroke: A Retrospective MR CLEAN Substudy

BACKGROUND AND PURPOSE

The absence of opacification on CTA in the extracranial ICA in acute ischemic stroke may be caused by atherosclerotic occlusion, dissection, or pseudo-occlusion. The latter is explained by sluggish or stagnant flow in a patent artery caused by a distal intracranial occlusion. This study aimed to explore the accuracy of CTA for differentiating pseudo-occlusion from true occlusion of the extracranial ICA.

MATERIALS AND METHODS

All patients from the Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Netherlands (MR CLEAN) with an apparent ICA occlusion on CTA and available DSA images were included. Two independent observers classified CTA images as atherosclerotic cause (occlusion/high-grade stenosis), dissection, or suspected pseudo-occlusion. Pseudo-occlusion was suspected if CTA showed a gradual contrast decline located above the level of the carotid bulb, especially in the presence of an occluded intracranial ICA bifurcation (T-occlusion). DSA images, classified into the same 3 categories, were used as the criterion standard.

RESULTS

In 108 of 476 patients (23%), CTA showed an apparent extracranial carotid occlusion. DSA was available in 46 of these, showing an atherosclerotic cause in 13 (28%), dissection in 16 (35%), and pseudo-occlusion in 17 (37%). The sensitivity for detecting pseudo-occlusion on CTA was 82% (95% CI, 57-96) for both observers; specificity was 76% (95% CI, 56-90) and 86% (95% CI, 68-96) for observers 1 and 2, respectively. The κ value for interobserver agreement was .77, indicating substantial agreement. T-occlusions were more frequent in pseudo- than true occlusions (82% versus 21%, P < .001).

CONCLUSIONS

On CTA, extracranial ICA pseudo-occlusions can be differentiated from true carotid occlusions.

Thrombectomy in posterior circulation stroke through persistent primitive trigeminal artery: A case report

We describe a case of intra-arterial treatment (IAT) of acute posterior circulation occlusion in a patient with a persistent primitive trigeminal artery (PPTA). The patient presented with an acute left sided hemiparesis and loss of consciousness (Glasgow coma score of 5). Computed tomography angiography showed an acute occlusion of the right internal carotid artery (ICA), the PPTA, distal basilar artery (BA), right posterior cerebral artery (PCA), and right superior cerebellar artery (SCA). Stent-retriever assisted thrombectomy was not considered possible through the hypoplastic proximal BA. After passage of the proximal ICA occlusion, the right PCA and SCA were recanalized through the PPTA, with a single thrombectomy procedure. Ten days after intervention patient was discharged scoring optimal EMV with only a mild facial and left hand paresis remaining. PPTA is a persistent embryological carotid-basilar connection. Knowledge of existing (embryonic) variants in neurovascular anatomy is essential when planning and performing acute neurointerventional procedures.

Advanced magnetic resonance imaging of the brain in patients treated with TNF-alpha blocking agents

OBJECTIVE

Neurological symptoms have been reported in patients treated with anti-TNF-alpha. In a pilot study we evaluated the effect of anti-TNF-alpha on cerebral parenchyma using advanced Magnetic Resonance (MR) techniques.

METHODS

Seven patients with a systemic inflammatory disease (5 rheumatoid arthritis, 2 psoriatic arthritis) had Magnetization Transfer Imaging, Diffusion Weighted Imaging (DWI) and Magnetic Resonance Spectroscopy (MRS) of the brain before and after administration of anti-TNF-alpha. Four patients were neuropsychologically evaluated.

RESULTS

After treatment with TNF-alpha blocking agents the Magnetization Transfer Ratio histogram Peak-heights (MTR-Pht) of the white and gray matter decreased (p < 0.01 and p < 0.05 respectively). The Apparent Diffusion Coefficient for the white and gray matter and the metabolite ratios in the centrum semiovale did not significantly change after therapy. Neuropsychological assessment showed no difference before and after anti-TNF-alpha.

CONCLUSION

The decrease of the MTR-Pht after anti-TNF-alpha therapy suggests loss of parenchyma integrity; however, these changes could not be attributed to inflammation or demyelination based on our complementary DWI and MRS data. The decrease of the MTR-Pht did not result in decreased cognitive function.