Search

Background and purpose Previous studies suggest that mechanisms and outcomes in patients with COVID-19-associated stroke differ from those in patients with non-COVID-19-associated strokes, but there is limited comparative evidence focusing on these populations. The aim of this study, therefore, was to determine if a significant association exists between COVID-19 status with revascularization and functional outcomes following thrombectomy for large vessel occlusion (LVO), after adjustment for potential confounding factors. Methods A cross-sectional, international multicenter retrospective study was conducted in consecutively admitted COVID-19 patients with concomitant acute LVO, compared to a control group without COVID-19. Data collected included age, gender, comorbidities, clinical characteristics, details of the involved vessels, procedural technique, and various outcomes. A multivariable-adjusted analysis was conducted. Results In this cohort of 697 patients with acute LVO, 302 had COVID-19 while 395 patients did not. There was a significant difference (p < 0.001) in the mean age (in years) and gender of patients, with younger patients and more males in the COVID-19 group. In terms of favorable revascularization (modified Thrombolysis in Cerebral Infarction [mTICI] grade 3), COVID-19 was associated with lower odds of complete revascularization (odds ratio 0.33, 95% confidence interval [CI] 0.23-0.48; p < 0.001), which persisted on multivariable modeling with adjustment for other predictors (adjusted odds ratio 0.30, 95% CI 0.12-0.77; p = 0.012). Moreover, endovascular complications, in-hospital mortality, and length of hospital stay were significantly higher among COVID-19 patients (p < 0.001). Conclusion COVID-19 was an independent predictor of incomplete revascularization and poor functional outcome in patients with stroke due to LVO. Furthermore, COVID-19 patients with LVO were more often younger and had higher morbidity/mortality rates.

Background: Radiomics features (RFs) extracted from CT images have the potential to provide valuable biological and structural information about ischemic stroke blood clots, which may inform mechanical thrombectomy (MT) outcome. Hypothesis: We aimed to identify RFs that predict MT outcome and interpret these findings using paired histological data. Methods: We extracted 293 RFs from co-registered non-contrast CT (nCCT) and CT angiography (CTA) images. RFs predictive of revascularization outcome (defined by first pass effect [FPE], which is near complete removal of occlusive clot in one pass of MT) were selected. We then trained and cross-validated (5-fold) a balanced logistic regression model to assess the RFs’ ability in FPE prediction. In a subset of cases, we performed digital histology on clots retrieved during MT. We computed 227 engineered clot histomic features from whole slide histopathology that represented clot structure and texture. Results: Following univariate feature selection and multi-variate feature ranking, we identified 6 RFs significantly associated with FPE (Figure, top rows). Continuity in lower pixel intensities, scattered higher intensities, and intensities with abrupt changes in texture were associated with successful revascularization outcome. The multi-variate model trained using the 6 RFs could predict FPE well, with AUC=0.832±0.031 and accuracy=0.760±0.059 in training and AUC=0.787±0.115 and accuracy=0.787±0.127 in cross-validation. Each of the 6 RFs were related to clot component organization in terms of red blood cell (RBC) and fibrin/platelet (FP) distribution. Clots with more diversity of components and with varying sizes of RBC and FP regions in the clot section were positively associated with RFs predictive of FPE. Conclusion: Clot RFs are potential structurally- and biologically-interpretable candidate biomarkers for accurate FPE prediction.

Background: Intracranial aneurysm (IA) rupture is associated with high morbidity and mortality rates. Identifying unstable, growing IAs can enable critical treatment before rupture. Hypothesis: RNA expression profiles in whole blood are different among cases of growing vs. stable IA, and thus may be candidate risk assessment biomarkers. Methods: We performed RNA sequencing on 69 blood samples from IA patients. To assess if IAs would grow or not, we calculated the published Predicted Aneurysm Trajectory (PAT) score. We dichotomized samples into stable and growing IAs by median PAT. Combat-seq removed potential batch effects and counts were normalized to TPMs. After randomly dividing the dataset into a training and testing cohort, differentially expressed (protein-coding) genes (DEGs) in the training dataset were defined as those with an FDR-corrected F-test ( q -value)0) in at least 50% of the samples. A Gaussian Naïve Bayes (GNB) model using the DEGs was trained and independently tested. To understand the pathobiology associated with the DEGs we performed gene ontology term enrichment analysis. Results: The PAT score predicted 34 IAs to be stable and 35 to be growing. Between these groups, we identified 36 DEGs that discriminated growing from stable IAs (Figure, A and B). Our GNB model using the DEGs achieved 86% accuracy in training (Figure, C). In the testing cohort, the DEGs also delineated growing from stable IAs (Figure, D), and the model had 70% accuracy (Figure, E). DEGs with increased expression in growing IAs were enriched for biological process (BP) terms including neutrophil activation and degranulation, while DEGs decreased in growing IAs were enriched for BP that included NK-cell immunity and cytotoxicity (Figure, F). Conclusion: Circulating transcriptomes can delineate IA growth based on the PAT score. Our biomarker model must be validated in larger cohorts and evaluated in longitudinal datasets.

Background: Intra-saccular hemodynamics play a key role in the pathobiological remodeling of the intracranial aneurysm (IA) wall. Evaluating the relationship between hemodynamic forces at the aneurysm wall and wall presentation can help us understand the complex processes of dysregulated vascular remodeling that occurs during IA natural history. Hypothesis: We hypothesized that greater hemodynamic insult, i.e., high wall shear stress (WSS) and high wall shear stress divergence (WSSD), results in endothelial damage and loss of internal elastic lamina, which presents as thin, translucent regions on interoperative imaging. Methods: From digital subtraction angiography images of 15 patients with IAs, aneurysm and surrounding vessels were segmented. Computational fluid dynamics (CFD) simulations were performed using generic boundary conditions for all the patients. Aneurysm walls were identified from 2D intraoperative images of IAs taken during clipping, and wall regions were classified as thick regions (white), normal (purple-crimson), and thin/translucent (red), using a semi-supervised machine learning algorithm. 2D wall classifications were mapped onto the 3D geometries for statistical analysis of differences between average hemodynamic properties across wall regions. Results: On an average, more than 30% of IA sac was visible on the 2D intra-operative image. Of the n =15 IAs, n =8, n =5 and n =2 showed pre-dominantly thick, thin, and normal wall types respectively. Average WSS and WSSD were significantly higher at thin regions of IAs as compared to both normal and thick regions. Thin wall regions also tended to co-locate with significantly lower RRT as compared to the thick and normal regions. Conclusion: Thin wall regions observed on intra-operative images of IAs are associated with greater hemodynamic insult (higher WSS and WSSD) and higher flow velocity (lower RRT).

Background: Determination of ischemic stroke etiology is essential for guiding post-operative stroke treatment and secondary prevention. Hypothesis: Canonical correlation analysis (CCA) of paired clot histomic and transcriptomic features can identify etiology-specific white blood cell (WBC) image-omic correlates. Methods: We analyzed paired H&E histology and transcriptome data from 31 clots (large artery atherosclerosis, LAA=5; cardio embolism, CE=18; stroke of other cause, OTH=3; cryptogenic, CR=5). A published computational pipeline for WBC segmentation and characterization was used to quantify 46 image features (e.g., textures representative of heterochromatin-euchromatin ratio) from >100,000 WBC instances. Features were used to classify WBCs into 1 of 10 validated classes and calculate clot class frequency distributions. SparseCCA was used to test associations among image features and expression signatures. Gene ontology (GO) enrichment analysis was used to identify biological process (BP) ontologies that were associated with positively correlated genes for each etiology. Results: Canonical covariates (e.g., histology canonical variable [CV] 1 and transcriptome CV1) were strongly and significantly correlated ( R >0.9, p Conclusion: We found significant structure-expression covariates representing biologically-interpretable traits that are different among stroke etiologies.

Background: Paired analysis of ischemic stroke clot histology and pre-treatment CT imaging could identify digital indicators of treatment outcome or stroke etiology. Yet, differences in scale limit our ability to interpret the biology of the clot from pre-treatment images. Hypothesis: Radiomic analysis of resected clots on microCT has the potential to identify important, biologically-interpretable features that are correlated among pre-treatment imaging modalities (CTA, nCCT) and post-treatment histopathology. Methods: We performed multimodal CT imaging of 10 stroke clots retrieved by mechanical thrombectomy (example in top of Figure). Clots were manually segmented from co-registered, pre-treatment CTA and nCCT images. For the same cases, portions of resected clots were stained with iodine, imaged on a ScanCo microCT 100 (4.9 μm resolution), and segmented using adaptive intensity thresholding. Radiomic texture features ( n =93 for each modality, 279 total) were extracted from the images using PyRadiomics. Pearson correlation analysis was used to test associations between CTA (or nCCT) and microCT radiomics. Statistical significance was considered at R ≥0.7 and q Results: 18 features were significantly correlated between CTA and microCT, and 5 features were significantly correlated between nCCT and microCT (bottom of Figure). Cluster Prominence ( R =0.96, q R =0.92, q R =0.90, q R =0.90, q Conclusion: Scale-invariant radiomic features can help bridge the gap between pre-treatment imaging and clot pathobiology.