Thrombus imaging characteristics within acute ischemic stroke: similarities and interdependence

Abstract

Background: The effects of thrombus imaging characteristics on procedural and clinical outcomes after ischemic stroke are increasingly being studied. These thrombus characteristics - for eg, size, location, and density - are commonly analyzed as separate entities. However, it is known that some of these thrombus characteristics are strongly related. Multicollinearity can lead to unreliable prediction models. We aimed to determine the distribution, correlation and clustering of thrombus imaging characteristics based on a large dataset of anterior-circulation acute ischemic stroke patients.

Methods: We measured thrombus imaging characteristics in the MR CLEAN Registry dataset, which included occlusion location, distance from the intracranial carotid artery to the thrombus (DT), thrombus length, density, perviousness, and clot burden score (CBS). We assessed intercorrelations with Spearman's coefficient (ρ) and grouped thrombi based on 1) occlusion location and 2) thrombus length, density and perviousness using unsupervised clustering.

Results: We included 934 patients, of which 22% had an internal carotid artery (ICA) occlusion, 61% M1, 16% M2, and 1% another occlusion location. All thrombus characteristics were significantly correlated. Higher CBS was strongly correlated with longer DT (ρ=0.67, p<0.01), and moderately correlated with shorter thrombus length (ρ=-0.41, p<0.01). In more proximal occlusion locations, thrombi were significantly longer, denser, and less pervious. Unsupervised clustering analysis resulted in four thrombus groups; however, the cohesion within and distinction between the groups were weak.

Conclusions: Thrombus imaging characteristics are significantly intercorrelated - strong correlations should be considered in future predictive modeling studies. Clustering analysis showed there are no distinct thrombus archetypes - novel treatments should consider this thrombus variability.

Keywords: CT; CT Angiography; Stroke; Thrombectomy; Thrombolysis.

Figure 1

Thrombus markers. A) Schematic overview of thrombus marker placement, axial view. (B) Schematic overview of thrombus marker placement, coronal view. (C) Axial view of CTA and NCCT scans showing a left-MCA occlusion and placed markers. To visualize the markers, a zoom-in of the occlusion is displayed and the markers are circled in their corresponding color in the image in the lower row. These circles do not represent the 1 mm radius spherical regions of interest used to compute density and perviousness. (D) Coronal view of the CTA and NCCT scans displayed in C showing a left-MCA occlusion and placed markers. A closer view of the occlusion is displayed in the image in the lower row, with the markers circled in their corresponding color. A1, anterior cerebral artery A1 segment; CTA, computed tomography angiography; ICA, internal carotid artery; M1, middle cerebral artery M1 segment; M2, middle cerebral artery M2 segment; MCA, middle cerebral artery; NCCT, non-contrast computed tomography.

Figure 2

Thrombus imaging characteristics. The diagonal shows the histograms of DT, thrombus length, density, perviousness and CBS. The graphs below the diagonal show bivariate scatter plots, Spearman’s correlation coefficient (ρ) and significance level of the correlation, where ** means p<0.01. The plots above the diagonal show bivariate scatter plots with a linear fit. CBS, clot burden score; DT, distance from the internal carotid artery terminus to the thrombus; HU, Hounsfield units; mm, millimeters; thr, thrombus.

Figure 3

Finding the optimal number of clusters. (A)SSE as a function of the number of clusters. The elbow point is found around 4–6, after which the decrease in SSE becomes linear. (B) Silhouette coefficients as a function of the number of clusters. The highest (and therefore, preferred) silhouette coefficients are found around 4–6. (C) Davies-Bouldin coefficients as a function of the number of clusters. Lower coefficients indicate better clustering. From k=4 onwards, Davies-Bouldin coefficients showed similar values, with two local minima at k=6 and k=9. (D) 3D visualization of thrombus imaging features clustered by 4-means unsupervised clustering. (E) 3D visualization of thrombus imaging features clustered by occlusion location. SSE, sum of the squared error.