Recanalization and clinical outcome of occlusion sites at baseline CT angiography in the Interventional Management of Stroke III trial

Abstract

Purpose: To use baseline computed tomographic (CT) angiography to analyze imaging and clinical end points in an Interventional Management of Stroke III cohort to identify patients who would benefit from endovascular stroke therapy.

Materials and methods: The primary clinical end point was 90-day dichotomized modified Rankin Scale (mRS) score. Secondary end points were 90-day mRS score distribution and 24-hour recanalization. Prespecified subgroup was baseline proximal occlusions (internal carotid, M1, or basilar arteries). Exploratory analyses were subsets with any occlusion and specific sites of occlusion (two-sided α = .01).

Results: Of 656 subjects, 306 (47%) underwent baseline CT angiography or magnetic resonance angiography. Of 306, 282 (92%) had arterial occlusions. At baseline CT angiography, proximal occlusions (n = 220) demonstrated no difference in primary outcome (41.3% [62 of 150] endovascular vs 38% [27 of 70] intravenous [IV] tissue-plasminogen activator [tPA]; relative risk, 1.07 [99% confidence interval: 0.67, 1.70]; P = .70); however, 24-hour recanalization rate was higher for endovascular treatment (n = 167; 84.3% [97 of 115] endovascular vs 56% [29 of 52] IV tPA; P < .001). Exploratory subgroup analysis for any occlusion at baseline CT angiography did not demonstrate significant differences between endovascular and IV tPA arms for primary outcome (44.7% [85 of 190] vs 38% [35 of 92], P = .29), although ordinal shift analysis of full mRS distribution demonstrated a trend toward more favorable outcome (P = .011). Carotid T- or L-type occlusion (terminal internal carotid artery [ICA] with M1 middle cerebral artery and/or A1 anterior cerebral artery involvement) or tandem (extracranial or intracranial) ICA and M1 occlusion subgroup also showed a trend favoring endovascular treatment over IV tPA alone for primary outcome (26% [12 of 46] vs 4% [one of 23], P = .047).

Conclusion: Significant differences were identified between treatment arms for 24-hour recanalization in proximal occlusions; carotid T- or L-type and tandem ICA and M1 occlusions showed greater recanalization and a trend toward better outcome with endovascular treatment. Vascular imaging should be mandated in future endovascular trials to identify such occlusions. Online supplemental material is available for this article.

Figure 1:

Diagram shows prespecified baseline CT angiography distribution of proximal occlusions (ICA, M1, basilar artery). The mRS distribution was not significantly different (generalized Wilcoxon test, P = .1068).

Figure 2:

Diagram shows post hoc baseline CT angiography distribution of any visible occlusions. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .011).

Figure 3a:

Diagrams show distribution of findings at baseline CT angiography. (a) Carotid T- or L-type occlusions or tandem ICA and M1 occlusions. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .02). (b) Proximal M1 occlusions without ICA involvement. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .93). (c) Distal M1 occlusions without ICA involvement. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .48). (d) M2 occlusion with or without ICA involvement. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .14).

Figure 3b:

Diagrams show distribution of findings at baseline CT angiography. (a) Carotid T- or L-type occlusions or tandem ICA and M1 occlusions. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .02). (b) Proximal M1 occlusions without ICA involvement. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .93). (c) Distal M1 occlusions without ICA involvement. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .48). (d) M2 occlusion with or without ICA involvement. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .14).

Figure 3c:

Diagrams show distribution of findings at baseline CT angiography. (a) Carotid T- or L-type occlusions or tandem ICA and M1 occlusions. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .02). (b) Proximal M1 occlusions without ICA involvement. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .93). (c) Distal M1 occlusions without ICA involvement. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .48). (d) M2 occlusion with or without ICA involvement. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .14).

Figure 3d:

Diagrams show distribution of findings at baseline CT angiography. (a) Carotid T- or L-type occlusions or tandem ICA and M1 occlusions. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .02). (b) Proximal M1 occlusions without ICA involvement. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .93). (c) Distal M1 occlusions without ICA involvement. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .48). (d) M2 occlusion with or without ICA involvement. The mRS distribution was not significantly different (generalized Wilcoxon test, P = .14).