Imaging the clot: does clot appearance predict the efficacy of thrombolysis?

See related article, pages 2379–2383. Unlike in acute myocardial infarction, the underlying pathophysiological mechanism of vascular arterial occlusion in acute stroke is heterogeneous. Composition of cerebral embolic material may vary, depending on specific endothelial and flow conditions of the embolic source. Old, platelet-rich, and well-organized thrombi formed under flow conditions have been shown to be more resistant to thrombolysis than fresh, fibrinand red cell–rich clots formed under conditions of stasis.1 Moreover, clot structure may differ depending on whether the embolic source is a thrombus engrafted in a proximal atherosclerotic lesion or a clot formed in cardiac cavities. In this context, stroke subtypes may represent a surrogate of the composition of offending clot. Efforts to image intravascular thrombus in acute ischemic stroke have been increasingly done in the last years. In acute ischemic stroke, the presence of hyperattenuated middle cerebral artery sign on computed tomography indicates intraluminal clot with a high specificity but low sensitivity (47%).2 On MRI, vessel signs of arterial occlusion have been described as hyperintense vessel sign on fluid-attenuated inversion recovery images and as susceptibility vessel sign on gradient-echo (GRE SVS) images.3,4 The basis for the detection of GRE SVS in patients with an acute intracranial artery occlusion is paramagnetic deoxyhemoglobin causing signal loss. In patients with acute ischemic stroke imaged at 6 hours of stroke onset, GRE SVS has been shown to have a high sensitivity for detecting an acute intracranial occlusion compared with vessel status on magnetic resonance (MR) angiography.3 Sensitivity and specificity of GRE SVS increase over time because of an increase in deoxihemoglobine content and thrombus retraction as a result of clot aging. In the accompanying article, Cho et al5 studied, retrospectively, 95 patients with acute ischemic stroke attributable to a major intracranial artery occlusion who underwent DWI, GRE, and MR angiography 24 hours of stroke onset. Twenty patients received thrombolysis. Stroke subtypes were independently assessed using Trial of Org 10172 in Acute Treatment criteria. Recanalization on follow-up MR angiography was assessed in 66 patients. The authors found that GRE SVS was associated with cardioemblic (CE) stroke and independently predicted recanalization on MR angiography. Although exploring clot structure and embolic source based on the appearance of intravascular thrombus on GRE is an attractive approach with potential diagnostic and therapeutic implications, the observations of Cho et al.5 raise several considerations.