Ischemic Stroke of Suspected Cardioembolic Origin Despite Anticoagulation: Does Thrombus Analysis Help to Clarify Etiology?

INTRODUCTION: Despite sufficient oral anticoagulation (OAC) to prevent cardioembolism, some patients suffer from cerebral ischemic strokes of suspected cardioembolic origin. Reasons for that are not clarified yet. In certain cases, the suspected cardioembolic origin of stroke is questioned. This study aimed to understand the thrombi origin and pathophysiology in patients suffering from stroke despite OAC by the analysis of histologic thrombus composition and imaging characteristics. MATERIALS AND METHODS: On two distinct cohorts, we retrospectively analyzed histologic (n = 92) and imaging features (n = 64), i.e., thrombus perviousness in admission CT imaging, of cerebral thrombi retrieved by the endovascular treatment for a large vessel occlusion of the anterior circulation. In each group, patients with non-cardioembolic strokes and suspected cardioembolic strokes with or without anticoagulation were compared. RESULTS: Fibrin/platelet content of suspected cardioembolic thrombi (mean/SD 57.2% ± 13) is higher than in non-cardioembolic thrombi (48.9% ± 17; p = 0.01). In suspected cardioembolic thrombi, the fibrin/platelet content does not differ in the subgroups of patients with (57.3% ± 13) and without prior OAC treatment (56.6% ±13; p = 0.8), both with higher values than non-cardioembolic thrombi. Thrombus perviousness (ε) of suspected cardioembolic OAC thrombi (mean/SD: 0.09 ± 0.06) differs significantly from non-cardioembolic thrombi (0.02 ± 0.02; p < 0.001). Further, ε is higher in suspected cardioembolic thrombi with OAC than in cardioembolic thrombi without OAC (0.06 ± 0.03; p = 0.04) and with insufficient OAC (0.04 ± 0.02; p = 0.07). CONCLUSION: Thrombi of the suspected cardioembolic origin of patients with prior OAC do not differ in their histologic composition from those without prior OAC, but both differ from non-cardioembolic thrombi. These histologic results make a non-cardioembolic etiology for strokes despite prior OAC rather unlikely but favor other reasons for these ischemic events. Perviousness assessment reinforces the histologic findings, with additional information about the OAC thrombi, which present with higher perviousness. This suggests that OAC would not affect the relative histologic thrombus composition but may alter the microstructure, as reflected by perviousness.