True versus pseudo-occlusion of the cervical internal carotid artery in acute stroke: A multicenter MR angiography study.

Introduction: Differentiating true from pseudo-occlusion of the cervical internal carotid artery (ICA) in acute ischemic stroke patients undergoing thrombectomy is crucial but challenging. We aimed to investigate the ability of contrast-enhanced magnetic resonance angiography (CE-MRA) to differentiate true from pseudo-occlusion (defined as an isolated thrombus of the intracranial ICA suppressing ascending blood flow) of the cervical ICA in acute ischemic stroke patients.

Materials and methods: Multicenter, retrospective analysis of acute ischemic stroke patients with true or pseudo-occlusion of the cervical ICA and subsequent thrombectomy. Patients with preprocedural CE-MRA showing a lack of contrast filling in the cervical ICA on the symptomatic side were included. Six readers (three radiology fellows and three board-certified radiologists) independently evaluated the CE-MRA images for true or pseudo-occlusion of the cervical ICA using a rating scheme. Their assessments were compared with DSA results as the reference standard. Diagnostic accuracy measures, as well as inter- and intra-reader reliability for detecting pseudo-occlusion, were calculated and compared between subgroups.

Results: A total of 41 patients were included. The median age was 73 years, and 39% were female. According to the reference standard, 16 of 41 (39%) patients had a pseudo-occlusion of the cervical ICA, while the remainder had a true occlusion. The aggregated sensitivity and specificity from all readers were 72% (95% confidence interval [CI]: 62%-81%) and 86% (95% CI: 79%-91%), respectively. Board-certified radiologists performed better, with a sensitivity of 81% (95% CI: 67%-91%) and specificity of 92% (95% CI: 83%-97%). Overall, inter-reader agreement was moderate (κ = 0.48; 95% CI: 0.31-0.65) and reached substantial agreement within the board-certified radiologists subgroup (κ = 0.65; 95% CI: 0.45-0.85).

Conclusion: Differentiating true occlusion from pseudo-occlusion of the cervical ICA using CE-MRA is feasible but requires training in specific imaging characteristics as well as experience in interpreting them, as evidenced by the higher diagnostic accuracy of board-certified radiologists. Correct distinction help in optimal material selection (e.g. size and type of guiding catheter) prior to endovascular treatment.