ABSTRACT NUMBER: ESOC2026A1940 OCCLUSION DYNAMICS BETWEEN CTA AND INITIAL DSA AFTER INTERHOSPITAL TRANSFER FOR MECHANICAL THROMBECTOMY

Abstract Background and aims In patients undergoing interhospital transfer for mechanical thrombectomy, occlusion status may change between initial computed tomography angiography (CTA) and the first intraprocedural digital subtraction angiography (DSA). During transfer, spontaneous reperfusion or changes in thrombus location may occur, resulting in discrepancies between referral imaging and angiographic findings. This study aims to systematically assess these dynamics. Methods This observational study included patients with CTA confirmed anterior circulation vessel occlusion and interhospital transfer for thrombectomy between 2018-08-10 and 2025-11-30. Primary outcome was the dynamic of the occlusion between CTA and initial DSA, categorized as spontaneous reperfusion, persistent occlusion, or change in occlusion pattern. Exploratory stratified analyses were performed for selected clinical and procedural variables. Results A total of 1,212 patients were included (median age 78 years, 51% female). In 21%, the initially planned thrombectomy was not performed and no DSA was obtained. Among patients undergoing DSA, occlusion location persisted in 58% and changed in 32%. Spontaneous reperfusion was observed in 10%. The occlusions that most frequently persisted were located in the M1 segment (67%) and the extracranial ICA (60%). Median time from CTA to DSA was 193 minutes (IQR: 165.5-230.0). Patients with intravenous thrombolysis prior to transfer showed higher rates of occlusion dynamics. Conclusions Discrepancies between occlusion status on CTA and initial DSA are common after interhospital transfer for mechanical thrombectomy. Multiple factors are associated with these dynamics, indicating that differences cannot be attributed solely to imaging techniques or interrater variability. Occlusion dynamics should be considered when interpreting imaging and planning transfer for endovascular treatment. Conflict of interest All authors: nothing to disclose Figure 1 - belongs to Results