E-163 Long-term experience using the save technique for treatment of intracranial large vessel occlusions

Background The Stent Retriever Assisted Vacuum-locked Extraction (SAVE) technique has been introduced as an effective embolectomy method in stroke patients suffering from intracranial large vessel occlusion (LVO). We present our multicenter, long-term experience with SAVE. Methods Retrospective, core-team analysis of 200 patients undergoing embolectomy using the SAVE technique due to intracranial LVO at 4 German centers. Primary endpoints were first-pass and overall complete/near-complete reperfusion, defined as a modified Thrombolysis in Cerebral Infarction (mTICI) score of 2 c and 3. Secondary endpoints were number of passes, time from groin puncture to reperfusion, embolization to new territories (ENT), and post-interventional symptomatic intracranial hemorrhage (sICH). Results Median age was 78 years (IQR 68–85). Median NIHSS at admission was 16 (IQR 12–20). Occlusions sites were: ICA-T in 39/200 (19.5%), M1 in 126/200 (63%), M2 in 30/200 (15%), and others in 5/200 (2.5%) cases. The primary endpoints were documented in 114/200 (57% first-pass mTICI2c or 3) and 154/200 (77% overall mTICI2c or 3) patients, respectively. The overall median time from groin puncture to reperfusion was 34 min (IQR 25–52) with a median of 1 (IQR 1–2) attempts. ENTs were observed in 3 patients (1.5%) and the rate of sICH was 2.6%. The rate of successful reperfusion (mTICI ≥2 b) on final angiograms was 95%. Conclusion SAVE is an effective, fast and secure embolectomy method. First-line use of SAVE leads to high-rates of complete/near-complete reperfusion. Disclosures M. Psychogios: 3; C; Stryker, Phenox, Acandis, Penumbra, Siemens. S. Henkel: None. A. Riabikin: None. C. Riedel: None. D. Behme: 3; C; Stryker, Phenox, Penumbra. I. Tsogkas: None. A. Hesse: None. N. Abdullayev: None. O. Jansen: None. M. Wiesmann: 1; C; Abbott. 3; C; Stryker, Bracco, Medtronic, Siemens. A. Mpotsaris: 2; C; Neuravi, Penumbra, Sequent Medical. V. Maus: None.