Abstract 1122‐000138: De Novo Intracranial Stenosis after Mechanical Thrombectomy with Stent Retrievers

Introduction : Increased vascular damage with the use of stent‐retrievers (SR) has been shown on histopathological analysis of the vascular tissue immediately after mechanical thrombectomy (MT) in animal models. We hypothesized that intraoperative endovascular damage‐intimal injury could result in fibrosis and de novo vascular stenosis (dnVS). The purpose of the study is to identify de novo or worsening intracranial stenosis (wICS) of the treated vessel(s) on patients who underwent MT for the treatment of acute ischemic stroke with SR, on follow‐up vascular imaging (FVI). Methods : This was a retrospective chart review. Patients who underwent MT with SR at two centers from January 2015‐December 2020, who had FVI (CTA, MRA or cerebral angiogram) were included. Patient characteristics, procedural details, timing for FVI and clinical outcomes were collected. Two neuroradiologists reviewed baseline angiograms and FVI to assess for the presence of dnVS or wICS, and graded each stenosis and collateral scores (CS), when stenosis was present. CS were calculated using the multiphase CT angiography collateral score (mCTA). Fischer exact test and Mann‐Whitney U test were used to assess for differences in categorical and continuous variables, respectively. Statistical analysis was performed using SPSS 28.0 (IBM Corp.). Results : Forty‐six patients within this cohort had FVI with 9 patients developing dnVS or wICS in the follow‐up period (19.6%) with a median follow‐up of 113 days. Five of these patients demonstrated a complete occlusion of the target vessel on FVI. Of the remaining 4 patients, mean degree of stenosis was 55%. Only 2 of these patients had underlying stenosis on baseline post‐treatment angiogram: one with 44% stenosis which progressed to 95% in 2 months. Another with mild stenosis that progressed to complete occlusion in 50 days. Adequate revascularization, defined as TICI score >2b was achieved in 88.8% of patients with dnVS or wICS, and in 89.2% of patients with stable FVI. No significant differences were observed in baseline demographics, NIHSS score at presentation or initial ASPECTS. Median number of passes was identical between patients who developed dnVS or wICS (median 1, IQR [1, 2], p = 0.683). Mean CS for dnVS or wICS was 3. No significant differences were observed in discharge or follow‐up NIHSS scores, mRS, mortality, or recurrent stroke or TIA between the two cohorts. Conclusions : MT with SR can be associated with dnVS or wICS in some patients. The number of passes with SR did not seem to have an impact on this. Patients with dnVS or wICS did not have a higher incidence of recurrent stroke or TIA. This could be due to the development of new collaterals in this population. Our study is limited by a small cohort, however, larger studies might be challenging as standardized radiological follow up of these patients has not been implemented.