Plaque features for the prediction of in-stent restenosis in symptomatic intracranial atherosclerotic stenosis from a retrospective single-center study.

Abstract

This retrospective single-center study evaluates whether high-resolution magnetic resonance vessel wall imaging (HR-VWI) plaque features can predict in-stent restenosis (ISR) before stent implantation. Consecutive intracranial atherosclerotic disease (ICAD) patients undergoing HR-VWI and endovascular treatment from March 2018 to June 2024 were analyzed. ISR was defined as > 50% stenosis within or near (< 5 mm) the stent and > 20% absolute luminal loss. Multivariate logistic regression identified independent ISR risk factors, developing a prediction model and diagnostic nomogram. ISR-free survival was evaluated using Kaplan-Meier curves. Among 217 patients (mean age 64.5 ± 10.2 years; 73.2% male), 50 (23.0%) developed ISR within 1 year. Plaque eccentricity (odds ratio (OR), 32.661; 95% confidence interval (CI): 3.726-286.334; p = 0.002), higher plaque-to-pituitary stalk contrast ratio (CR) (OR, 2.702; 95% CI: 1.098-6.649; p = 0.03), and lower lymphocyte-to-monocyte ratio (OR, 0.251; 95% CI: 0.106-0.594; p = 0.002) were independent ISR predictors. The model's AUC was 0.783 (95% CI: 0.683-0.883) for modeling and 0.827 (95% CI: 0.717-0.936) for validation, with a 0.3 cutoff. Kaplan-Meier analysis showed higher 12-month restenosis-free rates for patients with plaque eccentricity < 0.57 (68.5% vs. 27.7%; p < 0.001), CR < 38.22 (76.6% vs. 34.4%; p < 0.001) and lymphocyte-to-monocyte ratio ≥ 3.27 (70.4% vs. 37.6%; p < 0.001). In symptomatic ICAD patients, ISR was linked to plaque eccentricity, elevated CR on HR-VWI, and reduced lymphocyte-to-monocyte ratio. These single-center findings may optimize therapeutic strategies and reduce restenosis risk, though further validation is needed to confirm their generalizability and reliability. Question Can clinical and imaging biomarkers predict in-stent restenosis (ISR) in symptomatic intracranial atherosclerotic stenosis (ICAD) to guide preprocedural risk stratification? Findings Plaque eccentricity, high plaque-to-pituitary stalk contrast ratio, and low lymphocyte-to-monocyte ratio independently predict ISR; our nomogram achieved 75% sensitivity and 82.6% specificity. Clinical relevance Preprocedural ISR prediction enables personalized strategies (e.g., stent optimization, targeted anti-inflammatory therapy), potentially reducing restenosis and stroke recurrence risk in ICAD patients undergoing stenting.