Incremental value of pericarotid adipose tissue radiomics in predicting in-stent restenosis after carotid artery stenting
- 21 October 2025
- journal article
- Published by BMJ in Journal of NeuroInterventional Surgery
- Vol. 18 (10) , 2064-2072
Abstract
Objective To evaluate the potential of pericarotid adipose tissue radiomics to improve the prediction of in-stent restenosis (ISR) after carotid artery stenting (CAS). Methods This retrospective study included 191 patients who underwent carotid CT angiography (CTA) and CAS within 1 week at two centers from September 2019 to December 2023. ISR was defined as ≥50% stenosis on follow-up Doppler ultrasound or CTA. Three predictive models were developed and defined as follows: Model A (Clinical), Model B (Clinical + Imaging), and Model C (Clinical + Imaging + Radiomics) using receiver operating characteristic (ROC) analysis, calibration curves, and decision curve analysis. Results ISR occurred in 44 patients with a mean time interval of 11.3 months. Multivariate Cox regression analyses identified diabetes, fibrinogen, systolic blood pressure, calcified plaque volume, and pericarotid adipose tissue radiomics as independent predictors of ISR. The radiomics score, derived from 15 significant characteristics, outperformed conventional imaging markers. In the training set, Model C (AUC=0.881) significantly outperformed Model A (AUC=0.664) and Model B (AUC=0.840), with statistically significant differences (Model A vs Model C: P=0.001; Model B vs Model C: P=0.0246). This trend was consistent in the validation sets. Calibration curves showed good agreement between predicted and actual ISR probabilities, and decision curve analysis indicated that Model C provided greater net benefits. Conclusion The radiomic characteristics of pericarotid adipose tissue provide incremental value in predicting ISR after CAS and serve as a valuable biomarker for restenosis risk assessment.Keywords
Funding Information
- Liaoning Province Applied Basic Research Program (2023JH2/101300062)
- National Natural Science Foundation of China (82471983)
- Liaoning Province Science and Technology Plan Joint Fund Project (2023-MSLH-115)
This publication has 44 references indexed in Scilit:
- Enhancing Radiologists’ Performance in Detecting Cerebral Aneurysms Using a Deep Learning Model: A Multicenter StudyAcademic Radiology, 2024
- Prediction value of pericoronary fat attenuation index for coronary in-stent restenosisEuropean Radiology, 2024
- A new predictor of coronary artery disease in acute ischemic stroke or transient ischemic attack patients: pericarotid fat densityEuropean Radiology, 2023
- Associations between preprocedural carotid artery perivascular fat density and early in-stent restenosis after carotid artery stentingHeliyon, 2023
- Editor's Choice – European Society for Vascular Surgery (ESVS) 2023 Clinical Practice Guidelines on the Management of Atherosclerotic Carotid and Vertebral Artery DiseaseEuropean Journal of Vascular and Endovascular Surgery, 2022
- Identification of high-risk carotid plaque by using carotid perivascular fat density on computed tomography angiographyEuropean Journal of Radiology, 2022
- Prediction of acute coronary syndrome within 3 years using radiomics signature of pericoronary adipose tissue based on coronary computed tomography angiographyEuropean Radiology, 2021
- Perivascular Adipose Tissue and Coronary Atherosclerosis: from Biology to Imaging PhenotypingCurrent Atherosclerosis Reports, 2019
- Plasma fibrinogen levels and restenosis after primary percutaneous coronary interventionJournal of Thrombosis and Thrombolysis, 2011
- Crosstalk between perivascular adipose tissue and blood vesselsCurrent Opinion in Pharmacology, 2010