Abstract / Summary
Abstract Objective To develop and internally validate a model using contralateral-controlled thigh muscle biomarkers, measured on routine computed tomography angiography (CTA), to predict clinically driven, DSA-confirmed 1-year restenosis after endovascular therapy for unilateral peripheral artery disease (PAD). Materials and methods In this retrospective cohort study, 378 patients with unilateral PAD undergoing endovascular therapy and having baseline and 1-year follow-up CTA were included. Mid-thigh skeletal muscle cross-sectional area and whole-thigh muscle volume were quantified bilaterally with semi-automated software. Contralateral-controlled biomarkers (affected minus contralateral) were derived to isolate limb-specific ischemic effects. The primary endpoint was clinically driven 1-year restenosis (≥ 50%) confirmed by DSA. Results Among 378 patients, 162 (43%) developed restenosis at a median follow-up of 12.5 months. Contralateral-controlled muscle asymmetry was independently associated with restenosis after adjustment for lesion length, baseline stenosis degree, Fontaine stage, disease duration, and TASC II classification (OR 1.43, 95% CI: 1.08–1.89, p = 0.01). Adding muscle asymmetry to a severity-only model improved discrimination from AUC 0.76 to 0.79 (DeLong p = 0.04). A baseline predictive model combining lesion length and muscle asymmetry achieved AUC 0.74; a full explanatory model incorporating longitudinal muscle changes reached AUC 0.88, with the greatest net benefit on decision curve analysis. Conclusion Thigh muscle biomarkers, acquired opportunistically from routine CTA scans, improved the prediction of 1-year restenosis after endovascular therapy and may support targeted surveillance in patients with unilateral PAD. Key Points Question Can contralateral-controlled thigh muscle biomarkers on routine CT angiography improve the prediction of 1-year restenosis after endovascular therapy for unilateral peripheral artery disease? Findings In 378 PAD patients, baseline contralateral thigh muscle asymmetry independently predicted clinically driven, DSA-confirmed restenosis and improved AUC from 0.76 to 0.79. Critical relevance statement Opportunistic contralateral-controlled thigh muscle analysis on routine CT angiography provides a no-cost imaging biomarker that refines post-endovascular restenosis risk stratification, supporting targeted angiographic surveillance and advancing personalized vascular imaging practice in peripheral artery disease.