Polyvascular disease (PVD), characterized by the simultaneous involvement of multiple arterial territories, poses a significant challenge in vascular pathology. Its increasing prevalence, particularly among elderly patients with multimorbidity, underscores the need for an integrated diagnostic and therapeutic approach. This study aimed to evaluate the prevalence of PVD and identify predictive risk factors, integrating artificial intelligence (AI) to achieve more precise cardiovascular risk stratification

A cross-sectional study was conducted on 302 patients hospitalized for coronary artery disease. Clinical, biological, and angiographic data were collected. PVD was defined as significant involvement in at least two vascular territories (coronary, carotid, peripheral arteries). Statistical analyses were performed to identify risk factors associated with PVD. An AI-based algorithmic approach was explored to predict the risk of PVD from clinical and biological data

The prevalence of PVD in our cohort was 13.24%. Significantly associated risk factors included advanced age, diabetes (OR = 3.83; P < 0.001), hypertension (OR = 2.08; P = 0.011), dyslipidemia (OR = 2.00; P = 0.007), and the presence of carotid stenosis (OR = 3.72; P < 0.001). The AI-based algorithm improved PVD risk prediction compared to traditional models, with an area under the ROC curve of 0.85.

PVD is a common entity among coronary artery disease patients, associated with a high cardiovascular risk profile. The integration of AI enhances risk stratification and optimizes prevention and management strategies, especially in elderly patients with multimorbidity.

These results highlight the importance of a comprehensive and personalized approach to reduce the burden of PVD.