Identification of symptomatic carotid artery plaque: a predictive model combining angiography with optical coherence tomography

Abstract

ObjectiveSymptomatic carotid artery disease is indicative of an elevated likelihood of experiencing a subsequent stroke, with the morphology of plaque and its specific features being closely linked to the risk of stroke occurrence. Our study based on the characteristics of carotid plaque assessed by optical coherence tomography (OCT), the plaque morphology evaluated by digital subtraction angiography (DSA) and clinical laboratory indicators were combined, develop a combined predictive model to identify symptomatic carotid plaque.MethodsPatients diagnosed with carotid atherosclerotic stenosis who underwent whole-brain DSA and OCT examination at the Affiliated Hospital of Jining Medical University from January 2021 to November 2023 were evaluated. Clinical features, as well as DSA and OCT plaque characteristics, were analyzed for differences between symptomatic and asymptomatic cohorts. An analysis of logistic regression was carried out to identify factors associated with the presence of symptomatic carotid plaque. A multivariate binary logistic regression equation was established with the odds ratio (OR) serving as the risk assessment parameter. The receiver operating characteristic curve was utilized to assess the combined predictive model and independent influencing factors.ResultsA total of 52 patients were included in the study (symptomatic: 44.2%, asymptomatic: 55.8%). Symptomatic carotid stenosis was significantly linked to four main factors: low-density lipoprotein-cholesterol >3.36 mmol/L [OR, 6.400; 95% confidence interval (CI), 1.067–38.402; p = 0.042], irregular plaque (OR, 6.054; 95% CI, 1.016–36.083; p = 0.048), ruptured plaque (OR, 6.077; 95% CI, 1.046–35.298; p = 0.048), and thrombus (OR, 6.773; 95% CI, 1.194–38.433; p = 0.044). The combined predictive model generated using four indicators showed good discrimination (Area Under Curve, 0.924; 95% CI, 0.815–0. 979). The p value was <0.05 with 78.26% sensitivity and 93.10% specificity.ConclusionOCT is valuable in evaluating the plaque characteristics of carotid atherosclerotic stenosis. The combined predictive model comprising low-density lipoprotein-cholesterol >3.36 mmol/L, irregular plaque, ruptured plaque, and thrombus could help in the detection of symptomatic carotid plaque. Further research conducted on additional independent cohorts is necessary to confirm the clinical significance of the predictive model for symptomatic carotid plaque.