Progression of moderate-to-severe carotid disease

Abstract

Our goals were to investigate the degree to which patient demographics, risk factors, laboratory data, and medications influence moderate carotid disease progression among patients with asymptomatic moderate carotid disease and whether such associations are solely based on how progression is defined. In addition, we aimed to establish optimal threshold criteria to categorize patients at high risk of progression. In this retrospective study, 621 arteries were evaluated for internal carotid artery (ICA) stenosis between January 1997 and January 2014 and were determined to have moderate (50%-79%) stenosis via color duplex ultrasonography. "Moderate stenosis" was defined as an ICA peak systolic velocity (PSV) ≥120cm/s and a diastolic ICA velocity<140cm/s. Kaplan-Meier analysis of the time to progression was conducted using three independent end points: PSV ≥230cm/s (liberal criterion); ICA/common carotid artery (CCA) ratio ≥4.0 (moderate criterion), and diastolic ICA velocity ≥140cm/s (strict criterion). Kaplan-Meier survival curves were generated, and multivariate analysis was performed using Cox regression models. Risk stratification criteria were based on optimal sensitivity and specificity generated from receiver operating characteristic (ROC) curve analysis. The overall rate of progression was 28.5%, 21.1%, or 5.1% of study-eligible arteries over 5years using liberal, moderate, or strict criterion, respectively. Using liberal criterion, multivariate analysis suggested that initial PSV ≥200cm/s, ICA/CCA ratio ≥3, and male gender were significantly associated with progression. Using the moderate criterion, multivariate analysis revealed that initial PSV ≥200cm/s, ICA/CCA ratio ≥3, age, and male gender were significantly associated with progression. Using the strict criterion, multivariate analysis revealed that initial PSV ≥200cm/s was the only statistically significant predictor of progression. No additional patient demographics, comorbidities, initial laboratory values, or medications consistently influenced disease progression across any criteria in our study. ROC analysis suggests PSV ≥165cm/s is an ideal threshold value for the categorization of high risk patients, as this resulted in an optimal screening sensitivity of nearly 91% and a specificity of 59% over 2years. The timing and incidence of carotid disease progression depends on the definition of disease progression. Among all three criteria, only severity of disease at initial presentation reliably predicted progression. Based on the results of our ROC curve analysis, we propose that an initial ICA PSV ≥165cm/s (sensitivity: 90.7%, specificity: 58.7%) represents a reasonable value for defining high progression risk over a 2-year interval.