Quantitative analysis and predictors of embolic filter debris load during carotid artery stenting in asymptomatic...

The objective of this study was to perform a quantitative analysis and to identify predictors of embolic filter debris (EFD) load during carotid artery stenting (CAS) in asymptomatic patients.

All patients with asymptomatic carotid stenosis >70% undergoing CAS between 2008 and 2016 were included in a prospective database. A distal filter protection device was used in all patients. At the end of the procedure, the filter was fixed in formalin and then analyzed with a stereomicroscope. Morphometric analysis was performed with Image-Pro Plus software (Media Cybernetics, Rockville, Md). The total area of the filter membrane and the area covered by particulate material were quantified. The quantity of membrane occupied by debris was expressed as percentage of covered surface area. Anatomic and clinical variables were evaluated for their association with EFD load using multiple logistic regression.

Among the 278 patients undergoing CAS, an open-cell stent was implanted in 211 patients (76%); 67 patients (24%) received a closed-cell stent. Overall technical success and clinical success were both 99%; no perioperative death was reported. Stroke rate was 1.8% (major, n = 1 [0.4%]; minor, n = 4 [1.4%]); transient ischemic attacks occurred in 5% of cases (n = 14). The quantitative analysis of the filter revealed that EFD was present in 74% of cases (n = 207). The mean EFD load was 10% of the filter surface (median, 1; range, 0-80); it was <10% in 203 patients (73%), between 11% and 20% in 39 patients (14%), between 21% and 30% in 14 patients (5%), and >31% in 22 (8%). Patients with any type of ischemic neurologic event after CAS (stroke and transient ischemic attack) had a significantly higher mean EFD load compared with uneventful cases (26.7% ± 19.0% vs 8.5% ± 13.5%; P < .001). The observational frequency distribution analysis identified the presence of >12.5% EFD load as the optimal cutoff for the association with clinically relevant perioperative ischemic events (sensitivity, 78%; specificity, 77%; area under the curve, 0.81). The multivariate analysis demonstrated that age >75 years (odds ratio [OR], 2.56; P = .003), pre-existing ipsilateral ischemic cerebral lesions (OR, 2.09; P = .047), hypoechogenic plaque on the preoperative duplex ultrasound examination (OR, 6.05; P < .001), and plaque length >15 mm (OR, 1.79; P = .049) were independent predictors of EFD load >12.5%.

The majority of asymptomatic carotid stenoses treated with CAS have detectable embolic debris in the protecting filter. Age >75 years, pre-existing ipsilateral cerebral ischemic lesions, hypoechogenic plaque, and plaque length >15 mm should be taken into consideration as independent predictors of clinically relevant embolic debris during the procedure.