Assessing Neurocognition After Cerebrovascular Intervention

Patients are being asked to partake in this study because the PI has decided on clinical grounds as outlined in the clinical trial, NASCET, (either asymptomatic 70% carotid stenosis or symptomatic >50% carotid stenosis) that CAS or CEA is indicated. The only aspect of the study that is research is the cognitive testing, the inflammatory blood markers, and components of the neuroimaging.

In this pilot study 20 patients, undergoing CAS for CS, will undergo cognitive testing at 2 time points before intervention (1 months, and 2 months) and at 2 time points after invention (1 month and 2 months). The two preoperative time points will serve as the control group and the two postoperative time points will serve as the intervention group.

Patients, as outlined by NASCET criteria, will have between 50% symptomatic stenosis or 70% asymptomatic stenosis of the carotid artery. If patients present acutely or are otherwise clinically determined to be unstable (e.g. crescendo TIAs, intolerance of normal physiologic blood pressure, etc.) due to their carotid disease, they will excluded from the study.

Patients will serve as their own controls, classifying the study as a within-subjects design. Patients will also have venous blood drawn to ascertain the change in serum markers associated with endothelial dysfunction and inflammation before and after intervention. Finally, patients will have magnetic resonance imaging (MRA) one month before intervention and 2 months after intervention in order to assess changes in global cerebral blood flow.

The following neurocognitive tests will take place - the duration of all typically takes one hour to complete.

1. Memory measured with the Hopkins Verbal Learning Test-Revised (HVLT-R) and Rey Complex Figure
2. Executive function assessed with Trails A and B
3. Attention/processing speed measured with Wechsler Adult Intelligence Scale Digit Span and Coding subtests
4. Language assessed with the Multilingual Naming Test and Verbal Fluency
5. Visuospatial abilities measured with Block Design
6. Global cognition measured with Mattis Dementia Rating Scale
7. Estimated premorbid intelligence measured with the American version of the National Adult Reading Test (preoperatively only)
8. Mood assessed with the Beck Depression Inventory-II (BDI-II) and Beck Anxiety Inventory (BAI) Biomarkers, as a proxy for systemic inflammation will also be collected. The following biomarkers are associated with inflammation-TNF-α, IL-6, IL-1β, CRP, IFN-γ and endothelial cell activation-sICAM sVCAM will be obtained from venous blood (5 ml) and stored at -80° C until the time of electrochemiluminescence assay. As an accompanying measurement of inflammation, complete blood counts will be assessed.

R statistical software will be used for the statistical analysis. Fisher exact, chi square, and logistic regression testing will be used to assess categorical data while paired t tests and linear regression models will be used to assess continuous variables. Multivariable univariate logistic and linear regression will be performed to assess if carotid intervention is associated with statistically significant changes in neurocognitive performance or changes in serum biomarker levels.

Additional predictors to assess neurocognitive performance will include age, gender, race, degree of carotid stenosis, side of stenosis, prior history of stroke, prior history of neurosurgical or endovascular interventions, presence of coronary artery disease, active malignancy, vasculopathy including diabetes, hypertension, hyperlipidemia, obesity, systemic vascular disease, intracranial atherosclerosis, ultrasound perfusion of intracranial vessels, prior neurocognitive decline: ie prior history of Alzheimer's, degenerative neurological disease. Outcome variables will include standardized neuropsychological measures. The investigators hypothesize a more exuberant neurocognitive response in patients with an incomplete circle of Willis.

A dropped model will be performed to prevent overfitting of the regression analysis. A p value less than 0.05 will be used to define statistical significance.

The investigators predict that patients will not have a statistically significant change in neurocognitive performance during the three-month and six-week time points prior to intervention-and this will serve as the control period. We predict that 35% of patients will have some improvement in neurocognition after intervention. The sample size calculator to identify a difference in groups at a p<0.05 with a power of 0.8 predicts with 5% attrition determines that 20 patients will be required for the analysis.