Introduction
: Current imaging modalities might underestimate the presence and severity of intracranial atherosclerosis (ICAD). High resolution vessel wall imaging (HR‐VWI) MRI emerged as a powerful tool to diagnose plaques not detected on routine imaging. We aim to compare different imaging modalities (HR‐VWI MRI; digital subtraction angiogram (DSA); Time‐of‐flight (TOF) MRA; and CTA) in the identification and characterization of intracranial atherosclerotic culprit plaques.
Methods
: Patients diagnosed with ICAD were prospectively imaged with HR‐VWI MRI. Culprit plaques were identified based on the likelihood of causing the stroke. Using cross‐sectional images of intracranial vessels, regions of interest (ROI) were delineated. Then, diameters and ROI areas were measured for the purpose of calculating the following variables: degree of stenosis (DS) at the plaque level, plaque burden (PB), and remodeling index (RI). Additional imaging modalities (DSA, TOF MRA, and CTA) were identified retrospectively for each patient. The sensitivity of detecting a culprit plaque as well as the correlations between the different variables were analyzed for each modality. Linear regression analysis was used to determine the association of DS with PB and RI. Interobserver agreement on the determination of a culprit plaque on every imaging modality was evaluated.
Results
: A total of 44 patients who underwent HR‐VWI had ICAD and were included in the final analysis. Of those, 34 had CTA, 18 had TOF‐MRA, and 18 had DSA. Using HR‐VWI as gold standard, the sensitivity for culprit plaque detection was 88% for DSA, 78% for TOF MRA, and 76% for CTA. We found no difference between the DS in all four modalities using measured cross‐sectional diameters, but difference was found when measuring ROI areas to calculate DS. There was a significant positive correlation between PB and DS on HR‐VWI MRI (p<0.001), but not on the DSA (p = 0.168), MRA (p = 0.144), or CTA (p = 0.253), and a significant negative correlation between RI and DS on HR‐VWI MRI (p = 0.003), but not on DSA (p = 0.783), MRA (p = 0.405), or CTA (p = 0.751). PB and RI predicted the degrees of stenosis on HR‐VWI, but not on the other modalities. There was good inter‐rater agreement for culprit plaque detection on HR‐VWI (k = 0.48, p = 0.001), but no agreement was found on the other modalities.
Conclusions
: HR‐VWI MRI can locate otherwise undetectable plaques on conventional imaging through the ability to measure plaque burden, an essential component for characterization of plaques severity and a strong predictor of stenosis. HR‐VWI also showed more accurate measurements of degree of stenosis through measurement of ROI areas, and had good inter‐rater agreement for accurate plaque detection, compared to DSA, MRA, and CTA.
701-2 The Final % Cross-sectional Narrowing (Residual Plaque Burden) is the Strongest Intravascular Ultrasound Predictor of Angiographic Restenosis
