Serial MRI of carotid plaque burden: influence of subject repositioning on measurement precision.

MRI has the potential to track changes in the size of carotid atherosclerotic plaques for patient monitoring or in clinical trials. For either application, highly reproducible serial measurements are critical for drawing accurate conclusions. This study investigates the role of inconsistent repositioning of the artery from scan to scan in overall measurement variability. The total variability of cross-sectional area measurements is modeled as a combination of variability due to segmentation differences, changes in orientation of the artery, and longitudinal offsets of the image location. Model parameters were generated using measurements from carotid arteries imaged in vivo. Comparison with actual reproducibility measurements for the lumen and outer wall boundaries showed that the errors were accurately predicted by the model, including a strong correlation between lumen and wall measurement differences. Orientation variability was the single most important factor that affected reproducibility, which suggests that methods are needed to obtain consistent orientation of the artery relative to the image plane.