Hemodynamic Assessment of Carotid Artery Stenosis Using 4D Flow MRI: Complementary Insights Beyond Morphological Severity.

RATIONALE AND

OBJECTIVES

To investigate the association between hemodynamic parameters derived from four-dimensional flow magnetic resonance imaging (4D flow MRI) and carotid artery stenosis severity, to evaluate their diagnostic performance, and to perform an exploratory assessment of morphology-related hemodynamic variation.

MATERIAL AND

METHODS

A total of 65 patients with carotid artery stenosis underwent 4D flow MRI and time-of-flight MR angiography (TOF-MRA). Stenosis rate was calculated using the North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria. Hemodynamic parameters including peak velocity, average wall shear stress (WSS), and total forward flow volume were extracted at the most stenotic segment. Associations with stenosis rate were evaluated using Spearman correlation. Multivariable logistic regression and receiver operating characteristic (ROC) analyses were performed to assess predictive performance. Supplementary exploratory analyses examined the relationship between morphology-based classification and hemodynamic parameters.

RESULTS

Peak velocity showed a weak positive correlation with stenosis rate, whereas average WSS and total forward flow volume showed weak or negligible correlations. In multivariable analysis, peak velocity was identified as an independent predictor of significant stenosis (NASCET ≥50%). ROC analysis demonstrated moderate diagnostic performance, with the combined model achieving the highest area under the curve (AUC ≈ 0.72). Exploratory morphology-based analyses showed qualitative differences in flow patterns, but no significant quantitative differences in the evaluated hemodynamic parameters across morphological groups.

CONCLUSION

4D flow MRI-derived hemodynamic parameters, particularly peak velocity, may provide information on local flow acceleration that is not fully captured by morphology-based stenosis grading alone. The findings also suggest that morphology-based grading alone may not fully reflect local hemodynamic burden.