CT-based multi-threshold calcium scoring for aortic valve assessment: comparable predictive performance across HU...

Area of Science:

• Cardiovascular Imaging
• Interventional Cardiology
• Radiology

Background:

• Severe aortic stenosis (AS) requires accurate hemodynamic assessment.
• Computed tomography-derived aortic valve calcium (CT-AVC) scoring is used to evaluate AS severity.
• The optimal Hounsfield unit (HU) threshold for CT-AVC scoring in high-flow/high-gradient AS is debated.

Purpose of the Study:

• To evaluate if higher HU thresholds for CT-AVC scoring offer added hemodynamic or discriminatory value beyond the conventional 130-HU Agatston score.
• To assess the association of varying CT-AVC HU thresholds with transvalvular gradients and aortic valve area in severe AS.
• To determine the discriminatory performance of different HU thresholds for identifying very severe hemodynamic burden.

Main Methods:

• Retrospective analysis of 63 pre-transcatheter aortic valve implantation (TAVI) patients with severe AS.
• CT-AVC quantification at 130, 200, 500, 800, and 1000 HU.
• Correlation and multivariable linear regression to assess associations with peak/mean gradients and aortic valve area.
• Receiver Operating Characteristic (ROC) analysis to compare discrimination of very severe hemodynamic burden.

Main Results:

• CT-AVC at all HU thresholds was independently associated with higher peak and mean gradients.
• No statistically significant associations were found between CT-AVC and aortic valve area.
• Discrimination of very severe hemodynamic burden was modest and comparable across all HU thresholds (AUC 0.65-0.67).

Conclusions:

• In high-flow/high-gradient severe AS, CT-AVC consistently associates with transvalvular gradients across different HU thresholds.
• Higher HU thresholds do not outperform the conventional 130-HU score for hemodynamic stratification.
• Higher HU thresholds should be considered complementary densitometric analyses, not alternative diagnostic cut-offs.