How is the severity of aortic stenosis classified by computed tomography (CT) scan?

CT-Based Severity Classification of Aortic Stenosis

CT severity classification of aortic stenosis primarily relies on aortic valve calcium scoring using the Agatston method, with sex-specific thresholds defining severe AS as ≥2000 Agatston units in men and ≥1200 Agatston units in women, while CT-derived aortic valve area measurements have poor discriminatory performance and should not be used as primary criteria. 1

Calcium Score Classification (Primary CT Method)

The European Society of Cardiology and American Society of Echocardiography provide specific calcium score thresholds for severity assessment using multislice CT with the Agatston method 1:

Severe AS Classification by Calcium Score:

• Men: ≥2000 Agatston units makes severe AS likely; ≥3000 makes it very likely 1
• Women: ≥1200 Agatston units makes severe AS likely; ≥1600 makes it very likely 1

Exclusion of Severe AS:

• Men: <1600 Agatston units makes severe AS unlikely 1
• Women: <800 Agatston units makes severe AS unlikely 1

Critical caveat: A high calcium score confirms severe AS, but a low score makes severe AS highly unlikely. However, intermediate scores create a "grey zone" where calcium scoring alone cannot provide definitive answers and must be integrated with other parameters. 1

CT-Derived Aortic Valve Area (Limited Role)

CT-derived AVA measurements have poor discrimination for AS severity and should not be relied upon as primary diagnostic criteria. 2 The evidence demonstrates significant limitations:

Performance Characteristics:

• Direct planimetry AVA by CT (AVA_CT) <1.2 cm² shows only 85% sensitivity and 26% specificity for severe AS 2
• Using AVA_CT <1.2 cm² threshold produces 27% misclassification rate compared to calcium scoring 2
• AVA_CT <1.0 cm² performs even worse with 42% misclassification 2

Why CT-AVA Fails:

The poor performance stems from measurement variability, flow-dependent changes, and inability to distinguish true-severe from pseudo-severe AS across different hemodynamic patterns. 2 Research shows that 9.8% of high-gradient severe AS, 27.8% of classic low-flow/low-gradient AS, and 14.5% of paradoxical low-flow/low-gradient AS are reclassified as moderate when using CT-derived AVA. 3

Clinical Integration Algorithm

When using CT for AS severity assessment, follow this stepwise approach 1, 4:

1. Obtain non-contrast cardiac CT for calcium scoring using Agatston method 1

2. Apply sex-specific calcium thresholds:

   • High scores (men ≥3000, women ≥1600): Confirms severe AS 1
   • Intermediate scores (men 1600-3000, women 800-1600): Requires integration with echocardiographic parameters 1
   • Low scores (men <1600, women <800): Severe AS highly unlikely 1

3. For discordant cases (low-gradient AS with AVA <1.0 cm² but mean gradient 30-40 mmHg):

   • Calcium scoring becomes particularly valuable 1
   • Integrate with stroke volume index, left ventricular hypertrophy, and reduced longitudinal function 1
   • Consider dobutamine stress echocardiography if calcium score is intermediate 1

4. Avoid using CT-derived AVA as standalone criterion given its poor specificity and high misclassification rates 2

Common Pitfalls to Avoid

• Do not use AVA_CT <1.0 cm² as a threshold for severe AS - this produces 55% misclassification compared to calcium scoring 2
• Do not rely on CT measurements in isolation - calcium scoring must be integrated with echocardiographic hemodynamics, flow status, and clinical presentation 1, 4
• Recognize that calcium scoring has limitations in the intermediate range - it provides definitive answers only at the extremes (very high or very low scores) 1
• Remember sex-specific differences - women have lower calcium score thresholds than men for equivalent AS severity 1