Doppler echocardiography is the first-line standard for aortic stenosis (AS) severity — but up to 40% of resting echocardiographic assessments are discordant, creating clinical uncertainty about true disease severity. CT-AVC has emerged as a validated, load-independent alternative that directly quantifies the calcium burden driving the stenosis rather than measuring hemodynamic consequences.
Key reference: Pawade T, Sheth T, Guzzetti E, Dweck MR, Clavel MA. Why and How to Measure Aortic Valve Calcification in Patients With Aortic Stenosis. JACC: Cardiovascular Imaging. 2019;12(9):1835–48. DOI: 10.1016/j.jcmg.2019.01.045
1 Why measure aortic valve calcification?
Aortic stenosis progresses through a vicious cycle: valve calcification → mechanical stress → endothelial injury → osteoblast-like cell transformation → more calcification. Critically, the baseline calcium burden is the single most consistent predictor of disease progression, regardless of how it is measured.
This explains why CT-AVC is useful:
- Load-independent — unlike Doppler gradients and velocities, CT-AVC is not affected by low flow, low cardiac output, or reduced EF
- Anatomically validated — CT-AVC correlates excellently with actual calcium weight of explanted valves (r = 0.96, r² = 0.98; p < 0.001)
- Complementary prognostic information — CT-AVC independently predicts mortality and adverse events beyond standard echocardiographic parameters
When to use CT-AVC:
- Echo measurements are discordant (e.g., low AVA with low gradient or low flow)
- True severity is in doubt, and stress echo is inconclusive or not feasible
- Integrated into pre-TAVR CT as an add-on to the standard planning protocol
2 Scan acquisition protocol
CT-AVC uses the same Agatston method originally developed for coronary calcium scoring. Strict acquisition parameters are required:
| Parameter | Specification |
|---|---|
| Gating | ECG-gated; reconstruction at 60–80% of the RR interval (diastole) |
| kV | 120–140 kV |
| mAs | 30–80 mAs according to patient body weight |
| Acquisition | Spiral or volumetric |
| Pitch | 0.15–0.25 (scanner-dependent) |
| Slice thickness | 3 mm |
| Matrix | 512 × 512 |
| HU threshold | 130 HU for calcium detection |
Radiation dose for CT-AVC acquisition is approximately 1 mSv — modest and should not be a concern in the TAVR-eligible population. Pre-TAVR CT offers an efficient opportunity to acquire CT-AVC as an add-on at minimal incremental dose.
Contrast-enhanced CT: Current evidence favors noncontrast acquisitions for CT-AVC measurement. Contrast-enhanced CT introduces variability because opacification alters apparent calcium density; the threshold has not been standardized across scanners and studies.
3 Agatston scoring method
The Agatston score applies a density-weighted area method on axial 3 mm slices:
- Every lesion ≥ 130 HU and ≥ 1 mm² in the aortic valve is identified
- A density weighting factor is assigned based on peak attenuation in the lesion:
| Peak attenuation (HU) | Weighting factor |
|---|---|
| 130–199 | 1 |
| 200–299 | 2 |
| 300–399 | 3 |
| ≥ 400 | 4 |
- Score = weighting factor × lesion area, summed across all lesions on all slices
Reproducibility: Intraobserver variability 2.6 ± 2.8%; interobserver variability 4.3 ± 4.2% to 6.6 ± 7.0% — excellent for clinical use.
Calcium volume score (pixels with HU > 130, no density weighting) is an alternative with improved interscan reproducibility, though the vast majority of CT-AVC clinical data uses the Agatston score.
4 Identifying the aortic valve region
Quantification must include both valve leaflet and annular calcification, on axial 3 mm slices from the LVOT to the aortic direction. Always exclude:
| Structure | Why it must be excluded |
|---|---|
| LV outflow tract (LVOT) | Calcification may be continuous with aortic valve but is not valvular |
| Aortic sinuses / root | Aortic wall calcium is not valve calcium |
| Right coronary artery ostium | On axial slices, the aortic root and ostial RCA may appear on the same slice as the left coronary cusp — this is not aortic valve calcium |
| Mitral annulus | Posterior mitral annular calcium extends toward but is not part of the aortic valve |
| Coronary arteries | Calcified coronaries near the annulus |
Practical tip: Start measuring in the center of the valve and track calcium in continuity on adjacent axial slices in both the caudal and cranial directions. Use multiplanar reconstructions to confirm the valve level.
LVOT and mitral annulus boundaries can be challenging when there is continuous calcification between structures. In practice, the vast majority of patients with this pattern have very severe aortic valve calcification, and clinical decision-making is rarely hampered.
5 Common pitfalls
| Pitfall | How to avoid |
|---|---|
| Including ostial RCA calcium | Right and left aortic leaflets are never visualized on the same axial slice; any calcium near the right coronary ostium on an axial view is not valvular |
| Including aortic sinus calcification | Use coronal and sagittal reformats to confirm continuity with the leaflet |
| Using “en face” view for scoring | En face reconstruction reduces CT-AVC by up to 37% in an inconsistent and unpredictable manner (0–50% variation) — en face is useful for identifying valve region and leaflet anatomy but must never be used for quantification |
| Including mitral annular calcium | More common in women; if continuous with aortic valve, careful assessment of the extremity of the calcification and slice orientation helps determine boundaries |
| Contrast-enhanced CT | Not yet standardized for AVC measurement; use noncontrast acquisitions |
6 Sex-specific thresholds for severe AS
CT-AVC thresholds differ meaningfully by sex, reflecting that women develop hemodynamically severe AS with less calcium — likely because they have more fibrotic (noncalcific) leaflet thickening contributing to obstruction.
Validated sex-specific thresholds (Agatston units):
| Metric | Women | Men | Sensitivity | Specificity |
|---|---|---|---|---|
| AVC score (AU) — primary threshold | ≥ 1,274 | ≥ 2,065 | 89% / 80% | 81% / 82% |
| AVC score — validated multicenter | ≥ 1,377 | ≥ 2,062 | 87% / 80% | 84% / 82% |
| Rounded practical thresholds | > 1,300 | > 2,000 | — | — |
| AVC density (AU/cm²) | ≥ 292 (primary) / ≥ 420 (validated) | ≥ 476 / ≥ 527 | 96%/88% | 75%/88% |
These thresholds were derived in 646 patients and subsequently validated in an independent multicenter cohort of 918 patients across 9 centers in Europe and North America, representing over 1,168 patients with concordant echocardiographic measurements.
AVC density (AVC score ÷ aortic annulus area by echocardiography) is particularly useful in patients with very small or very large aortic annuli, where body-size differences make the raw score harder to interpret.
7 Interpreting the result
Severe CT-AVC (above sex-specific threshold) + severe CT-AVC density → aortic stenosis is severe with high confidence, even when echo is discordant.
Non-severe CT-AVC + non-severe density → stenosis will generally not be severe, although careful evaluation is required — especially in young women with bicuspid valves, who may have predominantly fibrotic (noncalcific) obstruction.
Important limitation: CT-AVC ignores noncalcific leaflet thickening. In a minority of patients (particularly younger women with bicuspid morphology), fibrosis rather than calcification is the dominant pathologic process; a low CT-AVC score does not exclude severe stenosis in these cases.
8 Prognostic value
CT-AVC is a powerful independent predictor of adverse outcomes, outperforming standard echocardiographic parameters:
- Severe AVC → 1.75× higher all-cause mortality (HR 1.75; 95% CI 1.04–2.92; p = 0.03) under medical treatment
- Severe AVC density → 2.44× higher mortality (HR 2.44; 95% CI 1.37–4.37; p = 0.002)
- Event-free survival (valve-related events): HR 1.62 (p = 0.04) for severe AVC; HR 3.90 (p < 0.001) for severe AVC density in discordant echo patients
- Adding CT-AVC to a comprehensive model provided a net reclassification index of 12.5% (p = 0.04) for predicting 1-year mortality
Disease progression: Severe CT-AVC is associated with a 3-fold faster rate of mean gradient progression — making CT-AVC useful for timing follow-up intervals. Patients with close-to-severe CT-AVC may warrant more frequent clinic follow-up.
9 When to use CT-AVC in clinical practice
CT-AVC is now incorporated into the 2017 ESC/EACTS guidelines for valvular heart disease as a complementary assessment of AS severity. Its crucial advantage is independence from hemodynamic status.
Most common clinical scenario:
A patient with AVA in the severe range (≤ 1 cm² or indexed ≤ 0.6 cm²/m²) but mean gradient or peak velocity suggesting only moderate disease (< 40 mm Hg, < 4 m/s). In this low-flow, low-gradient pattern:
- Stress echocardiography is the standard adjunct — but may be inconclusive in low-flow states
- CT-AVC offers a faster, feasible, and often conclusive alternative
- Severe CT-AVC + severe CT-AVC density → conclude severe AS and consider intervention
- Nonsevere CT-AVC + nonsevere density → stenosis likely not truly severe; manage medically with follow-up
CT-AVC should now be considered in the diagnostic evaluation of patients with discordant echocardiographic markers of aortic stenosis, with further work needed to establish its utility in asymptomatic patients and those with moderate-to-severe disease.