AS severity criteria: the four parameters and how to read them
ACC/AHA VHD 2020 evaluates AS severity by integrating AVA, Vmax, mean gradient, DVI, and flow state. Understanding what each threshold means — and why the parameters can disagree — is the foundation for correct severity classification.
ACC/AHA VHD 2020 evaluates AS severity by integrating AVA, Vmax, mean gradient, DVI, and flow state. Understanding what each threshold means — and why the parameters can disagree — is the foundation for correct severity classification. The AS Severity Tool classifies the hemodynamic pattern from the values entered. This page explains what each threshold means, why discordance occurs, and how low-flow states change interpretation.
Key takeaway
Severe AS is usually identified by AVA < 1.0 cm², Vmax ≥ 4.0 m/s, and mean gradient ≥ 40 mmHg. However, Vmax and mean gradient are highly flow-dependent. In low-flow states, truly severe AS may present with a gradient below 40 mmHg. DVI is useful as a cross-check because it avoids LVOT diameter measurement, but it should be interpreted with AVA, Vmax, mean gradient, SVI, LVEF, and valve calcification.
Key points
- AVA < 1.0 cm² suggests severe AS, but AVA is calculated by the continuity equation and is highly sensitive to LVOT diameter measurement error.
- Vmax ≥ 4.0 m/s and mean gradient ≥ 40 mmHg strongly support severe AS when flow is normal.
- In low-flow states, Vmax and mean gradient may fail to reach the severe range even when AS is truly severe.
- DVI < 0.25 supports severe AS and is useful when AVA and gradient disagree or LVOT diameter measurement is uncertain.
- SVI < 35 mL/m² defines low flow and changes how gradients should be interpreted.
- When AVA and gradient disagree, identify the flow state first — do not simply choose one number over the other.
- For perioperative planning, the AS phenotype matters: high-gradient AS, classical LFLG, paradoxical LFLG, or discordant AS.
When to read this
Read this when an echo report includes AS parameters and you want to understand what AVA, Vmax, mean gradient, DVI, and SVI mean — especially before using the AS Severity Tool.
The four core parameters for AS severity
| Parameter | Mild | Moderate | Severe | Key limitation / flow dependence |
|---|---|---|---|---|
| Aortic valve area (AVA) | > 1.5 cm² | 1.0–1.5 cm² | < 1.0 cm² | Calculated by the continuity equation; highly sensitive to LVOT diameter measurement error |
| Peak aortic velocity (Vmax) | < 3.0 m/s | 3.0–3.9 m/s | ≥ 4.0 m/s | Highly flow-dependent; low stroke volume may reduce Vmax even in severe AS |
| Mean pressure gradient | < 20 mmHg | 20–39 mmHg | ≥ 40 mmHg | Proportional to velocity squared; falls substantially in low-flow states |
| Dimensionless velocity index (DVI) | > 0.35 | 0.25–0.35 | < 0.25 | Ratio of LVOT VTI to aortic valve VTI; avoids LVOT diameter error but must be interpreted with other parameters |
| Stroke volume index (SVI) | — | — | — | SVI < 35 mL/m² defines low flow; it does not grade severity by itself but changes how gradients are interpreted |
Low-flow patterns: when standard thresholds are harder to interpret
| Pattern | SVI | LVEF | Key finding | Next check |
|---|---|---|---|---|
| Normal-flow high-gradient | ≥ 35 mL/m² | Any | AVA < 1.0 cm² and mean gradient ≥ 40 mmHg. Concordant severe AS | Proceed to intervention assessment for severe AS |
| Classical LFLG (low-flow low-gradient, reduced EF) | < 35 mL/m² | < 50% | AVA < 1.0 cm² with gradient < 40 mmHg. Reduced LV contractility; distinguish true severe AS from pseudo-severe AS | Confirm severity with low-dose dobutamine stress echo and/or CT aortic valve calcium scoring |
| Paradoxical LFLG (low-flow low-gradient, preserved EF) | < 35 mL/m² | ≥ 50% | Small hypertrophied LV with low SVI despite preserved LVEF | CT aortic valve calcium scoring is useful; DSE may be considered in selected cases |
| Discordant AS (normal flow, severe AVA, non-severe gradient) | ≥ 35 mL/m² | Any | AVA < 1.0 cm² but gradient < 40 mmHg despite preserved flow | Recheck LVOT diameter, assess DVI, and consider additional imaging when needed |
DVI as a cross-check
DVI is calculated by dividing LVOT VTI by aortic valve VTI. Because it does not use LVOT diameter, it is less affected by one of the most common sources of AVA calculation error.
A DVI below 0.25 supports severe AS, especially when AVA and gradient are discordant or when LVOT diameter measurement is uncertain. However, DVI should not be used as a standalone decision-maker. Interpret it together with AVA, Vmax, mean gradient, SVI, LVEF, and valve calcification.
Perioperative relevance: what these thresholds mean at the bedside
- High-gradient severe AS — Vmax ≥ 4.0 m/s or mean gradient ≥ 40 mmHg — indicates fixed obstruction to LV outflow. During induction and intraoperative management, maintaining preload, afterload, sinus rhythm, and coronary perfusion pressure is essential.
- Suspected LFLG AS before non-cardiac surgery should prompt cardiology review when feasible. These patients are vulnerable to hemodynamic instability under anesthesia; severity and intervention eligibility should be clarified before surgery when possible.
- A preoperative report showing AVA 0.9 cm² with mean gradient 28 mmHg should not be labeled moderate AS until SVI and LVEF have been assessed.
- For perioperative planning, the AS phenotype matters more than the severity grade alone. The AS Severity Tool returns patterns such as high-gradient AS, classical LFLG, paradoxical LFLG, or discordant AS.
The most common perioperative misread
A mean gradient of 28–35 mmHg is often interpreted as moderate AS. However, in a patient with SVI < 35 mL/m², a truly severe valve may generate only a modest gradient. If AVA is in the severe range but gradient is not, check SVI and LVEF before downgrading severity. Do not classify AS as moderate based on gradient alone.
- Otto CM, et al. 2020 ACC/AHA Guideline for the Management of Patients With Valvular Heart Disease. J Am Coll Cardiol. 2021;77(4):e25-e197.
- Baumgartner H, et al. Echocardiographic Assessment of Valve Stenosis: EAE/ASE Recommendations for Clinical Practice. J Am Soc Echocardiogr. 2009;22(1):1-23.
Related valve learning