Chapter 4 — Doppler Principles and Quantitative Doppler

📚 Guideline basis

ASE 2019 comprehensive TTE examination (JASE 2019;32:1–64); ASE/EACVI 2017 recommendations on non-invasive assessment of valve stenosis; ASE 2017 native valvular regurgitation (JASE 2017;30:303–71); ASE 2025 diastolic function update for acquisition specifications.

Mechanisms

The Doppler equation

Δf = (2 f₀ v cosθ) / c

Rearranged for the measured velocity:

v = (Δf · c) / (2 f₀ cosθ)

where Δf = Doppler shift (Hz), f₀ = transmitted frequency (Hz), v = blood velocity (m/s), θ = angle between beam and flow, c = 1540 m/s.

The cosθ term is the dominant clinical error source. Because cos 0° = 1, any misalignment underestimates velocity:

Insonation angle	cos θ	Velocity underestimate
0°	1.00	0%
10°	0.985	1.5%
20°	0.940	6%
30°	0.866	13%
45°	0.707	29%
60°	0.500	50%

Angles below 20° are conventionally accepted because the error remains under 6%. Angle correction is never used in cardiac Doppler — unlike vascular ultrasound, the true jet direction in three dimensions is unknown, and applying a correction converts a known underestimate into an unknown error in either direction.

The Bernoulli relation

The full Bernoulli equation relates a pressure gradient across a restriction to convective acceleration, flow acceleration, and viscous friction. The clinical simplified Bernoulli equation discards the latter two terms and the proximal velocity:

ΔP = 4v²

The discarded proximal velocity term matters. The full form is:

ΔP = 4(v₂² - v₁²)

Use the expanded form when v₁ > 1.5 m/s. This is not an academic point: in a patient with aortic stenosis and a high LVOT velocity (small ventricle, high output, or serial obstruction), ignoring v₁ overestimates the transaortic gradient. At v₁ = 1.5 m/s and v₂ = 4.0 m/s, the simplified equation gives 64 mmHg while the expanded gives 55 mmHg — a 14% overestimate that can misclassify severity.

The simplified equation is also invalid where viscous losses dominate: long, tunnel-like obstructions, and prosthetic valves with small effective orifice areas, where it systematically overestimates the gradient relative to catheter measurement (pressure recovery, see Chapter 18).

Nyquist limit and aliasing

Pulsed-wave Doppler samples intermittently; the maximum unambiguously resolvable frequency shift is half the pulse repetition frequency (PRF):

v_max = (c · PRF) / (4 f₀ cosθ)      and      PRF_max = c / (2D)

Consequences: aliasing worsens with greater depth (lower PRF), higher transmit frequency, and higher velocity. Remedies, in order of preference: shift the baseline; increase the velocity scale; reduce depth; lower transmit frequency; use high-PRF mode (accepting range ambiguity); switch to continuous-wave.

Modality selection

Modality	Measures	Range resolution	Velocity ceiling	Cardinal ICU use
Pulsed-wave (PW)	Velocity at a specified depth	Yes — sample volume	Nyquist-limited	LVOT VTI, mitral inflow, pulmonary vein, RVOT VTI
Continuous-wave (CW)	Peak velocity along the entire beam	No — range ambiguous	Effectively unlimited	TR jet, AS gradient, MR jet, prosthetic gradients
Colour flow	2D velocity map (mean, autocorrelation)	Yes	Nyquist-limited	Jet detection, alignment guide, PISA, shunt detection
Tissue Doppler (TDI)	Myocardial velocity (low velocity, high amplitude)	Yes	Not usually limiting	e′, s′, a′ velocities; TAPSE surrogate
Colour M-mode	Velocity vs depth vs time	Yes	Nyquist-limited	Propagation velocity Vp

Rule: colour Doppler is a locator, not a quantifier. Colour jet area is dependent on gain, Nyquist setting, machine, driving pressure, and chamber compliance, and it must not be used to grade regurgitation severity in the ICU (or anywhere).

The velocity–time integral

The VTI is the integral of the velocity envelope over one ejection period, with units of length (cm) — it is the distance a column of blood travels per beat, the "stroke distance":

VTI = ∫ v(t) dt   over the ejection period

Applied to a flow cross-section, it yields volume — the foundation of every quantitative calculation in Chapters 11, 18 and 20:

SV = CSA × VTI

The equation embeds three assumptions, each of which fails in identifiable ICU circumstances:

Assumption	Fails when	Effect
Cross-section is circular	LVOT is elliptical (present in most patients; CT/3D show the LVOT is ovoid)	2D method systematically underestimates CSA and therefore SV, by ~10–20% vs 3D/CT
Flow profile is flat (plug flow)	Severe LVOT obstruction, dynamic obstruction, sub-aortic membrane	Overestimates mean velocity
CSA and VTI measured at the same level	Sample volume placed at the annulus while diameter measured 1 cm below, or vice versa	Systematic mismatch; error can exceed 20%

The dominant practical rule: measure diameter and velocity at the same anatomical level, and, if serially monitoring the same patient, hold the diameter constant and track only the VTI. Because CSA does not change over hours, changes in VTI alone track changes in stroke volume, and the largest error source is eliminated.

Acquisition standards

Signal	View	Sample	Sweep speed	Averaging
LVOT VTI	Apical 5-chamber or 3-chamber	PW, 3–5 mm, 0.5–1.0 cm below annulus, at the level the diameter was measured	100 mm/s	≥3 beats sinus; ≥5 (ideally 10) in AF
Mitral inflow E/A	Apical 4-chamber	PW, 1–3 mm, at leaflet tips	100 mm/s	≥3 beats
Mitral annular e′	Apical 4-chamber, TDI preset	PW, 5–10 mm, septal and lateral annulus	100 mm/s	≥3 beats
TR peak velocity	Any window aligning the jet (RV-focused A4C, PSAX, subcostal)	CW, multiple windows	50–100 mm/s	Averaged over the respiratory cycle
Pulmonary vein S/D	Apical 4-chamber, anterior tilt	PW, 3–5 mm, 5–10 mm into right upper PV	100 mm/s	≥3 beats
IVRT	Apical long axis / 5-chamber	CW through LVOT capturing AV closure and MV opening	100 mm/s	≥3 beats

Low wall filter (100–200 Hz) and low gain are specified for spectral Doppler; over-gained spectra produce "bearding" or "feathering" that inflates measured peak velocity.

Signal quality: what to reject

A Doppler measurement must be discarded, not estimated, when:

The envelope is incomplete or the modal velocity is indistinct (common in TR jets — use agitated saline or an ultrasound enhancing agent to complete the envelope rather than guessing).
Spectral broadening obscures the modal velocity (measure the dense modal envelope, not the outer spray).
Alignment could not be verified from more than one window for a CW jet.
Beat-to-beat variability exceeds ~10% in sinus rhythm — this indicates respiratory or positional instability, not physiological variation, unless it is being deliberately measured (see respiratory variation of LVOT VTI, Chapter 12).

ICU-specific limitations

Confounder	Effect on Doppler	Mitigation
Positive-pressure ventilation	Cyclical changes in preload alter VTI beat-to-beat by design	Average across a full respiratory cycle, or record the cycle and analyse deliberately
Atrial fibrillation	Beat-to-beat variability in filling time and stroke volume	Average ≥5–10 consecutive beats; alternatively, use beats with matched preceding R-R intervals
Tachycardia	E and A fusion; shortened ejection; degraded time-interval measurement	Report E/A as unmeasurable if fused; use post-extrasystolic beats where E and A separate
High-dose vasopressors	Alters afterload, hence jet velocities and gradients	Record the dose with the study; do not compare serial studies across large dose changes without noting it
Mechanical support (IABP, Impella, VA-ECMO)	Non-physiological flow; VTI no longer equals total systemic flow	Interpret against the circuit — see Chapter 35

🛑 Critical pitfall: Grading a regurgitant lesion from colour jet area in a hypotensive patient. Low systemic pressure reduces the driving gradient across a mitral regurgitant orifice, shrinking the colour jet and causing severe acute MR to look moderate. Severity must be judged by mechanism, vena contracta, pulmonary venous flow reversal, and the clinical picture.

🛑 Critical pitfall: Applying ΔP = 4v² to a prosthetic aortic valve without accounting for pressure recovery and the small proximal-to-distal area ratio. Doppler routinely reports higher gradients than simultaneous catheterisation across mechanical bileaflet valves; the dimensionless index and effective orifice area are more reliable (Chapter 23).

💡 Clinical pearl: Interrogate every CW jet from at least two windows and take the highest velocity. Doppler underestimates but never overestimates by misalignment, so the highest well-aligned signal is the most accurate one.
💡 Clinical pearl: If a TR envelope is incomplete, agitated saline contrast will usually complete it. An incomplete envelope traced to its brightest edge produces a fabricated pulmonary artery systolic pressure.
💡 Clinical pearl: When serially monitoring cardiac output, freeze and store the parasternal LVOT diameter image once, then re-use that diameter for every subsequent study on that patient. Re-measuring it each time introduces variance that swamps the physiological signal you are trying to detect.

References

Mitchell C, Rahko PS, Blauwet LA, et al. Guidelines for performing a comprehensive transthoracic echocardiographic examination in adults. J Am Soc Echocardiogr 2019;32:1–64.
Baumgartner H, Hung J, Bermejo J, et al. Recommendations on the echocardiographic assessment of aortic valve stenosis: a focused update from the EACVI and ASE. J Am Soc Echocardiogr 2017;30:372–92.
Zoghbi WA, Adams D, Bonow RO, et al. Recommendations for noninvasive evaluation of native valvular regurgitation. J Am Soc Echocardiogr 2017;30:303–71.
Nagueh SF, Sanborn DY, Oh JK, et al. Recommendations for the evaluation of LV diastolic function by echocardiography and for HFpEF diagnosis: an update from the ASE. J Am Soc Echocardiogr 2025;38:537–69.