AV fistula Doppler ultrasound: how to measure flow volume and detect stenosis

Doppler ultrasound is the first-line imaging test for the hemodialysis arteriovenous fistula (AVF): it confirms maturation, measures access flow volume (Qa) and finds stenoses before they thrombose the access. This guide sums up the scanning protocol, the flow volume formula, reference values and stenosis criteria, with their sources.

Updated October 10, 2026 · 10-minute read

What AV fistula Doppler is used for

The Spanish GEMAV guideline recommends Doppler ultrasound for every native fistula that is not developing well and makes it the first-line imaging test, in experienced hands, for any suspected significant stenosis; diagnostic fistulography is reserved for inconclusive scans [1,5]. In practice, it is used to:

• Assess maturation before the first cannulation (see the guide on the rule of 6s).
• Run access surveillance with serial Qa measurements.
• Work up dysfunction: difficult cannulation, high pressures, recirculation, prolonged bleeding or a falling Qa.
• Find complications: thrombus, aneurysm, hematoma or steal.

AV fistula ultrasound protocol, step by step

Use a high-frequency linear probe and follow the access from the artery to the heart, first in grayscale and then with color and pulsed-wave Doppler [1,3].

The radial artery distal to a radiocephalic anastomosis often shows reversed flow, because it carries blood from the palmar arch toward the fistula. This is usually an expected finding, not a technical error.

How to measure AV fistula flow volume (Qa) in the brachial artery

Qa is the volume of blood flowing through the access each minute. It is calculated from the time-averaged mean velocity (TAMV, also called TAMEAN) and the vessel area:

Qa (mL/min) = TAMV (cm/s) × π × (d/2)² (cm²) × 60

1. Find the brachial artery in the upper arm, above the elbow, in a straight segment without stenosis. This applies to forearm fistulas too [1,2].
2. In long axis, measure the inner diameter (intima to intima) at the same point where you will measure velocity.
3. Turn on pulsed-wave Doppler. Center the sample volume so it covers 50–70% of the lumen [2].
4. Steer the beam or heel the probe to an insonation angle of 60° or less, and align the angle-correction cursor with the vessel axis [1,2].
5. Record several stable cardiac cycles and use the TAMV, not the peak velocity: using PSV greatly overestimates flow [2].
6. Repeat the measurement and average at least 3 readings [1,2].

Why not measure flow in the fistula vein?

The GEMAV guideline explains it: the fistula vein is tortuous, changes diameter, is easily compressed by the probe, has side branches that take part of the flow, and its flow is turbulent with a very broad spectrum. The brachial artery, by contrast, is straight, has a constant diameter and laminar flow, and an adequate angle is easy to obtain near the elbow [1]. Qa can be measured in the radial artery, but it underestimates flow because part of the fistula's inflow comes from the ulnar artery via the palmar arch [1]. In grafts, which are rigid and have no side branches, it can be measured in the graft itself [2]. Some radiology protocols measure in the outflow vein about 10 cm from the anastomosis [3]; if your unit does, keep the same method so results are comparable.

Keep in mind that the brachial artery already carries some baseline flow without a fistula (roughly 30–120 mL/min before surgery [3]), so the brachial method slightly overestimates the true access flow.

AV fistula flow volume formula: worked example

First row worked out: radius 0.25 cm → area = 3.1416 × 0.25² = 0.196 cm²; 70 × 0.196 × 60 ≈ 825 mL/min. Diameter enters the formula squared, so it is the biggest source of error: measure carefully and always at the same spot.

Normal, low and high AV fistula flow

There is no universal definition of a high-flow AV fistula: thresholds differ between guidelines, and the Qa trend should be read together with the clinical picture (shortness of breath, edema, signs of steal). In surveillance, the most useful comparison is with the same patient's previous measurements, taken with the same technique.

AVF stenosis criteria on ultrasound

The GEMAV guideline defines a significant stenosis by combining morphological and functional criteria [1,5]:

Velocity thresholds are not identical across studies:

• PSV ratio: ≥ 2:1 within the vein and ≥ 3:1 at the anastomosis, where turbulence is normal [6]. Using a ratio ≥ 2, ultrasound detected ≥ 50% stenosis in fistulas with 93% sensitivity but only 60% specificity [7].
• Absolute PSV: > 400 cm/s is the most widely cited threshold [1]; in a study of 780 accesses, a PSV ≥ 500 cm/s predicted ≥ 50% stenosis with 89% sensitivity and a 99% positive predictive value, and in that study the velocity ratio was unreliable [8].
• Diameter reduction: stenoses are often eccentric, so percent narrowing is hard to reproduce; velocity ratios are often more useful [3].

Example: PSV of 130 cm/s 2 cm upstream and 460 cm/s at the stenosis → ratio 3.5; residual lumen 1.8 mm and brachial Qa 470 mL/min. If the lumen is also reduced by more than 50%, it meets the criteria for significant stenosis, and GEMAV recommends treating it without delay with angioplasty or surgery because of its high thrombosis risk [1].

AV fistula waveforms

RI is very sensitive but not very specific: a normal RI does not guarantee a well-functioning fistula [2]. The insonation angle does not affect RI, but it does affect velocities and Qa [1]. If you are unsure which vessel is which, practice with the artery or vein lesson.

Aliasing and other Doppler pitfalls

• Aliasing: occurs when velocity exceeds the Nyquist limit (half the pulse repetition frequency, PRF). On color it shows as a mosaic with wrapped colors; on pulsed-wave the peak is cut off and wraps to the other side of the baseline. Raise the scale (PRF), lower the baseline or use a lower-frequency probe. Focal aliasing at a stenosis is a useful clue, but only a clue [1].
• Insonation angle: velocity depends on the cosine of the angle. From the Doppler equation itself, a 5° angle-correction error gives a velocity error of about 9% at 45°, 15% at 60° and 24% at 70°. That is why you should not go above 60° [1].
• Vessel perpendicular to the beam: at 90° there is no Doppler signal. Tilt the probe or steer the color box.
• Probe pressure: it flattens the vein and mimics a stenosis or a low Qa. Use plenty of gel and rest your hand.
• Color gain: too high "bleeds" outside the vessel; too low leaves gaps that look like thrombus. Adjust it until color fills the lumen without spilling over.
• Diameter: measuring on the adventitia, or at a different point from where you measure velocity, changes Qa a lot (see the worked example).

Educational content. It does not replace supervised hands-on training or your unit's protocols.

References

1. Ibeas J, Roca-Tey R, Vallespín J, Moreno T, et al. Spanish Clinical Guidelines on Vascular Access for Haemodialysis (GEMAV). Nefrología. 2017;37(Suppl 1):1-191. doi.org
2. Aragoncillo I, Caldés S. Ecografía Doppler en el acceso vascular (in Spanish). Nefrología al día. nefrologiaaldia.org
3. Sharbidre KG, Alexander LF, Varma RK, et al. Hemodialysis access: US for preprocedural mapping and evaluation of maturity and access dysfunction. RadioGraphics. 2024;44(1):e230053. doi.org
4. Robbin ML, Chamberlain NE, Lockhart ME, et al. Hemodialysis arteriovenous fistula maturity: US evaluation. Radiology. 2002;225(1):59-64. doi.org
5. Roca-Tey R, Ibeas J, Moreno T, Gruss E, et al. Dialysis arteriovenous access monitoring and surveillance according to the 2017 Spanish Guidelines. J Vasc Access. 2018;19(5):422-429. doi.org
6. Saati A, Puffenberger D, Kirksey L, Fendrikova-Mahlay N. The role of hemodialysis access duplex ultrasound for evaluation of patency and access surveillance. Cardiovasc Diagn Ther. 2023;13(1):190-195. doi.org
7. Vardza Raju A, Kyin May K, Htet Zaw M, et al. Reliability of ultrasound duplex for detection of hemodynamically significant stenosis in hemodialysis access. Ann Vasc Dis. 2013;6(1):57-61. doi.org
8. Wo K, Morrison BJ, Harada RN. Developing duplex ultrasound criteria for diagnosis of arteriovenous fistula stenosis. Ann Vasc Surg. 2017;38:99-104. doi.org