The single-number temptation

Ask ten vascular labs how they grade an internal carotid artery (ICA) stenosis and you will often get ten versions of the same answer: report the peak systolic velocity and stop there. In many departments, a max PSV above a fixed cut-off is treated as the whole diagnosis. It is fast, it is familiar, and it is reproducible on a good day. It is also the single most common reason a carotid report is challenged in a multidisciplinary meeting.

Grading ICA stenosis well is not a one-parameter exercise. The velocity you measure sits on top of a set of assumptions – about which reference standard your threshold came from, what the contralateral side is doing, what the heart is doing, and how the plaque behaves. Get those wrong and a perfect PSV number still leads you to the wrong conclusion. This article compares the grading systems used around the world and then works through the variables that decide whether your number means what you think it means.

The world does not use one grading system

Two measurement conventions underpin almost every national guideline, and they do not measure the same thing.

NASCET measures the residual lumen at the tightest point against the diameter of the normal distal ICA, well beyond the bulb. ECST measures the same stenosis against an estimated original bulb diameter. Because the bulb is wider than the distal ICA, the same lesion reads as a larger percentage under ECST. The practical consequence is a shift of roughly one severity band: a lesion that is about 50% by NASCET is closer to 65% by ECST, and a 70% NASCET lesion corresponds to roughly 80-85% by ECST. If you quote a percentage without naming the method, you have not communicated a stenosis – you have communicated an ambiguity.

On top of that, labs apply different velocity thresholds. The common NASCET-aligned duplex criteria use a PSV above roughly 125 cm/s for a 50% stenosis and above roughly 230 cm/s for a 70% stenosis, supported by end-diastolic velocity and the ICA/CCA ratio. Other published criteria and individual laboratories use their own numbers, usually derived by correlating their velocities against angiography. None of these thresholds are universal constants. They are local calibration points, and they only hold for the technique they were derived with.

Where velocity starts to lie

Even inside one system, PSV is fragile. The value depends on the Doppler angle (keep it at 60 degrees or less), on sample volume placement in the tightest jet rather than in the pre-stenotic segment, and on the cardiac output state of the patient. A well-known trap is the near-occlusion or pseudo-occlusion: at the very highest grades of stenosis the jet can collapse, and a lesion that is nearly occluded may present with velocities that look only moderate. Chasing the highest number you can find across the vessel is not rigor; it is a route to over-calling.

This is why the sensible units report a set of parameters rather than one:

• Peak systolic velocity in the ICA
• End-diastolic velocity in the ICA
• The ICA/CCA peak systolic ratio
• The spectral waveform: broadening, post-stenotic turbulence, and the shape of the jet
• Plaque features – echogenicity, surface irregularity, ulceration, and how the lesion evolves over time

The velocity tells you the haemodynamic consequence. The waveform and the plaque tell you the character of the disease. Neither replaces the other.

Contralateral disease changes the picture

The contralateral ICA is not background information. A severe stenosis or an occlusion on the opposite side alters the flow environment on the side you are scanning, through compensatory flow and altered pressure gradients. Interpreting one carotid in isolation throws away the context that makes the number meaningful, and it also hides the clinical reality: the patient’s stroke risk is a function of both sides plus the intracranial circulation and the collateral supply, not of a single velocity measurement.

The same logic applies to the vertebral arteries. Posterior circulation findings, subclavian steal physiology, and vertebrobasilar contribution to the overall cerebral supply all belong in the assessment. A carotid report that never looks beyond the bifurcation is only partly finished.

The patient in atrial fibrillation: which PSV do you use?

This is the question that exposes the weakness of single-parameter grading most clearly. In atrial fibrillation, stroke volume varies beat to beat, so measured PSV swings from one cardiac cycle to the next. There is no single “true” PSV to report, and picking the highest beat inflates the grade while picking a low beat under-calls it.

The practical answer is to stop treating the measurement as a single value. Average across several consecutive cardiac cycles, state that the patient is in an arrhythmia, and lean harder on the parameters that are more stable and more physiological in this setting – end-diastolic velocity, the ICA/CCA ratio, and the spectral waveform – while recognising that all velocity criteria were validated in patients in sinus rhythm and degrade in arrhythmia. In a patient whose cardiac output is fluctuating, a velocity-only diagnosis is the least trustworthy reading you can produce. The honest report flags the uncertainty rather than hiding it behind a number.

A practical multi-parameter framework

Pulling this together, a defensible grading routine looks like this:

1. Name the method. State whether the percentage is NASCET or ECST referenced.
2. Measure across several cycles and average, especially in arrhythmia or high-output states.
3. Report the parameter set, not one value: PSV, EDV, ICA/CCA ratio, and waveform description.
4. Characterise the plaque, and compare with any prior study.
5. Assess the contralateral ICA, the vertebrals, and what you can infer about collateral supply.
6. Sanity-check the number against the waveform. If they disagree, the waveform usually wins the argument.
7. Put the finding in clinical context – symptoms, timing, and the treatment threshold that decision-making actually uses.

Every one of these steps is a judgement call, and judgement is what separates a grading service from a measurement service. Laboratories that train to a single cut-off produce reports that are easy to generate and easy to overturn.

Learn the full assessment, not the shortcut

The Carotid and Vertebral Arteries master course, https://abcvascular.com/courses/carotid-and-vertebral-arteries-new/ is built for exactly this problem. It walks through the world’s grading systems side by side, the velocity and waveform parameters that make up a complete reading, the pitfalls of near-occlusion, contralateral disease, and arrhythmia, and how to turn all of it into a report you can defend. It is accredited, self-paced, and written for practising professionals who need scanning skill they can apply on the next list, not a threshold they have to memorise.

If your current protocol begins and ends with max PSV, this is the course that shows you what you are missing – and why the answer to “is one parameter enough?” is, almost always, no.