The Ankle-Brachial Index: Why the Higher Arm Wins

The ankle-brachial index calculation is a single division between blood pressure readings: the systolic pressure measured at an ankle, divided by the higher of the systolic pressures measured at the two arms.

ABI = P_ankle ÷ max(P_arm left, P_arm right)

Every quantity in the formula is a systolic pressure in the same unit — mmHg on virtually every cuff — so the units cancel and the index itself carries none. Worked with the example from the ABI calculator: a Doppler exam records 120 mmHg at the ankle, 125 mmHg at the left arm and 122 mmHg at the right. The denominator is the higher arm — 125, not 122, and not the average of the two — so ABI = 120 ÷ 125 = 0.96.

That is the whole computation, but two conventions are folded into it, and both repay understanding. Each leg gets its own index — two ABIs per person, one per ankle — while both arms compete for a single shared denominator, and the higher one always wins. What any particular value implies about a particular set of arteries is a clinical question: the interpretation bands live in guideline documents such as the American Heart Association's 2012 scientific statement on the ABI and NICE's peripheral arterial disease guidance, and applying them belongs to a clinician looking at the whole record. This guide stays on the arithmetic side of that line — what feeds the ratio, why the higher arm was chosen, and how far the number drifts when the convention is ignored.

What actually gets measured

An ABI exam produces more raw numbers than the formula suggests. Standard protocols take systolic readings at both arms, then Doppler readings from two arteries at each ankle, and the defining publications specify which ankle reading enters the numerator for which purpose. The formula then compresses all of that into one ratio per leg. The calculator mirrors the compression step exactly: it asks for one ankle pressure and both arm pressures, applies the max-arm rule to the denominator, and returns a single leg's index. Which ankle artery's number you type in follows from the protocol you are working under, not from the tool.

Why the higher arm wins the denominator

The arm reading is not in the formula for its own sake. It stands in for central systemic pressure — the reference the ankle pressure is being compared against. In most people the two arms read within a few mmHg of each other and either would serve. But when one arm reads noticeably lower, the usual suspicion is a narrowing in the artery supplying that arm — a subclavian narrowing can suppress one arm's reading — which makes that number an underestimate of the central pressure it is supposed to represent. The max-arm rule handles this without asking anyone to adjudicate on the spot: take the higher arm, discard the lower one entirely.

The rule also has a built-in arithmetic property. Dividing by the largest available candidate produces the smallest possible quotient, so taking the higher arm can only ever make the index lower, never higher. Any other denominator choice errs exclusively in the flattering direction. The tool's worked example states the intent plainly: the convention exists to bias the test toward catching disease, not missing it.

How much the convention is worth

With the arms 125 − 122 = 3 mmHg apart, as in the opening example, the stakes look small: dividing by the lower arm instead gives 120 ÷ 122 = 0.98 against the correct 0.96 — a drift of 0.98 − 0.96 = 0.02. Widen the inter-arm gap and the stakes grow with it. Take an ankle at 110 mmHg with arms at 138 and 118 mmHg — a disagreement of 138 − 118 = 20 mmHg:

Three defensible-sounding conventions, three different answers, spanning 0.93 − 0.80 = 0.13 — six and a half times the 0.02 drift seen when the arms nearly agreed. Averaging looks like the diplomatic option, but if the lower reading is the artifact of a narrowed artery, the average is contaminated by exactly the number the protocol set out to exclude.

Notice what makes the convention quietly elegant: it matters most precisely when the arms disagree most, and a large inter-arm difference is itself the situation in which one arm's reading is most suspect. The rule is self-triggering. When the arms agree, all three conventions collapse toward the same answer and nothing was lost; when they diverge, the rule acts, and always in the cautious direction.

Two legs, one reference

The denominator is shared; the numerator is not. Staying with arms of 125 and 122 mmHg, suppose the right ankle reads 132 mmHg where the left read 120. The right leg's index is 132 ÷ 125 = 1.056, reported as 1.06; the left leg's stays at 0.96. A value above 1 records exactly what the arithmetic says — that ankle's pressure came out higher than the higher arm's — and nothing more mysterious than that. Two legs, two indexes, one systemic reference: this is why one leg's result says nothing about the other leg, and why the calculator computes one leg at a time rather than pretending a person has a single ABI.

What the ABI ratio does and doesn't mean

Three properties of the ratio are pure arithmetic, and safe to state without stepping into interpretation.

It is dimensionless. mmHg over mmHg cancels, so an index of 0.96 is 0.96 in any consistent pressure unit — provided numerator and denominator use the same one. Mixing units is the one way to break the calculation before it starts.

An index of exactly 1 is the equality point. Had the ankle matched the higher arm at 125 mmHg, the ratio would be 125 ÷ 125 = 1.00. Below 1, the ankle reading was lower than the reference; above 1, higher. By itself, the number asserts nothing else.

It is sensitive to small measurement error. Starting from 120 ÷ 125 = 0.96, shift the ankle reading up by just 5 mmHg and the index becomes 125 ÷ 125 = 1.00; shift the arm reading up by the same 5 mmHg instead and it becomes 120 ÷ 130 = 0.92. A handful of mmHg — cuff position, patient rest, timing — moves the index by 0.04 in either direction. That sensitivity is why measurement protocols insist on technique and repetition, and why those procedural judgments sit with the clinician performing the exam, not with any calculator.

Everything beyond those three properties — whether a 0.96, a 0.80 or a 1.06 is reassuring, borderline or artifactual for a given person — is interpretation. Conditions the tool's scope note flags, such as diabetes or arterial calcification, can change how much trust an ankle reading deserves in the first place. That judgment call needs the symptoms, the history and the trend, and it belongs to the person who has all three.

Where the calculator draws its line

The ABI calculator carries the division at high precision, rounds once at the output, and prints a model-scope note beside the result instead of a diagnostic band — deliberately, because attaching a verdict would claim a judgment the arithmetic cannot support. What it does guarantee is the convention: the max-arm rule applied every time, all three inputs visible, consistent with the measurement literature its page cites. Nothing in that pipeline is hidden, and the same open-working standard holds for every tool on Quanta — the page has to earn its 0.96 in front of you, not ask you to take it on trust. Past that point, the open questions are protocol questions, and they have owners: which ankle reading enters the numerator is set by the protocol you are working under, and whether any given reading deserves trust is the examining clinician's call. If it is the convention itself that still seems underspecified for your situation, write to us through the contact page — ambiguity at the boundary of a scope note is something we treat as a defect, not a detail.

Sources

• Stanford Medicine 25 — the ankle-brachial index at the bedside
• NICE guideline CG147 — Peripheral arterial disease: diagnosis and management