Coagulation Panel

Antithrombin III (AT III)

Most coagulation tests flag blood that clots too slowly. Antithrombin flags the opposite problem, and how far it falls matters far more than whether it falls at all.

Part of the Coagulation Panel — see all 12 values together, including Factor V Leiden, Factor VIII, Lupus Anticoagulant.

Most clotting numbers warn when the system does too little; this one warns when the body's own restraint on clotting goes missing. Prothrombin time and the aPTT tick upward when blood is slow to clot. Antithrombin III runs the other way. It is the protein that puts the brakes on clotting, and when it falls, blood clots too readily.

Picture the counterweight that keeps an elevator car from free-falling when the cable lets go. Antithrombin is that counterweight for your clotting cascade: it constantly neutralizes thrombin and other clotting factors so that a small, useful clot does not snowball into a vein-blocking one. The functional test reports how much of that braking capacity is actually working, as a percentage of normal. Adults and children usually run 80 to 120%, according to the ISTH laboratory-testing recommendation; newborns sit lower, around 60 to 90%.

Here is the part the standard "low equals clot risk" framing leaves out. The risk is not a switch that flips at the bottom of the range. It is a slope, and most of the steepness lives near the floor.

What antithrombin activity usually means

%
Severe deficiency ≤ 60

The high-risk zone. In pregnancy, activity at or below 60% carried an odds ratio near 49 for venous clots, per the hereditary-thrombophilia study.

Clear deficiency 61–79

Below the reference range and worth a full workup with a clinician.

Low-normal gray zone 80–89

Just under the 90% mark studied for mild deficiency, where the odds ratio was 2.0 and not statistically significant. Close to normal for most people.

Typical range 90–120
High > 120

Generally not a concern on its own. See the high section below.

What does low antithrombin mean?

A low result means the brake is weaker than it should be, so blood clots more easily. MedlinePlus describes the test as a way to investigate repeated blood clots, or to explain why a blood-thinning medicine does not seem to work. But the number behind the word "low" carries very different weight depending on how far it has dropped.

The clearest data come from antithrombin during pregnancy, when clot risk is already raised. In the hereditary-thrombophilia analysis, mild deficiency defined as activity under 90% carried an odds ratio of 2.0 for a pregnancy-related clot, with a confidence interval (0.9 to 4.93) that crossed 1, meaning the effect was not statistically significant. Translated into absolute terms, that group faced roughly a 0.2 to 0.3% chance of a clot. A reading of 75% or 70% sits in this gentler stretch of the slope rather than the cliff. Activity of 60% or lower was a different story: an odds ratio around 49, and an absolute risk that climbed from about 6.1% in women under 35 to 9.0% in those 35 and older. Same diagnosis on paper, two very different situations.

That gradient matters because inherited antithrombin deficiency is not rare in its consequences. The National Blood Clot Alliance reports that about half of people with the deficiency develop a clot in their lifetime, often before age 30, and that without treatment up to half of pregnancies see a clot while roughly 1 in 6 ends in early fetal loss. The inherited form is autosomal dominant, so a single altered copy of the gene from one parent is enough to pass it on.

There is also a reason two people with identical activity numbers can have different antigen results. Deficiency comes in two flavors. Type 1 is quantitative: less protein is made, so both the amount (antigen) and the activity fall together. Type 2 is qualitative: the body makes a normal amount, but the protein is built wrong and does not work, so the antigen looks fine while the activity is low. That is why a functional activity assay, not an antigen-only test, is the one that catches both.

What can push antithrombin low

  • Inherited deficiency

    An autosomal dominant trait that lowers either the amount or the function of the protein from birth.

  • Liver disease

    The liver makes antithrombin, so reduced production pulls the level down.

  • Disseminated intravascular coagulation

    Widespread clotting consumes antithrombin faster than it is replaced.

  • Nephrotic syndrome

    Damaged kidneys leak the protein into the urine.

  • Sepsis

Why heparin sometimes stops working

There is a practical twist that sends many people to this test. Heparin, one of the most common hospital blood thinners, does almost nothing by itself. It works by latching onto antithrombin and supercharging it, accelerating the antithrombin-thrombin reaction by roughly 1,000-fold according to the cardiac-surgery literature. The drug is a throttle on the brake you already have.

So when antithrombin is low, heparin has less to work with. Doctors raise the dose, the blood keeps clotting, and the dose climbs again. That pattern, heparin resistance, is most often traced to low or dysfunctional antithrombin, especially the type 2 defects that affect the heparin-binding site. Spotting it usually means switching strategy rather than simply pushing more drug.

What does high antithrombin mean?

A high result is the reassuring direction. MedlinePlus notes that higher-than-normal antithrombin can follow anabolic steroid use, hemophilia, a kidney transplant, or low vitamin K, and that it is generally not a cause for concern on its own. More braking capacity is not a clotting problem.

What can push antithrombin high

  • Anabolic steroid use

  • Hemophilia

  • Kidney transplant

  • Low vitamin K

If your antithrombin comes back low

Turning a low result into a plan

  1. 1

    Bring the number to a doctor first

    Antithrombin is interpreted alongside your clot history and family history, not in isolation. A hematologist usually leads this conversation.

  2. 2

    Ask whether the level should be repeated

    A single reading can be skewed by acute clotting, heparin, or pregnancy. Clinicians often confirm a true deficiency on a separate, stable sample.

  3. 3

    Ask whether it is inherited or acquired

    Liver disease, nephrotic syndrome, DIC, and sepsis can all lower antithrombin temporarily, which changes the plan entirely.

  4. 4

    Make sure the functional assay was run

    Because type 2 deficiency hides behind a normal antigen level, the activity test is what confirms or rules it out.

No supplement raises antithrombin, and the test is not something to manage at home. The value of knowing the number is that it reframes a vague worry about clotting into a specific, gradable risk your care team can act on.

Antithrombin rarely travels alone on a lab report. It usually sits inside the coagulation panel next to screening tests like the PT versus aPTT pair, the natural anticoagulant protein C, the clot-breakdown marker D-dimer, and clotting-tendency genetics such as factor V Leiden. If you are trying to make sense of the whole report, the guide to reading a coagulation panel walks through how these pieces fit, and the guide to reading flagged values explains why a flag near the edge of the range is not the same as one deep below it. With antithrombin, that distinction is the difference between a footnote and a diagnosis.

Sources

Written and reviewed by BloodSight Editorial Team · Last updated

Try BloodSight

See your Antithrombin III on one timeline.

BloodSight calibrates the reference range to your sex, age, and lab — and shows every value across every visit.

Antithrombin III 5 visits
100 % −38
Mar Apr May Jun Jul

In your personal range

Normal ranges

Group Range Unit
Adult Male 80–120 %
Adult Female 80–120 %

Reference ranges may vary by laboratory and individual factors.

Antithrombin III — Common Questions

Is 85 percent antithrombin bad, or any mildly low result in the 80 to 90% band?
For most people, a result like 85% sits close to normal in terms of clot risk. In the pregnancy study most often quoted, activity under 90% carried an odds ratio of 2.0 that did not reach statistical significance, and the absolute risk of a pregnancy-related clot was about 0.2 to 0.3%. Mildly low is worth a conversation, not alarm. Your doctor will weigh it against your personal and family history.
At what antithrombin level does clot risk really climb?
The danger concentrates at the deep end. In the same pregnancy data, activity of 60% or lower carried an odds ratio of roughly 49, and the absolute risk of a pregnancy or postpartum clot ran from about 6.1% in women under 35 to 9.0% in women 35 and older. That is a different category of result from a number in the high 80s.
Why did my doctor keep increasing my heparin dose and it still didn't work?
Heparin does not thin blood on its own. It works by binding and activating antithrombin, speeding that natural brake by roughly 1,000-fold. When antithrombin is low, there is less for heparin to grab, so the dose climbs and the blood still clots. This pattern, called heparin resistance, often prompts an antithrombin test in the first place.
Can low antithrombin be temporary rather than inherited?
Yes. Inherited deficiency is one cause, but the level also drops with liver disease, which reduces production, with disseminated intravascular coagulation, which consumes it, with nephrotic syndrome, which loses it through the kidneys, and with sepsis. A single low reading is often repeated and interpreted in context before anyone concludes it is genetic.
Does a normal antigen level rule out antithrombin deficiency?
No, and this is the catch. Type 1 deficiency lowers both the amount of protein and its activity, so an antigen test catches it. Type 2 deficiency leaves the amount normal but the protein does not work, so an antigen-only test reads normal while the functional activity assay is low. The activity assay is what catches both types.

Disclaimer

This content is for informational and educational purposes only. It is not intended as medical advice, diagnosis, or treatment recommendation. Reference ranges may vary by laboratory. Always discuss your results with a qualified healthcare professional.

Related Tests

Protein C

Protein C is one of the body's natural brakes on clotting. Reading it correctly depends almost entirely on when the blood was drawn.

→
Prothrombin Time PT

Prothrombin time is a stopwatch on your blood's clotting. The catch: the same sample can post different seconds at different labs, which is exactly why the INR was invented.

→
Activated Partial Thromboplastin Time aPTT

aPTT times one arm of the clotting cascade, the one heparin acts on. Its great paradox: a prolonged result can point to a clotting risk, not a bleeding one.

→
D-dimer

D-dimer is the fragment left when the body dismantles a clot. It is the rare test you hope reads negative, because that is the only answer it gives with real confidence.

→
Factor V Leiden

The most common inherited clotting mutation, and the one whose name frightens more people than it ever harms.

→