Antithrombin deficiency

INTRODUCTION — Deficiency of antithrombin (AT; antithrombin III) can be inherited or acquired; it is defined as an AT activity level that is consistently less than 80 percent of normal (or the lower limit of the assay's reference range).

In some patients, AT deficiency can be associated with an increased risk of thromboembolism. The availability of plasma-derived AT concentrates and recombinant AT has made replacement therapy possible, but the approved use of these products is limited to patients with hereditary AT deficiency.

This topic discusses the causes, evaluation, and management of inherited and acquired AT deficiencies.

Additional topic reviews discuss other inherited thrombophilias and their management.

●Factor V Leiden – (See "Factor V Leiden and activated protein C resistance".)

●Prothrombin G20210A – (See "Prothrombin G20210A".)

●Protein S deficiency – (See "Protein S deficiency".)

●Protein C deficiency – (See "Protein C deficiency".)

●Management of inherited thrombophilias during pregnancy/postpartum – (See "Inherited thrombophilias in pregnancy".)

PATHOPHYSIOLOGY

AT function — Antithrombin (AT, previously called AT III, also known as heparin cofactor I) is a natural anticoagulant. It inhibits thrombin (factor IIa), factor Xa, and other serine proteases in the coagulation cascade such as factor IXa (figure 1 and figure 2) [1]. (See "Overview of hemostasis", section on 'Control mechanisms and termination of clotting'.)

AT is a serine protease inhibitor (serpin), a specific type of enzyme inhibitor. AT has a reactive center at position Arg393-Ser394 that interacts with the active site serine residue of the coagulation factor protease, and a heparin-binding site, which is distinct from the reactive center. Following the administration of heparin (unfractionated or low molecular weight [LMW]) or fondaparinux, AT activity is accelerated dramatically due to a conformational change induced by heparin binding leading to enhanced exposure of the reactive center in AT. This conformational change converts AT from a slow inactivator of coagulation factors such as thrombin (factor IIa) to a rapid inactivator (1000-fold increase in AT activity) [2]; the specific coagulation factor(s) affected depend on the size of the heparin molecule (figure 3). It is thought that endogenous heparan sulfates in the intact endothelium provide this role in normal physiology, in turn localizing the inhibitory activity of AT to the endothelial surface of blood vessels and maintaining the fluidity of blood [3]. AT may also have other roles such as reducing platelet adhesion to fibrinogen [4].

The half-life of AT is approximately 2.8 to 4.8 days. Production and turnover of AT both are highly regulated:

●AT is produced in the liver as a 464 amino acid precursor from which a 32 amino acid signal peptide is cleaved. The signal peptide plays a crucial role in translocation of the protein into the endoplasmic reticulum where cotranslational processing takes place [5].

●AT is present in plasma in two forms, an active monomer and an inactive "latent" form [6]. One molecule of latent AT can inactivate one molecule of the active AT monomer through the formation of a heterodimer; in health, this process only contributes to the normal turnover of the protein in plasma. However, this process may account for increased thrombosis risk in carriers of rare unstable variants of AT during fever or illness.

●Thrombin cleaves the reactive center of AT, which is followed by the formation of an inactive complex that is rapidly cleared from the circulation.

Healthy newborns have approximately 50 percent AT activity; this reaches adult levels by approximately six months of age [7-9]. Plasma levels of AT appear to be similar across ethnic groups [10,11].

AT deficiency can be inherited or acquired. Inherited deficiencies are due to AT gene mutations (see 'Genetics' below). Acquired deficiencies can result from impaired production of functional AT (eg, due to liver disease, asparaginase therapy); protein losses (eg, in nephrotic syndrome) or accelerated consumption, as occurs in acute thrombosis or disseminated intravascular coagulation (DIC). (See 'Acquired deficiency' below.)

Genetics — Inheritance of AT deficiency is autosomal dominant with variable penetrance. Individuals who are heterozygous for a variant in SERPINC1 have AT deficiency; homozygosity for AT deficiency is not seen, likely because it is not compatible with life. Children of affected individuals have an approximately 50 percent chance of inheriting the condition. However, not all individuals with laboratory evidence of AT deficiency will develop thrombosis or require anticoagulation.

Hundreds of pathogenic variants, including point mutations, insertions, and deletions, have been identified in SERPINC1, the gene encoding AT [12]. These are collated online in various databases such as ClinVar [13]. The resulting phenotypes are divided into type I and type II deficiencies, corresponding to reduced levels of AT antigen and activity (type I), or to functionally defective AT present at normal antigen levels (type II) [14].

As might be expected, heterozygosity for a null variant (with complete lack of AT production) appears to be associated with more severe disease compared with missense variants [15].

●Type I deficiency – Type I hereditary AT deficiency (reduced AT level) is usually caused by a small deletion or insertion, a larger deletion, or a single base substitution (mutation). Most of these changes in genetic sequence lead to reduced synthesis of the protein. Other changes may affect the signal peptide and impair post-translational processing or alter AT protein stability [5,16-20]. Homozygous type I deficiency has not been described, and homozygous gene deletion in mice is lethal [3].

●Type II deficiency – Type II hereditary AT deficiency (functional deficiency) is caused by a number of pathogenic variants that affect different aspects of protein function. Plasma AT functional activity may be markedly reduced, whereas AT immunologic activity typically is normal. Three subtypes of the type II phenotype have been described [2].

•Reactive site (type II RS) – Variants with decreased activity in AT functional assays generally have mutations near the thrombin binding site at the carboxyterminal end of the molecule.

•Heparin binding site (type II HBS) – Heparin binding site variants are the most common type of hereditary AT deficiency. These variants are also associated with a lower risk of thromboembolic complications compared with other SERPINC1 variants. Several abnormalities in AT that affect the heparin binding site have been identified; these may lead to isolated reductions in heparin cofactor activity. Affected individuals generally have plasma AT-heparin cofactor activity measurements of approximately 50 percent and normal progressive AT activity. (See 'Choice of assay' below.)

•Pleiotropic effect (type II PE) – The third subtype involves a mutation at the carboxyterminal end of the AT molecule between amino acids 402 to 429. These mutations produce conformational changes in the protein and can exhibit multiple defects including reductions in both heparin binding and progressive AT activity [16-19,21-24].

The relative frequency of these genetic variants depends on patient selection.

Acquired deficiency — A variety of acquired conditions lower the concentration of AT in the blood. Many of these are also associated with other alterations in procoagulant and anticoagulant factors, making it challenging to determine what role (if any) the reductions in AT levels play in the associated thrombotic complications.

●DIC – Disseminated intravascular coagulation (DIC) is characterized by ongoing coagulation and fibrinolysis that consumes procoagulant and anticoagulant factors [25]. Reports have suggested that more severe AT deficiency in the setting of DIC or sepsis is predictive of worse outcomes [26]. However, this association should not be interpreted to imply causation; for these conditions, the principal treatment involves addressing the underlying disease process. (See "Evaluation and management of disseminated intravascular coagulation (DIC) in adults", section on 'Intravascular coagulation and fibrinolysis'.)

●Thrombosis – Acute thrombosis can reduce AT levels transiently (table 1) [27].

●Liver disease – Severe liver disease (primarily cirrhosis) may cause reduced synthesis of procoagulant and anticoagulant factors, including AT [28,29]. (See "Hemostatic abnormalities in patients with liver disease", section on 'Physiologic effects of hepatic dysfunction'.)

●Nephrotic syndrome – Nephrotic syndromes may be associated with urinary losses of AT as well as other anticoagulant factors including protein S and protein C [30,31]. (See "Hypercoagulability in nephrotic syndrome".)

●ECMO – Extracorporeal membrane oxygenation (ECMO) and hemodialysis have also been associated with reductions in plasma AT levels [32]. This may result in ineffective anticoagulation with heparin. Replacement of AT can be accomplished more effectively by administering AT concentrate rather than by giving Fresh Frozen Plasma (FFP); AT replacement has not clearly been shown to be effective in improving clinical outcomes or prolonging circuit life, however. Thrombotic risk in these settings may also be contributed to by the presence of a central catheter for vascular access.

●Surgery/trauma – Surgery and trauma may be accompanied by a variety of prothrombotic and bleeding effects. Major surgery is associated with a modest fall in AT levels, with a nadir around the third postoperative day, followed by an increase to normal around the fifth postoperative day [33].

●Medications

•Asparaginase – Asparaginase therapy reduces synthesis of proteins that contain the amino acid asparagine, including AT. Asparagine-containing chemotherapy regimens for acute lymphoblastic leukemia (ALL) are associated with an increased thrombotic risk, which may be due to reduced AT in combination with other factors such as glucocorticoid administration, immobilization, and/or the presence of a central venous catheter.

•Heparins – Heparins can lower AT activity levels by as much as 30 percent, presumably by causing increased clearance of AT [34]. These reductions are not considered to increase thrombotic risk; however, this phenomenon may produce a false positive result in a patient undergoing thrombophilia testing while receiving heparin (ie, it may suggest the patient has hereditary AT deficiency when in fact the reduction is due to heparin therapy) (table 1).

•Estrogens – Modest reductions in plasma AT concentration are found in users of oral contraceptives and in individuals receiving estrogen for other purposes, but the clinical importance of these changes is not established [35,36].

AT levels do not change substantially during normal pregnancies, but they may decrease significantly with pregnancy-induced hypertension, preeclampsia, or eclampsia [35,37,38]. (See "Maternal adaptations to pregnancy: Hematologic changes", section on 'Coagulation and fibrinolysis'.)

AT is not considered an acute phase reactant. (See "Acute phase reactants".)

Consequences of deficiency — The two major consequences of AT deficiency are increased thrombotic risk and insensitivity to heparin:

●Increased thrombotic risk may be seen in individuals with inherited or some forms of acquired AT deficiency such as asparaginase chemotherapy (see 'Thrombosis' below). The reduced level of AT (typically, approximately 50 percent of normal, range 40 to 60 percent) can lead to uncontrolled thrombin generation and fibrin deposition within the vasculature. This mechanism contrasts with coagulation factor deficiencies, in which heterozygosity for a mutation typically is associated with a benign carrier condition and reduction of plasma activity levels to approximately half of normal rarely have clinical consequences.

●Insensitivity to heparin is seen in some individuals with AT deficiency (see 'Heparin resistance' below). This occurs because heparins require AT to inactivate coagulation enzymes. In contrast, direct inhibitors of coagulation enzymes such as direct thrombin (factor IIa) inhibitors and direct factor Xa inhibitors do not require AT.

Antithrombin function is not assessed by routine coagulation tests such as the prothrombin time (PT), activated partial thromboplastin time (aPTT), and thrombin time (TT); thus, results of these tests are not affected by AT deficiency.

EPIDEMIOLOGY — Hereditary AT deficiency (OMIM #613118) is infrequent; it is significantly less common than the factor V Leiden mutation and the prothrombin G20210A mutation (table 2). The estimated prevalence in the general population is thought to be in the range of 0.02 to 0.2 percent (1 in 5000 to 1 in 500 individuals) [39-43]. Deficiency is equally common in males and females.

The following studies illustrate the range of prevalences seen in different clinical settings:

●In a series of 2132 individuals who presented with venous thromboembolism, 274 (13 percent) had a deficiency of protein S, protein C, or AT; AT deficiency was seen in 10 (0.5 percent of the total cohort, 3.6 percent of the deficiencies) [44].

●In a cohort of 319 children and adolescents with thromboembolism, 21 (6.6 percent) had AT deficiency [45]. Of note, slightly over half had an additional precipitating factor such as oral contraceptive use, prolonged immobilization, or another inherited thrombophilia.

●In the several studies of healthy blood donors, prevalences of AT deficiency have ranged from approximately 1 in 600 to 1 in 200 (0.2 to 0.6 percent) [46-49]. Of interest, in one study that recalled 59 individuals who initially tested as AT deficient, repeat testing revealed AT deficiency in only three, none of whom had a personal or family history of thrombosis; an additional three had experienced an acute event such as a deep vein thrombosis at the time of the initial sample and may have had transient, acquired AT deficiency that had resolved at the time of retesting [46]. (See 'Acquired deficiency' above.)

These observations support the practice of reserving laboratory testing for AT deficiency to individuals with a compelling personal or family history. (See 'Indications for testing' below.)

CLINICAL MANIFESTATIONS

Thrombosis — AT deficiency is associated with venous thromboembolism (VTE). Often the family history is remarkably positive with an autosomal dominant pattern [50,51]. (See 'Genetics' above.)

Arterial thrombosis has been reported but is not characteristic of AT deficiency [52].

VTE risk — Individuals with hereditary AT deficiency have an increased risk of VTE. Estimates of the degree of increased risk vary among populations. A meta-analysis of case-control and cohort studies showed that compared with controls, AT-deficient individuals had an odds ratio (OR) of VTE of 16.3 (95% CI 9.9-26.7); this was substantially higher than the VTE risk attributed to protein S or protein C deficiency (ORs 5.4 and 7.5, respectively) (table 2) [53].

In studies that focused on thrombophilic families in which at least one member had a history of thrombosis, the risk for VTE in relatives was approximately 1 percent annually [54,55]. Individuals who co-inherit a second thrombophilic defect are likely to have a greater risk. However, as noted above, some individuals with AT deficiency, especially those diagnosed incidentally (eg, during population screening) never have a VTE.

In some series of individuals with AT deficiency who have a VTE, more than half have a spontaneous event; the remaining cases are associated with an acquired risk factor such as pregnancy, oral contraceptive use, surgery, or trauma [55]. The most common sites of thrombosis are the deep veins of the legs, the iliofemoral veins, the mesenteric veins, and the pulmonary arteries [22,56-59]. Less frequently involved sites include the vena cava and the renal, cerebral, or hepatic veins (Budd-Chiari syndrome) [45,60-63].

Patients with AT deficiency appear to be at particularly high risk of thrombosis during pregnancy [64]. In a 2016 systematic review that included four case-control studies involving pregnancies in women with and without AT deficiency, the OR for VTE with AT deficiency was 6.1 (95% CI 1.6-23.4) [65]. In another study, the risk of VTE was evaluated in 78 women with inherited AT deficiency, protein S deficiency, or protein C deficiency [66]. In women with AT deficiency, deep vein thrombosis occurred in 18 percent during pregnancy and in 33 percent in the postpartum period. By comparison, thrombosis during pregnancy occurred in only 7 percent of patients with protein C deficiency and in none of the patients with protein S deficiency. Thrombosis in the postpartum period occurred in 19 percent of protein C and 17 percent of protein S deficient patients. This increased risk provides our rationale for administering AT concentrate to some individuals with hereditary AT deficiency who become pregnant, especially around the time of delivery. (See 'Pregnancy' below.)

Age at presentation — The age of presentation of venous thromboembolism (VTE) in hereditary AT deficiency varies widely. Although type I AT deficiency (see 'Genetics' above) is the most clinically penetrant of the inherited thrombophilias, some individuals never have a thromboembolic event in their lifetime. (See 'VTE risk' above.)

Examples of the range of ages at presentation include the following:

●In a cohort of 21 children and adolescents with confirmed hereditary AT deficiency, the mean age of the first thromboembolic event was 14 years (range, 0.1 to 17 years) [45].

●In another cohort of 29 children, the mean age at diagnosis was eight years; 10 patients were diagnosed following a venous thromboembolic event (34 percent) [43].

●In a review of AT replacement in the peripartum setting, two presentations involved individuals first diagnosed when they developed a venous thromboembolic event during pregnancy [67].

●Case reports of earlier presentations may reflect multiple thrombophilic defects. This has been reported in patients with AT deficiency as well as the factor V Leiden (FVL) mutation [68]. The genes for AT and factor V are both on the long arm of chromosome 1, and in families with both mutations, they often segregate together (ie, first-degree relatives will either have both mutations or neither). In a case report, an infant of consanguineous parents presented at 13 months of age with fever and cerebral vein thrombosis; he had homozygosity for a familial SERPINC1 mutation (with AT activity of 11 percent) along with heterozygosity for the FVL mutation [69].

●Some children may be protected from VTE during childhood due to higher levels of alpha-2-macroglobulin during the first two decades of life; alpha-2-macroglobulin also acts as a thrombin inhibitor [70]. Some studies have reported an increased frequency of thromboembolic events after puberty, with a peak between 15 and 35 years [57].

Renal disease — Rarely, individuals with hereditary AT deficiency have developed fibrin thrombi in the kidneys, leading to renal insufficiency. This creates challenges for managing hemodialysis. (See 'ECMO, CPB, or hemodialysis' below.)

Heparin resistance — In some individuals with hereditary or acquired AT deficiency, anticoagulation with heparin is ineffective and the aPTT cannot be adequately prolonged. This occurs because heparin is an indirect inhibitor of thrombin and factor Xa, and requires adequate levels of circulating AT in order be effective. (See 'AT function' above.)

Management of heparin resistance due to AT deficiency is discussed briefly below and in a separate topic review. (See 'Heparin resistance' below and "Heparin and LMW heparin: Dosing and adverse effects", section on 'Heparin resistance/antithrombin deficiency'.)

DIAGNOSTIC EVALUATION

Indications for testing

Patients with VTE — We do not perform AT testing in the majority of patients with venous thromboembolism (VTE). We may test the following individuals:

●Suspected inherited thrombophilia (eg, based on family history or atypical presentation) – (See "Evaluating adult patients with established venous thromboembolism for acquired and inherited risk factors".)

●Children with thromboembolism in the absence of a central venous catheter – (See "Thrombophilia testing in children and adolescents".)

●Suspected heparin resistance – (See "Heparin and LMW heparin: Dosing and adverse effects", section on 'Heparin resistance/antithrombin deficiency'.)

●Asparaginase therapy or extracorporeal membrane oxygenation – (See 'Patients receiving asparaginase' below and 'ECMO, CPB, or hemodialysis' below.)

Importantly, in individuals with a thromboembolic event for whom testing is indicated, AT levels should not be measured at the time of the acute event, because thrombosis may cause a transient reduction in AT levels that could be misinterpreted as an underlying inherited deficiency. If a low level is measured during an acute illness, this should be repeated when the individual has recovered. Testing using functional assays should not be performed while patients are taking certain anticoagulants. (See 'Timing of testing' below.)

In the event that a patient with an acute VTE treated with heparin appears to have heparin resistance (due to failure of the activated partial thromboplastin time [aPTT] to prolong despite very high doses of heparin) and a low AT level is measured, it may not be possible to determine whether the low level is due to inherited deficiency or to acquired deficiency (which in turn may be due to thrombosis and/or heparin). In such cases, we would repeat the testing when the individual's medical condition has returned to baseline (eg, several weeks to months later). (See 'Timing of testing' below.)

Patients without VTE — Testing for AT deficiency in asymptomatic individuals (individuals without VTE) is rarely appropriate. Exceptions may include the following:

●Family members of an individual with hereditary AT deficiency associated with thromboembolic complications. (See 'Testing of family members and genetic counseling' below and "Screening for inherited thrombophilia in asymptomatic adults".)

●Children or adults receiving asparaginase chemotherapy, if a finding of AT deficiency would be helpful in management. (See 'Patients receiving asparaginase' below.)

●Individuals with apparent heparin resistance in a non-VTE setting (eg, cardiac surgery). (See "Heparin and LMW heparin: Dosing and adverse effects", section on 'Heparin resistance/antithrombin deficiency'.)

Choice of assay — A functional assay for plasma AT activity (also called AT-heparin cofactor assay) is the best first test for AT deficiency [50,51]. The assay measures the ability of heparin to inhibit a coagulation factor (thrombin [factor IIa] or factor Xa), which requires AT activity (figure 3). Guidance from the International Society for Thrombosis and Haemostasis recommends a chromogenic assay [51].

We prefer factor Xa inhibition to thrombin inhibition, due to the increased specificity of factor Xa inhibition for AT deficiency. The rationale is that factor Xa is inhibited by AT (heparin cofactor I) alone via a heparin-dependent mechanism, whereas thrombin can also be inhibited by heparin cofactor II [71-73]. Heparin cofactor II is an efficient inhibitor of thrombin in the presence of heparin concentrations used in the AT-heparin cofactor assay [72]. As a result, an assay based on thrombin inhibition can be affected by differences in heparin cofactor II activity. The increased specificity of the factor Xa-based assay was illustrated in a study of 67 members of a family with type II AT deficiency (see 'Genetics' above) [74]. In members with true AT deficiency based on genetic testing, the factor Xa-based assay was more reliable than the thrombin-based assay; six individuals might have been reported incorrectly as negative for the defect when tested with the thrombin-based assay.

The factor Xa-based assay may be inaccurate in individuals with a specific AT mutation known as AT Cambridge (AT A384S). Factor Xa-based assays cannot be used to diagnose these individuals, and results from thrombin-based assays are only mildly decreased [75]. The clinical relevance of this mutation remains unclear.

Most laboratories set the lower limit of normal for AT activity at approximately 80 percent (normal range, approximately 80 to 120 percent); the values are typically calculated as two standard deviations from the mean. Laboratory-specific values should be used to take into account instrument and assay variability. Patients with hereditary AT deficiency due to heterozygosity for a SERPINC1 mutation usually have AT activity levels in the range of 40 to 60 percent.

Assays for AT protein levels (rather than function) such as enzyme-linked immunosorbent assays (ELISAs) are also available [1,51,76]. However, these are less useful for initial testing because they do not reflect AT function and will fail to identify individuals with a type II defect (see 'Genetics' above). In some cases this testing can be done to distinguish between type I and type II defects in an individual with deficient AT activity.

If needed, additional testing can be performed in specialized coagulation laboratories. As an example, a few specialized laboratories can perform the progressive AT activity assay, which quantifies the capacity of AT to neutralize the enzymatic activity of thrombin in the absence of heparin. This assay can be used to distinguish functional deficiency produced by defects in the AT thrombin binding site versus the heparin binding site. This additional testing is usually reserved for research purposes and is not widely available.

Genetic testing is not routinely done but may be obtained from a specialty laboratory in selected cases [51]. (See 'Genetic testing' below.)

Timing of testing — An important consideration in the laboratory evaluation of patients with suspected AT deficiency is the timing of testing. Erroneous diagnoses can be made due to the influence of acute thrombosis, comorbid illness, and/or anticoagulant therapy on AT activity levels [51]. (See 'Acquired deficiency' above.)

For most patients with an acute thromboembolic event or acute illness, testing should be delayed until the patient recovers or is no longer receiving an anticoagulant. If AT activity was measured in the setting of an acute thromboembolic event or systemic illness, it should be repeated when the patient is asymptomatic. (Related Lab Interpretation Monograph(s): "Low antithrombin in adults".)

Exceptions may include the following:

●Family members of individuals with AT deficiency for whom rapid diagnosis is important for care (eg, in the setting of emergency surgery or acute thromboembolism).

●Individuals receiving heparin therapy for whom a therapeutic aPTT cannot be achieved and/or additional thrombosis develops, and for whom AT replacement therapy or a change in anticoagulant would be used if AT deficiency was identified.

●Individuals receiving asparaginase for whom AT replacement would be used if AT deficiency was identified.

For individuals on direct oral or parenteral anticoagulants who require a more rapid evaluation or for whom it is not considered safe to discontinue the drug, testing may be accomplished by using an alternative assay (eg, factor Xa-based assay in a patient receiving a direct thrombin inhibitor, thrombin-based assay in a patient receiving a direct factor Xa inhibitor). (See 'Choice of assay' above.)

The rationale for avoiding testing during anticoagulation is that anticoagulants in some cases can alter the results of testing:

●Heparin can lower AT activity, potentially giving the false impression that a patient has AT deficiency (table 1). This effect is best documented for unfractionated heparin but is also theoretically possible for low molecular weight (LMW) heparins.

●Direct factor Xa inhibitors (eg, rivaroxaban, apixaban, edoxaban) can lead to an overestimation of AT activity in a factor Xa-based functional assay, leading to the false impression that a patient with AT deficiency has normal AT levels.

●Direct thrombin inhibitors (eg, argatroban, dabigatran) can lead to an overestimation of AT activity in a thrombin-based assay, leading to the false impression that a patient with AT deficiency has normal AT levels.

In contrast to these other anticoagulants, warfarin does not appear to raise AT levels [77]. Prior reports of increased AT levels in individuals taking warfarin relied on methods for assaying AT that are no longer in common use.

Genetic testing — Genetic testing for AT deficiency may be ordered from a specialty laboratory, but we do not obtain it in our routine clinical practice [51].

Diagnosis — The diagnosis of AT deficiency requires demonstration of a reduced plasma AT activity level in a patient who is not in the midst of an acute illness or surgery that could cause transient reduction in AT activity. In patients with hereditary AT deficiency, AT activity is generally well below the lower limit of the normal reference range of the laboratory performing the test.

In some cases where an abnormal result is obtained, repeat testing for confirmation at a later time may be required.

As noted, the diagnosis of hereditary AT deficiency requires exclusion of acquired causes; confirmation of a familial disorder requires investigation of related family members [51]. Genetic testing and demonstration of a SERPINC1 mutation is confirmatory but not required. (See 'Genetics' above.)

In the vast majority of patients, the diagnosis is considered excluded if the activity level is in the normal range.

Differential diagnosis — The differential diagnosis of VTE in association with AT deficiency includes other causes of thromboembolism and heparin resistance.

●Inherited thrombophilias – Other inherited thrombophilias include the factor V Leiden mutation, prothrombin G20210A mutation, and deficiencies of protein S or protein C. Like hereditary AT deficiency, these may be associated with a familial syndrome of increased thromboembolic disease. Unlike AT deficiency, these are not associated with low AT levels unless the patient has had a recent thrombosis. (See "Factor V Leiden and activated protein C resistance" and "Prothrombin G20210A" and "Protein S deficiency" and "Protein C deficiency".)

●Acquired thrombophilias – Other acquired thrombophilias include a range of conditions such as nephrotic syndrome, some hemolytic anemias (eg, paroxysmal nocturnal hemoglobinuria [PNH]), myeloproliferative neoplasms and other malignancies, and antiphospholipid syndrome. Other transient associations with increased thromboembolic risk include hospitalization for an acute medical illness or surgery or the presence of a central venous catheter. Like AT deficiency, these conditions may increase the risk of thromboembolism. Unlike AT deficiency, these more commonly present in adulthood, and AT levels will be normal in these disorders unless the patient has had a recent thrombosis or other acute illness. (See "Overview of the causes of venous thrombosis" and "Cerebral venous thrombosis: Etiology, clinical features, and diagnosis".)

●Causes of heparin resistance – Other causes of heparin resistance include increased clearance, increased levels of heparin-binding proteins, elevations of fibrinogen or factor VIII, and certain medications (eg, aprotinin). Like AT deficiency, these conditions lead to a requirement for unusually large doses of heparin. Unlike AT deficiency, these conditions often can be overcome by raising the heparin dose, and they do not improve with AT administration. (See "Heparin and LMW heparin: Dosing and adverse effects", section on 'Heparin resistance/antithrombin deficiency'.)

MANAGEMENT

Overview of management — Anticoagulation is appropriate for individuals with hereditary AT deficiency who develop a thromboembolic event (typically, venous thromboembolism [VTE]) (algorithm 1). (See 'VTE treatment (hereditary deficiency)' below.)

In addition, prophylactic anticoagulation may be used in certain high-risk settings such as pregnancy or surgery. Details of the timing, duration, and choice of anticoagulant are discussed separately.

●Pregnancy – (See "Inherited thrombophilias in pregnancy", section on 'Our approach to patients with high-risk thrombophilias'.)

●Surgery – (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients".)

AT replacement may be used in addition to anticoagulation in certain settings in some individuals with hereditary AT deficiency (or rarely, acquired AT deficiency such as from asparaginase therapy), such as acute thromboembolism and/or heparin resistance. (See 'VTE treatment (hereditary deficiency)' below and 'VTE treatment (asparaginase)' below.)

AT replacement alone (without anticoagulation) may be appropriate as prophylaxis in certain settings (eg, peripartum, surgery, asparaginase) where even prophylactic anticoagulation is temporarily contraindicated; it may also be used in some cases of asparaginase therapy, but this use is less well studied. (See 'VTE prophylaxis (hereditary deficiency)' below and 'VTE prophylaxis (asparaginase)' below.)

VTE treatment (hereditary deficiency)

Anticoagulation — Individuals with hereditary AT deficiency with a new VTE (deep vein thrombosis [DVT] or pulmonary embolism [PE]) require treatment with anticoagulation. Most experts treat indefinitely.

The choice of anticoagulant requires careful consideration of the risks and benefits.

●Initial parenteral anticoagulation is the mainstay of therapy, especially for severe thromboses. Data are limited, but clinical experience suggests that LMW heparin appears to be effective [50]. AT replacement is used in some cases. (See 'AT replacement' below.)

●For oral anticoagulation, options include warfarin or a direct oral anticoagulant (DOAC). While the choice is individualized depending on a number of factors (severity of thrombosis, adherence, patient preference, potential dietary or drug interactions), we generally prefer LMW heparin followed by warfarin rather than a DOAC for patients with VTE in association with known hereditary AT deficiency.

Data on the efficacy of DOACs in individuals with AT deficiency are sparse, with only a few case reports of acute VTE treatment [50,78,79]. Most trials evaluating DOACs did not routinely screen participants for hereditary thrombophilia, so there may have been individuals with AT deficiency enrolled in these trials [80]. (See "Direct oral anticoagulants (DOACs) and parenteral direct-acting anticoagulants: Dosing and adverse effects".)

Thus, the rationale supporting the use of warfarin includes more data and clinical experience and the ability to effectively monitor its anticoagulant effect using coagulation testing (prothrombin time [PT, with international normalized ratio [INR]). Furthermore, if needed, a higher INR goal can be targeted in highly thrombosis-prone individuals. Furthermore, if the individual has a recurrent event while receiving warfarin, it is readily possible to establish whether the individual was taking the anticoagulant and whether the level was in the desired therapeutic range.

However, the DOACs are mechanistically appealing for acute VTE treatment and secondary prophylaxis, given that they act independently of AT. The DOACs apixaban and rivaroxaban are given at increased dose initially as "monotherapy" using the regimens approved for the treatment of DVT and PE. Individuals treated with dabigatran or edoxaban are treated with parenteral anticoagulation for 5 to 10 days before initiating the oral anticoagulant (table 3).

If it is decided to use a DOAC for the long-term prevention of recurrent VTE, we generally continue a higher dose regimen (eg, rivaroxaban 20 mg once daily or apixaban 5 mg twice daily rather than rivaroxaban 10 mg once daily or apixaban 2.5 mg twice daily), assuming the individual's bleeding risk is not excessive. This is because AT deficiency is one of the more thrombophilic of the hereditary thrombophilias. The prescribing physician and patient should understand that evidence for the optimal dose in such patients is lacking.

AT deficiency is the most clinically penetrant of the hereditary thrombophilias; although there is considerable variation in its expression, some patients with this disorder, in our experience, can be extremely thrombosis-prone and sustain recurrent VTE events despite therapeutic anticoagulation [81].

AT replacement — In some cases, AT replacement has been used in individuals receiving unfractionated or LMW heparin, especially for those with unusually severe thrombosis, recurrent thrombosis despite adequate anticoagulation, or heparin resistance (ie, patients receiving heparin for whom a therapeutic aPTT cannot be achieved despite the administration of very large doses of unfractionated heparin) [82,83].

Alternatively, a parenteral direct thrombin inhibitor such as argatroban, which does not require AT function, may be used. Decisions regarding therapy are individualized depending on whether the patient is anticoagulated at the time the VTE develops (which would favor use of AT replacement) and the initial aPTT response to heparin therapy (heparin resistance would favor AT replacement; heparin resistance with AT being unavailable would favor switching to a non-heparin anticoagulant such as argatroban).

When AT replacement is used, dosing is based on the patient’s baseline AT activity level and a target peak level of 120 percent of normal post-infusion. AT replacement can be stopped once the patient is stably anticoagulated with an oral anticoagulant and heparin is discontinued.

Available AT products and dosing — Available AT replacement products include an antithrombin concentrate derived from pooled human plasma (Thrombate III) and a recombinant human antithrombin produced from the milk of transgenic goats (rhAT, ATryn). (See "Plasma derivatives and recombinant DNA-produced coagulation factors", section on 'Antithrombin'.)

AT is dosed in units of activity; one unit is defined as the amount of AT in one milliliter of pooled normal human plasma. The dose and schedule depend on the product used (once daily dosing for plasma-derived AT, continuous infusion for recombinant AT), the patient’s baseline AT activity level and body weight (used for initial dose calculation), and the clinical setting (dosing of the concentrate differs in pregnancy versus surgery). Institutional guidelines and product information or the Lexicomp drug information on antithrombin (concentrate and recombinant) included with UpToDate should be consulted [84,85].

●Plasma-derived AT concentrate – AT concentrate is given as an initial intravenous dose over a 10 to 20 minute period. The first AT activity measurement is made two hours after the initial dose, and subsequent dosing is adjusted based on AT levels. Dosing interval is approximately every 24 hours, although this may change depending on AT activity levels [82]. Once daily dosing is possible for the plasma-derived AT concentrate due to its half-life, which is in the range of 43 to 77 hours (approximately two to three days) [86].

●Recombinant AT – Recombinant AT is administered as an initial intravenous dose followed by a continuous maintenance infusion [87]. AT activity monitoring is required for proper maintenance dosing. Continuous infusion is used because the half-life of the recombinant product is approximately 10 hours; the shorter half-life compared with plasma-derived AT is due to differences in glycosylation [67].

AT replacement products generally are well tolerated. The plasma-derived concentrate has a theoretical risk of transmitting an infectious disease, although these products undergo numerous steps to inactivate infectious agents and no infectious transmission has been confirmed [67]. The recombinant product has been associated with hypersensitivity reactions, and a registry has been created to monitor for antibody development. Cost is a major consideration for both products.

For individuals who do not have access to AT concentrates or recombinant AT, plasma products such as Fresh Frozen Plasma (FFP) may provide some AT. However, the doses required may confer a significant volume load as well as risks for transfusion reactions.

The concentrate and the recombinant product have not been compared directly to each other in a randomized trial. However, several large observational studies and trials comparing these products with placebo have reported that they are effective in facilitating AT replacement in individuals with heparin resistance and/or AT deficiency, and premarketing studies demonstrated a low rate of thrombosis when they were used in the perioperative or peripartum settings when anticoagulation was withheld [67,88-94]. As noted above, the half-life of the concentrate is significantly longer than the recombinant product.

VTE prophylaxis (hereditary deficiency) — In patients with incidentally discovered AT deficiency (ie, without a personal or family history of thrombosis), we would use prophylactic anticoagulation only during periods of increased thrombotic risk (eg, following major surgery); we generally would not undertake prophylactic AT administration.

However, some individuals with hereditary AT deficiency may benefit from anticoagulation and/or AT administration in certain settings, such as pregnancy or surgery. (See 'Pregnancy' below and 'Surgery' below.)

In contrast, secondary prophylaxis following VTE (eg, indefinite anticoagulation to prevent further VTE events) is usually appropriate.

Individuals with hereditary AT deficiency should avoid the use of oral contraceptives, especially if the family history is positive for thromboembolism. Alternative forms of contraception are presented separately. (See "Contraception: Counseling for women with inherited thrombophilias".)

Additional counseling and educational materials that discuss risk reduction strategies are available from the National Alliance for Thrombosis and Thrombophilia and the National Organization for Rare Disorders.

Pregnancy — Pregnancy is associated with an increased risk of thromboembolism, and AT deficiency further elevates this risk. The majority of women with hereditary AT deficiency will already be receiving thromboprophylaxis during pregnancy (typically with low molecular weight [LMW] heparin), regardless of their personal or family history of thrombosis. (See "Inherited thrombophilias in pregnancy".)

Details of the heparin product, dose, schedule, and monitoring are presented separately. (See "Use of anticoagulants during pregnancy and postpartum".)

Various approaches to the administration of AT during pregnancy and delivery have been used, including administration restricted to the time of delivery, for a few days before and after delivery, for six weeks postpartum, or for longer periods antepartum. AT may be continued during delivery to provide thromboprophylaxis when anticoagulation is withheld.

Our approach to AT administration depends on the patient’s personal and family history of thromboembolism (algorithm 1):

●For women with hereditary AT deficiency who develop a thromboembolism during pregnancy, we use therapeutic dose anticoagulation with LMW heparin for the remainder of the pregnancy, as discussed in detail separately. (See "Venous thromboembolism in pregnancy and postpartum: Treatment".)

●For women with hereditary AT deficiency and either a personal history of thrombosis or a compelling family history (eg, multiple family members with thrombosis, family member with unprovoked thrombosis), we suggest administering AT just prior to and at the time of delivery when anticoagulation cannot be administered (ie, from the time LMW or unfractionated heparin is discontinued in anticipation of delivery, until postpartum anticoagulation is resumed). This practice is based on the high risk of VTE during pregnancy in such patients and the efficacy of AT replacement described in case reports and our experience [67]. Anticoagulation should be resumed as soon as possible postpartum. (See "Use of anticoagulants during pregnancy and postpartum", section on 'Postpartum and breastfeeding'.)

Women with no personal history of thrombosis often can be managed with a single dose of AT concentrate just prior to delivery, while women with recurrent VTE or VTE during pregnancy may need additional doses postpartum along with therapeutic anticoagulation.

●Women with hereditary AT deficiency without a personal or family history of thromboembolism may be treated with anticoagulation alone.

Dosing during pregnancy is as follows:

●Plasma-derived AT concentrate:

•Maintenance dose (in units, given once every 24 hours): Approximately 60 percent of the initial dose

•Dose adjustments: The dose or interval is adjusted to keep activity levels between 80 and 120 percent

●Recombinant AT:

•Dose adjustments: Dose adjustments are made based on the AT activity level two hours after initial treatment

Surgery — Surgery is associated with a transient reduction in AT levels as well as increased thromboembolic risk, and AT replacement has been used in some individuals with hereditary AT deficiency to allow adequate thromboprophylaxis with heparin and/or heparin bridging during interruption of warfarin for the procedure [88,89]. Therapy with AT may be initiated just before a planned procedure and can be continued until anticoagulation can be instituted postoperatively; this is based on the type of surgical procedure performed and the thrombotic history of the patient.

Perioperative dosing is as follows:

●Plasma-derived AT concentrate:

•Maintenance dose (in units, given once every 24 hours): Approximately 60 percent of the initial dose

•Dose adjustments: The dose or interval is adjusted to keep activity levels between 80 and 120 percent

●Recombinant AT:

•Dose adjustments: Dose adjustments are made based on the AT activity level two hours after initial treatment

Evidence for efficacy in the surgical setting is limited; there are no randomized trials. In a single arm study that included six surgical patients, use of recombinant AT (typically given together with postoperative anticoagulation for thromboprophylaxis) was associated with good outcomes [89]. There were no VTE events, and therapy was generally well tolerated. In a retrospective review of 23 patients undergoing various surgical procedures, mostly gynecologic, use of AT concentrate was associated with a reduced incidence of thromboembolic complications; however, numbers were small, and many of the patients who did not receive AT were given thromboprophylaxis with dextran, which is no longer used [95].

Patients receiving asparaginase — The optimal management of patients with AT deficiency (or suspected deficiency) due to asparaginase therapy is unknown.

Guidance from the International Society of Haemostasis and Thrombosis (ISTH) has issued recommendations based on a relatively weak data set [96]. However, decisions regarding anticoagulation or AT replacement in individuals with acute lymphoblastic leukemia (ALL) are generally made by the clinician treating the leukemia, many of whom do not use anticoagulation, especially in the absence of other risk factors for thrombosis (see "Induction therapy for Philadelphia chromosome negative acute lymphoblastic leukemia in adults", section on 'Supportive care'). Institutional protocols should be followed.

The ISTH recommendations include the following:

●Provide thromboprophylaxis with LMW heparin during asparaginase treatment. (See 'VTE prophylaxis (asparaginase)' below.)

●Monitor AT levels during asparaginase therapy. Once weekly monitoring is considered appropriate. (See 'Choice of assay' above.)

●Administer AT concentrate for AT levels below 50 to 60 percent. The optimal target is unknown; a level in the range of 80 to 120 percent is reasonable. (See 'AT replacement' above.)

●Treat acute VTE with anticoagulation. Choice of therapy, need for AT concentrate, and continuation versus stopping asparaginase are discussed below. (See 'VTE treatment (asparaginase)' below.)

Mechanisms and magnitude of risk with asparaginase — The mechanism of VTE with asparaginase is multifactorial and involves decreased synthesis of asparagine-containing proteins such as AT as well as other global effects on protein synthesis [97]. Plasma levels of several procoagulant and anticoagulant factors are reduced, and these reductions are thought to be at least partially responsible for the increased risks of bleeding and thrombosis, respectively. Affected factors besides AT include prothrombin; factors V, VII, VIII, IX, X, XI; fibrinogen; protein C; protein S; von Willebrand factor (VWF); and plasminogen [97-100]. However, the correlation between VTE risk and plasma levels of AT, protein S, or protein C is weak, suggesting other contributing factors [101]. Asparaginase causes depletion of asparagine by catalyzing its hydrolysis to aspartic acid.

Asparaginase also increases the risk of bleeding, which may be due to a combination of effects. There may be prolongation of the prothrombin time (PT), activated partial thromboplastin time (aPTT), and thrombin time (TT), as well as hypofibrinogenemia with fibrinogen levels often <100 mg/dL. Chemotherapy-induced thrombocytopenia may contribute as well.

The risk of thrombosis with L-asparaginase is increased in both children and adults:

●Children – In a series of 1547 children receiving L-asparaginase as part of induction chemotherapy for ALL, thrombotic and/or hemorrhagic complications occurred in 18 (2.1 percent) [102]. Fourteen of these events were in the central nervous system (five intracranial thromboses, five intracranial hemorrhages, four brain infarctions with secondary hemorrhage); there were six leg vein thromboses.

●Adults – In a series of 548 patients receiving L-asparaginase for ALL induction that included 35 adults, the incidence of VTE increased with increasing age [103]. There were four VTE events in 16 individuals 21 to 30 years of age (25 percent) and eight VTE events in 19 individuals >30 years of age (42 percent). Sites of VTE were similar in adults as in children, with a trend for adults to experience more pulmonary emboli (PE). The risk of thromboembolic events appears to be greater in adults and individuals with high-risk ALL [104].

It is not clear whether different formulations of asparaginase are associated with different degrees of thromboembolic risk [105]. Other forms of asparaginase may not have as profound effects on coagulation as L-asparaginase produced from Escherichia coli. Erwinia asparaginase is an alternate preparation that may have fewer effects. In a series of 11 adults with acute lymphoblastic leukemia (ALL) who were treated with Erwinia asparaginase, there was significant lowering of antithrombin levels, but vitamin K-dependent procoagulant factors II, VII, and X remained within normal ranges [106].

Other factors that may affect risk include the age of the patient, other coagulation factor deficiencies induced by the drug, or other prothrombotic conditions (eg, presence of a central venous catheter, vascular inflammation, leukemia cell lysis). (See "Cancer-associated hypercoagulable state: Causes and mechanisms".)

Clinical presentation (asparaginase-associated VTE) — Thrombosis may occur during induction chemotherapy or postinduction. In the series of 548 adults cited above, the median time to development of VTE was 3.5 months (range 0.5 to 10.1 months) [103].

The most frequent site of thrombosis is intracranial (eg, dural sinus thrombosis); however, deep venous thrombosis (DVT) and pulmonary embolism (PE) can also occur [102,107-110]. Central nervous system thrombosis may also be complicated by hemorrhage. The predilection for central nervous system thrombosis in children remains unexplained [101].

Presenting findings of dural sinus thrombosis may include headache, focal neurologic findings, or encephalopathy. Appropriate laboratory testing and neuroimaging are presented separately. (See "Cerebral venous thrombosis: Etiology, clinical features, and diagnosis", section on 'Clinical aspects'.)

VTE prophylaxis (asparaginase)

●Prophylactic anticoagulation – Thromboprophylaxis with LMW heparin can be administered to individuals treated with asparaginase, as long as their platelet count is ≥30,000/microL, as outlined in guidance from the ISTH [96]. Outpatient thromboprophylaxis may be appropriate in selected individuals who are especially high-risk.

A retrospective study involving 214 adults with Philadelphia chromosome negative acute lymphoblastic leukemia (Ph- ALL) treated with asparaginase compared outcomes in 125 who received prophylactic LMW heparin versus 99 individuals who did not receive prophylactic anticoagulation [111]. The VTE risk was substantially reduced with anticoagulation (14 percent in the prophylaxis group, versus 27 percent in the group that did not receive LMW heparin). No episodes of major bleeding were reported.

●Prophylactic antithrombin – Practices differ among experts regarding monitoring of AT levels and prophylactic AT administration in a patient receiving asparaginase who has not had thromboembolism. Some experts measure AT levels and give prophylactic AT; others do not give AT prophylactically. (See "Thromboembolism in children with cancer", section on 'Prevention'.)

We do not administer AT for primary prophylaxis in patients who have not had thrombosis. However, we do administer AT to patients who have had a thromboembolic event while receiving asparaginase and who require continued treatment with the drug. (See 'VTE treatment (asparaginase)' below.)

Evidence for a potential benefit from prophylactic AT administration is limited.

•Children, randomized trial – The Prophylactic Antithrombin Replacement in Kids with ALL treated with L-Asparaginase (PARKAA) trial randomly assigned 85 children and adolescents to receive or not receive prophylactic AT replacement during induction chemotherapy [112]. Those assigned to the AT arm received AT concentrate once weekly for four weeks, dosed to produce a peak of supratherapeutic plasma AT activity (to 300 to 400 percent). Children who received AT replacement had trend toward a lower incidence of thrombosis (7 of 25 versus 22 of 60; 28 versus 37 percent) that did not reach statistical significance. There were no major adverse events attributed to therapy. This dose interval of once weekly was deemed not optimal because a quarter of patients in the AT arm had plasma trough levels drop below the lower limit of normal. A larger multicenter trial (Thrombotect) was initiated but results are unavailable [113,114].

•Children, observational study – A cohort of 112 children with acute lymphoblastic leukemia (ALL) receiving asparaginase was treated with AT supplementation (target, plasma AT levels >50 percent) plus prophylactic low molecular weight heparin [115]. Compared with historical controls with ALL on asparaginase who received AT alone, the children who received combined prophylaxis had a lower risk of thromboembolism (0 versus 13 percent). Over half of the combined therapy cohort required one, two, or three doses of AT to maintain plasma levels >50 percent.

As an alternative approach, some experts measure plasma fibrinogen levels and administer Fresh Frozen Plasma (FFP) or Cryoprecipitate to reduce the risk of bleeding as well as thrombosis; FFP will supply a small amount of AT but is unlikely to raise levels to the normal range. The concentration of AT in Cryoprecipitate is similar to FFP. In a retrospective comparison of two cohorts of pediatric patients who were receiving asparaginase-containing therapy for ALL, the rate of cerebral vein thrombosis was lower in those who were treated with FFP or Cryoprecipitate compared with those who were not (0 of 240 versus of 7 of 479; 0 versus 1.5 percent) [116]. In the replacement cohort, criteria for giving FFP or Cryoprecipitate were an AT activity <50 U/mL and/or a fibrinogen level <100 mg/dL; this resulted in AT administration to nearly all of the patients. FFP was given to 37 percent and Cryoprecipitate was given to 68 percent; some children received both products. Chemotherapy was identical for the two cohorts, and other outcomes (remission rates, bleeding) were similar. The finding of similar remission rates makes the theoretical risk of reduced chemotherapy efficacy due to asparagine in plasma less of a concern. The authors suggested that prophylaxis might be most appropriate for patients with high-risk disease during induction chemotherapy. However, small series in which FFP was administered have not observed beneficial effects on in vitro coagulation parameters [117,118].

Some patients may not be able to receive AT replacement due to cost. Further studies of the role of prophylactic AT administration are needed.

VTE treatment (asparaginase)

●Anticoagulation – Anticoagulation is appropriate for patients with VTE in the setting of asparaginase. The approach to anticoagulation is similar to other patients with VTE, with the caveat that bleeding risk may also be increased by thrombocytopenia and/or asparaginase-induced reductions in coagulation factors. These may be addressed by transfusion of platelets or of plasma products, respectively. (See "Platelet transfusion: Indications, ordering, and associated risks", section on 'Leukemia, chemotherapy, and HSCT'.)

The choice of anticoagulant and duration of therapy are addressed separately. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)" and "Cerebral venous thrombosis: Treatment and prognosis" and "Anticoagulation therapy for venous thromboembolism (lower extremity venous thrombosis and pulmonary embolism) in adult patients with malignancy" and "Venous thromboembolism: Initiation of anticoagulation".)

The use of a direct thrombin inhibitor rather than heparin for initial therapy has been proposed; the rationale is that the activity of direct thrombin inhibitors is not influenced by plasma AT levels (see 'Pathophysiology' above) [116]. However, evidence for the safety of this approach is lacking in many settings, especially for children.

The ISTH suggests LMW heparin for individuals who have (or are likely to develop) severe thrombocytopenia (platelet count <50,000/microL) [96]. Monitoring of anti-factor Xa activity is recommended due to variability of plasma AT concentrations. A DOAC is generally reasonable after thrombocytopenia resolves, although drug interactions should be reviewed. (See "Direct oral anticoagulants (DOACs) and parenteral direct-acting anticoagulants: Dosing and adverse effects".)

Anticoagulation is continued for at least six months in most cases [96].

●AT replacement – AT replacement is appropriate in addition to anticoagulation for individuals who develop VTE and for whom an adequate aPTT cannot be reached due to AT deficiency, or for those with high-risk VTE such as cerebral vein or cerebral sinus thrombosis [96]. (See 'Heparin resistance' below.)

For those who are adequately anticoagulated, the role of concomitant AT replacement is unclear. We suggest AT replacement, especially if it facilitates the completion of potentially curative chemotherapy. Other experts use FFP or Cryoprecipitate as a source of AT, and others use anticoagulation alone (without AT replacement), based on lack of high quality evidence that administering AT improves outcomes. (See "Thromboembolism in children with cancer", section on 'Asparaginase management'.)

Asparaginase should be held temporarily for high-risk thrombosis (cerebral venous or cerebral sinus thrombosis, large PE, proximal DVT, arterial thrombosis) [96]. In some cases, asparaginase may be resumed after the patient has been stabilized (approximately four weeks delay is appropriate). This decision requires clinical judgment, and input from an individual with expertise in hemostasis and thrombosis is advisable.

ECMO, CPB, or hemodialysis — Patients undergoing extracorporeal membrane oxygenation (ECMO) or cardiopulmonary bypass (CPB) have been treated with AT products to improve heparinization and reduce the risk of thromboembolism. As an example, in a multicenter cohort of 4210 children who received AT concentrate over a 10-year period, ECMO was the most common clinical setting in which AT was used, representing 44 percent of the cohort [119]. In this study, the age group most likely to receive AT concentrates was neonates younger than 30 days of age.

The appropriate dose of AT in the setting of ECMO is unknown. In a study of five pediatric patients receiving ECMO who were treated with recombinant AT (median age one month, range one day to three years), starting doses in the range of approximately 100 to 500 units/kg/day were used, but it took approximately 12 hours to raise AT activity levels to 80 to 120 percent, leading the authors to suggest that higher starting doses may be appropriate [120].

Rarely, patients with chronic kidney disease requiring dialysis will have concurrent AT deficiency. This may present a challenge in the management of hemodialysis anticoagulation because heparin may not effectively anticoagulate these individuals during the dialysis procedure (see 'Heparin resistance' above). In such patients, one option is to use a non-heparin anticoagulant such as argatroban. Case series have suggested that argatroban is well tolerated and provides adequate anticoagulation in individuals with hereditary AT deficiency who require hemodialysis [121,122]. A variety of other approaches to thromboembolism prophylaxis and/or treatment have been reported, including AT replacement, plasma transfusion, thrombolytic therapy, or substitution of peritoneal dialysis [123-125].

Use of AT in CPB is discussed separately. (See "Blood management and anticoagulation for cardiopulmonary bypass", section on 'Heparin resistance'.)

Heparin resistance — Heparin resistance is most commonly seen in patients receiving ECMO or undergoing cardiac surgery with CPB (see 'ECMO, CPB, or hemodialysis' above). AT replacement may be appropriate in this setting if it is in the best interest of the patient to receive a heparin anticoagulant [92,93,126]. (See "Blood management and anticoagulation for cardiopulmonary bypass", section on 'Heparin resistance'.)

Alternatively, some patients may be treated with a direct thrombin inhibitor (eg, argatroban), which does not require AT for its anticoagulant effect.

Some patients with heparin resistance due to AT deficiency may benefit from AT replacement. This was illustrated in a cohort of nine children with hereditary AT deficiency who were treated with heparin for a thromboembolic event; two required AT replacement therapy for anticoagulation to reach therapeutic levels [43]. AT replacement has also been reported in patients undergoing extracorporeal membrane oxygenation or cardiopulmonary bypass [92,93,126].

The decision to use AT replacement versus an alternative must be individualized according to the patient’s age, clinical status, and the reason for anticoagulation. Early involvement of the consultant with expertise in anticoagulation is advised. Management of heparin resistance is discussed in more detail separately. (See "Heparin and LMW heparin: Dosing and adverse effects", section on 'Heparin resistance/antithrombin deficiency'.)

Testing of family members and genetic counseling — Hereditary AT deficiency is an autosomal dominant condition, and in most families, siblings or children of an affected individual have approximately a 50 percent chance of inheriting the deficiency. (See 'Genetics' above.)

Testing of family members is appropriate if the family history is positive for thromboembolic events, especially if multiple family members have been affected or if thromboembolism occurred at an early age. We generally perform this testing around the time of puberty or later, and before starting oral contraceptives in women. Additional details are presented separately. (See "Screening for inherited thrombophilia in asymptomatic adults" and "Thrombophilia testing in children and adolescents".)

In contrast, we generally do not test family members of an individual who was found to have AT deficiency as an incidental finding when there is no family history of thrombosis.

PROGNOSIS — Overall, hereditary AT deficiency is not thought to be associated with a higher mortality rate than the general population, despite the increased risk of thromboembolic events and possible requirement for anticoagulant therapy [127,128].

Some reports have suggested that AT deficiency in the setting of disseminated intravascular coagulation (DIC) or sepsis confers a worse prognosis. However, we do not perform AT testing in these settings, nor do we consider AT levels an important prognostic factor; prognosis is much better estimated using other clinical variables and much better improved by appropriate therapy directed at the underlying disease.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Anticoagulation".)

SUMMARY AND RECOMMENDATIONS

●Definition and biology – Antithrombin (AT, previously called AT III, also known as heparin cofactor I) is a natural anticoagulant that inhibits thrombin (factor IIa), factor Xa, and other serine proteases (figure 1 and figure 2). Hereditary AT deficiency is autosomal dominant with variable penetrance. Several acquired conditions may lower AT levels, but the clinical significance of this finding is often limited. (See 'Pathophysiology' above.)

●Prevalence – Hereditary AT deficiency is infrequent, both overall and among inherited thrombophilias. The estimated general population prevalence is approximately 0.02 to 0.2 percent (table 2). The prevalence in individuals with venous thromboembolism (VTE) is approximately 1 to 7 percent. (See 'Epidemiology' above.)

●Clinical findings – Hereditary AT deficiency confers a greater risk of thrombosis than other hereditary thrombophilias. Some individuals never have a VTE; others may have life-threatening VTE as a child, young adult, or with a first pregnancy. Other individuals may come to medical attention because of kidney disease or heparin resistance. (See 'Clinical manifestations' above.)

●Evaluation – We do not test for AT deficiency in most patients with VTE. We may test individuals with a positive family history, thrombosis in childhood or early adulthood, heparin resistance, asparaginase therapy, or extracorporeal membrane oxygenation. When testing is done, plasma AT activity (also called AT-heparin cofactor assay) is the best first test. Testing is best performed when the patient has recovered from an acute VTE and is not receiving an anticoagulant. Plasma AT activity below the normal range (typically <80 percent) is consistent with AT deficiency. The differential diagnosis includes other inherited and acquired thrombophilias. (See 'Diagnostic evaluation' above.)

●Management

•Anticoagulation – Therapeutic anticoagulation is appropriate for individuals with hereditary AT deficiency who have a VTE (algorithm 1). Prophylactic anticoagulation may be used in certain high-risk settings such as pregnancy or surgery. (See 'VTE treatment (hereditary deficiency)' above.)

•AT concentrate – In some cases, AT replacement is also appropriate; available products include an AT concentrate purified from human plasma and recombinant human AT produced from the milk of transgenic goats.

•Specific populations

-Pregnancy – If an individual with hereditary AT deficiency develops a VTE during pregnancy, therapeutic dose anticoagulation is generally used with low molecular weight (LMW) heparin. (See 'Pregnancy' above and "Venous thromboembolism in pregnancy and postpartum: Treatment".)

-History of VTE – If an individual with a personal history of thrombosis or compelling family history (multiple individuals with thrombosis, individual with unprovoked thrombosis), we suggest administering AT concentrate just prior to and at delivery when anticoagulation cannot be administered (Grade 2C).

-Asparaginase – Optimal management during asparaginase therapy is unknown and may include prophylactic anticoagulation and/or AT replacement. (See 'Patients receiving asparaginase' above.)

-ECMO, dialysis, heparin resistance – Individuals with AT deficiency undergoing ECMO or hemodialysis, or those with heparin resistance, may benefit from AT replacement along with anticoagulation. (See 'ECMO, CPB, or hemodialysis' above and 'Heparin resistance' above.)

●First-degree relatives – Hereditary AT deficiency is autosomal dominant. First-degree relatives of an affected individual have a 50 percent chance of inheriting the variant. Testing first-degree relatives is appropriate if the family history is positive for VTE, especially if multiple family members are affected or if VTE occurred at an early age. (See 'Testing of family members and genetic counseling' above.)