Antithrombin Iii Deficiency

A number sign (#) is used with this entry because susceptibility to thrombophilia can be conferred by variation in the SERPINC1 gene, which encodes antithrombin III (AT3; 107300), on chromosome 1q25.

Description

Deficiency of antithrombin III is a major risk factor for venous thromboembolic disease. Two categories of AT-III deficiency have been defined on the basis of AT-III antigen levels in the plasma of affected individuals. The majority of AT-III deficiency families belong in the type I (classic) deficiency group and have a quantitatively abnormal phenotype in which antigen and heparin cofactor levels are both reduced to about 50% of normal. The second category of AT-III deficiency has been termed type II (functional) deficiency. Affected individuals from these kindreds produce dysfunctional AT-III molecules; they have reduced heparin cofactor activity levels (about 50% of normal) but levels of AT-III antigen are often normal or nearly normal (summary by Bock and Prochownik, 1987).

The 2 categories of antithrombmin III deficiency have been classified further. Type I (low functional and immunologic antithrombin) has been subdivided into subtype Ia (reduced levels of normal antithrombin), and type Ib (reduced levels of antithrombin and the presence of low levels of a variant). Type II (low functional but normal immunologic antithrombin) has been subdivided into subtype IIa (functional abnormalities affecting both the reactive site and the heparin-binding site of AT3); subtype IIb (functional abnormalities limited to the reactive site); and subtype IIc (functional abnormalities limited to the heparin-binding site) (summary by Lane et al., 1992).

Clinical Features

Egeberg (1965) described a pedigree in which persons in 3 generations had florid thrombophlebitis and other thrombotic disease associated with about half-normal levels of antithrombin III. He suggested that antithrombin III may be the same as heparin cofactor. Antithrombin deficiency in individual patients with severe venoocclusive disease and an impressive family history was also reported by Penick (1969) and by Nesje and Kordt (1970).

Marciniak et al. (1974) described a large kindred from eastern Kentucky that had an extensive history of recurrent venous thrombosis and pulmonary embolism. Nine persons in 3 generations showed low antithrombin III levels (26 to 49% of normal). Five others were suspected of having the biochemical defect. Male-to-male transmission was noted. They concluded that antithrombin III is the sole blood component through which heparin exerts its anticoagulant effect.

Tullis and Watanabe (1978) described the seventh reported family and suggested that familial hypercoagulability may be due, in some instances at least, to platelet antithrombin deficiency (with the serum deficiency representing a secondary defect).

A CRM+ form of antithrombin III deficiency was described by Sas et al. (1980). Not only does heparin require AT-III for its anticoagulant effect, but it also increases the turnover rate of AT-III. Both normal persons and persons with AT-III deficiency show a decrease in plasma AT-III levels when given heparin intravenously. In persons with AT-III deficiency the effect may lead to recurrent thrombosis despite heparin therapy.

The many special problems of pregnancy in women with AT-III deficiency were discussed by Nelson et al. (1985).

Wilson et al. (1987) found that 16 of 123 patients with acute mesenteric infarction (13%) had mesenteric venous thromboses. Of these, 6 patients could be studied for antithrombin III deficiency; deficiency was found in 3.

Aiach et al. (1987) described a family with a variant form of AT-III (107300.0012) that was apparently not associated with an increased incidence of venous thrombosis.

Johnson et al. (1990) described 2 sisters who at ages 27 and 40 had serious peripheral and CNS arterial thrombotic disease. Cigarette smoking was the only clear additional risk factor.

Rosendaal et al. (1991) found no evidence of excess mortality in 171 individuals from 10 families with either proven deficiency of AT-III or a 50% probability of being affected. They suggested, therefore, that a policy of prophylactic anticoagulation for patients with AT-III deficiency cannot be recommended.

Mitchell et al. (1991) proposed that the lower risk of thromboembolic complications in AT-III-deficient children may be due in part to a protective effect of elevated levels of alpha-2-macroglobulin (A2M; 103950) during childhood.

Heijboer et al. (1990) investigated the prevalence of isolated deficiencies of antithrombin III, protein C, protein S, and plasminogen in 277 consecutive outpatients with venographically proved acute deep vein thrombosis, as compared with 138 age-matched and sex-matched controls without deep vein thrombosis. They found deficiencies of 1 of these proteins in 23 (8.3%) of the patients as compared with 2.2% of controls. The positive predictive values for the presence of an isolated protein deficiency in patients with recurrent, familial, or juvenile deep-vein thrombosis, defined as the proportion of patients with the clinical finding who had a deficiency of 1 or more of the proteins, were 9%, 16%, and 12%, respectively. Heijboer et al. (1990) concluded that acute venous thrombosis in most outpatients cannot be explained by abnormalities of coagulation-inhibiting and fibrinolytic proteins and that information from the medical history concerning recurrent or familial venous thrombosis or the onset at an early age is not useful for identifying patients with protein deficiencies.

Pabinger et al. (1994) found that the probability for thrombosis was significantly higher in AT3-deficient females taking an oral contraceptive compared to AT3-deficient females who were not. In patients with protein C and protein S deficiency, there was no significant difference between the contraceptive and noncontraceptive groups. Pabinger et al. (1994) suggested that all contraceptives should be strictly avoided in these females and that AT3 measurement should be mandatory in female relatives of known AT3-deficient patients before starting contraceptives.

Mapping

Lovrien et al. (1978) found linkage of AT3 deficiency and Duffy blood group (FY; 110700) on chromosome 1 (lod score of 1.2 at a recombination fraction of 0.1 in males and 0.3 in females). Bishop et al. (1978) presented corroborating data on linkage with Duffy. The provisional assignment of antithrombin III deficiency to chromosome 1 by linkage to the Duffy blood group locus was confirmed (Bishop et al., 1982; Winter et al., 1982). For the linkage of AT3 deficiency and FY, Winter et al. (1982) found a combined maximum lod score of 4.2 at recombination fractions around 0.1. Two patients with deletions of 1q had half-normal levels of antithrombin III, suggesting that the AT3 locus lies in bands 1q22-q25.

Inheritance

Grundy et al. (1991) pointed out that although AT-III deficiency usually follows an autosomal dominant pattern of inheritance, a few patients with defective heparin binding have been shown to be homozygous for a lesion in the arg47 residue (see 107300.0003, 107300.0015).

Molecular Genetics

In the first family reported with thrombophilia due to AT-III deficiency by Egeberg (1965), Hultin et al. (1988) identified a mutation in the AT3 gene (107300.0001). The AT-III protein in patients with AT-III Oslo is decreased in both the immunologic and the functional assay.

Borg et al. (1988) identified a novel AT-III variant that showed defective heparin binding (107300.0016). The mutation was not associated with thrombophilia. This and other mutant forms of AT-III that showed a heparin-binding defect (e.g., 107300.0003 and 107300.0015) suggested that arginine-47 is a prime heparin-binding site in antithrombin. Borg et al. (1990) studied the basis of reduced heparin affinity.

In a patient with recurrent thrombophlebitis and AT-III deficiency, Koide et al. (1984) identified homozygosity for an arg47-to-cys mutation in the AT3 gene (107300.0003). Members of the family who were heterozygous for the mutation were asymptomatic.

In a patient presenting with recurrent venous thromboembolism, Aiach et al. (1988) identified a reactive site variant (107300.0018) in the AT3 gene. The mutation resulted in defective serine protease inhibition.

Population Genetics

Rosenberg (1975) placed the prevalence of AT-III deficiency at 1 per 2,000 and the frequency among hospitalized patients with recurrent or extensive thrombosis at 2 to 3%.

Harper et al. (1991) concluded that the frequency of antithrombin deficiency is about 5% among patients who present with venous thrombosis before the age of 40 years. About 2% of all such patients have a dysfunctional variant of AT-III.

In a survey of over 4,000 Scottish blood donors with a sensitive heparin cofactor assay, Tait et al. (1991) found an incidence of hereditary AT-III deficiency of 1 in 350 donors, most of whom were clinically asymptomatic.

Perry and Carrell (1996) estimated that AT-III deficiency has a prevalence of 1:630 in the general population and is found in 3 to 5% of patients with thrombotic disease.