|
|||||||||
|
|
|
|
|
|
|
|
|
|
|
CORRESPONDENCE Antithrombin is a single-chain glycoprotein of 432 amino acids.
It is the most important endogenous coagulation inhibitor and inhibits
several coagulation factors: IXa, Xa, XIa, XIIa, and thrombin (of which
the inhibition of thrombin probably is the most important). In 1992 Blajchman et al1 reported the Antithrombin
Stockholm point mutation in codon 392 (Gly to Asp) as a reactive site
(RS) type of mutation that results in a type II antithrombin
deficiency. It was found in a young woman on oral contraceptives who
presented with pulmonary embolism. The authors reported low
antithrombin activity measured with thrombin inhibition-based (72%-74%) and factor Xa inhibition-based (53%-56%)
methods using chromogenic peptide substrate assays.1
Several years later, much to our surprise, the proband and her 2 siblings had normal antithrombin activity in a routine factor Xa
inhibition-based test. Her 2 siblings had previously
presented with low levels but never suffered from any symptoms. Our present investigation shows that the 3 patients have normal antigen
levels measured with Lia antigen test.2 The
factor Xa inhibition-activity assay also gave normal values,
but decreased antithrombin activity values (about 60%) were found with
a thrombin inhibition-based activity test using bovine thrombin (Table
1). Both activity methods were based on
chromogenic substrates and were carried out in the presence of excess
heparin. Our results differ from the original Canadian
report1 especially concerning the factor Xa
inhibition-based test.
Chromogenic peptide substrate assays have been used for many years to measure antithrombin activity. These assays are based on either thrombin inhibition3 or factor Xa inhibition.4 For thrombin-based tests bovine thrombin should be preferred because human thrombin also reacts with heparin cofactor II and may lead to overestimation of antithrombin.5,6 In our present study we used bovine thrombin (Sigma Aldrich, St Louis, MO) and substrate S-2238 (Haemochrom Diagnostica, Molndal, Sweden) for the thrombin inhibition assay, which was modified for the Cobas Mira analyzer (Roche Diagnostics, Basle, Switzerland). For factor Xa inhibition we used the Coamatic LR kit (Haemochrom Diagnostica) with S-2772 and the automatic method for Thrombolyzer (Benhk Elektronik, Norderstedt, Germany). The Canadian study used human thrombin in the thrombin inhibition assay, which can explain the slight difference compared with our results. Other factors that can contribute to differences in test results are plasma amount, incubation time, and heparin concentration in the assays. These factors however can hardly account for a discrepancy between 55% (Canadian study) and 110% (present study) as in the case of the factor Xa inhibition method (Table 1). In the present study with the factor Xa-based method normal antithrombin activity was found in the patient with type II deficiency; her siblings also had low levels with the thrombin-based test. We would therefore like to stress that there is a possibility of not detecting all type II deficiencies with a factor Xa inhibition-based test, which is the most widespread routine method. In our investigation the factor Xa-based method overestimated antithrombin activity levels. According to our knowledge this has not previously been described and can cast some doubts about the validity of the use of this test as the test of choice when screening for antithrombin deficiency. Therefore we believe that both the bovine thrombin- and factor Xa inhibition-based tests, together with an antigen method, should be performed in patients with suspected antithrombin deficiency type II.
Johanna S. Ungerstedt, Sam Schulman, Nils Egberg, Jovan Antovic, and Margareta Blombäck
References
1.
Blajchman MA, Fernandez-Rachubinski F, Sheffield WP, Austin RC, Schulman S.
Antithrombin-III-Stockholm: a codon 392 (Gly-Asp) mutation with normal heparin binding and impaired serine protease reactivity.
Blood.
1992;79:1428-1434 2. Antovic JSJ, Karlman B, Blomback M. Evaluation of a new immunoturbidimetric test (liatest antithrombin III) for determination of antithrombin antigen. Clin Lab Haematol. 2001;23:313-316[Medline] [Order article via Infotrieve]. 3. Abildgaard U, Lie M, Odegard OR. Antithrombin (heparin cofactor) assay with "new" chromogenic substrates (S-2238 and Chromozym TH). Thromb Res. 1977;11:549-553[CrossRef][Medline] [Order article via Infotrieve]. 4. Odegard OR, Lie M, Abildgaard U. Antifactor Xa activity measured with amidolytic methods. Haemostasis. 1976;5:265-275[Medline] [Order article via Infotrieve]. 5. Demers C, Henderson P, Blajchman MA, et al. An antithrombin III assay based on factor Xa inhibition provides a more reliable test to identify congenital antithrombin III deficiency than an assay based on thrombin inhibition. Thromb Haemost. 1993;69:231-235[Medline] [Order article via Infotrieve]. 6. Conard J, Bara L, Horellou MH, Samama MM. Bovine or human thrombin in amidolytic at III assays: influence of heparin cofactor II. Thromb Res. 1986;41:873-878[CrossRef][Medline] [Order article via Infotrieve]   CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati What's this?
Related Article in Blood Online:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|||||||
 |
 |
 |
 |
 |
 |
 |
 |
||
| Copyright © 2002 by American Society of Hematology Online ISSN: 1528-0020 | |||||||||
