Factor V Leiden (FV R506Q) in families with inherited antithrombin deficiency

Thrombosis and Haemostasis © F. K. Schattauer Verlagsgesellschaft mbH (Stuttgart) 75 (3) 417-21 (1996)

pactor V Leiden (FV R506Q) in Families

\vith Inherited Antithrombin Deficiency

H.H. van Boven

, P.H. Reitsma

, F.R. Rosendaal

-

, T.A. Bayston

, V. Chowdhury

,

K,A. Bauer

,1. Scharrer

, J. Conard

, D.A. Lane

From the 'Department of Clinical Epidemiology, University Hospital Leiden, 2Hemostasis and Thrombosis Research Center, University Hospital Leiden, The Netherlands; 3Department of Haematology, Charing Cross and Westminster Medical School, London, Institute of Molecular Medicine, John Radcliffe Hospital, Oxford, UK; 5Beth Israel Hospital, Boston; USA; 0Johann Wolfgang Goethe Universität,

Frankfurt am Main, Germany; 7Hötel Dieu, Paris, France

Summary

We investigated the presence of the gene mutation of factor V, FV ^l)6Q or factor V Leiden, responsible for activated protein C resis-nce, in DNA samples of 127 probands and 188 relatives from 128 jinhes with antithrombin deficiency. The factor V mutation was iden-ned in 18 families. Nine families were avaiiable to assess the mode of

heiitance and the clinical relevance of combined defects.

The factor V and antithrombin genes both map to chromosome l. viregation of the defects on opposite chromosomes was observed in -ce families. Co-segregation with both defects on the same chromo-•;ne was demonstrated in four families. In one family a de novo muta-Ί1 of the antithrombin gene and in another a crossing-over event were ,· most likely explanations for the observed inheritance patterns.

In six families with type I or II antithrombin deficiency (reactive site pleiotropic effect), 11 of the 12 individuals with both antithrombin ..iiciency and the factor V mutation developed thrombosis. The me-jii age of their first thrombotic episode was 16 years (ränge 0-19); this Ow compared with a median age of onset of 26 years (ränge 20-49) 15 of 30 camers with only a defect in the antithrombin gene. One of e subjects with only factor V mutation experienced thrombosis at - '\ears of age. In three families with type II heparin binding site defi--•ncies, two of six subjects with combined defects experienced ""inbosis; one was homozygous for the heparin binding defect.

Our results show that, when thrombosis occurs at a young age in iihrombin deficiency, the factor V mutation is a likely additional risk Λν Co-segregation of mutations in the antithrombin and factor V ••"•i·, provides a molecular explanation for severe thrombosis in

>:ral generations. The findings Support that combinations of genetic ~i\ tactors underly differences in thrombotic risk in families with

'"inbophilia. itroduction

v\eral studies have established resistance to activated protein C äs "idependent and strong risk factor for thrombosis (1-4). Activated ic'n C resistance is associated with the same point mutation of the • '"r V gene, a G—>A Substitution at nucleotide position 1691 that

pre-TOfondeuce to: Dr. H. H. van Boven, Department of Clinical Epide-"->· University Hospital, Building l CO-P45, P.O. Box 9600, 2300 RC -" The Netherlands - FAX Number: +31715248122

dicts replacement of Arg(R) 506 by Gln(Q) (factor V Leiden) (5-7). Unlike other genetic risk factors for thrombosis, the frequency of the factor V Leiden alleie is high in the general population: 2-6% of normal individuals carry the factor V mutation (5, 8,9).

It is becoming increasingly clear that the combination of two (or more) mutant alleles of genes that encode proteins involved in the clot-ting mechanism determines the thrombotic risk in individuals (10). In homozygous individuals this combination consists of two mutant al-leles of one and the same gene. It was recently found that the homozy-gous state for the factor V mutation gives a high, 80-fold, increased risk of thrombosis (11). Likewise, the presence of a second genetic risk fac-tor might explain higher numbers or severe thrombotic events observed in some families with one other inherited risk factor. In a study of fam-ilies with a combination of protein C gene mutation and the factor V mutation, 73% of the carriers of two risk factors had experienced a thrombotic episode, compared to 31% (protein C) and 13% (factor V) of related carriers of a single defect (12). Finally, several families are reported in the literature in which the combination of two (or three) independent segregating risk factors for thrombosis is associated with severe thrombotic episodes (13,14,15).

Inherited antithrombin deficiency arises from an uncommon autoso-mal disorder and is associated with an increased risk of venous throm-boembolic disease. The gene encoding antithrombin has been localised to chromosome lq23-25 (16). With the human locus for the factor V gene mapped to chromosome lq21-25, only an estimated distance exists between 3 and 11 cM from the antithrombin gene (derived from the NIH/CEPH Collaborative Mapping Group linkage map and a recent study) (5, 17, 18). Antithrombin deficiency may be classified in two major types, based upon the results of functional and immunological assays. Type I deficiency is characterised by reduced functional and im-munological antithrombin levels, both approximately 50% of normal. Type II deficiency results from the presence of a functionally inactive protein, usually with approximately normal antigen levels. Type II deficiency may be subclassified into variants with reactive site (type II RS), heparin binding site (type II HBS) and pleiotropic effect (type II PE) defects (19). This classification is clinically important. The type II HBS variants are associated with only a mildly increased thrombotic risk, except when present in homozygous state (20). Excluding this sub-type (sub-type II HBS), about 50% (ränge 15 to 100%) of individuals with antithrombin deficiency are symptomatic in reported families (21).

Because of the high alleie frequency of the mutated factor V gene, it is likely that many members of families with antithrombin deficiency also carry this alleie. We therefore investigated DNA samples of a large

group of kindreds, registered with antithrombin deficiency in our cen-tres, for the presence of this combination. Detailed data of nine families were available to investigate the effect of the chromosomal location on the segregation pattern and to assess the clinical relevance of the com-bination of risk factors.

Patients and Methods

Patients and families. DNA of 127 probands and i 88 relatives was available

from 128 unrelated families with inherited antithrombin deficiency. 111 fami-lies (l 10 probands) presented with antithrombin type I or type II RS/PE defi-ciency and 17 families (17 probands) with antithrombin type II HBS deficien-cy. Diagnosis of inherited antithrombin deficiency was initially based on anti-thrombin activity levels of less than 80 %, and in the majority the genetic defect has been identified (see reference 22). Since each of the centres is a specialized reference centre for diagnosis and treatment of familial thrombophilia, most of the families registered were referred because of unexplained, recurrent throm-bosis occurring at young age. Notably, the probands and family members had been diagnosed over a period of time before the identification of the factor V mutation. The therefore unbiased Information of the occurrence of venous thrombosis was obtained by routine medical history, and included self-reported thrombophlebitis, deep venous thrombosis and pulmonary embolism.

Detection of the factor V Leiden mutation andmutations in the antithrombin '-»

gene. For all 315 DNA samples a 220 bp fragment of exon 10/intron 10 of !jf

the factor V gene was amplified, followed by digestion with the restric- f tion enzyme Mnl I, äs described previously (5, 12). The 220 bp fragment ' of a normal factor V gene is cleaved by Mnl I resulting in fragments of 37,67, i and 116 bp. When the factor V mutation is present, one Mnl I site is lost resulting in a 67 and 153 bp fragment. These fragments were visualised follow-ing electrophoresis on a 2 % agarose gel. The antithrombin gene mutations were investigated by PCR and genomic sequencing of affected exons, äs de-scribed (23,24).

Results

Screening for the factor V Leiden mutation was performed in DNA samples of 127 probands and 188 related family members. The factor V mutation was found to be present in 18 families. In 15 of these families the proband carried the factor V mutation in combination with a defect in the antithrombin gene. In three families the factor V mutation was identified in relatives of the proband.

Detailed medical histories with emphasis on manifestations of thrombosis were available for 47 individuals from nine families with both the factor V mutation and an antithrombin defect. These nine

Βτθ

θ

1 / 2

S1 i

.2 1 2 3 /" 4

Fig. l Pedigrees of nine families, in which both an antithrombin gene mutation (phenotypically in families two and three) and the factor V Leiden mutation are

segregating. Thrombotic Symptoms are indicated by a hatched right half of the Symbols. The presence of an antithrombin mutation by a solid lower left of the sym-bol, the presence of the factor V Leiden mutation by a solid upper left of the symbol. Homozygosity for the antithrombin mutation is indicated by a solid lower half of the symbol·. Individuais not tested are indicated by NT in the left of the symbol. Probands are indicated by an arrow

·.·· *>··£

jniilies coraprised six families with type I or type IIRS or PE deficien-ies (35 subjects) and three famildeficien-ies with type II HBS deficiency \l subjects). Eight of these 47 subjects carried an antithrombin -luiation, five the factor V mutation, 18 the combination, and 16 sub-Ais had no defects.

[he inhentance patterns of the factor V mutation and antithrombin eticiency are shown in Fig. 1. In pedigree one, two and presumably iree the defects segregated independently, apparently due to their pres-se on opposite chromosomes. Conpres-sequently, family members with \nh defects were observed only in one generation. In pedigree four, \e and six, the factor V mutation and the mutant antithrombin gene i-segregated over at least two generations. Similarly, in pedigree c\en two daughters appear to have inhented a chromosome from their

iher. (hat carries both defects.

Pedigrees eight and nine show complicated patterns of segregation. rslly, a de novo antithrombin mutation has occured in pedigree eight mdividual II. 1(19). It will become certain whether or not the new de-.Λ is linked to the factor V Leiden allele when family members in the ,\i generation are investigated. In pedigree nine, II. l inherited the two Stifts on opposite homologous chromosomes. Of the daughters, III. l 'icnted the chromosome carrying the factor V mutation and III.2 ,· Lhromosome with the antithrombin gene defect. In contrast, III.3 V'nted a chromosome (hat carries the factor V mutation äs well äs the iilhrombin gene defect, an A insertion in codon 169. The simplest •Variation for this inhentance pattern assumes a crossing over event at JOMS in the father, resulting in recombination onto one chromosome. In Table l the underlying defects and available clinical details are - cd of the nine families with combined defects. To assess the clinical ovance of the combined defects, we compared the histories of throm-M\ m the six families, with the type I or II (RS or PE) deficiencies, ih the histories from 30 subjects of nine other families. These nine •uilies were selected on the basis of the availability of clinical data. Λ were charactensed by antithrombin deficiency only, without con-'intant activated protein C resistance. In the nine families eight Jorlymg antithrombin gene defects are known (16, unpublished). In ^c families 12 of 23 camers of an antithrombin defect were sympto-ΊΙ and seven individuals with no mutations asymptomatic. The six '"!.es with type I and type II RS or PE deficiencies and factor V Λιΐιοη comprised 16 symptomatic family members: four of seven "ws of an antithrombin defect, one of five carriers of factor V mu-"n and 11 of 12 individuals with both defects experienced

thrombo-•Ml 11 family members with no defects were asymptomatic. ! l of 12 individuals with combined antithrombin and factor V gene AIS had developed thrombosis (92%). In pedigree four IV.2 inherit-Ίι>ιΙι defects, but she was asymptomatic at a current age of five years. -' median age at the first thrombotic episode for the 11 individuals

) both defects is 16 years (ränge 0-19). In the families 16 of the in -1 '0 antithrombin deficient individuals had experienced a thrombot-^isode (54%). The median age for 15 of these 16 carriers (for one •ige of onset is not known) of a single defect in the antithrombin "-' u äs 26 years (ränge 20-49). In Fig. 2 the age of the first

thrombot-'1|sude is shown for these 26 symptomatic individuals. One of five "«-'rs of only factor V mutation experienced thrombosis at the age of

^irs

1 'he three families with heparin binding defects, pedigrees five, 'mtl seven, two of six family members with both defects were ''ntomatic. One had experienced thrombosis at 28 years. The other '•-'J the factor V mutation but was also homozygous for the heparin ">v~ defect. It has been reported that at 14 months he developed

Table l Details of individuals from nine families with factor V Leiden and

antithrombin dceficiency Pedigree number 1 2 3 4 5 6 7 8 9 I I 12 Π.1 III 1 II.2 113 114 m.i III 2 11 12 Π.2 I 1 II I III 1 ra.3 m.6 rv.i W.2 rv.4 u 12 II. 1 M Π.1 Π.1 Π.4 I 1 EU I 1 Π.1 113 III 1 III 2 m.3 Sei M F F F M F F M M M F M M M M M M M F M M F M F F F F M M M M M F F M

Antithrombin deficiency Factor V Leiden* Narural history Reference assays or genetic defect Age first thrombosis

act% ag% (NR 80-120«) Type II (reactive site defect)

Arg393Cys normal Arg393Cys Arg393Cys Typel 41 50 117 104 57 51 50 59 69 57 Type! NT1 94 88 68 48 Type! NT NT 9788 G-A 9788 G-A 9788 G-A 9788 G-A 9788 G-A 9788 G-A Type H (hepann binding site)

Phe99Leu Phe99Leu Phe99Leu (homozygous) Type II (hepann bmding site)

Arg47Cys Arg47Cys Type II (heparin binding site)

Arg47Cys Arg47Cys Type II (pleiotropic cffects)

normal Phe402Ser Typel NT 5501+ A 5501 + A normal 5501+ A 5501+ A GG GA GA GG GG AG GG AG GG NT AA AG NT NT GA GA GA GA GA GA GA GG GA GA GA GA GA GA GA AG GG AG GG AG (28) recurrent ulcers no thrombosis recurrent DVTi, 10 years no thrombosis

(23)

DVT+PE5, 49 years DVT, 40 years

no thrombosis, current age 40 years DVT+PE, 16 years

DVT+PE, 26 years

(unpublished) recurrent thrombosis, 20 years no thrombosis, current age 44 years DVT+mesentenal thrombosis, 15 years

(24)

CVA**, 29 years phtebitis, MItt. PE, 29 years recurrent DVT, 17 years recurrent DVT, 13 years DVT, PE, 14 years sagittal smus thrombosis, 10 days no thrombosis, current age 5 years large DVT, 15 months (25) no thrombosis no thrombosis DVT+CVA. 14 months (unpublished) no thrombosis, current age 58 years no thrombosis, current age 28 years (unpublished) no thrombosis, current age 36 years recurrent thrombosis, 28 years

(19)

no thrombosis DVT, 16 years

(23)

Age in years at first thrombosis according to deficiency 50 40 30 20 10 26 YR OO· 00

Antithrombin FVLeiden + Äntithrombin deficiency deficiency

(15/30) (11/12)

Fig. 2 Age of first thrombotic episode of symptomatic subjects of

investigat-ed families with type I or type II (RS. PE) antithrombin deficiency. Solid sym-bols are symptomatic famiiy members of six families in which both defects seg-regated. Open Symbols for symptomatic famiiy members of nme families with type I deficiency only. The number of respectively symptomatic mdividuals and asymptomatic individuals are in parentheses. One of five symptomatic car-ners of factor V mmation experienced thrombosis at age of 40 years

Discussion

We screened famiiy members of 128 families with inherited anti-thrombin deficiency for the factor V Leiden mutation and identified 18 families in which both defects were present. Nine of these families were selected for further study on the basis of the ready availability of pedigree and clinical data. Two distinct inheritance patterns could be discerned for transmission of the defects. These differences arise from the presence of the two defects on either a single chromosome l, or on separate copies of this chromosome in the diploid somatic cell. If the defects segregate independently due to their presence on separate chro-mosomes, carriers of both defects are limited to one generation. If the defects are present on one chromosome, however, they co-segregate and individuals will inherit two risk factors in successive generations; this is vividly illustrated in pedigree four (Fig. 1). The co-segregation provides a molecular explanation for severe familial thrombosis extending for several generations.

As both genes are linked to the same region of chromosome l, re-combination can generate or terminate co-segregation of both defects. Nevertheless, it is unlikely that recombination occurred twice in pedi-gree seven, and we have concluded that co-segregation of both defects occurred with a single chromosome from individual 1.1. In pedigree

three, it can not been excluded that a de novo mutation in the antithrom-bin gene occurred in II. l. However, 1.1 experienced recurrent thrombo-sis, and received life-long prophylaxis with anticoagulants. Although he was not tested, it is more likely that II. l inherited an antithrombin gene defect from 1.1.

Individuals who are heterozygous for the heparin binding variant are not reported to have an increased risk of developing venous thrombosis (26). The numbers involved in the families with this deficiency subtype are too small to allow any statistical estimate of the risk for thrombosis due to the presence of the factor V mutation. However, the factor V mu-tation was present in both symptomatic famiiy members and it seems a potential additional risk factor in this deficiency subtype. Still, four of the six carriers of both defects were asymptomatic. This suggests either a moderate risk for these famiiy members, constituted of the risk of the heparin binding defect itself and the factor V mutation, or the presence of additional risk factors in the two symptomatic members. In support of this, the exceptional young age (14 months) and severity of thrombo-sis in one famiiy member is explained by the presence of such a third risk factor, the homozygosity for the heparin binding defect.

Our fmdings clearly support the contention that the presence of simultaneous genetic defects increased the risk for thrombosis in anti-thrombin type I and II RS or PE: 92% of carriers of two risk factors had experienced a thrombotic episode, compared to 54% of carriers of a single antithrombin gene defect. Our observed one of five sympto-matic individuals with only factor V mutation is in accordance with an extended famiiy study in which approximately 25% of the relatives with APC resistance were reported to have experienced thrombosis (27). Very striking was the early median age of onset for the camers of both defects, being 16 years (ränge 0-19). By comparison, in the families with antithrombin type I or type II (RS, PE) deficiency, the median age was 26 years (ränge 20-49) for symptomatic famiiy mem-bers. The recognition that combinations of distinct genetic risk factors might be responsible for severe thrombosis is relatively recent. While the combination of genetic risk factors has been described, it is worth noting that the present study shows a combination of defects that alter the function of both the heparan sulphate-antithrombin and the protein C/protein S/factor V pathways. It appears from the results presented here that simultaneous defects in these pathways result in severe manifestations of thrombotic disease.

The increased risk for thrombosis caused by the presence of two risk factors resulted in a familial tendency for thrombosis and therefore re-ferral and study of these families in our centres. It is for further study to unravel whether additional genetic and/or environmental risk factors determine the risk of thrombosis in the families without the factor V mutation, which were referred to our centres. In conclusion, when the factor V Leiden mutation segregates in antithrombin deficient families, thrombosis may occur at a very young age in individuals with both genetic defects. If factor V gene mutation is detected in families with antithrombin deficiency, famiiy investigation should be performed to determine whether both defects segregate on one chromosome because of a strongly increased risk for famiiy members in the next generation.

Acknowledgements

The work was supported by grants from the Special Trustees of Charing Cross and Westminster Hospitals and Medical School, and from the Wellcome Trust.

We are indebted to Dr. E. Briet and Dr. J.P. Vandenbroucke for the critical reading and discussion. We thank Dr. H. R. Büller and Dr. J. W. ten Cate (Department of Haemostasis and Thrombosis, Amsterdam) for permitting us to study two families with inherited antithrombin deficiency.

References

l Dahlbäck B, Carlsson M, Svensson PJ. Familial thrombophilia due to a p\iously unrecognized mechanism characterized by poor anticoagulant re-iponse to activated protein C: Prediction of a cofactor to activated protein C. PNAS 1993; 90: 1004-8.

1 Gnffin JH, Evatt B, Wideman C, Fernändez JA. Anticoagulant protein C pathway defective in majority of thrombophilic patients. Blood 1993; 82: 1989-93.

ι Svensson PJ, Dahlbäck B. Resistance to activated protein C äs a basis for

\enous thrombosis. N Engl J Med 1994; 330: 517-22.

\ Koster T, Rosendaal FR, de Ronde H, Briet E, Vandenbroucke JP, Bertina

RM. Venous thrombosis due to poor anticoagulant response to activated protein C: Leiden Thrombophilia Study. Lancet 1994; 342: 1503-6. i Bertina RM, Koeleman BPC, Koster T, Rosendaal FR, Dirven RJ, de Ronde

H, van der Velden PA, Reitsma PH. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature 1994; 369: 64-7.

h Greengard JS, Sun X, Xu X, Fernändez JA, Griffin JH, Evatt B. Activated

proteui C resistance causes by Arg 506 Gin mutation in factor Va. Lancet 1994; 343: 1361-2.

" Voorberg J, Roelse J, Koopman R, Büller H, Berends F, ten Cate JW, Mert-ens K, van Mourik JA. Association of idiopathic venous thromboembolism with single point mutation at Arg™ of factor V. Lancet 1994; 343: 1535-6.

- Ridker PM, Hennekens CH, Lindpaintner K, Stampfer MJ, Eisenberg PR, Miletich JP. Mutation in the gene coding for coagulation factor V and the risk of myocardial infarction, stroke, and venous thrombosis in apparently healthy men. N Engl J Med 1995; 332· 912-7.

1 Kontula K, Ylikorkala A, Miettinen H, Vuorio A, Kauppinen-Makelin R,

Hamäläinen L, Palomäki H, Kaste M. Arg506Gln Factor V muta-uon (Factor V Leiden) in patients with ischaemic cerebrovascular di-•.ease and survivors of myocardial infarction. Thromb Haemost 1995; 73: 558-60.

1 Miletich JP, Prescott SM, White R, Majerus PW, Bovill EG. Inherited pre-disposition to Thrombosis. Cell 1993;72:477-80.

Rosendaal FR, Koster T, Vandenbroucke JP, Reitsma PH. High risk of ihrombosis in patients homozygous for factor V Leiden (Activated Protein C-resistance). Blood 1995; 85: 1504-8.

- Koeleman BPC, Reitsma PH, Allaart CF, Bertina RM. Activated protein C lesistance äs an additional risk factor for thrombosis in protein C-deficient tamilies.Bloodl994;84: 1031-5.

; Gouault-Heilmann M, Quetin P, Dreyfus M, Gandrille S, Emmerich J, Leroy-Matheron C, Guesnu M. Massive thrombosis of venous cerebral Mnuses in a 2-year-old boy with a combined mherited deficiency of anti-thrombin III and protein C. Thromb Haemost 1994; 72: 782-6.

- Koller H, Stoll G, Sitzer M, Burk M, Schöttler B, Freund H-J. Deficiency of both protein C and protein S in a family with ischaemic strokes in young Jdults. Neurology 1994; 44: 1238-40.

15. Jobin F, Lessard M. Two cases of inherited triple deficiency in a large kin-dred with thrombotic diathesis and deficiencies of antithrombin UJ, heparin cofactor II, protein C and protein S. Thromb Haemost 1991; 66: 295-9. 16. Bock SC, Harris JF, Balazs I, Trent JM. Assignment of the human

antithrombin III structural gene to chromosome lq23-25. Cytogenet Cell Genet 1985; 39:67-9.

17. NIH7CEPH Collaborative Mapping Group. A Comprehensive genetic link-age map of the human genome. Science 1992; 258: 67-86.

18. Wang H, Riddell DC, Guinto ER, MacGillivray RT, Hamerton JL. Local-ization of the gene encoding human factor V to chromosome lq21-25. Genomics 1988; 2: 324-8.

19. Lane DA, Olds RJ, Conard J, Boisclair M, Bock SC, Hultin M, Abildgaard U, Ireland H, Thompson E, Sas G, Horellou MH, Tamponi G, Thein SL. Pleiotropic effects of antithrombin Strand l C Substitution mutations. J Clin Invest 1992; 90: 2422-5.

20. Finazzi G, Caccia R, Barbui T. Different prevalence of thromboembolism in the subtypes of congenital antithrombin III deficiency: review of 404 cases. Thromb Haemost 1987; 58: 1094.

21. Demers C, Ginsberg JS, Hirsh J, Henderson P, Blajchman MA. Thrombosis in antithrombin-III-deficient persons. Report of a large kindred and litera-ture review. Ann Intern Med 1992; 116: 754-61.

22. Lane DA, Olds RJ, Boisclair M, Chowdhury V, Thein SL, Cooper DN, Blajchman MA, Perry DJ, Emmerich J, Aiach M. Antithrombin III muta-tion database: First Update. Thromb Haemost 1993; 70: 361-9.

23. van Boven HH, Olds RJ, Thein SL, Reitsma PH, Lane DA, Briet E, Vandenbroucke JP, Rosendaal FR. Hereditary antithrombin deficiency: heterogeneity of the molecular basis and mortality in Dutch families. Blood 1994; 84: 4209-13.

24. Chowdhury V, Olds RJ, Lane DA, Conard J, Pabinger I, Ryan K, Bauer KA, Bhavnani M, Abildgaard U, Finazzi G, Castaman G, Mannucci PM, Thein SL. Identification of nine novel mutations in type I antithrombin deficiency by heteroduplex screening. Br J Haematol 1993; 84: 656-61. 25. Chowdhury V, Lane DA, Mille B, Auberger K, Gaudenberger-Bachem S,

Pabinger I, Olds RJ, Thein SL. Homozygous antithrombin deficiency: report of two new cases (99 Leu to Phe) associated with arterial and venous thrombosis. Thromb Haemost 1994; 72: 198-202.

26. Girolami A. Lazzaro AR, Simioni P. The relationship between defective heparin cofactor activities and thrombotic phenomena in AT III abnormal-ities. Thromb Haemost 1988; 59: 121.

27. Zöller B, Svensson PJ, He X, Dahlbäck B. Identification of the same factor V gene mutation in 47 of 50 thrombosis-prone families with inherited resis-tance to activated protein C. J Clin Invest 1994; 94:2521.

28. Ireland H, Lane DA, Thompson E, Olds R, Thein SL, Hach Wunderle V, Scharrer I. Antithrombin Frankfurt l: arginine to cysteine Substitution at the reactive site and formation of a variant antithrombin-albumin covalent com-plex. Thromb Haemost 1991; 65: 913.

Received September 6,1995 Accepted after revision November 24,1995