Thrombosis and Haemostasis © F. K. Schattauer Verlagsgesellschaft mbH (Stuttgart) 76 (6) 824-34 (1996)

Review Articles

Inherited Thrombophilia*: Part 2

David A. Lane1, Pier M. Mannucci2, Kenneth A. Bauer3, Regier M. Bertina4,Nikolay P. Bochkov5, Victor Boulyjenkov6, Mammen Chandy7, Björn Dahlbäck8,E. K. Ginter9, Joseph P. Miletich10, Frits R, Rosendaai4, Uri Seligsohn11

From the 'Charing Cross and Westminster Medical School, London, UK; 21RCCS Maggiore Hospital, Milan, Italy;3Beth Israel Hospital, Boston USA; 4Universitv Hospital Leiden, The Netherlands; 5Research Centre of MedicalGenetics, Moscow, The Russian Federation; 6WHO, Switzerland; 7Christian Medical College Hospital, Vellore, India;8University of Lund, Malmö, Sweden; 'Institute of Clinical Genetics, Moscow, The Russian Federation; 10WashingtonUniversity, St Louis, USA; ''Institute of Thrombosis and Hemostasis, Tel Hashomer, Israel

This article continues the review begun in a previous issue of theJournal, which considered the molecular basis and epidemiology ofthrombophilia.

Laboratory Investigation of Inherited Thrombophilia

What Laboratory Assays Should be Included?

In order to evaluate the contribution of genetic defects in the patho-genesis of thrombosis, the laboratory investigation should include afunctional APC-R lest, determination of the factor V genotype andmeasurements of plasma concentrations of PC, PS (total and free PS)and antithrorabin. In addition, a thrombin time will ensure that rarecases of dysfibrinogenemia are not overlooked. At present, availableresults suggest fibrinolytic parameters not to be of value in the evalua-tion of individual patients (2).

Before selecting the analytical procedure in the laboratory a numberof questions should be addressed such äs what are the sensitivity andspecificity of the laboratory assay for the genetic defect to be detected?Moreover, the assay performance such äs inter- and intra-day Variationare important factors to take into account. Quality control should be ofhigh priority. In the choice of assay it is cracial to consider possibleinfluence by other defects, e.g. APC-R influence on coagulation-basedfunctional assays for PC and PS (see below). An important parameter tobe taken into consideration is the predictive value of a positive and anegative lest. This is determined not only by the specificity and sensi-tivity of the assay but also by the prevalence in the population to bestudied for the particular defect. Thus, unless the specificity is close to100%, the predictive value of a positive lest is usually quite low fordefects which are rare in the population. Examples of this will be givenbelow.

Correspondence to: Dr DA Lane, Department of Haematology, CharingCross and Westminster Medical School, Hammersmith, London W6 8RF,UK - FAX: +181 846 7111; e mail: d.lane@cxwms.ac.uk

* Based upon the World Health Organisation report of a Joint World HealthOrganisation/International Society on Thrombosis and Haemostasis Meetingheld in Geneva, 6-8 November 1995. Published with permission of the WorldHealth Organisation. 'Meeting Rapporteur; 2ISTH Secretariat. 6WHO Secrelanat

Anticoagulant Protein Deficiencies

A detailed evaluation of the possible contribution of anticoagulantprotein deficiencies in the pathogenesis of thrombosis should includeanalysis of both protein levels with immunological assays and of thefunctional activity of each of these proteins. However, this is notalways possible and practical compromises have to be made. In gener-al, deficiencies of type I (low protein concentration) of any of the anti-coagulant proteins are much more common than type Π deficiencies(low functional activity but normal protein concentration). An optimalassay for each of these proteins is an easy and cheap functional assaywhich detects both type I and type II deficiencies. In the case of anti-thrombin, there are several commercially available functional assays(based on Inhibition of either thrombin or factor Xa) which fulfill thesecriteria. An International Standard for antithrombin is available.

The Situation for PC and PS is more complicated because the avail-able coagulation based functional assays are complicated to performand several of them have low sensitivity and specificity. Assaysdevised to measure the protein concentrations are often cheaper andeasier to perform than several of the commercially available functionalassays, and consequently such assays are used in most laboratories inthe primary evaluation of the patients. Both sensitivity and specificityof immunological äs well äs of functional assays are lower when usedto lest individuals on oral anticoagulation than when used to testpatients without anticoagulant therapy.

In the case of PC, depending on the site of the mutation there aremany different possible types of functional defects. The interactionwith the thrombin-thrombomodulin complex may for instance beimpaired with a resulting poor activation; mutations in the vitaminK-dependent region may lead to poor phospholipid binding ability;mutations in other parts of PC may lead to functional defects due topoor interactions with either Substrates or cofactors. There is at presentno easily available functional assay which monitors all functionalaspects of PC and consequently, the laboratory evaluation is a practicalcompromise. A functional, easy, assay is available for PC and it is usedin many laboratories äs an alternative to immunological assays in theprimary screening of the thrombosis patients. The assay includes acti-vation of PC in plasma with a PC activator (Protac) isolated from asnake venom and measurement of the proteolytic activity of the activat-ed PC against a synthetic Substrate. The assay, which is readily auto-mated, is available from several commercial sources. It shouid be bornein mind that the assay does not detect several of the type II detects. As

824

there is an overlap m the PC concentrations between normals and thosewith the genetic defects (3), sensitivity and specificity of all availablefunctional or immunological assays for the genetic defects causmg PCdeficiency are below 100% The Protac based assays perform quite welland have coefficients of Variation below 2% The sensitivity has beenestimated to be 85%, whereas the specificity is estimated to be 95% (3)When this assay is used to study a population m which the prevalenceis around 0 3%, the predictive value of a positive test is 5% whereas thepredictive value of a negative test 99 9% Thus, m screenmg of thegeneral population this assay is excellent to exclude PC deficiency,whereas a positive test is caused by a genetic defect m only 5% of thecases When the assay is used to test patients with thrombosis (preva-lence of genetic defect approximately 3%), the predictive value of anegative test is 34% whereas the predictive value to a positive test is99 6% The predictive value of a positive test is negatively related to thePC level, t e the lower the PC level the higher the predictive valueThere is an International Standard for PC available

At present, there are no functional PS assays which fulfill therequirements that can be put on an assay which is to be used m thepnmary evaluation of thrombosis patients The low specificity of theavailable functional assays is a major problem äs it has recently beenshown that functional assays for PS are mfluenced by the APC-R asso-ciated with the presence of the factor V 506Arg to Gin mutation (4) AsPS circulates m plasma both äs free protem (approximately 30%) and incomplex with the C4b-bmdmg protem (5), a relevant question is whichfraction of PS should be measured with the immunological assays, thefree form, the bound form or total PS The current provisional conclu-sion is that the free PS assays have higher sensitivity and specificity forthe genetic defects causmg PS deficiencies than assays for total PSThis is because there is a great overlap m the total PS levels betweennormals and those with the genetic defect (6) Although the prevalenceof PS deficiency m the general population is not well estabhshed it isreasonable to suspect that it is sirmlar to that of PC and the discussionabout the predictive value for a negative and positive test (see above) istherefore relevant also for PS There are several commercially availableassays for both free and total PS and an International Standard willbecome available

The APC-R Test

So far, practical chmcal expenence has mamly been obtamed withthe APTT-based test for APC-R (7) In an APTT reaction, the additionof APC results m degradation of both factor Villa and factor Va, whichdelays thrombin generation and hence the formation of the fibnnclot The factor V 506Arg to Gin mutation is associated with partialresistance to degradation of mutated factor Va by APC but themutated factor Va expresses normal procoagulant properties (8-10)Consequently, thrombin generation is not properly impaired m the pres-ence of APC, which results m less Prolongation of the clottmg time

As the APC-R test is based on a Standard APTT-reaction it is easy toperform The APTT-reaction is run m the presence and absence of acarefully standardised amount of APC, which is mcluded in the calciumchlonde solution used to mitiate the clottmg reaction (7) The clottmgtime is measured m the presence (a) and absence (b) of APC and a ratio(the APC-raüo) between the two clottmg times (a/b) is calculated If theassay is always done on the same Instrument and also m other respectsperformed under stnctly standardised conditions, the resultmg APC-ratios can be used äs they are (l l, 12) However, it has been observedthat different Instruments give different clottmg times and APC-ratiosobtamed on one type of Instrument cannot be directly compared with

those from another type (13) If results from different laboratones areto be compared, it is beneficial to normahse the APC-ratios agamstthe APC-ratio of a normal plasma pool (14) There are some nnportantconsiderations to take mto account m relation to such a normalisationprocedure Due to the high prevalence of APC-R m the general popula-tion a pool of normal plasmas is fairly hkely to contam plasma frommdividuals with APC-R and m accordance with this, it has mdeed beenfound that APC-ratios of normal plasma pools are always lower thanthe mean of the mdividual APC-ratios For laboratones not havmg thepossibihty to perform factor V genotypmg, a practical solution to theproblem may be to exclude plasmas with extreme APC-ratios from thepool. eg the 5 or 10% with the highest and the 5 or 10% with thelowest APC-ratios In this context, it is obvious that an InternationalStandard plasma would be useful

The handlmg of the plasma samples is probably of importance andthe quahty of the results depend on stnct standardisaüon The centnfu-gation of blood should ensure that the resultmg plasma is platelet poor(<1% of normal platelet count), because e\en a small contammation otplatelets affects the APC-ratio, m particular after freezmg and thawmgof the plasma, platelet contammation will lower the APC-ratio (15) Forroutme purposes it is recommended to centnfuge the plasma at 2000 gfor 20 mm at room temperature When the plasma is pipetted off, onlythe middle portion should preferably be utilised to avoid plateletcontammation by native or disrupted platelets

The APC-R test can be performed on both fresh and frozen plasmabut "like should be compared with like", i e \vhen fresh plasma is ana-lysed, the normal ränge should have been determmed with fresh plasma, when samples are frozen at -70° C or at -20° C they should becompared with normal controls handled in the same way In suppoit foithis concept, it was recently reported that freezmg of plasma results mdecreased APC-R (Coatest® APC™ Resistance from ChromogemxMolndal, Sweden, was used m the study) but of practical impoitancethe authors found no differences between freezmg tempeiatures(-70° C and -30° C tested) (14) When all variables are stnctly con-trolled, the APC-R test is rehable and gi\es good discnminationbetween normal and APC-R mdividuals The reported sensitivity andspecificity of the APC-R test for factor V 506Arg to Gin aie 85-95% orbetter Variation m endogenous levels of PC has of couise, no mflu-ence on the APC-R because a standardised amount of APC is added Inaddition, Variation of the endogenous free PS level withm the rängeexpected for heterozygous PS deficiency and normals has no or only avery mmor mfluence on the APC response m the APTT-based assay(7,14)

The original APC-R test has a sensiti\ it\ of 94% tor detection offactor V 506 Gin allele, whereas the specificin is approximately 85%(sirmlar results have been found foi the commercial assay from Chi omo-gemx) If this test is used for scieenmg ot a population m which theprevalence is around 10%, the negative predictive value is 99 7%whereas the positive predictive value is onh 16% In screenmg a popu-lation with thrombosis the negative predicme \dlue is 97% whereasthe positive predictive value is around 73% Thus a positive test shouldalways be confirmed by factor V genonpmg The sensitivity andspecificity of the APC-R test appears to be lelated to the quahty of theAPTT reagent

The original APC-R test is not lehable tor analvsis ot plasmas hommdividuals receivmg oral anticoagulants or hepann 01 it they aredenved from mdividuals with other coagulation detects such äs lupusanticoagulants or coagulation factor dehcienues ( 1 1 14) In plasmafrom patients on oral vitamm K antagomsts the dottmg times m thepresence of APC are usually quite long (clottmg mav not even occur m

825

such plasmas), which results in high ratios In order to allow analysis ofsuch plasmas, a modified APC-R test, m which sample plasma isprediluted with factor V deficient plasma before assay, was suggested(16-18) A predüution of l m 4 of sample plasma with factor Vdeficient plasma seems to provide sufficient amounts of VitaminK-dependent coagulation factors in the assay to yield normal basalAPTT This modification gives vahd results not only for patients onoral anticoagulation but has been found also to provide an improveddiscnmmation for the factor V 506 Arg to Gin mutation For some timethis modification has been evaluated, usmg plasmas from mdividualswith or without oral anticoagulant therapy and a 100% sensitivity andspeciftcity for the presence ot the factor V 506 Gin allele was obtamed,irrespective of the plasma origm Thus, the modified test is an excellentscreenmg test for the presence of the factor V 506 Gin allele

Analysis for the Factor V 506Ar g to Gin Mutation

The codon for Arg506 is positioned close to the exon-mtronboundary in exon 10ofthefactorVgene(19) Determination ot the G toA mutation mvolves amphfication of this nucleotide region eithei fromgenomic DNA or from mRNA The detection of the point mutation canbe made m many ways, e g by nucleotide sequencmg, by differenthybndisation techniques, by restnction enzyme cleavage or by allelespecific amplification (20-25) The methodology can be optirmsed toallow analysis of a large number of samples every day The rate limit-mg step is usually the preparation of patient DNA, even though rathersimple extraction procedures can be used It is important to recogmsethe nsk of contammation of PCR-based assay s and hence it is of utmostimportance to orgamse the work carefully and include both positive andnegative controls

The APC-R Test vet ms the Factor V Gene Mutation Test

Optimal evaluation of a smgle patient requires both the functionalAPC-R test and the factor V gene mutation analysis to be used, becausethe two methods provide complementary Information However, foreconomical and practical reasons it will not always be possible to doboth tests on all mdividuals Until more expenence is obtamed bymdividual mvestigators it seems that a rational approach would be toperform parallel APC-R tests m the absence and presence of factor Vdeficient plasma until confidence is obtamed with the latter Apart fiomallowmg analysis of plasma from mdividuals on oral anticoagulanttherapy, this approach will also decrease the need foi confirmatorygenetic testine

the other genetic defects are less mteresting to evaluate for prophylac-tic screenmg are the relatively low levels of sensitivity and specificitywhich gives low predictive value of a positive test Cost-benefitanalysis are required m the evaluation of prophylactic testmg

To date it has not been recommended to perform the laboratorvmvestigation m association with the acute thrombotic event mainlybecause most assays are difficult to Interpret and the therapeuticregimes may mfluence the results of the assays This recommendationmay need to be re-evaluated with the availabihty of the modifiedhighly specific and sensitive APC-R test (with factor V deticient plas-ma) and factor V genotypmg

Wheie Should the Laboiatoi~\ [nvestigatwn be Perfouned7

The specialised assays for anticoagulant protem deficiencies areprefeiably performed m specialised coagulation umts whereas it is amatter of debate where the APC-R testmg should be done Time willgive the APC-R testmg its nght role m practical medicme and it will atthis stage become obvious whether the testmg should be periormed notonly m specialised laboratories but also m major or smaller hospitals

Who Should be Tested?

This question is related to the questions what to test and why It isrecommendable that testmg is performed on patients with thrombosisand possibly also on their first degree family members It is vahd toconclude that theie aie at piesent no data to support general screenmgof the population for APC-R Whether prophylactic APC-R testmg isbeneficial m association with nsk situations is yet to be evaluated

Whv Should We Test for Genetic Defects Piedisposmgfot Thrombosis'

For each category of mdividuals it is impoitant to ask the questionwhy the testmg is performed The most pertment question is whetherthe results of the assays will affect the handlmg of the patients This istrue for both general screenmg and for mvestigation of thrombosispatients The lapid expansion of knowledge m this field will bring ne\vtherapeutic lecommendations for mdividuals with genetic predisposi-tion for thrombosis There are two major leasons for testmg of farmhmembers of patients with thrombosis the prophylactic and diagnosticpurposes The prophylactic aspects lelate to the possibilities to giveadequate advice to family members and the diagnostic purposes tomake rehable diasnosis of mhented delects

When Is itAppropnate to Perfotm the Laboratorv Evaluation7

At present most testmg is performed after a thrombotic episode andalso after discontmuation ot the oral anticoagulation Most of the assayshave their highest sensitivity and specificity for the genetic defects ifthis is stnctly followed Testmg for PC and PS is not rehable durmg oralanticoagulation and it is pieterable that the patient has been otf this kmdot therapy for at least 10 days In raie cases it is possible to discontmuethe therapy for 10 days dm mg which the patient receives heparm (Stan-dard or low molecular weight hepann) A sample can be drawn on themornmg ot the tenth day betöre the hepann admimstration

It will be mteresting to evaluate whether piophylattic testmg otAPC-R betöre surgeiy oral contraception, hospitahsation will bebeneticial Highly sensitive and specific assays for APC-R with highpiedictive value aie available and mexpensive The mam leasons why

Clinical Manifestations of Inhented Thrombosis

The most common climcal problem is deep vein thrombosis of thelower hmbs, with 01 without pulmonarv embohsm, accountmg forapproximately 90% of all the thrombotic episodes, Table l Unusualsites ot venous thrombosis, such äs those m the mesentenc or cerebralvems, account tor less than 5% of the total episodes in patients withantithiorabm PC 01 PS deficiencies m patients with APC-R thrombo-sis seems to occur less tiequently at these sites Supeihcial thiombo-phlebitis is moie trequent m patients with PC or PS dehciency andAPC R than m antithiombm deficient patients Even though a t öle foiPS deliciency äs a nsk tactoi tor artenal thiombosis has been postulat-ed, there is little current evidence that this or other heterozygousdetects ot the antiooagulant Systems increase the nsk öl artenal throm-bosis (28, 29)

826

Table l Clmical features of patients with inhented thrombophiha dueto defects m the anticoagulant pathways

Venous thromboembohsm (>90% of cases)- Deep vem thrombosis of lower hmbs (common)- Pulmonary embolism (common)- Superficial thrombophlebitis- Mesentenc vem thrombosis (rare but charactenstic)- Cerebral vem thrombosis (rare but charactenstic)Family history of thrombosis*First thrombosis usually at young age (<45 yrs )x

Frequent recurrences*Neonatal purpura fulmmans(homozygous protein C and protein S deficiency)

- All these features are less evident m patients with APC-R, whoappear to be less severely affected climcally

There is history of thrombosis at diagnosis m 50-60% of mdividualswith antithrombm, PC and PS deficiencies, with a 50% recurrence rate,the first thrombotic episode occurs äs early äs before 40 years inapproximately 80% of patients In antithrombm deficiency, the Overallnsk of venous thrombosis is considered greater than m PC or PSdeficiency (30), but contradictory results have been obtamed (29) Onthe other hand, mdividuals with APC-R appear to have a lesser tenden-cy to thrombosis than those with the defects of the naturally occumnganticoagulants There is history of thrombosis m only 23% to 31% ofcases with APC-R (l l, 24) and only 30% of them develop thrombosisbefore age 45 (31) The presence of APC-R however magmfies the nskof thrombosis m patients with antithrombm, PC and PS deficiencies(32-34), äs discussed earlier

In antithrombm, PC and PS deficiencies 32% to 50% of the venousthrombotic episodes occur when other nsk factors are concomitantlypresent (surgery, pregnancy, immobihzation) (26-29) In mdividualswith APC-R, the need of the existence of such nsk factors to tnggerthrombotic episodes appears to be greater (62%) than for the otherthrombophilic syndromes (24)

The nsk factors that are more often associated with the occurrence ofthrombosis are pregnancy, the puerpenum and surgery (24,26,27,29)In women with antithrombm deficiency, the frequency of thrombosisdunng pregnancy and the puerpenum is between 37% and 44%, m PCor PS deficiency, between 12% and 19% (35,36), m APC-R, 28% (37)Thrombotic episodes occur most frequently during the puerpenum,accounting for 60-75% of all the episodes comphcating pregnancy (35,36) A retrospective analysis of a large number of antithrombm, PC orPS deficient mdividuals gave an overall frequency of venous thrombo-sis comphcating surgery of 22%, with no sigmficant differences due tothe type of deficiency or surgical procedure (36) Intake of oral contra-ceptives is associated with an mcreased thrombotic nsk, parücularly mwomen with antithrombm deticiency and APC-R (38, 39) Patientswith dysfunctional defects of antithrombm, PC and PS have thrombot-ic nsks similai to those of the correspondmg quantitative defectsA notable exception is the antithrombm HBS type II subtype, with aprevalence ot thrombosis m these (heterozygous) cases of only 6%,contrasting with 52-68% m patients with other types of antithrombmdeficiencies (40)

Homozygous antithrombm deficiency is extremely rare and almostexclusively reported m patients with HBS defects These mdividualshave a severe thrombotic history of early onset, often affecting artenes

(41) Homozygous type I antithrombin deficiency is, presumably, m-compaüble with hfe in one report, two brothers with this defect diedwithm three weeks of birth (42) Homozygous PC deficiency has pecu-har phenotypic and chnical expressions [reviewed m (43)] In patientswith unmeasurable PC, purpura fulmmans. due to thrombosis of smallvessels with cutaneous and subcutaneous ischemic necrosis, may occursoon after birth or m the first year of hfe (44,45) In patients with verylow but measurable PC (5-20%), chnical mamfestations are milder andgenerally similar to those for heterozygous deficiency (46) Homo-zygous PS deficiency has been rarely reported, but is also associatedwith neonatal purpura fulmmans (47) Due to the high frequency of themutant factor V m the general population, homozygous APC-R isrelatively frequent, ~1 5000(48) Whether homozygotes have a nsk ofartenal thrombosis greater than that of the general population remamsto be elucidated (49,50)

Management of Inhented Thrombophilia

Acute Events

The management of acute venous thrombosis or pulmonary embo-lism m patients with inhented thrombophiha is generally not differentfrom other patients Thrombolytic therapy can be used m patients withmassive acute venous thrombosis or pulmonary embolism Hepannshould be mitiated with an intravenous bolus of 5,000 umts followed byan mfusion of 1400 umts per hour (51), or if a weight adjusted regimenis used, a bolus of 80 umts per kg body weight followed by an mfusionof 18 umts per kg per hour (52) The APTT should be performedapproximately 6 hours after therapy is mitiated and at least daily there-after to mamtam the clotting time in the therapeutic ränge For manycommercial APTT reagents, this conesponds to an APTT that is l 8 to3 0 times the mean of the noimal ränge or an anti-factor Xa heparmlevel of 0 3 to 0 7 umts per ml (53) For less sensitive APTT reagents,the therapeutic APPT ratio is l 5 to 2 0 (54) Warfarm can be startedwithm the first 24 h Heparm is contmued for at least 5 days (55) oruntil the prothrombm time is m the therapeutic ränge, namely an Inter-national Normahzed Ratio (INR) of 2 0 to 3 0

Patients with antithrombm deficiency can usually be treated success-fully with intravenous heparm (56), though m some situations unu-sually high doses of the drug are required to achieve adequate anticoag-ulation In antithrombm deficient patients receiving heparm for thetreatment of acute thrombosis, the adjunctue role of antithrombmconcentrate punfied from human plasma is not clearly defmed, äscontrolled tnals have not been performed (56) This product shouldprobably be admimstered when difficulty is encountered m achievmgadequate heparmization, or recurrent thrombosis is observed despiteadequate anticoagulation It is also reasonable to treat antithrombmdeficient subjects with concentrate before major surgenes or m ob-stetncal situations where the nsks of bleedmg from anticoagulation areunacceptable The manufactunng processes used to prepare antithrom-bm concentrate result m a product that is greater than 95% pure, theyalso mactivate the hepatitis B and C viruses and human immuno-dehciency virus I (57 58) Hence, it is preterable to admimster anti-thrombin concentrate rather than tresh trozen plasma

The mfusion ot 50 umts of antithrombm concentrate per kilogramot body weight (one umt is defmed äs the amount ot antithrombm mone ml of pooled noimal human plasma) \ \ i l l usually raise the plasmaantithrombm level to approximately 120rr m α congemtally deficientmdividual with a baselme level ot 50% (58-62) Plasma levels shouldbe momtored to ensuie that they remam above 80%, the admmistration

827

of 60% of the initial dose at 24 h mtervals is recommended to mamtainmhibitor levels in the normal ränge (62)

Due to the intrequent occurrence of coumann-mduced skin necrosis,it may be advisable to take special precautions when mitiating oral anti-coagulant treatment m a patient who is previously known or likely tohave PC deficiency Warfann should be started only when the patient isfully hepannized and the dose of the drug should be mcreased gradual-ly, stanmg from a relatively low level (e g , 2 mg for the first 3 days andthen m increasmg amounts of 2 to 3 mg until therapeutic anticoagula-tion is achieved) Patients with heterozygous PC deficiency anda history of wartann induced skin necrosis have been successfullyretreated with oral anticoagulants Here PC admmistration either in theform of fresh frozen plasma or PC concentrate provides protectionagamst the development of recurrent skin necrosis until a stable level ofanticoagulation is achieved (63,64)

After an episode of venous thrombosis or pulmonary embohsm, pa-tients are usually contmued on oral anticoagulants for 3 to 6 monthsRecent data indicates that the nsk of recurrence is greater m patientswith permanent äs opposed to temporary nsk factors for thrombosis(65, 66) and it is therefore appropnate that warfarm should be contm-ued for at least 6 months at an INR of 2 0 to 3 0 in patients withinhented thrombophilia

After 6 months of anticoagulant treatment for an acute thromboticevent, an assessment must be made äs to the relative benefit conferredby long-term anticoagulant therapy in preventmg future thrombo-embohc comphcations versus the potential side-effects, cost, and m-convemence for the patient Unfortunately there is a paucity of rehabledata regardmg the magmtude of the thrombotic nsk or the benefit ofanticoagulant treatment m patients with deficiencies of antithrombm.PC, or PS äs these are relatively uncommon disorders Due to the highfrequency of APC-R m patients presentmg with a first episode ofvenous thrombosis, rehable data is just emergmg regardmg the nsk ofrecurrence (67) At this time however. only general guidehnes (ratherthan rigid lecommendations) for managmg patients with the vanoushereditary defects predisposmg to thrombosis will be proposed

Inhented Tlu ombotic Disoideis and Recommendatwns Pertamingto Duratwn of Anticoagulant Tieatment

When a heterozygous patient with one of the hereditary thromboticdisorders is identified. tamily studies should be conducted sinceapproximatelv half of their first-degree relatives will be affectedAffected asyrnptomaüc mdividuals should receive counselmg regard-mg the imphcations of the diagnosis and advice regardmg Symptomsthat require immediate medical attention In women of child-beanngage, oral contraceptives are generally contramdicated m view of themcreased thrombotic nsk associated with the use of these medications,although individual circumstances need to be taken mto considerationThe replacement dose of estrogens that is admmistered to postmeno-pausal women is much lower than the contraceptive dose and has notbeen shown to mcrease the nsk of venous thrombosis in the generalpopulation (68) As theie are not currently data mdicatmg that thesemedications mcrease the nsk ot thrombosis m patients with a hereditarythrombotic disoider, postmenopausal estrogen leplacement is notabsolutely contramdicated m women who have a strong mdication forreplacement therapy

All biochemicallv atteited mdividuals should be carefully evaluatedpnor to surgical, medical or obstetnc protedures that carry an mcreasedthrombotic nsk These subjects should then receive appropnate prophy-lactic anticoagulation regimens 1t specitic concentrates are available

tor the patient's deficiency state, under some restncted circumstancesthese might also be admmistered to raise the plasma levels of the pro-tem to the normal ränge dunng the pen-operative penod

In patients with an inhented thrombotic disorder, the occurrence oftwo or more spontaneously occumng thromboembohc episodes oftenleads to the contmuation of oral anticoagulants for life. even thoughsome feel that the nsks of bleedmg could exceed those of recurrence ofthrombosis

Given that future events in an asymptomatic patient or m an individ-ual with only one pnor thrombotic episode cannot currently be accu-rately predicted and that there is a fimte nsk of bleedmg associated withwarfann therapy, recommendations relatmg to long-term anticoagula-tion are best mdividuahzed at the current time (69, 70) The climcalfeatures that should be considered in making this decision mclude1 Whether the thrombotic episodes were spontaneous or whether pre-

cipitatmg factors were present, e g , if a precipitatmg event such äs amajor abdominal Operation was present, it would be reasonable tomanage the patient without long-term oral anticoagulation after theacute episode was adequately treated

2 The sex and hfestyle of the individual, e g , situations where thesefactors may influence the decision-making process mclude womenof child-beanng age planmng to conceive, occupations that entailprolonged penods of immobihzation and thereby might be associat-ed with an mcreased nsk of thromboembohsm, jobs with higher thanaverage chance of trauma that might lead to thrombotic or bleedmgcomphcations

3 A history of thromboembohsm m other biochemically affectedmembers of the family, though marked mtra- and mter-familialheterogeneity has been observed m the phenotypic expression of themhented thrombotic disorders

4 The number, Sites, and seventy of thrombosis, e g , a patient whopieviously sustamed a massive pulmonary embolus is more likely toreceive long-term warfann than a subject who developed deepvenous thrombosis m a calf vem

Management ofPtegnancM

The management of pregnancies in women with hereditary throm-botic disorders poses special problems (71) äs formal studies evaluat-mg treatments and exammmg decision analyses are unavadable Theincidence ot thrombotic comphcations dunng pregnancy and the post-partum penod appears to be greater in women with antithrombm deti-ciency than m those with deficiencies of PC or PS (35) Recent dataalso indicates that 60% of women who develop a first episode of venousthrombosis dunng pregnancy have a diagnosis of APC-R (10) Durmgpregnancy, adjusted-dose heparm admmistered by the subcutaneousroute is the anticoagulant of choice m many Centres because its effica-cy and safety for the fetus are estabhshed (72) (Centres with increasmgexpenence of low MW hepanns, howe\ er, might prefer to use this classof agent, m view of potentially simpler laboratory momtonng) Patientswith a history ot thrombotic episodes should receive treatment through-out pregnancy, while affected women with antithrombm deficiencvwho have not yet expenenced such events should probably receivetreatment Treatment of asymptomatic women with other hereditarythrombotic disorders should be considered on an individual basis

The dose and duration of heparm therapy m pregnancy (and mdeedeven its overall benetit) is uncertam äs uppropnately designed climcaltnals have not been pertormed m these patient populations Patientsconsidered to be at high nsk should lecene tull-dose heparm by sub-cutaneous mjection every 12 h tor the duration ot pregnancy The dose

828

of hepann should be adjusted to mamtain the 6 h postinjection-APTTl 5 times the control value In woraen considered to be at intermediatensk, lower doses of heparm can be used (5,000 to 10,000 units subcuta-neously every 12 h) and therapy can be started dunng the second orthird trimester and contmued for approximately 6 weeks mto the post-partum penod Low nsk patients can be observed closely throughoutthe pregnancy with duplex ultrasound imagmg of the leg vems at regu-lär mtervals

In women who are planning pregnancy while chronically taking oralanticoagulants, several approaches can be taken to mmimize the nsk ofboth thrombotic comphcations and warfann erabryopathy One is tostop warfann and commence subcutaneous hepann therapy this poten-tially exposes the patient to many months of hepann therapy and thensk of osteoporosis while she is trymg to conceive An approach inwomen with antithrombm deficiency is to use antithrombm concen-trates until conception This product, however, is costly and needs to beadmimstered mtravenously at frequent mtervals Fmally, warfann therapy could be contmued with the performance of pregnancy tests on afrequent basis As soon äs pregnancy is diagnosed, and pnor to the sixthweek of gestation, oral anticoagulants must be discontmued and hepar-m therapy mitiated Though the nsk of warfann embryopathyappears to be quite small dunng the first six weeks of pregnancy (73),even the small nsk of this complication makes this the least preferableof the three approaches

Coumann-mduced Skm Necrosis and Neonatal Purpurn fulmmans

A clear association has been estabhshed between the rare complica-tion of coumann-induced skm necrosis and hereditary PC deficiency(63,74) About a third of patients with coumann-mduced skm necrosiswill prove to have hereditary PC deficiency (75) This complication hasalso been descnbed m a patient with heterozygous PS deficiency (76)As coumann-induced skm necrosis is a rare complication, therapy hasbeen guided pnmanly by knowledge regardmg its pathogenesis Thediagnosis should be suspected m patients with painful, red skm lesionsdeveloping withm a few days after the Initiation of the drug and imme-diate Intervention is required to prevent rapid progression and reducecomphcations Therapy should consist of immediate discontmuaüon ofwarfann, admmistration of Vitamin K, and mfusion of hepann at thera-peuüc doses Lesions, however, have been reported to progress despiteadequate anticoagulation with hepann In patients with hereditary PCdeficiency, the admmistration of a source of PC should be senouslyconsidered, and it may also be appropnate m other patients with warfarin-mduced skm necrosis äs they mvanably have reduced plasma levelsof functional PC when the skm lesions first appear Fresh frozenplasma has been used, but improved results can be expected with theadmmistration of a highly punfied PC concentrate, which facilitates therapid and complete normahzation of plasma PC levels (77)

The management of neonatal purpura fulmmans m association withhomozygous or doubly heterozygous PC deficiency is more complicat-ed and hepann therapy äs well äs antiplatelet agents have not beenshown to be effective (44,78-80) The admmistration of a source of PCappears to be cntical m the initial treatment of these patients Freshfrozen plasma has been used with success to treat these mfants How-ever, the half-life of PC m the circulation is only about 6 to 12 hours(81,82), and the admimstration of plasma on a frequent basis is limitedby the development of hyperprotememia, hypertension loss of venousaccess, and the potential for exposure to mfectious viral agents A high-ly punfied concentrate of PC has been developed and is efficacious mtreatmg neonatal purpura tulminans (83) Wartarm has been admims-

tered to these mfants without the redevelopment of skm necrosis dunngthe phased withdrawal of fresh frozen plasma mfusions (44,80,84-86),and this medication has been used chronically to control the thrombot-ic diathesis A 20 month old child with hver failure and homozygousPC deficiency has undergone successful hver transplantation whichnormalized his PC levels and resolved the thrombotic diathesis (87)Neonatal purpura fulmmans has been descnbed in association withhomozygous PS deficiency (88, 89)

Prenatal Diagnosis in Inhented Thrombophilia

Prenatal diagnosis can only be considered appropnate m possiblecases of severe thrombophiha where the fetus is expected to be homo-zygous or compound heterozygous for mhibitor deficiency Only m afew mstances has prenatal diagnosis been performed, m attempts toavoid severe thrombophiha The initial procedure used was fetal bloodsamplmg by cordocentesis dunng the 19th week of gestation m awoman who had previously delivered two newborns who died ofmassive thrombosis and purpura fulmmans due to homozygous PCdeficiency (90) At this stage of gestation the mean level of PC antigenis about 10 U/dl (90-92), and thus, care should be taken m distmguish-mg between results consistent with heterozygosity with a PC level ofabout 5 U/dl and homozygosity In two reports the mean level of fieePS antigen at 15-23 weeks of normal gestation was 27 and 38 U/dl,respectively (91, 92), and hence the diagnosis of homozygous PS defi-ciency by fetal blood samplmg should be easier than the diagnosis ofsevere PC deticiency Such a procedure has not so far been camed outm cases of suspected severe PS deficiency

A more direct and precise approach to prenatal diagnosis of severedeficiencies of PC, PS and antithrombm depends pnmanly on the identification of the mutaüon or mutations responsible for the disease m theproband and on devismg an easy method for its detection, i e , poly-merase cham reaction and restnction analysis or Southern analysis Suchwork can be performed m highly speciahzed centers that have a particular mterest m this issue For Identification of the mutaüon causmg PCdeficiency a scannmg method was devised usmg denaturant gradientgel electrophoresis (DGGE) of 13 PCR amphfied fragments that coverexon I and most of the PC codmg regions (93) A sirmlar approach wasalso used for detection of mutations causmg PS deficiency (94) Oncethese prerequisites have been met, prenatal diagnosis can be piovideddunng the 9-10th week of gestation by chonomc vülus samplmg andDNA analysis Such an approach was recently used m a family affect-ed by severe PC deficiency (95) and also m a tamily with antithrombmdeficiency type IIHBS, m which the fetus was potentially homozygousfor a 99 Leu to Phe Substitution and heterozygous for factor V 506Argto Gin (96) Prenatal exclusion of a severe thrombophiha may also becamed out by mdirect restnction fragment length polymorphism(RFLP) trackmg Such a procedure was used m a family m which aprevious mfant died of bilateral renal vem thrombosis and both parentswere found to be heterozygotes for PS deficiency (97)

linherited Thrombophiha in the Developing Countnes

The prevalence and morbidity of thrombophiha in different geo-graphic regions of the world may be difterent because of genetic racialdifferences or because the phenotypic expression of the disease is al-tered by environmental factors While several studies on mhentedthrombophiha have been published, these should be mterpreted withcaution, due to many problems mvolved m the execution ot the workFor example, there are concerns with quahtv control ot laboratory

829

assays, with objective diagnosis of thrombosis and with study designFurthermore, no large epidemiological studies on groups such äs blooddonors have been performed m developmg countnes and there aretherefore no reasonable estimates of the prevalence of the mhibitordeficiencies

Despite these reservations, a number of published reports haveattempted to address the issues of frequency of venous thrombosisfollowmg surgery (98-102) and of hereditary throbophiha (103-105)The avaüable data suggests that the prevalence of deep venous throm-bosis may be less m the developmg than m the developed countnesHowever, there are suggestions that patients who do get venousthrombosis are more hkely to have an underlymg mhibitor deficiencyAnother potentially mterestmg aspect is that early onset of stroke maybe associated with mhibitor dehciency (106) Further work is clearlyrequired to consohdate these vanous observations

Conclusions

1 At the present time, mutations in four genes are clearly Imked toincreased nsk for venous thromboembolism Many discrete mutationscause deficiencies of antithrombm, PC and PS that dimimsh the capac-ity to balance procoagulant activity One specific mutation m factor V506Arg to Gin has a similar impact by rendermg this procoagulantfactor resistant to proteolytic degradation With these four estabhshednsk factors for thrombosis, roughly 50% of the familial thrombophihascan be explamed Apparently a number of other genetic nsk factors sofar has escaped detection It is however unhkely that these will be foundamong deficiencies of plasminogen, heparm cofactor II, tissue factorpathways mhibitor or ß2 glycoprotem I Other candidates will needfurther evaluation (for mstance, dysfibnnogenemia thrombomodulmdefects and mhented hyperhomocystememia) It is mcreasmgly appar-ent that co-mhentance of more than one relatively mild thrombophilicnsk factor causes more severe clmical expression

2 Increasmgly, attempts are bemg made to assign a magmtude togenetic and acquired nsk factors The nsks estimates that have beenfound m various studies depend heavily on the way the subjects forthose studies were selected and do not necessanly apply to mdividualsselected differently In particular, results from studies among selectedfamihes with stnking thrombophiha are probably overestimatmg thensk when apphed to mdividuals who were found because of a smglethrombotic event The most strmgent selection cntena have been usedto ascertam farmlies with the rare abnormahties (PC, PS or antithrom-bm), which more readily explams the differences found m family stud-ies on these disorders äs compared to APC-R than a true difference mseventy Fmally, when gene-gene and gene-environment mteractionsare required to bring about thrombosis, there will be differences withmfamihes äs much äs between famihes

3 Laboratory evaluation of thrombophiha should mvolve the use ofassays with highest possible sensitivity and specificity for the geneticdefect that is to be detected Assays can be of immunological and functional nature, m the case of immunological assays it should be realizedthat cases with truly functional defective protems may not be detectedA practically useful approach should be taken and the selection ofanalytical procedures should be govemed by the aim of the mvestigation äs well äs of locally determmed tactors such äs prevalence ot thegenetic defects to be detected and availabihty ot techmcal suppoitBased on avaüable scientific Information, the laboiatory evaluationshould mclude measurements ot PC, total and tree PS, antithrombmand a functional APC R test which is sensitive and specific for the pres-ence of the tactor V 506 Gin allele

In the case of PC, assays which are based on its activation with thePC activator Protac and measurements of the active enzyme withsynthetic Substrate, fulfil required quality cntena At present no func-tional PS assays can be recommended for general screenmg of throm-bophilic patients Immunological assavs of total äs well äs ot free PSare recommended Recently published results mdicate free PS to be thebest marker for genetically determmed PS deficiency but furtherstudies are required before a recommendation only to measure free PScan be made The functional assays for antithrombm which are basedon hepann stimulated Inhibition of factor Xa are recommended forscreenmg of thrombophilic patients For initial screenmg of APC-R,functional tests are recommended The test can be impro\ed by dilutmgthe patient plasma m FV-deficient plasma As assays for PC and PShave distmctly lower sensitivity and specificity for the presence ofmhented deficiency durmg the acute thrombotic epidsode and oral anti-coagulation, it is at present recommended to perform the laboratorymvesügation for these components after discontmuation of the therapy

4 The clmical mamfestations of the defects of naturally-occumnganticoagulant Systems (antithrombm, PS and PC deficiencies, APC R)are similar In heterozygotes, typical mamfestations are those of venousthromboembolism, such äs deep-vem thrombosis of the legs, pulmonaryembohsm and superficial thrombophlebitis Visceral and cerebral veinthrombosis are rarer but quite typical for mhented thrombophiha Pa-tients with homozygous defects have usuallv more severe clmical mamfestations with an earlier age of onset Some mamfestations are quite typ-ical for homozygous defects, such äs skm necrosis and widespread neonatal thrombosis m PC and PS deficiencies There are prehminaiy datasuggesting that some homozygous defects (antithrombm type IIHBS de-ficiency) may be also associated with an increased nsk for artenal thrombosis m the young but moie data on this and other homozygous deficien-cies are warranted to establish am relationship with artenal disease

5 When a symptomatic patient with mhented thrombophiha dueto a known genetic defect is identified family studies should beconducted smce approximately half ot their first-degree relatives willbe affected Asymptomatic mdmduals who carry the genetic defectshould receive counsellmg regaidmg the imphcations of the diagnosisand Symptoms that reqmre medical attention In general, the management of symptomatic mdividuals with the genetic defect is similar tothat for symptomatic patients without an identifiable genetic defect Anexception is patients with neonatal purpura fulmmans m associationwith homozygous or doubly heterozygous PC deficiency m whom theadmimstration of a source of PC is cntical m initial treatment Giventhat future thrombotic events m patients with mhented thrombophihacannot be accurately predicted and there is a fimte nsk of bleedmgassociated with anticoagulant therapy recommendations relatmg tolong-term treatment are best mdividuahzed at the current time

6 Once an mdividual is defmed äs bemg affected by hereditarythrombophiha äs many family members äs possible are exammed forthe particulai defect detected m the proband and a pedigree is construct-ed Familv members who are found to be affected are counselled aboutthe nsk of thrombosis An evaluation of the potential nsk of birth ofseverely atfected newborns is usually carned out m famihes in whichmteimarriage is practised, and consequemlv counsellmg, extensivecamer detection and prenatal diagnosis is planned and executed

The target tamihes for prenatal diagnosis ot hereditary thrombophil-las are those famihes that had already been atflicted by mfants withseveie thrombosis due to homozygositv 01 compound heterozygosityfor PC PS or antithrombm detiueni\ äs well äs those famihesmentioned above In these target tamihes an attempt is made to detectthe responsible mutation(s), and to de\ise α simple method for their

830

detection, e g , PCR and restnction analysis or Southern analysis Thisis follovved by an extensive study of family members m the child-bear-ing age for camership and äs a consequence counselhng is providedPrenatal diagnosis is then based on DNA obtamed by chonomc villussampling or amniocentesis

In mstances m which the mutation cannot be identified, specificRFLP trackmg m family members and eventually in fetal DNA is analternative approach In still other mstances of PC or PS deficiencytetal blood sampling at 18-22 weeks of gestation is an Option smce dataon the levels ot these components in normal fetuses are available Forantithrombm such values are unavailable and thus at present fetal bloodsampling for affected famihes cannot be offered, unless the specificmutation has been identified

7 Deep vem thiombosis and pulmonary embohsm have a lowermcidence m the developing world when compared to the West, and thismay be due to a combmation of racial and environmental factors Thereare hmited studies with complete laboratory evaluation on inhentedthrombophiha from the developing countnes It appears that there is ahigher Chance of finding an underlymg genetic defect (PC PS and antithiombm deticiency) m patients with thrombosis m the developingworld Prelimmary data on APC-R suggests that this defect is rare mAsians, Afncans and Chinese

Recommendations

1 Recent reports m the hteratuie support the hypothesis that famihalthrombophiha is a multiple gene disorder and that the penetrance ofthe disease is higher m camers of multiple gene defects In relationto this it ii> to be expected that prophylactic and therapeutic measures need to be adjusted to the number of mdependent nsk factorspresent in an individual Therefore efforts should be intensified toidentify those genetic nsk factors that so far have escaped detection,so that these can be mcluded m diagnostic screenmg procedures

2 Guidelmes need to be developed for the use of specific laboratorytests m screenmg procedures aimmg at the Identification of individ-uals who carry a genetic nsk factor for venous thrombosis

3 Given the large number of patients that must be followed to detectnew abnormalities, collaborative international mvestigations withstandardized recruitment protocols should be encouraged

4 More Information Imking genetic causes of hvperhomocystememiato venous thromboembohsm is necessary

5 An estimate of the prevalence of PS deficiency m the populationwould help evaluate its relative nsk

6 More specific recommendations need to be developed for theclassification of hereditary PS deficiencv Within this context itneeds to be established whether so called type III PS deficiency isan mdependent nsk factor for venous thrombosis or a different phe-notype of type I PS deficiency

7 As recent data suggest that measuiement ot free PS may be morevaluable than total PS assays toi the diagnosis of PS deficiency, itis recommended that studies are pertormed which evaluate theperformance of different methods tor determmation of free PS

8 General screenmg of the population tor genetic defects of PC, PSand antithrombm can at present not be lecommended mainly basedon the low prevalence of these detects m the population and onthe low predictive value of a positive test It is recommended thatstudies are performed which address the question whether generalscreenmg for APC R (FV 506 Gin allele) e g betoie oral contra-ception, surgery or hospitahzation is beneticial toi the decision ontherapeutic and prophylaUic legimes

9 Further climcal studies should be orgamzed to evaluate whetherdetects causmg mhented thrombophiha are a cause of or contnbuteto artenal thrombosis

10 Heterozygous deficiencies of PC, PS or antithrombm should not bethe target ot prenatal diagnosis In view of the very low expectedtrequency of the severe homozygous or compound heterozygousdefects m PC, PS and antithrombm m the general population it isnot recommended to carry out population screenmg for camership

11 It would be preferable to establish reference centres where workup of the mutations and DNA based prenatal diagnosis can beperformed in cases with severe defects

12 It is necessary to obtam accurate data on the frequency and impactof thrombophiha in the developing world In order to do this mdi-vidual laboratones should be identified m different regions andthese should develop the necessary expertise to screen for anddocument the genetic defect responsible m association with WHOCollaboratmg Centres

13 For the present screenmg should be done on all patients with venousthromboembohsm m the context of a study to determme the per-centage of patients with hereditary thrombophiha m the developingworld where thrombotic disorders m general appear to have a lowprevalence Family studies should be done m all patients m whom agenetic defect is documented

14 Smce facilities for screenmg may be available only m referencecenters treatment should be mitiated without delay where appropnate and tests performed m the reference center after anticoagulantshave been discontmued

15 Data on the thrombotic nsk in patients and asymptomatic familymembeis with thrombophiha should be collected to determmewhether the nsk profile is different m the developing world

16 Careful documentation of the nsk of hemorrhage on anticoagulantsis necessary m order to determme the nsk benefit ratio of therapeu-tic Intervention in patients with thrombophiha

17 Recommendations for screenmg for hereditary thrombophiha at anational level m health care Services can be made only afteradequate data on the epidemiology, nsk of thrombosis and lesult oftherapeutic Intervention is available

18 Smce there is a higher mcidence of stroke m the young m thedeveloping world this population needs to be studied in order todetermme the number of patients who have thrombophiha äs theunderlymg cause

19 In order to mcrease awareness of mhented thrombophiha m devel-oped and developing countnes, it is proposed, m cooperation withthe ISTH, to select two centres for designation äs WHO Collaborat-mg Centres to improve diagnosis, climcal recognition and treatmentof related thrombosis These Centres will serve äs reference centresfor the appropnate WHO Regions and will improve education ofboth health professionals and the general pubhc

20 The next meetmg of WHO/ISTH experts should be orgamzed atone of the proposed centres for designation äs a WHO Collaborat-mg Centre, and focus on the climcal problems ot mhented thrombo-phiha A traming course on the diagnosis for thiombophilia forlocal doctois and invited participants from countnes withm theWHO Region should also be foieseen at the samc time

References

l Lane DA Mannticci PM Bauer KA Bertina RM Bochkcn NPBoulyjenkov V Chandy M Dahlback B Gmter EK Miletich JP Rosenddiil FR, Sdigsohn U Inhented thrombophiha Pait l Thiomb Haemost1996 76 651 62

811

2 Malm J, Aurell M, Nilsson IM, Dahlback B Thromboembohc disease-cntical evaluation ot laboratory mvestigation Thromb Haemost 1992.687-13

3 Allaart CF, Poort SR, Rosendaal FR, Reitsma PH, Bertma RM. Bnet EIncreased risk ot venous thrombosis in camers of hereditary protem Cdeficiency defect Lancet 1993, 341 134-8

4 Faioni EM Franchi F Asti D, Sacchi E. Bernardi F, Mannucci PM Re-sistance to activated protem C m mne thrombophilic famüies mterferencein a protem S tunctional assay Thromb Haemost 1993 70 1067-71

5 Dahlback B Protein S and C4b-bmding protem components mvolved mthe regulation ot the protem C anticoagulant sybtem Thromb Haemost1991,66 49-61

6 Zoller B Garcia de Frutos P, Dahlback B Evaluation of the relationshipbetween protem S and C4b-bmdmg protem isoforms m hereditary protemS deticiency demonstratmg type I and type III deficiencies to be pheno-typic vanants of the same genetic disease Bloodl995,85 3524-31

7 Dahlback B, Carlsson M, Svensson PJ Famihal thrombophilia due to apreviously unrecogmzed mechamsm charactenzed by poor anticoagulantlesponse to activated protem C prediction of a cofactor to activatedprotem C Proc Natl Acad Sei U S A 1993, 90 1004-8

8 Kalafatis M, Bertina RM, Rand MD, Mann KG Charactenzation of themolecular defect m factor VR506Q J Biol Chem 1995, 270 4053-7

9 Heeb MJ, Kojima Y, Greengard JS, Griffin IH Activated protem Cresistance molecular mechamsms based on studies usmg punfied Gln506-factorV Bloodl995, 85 3405-11

10 Dahlback B Physiological anticoagulation Resistance to activatedprotem C and venous thromboembohsm J Clm Invest 1994,94 923-7

11 Svensson PJ, Dahlback B Resistance to activated protem C äs a basis forvenous thrombobis N Engl J Med 1994 330 517-22

12 Koster T, Rosendaal FR, de Ronde H, Briet E, Vandenbroucke JP, BertmaRM Venous thrombosis due to poor anticoagulant response to activatedprotem C Leiden Thrombophilia Study Lancet 1993, 342 1503-6

13 Rosen S, Johansson K, Lindberg K, Dahlback B Mulücenter evaluation ofa kit for activated protem C resistance on vanous coagulation Instrumentsusmg plasmas from healthy mdividuals The APC Resistance StudyGioup Thromb Haemost 1994,72 255-60

14 de Ronde H, Bertina RM Laboratory diagnosis of APC-resistance acnti-cal evaluation of the test and the development of diagnostic cntenaThromb Haemost 1994,72 880-6

15 Shizuka R, Kanda T, Amagai H, Kobayashi I False-positive activated pro-tem C (APC) sensitivity ratio caused by freezing and by contarmnation ofplasma with platelets Thromb Res 1995,78 189-90

16 Jorquera JI, Montoro JM, Fernandez MA Aznar JA, Aznar J Modifiedtest for activated protem C resistance Lancet 1994, 344 1162-3

17 Trossaert M, Conard J, Horellou MH Samama MM, Ireland H, BaystonTA, Lane DA Modified APC resistance assay for patients on oral anti-coagulants Lancet 1994 344 1709

18 Trossaeit M, Conard J, Horellou MH, Samama MM Influence of storageconditions on activated protem C resistance assay Thromb Haemost 1995,73 163-4

19 Cnpe LD, Moore KD, Kane WH Structure ot the gene for human coagu-lation factor V Biochemistry 1992,31 3777 85

20 Beitina RM, Koeleman BP, Koster T. Rosendaal FR, Dirven RJ,de Ronde H, van der Velden PA, Reitsma PH Mutation in blood coagula-tion tactor V associated with resistance to activated protem C Nature1994 369 64-7

21 Gieengard JS, Sun X, Xu X, Fernandez JA Gutfm JH, Evatt B Activatedprotem C resistance caused by Arg 506 Gin mutation in tactor Va Lancet1994 343 1361-2

22 Zoller B, Dahlback B Lmkage between mherited resistance to activatedprotem C and lactor V gene mutation m venous thrombosis Lancet 1994,343 15368

23 Vooibeig J Roelse J Koopman R Buller H Beiends F, ten Cate JW,Meitens K, van Mounk JA Assouation ot idiopathic venous thiombo-

embohsm with smgle pomt-mutation at Arg 506 of factor V Lancet 1994,343 1535-6

24 Zoller B, Svensson PJ, He X. Dahlback B Identification of the samefactor V gene mutation m 47 out of 50 thrombosis-prone famihes withmhented resistance to activated protem C J Clm Invest 1994,94 2521-4

25 Gandnlle S Alhenc Gelas M, Aiach M A rapid screemng method for thefactor V Arg 506 to Gin mutation Blood Coag Fibnnolysis 1995 6245-8

26 Bnet E. Broekmans AW Hereditary protem S deficiency In Protein Cand related protems Bertina RM (eds) Churchill Livmgstone, Edinburgh1988 pp 203

27 Broekmans AW, Conard J Hereditary protem C deficiency In Protein Cand related protems Bertma RM (eds) Churchill Livmgstone, Edinburgh1988, pp 160

28 Demers C. Ginsberg JS, Hirsh J, Henderson P, Blajchman MA Thrombo-sis m antithrombin-III-deficient persons Report of a large kmdred andliterature review Ann Intern Med 1992,116 754-61

29 De Stefano V, Leone G, Mastrangelo S, Tnpodi A, Rodeghiero F,CastamanG Barbui T, Fmazzi G, Bizzi B, Mannucci PM Clmicalmam-festations and management of mhented thrombophilia retrospectiveanalysis and follow-up after diagnosis of 238 patients with congemtaldeficiency of antithrombm III, protem C, protem S Thromb Haemost1994,72 352-8

30 Fmazzi G, Barbui T Different mcidence of venous thrombosis in patientswith mhented deficiencies of antithrombm III, protem C and protem SThromb Haemost 1994, 71 15-8

31 Martmelli I, Faioni EM, Monzam ML, Mannucci PM Inhented resistanceto activated protem C (APC) clmical and laboratory features Blood 1994,84 supp l 84a

32 Koeleman BP, Reitsma PH Allaart CF, Bertina RM Activated protem Cresistance äs an additional nsk tactor for thrombosis m protem C-deficientfamihes Blood 1994, 84 1031-5

33 Zoller B, Berntsdotter A Garcia de Frutos P Dahlback B Resistance toactivated protem C äs an additional genetic nsk factor in hereditary defi-ciency of protem S Blood 1995 85 3518-23

34 van Boven HH, Reitsma PH, Rosendaal FR Bavston TA, Chowdhury V,Bauer KA, Scharrer I, Conard J, Lane DA Factor V Leiden (FV R506Q)m famihes with mhented antithrombm deficiency Thromb Haemost1996, 75 417-21

35 Conard J, Horellou MH Van Dreden P, Lecompte T, Samama M Throm-bosis and pregnancy m congemtal deficiencies m AT III, protem C orprotem S study of 78 women Thromb Haemost 1990, 63 319-20

36 De Stetano V, Leone G Mastrangelo S, Tnpodi A, Rodeghiero F, Casta-man G, Barbui T, Fmazzi G. Bizzi B. Mannucci PM Thrombosis dunngpregnancy and surgery m patients with congemtal deficiency of anti-thrombmlll protem C protem S Thromb Haemost 1994,71 799-800

37 De Stefano V, Mastrangelo S Paciarom K. Ireland H, Lane DA, Scirpa P,Bizzi B Leone G Thrombotic nsk dunng pregnancy and puerpenum inwomen with APC-resistance Effective subcutaneous hepann m prophy-laxis m a pregnant patient Thromb Haemost 1995,74 793-4

38 Pabmger I, Schneider B Thrombotic nsk ot women with hereditaryantithrombm III-, protem C and protem S-deticiency takmg oral contra-ceptive medication The GTH Study Group on Natural Inhibitors ThrombHaemost 1994, 71 548-52

39 Vandenbroucke JP, Koster T Bnet E. Reitsma PH, Bertma RM, Rosendaal FR Increased nsk ot \ enous thrombosis m oral-contraceptive userswho are camers of factor V Leiden mutation Lancet 1994, 344 1453-7

40 Fmazzi G, Caccia R Barbui T Different pre\ alence ot thromboembohsmm the subtypes of congemtal antithrombm III deticiency review of 404cases Thromb Haemost 1987 58 1094

41 Chowdhury V Lane DA. Auberger K Gandenberger Bachern S, PabmgerI, Olds RJ,ThemSL Homozvgous antithrombm deticiency reportoftwonew cabes (99 Leu to Phe) associated with artenal and venous thrombosisThromb Haemost 1994 72 198-202

832

42 Hakten M, Deniz U, Ozbag G, Ulutm ON Two cases of homozygousantithrombin III deficiency m a family with congenital deficiency ofATHI In Thrombosis and haemorrhagic disorders Senzmger H, Vmaz-zerH(eds) Schmut and Meyer GmbH, Wurzberg 1989 pp 177-81

43 Tnpodi A, Franchi F, Krachmalnicoff A, Mannucci PM Asymptomatichomozygous protein C deficiency Acta Haematol 1990. 83 152-5

44 Branson HE, Katz J, Marble R, Griffin JH Inhented protein C deficiencyand coumann-responsive chronic relapsmg purpura fulmmans m a new-bora infam Lancetl983,2 1165-8

45 Marlar RA, Montgomery RR, Madden RM Homozygous protein Cdeficiency In Protein C and related protems Bertina RM (eds) ChurchillLivmgstone, Edinburgh 1988, pp 182

46 Sharon C, Tinndelli MC, Mannucci PM, Tnpodi A, Mariam G Homozy-gous protein C deficiency with moderately severe climcal SymptomsThromb Res 1986,41 483-8

47 Mahasandana C, Suvatte V, Marlar RA, Manco Johnson MJ, Jacobson LJ,Hathaway WE Neonatal purpura fulmmans associated with homozygousprotein S deficiency Lancet 1990, 335 61-2

48 Rosendaal FR, Koster T, Vandenbroucke JP, Reitsma PH High nsk ofthrombosis m patients homozygous for factor V Leiden (activated proteinC resistance) Blood 1995, 85 1504-8

49 Holm J, Zoller B, Svensson PJ, Berntorp E, Erhardt L, Dahlback BMyocardial mfarction associated with homozygous resistance to activatedprotein C [letter] Lancet 1994, 344 952-3

50 Ridker PM, Hennekens CH, Lmdpamtner K, Stampfer MJ, Eisenberg PR,Müetich JP Mutation m the gene codmg for coagulation factor V and thensk of myocardial mfarction N Engl J Med 1995,332 912-7

51 CruickshankMK,LevmeMN,HirshJ,RobertsRS, SiquenzaM Astand-ard hepann monogram for the management of hepann therapy ArchIntern Med 1991,151 333-7

52 RaschkeRA,ReillyBM,GmdryJR,FontanaRR, SnmvasS Theweight-based hepann dosmg nomogram compared with a Standard care nomogram Ann Intern Med 1993,199 874-81

53 Bnll-Edwards P, Ginsberg JS, Johnstone M, Hirsh J Establishing a thera-peutic ränge for hepann therapy Ann Intern Med 1993,119 104-9

54 Bjornsson JD, Nash PV Vanabihty m hepann sensitwity of APTTreagents Am JChnPath 1986, 86 199-204

55 Hüll RD, Raskob GE, Rosenbloom D Hepann for 5 days äs comparedwith 10 days m the initial management of proximal venous thrombosis NEngl J Med 1990, 322 1260-4

56 Schulman S, Tengborn L Treatment of venous thromboembohsmm patients with congenital deficiency of antithrombin III ThrombHaemost 1992, 68 634-6

57 Hoffman DL Punfication and large scale preparation of antithrombin IIIAm J Med 1989, 87 23S-26S

58 Menache D, O'Malley JP, Schorr JB, Wagner B, Williams C, Alvmg BM.Ballard JO, Goodmght SH, Hathaway WE, Hultm MB Evaluation of thesafety, recovery, half-life, and climcal efficacy of antithrombin III(human) in patients with hereditary antithrombin III deticiency Coopera-tive Study Group Blood 1990 75 33-39

59 Mannucci PM, Boyer C, Wolf M, Tnpodi A, Larneu MJ Treatment ofcongenital antithrombin III deficiency with concentrates Br J Haematol1982, 50 33-9

60 Winter JH Transfusion studies m patients with famihal antithrombin defi-ciency half disappearance time of mfused antithrombin III and mfluenceot such mfusions on platelet hfe span Br J Haematol 1981, 49 449-53

61 Brandt P Observations dunng treatment of antithrombin III dehuentwomen m hepann and antithrombin III dunng pregnancy, partuntion, andabortion Thromb Res 1981,22 15-24

62 Schwanz RS, Bauer KA, Rosenberg RD, Kavanaugh EJ, Davies DC,Bogdanoff DA Climcal expenences with antithrombin III concentrate intreatment of congenital and acquired deticiency ot antithrombin Am JMed 1989, 87 53S-60S

63 Zauber NP, Stark MW Successful warfann anticoagulation despite pro-tein C deficiency and a history of wartarm necrosis Ann Intern Med 1986,104 659-60

64 De Stefane V, Mastrangelo S, Schwarz HP, Pola P, Flore R, Bizzi B,Leone G Replacement therapy with a punfied protein C concentratedunng mitiation of oral anticoagulation m severe protein C congenitaldeficiency Thromb Haemost 1993,70 247-9

65 Schulman S, Rhedm AS, Lmdmarker P, Carlsson A Lators G, Nicol PA companson of six weeks with six months of oral anticoagulation aftera first episode of venous thromboembohsm N Engl J Med 1995, 3321661-5

66 Levme MN, Hirsh J, Gent M Optimal duration of oral anticoagulanttherapy a randomised tnal companng four weeks with three months ofwarfann m patients with proximal deep venous thrombosis ThrombHaemost 1995,74 606-11

67 Ridker PM, Miletich JP, Stampfer MJ, Goldhaber SZ, Lmdpamtner K,Hennekens CH Factor V Leiden and nsks of recurrent idiopathic venousthromboembohsm Circulation 1995,92 2800-2

68 Devor M, Barrett Connor E, Renvall M, Feigal D, Ramsdell J Estrogenreplacement therapy and the nsk of venous thrombosis Am J Med 1992,92 275-85

69 Hirsh J, Pnns MH Samama M Approach to the thrombophihc patientIn Hemostasis and Thrombosis basic pnnciples and climcal practiceColman RW, Hirsh J, Marder VJ, Salzman EW (eds) Lippmcott, Phila-delphia 1994, pp 1543-61

70 Bauer KA Hypercoagulable states In Hematology basic pnnciples andclimcal practice Hoffman R, Benz EJ, Shattü SJ (eds) Churchill Livmg-stone, New York 1995, pp 1781-95

71 Hirsh J, Piovella F, Pmi M Congenital antithrombin III deficiency Inci-dence and climcal features Am JMed 1989, 87 (suppl 3B) 34S-38S

72 Ginsberg JS, Hirsh J Anticoagulants dunng pregnancy Ann Rev Med1989,40 79-86

73 Iturbe Alessio I Foneseca MC, Mutchimk 0, Santos MA Zajanas A,Salazar E Risks of anticoagulant therapy in pregnant women with artifici-al heart valves N Engl J Med 1986, 315 1390-3

74 Broekmans AW, Bertina RM, Loehger EA, Hofmann V, Klingemann HGProtein C and the development of skm necrosis dunng anticoagulant thera-py Thromb Haemost 1983,49 251

75 Broekmans AW, Teepe RGC, van der Meer FJM, Briet E Bertina RMProtein C (PC) and coumann-mduced skin necrosis Thromb Res 1986,6137 (abstract)

76 Friedman KD, Marlar RA, Houson JG, Montgomery RR Warfann-mduced skin necrosis m a patient with protein S deficiency Blood 1986,68 (suppl) 333a (abstract)

77 Schramm W, Spannagl M, Bauer KA, Rosenberg RD, Birkner B, LmnauY, Schwarz HP Treatment of coumann-mduced skin necrosis with amonoclonal antibody punfied protein C concentrate Arch Dermatol 1993,129 753-6

78 Estelles A, Garcia Plaza I, Dasi A, Aznar J, Duart M, Sanz G, PerezRequejo JL, Espana F, Jimenez C, Abeledo G Severe inhented "homozy-gous" protein C deficiency m a newborn mtant Thromb Haemost 1984,52 53-6

79 Sills RH, Marlar RA, Montgomery RR, Deshpande GN, Humbert JRSevere homozygous protein C deficiency J Pediatr 1984, 105 409-13

80 Yuen P, Cheung A, Lm HJ, Ho F, Mimuro J, Yoshida N, Aoki N Purpurafulmmans m a Chinese boy with congenital protein C detiuency Pedi-atncs 1986,77 670-6

81 Comp PC, Nixon RR Esmon CT Determination ot tunctional levels ofprotein C, an antithrombotic protein usmg thrombm-thrombomoduhncomplex Blood 1984, 63 15-21

82 Vigano S, Mannucci PM, Solmas S Bottasso B Mariam G Detrease mprotein C antigen and formation ot an abnormal protein soon atter startmgoral anti<.oagul.int therapy BrJ Haematol 1984 57 213-20

83 Dreyfus M. Magny JF, Bndey F, Schwarz HP Planche C, Dehan M,Tchernu G Treatment of homozygous protein C dehuency and neonatal

833

purpura fulmmans with a punfied protem C concenirate N Engl J Med1991,325 1565-8

84 Peters C, Casella JF, Marlar RA, Montgomery RR. Zmkham WH Homo-zygous protem C deficiency observations on the nature of the molecularabnormaluy and the effectiveness ot warfann therapy Pediatncs 1988 812726

85 Garaa Plaza I. Jimenez Astorga C, Borrego D Marty ML Coumarm pro-phylaxis tor tulmmant purpura syndrome due to homozygous protem Cdeficiency Lancet 1985, l 634-5

86 Hartman KR. Manco Johnson M. Rawlmgs JS Bower DJ, Marlar RAHomozygous protem C deficiency early treatment with warfann Am JPediatr Hematol Oncol 1989 11 395-401

87 Casella JF, Lewis JH, Bontempo FA, Zitelh BJ, Markel H, Starzl TESuccessful treatment of homozygous protem C deficiency by hepatictransplantation Lancet 1988, l 435-8

88 Pegelow CH. Ledford M, Young JN, Zilleruelo G Severe protem S defi-ciency m a newborn Pediatncs 1992, 89 674-6

89 Mahasandana C, Suvatte V, Chuansumnt A, Marlar RA, Manco JohnsonMJ, Jacobson LJ, Hathaway WE Homozygous protem S deficiency m anmfant with purpura fulmmans J Pediatr 1990 117 750-3

90 Mibashan RS, Millar DS, Rodeck CH, Nicolaides KH, Berger A, Sehg-sohn U Prenatal diagnosis of hereditary protem C deficiency [letter]N Engl J Med 1985, 313 1607

91 Mehssari E. Nicolaides KH, Scully MF, Kakkar W Protein S and C4b-bmdmg protem m fetal and neonatal blood Br J Haematol 1988, 70199-203

92 Moalic P, Gruel Y, Body G, Foloppe P, Delahousse B Leroy J Levels andplasma distnbution of free and C4b-BP-bound protem S m human fetusesandfull-teimnewborns Thromb Res 1988,49 471-80

93 Gandnlle S, Goossens M, Aiach M Scannmg method to estabhsh themolecular basis of protem C deficiencies Hum Mutat 1994,4 20-30

94 Gandrille S. Borgel D, Eschwege Gufflet V, Aillaud M, Dreyfus M.Matheron C. Gaussem P, Abgrall JF, Jude B, Sie P Toulon P, Aiach MIdentification of 15 ditferent candidate causal pomt mutations and threepolymorphisms m 19 patients with protem S deticiency using a scanmngmethod for the analysis ot the protem S active gene Blood 1995, 85130-8

95 Millar DS. Allgrove J. Rodetk C, Kakkar W, Coopei DN A homozy-gous deletion/msertion mutation m the protem C (PROC) gene causmgneonatal Purpura fulmmans prenatal diagnosis m an at-nsk pregnancyBlood Coagul Fibrmolvsis 1994, 5 647-9

96 Lane DA. Auberger K, Ireland H. Röscher AA, Them SL Prenatal dia-gnosis m combmed antithrombm and factor V mutation Bnt J HaematolΪ996, 94 753-5

97 Formstone CJ, Voke J. Tuddenham EGD Jenkms E. Clark J Layton M,Nicolaides K. Kakkar W, Cooper DC Prenatal exclusion ot se\ereprotem S deficiency by mdirect RFLP anaiv sis Thromb Haemostas 199369 931

98 Cunnmgham IGE. Young NK The mudence ot post operative deep \emthrombosis m Malaysia Br J Surg 1974, 61 482

99 Osime U Incidence ot post operative deep \em thrombosis m Nigenansusmg 125 I-labelled fibnnogen BrMed J 1978. 2 1607

100 Nandi P. Wong KP, Wei WI. Ngan H. Ong GB Incidence of postopera-tive deep vem thrombosis m Hong Kong Chinese Br J Surg 1980, 67251-3

101 Hassan MA, Rahman EA, Rahman IA Prostatectomy and deep vemthrombosis m Sudanese patients BrJ Surg 1974 61 650

102 Chummjarakij T Poshyachmda V Post operative thrombosis m Thaiwomen Lancet 1975. l 1357

103 Awidi AS, Abu Klialat M, Herzallah U, Abu Rajab A, Shannak MM, AbuObeid T, al Taher I, Anshasi B Hereditary thrombophilia among 217consecutive patients with thromboembohc disease m Jordan Am J Hema-tol 1993.44 95-100

104 Lm HW, Kwong YL Bourke C, Lam CK. Lie AK. Wei D, Chan LC Highmcidence of thrombophilia detected m Chinese patients with venousthrombosis Thromb Haemost 1994, 71 416-9

105 Chafa 0, Fischer AM Sternberg C. Menane F Millien C, Benabadji MThe relative frequency ot mhented mhibitor deficiency among 93 voungpatients with thrombophilia m Algeria [letter] Blood Coagul Fibrmolv sis1993,4 639-40

106 Martmez HR, Rangel Guerra RA, Marfil LJ Ischemic stroke due to deficiency of coagulation mhibitors Report ot 10 voung adults Stroke 199324 19-25

Received Apnl 16, 1996 Accepted after revision Julv 30, 1996

834