Postgrad Med J (1991) 67, 848 - 855 ©) The Fellowship of Postgraduate Medicine, 1991

Thrombosis Research Institute Workshop

Heparin and its low molecular weight derivatives: clinical potentialProceedings of a Thrombosis Research Institute Workshop, heldon 17 December 1990 at the Thrombosis Research Institute,Emmanuel Kaye Building, Manresa Road, London SW3 6LR

Introduction

M.F. Scully

Thrombosis Research Institute, Emmanuel Kaye Building, Manresa Road, Chelsea, London SW3 6LR, UK

The efficacy of heparin in the prophylaxis andtreatment of thrombosis is well established.' Aconsensus conference organized by the NationalInstitute of Health in 1986 concluded that heparinis the drug of choice for the prophylaxis ofpostoperative thrombosis2 and recent analysis ofaccumulated clinical data has vindicated its use inthe prevention of postoperative pulmonary embo-lism.3 Although heparin has been in use clinicallyfor 50 years its pharmacological properties areassociated with drawbacks and undesired sideeffects. The short half-life of heparin in vivorequires frequent administration - repeated dailydosages and continuous infusion. In a number ofclinical situations it is not completely effective,particularly in the case of orthopaedic surgery.4Side effects include the infrequent occurrence ofdirect and immune-related effects on platelets,5 andhaemorrhage. These arise from heparin's antico-agulant and antiplatelet properties and mustalways be considered especially in the surgical andpostsurgical situation.6

Heparin acts by controlling the activity of theproteolytic class of coagulation factors. Normallythe activity of these proteases is controlled by aplasma inhibitor, antithrombin III as indicated bythe numerous reports of a history of thromboticdisease in patients with deficiency of this protein.Heparin acts to potentiate the rate of interactionbetween antithrombin III and coagulation pro-teases and in this it is uniquely potent, causing, atconcentrations ofbetween 0.1 to 1.0 gg/ml, acceler-ation of the rate of interaction between antithrom-bin III and thrombin by 10,000 fold7 and betweenantithrombin III and factor Xa by 4500 fold.8

Unfractionated heparin is a mixture ofsulphatedcarbohydrate molecules and it was during earlyinvestigation of this molecular heterogeneity that afunctional heterogeneity was noted which led to theproposal that heparin therapy could be improvedby using low molecular weight forms. The observa-tion made by Andersson et al.9 was that, uponfractionation according to molecular weight, lowmolecular weight fractions had a reduced antico-agulant potency [measured as the ability to prolongthe activated partial thromboplastin time (APTT)or thrombin time of plasma] while retaining theability to prolong clotting times induced by addi-tion of factor Xa. Since factor Xa had been shownin vivo to be more thrombogenic on a weight forweight basis than thrombin,'" low molecular weightheparin could be considered to have an antithrom-botic potency (antifactor Xa activity) equivalent tonormal heparin but with reduced anticoagulantpotency (equated to reduced risk of bleeding).These considerations stimulated the developmentof low molecular weight heparin for therapeuticusell and numerous clinical trials have since beenreported.From the evidence to date a principal advantage

for using low molecular weight heparin resides inits pharmacokinetic properties: the clearance rate isslower than unfractionated heparin and a singledaily dose of subcutaneous (s.c.) heparin has beenproposed in most situations.'2 The efficacy wouldappear to be similar to unfractionated heparin.Current opinion as regards a reduction in bleedingrisk when using low molecular weight heparin isequivocal. However, the methods ofmanufacture ofthese forms of heparin differ from one pharma-ceutical company to another and the compositionofthe products varies; opinions about the tendencyof a particular product to cause less bleeding arecertain to form as the low molecular weightheparins are introduced for general clinical use.Correspondence: M.F. Scully, Ph.D.

by copyright. on A

ugust 21, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.67.791.848 on 1 S

eptember 1991. D

ownloaded from

THROMBOSIS RESEARCH INSTITUTE WORKSHOP 849

Although low molecular weight forms ofheparinhave been used in a number of European countriesfor some time it is only recently that they have beenregistered and become available for use in Britishclinics. For this reason the British Society forHaemostasis and Thrombosis proposed the organ-ization of a meeting under the title of 'Heparin andits low molecular weight derivatives: clinical poten-tial' which was held at the Thrombosis ResearchInstitute on December 17th, 1990.

In the following abstracts from that meeting it isas well to point out that dosage of these new low

molecular weight (LMW) heparins is on the basisof weight or as anti-factor Xa units (that is, asmeasured by the ability to increase the rate ofinhibition of coagulation factor Xa by plasma).This is unlike conventional, unfractionated (UF)heparin whose unit of activity is measured accor-ding to the ability to prolong the clotting time oftotally activated blood or plasma (whole bloodclotting time or APTT). LMW heparin productsare known by a variety of registered names whosesuppliers are given in the text.

Referemees

1. Ockelford, P. Heparin 1986. Indications and effective use.Drugs 1986, 31: 81-92.

2. Consensus conference: Prevention of deep vein thrombosisand pulmonary embolism. JAMA 1986, 256: 744-747.

3. Collins, R., Scrimgeour, A., Yusuf, S. & Peto, R. Reductionin fatal pulmonary embolism and venous thrombosis byperioperative administration ofsubcutaneous heparin. Over-view of results of randomized trials in general, orthopaedicand urologic surgery. N Engl J Med 1988, 318: 1162-1173.

4. Hampson, W.J.G., Harris, F.C., Lucas, H.H. et al. Failure oflow-dose heparin to prevent deep vein thrombosis after totalhip replacement. Lancet 1974, 1 795-797.

5. King, D.J. & Kettom, J.G. Heparin-associated throm-bocytopenia. Ann Intern Med 1984, 100: 535-540.

6. Morabia, A. Heparin doses and major bleedings. Lancet1986, i: 1278-1279.

7. Hoylaerts, M., Owen, W.G. & Collen, D. Involvement ofheparin chain length in the heparin catalyzed inhibition ofthrombin by antithrombin III. J Biol Chem 1984, 259:5670-5677.

8. Ellis, V., Scully, M.F. & Kakkar, V.V. The relative molecularmass dependency of the antifactor Xa properties of heparin.Biochem J 1986, 2238: 329-333.

9. Andersson, L.-O., Barrowcliffe, T.W., Holmer, E., Johnson,E.A. & Sims, G.E.C. Anticoagulant properties of heparinfractionated by affinity chromatography on matrix boundantithrombin III and by gel filtration. Thromb Res 1976, 9:575-583.

10. Wessler, S. & Yin, E.T. Experimental hypercoagulable stateinduced by factor X: comparison of the non-activated andactivated forms. J Lab Clin Med 1968, 72: 256-263.

11. Kakkar, V.V., Djazaeri, B., Fok, J., Fletcher, M., Scully,M.F. & Westwick, J. Low molecular weight heparins and theprevention of deep vein thrombosis. Br Med J 1982, 284:375-379.

12. Holmer, E. Low molecular weight heparin. In: Lane, D.A. &Lindahl, U. (eds) Heparin. Chemical and Biological Proper-ties, Clinical Applications. Edward Arnold, London, 1990,pp. 575-596.

Contributors

PROFESSOR U. ABILDGARRD, Medical Department, AkerHospital, 0154 Oslo 3, Norway.PROFESSOR L.-O. ANDERSSON, Kabi Pharmacia, SI 1287Stockholm, Sweden.DR T.W. BARROWCLIFFE, National Institute for Bio-logical Standards and Control, Blanche Lane, SouthMimms, Potters Bar, Hertfordshire EN6 3QF, UK.PROFESSOR B. BONEU, Laboratoire d'Hemostase, Centrede Transfusion Sanguine, 31052 Toulouse Cedex, France.PROFESSOR J. HARENBERG, Department of Medicine,Faculty of Clinical Medicine, Klinikum Mannheim,University of Heidelberg, Mannheim, FRG.PROFESSOR H.C. HEMKER, Department of Biochemistry,University of Limburg, Maastricht, The Netherlands.

PROFESSOR V.V. KAKKAR, Thrombosis Research Insti-tute, Manresa Road, Chelsea, London SW3 6LR, UK.PROFESSOR D.A. LANE, Department ofHaematology andMedicine, Charing Cross & Westminster Medical School,London, UK.PROFESSOR M. SAMAMA, Laboratoire Central d'Hema-tologie, Hotel-Dieu, Paris, France.DR M.F. SCULLY, Thrombosis Research Instititute,Manresa Road, Chelsea, London SW3 6LR, UK.DR D.P. THOMAS, National Institute for BiologicalStandards and Control, Blanche Lane, South Mimms,Potters Bar, Hertfordshire EN6 3QF, UK.

Abstracts

Heparin molecular heterogeneity - the developmentof the antifactor Xa concept

L.-O. Andersson

Normal, clinically used heparin is a heterogeneous sub-stance. It is heterogeneous both with respect to structure

and to molecular weight, containing components withMW from 5000 to 25000.

In 1976, two important discoveries were made. One wasthat heparin is composed of one third of molecules thathave high affinity (HA) heparin for the serpin antithrom-bin III and two thirds ofmolecules with low affinity (LA)heparin for antithrombin.'13 The HA heparin fractionhas very high anticoagulant activity and the LA fraction

by copyright. on A

ugust 21, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.67.791.848 on 1 S

eptember 1991. D

ownloaded from

850 THROMBOSIS RESEARCH INSTITUTE WORKSHOP

very low anticoagulant activity. In animal thrombosismodel experiments it was also later shown4 that the HAheparin fraction had strong antithrombotic activitywhereas the LA fraction was essentially inactive. Theother discovery3 was that heparin fractions of molecularweights below 5000 did have anticoagulant activity whenmeasured as ability to inhibit activated factor X. Pre-viously they were thought to be inactive as they did notaffect the activated partial thromboplastin time (APTT).This discovery provided the possibility to test a hypo-thesis, put forward earlier, that anti-factor Xa activitywould be a better measure of antithrombotic activitythan APTT and other global coagulation tests. Lateranimal studies5 showed, however, that anti-factor Xaactivity was a necessary but not sufficient condition forgood antithrombotic activity.

References

1. Lam, L.M., Silbert, J.E. & Rosenberg, R.D. The separation ofactive and inactive forms of heparin. Biochem Biophys ResCommun 1976, 69: 570-577.

2. Hook, M., Bjork, I., Hopwood, J. & Lindahl, U. Anti-coagulant activity of heparin: separation of high activity andlow activity species by affinity chromatography on immobil-ized antithrombin. FEBS Lett 1976, 66: 90-93.

3. Andersson, L.-O., Barrowcliffe, T., Holmer, E., Johnson, E.A.& Sims, G.E.C. Anticoagulant properties of heparin frac-tionated by affinity chromatography on matrix-bound anti-thrombin III and by gel filtration. Thromb Res 1976, 9:575-583.

4. Thomas, D.P., Merton, R., Barrowcliffe, T.W., Thunberg, L.& Lindahl, U. Effects of heparin oligosaccharides with highaffinity for antithrombin III in experimental venous throm-bosis. Thromb Res 1982, 47: 244-248.

5. Holmer, E., Mattsson, C. & Nilsson, S. Anticoagulant andantithrombotic effects ofheparin fragments in rabbits. ThrombRes 1982, 25: 475-485.

Action of heparins on the coagulation cascade

H.C. Hemker and S. Beguin

Heparins, at pharmacologically relevant concentrations,potentiate antithrombin III. They therefore enhance theinactivation of the serine proteases of the coagulationcascade. Factor VIIa (FVIIa), which is active only in atight complex with thromboplastin, is not inhibited,probably because only free proteases are attacked. Someinhibition of the contact factors does occur, but not to anextent that prevents activation of sufficient factor IX. Theinhibitions of possible importance for the action ofheparins in vivo, therefore, are those of thrombin, factorXa and factors IXa.

Unfractionated (UF) heparin, enhances the inactiva-tion of all three, but its action on factor Xa does not leadto a significant reduction of the velocity or prothrombinconversion, because factor Xa is formed in excess and theamount of prothrombinase is determined by the amountof factor Va. In the intrinsic system (APTT) the ratelimiting process is the feedback activation offactor VIII bythrombin. This process is inhibited by heparins because oftheir antithrombin action. This prolongs the lag time ofthe thrombin burst. During this lag time, factor IXa isscavenged, which inhibits factor Xa generation. The

combined action on factor Xa generation and inhibitionis important enough to cause prothrombinase inhibition.The main action of heparin therefore is on thrombin, itsanti-factor Xa action is hardly important and its anti-factor IXa action is secondary to inhibition of factor VIIIactivation, i.e. to an antithrombin effect.Low molecular weight (LMW) heparins come in two

kinds, those that are above the critical chain length of 18sugar units (ACLH), that catalyse both thrombin andfactor Xa inactivation, and the ultra low molecularweight heparins (ULM; 5-18 sugar units) that act onfactor Xa only. The specific activity of the ACLHsincreases not more than twofold with increasing chainlength. When tested in the presence ofphysiological CaI +concentrations the ratio of anti-factor Xa to antithrom-bin action varies between 0.3 and 0.5 for unfractionated(UF) heparin and various commercially availableLMWHs. In platelet-poor plasma (PPP), the mechanismof action of UF heparin and these LMW heparins isessentially similar. The LMW heparins differ con-siderably from UF heparin as to the content of materialwith a high affinity for antithrombin III (HAM), thelower the mean molecular weight (MMW), the lower theHAM content. The properties of a LMW heparin arelargely determined by its ACL-HAM content. Theadmixture ofULM hardly influences the mode of actionof a heparin in clotting PPP but it will cause higher valuesin specific anti-factor Xa tests, especially in the absence ofCaI + ions and in samples that contain platelet factor 4(pf4), because PF4 preferentially neutralizes HAM. Onlypractically pure ULMs, like pentasaccharide, will act viatheir anti-FXa properties.The important difference between LMW heparins and

UF heparin, apart from pharmacokinetic properties thatwill not be discussed here, is the capability of LMWheparins to maintain their action in the presence ofactivated platelets. This is a consequence of their contain-ing relatively large amounts ofnon-HAM that binds PF4,so that part of the HAM present escapes neutralization.

Pharmacokinetics of low molecular weight heparins

B. Boneu

After parenteral injection, heparin is removed from theblood via 2 mechanisms, saturable (reticulo-endothelialand endothelial cells) and non-saturable (renal excre-tion).' At the doses currently used for therapeutic pur-pose, unfractionated (UF) heparin is mainly removed viathe saturable mechanism. This accounts for the dose-dependent half-life of UF heparin, the concave-convexpattern of disappearance of the biological activity afterintravenous injection of a high dose, and the apparentpoor bioavailability when low doses of UF heparin aredelivered subcutaneously, while this bioavailability tendstowards 100% at higher doses.Due to a lower affinity for endothelial cells2 low

molecular weight (LMW) heparin is mainly removed viathe non-saturable (renal) mechanism. This propertyaccounts for the longer half-life of LMW heparin whichremains constant at any dose, the 100% bioavailabilityafter subcutaneous administration of low doses, and for

by copyright. on A

ugust 21, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.67.791.848 on 1 S

eptember 1991. D

ownloaded from

THROMBOSIS RESEARCH INSTITUTE WORKSHOP 851

the prolongation of the half-life in cases of chronic renalinsufficiency.Our current pharmacokinetic/pharmacodynamic know-

ledge of LMW heparin mainly derives from the disap-pearance of the anti-factor Xa activity after parenteralinjection. Other studies3 indicate that the clearance of theanti-factor Ila activity of a LMW heparin is twice as highthan that of the anti-factor Xa activity. The result is thatafter a subcutaneous injection, the ex vivo anti-factorXa/anti-factor Ila ratio of a LMW hepann is higher thanin vitro and continuously increases until the anti-factorIha activity becomes undetectable. Chronometric assayshave been proposed to determine anti-factor Xa activityin the plasma. These assays are also sensitive to anti-factor Ila activity of LMW heparin and therefore mayunderestimate the circulating anti-factor Xa activity aftersubcutaneous injection.

References

1. Boneu, B., Caranobe, C. & Sie, P. Pharmacokinetics ofheparinand low molecular weight heparin. In: Baillieres ClinicalHaematology, vol. 3, 1990. Bailliere-Tindall, pp. 531-544.

2. Barzu, T., Van Rijn, J., Petitou, M., Molho, P. & Tobelem, G.Endothelial binding sites for heparin. Specificity and role inheparin neutralization. Biochem J 1986, 238: 847-854.

3. Briant, L., Caranobe, C., Saivin, S. et al. Unfractionatedheparin and Cy216: Pharmacokinetics and bioavailability ofthe antifactor Xa and antifactor Ila effects after intravenousand subcutaneous injection. Thromb Haemostas 1989, 61:348-353.

Antithrombotic properties of low molecular weightheparin - laboratory assessment

T.W. Barrowcliffe, E. Gray, R.E. Merton andD.P. Thomas

The main difference between low molecular weight(LMW) heparins and unfractionated (UF) heparin in invitro assays is that their activity in anti-Ila or activatedpartial thromboplastin time (APTT) assays is muchlower than their anti-Xa activity. A major problem in invitro assays ofLMW heparin has been variability betweendifferent laboratories and different methods, so that apreparation assayed at 7,500 units in one laboratorycould be assessed as 3,000 units in another laboratory.This problem has now been minimized by the establish-ment of a separate International Standard for LMWheparin.The second, and most important, is, which of the in

vitro activities ofLMW heparins are most relevant to itsantithrombotic action? This is a complex issue which isstill a subject of controversy. The importance of anti-factor Xa activity was studied by several groups usingfragments below MW 5,000, which have anti-factor Xabut not anti-factor Ila activity. In animal models it wasfound that such fragments, including the synthetic penta-saccharide, were much less potent antithrombotic agentsthan UF heparin, despite their high anti-factor Xaactivity. A possible reason for this discrepancy which wasexplored in our laboratory was the method of measure-

ment of anti-Xa activity. The normal method uses freeenzyme, and our studies showed that, when inhibition offactor Xa was measured in the presence of Ca' + ions, orin the prothrombinase complex, the anti-factor Xaactivities of the fragments relative to UF heparin weremuch lower, and correlated better with antithromboticactivities.Other groups have emphasized the importance of

inhibition of thrombin, and in particular of the thrombinfeedback loops (activation of factors V and VIII), andseveral studies have demonstrated the antithromboticeffectiveness of hirudin and dermatan sulphate, whichhave anti-factor Ha but no anti-factor Xa activity. In ourstudies, 10 times less hirudin than UF heparin wasrequired for inhibition of thrombin activation of factorVIII, but in a rabbit stasis model, UF heparin and hirudinwere equally effective as antithrombotic agents, sugges-ting that the action of UF heparin is not limited toinhibition of this particular pathway. Several animalstudies have found a good correlation between anti-factorXa and antithrombotic activities of LMW heparinsproduced for clinical use, and this has been substantiatedin many clinical studies of prophylaxis and treatment ofDVT and in haemodialysis.The activities of LMW heparin in vivo may differ in

many respects from those measured in vitro, and lab-oratory assessment can only be at best an approximationof antithrombotic action. Nonetheless, the evidence todate suggests that the anti-Xa activities ofLMW heparinare the best marker for their antithrombotic activities invivo.

Low molecular weight heparin and cardiovasculardisease

V.V. Kakkar

Recent accumulating evidence strongly suggests thatanticoagulant therapy has several potential benefits inpatients suffering from cardiovascular disease. Theresults of randomized, controlled trials indicate thatanticoagulant therapy decreases morbidity, systemicembolism and prevents clinically important venousthromboembolism. It may also play an important part inpreventing re-occlusion following successful coronarythrombolytic therapy and angioplasty.

Recently, a new generation of antithrombotic agentshas been developed which may overcome many of thedisadvantages of commercially available heparin. Theseinclude low molecular weight (LMW) fragments orfractions ofheparin. There is now overwhelming evidencethat the anticoagulant action of heparin is primarily dueto its ability to bind tightly with the main inhibitor ofcoagulation, antithrombin III (ATIII). Such a complexwith ATIII transforms a slow, progressive inhibitor ofserine proteases into a potent, fast inhibitor. Variousfractions or fragments ofheparin have now been commer-cially prepared which are known to have an essentialATIII binding sequence. The LMW heparins also possessa number ofunique properties which may be particularlyuseful in treating patients suffering from cardiovasculardisease. These include high bioavailability, long duration

by copyright. on A

ugust 21, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.67.791.848 on 1 S

eptember 1991. D

ownloaded from

852 THROMBOSIS RESEARCH INSTITUTE WORKSHOP

of action after subcutaneous administration, reducedeffect of lipolysis and significant effect on whole bloodviscosity. The value of LMW heparin as an anti-thrombotic agent in cardiovascular disease patients willbe discussed in the paper.

Treatment ofthrombotic disease using lowmolecular weight heparin

U. Abildgaard

Heparin is the main drug for the treatment of acutethrombosis. Low molecular weight (LMW) heparinsdiffer from standard heparin regarding pharmacokin-etics, interaction with platelets, and anticoagulant profile.In terms of clinical effects this could mean that LMWheparin injected subcutaneously (s.c.) would need littlelaboratory control, and possibly cause less bleeding. Alimited number of trials have compared LMW heparin(CY216, ChoayPharmaceuticals, France; Fragmin0KabiPharmacia, Sweden; LogiparinO, Novo Nordisk, Den-mark) and standard heparin in the treatment of acutedeep venous thrombosis (DVT).Intravenous administration Albada et al.' found iden-tical antithrombotic efficacy, and identical mean plasmaanti-factor Xa effect in the two patient groups. Atendency towards less bleeding with Fragmin was notsignificant. The mean LMW heparin dose was 205anti-factor Xa U/kg/day, and the mean dose of standardheparin was 425 anti-factor Xa U/kg/day.Subcutaneous LMW heparin versus intravenous (i.v.)standard heparin In such comparisons, SC LMW hep-arins have given at least as good results as the conven-tional i.v. standard heparin. Dose adjustment was lessoften needed with subcutaneous Fragmin than withstandard heparin.2 Moreover, CY216 given in fixed dosesper kg bodyweight had the same efficacy as conventionali.v. standard heparin therapy adjusted according toAPTT values.3Subcutaneous LMW heparin versus i.v. standard hep-arin Similar efficacy and safety was observed in doubleblind trials comparing Fragmin4 and CY2165 with i.v.standard heparin. The need for dose adjustment wasagain reduced in the LMW heparin group.4Dosefinding trials with LMW heparin The comparativetrials suggested that LMW heparin in fixed doses (about200 anti-factor Xa U/kg/d) compare favourably with themore complicated regimen of standard heparin by i.v.infusion and dose adjustment. Subsequent studies haveaddressed the size ofthe dailyLMW heparin dose and thenumber ofdaily injections. As the various LMW heparinsare not identical, results from one type cannot beextrapolated to another. Moreover, the producers haveemployed different unit definitions. Given these limita-tions, it seems that with daily doses of 240 anti-factor XaU/kg/ of Fragmin, the antithrombotic efficacy might besuperior to that of conventional i.v. standard heparin.With 480 anti-factor Xa U/kg, the bleeding risk isincreased. Blood tests show that 8-hourly injectionsproduce remarkably stable anti-factor Xa levels. Butclinical studies comparing one and two daily injections ofrelatively low doses of LMW heparin (150-200 anti-factorXa U/kg) have failed to show clinical advantages of

dividing the daily dose, both for Fragmin6 and forlogiparin.

Experience with such simplified regimens is as yetlimited, particularly in ileofemoral deep venous throm-bosis and other severe thrombotic conditions. For themajority of patients with deep venous thrombosis, how-ever, the new results suggest that one daily injection ofLMW heparin is sufficient. Laboratory monitoring maybe reserved to patients who do not respond or who bleed.

References

1. Albada, J., Nieuwenhuis, H.K. & Sixma, J.J. Treatment ofacute venous thromboembolism, with low molecular weightheparin (Fragmin). Results of a double-blind randomizedstudy. Circulation 1989, 80: 935-940.

2. Aiach, M., Fiessinger, J.N., Vitoux, J.F. et al. Traitement desthromboses veineuses profondes constituees. Etude com-parative d'un fragment d'heparine de bas poids moleculaire(Fragmine) administree par voie sous-cutanee et de l'heparinestandard administree par voie intraveinuse continue. Etudemulticentrique. Rev Med Interne 1989, 10: 375-381.

3. Dutoux, P. Treatment of proximal deep vein thrombosis(DVT) of the lower limbs by CY216 (LMWH) versus unfrac-tionated heparin (UFH). Thromb Haemost 1987, 58: 437(Abstract).

4. Holm, H.A., Handeland, G.F., Abildgaard, U. et al. Sub-cutaneous heparin treatment of deep venous thrombosis: Acomparison of unfractionated and low molecular weightheparin. Haemostasis 1987, 16: 30-37.

5. Parker, C.J., Huber, D.E., Hedges, A.R. & Kakkar, V.V. Atrial of low molecular weight (LMW) heparin and unfrac-tionated (UF) heparin in the treatment of established deepvenous thrombosis (DVT). Thromb Haemost 1987, 58: 1390(Abstract).

6. Holmstr6m, M., T6rnebohm, E., Berglund, M.-C., Granqvist,S. & Lockner, D. Fragmin (Kabi) subcutaneously once ortwice daily in the treatment of deep venous thrombosis.International Congress on Internal Medicine 1990, AbstractNo. 123, Stockholm.

Prophylaxis with low molecular weight heparins intotal hip replacement

M. Samama

The risk of postoperative thrombosis, and speciallyproximal deep vein thrombosis, is significantly higher inhip surgery than in abdominal surgery in the absence ofprophylaxis. The North American consensus conferencehas recommended the use of adjusted doses of sub-.utaneous heparin which are more efficacious than stan-lard doses of heparin in the prevention of postoperativeLhromboembolism in these patients. More recently, anumber of low molecular weight (LMW) heparins havebeen evaluated in patients undergoing total hip replace-ment. These trials have compared a LMW heparin to aplacebo or to a standard dose of unfractionated heparin'with or without dihydroergotamine) or to a dose ofhieparin adjusted on activated partial thromboplastinime (APTT). After some dose finding trials, optimalloses have been defined: Clexane®& (Rhone-Poulenc,France) 40 mg once a day, FragminO (Kabi Pharmacia,Sweden) 5,000 anti-factor Xa units once a day or 2,500tnti-factor Xa units twice a day, FraxiparinO (Sanofi,

by copyright. on A

ugust 21, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.67.791.848 on 1 S

eptember 1991. D

ownloaded from

THROMBOSIS RESEARCH INSTITUTE WORKSHOP 853

France) 42 anti-factor Xa units per kg bodyweight fromDay 1 to Day 3 and 63 anti-factor Xa units per kgbodyweight from Day 4. Lomoparan@ (Org 10172Organon, Netherlands) has been used at a dose of 750anti-factor Xa units twice a day.The first injection of the drug was administered 12

hours or 2 hours before surgery, or postoperatively in thedifferent controlled studies. Bilateral phlebography wasperformed on the 10-12th postoperative day. In thissmall number of well conducted studies the rate of totaland proximal deep venous thrombosis was significantlyreduced in patients receiving Clexane or Lomoparan ascompared to placebo. Minor and major bleeding was notsignificantly increased. The different LMW heparins werefound to possess at least a similar efficacy and safety as thedifferent regimens ofunfractionated heparin used in thesestudies.These very encouraging results suggest that a standard

dose, or a dose adapted to bodyweight, ofLMW heparincould be as effective as unfractionated heparin given atdoses adjusted on APTT results which is considered as theprophylactic treatment of choice in hip surgery.

Low molecular weight heparins in haemodialysis

H. Ireland, D.A. Lane and J.R. Curtis

Haemodialysis and haemofiltration are used to clearuraemic toxins from the blood of patients with end stagerenal failure. Ideally, membranes and tubing used in theextracorporeal circuit would be sufficiently biocompati-ble to enable their use without the complication ofanticoagulants being added to the blood that passesthrough them. New membranes are continuously beingdeveloped but none so far is able to dispense withanticoagulant for routine dialysis of patients with chronicrenal failure.

Heparin has been the anticoagulant of choice inhaemodialysis. Most patients dialyse using heparin 2 or 3times a week for 3-6 hours on each occasion. A patientmay therefore be anticoagulated with heparin for up to900 hours/year. In view of this, it might be expected thatthere would be established guidelines to the use ofheparin. In practice, there is considerable variation in thetotal amount of heparin actually used, and in the mode inwhich it is administered intravenously. While most cen-tres anticoagulate by giving heparin as an initial bolusinjection, followed by a continuous infusion up untilapproximately 30 minutes before the end ofdialysis, thereis up to a 4-fold variation between different centres in thetotal amount of heparin administered during dialysis.This is probably due in part to the lack of a convenientobjective method of determining the in vivo effect ofheparin in dialysis.

Heparin is a heterogeneous glycosaminoglycan andcommercial preparations have a wide range in molecularweight (MW), 5000-30,000. Heparin expresses its anti-coagulant properties mainly by interacting with andaccelerating the action of the inhibitor, antithrombin III.Commercial (unfractionated) heparin accelerates theaction of antithrombin III against almost all coagulationenzymes, particularly thrombin and factor Xa. Theanticoagulant and non-anticoagulant properties of hepa-

rin are often MW dependent. As the MW of the heparinfalls below 5,400, its ability to accelerate thrombininhibition is lost, so that heterogeneous low molecularweight (LMW) heparin preparations that are nowavailable for routine use have higher anti-factor Xa thanantithrombin activities. LMW heparins also have re-duced ability to induce release of lipoprotein lipase;reduced interaction with blood neutralising proteins andhave reduced interaction with platelets; this last propertymay explain the reduced haemorrhagic effect of LMWheparins seen in experimental models. An importantproperty of low MW heparins is their altered phar-macokinetics: following subcutaneous injections they arebetter absorbed than heparin, and following intravenousinjection the half-life of elimination is prolonged at least2-fold. Because clearance ofLMW heparins mainly takesplace through the kidneys, the half-life may be evenfurther prolonged in patients with renal failure.As each of these properties may be of importance

during use of LMW heparins in haemodialysis andbecause most LMW heparins differ in terms of their meanMW and MW distribution, detailed dose finding studiesshould be performed with all LMW heparins before theycan be considered for routine use as anti-coagulants inthis clinical situation.A number of dose finding studies have now been

published.1-9 In some of these, attempts have been madeto use objective means of comparing the in vivo anti-coagulant effect (clot inhibitory) effects ofLMW heparinswith those of heparin: the most effective method has beenthe use of the FPA assay to detect the inhibition of theearly conversion of fibrinogen in the circulating blood.'0use of an objective test such as this might be consideredprudent in view of the variation in routine use of heparin,noted above. The results ofthese dose finding studies havesuggested three potential advantages ofLMW heparin inroutine haemodialysis.(1) LMW heparins may be given as a single initial bolus

injection rather than bolus injection followed byinfusion, provided dialysis is not greatly prolonged.

(2) LMW heparin has reduced lipolytic effects duringdialysis."

(3) LMW heparin may have less effect upon haemo-stasis.

References

1. Lane, D.A., Ireland, H., Flynn, A.M., Anastassiades, E. &Curtis, J.R. Haemodialysis with LMW heparin: dosagerequirements for the elimination of extra-corporeal fibrinformation. Nephrol Dial Transplant 1986, 1: 179-187.

2. Ireland, H, Lane, D.A., Flynn, A., Anastassiades, E. &Curtis, J.R. The anticoagulant effect ofheparinoid Org 10172during haemodialysis: an objective assessment. ThrombHaemost 1986, 55: 271-275.

3. Ireland, H., Lane, D.A., Flynn, A., Pegrum, A.C. & Curtis,J.R. Low molecular weight heparin in haemodialysis forchronic renal failure: a dose finding study ofCY222. ThrombHaemost 1988, 5: 240-247.

4. Ryan, K., Lane, D.A., Flynn, A., Sheppard, J., Ireland, H.A.& Curtis, J.R. Dose finding study of a low molecular weightheparin, Innohep, in haemodialysis. Thromb Haemost, inpress.

5. Schrader, J., Stibbe, W., Armstrong, V.W. et al. Comparisonof low molecular weight heparin to standard heparin inhaemodialysis/haemofiltration. Kidney Int 1988, 33:890-896.

by copyright. on A

ugust 21, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.67.791.848 on 1 S

eptember 1991. D

ownloaded from

854 THROMBOSIS RESEARCH INSTITUTE WORKSHOP

6. Ljunberg, B., Blomback, M., Johnson, H. & Lins, L.-E. Asingle dose of a low molecular weight heparin fragment foranticoagulation during dialysis. Clin Nephrol 1987, 27:31-35.

7. Anastassiades, E., Lane, D.A., Ireland, H., Flynn, A. &Curtis, J.R. A low molecular weight heparin ('Fragmin') forroutine haemodialysis: a crossover trial comparing three doseregimes with a standard regimen of commercial unfrac-tionated heparin. Clin Nephrol 1989, 32: 290-296.

8. Anastassiades, E., Ireland, H., Flynn, A., Lane, D.A. &Curtis, J.R. A low molecular weight heparin (Kabi 2165,'Fragmin') in repeated use for haemodialysis: prevention ofclotting and prolongation of the venous compression time incomparison with commercial unfractionated heparin. Neph-rol Dial Transplant 1990, 5: 135-140.

9. Suzuki, T., Ota, K., Naganuma, S. et al. Clinical applicationof Fragmin (FR-860) in haemodialysis: multicentre coopera-tive study in Japan. Semin Thromb Hemost 1990, 16 (Suppl):46-51.

10. Ireland, H., Lane, D.A. & Curtis, J.R. Objective assessmentof heparin requirements for haemodialysis in humans. J LabClin Med 1984, 103: 643-652.

11. Schrader, J., Andersson, L.A., Armstrong, V.W., Kundt, M.,Stibbe, W. & Scheler, F. Lipolytic effects of heparin and lowmolecular weight heparin and their importance in haemo-dialysis. Semin Thromb Hemost 1990, 16 (Suppl): 4147.

Treatment of acute venous thrombosis with sub-cutaneous low molecular weight heparin

J. Harenberg, K. Huck, G. Stehle, K. Mall, M.Blauth and D.L. Heene

Heparin is the drug of choice for prophylaxis andtreatment of thrombo-embolism. Low doses are emp-loyed for prophylaxis and high doses for treatment ofthromboembolism. High-dose regimens especially areassociated with an appreciable risk ofhaemorrhage.I Thishas initiated the development of drugs with an improvedbenefit-to-risk ratio. Heparin is a mixture of polydispersemucopolysaccharides with molecular weights rangingfrom 3,000 to 30,000 daltons.2 These mucopolysac-charides bind to antithrombin III via a specific pentasac-charide subunit of the molecules.3 Fractionation ofheparin according to molecular size demonstrated that,with decreasing the molecular weight, factor Xa inhibi-tion remained unchanged whereas thrombin inhibitiondecreased, resulting in an increased anti-Xa/antithrom-bin ratio.4 The low molecular weight (LMW) heparinfractions have been shown to possess an improvedpharmacological profile as compared with unfractionatedheparin (UF).5'6

It remains open whether LMW heparins also presentextension of acute venous thrombosis and to what extentthey induce thrombolysis. Therefore, we conducted aclinical study to assess safety and efficacy of a LMWheparin preparation as compared with UF heparin in thetreatment of patients with acute deep vein thrombosis.

Patients of both sexes between 20 and 80 years of age,with acute deep vein thrombosis were randomized for theclinical trial. The study was designed prospectively,randomized and controlled using UF heparin. Allpatients had to present with symptoms of acute deep veinthrombosis ofthe lower or upper extremity which was lessthan 4 weeks. LMW heparin (n = 24) was administered

twice daily subcutaneously at a dose of 2 x 150 anti-factor Xa units/kg body weight, and UF heparin (n = 26)was given intravenously by continuous infusion at a doseof 450 anti-factor Xa units/kg body weight daily for 10days. Doses were adjusted to maintain peak anti-factorXa levels between 0.5 and 1.0 anti-factor Xa units per ml.One patient in the UF heparin group and 2 patients in theLMW heparin group suffered from bleeding complica-tions. Two patients on LMW heparin and 12 patients onUF heparin had high evidence of pulmonary embolismbased on defects on ventilation-perfusion scintigraphy.Control phlebography and Duplex sonography demon-strated a significant improvement during both treatmentregimens. Using Duplex sonography, complete andincomplete dissolution of thrombi of vena poplitea, venafemoralis, or vena iliaca were seen in 9 of 13 patients onUF and 4 of9 patients on LMW heparin. No change wasobserved in 4 of 13 versus 5 of 9 patients.

Ascending phlebography at the end of the treatmentperiod was performed in 13 and 15 patients on UF andLMW heparin respectively. A regression of thrombi wasseen in 10 of 13 and 13 of 15 patients in both treatmentgroups. Progression was verified in 0 and I patientsrespectively. No change was seen in 3 and 1 patients inboth treatment groups. The Marder score improved in thegroup treated with UF heparin (from 25.8 to 18.1,P<0.01) and in the LMW heparin group (from 22.0 to11.0, P< 0.001). Reperfusion ofthe deep vein system was70% with LMW heparin and 75% with UF. The anti-factor Xa levels were significantly higher in the LMWheparin group, and activated partial thromboplastin andthrombin clotting times were significantly higher in thegroup receiving UF heparin. Thrombin/antithrombin IIIcomplexes and D-dimer concentration decreased duringthe treatment, but did not differ between the tworegimens. At the end of the treatment period with LMWheparin, protein C and antithrombin III weresignificantly higher.The study demonstrated that administration ofLMW

heparin twice daily by the subcutaneous route is aseffective as continuous intravenous infusion of UFheparin. Special attention has to be paid to the dose ofLMW heparin which should not exceed 15,000 anti-factorXa units per day. The subcutaneous treatment offersadvantages for patients and hospital staff. It facilitates themobilization of the patient, minimizes the risk of localand systemic infections, and makes outpatient treatmentpossible.7

References

I. Kelton, J.G. & Hirsh, J. Bleeding associated with antithrom-botic therapy. Semin Hematol 1980, 17: 259-291.

2. Harenberg, J. & De Vries, J.X. Characterization ofheparins byhigh performance size exclusion liquid chromatography. JChromatogr 1983, 261: 287-292.

3. Choay, J., Petitou, M., Lormeau, J.C., Sinay, P., Casu, B. &Gatti, G. Structure-activity relationship in heparin. A syn-thetic pentasaccharide with high affinity for antithrombin III.Biochem Biophys Res Commun 1983, 116: 492-499.

4. Barrowcliffe, T.W., Johnson, E.A., Eggleton, C.A. & Thomas,D.P. Anticoagulant activities of lung and mucous heparins.Thromb Res 1977, 12: 27-36.

by copyright. on A

ugust 21, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.67.791.848 on 1 S

eptember 1991. D

ownloaded from

THROMBOSIS RESEARCH INSTITUTE WORKSHOP 855

5. Bratt, G., T6rnebohm, E., Widlund, L. & Lockner, D. Lowmolecular weight heparin (Kabi 2165, Fragmin). Phar-macokinetics after intravenous and subcutaneous administra-tion in human volunteers. Thromb Res 1986, 42: 613-620.

6. Harenberg, J. & Heene, D.L. Pharmacology and specialclinical applications of low molecular weight heparins. Am JHematol 1988, 29: 233-240.

7. Harenberg, J., Huck, K., Bratsch, H. et al. Therapeuticapplication of subcutaneous low molecular weight heparin inacute venous thrombosis. Haemostasis 1990. In press.

by copyright. on A

ugust 21, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.67.791.848 on 1 S

eptember 1991. D

ownloaded from