Original Paper

Haemostasis 2000;30:196–203

Antithrombotic Activity of OrallyAdministered Low Molecular WeightHeparin (Logiparin) in a Rat Model

Linda M. Hiebert Tilly Ping Sandra M. Wice

Department of Veterinary Physiological Sciences, University of Saskatchewan, Saskatoon,Canada

Received: June 9, 2000Accepted in revised form: July 27, 2000

Dr. Linda M. HiebertDepartment of Veterinary Physiological Sciences, 52 Campus DriveUniversity of Saskatchewan, Saskatoon, S7N 5B4 (Canada)Tel. +1 306 966 7358, Fax +1 306 966 7376E-Mail linda.hiebert@usask.ca

ABCFax + 41 61 306 12 34E-Mail karger@karger.chwww.karger.com

© 1998 S. Karger AG, Basel0301–0147/00/0304–0196$17.50/0

Accessible online at:www.karger.com/journals/hae

Key WordsHeparin W Logiparin, oral W Thrombosis W

Activated partial thromboplastin time

AbstractPrevious studies in rats demonstrated thatorally administered, unfractionated bovinelung heparin is absorbed and has a dose-dependent antithrombotic effect. The objec-tive of this study was to determine if an orallow molecular weight heparin had a similarantithrombotic effect in the same model.Thrombosis was induced in rats by applica-tion of 10% formalin in 65% methanol to theexposed jugular vein. Immediately follow-ing, saline, unfractionated heparin (3.3–60 mg/kg) or the low molecular weight hepa-rin, Logiparin (0.025–15 mg/kg; 20–30 ratsper group) was placed in the stomach and 4 hlater the jugular vein was inspected for athrombus. Compared to saline, oral Logipar-in reduced the incidence of thrombosis at all

doses with a dose-dependent effect suggest-ed. A significant increase was observed inthe activated partial thromboplastin time andin plasma heparin concentrations, deter-mined by AccuclotTM Heptest® and anti-fac-tor Xa chromogenic assay for rats given oralLogiparin versus saline. A dose-dependentincrease in plasma heparin concentrationwas observed when estimated by the anti-Xachromogenic assay. Heparin was recoveredin 9% of aortic endothelial samples when60.8 mg/kg Logiparin was administered. A50% reduction in thrombosis was observedat 0.1 mg/kg for oral Logiparin versus 7.5 mg/kg for unfractionated bovine lung heparinindicating that oral Logiparin is an effectiveantithrombotic agent at doses lower than un-fractionated heparin. Orally administeredlow molecular weight heparin may be usefulfor the prevention and treatment of thrombo-sis.

Copyright © 2001 S. Karger AG, Basel

Dow

nloa

ded

by:

Uni

v. o

f Mic

higa

n, T

aubm

an M

ed.L

ib.

141.

213.

236.

110

- 8/

8/20

13 9

:50:

06 A

M

Antithrombotic Activity of Oral LowMolecular Weight Heparin

Haemostasis 2000;30:196–203 197

Introduction

Heparin has traditionally been adminis-tered parenterally and is believed to be inef-fective when given by the oral route [1]. Ourrecent studies challenged this assumption andshowed that unfractionated bovine lung hepa-rin or dextran sulphates Mr 8,000, are recov-ered from plasma and endothelium followingoral administration. At 6 min, 45% of heparinplaced in the stomach was found with endo-thelium while 0.18% was in plasma thus indi-cating that heparin was rapidly absorbed withmost of the drug being sequestered by theendothelium of the vasculature [2, 3]. Thesame heparin or dextran sulphate Mr 8,000,administered orally, prevented thrombosis ina rat model with a dose-dependent effect ob-served when unfractionated heparin was ad-ministered at doses of 3.25–60 mg/kg [2, 4].

Derivatives of heparin have been devel-oped with low average molecular weights andlower degrees of polydispersity [5]. Thesecompounds have been recommended for clin-ical use in the prevention and treatment ofmany conditions which formally used unfrac-tionated heparin [1]. In this study we wishedto determine if a low molecular weight(LMW) heparin, Logiparin, administeredorally, would prevent thrombosis in a ratmodel as has been observed for oral unfrac-tionated heparin [2, 4]. In addition to examin-ing the incidence of thrombosis with the doseof Logiparin administered, we determinedchanges in coagulation by the activated par-tial thromboplastin times (APTT), plasmaheparin concentration by measuring anti-fac-tor Xa activity using the Heptest and a chro-mogenic assay, and the chemical amount ofheparin with endothelium 4 h after drug ad-ministration. Thrombosis incidence withLogiparin treatment versus unfractionatedbovine lung heparin, in this rat model, is alsocompared.

Materials and Methods

HeparinLMW heparin, Logiparin (anti-Xa activity

90.7 IU/mg, peak maximum molecular mass 5,600daltons) obtained from porcine mucosal heparin, wasgenerously donated by Novo Nordisk, Denmark. Logi-parin was dissolved in water at concentrations of 100,50, 25 and 2 mg/ml. Unfractionated bovine lung hepa-rin, 150 units/mg, lot No. ZX320, was obtained fromUpjohn, Kalamazoo, Mich. and dissolved in water at aconcentration of 100 mg/ml.

AnimalsThree hundred twenty-one male Wistar rats,

weighing 310 B 39 g (B SD), were handled andhoused according to the Principles of Animal Care setout by the Canadian Federation of Biological Societies.The animals were fasted overnight prior to treatmentand were anesthetized with barbital and methoxyflu-rane for experimental procedures.

Thrombosis TestThe thrombosis test was performed as described by

Blake et al. [6]. Several drops of 10% formalin in 65%methanol were applied to the exposed right jugularvein. The incision was closed and immediately, saline(30 rats), unfractionated heparin at 60, 30, 15, 7.5 or3.3 mg/kg, or LMW heparin at 15, 7.5, 3.3, 1.7, 0.8, 0.1or 0.025 mg/kg (20–25 rats per dose) was introducedinto the stomach by stomach tube. Saline (0.2 ml) wasadded to the measured quantity of stock solution (vol-ume 0.01–0.2 ml; average 0.05 ml) in the syringe togive a final administered volume of 0.2–0.4 ml. Saline(0.2 ml) was given to control rats. The heparin dose orsaline was assigned randomly with 5–6 rats treated perday. After 4 h, the rat was again anesthetized and theexposed jugular vein examined for the presence of avascular plug by pressing gently with a cotton pledget.The result was scored as negative if blood was seen toflow when the pledget was removed and positive ifflow was obstructed indicating the presence of a vascu-lar plug. Results were scored as positive/negative if theplug was unstable when pressure with the cotton pled-get was applied. Percent incidence of nonpatency wascalculated.

Collection of Blood and Blood VesselsImmediately after examination of the jugular vein,

a laparotomy was performed and a blood sample ofapproximately 10 ml (9 parts blood to 1 part 3.8%sodium citrate) was taken from the abdominal aorta.

Dow

nloa

ded

by:

Uni

v. o

f Mic

higa

n, T

aubm

an M

ed.L

ib.

141.

213.

236.

110

- 8/

8/20

13 9

:50:

06 A

M

198 Haemostasis 2000;30:196–203 Hiebert/Ping/Wice

Plasma was prepared. As a source of endothelium, thethoracic aorta was removed and placed in saline.

Harvesting of EndotheliumEndothelium was removed from blood vessels ac-

cording to the method of Hiebert and Jaques [7]. Thethoracic abdominal aorta was removed and slit open,pinned to dental wax lumen side up, and rinsed insaline. Cellulose acetate paper was applied to thelumenal surface and when lifted, endothelium wasremoved. The length and width of the imprint weremeasured to the nearest millimetre (mean area 2.6 B0.8 cm2 SD).

Determination of Plasma HeparinPlasma heparin was measured by the APTT (Sig-

ma, St. Louis, Mo.), AccuclotTM Heptest® (procedureNo. CRS114, Sigma) and anti-factor Xa chromogenicassay (AccucolorTM procedure No. CRS106, Sigma),when plasma was derived from blood samples ob-tained with a minimum of tissue trauma.

Determination of Heparin with EndotheliumCellulose acetate paper was removed from endo-

thelium by dissolving in cold acetone followed by cen-trifuging and discarding the supernatant. The processwas repeated to give a dry tissue powder [7]. Agarosegel electrophoresis was used to identify and measureheparin in endothelial extracts [8]. The dried powders,dissolved in suitable volumes of water, were applied toagarose gel slides along with the administered heparinused as a reference. Following electrophoresis, gelswere fixed in 0.1% hexadecyltrimethylammoniumbromide and air-dried. Slides were stained with 0.04%toluidine blue in 80% acetone and background colourwas removed with 1% acetic acid. Heparin was identi-fied by electrophoretic migration as compared to refer-ence material and amounts determined by densitome-try.

Data AnalysesConfidence intervals (95%) for the incidence of

thrombosis were calculated using confidence belts forproportions [9]. Values for the amount of heparin onplasma and endothelium are presented as means andstandard errors and as individual values. To expressheparin concentration in Ìg/cm3 the heparin foundwith endothelium in Ìg/cm2 is divided by 2 ! 10–5

(the average endothelial thickness). Data were ana-lyzed using all samples, with 0 assigned to samplesbelow the limit of detection (240 Ìg/cm3). Student’s ttest was used to evaluate the significance of differencesbetween means of heparin-treated versus control rats.

A non-parametric test, Spearman’s coefficient of rankcorrelation, with Student’s t test was used to determinethe effect of dose on plasma and endothelial heparinconcentrations (excluding controls).

Results

The effect of oral Logiparin on thrombosisis shown in figure 1 and includes the inci-dence of thrombosis and the proportion ofstable (hard clot) and unstable (soft clot)plugs. Confidence intervals (95%) are shownfor total antithrombotic events. The incidenceof thrombosis was 60% at 0.025 mg/kg, com-pared to 83.3% in rats given oral saline, and40% at 0.1 mg/kg. From 0.8 to 15 mg/kg theincidence of thrombosis varied from 5 to24%. Only stable plugs were seen in controlanimals. A greater incidence of hard throm-botic plugs was seen at doses less than 0.1 mg/kg versus doses equal or greater than 0.8 mg/kg oral Logiparin. These results suggest adose-dependent effect.

APTT and the plasma heparin concentra-tion as determined by AccuclotTM Heptest®,4 h after drug administration, were signifi-cantly increased when comparing plasmasamples from all heparin- versus saline-treated rats, although a dose-dependent effectwas not evident (fig. 2a, b). The plasma hepa-rin concentration as determined by a chromo-genic assay (Accucolor) was significantly in-creased at all doses of Logiparin administra-tion as compared to control values with adose-dependent effect observed (table 1) asdetermined by Spearman’s coefficient of rankcorrelation. Plasma heparin concentrationsdetermined by this method were always lessthan 1 Ìg/ml.

When endothelial samples were examined4 h after administration, heparin was recov-ered from 9% of samples when doses greateror equal to 0.8 mg/kg were administered as

Dow

nloa

ded

by:

Uni

v. o

f Mic

higa

n, T

aubm

an M

ed.L

ib.

141.

213.

236.

110

- 8/

8/20

13 9

:50:

06 A

M

Antithrombotic Activity of Oral LowMolecular Weight Heparin

Haemostasis 2000;30:196–203 199

Fig. 1. Prevention of experimentalthrombosis by oral administrationof LMW heparin (Logiparin). Sa-line or heparin was administeredby stomach tube immediately afterinitiation of a thrombus by applica-tion of 10% formalin in 65% meth-anol to the jugular vein. The veinwas examined for the presence of athrombus 4 h later [30 rats, 0 mg/kg (saline); 20–25 rats per group forall other doses]. Error bars repre-sent 95% confidence intervals.‘Hard clot’ shows % incidence ofstable plugs, ‘soft clot’ unstableplugs.

Fig. 2. Changes in plasma heparinas determined by Heptest (a) andAPTT (b) after oral administrationof LMW heparin (Logiparin). Plas-ma samples were taken 4 h afterdrug administration by stomachtube. a Means B SE are shown.b Geometric means are shown. Asignificant increase was observedin plasma heparin concentrationsand APTT from heparin-treatedrats versus saline-treated rats: p =0.006 and p !! 0.001, respectively(Student’s one-tailed t test). A dose-dependent effect was not evident ineither plasma heparin concentra-tions or APTT versus log dose us-ing Spearman’s coefficient of rankcorrelation: p 1 0.1 and p 1 0.5,respectively.

Dow

nloa

ded

by:

Uni

v. o

f Mic

higa

n, T

aubm

an M

ed.L

ib.

141.

213.

236.

110

- 8/

8/20

13 9

:50:

06 A

M

0.00

200 Haemostasis 2000;30:196–203 Hiebert/Ping/Wice

Fig. 3. Prevention of thrombosiswith oral Logiparin versus oral un-fractionated bovine lung heparin.Saline or heparin was administeredby stomach tube immediately afterinitiation of a thrombus by applica-tion of 10% formalin in 65% meth-anol to the jugular vein. The veinwas examined for the presence of athrombus 4 h later. Error bars re-present 95% confidence intervals.A 50% reduction in thrombosis in-cidence was achieved at 0.1 mg/kgfor oral Logiparin versus 7.5 mg/kgfor oral unfractionated heparin.

Table 1. Logiparin with plasma and aortic endothelium 4 h after oral administration

Dosemg/kg

Plasma1

samplespositive

concentrationÌg/ml

Endothelium

samplespositive

concentrationÌg/cm3

0/5 0 0/16 00.025 1/9 0.04B0.04 0/20 0

0.350.1 2/12 0.03B0.02 0/20 0

0.26, 0.060.8 1/6 0.03B0.04 3/25 711B344

0.19 893, 2,381, 5,4051.7 3/12 0.03B0.02 1/25 444B453

0.06, 0.04, 0.21 11,0003.25 4/10 0.04B0.02 2/25 250B177

0.06, 0.07, 0.08, 0.15 3,125, 3,2607.5 3/6 0.02B0.01 2/25 216B157

0.04, 0.01, 0.05 2,083, 3,33015 5/9 0.05B0.02 2/18 1,065B869

0.10, 0.07, 0.1, 0.04, 0.1 14,814, 4,348

Italicized numbers show means and standard errors; those in standard type show individu-al positive values.1 Plasma heparin concentrations were determined by anti-Xa chromogenic assay Accu-color.

Dow

nloa

ded

by:

Uni

v. o

f Mic

higa

n, T

aubm

an M

ed.L

ib.

141.

213.

236.

110

- 8/

8/20

13 9

:50:

06 A

M

Antithrombotic Activity of Oral LowMolecular Weight Heparin

Haemostasis 2000;30:196–203 201

shown in table 1. Heparin was not observedon control endothelium or when 0.025 and0.1 mg/kg were administered. Endothelialheparin concentrations were dose-dependentwhen analyzed by Spearman’s coefficient ofrank correlation. Average heparin concentra-tions in endothelium (Ìg/cm3) versus plasmaconcentrations (Ìg/ml) were on the order of1,000 times greater when doses equal or great-er than 0.8 mg/kg were considered.

The percent incidence of thrombosis, with95% confidence intervals, when varying dosesof Logiparin are given by stomach tube, wascompared to the incidence of thrombosis fol-lowing administration of unfractionated bo-vine lung heparin by the same route (fig. 3).For unfractionated bovine lung heparin,3.3 mg/kg resulted in a percent thrombosisincidence equal to that when rats were givensaline (80 vs. 83.3%); for Logiparin, 0.025 mg/kg approached the percent thrombosis inci-dence in control rats but was still lower (60 vs.83.3%). A much lower dose was required toreduce the incidence by 50% for Logiparin(0.1 mg/kg) versus unfractionated bovine lungheparin (7.5 mg/kg). At 15 mg/kg the percentincidence of thrombosis was equal for bothoral unfractionated heparin and Logiparin.

Discussion

Our results indicate that like unfraction-ated bovine lung heparin, the LMW heparin,Logiparin, administered orally, decreases theincidence of thrombosis in a rat model in adose-dependent manner suggesting that oralLogiparin is absorbed. APTT and plasmaheparin concentrations were elevated in Logi-parin-treated rats compared to rats given sa-line also supporting the hypothesis that orallyadministered Logiparin is absorbed. A dose-dependent effect was seen when plasma hepa-rin was measured by the chromogenic assay

with the incidence of recovery increasing withdose (table 1). Low plasma levels followingoral administration are in agreement withprevious studies where less than 1% of theadministered dose was observed followingoral unfractionated heparin in rats [3]. Plas-ma heparin concentrations are on the order of1,000 times lower than endothelial concentra-tions, which is similar to the plasma/endothe-lial ratios observed when using unfraction-ated heparin [2, 3].

Heparin was found on aortic endotheliumof a small number of rats when oral Logiparinwas administered. It was not detected on en-dothelium of saline-injected rats or at doses of0.1 and 0.025 mg/kg Logiparin. The inabilityto detect endothelial heparin at these lowdoses and low incidences of recovery fromendothelium is not surprising. At 0.1 mg/kg, if100% of the administered heparin was foundwith endothelium and was distributed equallythroughout all the vascular beds, only 450 Ìg/cm3 would be recovered from an endothelialsample equal to the mean area examined inthis study (2.6 cm2). The limit of detection forour method is 240 Ìg/cm3. Previous studieswith oral heparin have demonstrated thatpeak endothelial concentrations, 45% of theadministered dose, are found within 15 minof administration [2, 3]. In this study we aresampling endothelium at 4 h after administra-tion where endothelial concentrations havepreviously been observed to be much lowerand therefore would be below the limit ofdetection. Despite the low incidence of detec-tion a dose-dependent increase in endothelialheparin concentrations is observed when ana-lyzed by non-parametric statistics suggestingthat heparin is absorbed and heparin endothe-lial concentrations are important for this pro-tective effect.

When the antithrombotic activity of Logi-parin versus unfractionated bovine lung hepa-rin was compared, a 50% reduction in throm-

Dow

nloa

ded

by:

Uni

v. o

f Mic

higa

n, T

aubm

an M

ed.L

ib.

141.

213.

236.

110

- 8/

8/20

13 9

:50:

06 A

M

202 Haemostasis 2000;30:196–203 Hiebert/Ping/Wice

bosis relative to saline-treated rats was ob-served at 0.1 mg/kg Logiparin compared to7.5 mg/kg for unfractionated bovine lung hep-arin (see fig. 3). In this model therefore, oralLogiparin is greater than 70 times more effec-tive than oral unfractionated bovine lung hep-arin when compared on a weight basis. Theincreased effectiveness of Logiparin in pre-venting thrombosis may be related to an in-crease in absorption but could also be a func-tion of plasma or endothelial distribution orits effect on coagulation since the molecularweight distribution of heparin chains in logi-parin and unfractionated heparin is incom-parable. The observation that a lower molecu-lar weight heparin may be more easily ab-sorbed is in agreement with observations bySue et al. [10] who showed that lower molecu-lar weight fractions of heparin produced byalcoholic fractionation increased the wholeblood clotting time when compared to unfrac-tionated heparin when 40 mg/kg was intro-duced into mouse duodenum and by Lasker[11] who showed increased anti-Xa activity infasted rats given LMW heparin versus unfrac-tionated heparin. Compared to our studiesusing oral unfractionated heparin we were notable to completely eliminate the thromboticevent using oral Logiparin. A 16% incidenceof thrombosis was observed when 15 mg/kgoral Logiparin was administered, a dose 150times the ED50 (0.1 mg/kg). When using oralunfractionated heparin, we were able to elimi-nate thrombus formation at 60 mg/kg, a doseonly 8 times the ED50 (7.5 mg/kg). It is possi-ble that at higher doses, unfractionated hepa-rin in contrast to Logiparin, is exerting itsanticoagulant activity through heparin cofac-tor II as well as antithrombin III resulting in amore complete antithrombotic effect.

Our observations demonstrate that thequality of the thrombus differs when Logipar-in is administered versus unfractionated hep-arin with most thrombi consisting of a stable

plug when unfractionated heparin is adminis-tered (data not shown) while thrombotic plugswith Logiparin are unstable. At 15 mg/kg theincidence of thrombosis was identical whenLogiparin and unfractionated heparin arecompared; however, unstable plugs were seenwith Logiparin (fig. 3) while stable plugs wereobserved with unfractionated heparin. Thereason for these differences remain to be ex-plored.

These data agree with our previous obser-vations that heparin is absorbed when admin-istered orally and has an antithrombotic ef-fect. These data suggest that LMW heparinmay be a more effective oral drug than unfrac-tionated heparin. Since LMW heparins areprepared by a variety of techniques, it isunwise to assume that all LMW heparins arebioavailable and are effective antithromboticagents when administered orally; therefore,further studies on the effects of LMW hepa-rins when administered orally are required.

Acknowledgment

This study was supported by a grant from the Heartand Stroke Foundation of Saskatchewan.

Dow

nloa

ded

by:

Uni

v. o

f Mic

higa

n, T

aubm

an M

ed.L

ib.

141.

213.

236.

110

- 8/

8/20

13 9

:50:

06 A

M

Antithrombotic Activity of Oral LowMolecular Weight Heparin

Haemostasis 2000;30:196–203 203

References

1 Turkstra F, Koopman MMW, Bül-ler HR: The treatment of deep veinthrombosis and pulmonary embo-lism. Thromb Haemost 1997;78:489–496.

2 Jaques LB, Hiebert LM, Wice SM:Evidence from endothelium of gas-tric absorption of heparin and ofdextran sulphates 8000. J Lab ClinMed 1991;117:122–130.

3 Hiebert LM, Wice SM, McDuffieNM, Jaques LB: The heparin targetorgan – The endothelium. Studies ina rat model. Q J Med 1993;86:341–348.

4 Hiebert LM, Wice SM, Jaques LB:Antithrombotic activity of oral un-fractionated heparin. J CardiovascPharmacol 1996;28:26–29.

5 Fareed J, Hoppensteadt DA, Walen-ga JM: Current perspectives on lowmolecular weight heparins. SeminThromb Hemost 1993;19(suppl1):1–11.

6 Blake OR, Ashwin JG, Jaques LB:An assay for the antithrombotic ac-tivity of anticoagulants. J Clin Pa-thol 1959;12:118–122.

7 Hiebert LM, Jaques LB: The obser-vation of heparin on endotheliumafter injection. Thromb Res 1976;8:195–204.

8 Jaques LB, Wice SM, Hiebert LM:Determination of absolute amountsof heparin and of dextran sulfate inplasma in microgram quantities. JLab Clin Med 1990;115:422–432.

9 Steel RGD, Torrie JH: Principlesand Procedures of Statistics. NewYork, McGraw-Hill Book, 1980, pp599–603.

10 Sue TK, Jaques LB, Yuen E: Effectsof acidity, cations and alcoholicfractionation on absorption of hepa-rin from gastrointestinal tract. Can JPhysiol Pharmacol 1976;54:613–617.

11 Lasker SE: Low molecular weightheparin-like preparations with oralactivity. Semin Thromb Hemost1985;11:37–39.

Dow

nloa

ded

by:

Uni

v. o

f Mic

higa

n, T

aubm

an M

ed.L

ib.

141.

213.

236.

110

- 8/

8/20

13 9

:50:

06 A

M