SEARCH

SEARCH BY CITATION

Keywords:

  • Antithrombin;
  • coagulopathy;
  • xenotransplantation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Delayed rejection of pig kidney xenografts by primates is associated with vascular injury that may be accompanied by a form of consumptive coagulopathy in recipients. Using a life-supporting pig-to-baboon renal xenotransplantation model, we have tested the hypothesis that treatment with recombinant human antithrombin III would prevent or at least delay the onset of rejection and coagulopathy. Non-immunosuppressed baboons were transplanted with transgenic pig kidneys expressing the human complement regulators CD55 and CD59. Recipients were treated with an intravenous infusion of antithrombin III eight hourly (250 units per kg body weight), with or without low molecular weight heparin. Antithrombin-treated recipients had preservation of normal renal function for 4–5 days, which was twice as long as untreated animals, and developed neither thrombocytopenia nor significant coagulopathy during this period. Thus, recombinant antithrombin III may be a useful therapeutic agent to ameliorate both early graft damage and the development of systemic coagulation disorders in pig-to-human xenotransplantation.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Transplantation of organs and cells from pigs holds enormous promise as a means of solving the critical shortage of donated human tissues, but there are many immunological barriers that stand in the way of successful clinical application (1). Pig organs transplanted into unmodified humans or nonhuman primates usually undergo hyperacute rejection within minutes to hours. In this process, pre-existing anti-pig antibodies bind to the vasculature of the xenograft and activate the complement system, leading to hemorrhage, edema and intravascular thrombosis. Some progress has been made towards prolonging xenograft survival (2). For example, organs from pigs genetically modified to express human complement regulatory proteins are capable of controlling activation of primate complement and are thus protected from hyperacute rejection in primates (3–5). However, all pig-to-primate vascularized grafts are ultimately rejected within days to months, depending on the treatment protocol, by a process termed acute vascular rejection or delayed xenograft rejection (2). The mechanisms responsible for delayed rejection are not clear, although it appears that activation of vascular endothelial cells with accompanying prothrombotic and inflammatory changes is a key event.

Under physiological conditions, the quiescent vascular endothelium maintains an anticoagulant surface by expressing anticoagulant and platelet anti-aggregatory proteins (6). Several of the anticoagulants act by limiting the generation and procoagulant activity of thrombin, the main effector protease of the coagulation cascade. These molecules include antithrombin III (ATIII) and tissue factor pathway inhibitor (TFPI), which retain proximity to the endothelial cell surface by associating with heparan sulfate, and thrombomodulin (TM), which is membrane bound. In the xenograft setting, several events converge to cause disordered coagulation (Figure 1). Extravascular tissue factor (TF) expressed on the subendothelial matrix is exposed to circulating clotting factors, generating thrombin, when vascular endothelial cells are destroyed, injured or activated by binding of antidonor antibodies and/or complement. Additional intravascular TF is expressed by monocytes adhering to activated platelets and endothelial cells at the site of injury, as well as by the activated endothelial cells themselves (7, 8). Endothelial cell activation also results in the loss of heparan sulfate and its associated molecules ATIII and TFPI, the disappearance of TM from the cell surface (9), and the expression of inflammatory mediators such as platelet-activating factor (10). Cross-species molecular incompatibilities between activated coagulation components and their inhibitors may further tip the balance towards activation of coagulation (9). Porcine TM fails to efficiently bind human thrombin and hence fails to catalyze the generation of the activated human protein C, which is a potent anticoagulant (11). Porcine TFPI does not efficiently neutralize human factor Xa (12). Finally, in addition to initiating clotting, thrombin may stabilize clots and can trigger further endothelial cell activation and platelet aggregation and activation (8, 9).

image

Figure 1. Proposed events leading to vascular rejection of a pig-to-primate xenograft. Endothelial cells (EC) lining graft blood vessels are activated by the binding of xenoreactive natural antibodies (XNA) and/or activated complement. Upon activation the cells retract from each other, causing loss of barrier function and exposure of coagulation-triggering molecules including tissue factor (TF). Activation also causes loss of cell surface heparan sulfate (HS) with its associated anticoagulants antithrombin III (ATIII) and tissue factor pathway inhibitor (TFPI), down-regulation of thrombomodulin and the platelet anti-aggregatory protein vascular NTPDase (not shown), up-regulation of TF and cell adhesion molecules (CAM), and secretion of cytokines including platelet activating factor (PAF). The net result of these changes is platelet (Pt) aggregation and microvascular thrombosis, which may be worsened by molecular incompatibilities between activated primate coagulation components and their corresponding porcine inhibitors.

Download figure to PowerPoint

The consequences of intravascular coagulation are not limited to loss of the xenograft. Many primate recipients of porcine vascularized organs or cellular grafts exhibit a profound consumption of platelets (thrombocytopenia) and clotting factors that can evolve into a life-threatening condition in keeping with disseminated intravascular coagulation (DIC) (9).

Thrombin is clearly an important effector in xenograft rejection and is thus an attractive target for therapeutic intervention. ATIII is the physiological regulator of thrombin and other serine proteases generated during coagulation (13). ATIII neutralizes thrombin by binding it in an equimolar, irreversible complex (14), and its anticoagulant activity is potentiated by unfractionated heparin and to a lesser extent by low molecular weight heparin (LMWH) (15). Results from our initial pig-to-baboon renal xenotransplantation study (3) are consistent with the notion that depletion of ATIII contributes to xenograft damage and coagulopathy. In this model, transgenic pig kidneys expressing human CD55 and CD59 were transplanted into untreated baboons from which both native kidneys had been removed. Hyperacute rejection was prevented but recipients rapidly developed symptoms of DIC, including thrombocytopenia, delayed clotting times, and accumulation of fibrin degradation products. Treatment with LMWH was moderately successful in maintaining platelet number but had no effect on the other indicators of DIC, and did not prevent the progressive deterioration of graft function after 2 days. Analysis of a limited number of stored plasma samples demonstrated a decline in ATIII activity coinciding with graft rejection and the development of coagulopathy (data not shown), perhaps explaining the relative lack of efficacy of LMWH.

ATIII concentrate has been shown to prevent DIC in a porcine sepsis model (16), and recombinant human ATIII (rhATIII) attenuated both the coagulation and inflammatory responses in a baboon sepsis model (17). We therefore decided to test whether treatment with high doses of rhATIII would prevent coagulopathy and protect renal xenografts from early injury in the pig-to-baboon model.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

RhATIII

rhATIII produced in the milk of transgenic goats was provided in lyophilized form by Genzyme Transgenics Corporation (Framingham, MA, USA). It was reconstituted before use in sterile water for injection, and the resulting solution was 58 mg/mL (406 units/mL) rhATIII in 10 mm sodium citrate, glycine, 135 mm sodium chloride buffer, pH 6.8–7.2.

Donors and recipients

Kidneys were harvested from heterozygous transgenic pigs expressing high levels of human CD55 and moderate levels of human CD59 (3). The pigs were aged from 12 to 16 weeks and weighed between 34 and 56 kg. Kidneys were transplanted into 11–15 kg adult female baboons (Papio hamadryas) supplied by the National Baboon Colony, Royal Prince Alfred Hospital, Sydney, Australia.

Surgery

Surgical techniques were as previously described (3) except that pig kidneys were perfused after harvest with 100 units of rhATIII in 10 mL normal (0.9%) saline and were not biopsied following transplantation. Briefly, the donor pig left kidney was removed and placed in cold normal saline with sterile ice. After a cold ischemia time of approximately 2 h, the kidney was transplanted into the recipient baboon, from which both kidneys had been removed. Recipients were fitted with a primate jacket and tether (Lomir Biomedicals, Quebec, Canada), and a central venous catheter was passed through the tether and connected to an infusion pump via a swivel mechanism (Lomir) attached to the side of the cage. All procedures were approved by the Animal Ethics Committee of St. Vincent's Hospital and the Animal Welfare Committee of the Central Sydney Area Health Service.

Postoperative monitoring and treatment

Blood samples were taken every 8 h for full blood examination and to measure serum creatinine and urea, total protein, electrolytes, ATIII, APTT, INR, fibrinogen, and d-dimer. The Pathology Department of St. Vincent's Hospital, Melbourne, performed all analyses. ATIII was measured using a functional (chromogenic) assay. Activated partial thromboplastin time (APTT) and international normalized prothrombin ratio (INR) were assayed by standard methods. Fibrinogen was measured by the Clauss method using an STA-Fibrinogen 5 kit (Diagnostico Stago, Asnieres-sur-Seine, France). d-dimer level was measured using the Dimertest (Agen Biomedical, Brisbane, Australia). After each blood sampling, recipients were infused over a 1-h period with 250 units of rhATIII per kg of body weight in 100 mL sterile normal saline, followed by 17 units LMWH (Fragmin, Lilly, Indianapolis, IN, USA) per kg and finally 20 mg of the diuretic frusemide (Lasix, Hoechst Marion Roussel Australia).

Histopathology and immunopathology

It was considered unwise to biopsy anticoagulated recipients, so histological analysis was confined to tissue sections from grafts removed during or after rejection. Four-micrometer sections were cut from both fresh-frozen and fixed (10% formalin), paraffin-embedded samples. Paraffin sections were stained with hematoxylin and eosin using standard procedures and examined by light microscopy. Fresh-frozen sections were stained for platelet deposition using an anti-human platelet glycoprotein GPIIbIIIa monoclonal antibody (Lilly, Indianapolis, IN, USA) as previously described (3). For detection of infiltrating baboon T cells, B cells, and monocytes/macrophages, paraffin sections were stained as previously described (3) using anti-human CD3, CD20cy, and CD68 antibodies (Dako Australia, Botany, Australia), respectively. All anti-human antibodies were cross-reactive with baboon.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

RhATIII dose

Since ATIII had not been used previously in pig-to-primate xenotransplantation, the dose was initially based on a report that supplementation of rhATIII to between 10 and 20 times the physiological level (1 unit/mL) was necessary to prevent DIC in a baboon model of sepsis (17). The first recipient was treated by infusion of 500 units of rhATIII per kg of body weight immediately post-transplantation (t = 0 h), which was predicted to increase circulating ATIII to 10–15 units/mL. This was followed by infusion of 35 units LMWH per kg, a dose determined in our previous study (3). Unfortunately the baboon died due to pulmonary edema, although the relative contribution of anticoagulation and aggressive fluid replacement could not be determined. Nevertheless, as a precaution, the doses of both rhATIII and LMWH (n = 3) or rhATIII alone (n = 1) were halved for subsequent recipients and the first doses were not given until baboons had recovered from surgery (t = 2 h). The same doses were repeated every 8 h until the end of the experiment, resulting in ATIII levels of approximately 7 units/mL post-dose and 1.2–2 units/mL pre-dose. No bleeding complications were observed, and three of the four transplants performed were technically successful. The one exception, from the rhATIII + LMWH group, exhibited declining kidney function, which was found upon autopsy to be due to ureteric obstruction at the level of the bladder anastomosis; the data from this experiment were included in the analysis of coagulopathy but not of xenograft function.

Effect of rhATIII treatment on renal xenograft function

Renal xenograft function was monitored by measuring the serum creatinine concentration. Graft function in this study was compared with that of the 4 grafts from our previous study (3) using LMWH alone (dashed lines, Figure 2). Treatment with rhATIII significantly prolonged stable graft function. Whereas the historical group showed markedly diminished function by 2 days (≥ 65% reduction in estimated glomerular filtration rate), the rhATIII group (n = 3) showed no loss of function for 4–5 days (Figure 2). One recipient was electively euthanazed at day 4 with normal kidney function to investigate early histopathological changes. Interestingly, rhATIII did not require LMWH to exert this protective effect.

image

Figure 2. Renal xenograft function. The historical group treated with LMWH alone (recipients D17 ∅, D20 ×, D24 ⊕ and D31 ★, dashed lines) showed rises in serum creatinine, indicating kidney damage, as early as day 1–2. In contrast, recipients treated with rhATIII plus LMWH (D7 □ and D23 ▵) or rhATIII alone (E17 ○, thick line) showed stable graft function for at least 4 days. Note that recipient D23 was sacrificed at day 4 for histological analysis of the graft, which was still functional.

Download figure to PowerPoint

Effect of rhATIII treatment on the development of coagulopathy

We have previously shown that LMWH prevented profound thrombocytopenia in xenograft recipients but had no detectable effect on derangements to the extrinsic and intrinsic coagulation pathways, which were manifested as prolonged APTT and elevated INR, respectively (3). Treatment with rhATIII either alone or with LMWH attenuated the thrombocytopenia (Figure 3A), but in addition dramatically slowed the rises in APTT and INR (Figure 3B,C). Thrombin clotting time was also relatively normal in the rhATIII-treated recipients for 4–5 days (data not shown). However, the accumulation of soluble fibrin degradation products (d-dimer) was not affected by rhATIII (Table 1), indicating that fibrinolytic activation was still occurring. Plasma fibrinogen rose in all recipients for about 2 days following transplantation, probably as an acute phase reaction, and then gradually declined to approximately pretransplant levels (data not shown).

image

Figure 3. Coagulation function in xenograft recipients. A, platelet count; B, activated partial thromboplastin time (APTT), a measure of the extrinsic coagulation pathway; C, international normalized prothrombin ratio (INR), a measure of the intrinsic coagulation pathway. LMWH-treated recipient D20 □×, dashed line) maintained acceptable platelet numbers but showed marked disturbances to both the APTT and INR within 1–2 days. Treatment with rhATIII with (D7 □, D23 ▵, and B13 ◊) or without (E17 ○, thick line) LMWH prevented the development of coagulopathy for at least 4 days.

Download figure to PowerPoint

Table 1. :  Serum concentration of d-dimer (µg/mL)
 TreatmentTime post-transplant (h)
BaboonLMWHrhATIII0–218–2246–5274–7894–100108–110120–124
  1. ND, not done.

D20+< 0.250.5–1.0ND2.0–4.02.0–4.0> 4> 4
D7++< 0.250.5–1.01.0–2.01.0–2.02.0–4.02.0–4.0> 4
D23++0.25–0.50.25–0.52.0–4.0> 42.0–4.0  
B13++0.25–0.50.5–1.01.0–2.0> 4> 4  
E17+< 0.251.0–2.01.0–2.0> 4> 4> 4 

Histopathology of xenografts

Grafts removed from the rhATIII-treated group presented a quite different histological picture to those from the historical LMWH-treated group. The latter showed significant interstitial and glomerular hemorrhage (e.g. recipient D20, day 5, Figure 4A), along with widespread areas of segmental infarction (not shown). In contrast, grafts from rhATIII-treated recipients (e.g. recipient E17, day 5, Figure 4B) were edematous but contained no necrotic areas, and with one exception (recipient D7) showed only minor bleeding, confined largely to glomeruli. The graft from D7, which was removed at day 6 amid signs of rapidly developing rejection (Figures 2 and 3), contained massive and widespread hemorrhage, suggesting a sudden breach of the vasculature. All grafts showed uniform platelet deposition on glomerular vessels but not on small to medium vessels; intertubular sinusoids were mostly negative at day 4, but many displayed significant platelet deposits at days 5 and 6 (data not shown). Like the historical group, grafts from the rhATIII treatment group also contained an intense cellular infiltrate consisting predominantly of T cells and macrophages, with smaller numbers of B cells (data not shown).

image

Figure 4. Histopathology of xenografts removed at day 5 after transplantation. A, graft removed from recipient D20 (historical group, LMWH-treated), showing interstitial and glomerular hemorrhage and numerous protein-plugged tubules. B, graft removed from recipient E17 (rhATIII alone), showing only relatively minor bleeding. The graft is edematous and contains a pronounced cellular infiltrate that is particularly evident surrounding the blood vessel at top center. Original magnification, ×50.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Activation of coagulation with systemic consumption of clotting factors is a major problem in xenotransplantation, threatening both the integrity of the graft and the health of the recipient. Despite the development of strategies to deplete xenoreactive antibodies, inhibit complement activation, and suppress the cellular immune response, thrombotic complications in primate recipients of porcine solid organ xenografts are still frequently observed (18). Similarly, porcine cellular grafts (islets of Langerhans) have been shown to trigger the coagulation and complement cascades in primates, although damage was reduced by treatment with heparin and soluble complement receptor 1 (19). We demonstrate in this study that treatment with high doses of recombinant human ATIII protects renal xenografts from early injury due to coagulation, and delays the development of coagulopathy.

We reported previously that rejected grafts removed from recipients treated with LMWH showed a characteristic histological picture of wedge-shaped necrotic tissue interspersed with regions of significant interstitial hemorrhage (3). This and a gradual decline in graft function were suggestive of progressive thrombotic occlusion of small to mid-sized blood vessels, causing segmental ischemic necrosis. The consumption of clotting factors presumably continued to the point that DIC developed, with prolonged clotting times (which would not be expected to the same extent in untransplanted animals given this dose of LMWH) leading to uncontrolled bleeding within the graft. In contrast, grafts in rhATIII-treated recipients maintained normal function for 4–5 days before being rapidly rejected, with histology showing no segmentalinfarction. These results indicate that, despite the fact that the treatment regimen was not optimized and that there may have been some injury to grafts before the first treatment at t = 2 h, rhATIII reduced the extent of early coagulation and consequent injury, ensuring better graft function. A corollary of this effect was that platelets and clotting factors were not consumed faster than they could be replaced. The rise in d-dimer was not prevented, probably because d-dimer is a very sensitive index of fibrinolysis, a process which is not influenced in a negative fashion by ATIII. In fact, rhATIII treatment may predispose to high levels of d-dimer by blocking activation of the zymogen thrombin activated fibrinolysis inhibitor (TAFI) (20). More importantly, however, rhATIII delayed the increase in the APTT and INR, reducing the risk of life-threatening bleeding in the recipients. Nevertheless in the absence of additional treatment, the cellular immune response appeared to proceed unchecked and began to destroy the vasculature after 4 or 5 days, leading to edema, some bleeding and rapid graft loss.

The protective effect of rhATIII was observed even in the absence of LMWH, despite heparin being a dramatic potentiator of ATIII's anti-coagulatory activity. There are several possible explanations for this. First, there may be sufficient circulating or endothelium-bound heparin/heparin-like molecules naturally present in recipients to effectively interact with the elevated levels of ATIII. However, the physiological presence of heparin in plasma is controversial (21), and heparan sulfate is lost from the endothelial cell surface upon activation (22). Second, it has been proposed that high doses of ATIII interact with heparin-like glycosaminoglycans (GAGs) on the endothelium to promote the release of prostacyclin, which has well-known anti-inflammatory effects including vasodilatation and inhibition of platelet activation (23). In fact, circulating heparin blocks these effects by preventing ATIII from binding to cell surface GAGs (23), and for this reason rhATIII was always infused before LMWH in this study. The relative importance of the anticoagulant and anti-inflammatory properties of ATIII in the xenograft setting remains to be determined.

The success of pig-to-human xenotransplantation is likely to depend on a combination of strategies to deal with xenoantibody binding, complement activation, endothelial cell activation, and the cellular immune response. Even if these problems are solved, however, inappropriate activation of coagulation may persist as a consequence of molecularincompatibilities of nonimmunological origin (9). This study demonstrates that such intragraft thrombosis is not an intractable problem. Although long-term ATIII therapy is unlikely to be feasible because of the risks of systemic anticoagulation, there are many novel anticoagulant, antiplatelet, and thrombolytic agents yet to be tested. Alternatively, genetic modification of the donor pig to express human antithrombotic molecules such as TM and TFPI may prove to be a useful approach.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

The authors are grateful to Dr Sudhir Shah and Dr Yann Echelard from Genzyme Transgenics Corporation for helpful discussions and suggestions.

This study was supported in part by financial assistance from Genzyme Transgenics Corporation, Framingham, MA, USA.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    Nagayasu T, Platt JL. A perspective on xenotransplantation. Graft 1999; 2: S152S158.
  • 2
    Logan JS. Prospects for xenotransplantation. Curr Opin Immunol 2000; 12: 563568.
  • 3
    Cowan PJ, Aminian A, Barlow H et al. Renal xenografts from triple-transgenic pigs are not hyperacutely rejected but cause coagulopathy in non-immunosuppressed baboons. Transplantation 2000; 69: 25042515.
  • 4
    Cozzi E, Bhatti F, Schmoeckel M et al. Long-term survival of nonhuman primates receiving life-supporting transgenic porcine kidney xenografts. Transplantation 2000; 70: 1521.
  • 5
    Chen RH, Naficy S, Logan JS, Diamond LE, Adams DH. Hearts from transgenic pigs constructed with CD59/DAF genomic clones demonstrate improved survival in primates. Xenotransplantation 1999; 6: 194200.
  • 6
    Robson SC, Candinas D, Hancock WW, Wrighton C, Winkler H, Bach FH. Role of endothelial cells in transplantation. Int Arch Allergy Immunol 1995; 106: 305322.
  • 7
    Ruf W, Edgington TS. Structural biology of tissue factor, the initiator of thrombogenesis in vivo. FASEB J 1994; 8: 385390.
  • 8
    Coughlin SR. Thrombin signalling and protease-activated receptors. Nature 2000; 407: 258264.DOI: 10.1038/35025229
  • 9
    Robson SC, Cooper DKC, D'Apice AJF. Disordered regulation of coagulation and platelet activation in xenotransplantation. Xenotransplantation 2000; 7: 166176.
  • 10
    Biancone L, Cantaluppi V, Segoloni G et al. Role of platelet-activating factor in functional alterations induced by xenoreactive antibodies in porcine endothelial cells. Transplantation 2000; 70: 11981205.
  • 11
    Siegel JB, Grey ST, Lesnikoski B-A et al. Xenogeneic endothelial cells activate human prothrombin. Transplantation 1997; 64: 888896.
  • 12
    Kopp CW, Siegel JB, Hancock WW et al. Effect of porcine endothelial tissue factor pathway inhibitor on human coagulation factors. Transplantation 1997; 63: 749758.
  • 13
    Bauer KA, Rosenberg RD. Role of antithrombin III as a regulator of in vivo coagulation. Semin Hematol 1991; 28: 1018.
  • 14
    Perry DJ. Antithrombin and its inherited deficiencies. Blood Rev 1994; 8: 3755.
  • 15
    Weitz JI. Low-molecular-weight heparins. N Engl J Med 1997; 337: 688698.
  • 16
    Dickneite G, Leithauser B. Influence of antithrombin III on coagulation and inflammation in porcine septic shock. Arterioscler Thromb Vasc Biol 1999; 19: 15661572.
  • 17
    Minnema MC, Chang AC, Jansen PM et al. Recombinant human antithrombin III improves survival and attenuates inflammatory responses in baboons lethally challenged with Escherichia coli. Blood 2000; 95: 11171123.
  • 18
    Buhler L, Basker M, Alwayn I et al. Coagulation and thrombotic disorders associated with pig organ and hematopoietic cell transplantation in nonhuman primates. Transplantation 2000; 70: 13231331.
  • 19
    Bennet W, Sundberg B, Lundgren T et al. Damage to porcine islets of Langerhans after exposure to human blood in vitro, or after intraportal transplantation to cynomologus monkeys: protective effects of sCR1 and heparin. Transplantation 2000; 69: 711719.
  • 20
    Engelberg H. Evidence that endogenous heparin activity deficiency may be an important factor in atherogenesis. Semin Thromb Hemost 1997; 23: 159166.
  • 21
    Bajzar L, Manuel R, Nesheim ME. Purification and characterization of TAFI, a thrombin-activable fibrinolysis inhibitor. J Biol Chem 1995; 270: 1447714484.
  • 22
    Platt JL, Vercellotti GM, Lindman BJ, Oegema TR, Bach FH, Dalmasso AP. Release of heparan sulfate from endothelial cells. Implications for pathogenesis of hyperacute rejection. J Exp Med 1990; 171: 13631368.
  • 23
    Okajima K, Uchiba M. The anti-inflammatory properties of antithrombin III: new therapeutic implications. Semin Thromb Hemost 1998; 24: 2732.