The adhesion of labelled neutrophils on synthetic vascular grafts. An experimental porcine study

Eur J Vasc Surg 7, 257-262 (1993)

The Adhesion of Labelled Neutrophils on Synthetic Vascular Grafts. An Experimental Porcine Study* H~kan P~irsson, Wieslav Jundzill, Torbj6rn Jonung, Johan Th6rne and Lars Norgren Department of Surgery, Lund University, Sweden The adhesion of neutrophils onto different vascular grafts was studied in vivo in a pig model. In acute experiments autologous l~lln-labelled neutrophils were reinfused after end-to-side implantation of 5 cm, 6 mm internal diameter grafts. The dynamic deposition on each graft was determined for 300min in vivo. Static measurements in vitro concluded the study. The adhesion was greater in Dacron and collagen coated Dacron grafts compared to expanded polytetrafluoroethylene (ePTFE) and to Dacron grafts coated with a polymer. The segmental activity along all the grafts increased towards the distal anastomosis. The results suggest different inflammatory response to various graft materials.

Key Words: Vascular grafts; Inflammatory reaction; Neutrophils.

Introduction The implantation of a biomaterial elicits both an acute and a chronic inflammatory response. 1"2 The degree of the inflammatory response is an inverse function of the biocornpatibility of the material. The interaction of circulating blood elements with the nonbiological surfaces of prosthetic biomaterials activates the complement system resulting in adhesion of mainly platelets and neutrophils. 3-5 Attention has been paid to the coagulation system and to the platelets, as thrombosis is a major cause of graft failure. 6'7 Leukocytes may also be activated w h e n they pass through a vascular graft with subsequent release of lyric enzymes and toxic oxygen radicals, s A heavy polyrnorphonuclear leukocyte (PMN) infiltration of grafts has been demonstrated in the immediate posrimplant period. 9 In a recent study 1° the dynamic deposition of labelled leukocytes on polytetrafluoroethylene (ePTFE) grafts has been reported. Adherent PMN cells as well as macrophages may contribute to vascular graft occlusion by augumentarion of the co* Presented at the 6th Annual Meeting of the European Society for

Vascular Surgery, Athens, September 1992. Please address all correspondence to: Dr Hakan Parsson, Department of Surgery, Lund University, S-221 85 Lund, Sweden. Supported by a research grant from the University of Lund. 0950-821X/93/070257+06 $08.00/0 © 1993 Grune & Stratton Ltd.

agulation and platelet stimulation, releasing platelet activating factor. 11 The release of smooth muscle cell mitogens might further contribute to intimal hyperplasia. 12 Differences between various vascular graft materials regarding leukocyte activation and adhesion have been demonstrated in vitro but not in vivo. 13,14 The aim of this study was therefore to evaluate the acute deposition of 111In-labelled autologous neutrophils in different grafts in vivo utilizing an experimental animal model.

Materials and Methods Grafts

Five different kinds of grafts were used: (1) A knitted Dacron graft, Micron ® (Intervascular Inc., U.S.A.) (Dacron); (2) A knitted Dacron graft with an external surface polymer coating, Potygraft® (Intervascular Inc., U.S.A.) (PG); (3) A Polygraft® with heparin bonding (Intervascular Inc., U.S.A.) (PG-H); (4) A knitted Dacron graft with external collagen

258

H. P~irsson et aL

coating, Hemaguard ® (Intervascular Inc., U.S.A.) (HG); and (5) A thinwalled expanded polytetrafluoroethylene graft, Gore-Tex ® (W. L. Gore & Ass. Inc., U.S.A.) (ePTFE). For heparin bonding of PG-H grafts a complex of heparin with tridodecylmethylammonium chloride (TDMAC) was used. This results in hydrophobic association of hydrocarbon chains of the TDMAC with the graft, leaving the quaternary ammonium part of TDMAC exposed to the surface forming ionic bonds with the highly anionic sulphate groups of heparin. 15-17 All grafts were of 5cm length and with an internal diameter of 6 mm.

Surgical procedure Eighteen pigs of Swedish countrybreed (weight 2025 kg) were used in this study. They were fasted for 12 h before the experiment. Anaesthesia was effected with an intramuscular injection of azaperonum (Stresnil, ® Leo, Sweden) and repeated intravenous (i.v.) injections of metomidatum (Hypnodil, ® Janssen Pharmaceutical, Belgium) when necessary. The animal was intubated and mechanically ventilated with a gas mixture of 50% oxygen and 50% nitrous oxide using a Servo ventilator 900B (Siemens-Elema AB, Sweden). A central venous catheter was placed in the left jugular vein for withdrawal of blood and reinfusion of the labelled cells. Through a midline incision the abdominal aorta and iliac arteries were exposed. The grafts were randomly inserted in pairs and were not preclotted. Eight Dacron, seven PG, seven PG-H, seven HG and seven ePTFE grafts were implanted. The proximal anastomosis was placed in the common iliac artery just distal to the aortic bifurcation with an end-to-side technique. The distal end-to-side anastomosis was sutured to the common iliac artery above the internal iliac artery. The anastomoses were made with continuous 6/0 monofilament polypropylene sutures (Prolene, ® Ethicon Inc., U.S.A.). The intermediate artery was ligated and the arterial clamps were removed simultaneously. Graft patency was assessed by palpation. All pigs were given Dextran40 (Rheomacrodex, ® Kabi-Pharmacia AB, Sweden) i.v. at a rate of 50ml/h. The infusion started prior to clamping of the vessels and was finished at the end of the measurement. Eur J VascSurg Vol 7, May 1993

Leucocyte labelling Forty millilitres of blood was withdrawn from the i.v. line prior to the surgical procedure. Neutrophil (PMN) isolation and labelling using a methyl-cellulose Percoll-density gradient and 111In isotope 18'19 was done simultaneously with surgery. Fifty microlitres were saved for determination of the labelling efficiency. 2° The neutrophils were reinfused intravenously as soon as blood flow had been restored in the grafts and haemostasis achieved.

Nuclear imaging and data analysis The animal was placed under a scintillation camera (Maxicamera, General Electric, U.S.A.) and the activity was continuously measured for 300min. Sequential 300 s images were taken and stored on a computer (Gamma 11, Digital Equipment, U.S.A.). Blood sampling for white blood cell (WBC) count, differentiation count, platelet count and peripheral blood radioactivity was done prior to operation, immediately after injection of labelled neutrophils and then every 30 rain. All blood samples were collected in EDTA-tubes. White blood cell count and platelet count were measured by phase contrast microscopy. Differential WBC was determined in films stained with Giemsa-May-Grunewald dye. Fifty microlitres of blood was centrifuged in a capillary tube for 5 min at 1200 g. The activity in the cell fraction and cell-poor supernatant was measured as above. Finally the grafts with adjacent vessels were explanted and thoroughly rinsed in saline. One femoral artery of 5 cm length was also explanted and used as a reference vessel. The specimens were cut in three parts, one proximal, One mid and one distal segment. Each segment was measured in a NaI-well counter, dried and weighed. Corrections were made for different graft surfaces. After completion of the study, regions of interest (ROI) were selected over the grafts and time activity curves generated. The dynamic deposition of labelled neutrophils was calculated from these curves.

Thrombus free surface (TFS) Each graft was photographed immediately after explantation. The pictures were digitalized and planimetrically analysed by a computer.

Neutrophil Adhesion to Grafts

259

300

Statistical analysis D a t a w e r e e x p r e s s e d as m e a n + S.D. Statistical c o m parisons were performed by ANOVA and unpaired S t u d e n t ' s t-test.

,~ -~

HG Dacron PG PG-H

"~

e-PTFE Results < 100 ~ T h e P G a n d P G - H g r a f t s w e r e stiff, b u t did n o t c a u s e technical p r o b l e m s . F r o m all h e p a r i n i z e d grafts a c o n s i d e r a b l e b l e e d i n g w a s n o t i c e d b u t after slight manual compression the bleeding stopped. T h e labelling efficiency w a s 98 + 2% ( m e a n + S.D.). T h e m e a n i n j e c t e d activity w a s 10.8 + 1.2 MBq. T h e n u m b e r of circulating l e u k o c y t e s (WBC) w a s significantly r e d u c e d (p < 0.001) after b l o o d f l o w r e s t o r a t i o n (Fig. 1). T h e d i f f e r e n t i a t i o n c o u n t r e v e a l e d a 20

WBC X

I I

I

I "--k,

i

L N

i

ip < 0.001 I

i

i

I

i

I

i

I

i

I

i

J

I

I

i

I

I i

I

Aorta

I 300

Time (min) Fig. 2. Activity as function of time after operation. Curves are generated for the native aorta and implanted grafts, following reinfusion of rain-labelled neutrophils. Aorta represents aortic activity; e-PTFE is a thinwalled expanded polytetrafluoroethylene graft, Gore-Tex® (W. L. Gore & Ass. Inc., U.S.A.); PG-H is a knitted Dacron graft with an external surface polymer coating and heparinbonding Polygraft® (Intervascular Inc., U.S.A.); PG is a knitted Dacron graft with an external surface polymer coating, Polygraft® (Intervascular Inc., U.S.A.); Dacron is a knitted Dacron graft, Micron® (Intervascular Inc., U.S.A); HG is a knitted Dacron graft with external collagen coating, Hemaguard ® (Intervascular Inc., U.S.A.).

up < 0.001 0 C Operation

I

demonstrated significant d i f f e r e n c e s a m o n g the grafts in t h e s a m e r a n k of o r d e r as d e m o n s t r a t e d in t h e d y n a m i c m e a s u r e m e n t (Fig. 3). T h e r e g i o n a l dis-

.~ 10 8 0 r.)

0

~

I

I

300 Time (min) Fig. 1. Serial changes of white blood cell count (WBC), neutrophil (N) and lymphocyte (L) counts (106/ml) during the acute experiments. Values are expressed as mean + S.D.

P<

0

parallel r e d u c t i o n in n e u t r o p h i l s (p < 0.001) w h i c h w a s partially r e s t o r e d . T h e l y m p h o c y t e c o u n t w a s stable d u r i n g t h e e x p e r i m e n t (Fig. 1). T h e free r a d i o activity in p l a s m a (less t h a n 2 % ) w a s stable t h r o u g h o u t t h e e x p e r i m e n t , i n d i c a t i n g n o release of activity f r o m labelled n e u t r o p h i l s . Figure 2 s h o w s , in relative units, the i n c r e a s e in activity as a f u n c t i o n of t i m e in t h e grafts. A signific a n t i n c r e a s e in r a d i o a c t i v i t y o v e r all t y p e s of grafts w a s initially r e g i s t e r e d , a n d w a s stabilized after 9 0 1 2 0 m i n . D a c r o n a n d collagen t r e a t e d D a c r o n grafts w e r e significantly differing f r o m ePTFE grafts (p < 0.01). P o l y m e r c o a t e d grafts d i f f e r e d to a lesser d e g r e e (p < 0.05). T h e p e r i p h e r a l activity w a s stable during the measurement. In vitro m e a s u r e m e n t of the total graft activity

i 1

0.001

~ 1.0 :g

5

0

ePTFE HG DACRON PG PG-H A.FEM n=7 n=7 n=8 n=7 n=7 n=18 Fig. 3. Total activity of explanted grafts. Values represent %o of injected activity and are expressed as mean _+ sin. t r i b u t i o n a l o n g the grafts s h o w e d t h a t the activity w a s m o s t p r o n o u n c e d in the distal p a r t s of all k i n d s of grafts (Fig. 4). At e x p l a n t all grafts w e r e p a t e n t , c o v e r e d b y a g l i s t e n i n g r e d film a n d s o m e t h r o m b u s material, b u t without gross thrombus formation. The determin a t i o n of t h r o m b u s free surface (TFS) s h o w e d differEur J Vasc Surg Vol 7, May 1993

260

H. P~irsson et aL

phils augment coagulation by release of platelet activating factor,11" 29 thromboxane A23°'31 and by a direct interaction between neutrophils and platelets. 32 The present study confirms this immediate reaction with a rapid increase of neutrophil deposition during the first hour of observation and a subsequent steady-state on all graft materials. This pattern of reaction is in line with former observations on platelet interaction with both autologous vein, 33 ePTFE 7 and Dacron grafts. 6"7 The peripheral activity was stable and no increase of "free" circulating allIn was observed indicating a stable binding to neutrophils. 0 prox mid dlst rox mid dlst prox mid dlst prox mid dlst prox mld dlst The labelling procedure also gives a very high specifiePTFE HG DACRON PG PG-H city for neutrophils with a labelling efficiency of 98% Fig. 4. In vitro activity measurement in segments of explanted thus promoting a good picture resolution on the scingrafts representing %0of injected activityexpressed as mean + S.D. Prox, proximal third; med, medial third; dist, distal third of the tillation camera. graft. In vitro studies have shown significant differences regarding the acute inflammatory response between various vascular graft materials. 2'13 In this study ePTFE had less adhesion of neutrophils compared to Dacron. 14 A reduced activation of the complement system by ePTFE compared to Dacron has earlier been demonstrated. 9 The in vivo results in this study support the earlier findings from the in vitro 25 studies. r~ Preclotting of a Dacron graft implies less activation of the complemental system and reduced neutrophil adhesion in vitro, and evaluation by scanning electron microscopy has shown the adhesion to the Dacron fabric and not to the fibrin deposits. 14 The 0 PTFE HO DACRON PG PG-H addition of a polymer to the Dacron material (PG) n=7 n=7 n=8 n=7 n=7 reduced the adhesion of neutrophils as compared to Fig. 5. Thrombus free inner surface (TFS) of explanted graft seg- pure Dacron, which probably reflects a reduced ments. Values are expresses as mean + S.D. exposed surface area of the Dacron material. The macroscopic evaluation of the grafts showed ences between the graft materials with highest scores a high TFS for the ePTFE grafts and increased de-~ for ePTFE-grafts and lowest for pure Dacron (Fig. 5). posits of thrombus material in Dacron grafts, which All kinds of coated grafts had very similar TFS. also supports an increased interaction at the b l o o d graft interface. The addition of ionic-bonded heparin to the inner surface of the graft should possibly reduce the Discussion adherence of platelets and leukocytes. It has been shown in rats that heparin is a specific inhibitor of The accumulation of blood elements on different graft smooth muscle cell migration and intimal thickening materials has been extensively investigated both in after carotid artery balloon injury. 34 No beneficial acute models with labelled platelets 6'7 and in long- effect from heparin-bonding was observed in the term experiments with various morphologic tech- grafts regarding adhesion of neutrophils. A report by niques. 21-23 These experiments have focused on Kalman et al. 24 focuses on the role of complement platelet activation and subsequent adverse events. activation and platelet aggregation by air bubbles Recent works indicate complement and neutrophil trapped in Dacron grafts. The procedure by which activation by different vascular graft ma- heparin is bonded to TDMAC might imply excessive terials. 2"9"14'24 The activation of the complement activation of complement by such bubbles and also system promotes platelet and neutrophil cell ad- decreased anticomplement activation. Ionic bonding hesion and aggregation. 4"s'24-28 Activated neutro- has the disadvantage of heparin removal. In this

025 I

I

Eur J Vasc Surg Vol 7, May 1993

Neutrophil Adhesion to Grafts

study it was not possible to perform analysis of the amount of heparin b o n d e d to the grafts at explantation as this would have implied the addition of another isotope. Theoretically, the heparin might be rinsed off the surface shortly after implantation of the graft. A previous study, however, showed an initial loss of 10% and no further loss for 7 days in experiments in vivo. 35 By implantation it was noted that the grafts treated with heparin were permeable to blood, and slight manual compression of the graft was necessary for haemostasis. This might be due to the coating procedure with TDMAC-heparin and a subsequently increased porosity of the grafts despite the polymer coating. The higher level of neutrophil activity registered in the distal part of the grafts may indicate a lower shear stress in this region allowing neutrophils to adhere and aggregate on the graft s u r f a c e . 36 It is also possible that blood flow turbulence through the endto-side anastomoses promotes neutrophil activation and subsequent release of proteolytic enzymes and oxygen-derived free radicals. 37-4° The release of platelet activating factor by activated neutrophils might stimulate smooth muscle cells 12 and contribute to later intimal hyperplasia. Leukocyte count and differentiation count showed a significant reduction of circulating leucocytes and a parallel reduction of neutrophils during the surgical procedure. This decline was partially restored during the observation time and probably reflects leukocyte trapping in lungs and liver due to the surgical trauma 19 and sequestration in the peripheral capillary beds. Activation of neutrophils and the complement system by the graft can theoretically result in increased adhesion and sequestration in the microcirculation distal to the graft and subsequently in reduced peripheral perfusion. In conclusion, differences regarding the in vivo adhesion of labelled neutrophils on various vascular graft materials have been demonstrated. Knowledge of the biocompatibflity of a graft material not only includes studies of platelet reaction and coagulation but also of variables determining inflammatory response, such as neutrophil cell reactions and complem e n t activity. Differences between graft materials is evident but further studies are necessary to evaluate whether the acute reaction has any influence on longterm healing. There may also be differences between animal experiments and the clinical situation.

261

References 1 IKaTNERB, JOHNSTONA, LENKT. Biomaterial surfaces. J Biomed Mater Res 1987; 21: 59-87. 2 MARCHANTRE,MILLERKM, ANDERSONJM.Invivobiocompatibility studies. In vivo leukocyte interactions with Biomer. l Biomed Mater Res 1984; 18, 1169-1190. 3 BROTHERSTE, GRAHAMLM, TILL GO. Systemic effects of prosthetic vascular graft implantation. Surgery 1988; 104: 375-382. 4 LONG J, DESANTIS S, SHORS E, et al. Thrombogenicity and the interaction of proteins, platelets and white cells. Biomat Ned Dev Art Org 1983; 11: 63-72. 5 STEWARTnJ, LYNCHPR, REICHLEFA, RITCHIEWGM, SMITH A, SCHAUBRG. The adhesion of leucocytes, erythrocytes, and noncellular material to the luminal surface of natural and artificial blood vessels in vivo. Ann N Y Acad Sci 1977; 283: 179-207. 6 P.KRSSONH, JONSSON gA, NORGREN L, STRANDSE, THORNEJ. Experimental evaluation of small diameter synthetic arterial grafts. Inter Anglo 1988; 7: 60-64. 7 ALLEN BT, SICARD CA, WELCH MJ, MATHIAS CJ, CLARK RE. Platelet deposition on vascular grafts. Ann Surg 1986; 203: 318328. 8 SACKST, MOLDOW CF, CRADDOCKPR, BOWERSTK, JACOBHS. Oxygen radicals mediate endothelial cell damage by complement-stimulated granulocytes. An in vitro model of immune vascular damage. J Clin Invest 1978; 61: 1161-1167. 9 SHEPARD AD, GELFANDJA, CALLOWAD, O'DONNELL TF. Complement activation by synthetic vascular prostheses. J Vasc Surg 1984; 6: 829-836. 10 LUNDELLA, BERGQVISTD, LEIDES, LINDBLADB, LJUNGBERGJ. The effect of a combined Thromboxane receptor- and synthesisantoganist on platelet, fihrinogen, and leucocyte uptake in ePTFE grafts: an experimental study in sheep. Eur J Vasc Surg 1992; 6, 276-281. 11 OSTERMANG, TILLJ, THIELMANNK. Studies on the stimulation of human blood platelets by semi-synthetic platelet-activating factor. Thromb Res 1983; 30: 127-132. 12 KAHALEH MB, DELUSTROF, BOCKW, LERoY ED. Human monocyte modulation of endothelial cells and fibroblasts growth, possible mechanism for fibrosis. Clin Immunol Immunopahol 1986; 39: 242-255. 13 KRAUSETJ, ROBERTSSONFM, LIESCHJB, WASSERMANAJ, GRECO RS. Differential production of Interleukin 1 on the surface of biomaterials. Arch Surg 1990; 125: 1158-1160. 14 KOTTKE-MARCHANTK, ANDERSONJM, MILLER KIVI, MARCHANT RE, LAZARUSH. Vascular graft-associated complement activation and leucocyte adhesion in an artificial circulation. J Biomed Mater Res 1987; 21: 379-397. 15 ESQUIVELCO, BJORCKCG, BERGENTZSE, et al. Reduced thrombogenic characteristics of expanded polytetrafluoroethylene and polyurethane arterial grafts after heparin bonding. Surgery 1983; 95: 102-107. 16 HARVEYRA, KIN HC, PINCUSJ, TROOSKINSZ, WILCOXJN, GRECO RS. Binding of tissue plasminogen activator to vascular grafts. Thromb Haemost 1989; 28: 131-136. 17 FREEMANAJ, GARDNERCJ, DODDSMG. Animproved method for bonding heparin to intravascular cannulae. J Pharmacol Methods 1990; 23: 7-11. 18 THAKURML, COLEMANRE, WELCHMJ. Indium-Ill-labelled leucocytes for the localisation of abscesses: preparation, analysis, tissue distribution and comparison with gallium-67 citrate in dogs. ] Lab Clin Med 1977; 89: 217-227. 19 THORNEJ, BLOMQVlSTS, ELMERO, GRAFSTROMG,MARTENSSONL. Polymorphonnclear leucocyte sequestration in the lungs and liver following soft-tissue trauma: an in vivo study. J Trauma 1989; 29: 451-456. 20 TH~)RNEJ, OLSSONP-I, STRANDSE. An experimental method for in vivo studies of pulmonary platelet sequestration. Eur J Nucl Ned 1984; 9: 472-477. 21 KENNY DA, BERGERK, WALKERMW, et al. Experimental cornEur J Vasc Surg Vol 7, May 1993

262

22 23 24 25 26

27

28

29 30 31

H. P~irsson et al.

parison of the thrombogenicity of fibrin and PTFE flow surfaces. Ann Surg 1980; 191: 355-361. GOLDEN MA, HANSSON SR, KIRKMAN TR, SCHNEIDER PA, CLOWES AW. Healing of polytetrafluoroethylene arterial grafts is influenced by graft porosity. J Vasc Surg 1990; 11: 838-845. CLOWESAW, GOWNAM, HANSSONSR, REIDYMA. Role of cellular proliferation in early healing of PTFE prostheses. Am J Pathol 1984; 118: 43-54. KALMAN PG, McCULLOUGH DA, WARD CHA. Evacuation of microscopic air bubbles from Dacron reduces complement activation and platelet aggregation. ]Vasc Surg 1990; 11: 591-598. POOLEY MJ, NACHMAN RL. Human platelet activation by C3a and C3a des-Arg. J Exp Med 1983; 158: 603-607. CRADDOCKPR, FEHRJ, DALMASSOAP, BRIGHAMKL, JACOBHS. Hemodialysis leukopenia. Pulmonary vascular leukostasis resulting from complement activation by dialyzer cellophane membranes. J Clin Invest 1977; 59: 879-888. CRADDOCKPR, HAMMERSCHMIDTD, WHITE JG, DALMASSOAP, JACOBHS. Complement (C5a) induced granulocyte aggregation in vitro: a possible mechanism of complement mediated leucostasis and leucopenia. J Clin Invest 1977; 60: 260-264. O'FLAHERTYJT, KREUTZERDL, WARD PA. Chemotactic factor influences on the aggregation, swelling and foreign surface adhesiveness of human leucocytes. Am J Pathol 1978; 90: 537544. NIEMETZJ, MI;HLFELDERT, CHIEREGME, TROYB. Procoagulant activity of leukocytes. Ann NY Acad Sci 1982; 283: 208. SVENSSONJ, HAMBERGM, SAMUELSSONB. On the formation and effects of thromboxane A2 in human platelets. Acta Physiol Scand 1976; 98: 285. GOLDSTEINIM, MALMSTENCLO, KINDAHLH. Thromboxane generation by human peripheral blood polymorphonudear leucocytes. J Exp ivied 1978; 148: 787-791.

Eur J Vasc Surg Vol 7, May 1993

32 KORNECKIE, EHRLICHYH, EGBRINGR, et al. Granulocyte-platelet interactions and platelet fibrinogen receptor exposure. Am J PhysioI 1988 255; 651-658. 33 GOLDMAN M, NORCOTT HC, HAWKER RJ, HAIL C, DROLC Z, McCoLLUM CN. Femoropopliteal bypass grafts--an isotope technique allowing in vivo comparison of thrombogenicity. Br J Surg 1982; 69: 380-382. 34 FINGERLEJ, JOHNSONR, CLOWESAW, MAJESKYMW, REIDYMA. Role of platelets in smooth muscle cell proliferation and migration after vascular injury in rat carotid artery. Proc Natl Acad Sci 1989; 86: 8412-8416. 35 LARSSONRL, HJELTEMB, ERIKSSONJC, LAGERGRENH, OLSSONP. The stability of glutaraldehyde-stabilized 35-heparinized surfaces in contact with blood. Thromb Haemost 1977; 37: 262-273. 36 MAYROVlTZHN, WIEDEMANMP, TUMARF. Factors influencing leukocyte adherence in microvessels. Thromb Haemostas 1977; 38: 823-830. 37 FRITZ H, JOCHUM M, DUSWALD KH, et al. Granulocyte proteinases as mediators of unspecific proteolysis in inflammation. A review. In Proteinases in Inflammation and Tumour Invasion. New York: Walter De Gruyter, 1986; 1-24. 38 CARREL RW. ~l-Antitrypsin: molecular pathology, leucocytes and tissue damage. J Clin Invest 1986; 78: 1427-1431. 39 SMEDLEYLA, TONNESENMG, WORTHEN GS, et al. Neutrophilmediated injury to endothelial cells: enhancement by endotoxin and essential role of neutrophil elastase. J Clin Invest 1986; 77: 1233-1243. 40 WEISS SJ. Tissue destruction by neutrophils. New Eng J Med 1989; 320: 365-376.

Accepted 18 December 1992