Eur J VascSurg 6, 487-493 (1992)
Endothelial Cell Seeding of Damaged Native Vascular Surfaces: Prostacyclin Production M. M. Thompson, J. S. Budd, S. L. Eady, K. E. Allen, M. James, R. F. L. James and P. R. F. Bell
Department of Surgery, University of Leicester, Leicester, U.K. Endothelial cell seeding has been successfid in reducing the thrombogenicity of pros&etic vascular grafts in animal and clinical studies. The reduction in thrombogenici~ may be attributed to the intrinsic properties of endothelial cells themselves, and &eir abili~ to produce anti-thrombogenic mediators such as prostacyclin, and endothelium-derived relaxing factor. Endothelial seeding of damaged vascular surfaces produced during percutaneous transluminaI angioplasty and endarterectomy is an attractive possibili~ due to the excellent attachment characteristics of the sub-endothelial tissue exposed during these procedures. The ability of endothelial seeded damaged vascular surfaces to produce prostacyclin was measured in an in vitro model of vascular injury. Endothelial-seeded damaged surfaces produced sigr~_~'cantly higher prostacyclin release than did vessels damaged by balloon dilatation (265.Spgcm- 2 min- 1 and 87.Spgcm- 2 rain- 1 respectively). This study provides evidence that endothelial seeding of damaged native vascular surfaces is technically feasible and that seeding may reduce the thrombogenicity of vascular surfaces following balloon dilatation. Key Words: Endo&eliaI cell seeding; Angioplasty; Prostacyclin; Thrombosis.
Introduction Endothelial cell seeding has been proposed as a method of reducing the thrombogenicity of prosthetic vascular grafts. Endothelial cells possess a wide range of anti-thrombotic and anti-proliferative activities which may reduce graft thrombosis and myointimal hyperplasia in seeded prosthetic grafts. Endothelial cell seeding has successfully reduced the thrombogenicity of grafts in animal models, 1-3 but clinical trials have been largely disappointing, 4-7 probably because of inefficient cell harvest and poor cell retention on the graft surface. 8 Damage to the native endothelial cell surface occurs during angioplasty and endarterectomy. 9-~2 Consequent exposure of the sub-endothelial connective tissue promotes platelet adhesion and activation of the coagulation cascade, 13"14 which may result in thrombosis. Rapid restoration of the endothelial cell monolayer achieved by endothelial cell seeding may decrease the thrombotic tendency due to the formation of a mechanical barrier and also by increasing Please address all correspondence to: M. M. Thompson, Department of Surgery,ClinicalSciencesBuilding,LeicesterRoyalInfirmary, PO Box 65, Leicester,LE2 7LX,U.K. 0950-821X/92/050487+07 $08.00/0© 1992Grune & StrattonLtd.
production of anti-thrombotic mediators such as prostacyclin (PGI2), 15 and endothelium-derived relaxing factor (EDRF). 16 Endothelial cell seeding of damaged native vascular surfaces has advantages over seeding of prosthetic grafts; the damaged native vessel produces an ideal surface for endothelial cell attachment, 17 and the area to be seeded is usually smaller thus requiring fewer endothelial cells for the formation of a confluent monolayer. Seeded endothelial cells have been demonstrated to form a confluent thrombo-resistant monolayer on vascular endarterectomy sites,18 resulting in increased patency and a reduction in the degree of myointimal hyperplasia.12 In this study we examined whether endothelial cell seeding of a damaged vascular surface could increase local release of prostacyclin. Materials and Methods
Development of an in vitro model simulating damaged vascular surfaces Segments of long saphenous vein, obtained following coronary artery bypass grafting were immersed in
488
M.M. Thompson et aL
minimal essential medium (MEM; Flow Laboratories, U.K.) and threaded over a 5-mm balloon angioplasty catheter (Olbert Catheter, Meadox, U.K.). The angioplasty balloon was inflated to 9 atm pressure for 30 s, after which time the vein segment was removed from the catheter, opened longitudinally, pinned on to a silicone gel plate and fixed under tension in 4% paraformaldehyde (BDH, Atherstone, U.K.)/2% gluteraldehyde solution (Sigma Chemicals, U.K.). Segments of vein were prepared for light and scanning electron microscopy (SEM). Segments for light microscopy were dehydrated through graded alcohol, transferred to xylene for 180min and then immobilised in wax over a 4-h period. Serial sections of 4 txm thickness were subsequently stained with haematoxylin and eosin (H&E) and elastic van Gieson (EVG). Vein segments for SEM were dehydrated through graded ethanol followed by acetone solutions prior to carbon dioxide critical point drying and sputter coating with gold. Sections were viewed on a DS 130 scanning electron microscope. Dilated and non-dilated vein segments were independently reported by a pathologist (M.J.), who was blind to the sample type.
dothelial cells to damaged vascular surfaces at different seeding times and seeding densities, we would expect the cell attachment in this experiment to be 2 x 105 endothelial cells cm -2 of damaged vein. Non-seeded control veins, and damaged vein segments were both sham seeded for 30 min in the seeding chamber with complete culture medium prior to PGI2 assay. All vein segments were examined by both light and scanning electron microscopy.
Experimental design Prostacyclin production was measured in three vein groups: control, dilated, and dilated plus seeded veins. Vein segments obtained after aorto-coronary bypass grafting acted as controls. Within each experiment, the control, dilated and seeded veins were all derived from a single vein segment which allowed comparison of paired samples. Eight separate experiments were performed.
Prostacyclin measurement Endothelial cell seeding Endothelial cells were isolated from human umbilical veins and then subsequently grown in tissue culture as has previously been described. 19 In this experiment cells from the second and third passage ~were used. Prior to seeding, endothelial cells were harvested from tissue culture by brief trypsinisation (0.1% trypsin in EDTA; Sigma Chemicals, U.K.) and centrifugation (300 x g for 7min at 4°C), and t h e n resuspended in complete culture medium at a concentration of 1 x 107cells m1-1. Endothelial cell seeding was p e r f o r m e d by placing a 5-mm diameter punch biopsy of dilated saphenous vein on silicone gel in a 5-mm diameter seeding chamber. A 100 ~1 aliquot of the endothelial cell suspension was added to the seeding chamber so that the cells only came into contact with the luminal vein surface. The seeding density was 5 x 106 cells cm -2. After incubation in 95% air 5% CO2 for 30rain, the unattached cells were removed by washing the luminal vein surface three times with MEM. The vein segment was then removed from the seeding chamber and PGI2 release measured. From the results of previous experiments which have measured the attachment of i n d i u m - i l l labelled enEur J Vasc Surg Vol 6, September1992
A 5-mm punch biopsy of the vein segment was placed in a single well of a 24-well tissue culture plate which contained I ml of pre-warmed MEM (pH 7.4). The plate was incubated in a 95% air 5% CO2 atmosphere at 37°C. A sample of the medium (100 txl) was taken from the well 30 min after the vein segment was added, and this was assayed for basal PGI2 release. After 30 min, thrombin was added to the well to obtain a final concentration of lum1-1. A further sample of medium was taken 10min after the addition of thrombin, which was assayed for stimulated PGI2 release. Prostacyclin concentration in the samples was assayed by the measurement of 6-keto prostaglandin FI~, the stable metabolite of PGI2 breakdown at pH 7.4, by the use of a commercially available competitive radioimmunoassay kit (Amersham, U.K.).
Results
Light and electron microscopic examination Light microscopic examination of dilated saphenous vein segments revealed endothelial desquamation, rupture of the internal elastic lamina, tangential and
Endothelial Cell Seeding and PGI 2 Release
489
circumferential tears in the tunica m e d i a a n d stretching of the tunica m e d i a (Fig. 1). Scanning electron m i c r o s c o p y c o n f i r m e d complete endothelial d e s q u a m a r i o n o v e r the dilated area with s u b s e q u e n t exp o s u r e of sub-endothelial connective tissue on the luminal surface (Fig. 2), as well as revealing tears of the media. Control vein s e g m e n t s s h o w e d p a t c h y areas of endothelial cell loss o v e r the entire area of the vein surface, but no d a m a g e to t h e d e e p e r layers. Endothelial cell loss in control vein s e g m e n t s w a s m e a s u r e d at 35%. Vein s e g m e n t s that h a d b e e n s e e d e d with endothelial cells s h o w e d the p r e s e n c e of endothelial cells on their luminal surface. S o m e of the cells h a d b e c o m e attached a n d s p r e a d onto the subendothelial tissue, w h e r e a s other cells w e r e in the process of a t t a c h m e n t (Fig. 3).
Prostacyclin release
Fig. 1. Light micrograph of a transverse section through a dilated vein segment. There is a tear in the intima which extends through the media, which is stretched at this point. Elastic van Gieson stain.
Results for basal a n d stimulated PGI2 release by the three g r o u p s of vein s e g m e n t s are p r e s e n t e d in Figures 4 a n d 5. Results are e x p r e s s e d as a m e d i a n w i t h 95% confidence intervals (C.I.). Prostacyclin p r o d u c t i o n is m e a s u r e d in units of p g 6-keto prostaglandin FI~ cm -2 vein m i n -1. Basal PGI2 release w a s higher f r o m control vein s e g m e n t s t h a n f r o m d a m a g e d vein s e g m e n t s , but the difference w a s not significant at a p-value of less t h a n
Fig. 2. Scanning electron micrograph of a distended vein. The endothelial monolayer is absent and a tear is seen extending into the media. Eur J Vasc Surg Vol 6, September 1992
490
M.M. Thompson et al.
Fig. 3. Scanning electron micrograph showing seeded endothelial cells on the surface of a distended vein. The smaller, rounder endothelial cells are in the process of attachment. In the areas not covered by seeded cells, the sub-endothelium is exposed.
700
500
600
400 E
"2 500
E
?
? 300
400 v
o 300
200
G "6 {J
D
200
*6 o
EL
I00 I00
Control
Dilated
Seed
Control
Dileted
Seed
Fig. 4. Basal prostacyctin release from the three experimental vein groups (median _+ 95% C.I.).
Fig. 5. Stimulated prostacyclin release from the three experimental vein groups (median ± 95% C.I.).
0.05 (difference between medians = 48.8, z = -1.68, p = 0.093, 95% C.I. - 4 . 5 to 98.5, Wilcoxon paired rank test). Basal PGI2 release from seeded d a m a g e d vein segments was, however, significantly higher than from d a m a g e d vein (difference between medians = 262, z = -2.52, p = 0.012, 95% C.1.50.5 to 556, Wilcoxon paired rank test). Following thrombin stimulation, PGI2 release
from both control and seeded vein segments was significantly higher than from d a m a g e d segments (difference between medians = 112, 118, z = -2.52, -2.24, p = 0.012, 0.025, 95% C.I. 14 to 193, 19 to 257, respectively, Wilcoxon paired rank test). The results for control and seeded veins have not been directly compared as the endothelial cell identity is different in each case.
Eur J Vasc Surg Vol 6, September 1992
Endothelial Cell Seeding and PGI2 Release
Discussion
Percutaneous transluminal angioplasty (PTA) causes severe damage to the arterial wall, resulting in endothelial desquamation, splitting of the atheromatous plaque, splitting of the tunica media, and exposure of the sub-endothelial connective tissue elements to the bloodstream. 9"1°'2°-27 Similarly, surgical endarterectomy injures the tunica intima and media, which inevitably leads to the formation of a thrombogenic luminal flow surface. 2s-3° In this study we have used an in vitro model of vascular injury produced by hyperbaric distention of long saphenous vein segments. The histological changes thus produced encompass total endothelial desquamation with localised damage to the tunica media. These changes, which are similar to the damage produced to the arterial wall during PTA and endarterectomy, suggest that this may be a useful model for the in vitro study of damaged vascular surfaces. Damage to the endothelial cell monolayer and exposure of the sub-endothelium during PTA encourages platelet deposition 31' 32 which occurs rapidly to a degree that is determined by the extent of vascular damage. 27 Shortly after PTA, the damaged area is covered by a monolayer of platelets, some of which may aggregate and undergo their release reaction liberating vasoactive mediators, smooth muscle mitogens and thromboxane A2.33'34 These changes may contribute towards vasospasm, 35 intravascular thrombosis and intravascular coagulation which may occur at the angioplasty site resulting in acute reocclusion. The enhanced thrombogenicity of angioplasty and endarterectomy sites is due partly to exposure of the highly thrombogenic sub-endothelium, but may also be partly mediated by loss or damage to the endothelial cell monolayer. The endothelium exerts a fine control over the thrombotic process, and in the resting state is anti-thrombotic in action. 36,37 Regulation of the thrombotic process is partly achieved through a balance between prostacyclin (PGI2) production from the endothelial cells and thromboxane A2 production from circulating platelets. Prostacyclin is an arachidonic acid (AA) derivative that is generated by blood vessel microsomes. 3s The ability of blood vessels to synthesise PGI2 is greatest at the intimal surface and decreases progressively towards the adventitia. 39 Endothelial cells have the greatest capacity to produce PGI2, although smooth muscle cells also have synthetic capability. 4° Prostacyclin is a powerful vasodilator, 41 and is the most potent naturally occurring inhibitor of platelet aggregation. 42 Prostacyclin exhibits synergy with EDRF, 43 and
491
together these agents play a central role in the prevention of thrombosis in normal vessels. Several investigators have demonstrated that damage to the arterial wall causes an initial increase in PGI2 production followed by a later fall. 44 The initial increase may be due to increased mobilisation of AA with conversion to PGI2 in the sub-endothelium. However, PGI2 levels fall shortly after injury as the capacity of the sub-endothelium to synthesise PGI2 is expressed only briefly. Studies measuring the luminal PGI2 release in isolation have revealed that angioplasty causes a decrease in the basal and stimulated release of PGI2. 45 Obviously a decrease in local PGI2 production at sites of arterial damage will predispose to thrombosis, and any mechanism that increases local PGI2 production will negate this tendency. In this study, damage to the vessel by balloon dilatation is accompanied by a decrease in basal and thrombin stimulated 46 PGI2 release as compared to the control vessel. An initial release in PGI2 release was not observed. This was probably due to a l h time delay (30 rain sham seeding and 30 min incubation) between damage to the vessel and measurement of PGI2 release. The difference between basal PGI2 release from control and damaged veins was not statistically significant, and this may be attributable to endothelial damage on the control vessels caused during preparation for aorto-coronary bypass grafting. We have noted a 35% endothelial cell loss on control vein segments, and Angelini et al. 47 have shown that veins prepared for use in coronary artery surgery have a lower PGI2 production than nonmanipulated veins. Endothelial cell seeding significantly increased both basal and stimulated PGI2 release from damaged vein segments. There was no significant difference between basal and stimulated PGI2 release from seeded vein segments, in fact the basal release Was higher. This was attributed to centrifugation of the endothelial cells during preparation for seeding. Centrifugation is a powerful stimulus to PGI2 release 46 and would thus negate any further stimulus to PGI2 release by thrombin. In this study it has been shown that endothelial cell seeding of damaged vascular surfaces can increase the diminished PGI2 production that is associated with vascular damage. Reduced PGI2 release encourages platelet aggregation, release of vasoactive mediators, and vasospasm which in turn may be responsible for the 40% acute re-occlusion rate seen after PTA of long femoro-popliteal lesions. 48 Increasing PGI2 production by endothelial cell seeding may reduce the re-occlusion rate as well as having some Eur J VascSurg Vol 6, September1992
492
M.M. Thompson et al.
effect on late restenosis due to myointimal hyperplasia. It has been demonstrated in animal studies that the degree of myointimal hyperplasia caused by arterial damage is less in areas covered by regenerating endothelium 49'5° which may be due to synthesis of heparin proteoglycans which reduce smooth muscle proliferation following vascular injury. 51 Endothelial cell seeding of PTA sites has become possible with the development of specialised arterial catheters, and Nabel et al. 52 have successfully seeded genetically modified endothelial cells on to areas of vascular damage with reasonable cell retention. Endothelial cell seeding of PTA and endarterectomy sites is an attractive method for reducing early and late re-occlusions following vascular reconstruction. This study confirms the potential advantages of endothelial cell seeding to increase PGI2 production following PTA and endarterectomy.
11 12
13
14 15 16 17 18
Acknowledgements
19
M. M. Thompson would like to acknowledge the support of the British Heart Foundation, The Peel Medical Research Trust and the Wellcome Trust for supporting this study.
20
21
References 1 BUDD JS, ALLEN KE, HARTLEY G, BELL PRF. The effect of preformed confluent endothelial cell monolayers on the patency and thrombogenicity of small calibre vascular grafts. Eur J Vasc Surg 1991; 5: 397-405. 2 BURKEL WE, MINTER DW, FORD JW, KAHN RH, GRAHAM LM, STANLEYJC, Sequential studies of healing in endothelial seeded vascular prostheses. JSurg Res 1981; 30: 305-324. 3 ALLEN BT, LONG JA, CLARK RE, et al. Influence of endothelial cell seeding on platelet deposition and patency in small diameter Dacron arterial grafts. J Vasc Surg 1984; 1: 224-233. 4 FASOLR, ZILLAP, DEUTSCHM, GRIMMM, EISCHLEINT, LAUFERG. H u m a n and othelial cell seeding: evaluation of its effectiveness by platelet parameters after one year. J Vasc Surg 1989; 9: 432436. 5 HERRING MB, COMPTON RS, LEGRAND DR, GARDNER AL, MADISON DL, GLOVER JL. Endothelial seeding of polytetrafluoroethylene popliteal bypasses. J Vasc Surg 1987; 6: 114-118. 60RTENWALL P, WADENVIKH, KUTTIJ, RISBERGB. Endothelial cell seeding reduces thrombogenicity of Dacron grafts in humans. J Vasc Surg 1990; 11: 403-410. 70RTENWALL P, WADENVIKH, RISBERGB. Reduced platelet deposition on seeded versus unseeded segments of expanded polytetrafluoroethylene grafts: clinical observations after a 6-month follow up. J Vasc Surg 1989; 10: 374-380. 8 HERRING MB. Endothelial ceil seeding. J Vasc Surg 1991; 13: 731732. 9 BLOCKPC, BAUGHMANKL, PASTERNAKRC, FALLONJT. Transluminal angioplasty: correlation of morphologic and angiographic findings in an experimental model. Circulation 1980; 61: 778785. 10 BLOCKPC, MYLERRK, STERTZERS, FALLONJT. Morphology after Eur J Vasc Surg Vol 6, September 1992
22
23 24 25
26
27 28 29 30 31 32
transluminal angioplasty in h u m a n beings. N EGg l Med 198l; 305: 382-385. HERRING MB, COMPTON R, LEGRAND DR, GARDNER AL. Endothelial cell seeding in the management of vascular thrombosis. Semin Thromb Haemast 1989; 15: 200-204. BUSHHI, JAKUBOWSKI,SENTISSIJM, CULR GR, HAYESJA, DEYKIN D. Neointimat hyperplasia occurring after carotid endarterectomy in a canine model: effect of endothelial cell seeding versus perioperative aspirin, l Vasc Surg 1987; 5: 118-125. MINAR E, EHRINGER H, AHMADI R, DUDCZAK R, LEITHA T, KOPPENSTEINER R, JUNGM, STUMPFLENA. Platelet deposition at angioplasty sites and its relation to restenosis in h u m a n iliac and femoropopliteal arteries. Radiology 1989; i70: 767-772. POPE CF, EZEKOWITZMD, SMITH EO, et al. Detection of platelet deposition at the site of peripheral balloon angioplasty using i n d i u m - I l l platelet scintigraphy. Am J Cardio11985; 55: 495-497. WEKSLERBB, MARCUS AJ, JAFFE EA. Synthesis of prostaglandin 12 (prostacyclin) by cultured h u m a n and bovine endothelial cells. Proc Natl Acad Sci USA 1977; 74: 3922-3926. DUSTING GJ, MACDONALD PS. Prostacyclin and vascular function: implications for hypertension and atherosclerosis. Pharmacol Ther 1990; 48: 323-344. H o c n J , JARRELLBE, SCHNEIDERT, WILLIAMSSK. Endothelialcetl interactions with native surfaces. Ann Vasc Surg 1989; 3: 153159. SCHNEIDER PA, HANSON SR, PRICE TM, HARKER LA. Confluent durable endothelialisation of endarterectomised baboon aorta by early attachment of cultural endothelial cells. J Vasc Surg 1990; 11: 365-372. BUDD JS, BELL PRF, JAMES RFL. Attachment of i n d i u m - I l l labelled endothelial cells to pretreated polytetrafluoroethylene vascular grafts. Br J Surg 1989; 76: 1259-1261. SANBORNTA, FAXONDP, HAUDENSCHILDC, GOTTSMANSB, RYAN TJ. The mechanism of transluminal angioplasty: evidence for formation of aneurysms in experimental atherosclerosis. Circulation 1983; 5: 1136-1140. ZARINS CK, LU C, GEWERTZBL, LYON RT, RUSH DS, GLAGOVS. Arterial disruption and remodelling following balloon dilatation. Surq{ery 1982; 92: 1086-1094. FAXON DP, WEBER VJ, HAUDENSCHILD C, GOTTSMAN SB, McGOVERN WA, RYAN TJ. Acute effects of transluminal angioplasty in three experimental models of atherosclerosis. Arteriosclerosis 1982; 2: 125-133. FALEON JT. Pathology of arterial lesions amenable to percutaneous transluminal angioplasty. AJR 1980; 135: 913-916. BODROG I, MOHACSY J, URAI L. Light and electron microscopic study of pathomorphotogical changes on the arterial wall after transluminal angioplasty. Int Angiol 1986; 5: 13-19. FAXON DP, SANBORN TA, HAUDENSCHILD C, GOTTSMAN SB, McGoVERN WA, RYAN TJ. Restenosis following transluminal angioplasty in experimental atherosclerosis. Arteriosclerosis 1984; 4: 189-195. CASTANEDA-ZUNIGA WR, FORMANEK A, TADAVARTHY M, MLODAVAR Z, EDWARDS JE, ZOLLIKOFER C, AMPLATZ K. The mechanism of balloon angioplasty. Radiology 1980; 135: 564571. STEELE PM, CHESEBRO JH, STANSON AW, et aI. Balloon angioplasty. Natural history of the pathophysiological response to injury in a pig model. Circ Res 1985; 57: 105-112. KINLOUGH-RATHBONE RL, PACKHAM MA, MUSTARD JF. Vessel injury, platelet adherence, and platelet survival. Arteriosclerosis 1983; 3: 529-546. CALLOW AD. Healing of the endarterectomy wound. In: WATSON J, ed. Vascular Diseases. Orlando: Grune & Stratton, 1987; 407-419. STRATTON JR, ZIERLER RE, KAZMERS A. Platelet deposition at carotid endarterectomy sites in humans. Stroke 1987; 18: 722727. CASTANEDA-ZUNIGAER, SIBLEY R, AMPLAZ K. The pathologic basis of angioplasty. Angiology 1985; 35: 195-205. WILENTZJR, SANBORNTA, HAUDENSCHILDCC, VALERICR, RYAN TJ, FAXON DP. Platelet accumulation in experimental angio-
Endothelial Cell Seeding and PGI2 Release
plasty: time course and relation to vascular injury. Circulation 1987; 75: 636-642. 33 ELLIS EF, OLEZ D, ROBERTSLJ, et al. Coronary artery smooth muscle contraction by a substance released by ptatelets: evidence that it is thromboxane A2. Science 1976; 193: 1135. 34 Ross R, GLOMSETJ, KARIYAB, HARKERL. A platelet dependent serum factor that stimulates the proliferation of arterial smooth muscle cells in vitro. Proc Nat Acad Sci USA 1974; 71: 1207-1210. 35 LAM JYT, CHESEBROJH, STEELE PM, BADIMO L, FUSTER V. Is vasospasm related to platelet deposition: relationship in a porcine preparation in vivo. Circulation 1987; 75: 243-248. 36 ESMON CT. The regulation of natural anticoagulant pathways. Science 1987; 235: 1348-1352. 37 NAWORTHP, KISIELW, STERND. The role of endothelium in the homeostatic balance of haemeostasis. Clin Haematol 1985; 14: 531-545. 38 DUSTINGGJ, MACDONALDPS. Prostacyclin and vascular function: implications for hypertension and atherosclerosis. Pharmacol Ther 1990; 48: 323-344. 39 MONCADAS, HERMAN A, HIGGS EA, VANE JR. Differential formation of prostacyclin (PGX or PGI2) by layers of ther arterial wall. An explanation for the anti-thrombotic properties of vascular endothelium. Thromb Res 1977; 11: 323-324. 40 WEKSLERBB, MARCUSAJ, JAFFEEA. Synthesis of prostaglandin I~ (prostacyclin) by cultured human and bovine endothelial cells. Proc Natl Acad Sci USA 1977; 74: 3922-3926. 41 HIGGSGA, CARDINALDC, MONCADAS, VANE JR. Microcirculatory effects of prostacyclin (PGI2) in the hamster cheek pouch. Microvasc Res 1979; 18: 245-254. 42 MONCADAS, PALMERRMJ, HIGGSEA. Prostacyclin and endothelium derived relaxing factor: biologic implications and significance. Thromb Haemost 1987; 25: 597-618. 43 MACDONALDPS, READ MA, DUSTINGGJ. Synergistic inhibition
44 45 46 47
48
49 50
51 52
493
of platelet aggregation by endothelium derived relaxing factor and prostacyclin. Thromb Res 1988; 49: 437-449. BOEYNAEMSJM, GALANDN, KETELBANTP. Prostacyclin production by deendothelialised rabbit aorta. J Clin Invest 1985; 76: 714. MATTSSONE, BRUNKWALLJ, BERGQVISTD. Influence of transluminal angioplasty on the prostanoid release from the arterial wall. EurJ Vasc Surg 1990; 4: 11-17. BAENZIGERNL, FOGrRTYFJ, MERTZLF, CHERNUTALF. Regulation of histamine-mediated prostacyclin synthesis in cultured human vascular endothelial cells. Cell 1981; 24: 915-923. ANGELINI GD, BRECKENRIDGEIM, PSAILA JV, WILLIAMS HM, HENDERSONAH, NEWBYAC. Preparation of human saphenous vein for coronary artery bypass grafting impairs its capacity to produce prostacyclin. Cardiovasc Res 1987; 21: 28-33. JORGENSONB, MEISNERS, HOLSTEINP, TONNESENKH. Early rethrombosis in femoropopliteal occlusions treated with percutaneous transluminal angioplasty. Eur J Vasc Surg 1990; 4: 149152. HANDENSCHILDCC, SCHWARTZSM. Endothelial regeneration II. Restoration of endothelial continuity. Lab Invest 1979; 41: 407418. BJORKERUDS, BONDJERSG. Arterial repair and atherosclerosi after mechanical injury. Part 5. Tissue response after induction of a large superficial transverse injury. Atherosclerosis 1973; 18: 235-255. CASTELLOTJJ, ADDONIZIOML, ROSENBERGR, KARNOVSKYMJ. Cultured endothelial cells produce a heparinlike inhibitor of smooth muscle cell growth. Journal of Cell Bio11981; 90: 372-378. NABELEG, PLAUTZG, BOYCEFM, STANLEYJC, NABELGJ. Recombinant gene expression in vivo within endothelial cells of the arterial wall. Science 1989; 244: 1342-1344.
Accepted 5 May 1992
Eur J Vasc Surg Vol 6, September 1992
Endothelial Cell Seeding of Damaged Native Vascular Surfaces: Prostacyclin Production M. M. Thompson, J. S. Budd, S. L. Eady, K. E. Allen, M. James, R. F. L. James and P. R. F. Bell
Department of Surgery, University of Leicester, Leicester, U.K. Endothelial cell seeding has been successfid in reducing the thrombogenicity of pros&etic vascular grafts in animal and clinical studies. The reduction in thrombogenici~ may be attributed to the intrinsic properties of endothelial cells themselves, and &eir abili~ to produce anti-thrombogenic mediators such as prostacyclin, and endothelium-derived relaxing factor. Endothelial seeding of damaged vascular surfaces produced during percutaneous transluminaI angioplasty and endarterectomy is an attractive possibili~ due to the excellent attachment characteristics of the sub-endothelial tissue exposed during these procedures. The ability of endothelial seeded damaged vascular surfaces to produce prostacyclin was measured in an in vitro model of vascular injury. Endothelial-seeded damaged surfaces produced sigr~_~'cantly higher prostacyclin release than did vessels damaged by balloon dilatation (265.Spgcm- 2 min- 1 and 87.Spgcm- 2 rain- 1 respectively). This study provides evidence that endothelial seeding of damaged native vascular surfaces is technically feasible and that seeding may reduce the thrombogenicity of vascular surfaces following balloon dilatation. Key Words: Endo&eliaI cell seeding; Angioplasty; Prostacyclin; Thrombosis.
Introduction Endothelial cell seeding has been proposed as a method of reducing the thrombogenicity of prosthetic vascular grafts. Endothelial cells possess a wide range of anti-thrombotic and anti-proliferative activities which may reduce graft thrombosis and myointimal hyperplasia in seeded prosthetic grafts. Endothelial cell seeding has successfully reduced the thrombogenicity of grafts in animal models, 1-3 but clinical trials have been largely disappointing, 4-7 probably because of inefficient cell harvest and poor cell retention on the graft surface. 8 Damage to the native endothelial cell surface occurs during angioplasty and endarterectomy. 9-~2 Consequent exposure of the sub-endothelial connective tissue promotes platelet adhesion and activation of the coagulation cascade, 13"14 which may result in thrombosis. Rapid restoration of the endothelial cell monolayer achieved by endothelial cell seeding may decrease the thrombotic tendency due to the formation of a mechanical barrier and also by increasing Please address all correspondence to: M. M. Thompson, Department of Surgery,ClinicalSciencesBuilding,LeicesterRoyalInfirmary, PO Box 65, Leicester,LE2 7LX,U.K. 0950-821X/92/050487+07 $08.00/0© 1992Grune & StrattonLtd.
production of anti-thrombotic mediators such as prostacyclin (PGI2), 15 and endothelium-derived relaxing factor (EDRF). 16 Endothelial cell seeding of damaged native vascular surfaces has advantages over seeding of prosthetic grafts; the damaged native vessel produces an ideal surface for endothelial cell attachment, 17 and the area to be seeded is usually smaller thus requiring fewer endothelial cells for the formation of a confluent monolayer. Seeded endothelial cells have been demonstrated to form a confluent thrombo-resistant monolayer on vascular endarterectomy sites,18 resulting in increased patency and a reduction in the degree of myointimal hyperplasia.12 In this study we examined whether endothelial cell seeding of a damaged vascular surface could increase local release of prostacyclin. Materials and Methods
Development of an in vitro model simulating damaged vascular surfaces Segments of long saphenous vein, obtained following coronary artery bypass grafting were immersed in
488
M.M. Thompson et aL
minimal essential medium (MEM; Flow Laboratories, U.K.) and threaded over a 5-mm balloon angioplasty catheter (Olbert Catheter, Meadox, U.K.). The angioplasty balloon was inflated to 9 atm pressure for 30 s, after which time the vein segment was removed from the catheter, opened longitudinally, pinned on to a silicone gel plate and fixed under tension in 4% paraformaldehyde (BDH, Atherstone, U.K.)/2% gluteraldehyde solution (Sigma Chemicals, U.K.). Segments of vein were prepared for light and scanning electron microscopy (SEM). Segments for light microscopy were dehydrated through graded alcohol, transferred to xylene for 180min and then immobilised in wax over a 4-h period. Serial sections of 4 txm thickness were subsequently stained with haematoxylin and eosin (H&E) and elastic van Gieson (EVG). Vein segments for SEM were dehydrated through graded ethanol followed by acetone solutions prior to carbon dioxide critical point drying and sputter coating with gold. Sections were viewed on a DS 130 scanning electron microscope. Dilated and non-dilated vein segments were independently reported by a pathologist (M.J.), who was blind to the sample type.
dothelial cells to damaged vascular surfaces at different seeding times and seeding densities, we would expect the cell attachment in this experiment to be 2 x 105 endothelial cells cm -2 of damaged vein. Non-seeded control veins, and damaged vein segments were both sham seeded for 30 min in the seeding chamber with complete culture medium prior to PGI2 assay. All vein segments were examined by both light and scanning electron microscopy.
Experimental design Prostacyclin production was measured in three vein groups: control, dilated, and dilated plus seeded veins. Vein segments obtained after aorto-coronary bypass grafting acted as controls. Within each experiment, the control, dilated and seeded veins were all derived from a single vein segment which allowed comparison of paired samples. Eight separate experiments were performed.
Prostacyclin measurement Endothelial cell seeding Endothelial cells were isolated from human umbilical veins and then subsequently grown in tissue culture as has previously been described. 19 In this experiment cells from the second and third passage ~were used. Prior to seeding, endothelial cells were harvested from tissue culture by brief trypsinisation (0.1% trypsin in EDTA; Sigma Chemicals, U.K.) and centrifugation (300 x g for 7min at 4°C), and t h e n resuspended in complete culture medium at a concentration of 1 x 107cells m1-1. Endothelial cell seeding was p e r f o r m e d by placing a 5-mm diameter punch biopsy of dilated saphenous vein on silicone gel in a 5-mm diameter seeding chamber. A 100 ~1 aliquot of the endothelial cell suspension was added to the seeding chamber so that the cells only came into contact with the luminal vein surface. The seeding density was 5 x 106 cells cm -2. After incubation in 95% air 5% CO2 for 30rain, the unattached cells were removed by washing the luminal vein surface three times with MEM. The vein segment was then removed from the seeding chamber and PGI2 release measured. From the results of previous experiments which have measured the attachment of i n d i u m - i l l labelled enEur J Vasc Surg Vol 6, September1992
A 5-mm punch biopsy of the vein segment was placed in a single well of a 24-well tissue culture plate which contained I ml of pre-warmed MEM (pH 7.4). The plate was incubated in a 95% air 5% CO2 atmosphere at 37°C. A sample of the medium (100 txl) was taken from the well 30 min after the vein segment was added, and this was assayed for basal PGI2 release. After 30 min, thrombin was added to the well to obtain a final concentration of lum1-1. A further sample of medium was taken 10min after the addition of thrombin, which was assayed for stimulated PGI2 release. Prostacyclin concentration in the samples was assayed by the measurement of 6-keto prostaglandin FI~, the stable metabolite of PGI2 breakdown at pH 7.4, by the use of a commercially available competitive radioimmunoassay kit (Amersham, U.K.).
Results
Light and electron microscopic examination Light microscopic examination of dilated saphenous vein segments revealed endothelial desquamation, rupture of the internal elastic lamina, tangential and
Endothelial Cell Seeding and PGI 2 Release
489
circumferential tears in the tunica m e d i a a n d stretching of the tunica m e d i a (Fig. 1). Scanning electron m i c r o s c o p y c o n f i r m e d complete endothelial d e s q u a m a r i o n o v e r the dilated area with s u b s e q u e n t exp o s u r e of sub-endothelial connective tissue on the luminal surface (Fig. 2), as well as revealing tears of the media. Control vein s e g m e n t s s h o w e d p a t c h y areas of endothelial cell loss o v e r the entire area of the vein surface, but no d a m a g e to t h e d e e p e r layers. Endothelial cell loss in control vein s e g m e n t s w a s m e a s u r e d at 35%. Vein s e g m e n t s that h a d b e e n s e e d e d with endothelial cells s h o w e d the p r e s e n c e of endothelial cells on their luminal surface. S o m e of the cells h a d b e c o m e attached a n d s p r e a d onto the subendothelial tissue, w h e r e a s other cells w e r e in the process of a t t a c h m e n t (Fig. 3).
Prostacyclin release
Fig. 1. Light micrograph of a transverse section through a dilated vein segment. There is a tear in the intima which extends through the media, which is stretched at this point. Elastic van Gieson stain.
Results for basal a n d stimulated PGI2 release by the three g r o u p s of vein s e g m e n t s are p r e s e n t e d in Figures 4 a n d 5. Results are e x p r e s s e d as a m e d i a n w i t h 95% confidence intervals (C.I.). Prostacyclin p r o d u c t i o n is m e a s u r e d in units of p g 6-keto prostaglandin FI~ cm -2 vein m i n -1. Basal PGI2 release w a s higher f r o m control vein s e g m e n t s t h a n f r o m d a m a g e d vein s e g m e n t s , but the difference w a s not significant at a p-value of less t h a n
Fig. 2. Scanning electron micrograph of a distended vein. The endothelial monolayer is absent and a tear is seen extending into the media. Eur J Vasc Surg Vol 6, September 1992
490
M.M. Thompson et al.
Fig. 3. Scanning electron micrograph showing seeded endothelial cells on the surface of a distended vein. The smaller, rounder endothelial cells are in the process of attachment. In the areas not covered by seeded cells, the sub-endothelium is exposed.
700
500
600
400 E
"2 500
E
?
? 300
400 v
o 300
200
G "6 {J
D
200
*6 o
EL
I00 I00
Control
Dilated
Seed
Control
Dileted
Seed
Fig. 4. Basal prostacyctin release from the three experimental vein groups (median _+ 95% C.I.).
Fig. 5. Stimulated prostacyclin release from the three experimental vein groups (median ± 95% C.I.).
0.05 (difference between medians = 48.8, z = -1.68, p = 0.093, 95% C.I. - 4 . 5 to 98.5, Wilcoxon paired rank test). Basal PGI2 release from seeded d a m a g e d vein segments was, however, significantly higher than from d a m a g e d vein (difference between medians = 262, z = -2.52, p = 0.012, 95% C.1.50.5 to 556, Wilcoxon paired rank test). Following thrombin stimulation, PGI2 release
from both control and seeded vein segments was significantly higher than from d a m a g e d segments (difference between medians = 112, 118, z = -2.52, -2.24, p = 0.012, 0.025, 95% C.I. 14 to 193, 19 to 257, respectively, Wilcoxon paired rank test). The results for control and seeded veins have not been directly compared as the endothelial cell identity is different in each case.
Eur J Vasc Surg Vol 6, September 1992
Endothelial Cell Seeding and PGI2 Release
Discussion
Percutaneous transluminal angioplasty (PTA) causes severe damage to the arterial wall, resulting in endothelial desquamation, splitting of the atheromatous plaque, splitting of the tunica media, and exposure of the sub-endothelial connective tissue elements to the bloodstream. 9"1°'2°-27 Similarly, surgical endarterectomy injures the tunica intima and media, which inevitably leads to the formation of a thrombogenic luminal flow surface. 2s-3° In this study we have used an in vitro model of vascular injury produced by hyperbaric distention of long saphenous vein segments. The histological changes thus produced encompass total endothelial desquamation with localised damage to the tunica media. These changes, which are similar to the damage produced to the arterial wall during PTA and endarterectomy, suggest that this may be a useful model for the in vitro study of damaged vascular surfaces. Damage to the endothelial cell monolayer and exposure of the sub-endothelium during PTA encourages platelet deposition 31' 32 which occurs rapidly to a degree that is determined by the extent of vascular damage. 27 Shortly after PTA, the damaged area is covered by a monolayer of platelets, some of which may aggregate and undergo their release reaction liberating vasoactive mediators, smooth muscle mitogens and thromboxane A2.33'34 These changes may contribute towards vasospasm, 35 intravascular thrombosis and intravascular coagulation which may occur at the angioplasty site resulting in acute reocclusion. The enhanced thrombogenicity of angioplasty and endarterectomy sites is due partly to exposure of the highly thrombogenic sub-endothelium, but may also be partly mediated by loss or damage to the endothelial cell monolayer. The endothelium exerts a fine control over the thrombotic process, and in the resting state is anti-thrombotic in action. 36,37 Regulation of the thrombotic process is partly achieved through a balance between prostacyclin (PGI2) production from the endothelial cells and thromboxane A2 production from circulating platelets. Prostacyclin is an arachidonic acid (AA) derivative that is generated by blood vessel microsomes. 3s The ability of blood vessels to synthesise PGI2 is greatest at the intimal surface and decreases progressively towards the adventitia. 39 Endothelial cells have the greatest capacity to produce PGI2, although smooth muscle cells also have synthetic capability. 4° Prostacyclin is a powerful vasodilator, 41 and is the most potent naturally occurring inhibitor of platelet aggregation. 42 Prostacyclin exhibits synergy with EDRF, 43 and
491
together these agents play a central role in the prevention of thrombosis in normal vessels. Several investigators have demonstrated that damage to the arterial wall causes an initial increase in PGI2 production followed by a later fall. 44 The initial increase may be due to increased mobilisation of AA with conversion to PGI2 in the sub-endothelium. However, PGI2 levels fall shortly after injury as the capacity of the sub-endothelium to synthesise PGI2 is expressed only briefly. Studies measuring the luminal PGI2 release in isolation have revealed that angioplasty causes a decrease in the basal and stimulated release of PGI2. 45 Obviously a decrease in local PGI2 production at sites of arterial damage will predispose to thrombosis, and any mechanism that increases local PGI2 production will negate this tendency. In this study, damage to the vessel by balloon dilatation is accompanied by a decrease in basal and thrombin stimulated 46 PGI2 release as compared to the control vessel. An initial release in PGI2 release was not observed. This was probably due to a l h time delay (30 rain sham seeding and 30 min incubation) between damage to the vessel and measurement of PGI2 release. The difference between basal PGI2 release from control and damaged veins was not statistically significant, and this may be attributable to endothelial damage on the control vessels caused during preparation for aorto-coronary bypass grafting. We have noted a 35% endothelial cell loss on control vein segments, and Angelini et al. 47 have shown that veins prepared for use in coronary artery surgery have a lower PGI2 production than nonmanipulated veins. Endothelial cell seeding significantly increased both basal and stimulated PGI2 release from damaged vein segments. There was no significant difference between basal and stimulated PGI2 release from seeded vein segments, in fact the basal release Was higher. This was attributed to centrifugation of the endothelial cells during preparation for seeding. Centrifugation is a powerful stimulus to PGI2 release 46 and would thus negate any further stimulus to PGI2 release by thrombin. In this study it has been shown that endothelial cell seeding of damaged vascular surfaces can increase the diminished PGI2 production that is associated with vascular damage. Reduced PGI2 release encourages platelet aggregation, release of vasoactive mediators, and vasospasm which in turn may be responsible for the 40% acute re-occlusion rate seen after PTA of long femoro-popliteal lesions. 48 Increasing PGI2 production by endothelial cell seeding may reduce the re-occlusion rate as well as having some Eur J VascSurg Vol 6, September1992
492
M.M. Thompson et al.
effect on late restenosis due to myointimal hyperplasia. It has been demonstrated in animal studies that the degree of myointimal hyperplasia caused by arterial damage is less in areas covered by regenerating endothelium 49'5° which may be due to synthesis of heparin proteoglycans which reduce smooth muscle proliferation following vascular injury. 51 Endothelial cell seeding of PTA sites has become possible with the development of specialised arterial catheters, and Nabel et al. 52 have successfully seeded genetically modified endothelial cells on to areas of vascular damage with reasonable cell retention. Endothelial cell seeding of PTA and endarterectomy sites is an attractive method for reducing early and late re-occlusions following vascular reconstruction. This study confirms the potential advantages of endothelial cell seeding to increase PGI2 production following PTA and endarterectomy.
11 12
13
14 15 16 17 18
Acknowledgements
19
M. M. Thompson would like to acknowledge the support of the British Heart Foundation, The Peel Medical Research Trust and the Wellcome Trust for supporting this study.
20
21
References 1 BUDD JS, ALLEN KE, HARTLEY G, BELL PRF. The effect of preformed confluent endothelial cell monolayers on the patency and thrombogenicity of small calibre vascular grafts. Eur J Vasc Surg 1991; 5: 397-405. 2 BURKEL WE, MINTER DW, FORD JW, KAHN RH, GRAHAM LM, STANLEYJC, Sequential studies of healing in endothelial seeded vascular prostheses. JSurg Res 1981; 30: 305-324. 3 ALLEN BT, LONG JA, CLARK RE, et al. Influence of endothelial cell seeding on platelet deposition and patency in small diameter Dacron arterial grafts. J Vasc Surg 1984; 1: 224-233. 4 FASOLR, ZILLAP, DEUTSCHM, GRIMMM, EISCHLEINT, LAUFERG. H u m a n and othelial cell seeding: evaluation of its effectiveness by platelet parameters after one year. J Vasc Surg 1989; 9: 432436. 5 HERRING MB, COMPTON RS, LEGRAND DR, GARDNER AL, MADISON DL, GLOVER JL. Endothelial seeding of polytetrafluoroethylene popliteal bypasses. J Vasc Surg 1987; 6: 114-118. 60RTENWALL P, WADENVIKH, KUTTIJ, RISBERGB. Endothelial cell seeding reduces thrombogenicity of Dacron grafts in humans. J Vasc Surg 1990; 11: 403-410. 70RTENWALL P, WADENVIKH, RISBERGB. Reduced platelet deposition on seeded versus unseeded segments of expanded polytetrafluoroethylene grafts: clinical observations after a 6-month follow up. J Vasc Surg 1989; 10: 374-380. 8 HERRING MB. Endothelial ceil seeding. J Vasc Surg 1991; 13: 731732. 9 BLOCKPC, BAUGHMANKL, PASTERNAKRC, FALLONJT. Transluminal angioplasty: correlation of morphologic and angiographic findings in an experimental model. Circulation 1980; 61: 778785. 10 BLOCKPC, MYLERRK, STERTZERS, FALLONJT. Morphology after Eur J Vasc Surg Vol 6, September 1992
22
23 24 25
26
27 28 29 30 31 32
transluminal angioplasty in h u m a n beings. N EGg l Med 198l; 305: 382-385. HERRING MB, COMPTON R, LEGRAND DR, GARDNER AL. Endothelial cell seeding in the management of vascular thrombosis. Semin Thromb Haemast 1989; 15: 200-204. BUSHHI, JAKUBOWSKI,SENTISSIJM, CULR GR, HAYESJA, DEYKIN D. Neointimat hyperplasia occurring after carotid endarterectomy in a canine model: effect of endothelial cell seeding versus perioperative aspirin, l Vasc Surg 1987; 5: 118-125. MINAR E, EHRINGER H, AHMADI R, DUDCZAK R, LEITHA T, KOPPENSTEINER R, JUNGM, STUMPFLENA. Platelet deposition at angioplasty sites and its relation to restenosis in h u m a n iliac and femoropopliteal arteries. Radiology 1989; i70: 767-772. POPE CF, EZEKOWITZMD, SMITH EO, et al. Detection of platelet deposition at the site of peripheral balloon angioplasty using i n d i u m - I l l platelet scintigraphy. Am J Cardio11985; 55: 495-497. WEKSLERBB, MARCUS AJ, JAFFE EA. Synthesis of prostaglandin 12 (prostacyclin) by cultured h u m a n and bovine endothelial cells. Proc Natl Acad Sci USA 1977; 74: 3922-3926. DUSTING GJ, MACDONALD PS. Prostacyclin and vascular function: implications for hypertension and atherosclerosis. Pharmacol Ther 1990; 48: 323-344. H o c n J , JARRELLBE, SCHNEIDERT, WILLIAMSSK. Endothelialcetl interactions with native surfaces. Ann Vasc Surg 1989; 3: 153159. SCHNEIDER PA, HANSON SR, PRICE TM, HARKER LA. Confluent durable endothelialisation of endarterectomised baboon aorta by early attachment of cultural endothelial cells. J Vasc Surg 1990; 11: 365-372. BUDD JS, BELL PRF, JAMES RFL. Attachment of i n d i u m - I l l labelled endothelial cells to pretreated polytetrafluoroethylene vascular grafts. Br J Surg 1989; 76: 1259-1261. SANBORNTA, FAXONDP, HAUDENSCHILDC, GOTTSMANSB, RYAN TJ. The mechanism of transluminal angioplasty: evidence for formation of aneurysms in experimental atherosclerosis. Circulation 1983; 5: 1136-1140. ZARINS CK, LU C, GEWERTZBL, LYON RT, RUSH DS, GLAGOVS. Arterial disruption and remodelling following balloon dilatation. Surq{ery 1982; 92: 1086-1094. FAXON DP, WEBER VJ, HAUDENSCHILD C, GOTTSMAN SB, McGOVERN WA, RYAN TJ. Acute effects of transluminal angioplasty in three experimental models of atherosclerosis. Arteriosclerosis 1982; 2: 125-133. FALEON JT. Pathology of arterial lesions amenable to percutaneous transluminal angioplasty. AJR 1980; 135: 913-916. BODROG I, MOHACSY J, URAI L. Light and electron microscopic study of pathomorphotogical changes on the arterial wall after transluminal angioplasty. Int Angiol 1986; 5: 13-19. FAXON DP, SANBORN TA, HAUDENSCHILD C, GOTTSMAN SB, McGoVERN WA, RYAN TJ. Restenosis following transluminal angioplasty in experimental atherosclerosis. Arteriosclerosis 1984; 4: 189-195. CASTANEDA-ZUNIGA WR, FORMANEK A, TADAVARTHY M, MLODAVAR Z, EDWARDS JE, ZOLLIKOFER C, AMPLATZ K. The mechanism of balloon angioplasty. Radiology 1980; 135: 564571. STEELE PM, CHESEBRO JH, STANSON AW, et aI. Balloon angioplasty. Natural history of the pathophysiological response to injury in a pig model. Circ Res 1985; 57: 105-112. KINLOUGH-RATHBONE RL, PACKHAM MA, MUSTARD JF. Vessel injury, platelet adherence, and platelet survival. Arteriosclerosis 1983; 3: 529-546. CALLOW AD. Healing of the endarterectomy wound. In: WATSON J, ed. Vascular Diseases. Orlando: Grune & Stratton, 1987; 407-419. STRATTON JR, ZIERLER RE, KAZMERS A. Platelet deposition at carotid endarterectomy sites in humans. Stroke 1987; 18: 722727. CASTANEDA-ZUNIGAER, SIBLEY R, AMPLAZ K. The pathologic basis of angioplasty. Angiology 1985; 35: 195-205. WILENTZJR, SANBORNTA, HAUDENSCHILDCC, VALERICR, RYAN TJ, FAXON DP. Platelet accumulation in experimental angio-
Endothelial Cell Seeding and PGI2 Release
plasty: time course and relation to vascular injury. Circulation 1987; 75: 636-642. 33 ELLIS EF, OLEZ D, ROBERTSLJ, et al. Coronary artery smooth muscle contraction by a substance released by ptatelets: evidence that it is thromboxane A2. Science 1976; 193: 1135. 34 Ross R, GLOMSETJ, KARIYAB, HARKERL. A platelet dependent serum factor that stimulates the proliferation of arterial smooth muscle cells in vitro. Proc Nat Acad Sci USA 1974; 71: 1207-1210. 35 LAM JYT, CHESEBROJH, STEELE PM, BADIMO L, FUSTER V. Is vasospasm related to platelet deposition: relationship in a porcine preparation in vivo. Circulation 1987; 75: 243-248. 36 ESMON CT. The regulation of natural anticoagulant pathways. Science 1987; 235: 1348-1352. 37 NAWORTHP, KISIELW, STERND. The role of endothelium in the homeostatic balance of haemeostasis. Clin Haematol 1985; 14: 531-545. 38 DUSTINGGJ, MACDONALDPS. Prostacyclin and vascular function: implications for hypertension and atherosclerosis. Pharmacol Ther 1990; 48: 323-344. 39 MONCADAS, HERMAN A, HIGGS EA, VANE JR. Differential formation of prostacyclin (PGX or PGI2) by layers of ther arterial wall. An explanation for the anti-thrombotic properties of vascular endothelium. Thromb Res 1977; 11: 323-324. 40 WEKSLERBB, MARCUSAJ, JAFFEEA. Synthesis of prostaglandin I~ (prostacyclin) by cultured human and bovine endothelial cells. Proc Natl Acad Sci USA 1977; 74: 3922-3926. 41 HIGGSGA, CARDINALDC, MONCADAS, VANE JR. Microcirculatory effects of prostacyclin (PGI2) in the hamster cheek pouch. Microvasc Res 1979; 18: 245-254. 42 MONCADAS, PALMERRMJ, HIGGSEA. Prostacyclin and endothelium derived relaxing factor: biologic implications and significance. Thromb Haemost 1987; 25: 597-618. 43 MACDONALDPS, READ MA, DUSTINGGJ. Synergistic inhibition
44 45 46 47
48
49 50
51 52
493
of platelet aggregation by endothelium derived relaxing factor and prostacyclin. Thromb Res 1988; 49: 437-449. BOEYNAEMSJM, GALANDN, KETELBANTP. Prostacyclin production by deendothelialised rabbit aorta. J Clin Invest 1985; 76: 714. MATTSSONE, BRUNKWALLJ, BERGQVISTD. Influence of transluminal angioplasty on the prostanoid release from the arterial wall. EurJ Vasc Surg 1990; 4: 11-17. BAENZIGERNL, FOGrRTYFJ, MERTZLF, CHERNUTALF. Regulation of histamine-mediated prostacyclin synthesis in cultured human vascular endothelial cells. Cell 1981; 24: 915-923. ANGELINI GD, BRECKENRIDGEIM, PSAILA JV, WILLIAMS HM, HENDERSONAH, NEWBYAC. Preparation of human saphenous vein for coronary artery bypass grafting impairs its capacity to produce prostacyclin. Cardiovasc Res 1987; 21: 28-33. JORGENSONB, MEISNERS, HOLSTEINP, TONNESENKH. Early rethrombosis in femoropopliteal occlusions treated with percutaneous transluminal angioplasty. Eur J Vasc Surg 1990; 4: 149152. HANDENSCHILDCC, SCHWARTZSM. Endothelial regeneration II. Restoration of endothelial continuity. Lab Invest 1979; 41: 407418. BJORKERUDS, BONDJERSG. Arterial repair and atherosclerosi after mechanical injury. Part 5. Tissue response after induction of a large superficial transverse injury. Atherosclerosis 1973; 18: 235-255. CASTELLOTJJ, ADDONIZIOML, ROSENBERGR, KARNOVSKYMJ. Cultured endothelial cells produce a heparinlike inhibitor of smooth muscle cell growth. Journal of Cell Bio11981; 90: 372-378. NABELEG, PLAUTZG, BOYCEFM, STANLEYJC, NABELGJ. Recombinant gene expression in vivo within endothelial cells of the arterial wall. Science 1989; 244: 1342-1344.
Accepted 5 May 1992
Eur J Vasc Surg Vol 6, September 1992
