418

)ACC V .1 19. No 2 1`6rusry 199211.-12

REVIEWS Experimental Models of Coronary Artery Restenosis DAVID W . M . MULLER, MBBS, FRACP, STEPHEN G . ELLIS, MD, FACC . ERIC 7 . TOPOL, MD . FACC Ann Arhor . AM higtio

The study of potentially effective drug therapies and mechanical devices for the prevention of restenosis after percutaneous coronary revascutariration teas relied heavily on the use of experimental animal models . To date, >50 experimental studies have been reported and have suggested that at least nine different classes of pharmacologic agents inhibit the intimal proliferative response to arterial injury. However, no pharmacologic intervention has yet

stimulus to intimal proliferation and several practical and logistic . Finally, the reported efflcaeles of specific drug considerations therapies in the respective animal models and in humans were compared . This review suggests that significant interspecies and occasionally intraspecies differences do exist among the respective animal

been shown t o reprodncthly reduce the incidence of r tenosis

timal thickening. These differences appear to reflect not only differences in the nature and extent of the arterial injury, but also intrinsic differences among the species in drug and lipid metabo .

after coronary balloon angioplasty in humans . To Identify the reasons for the apparent nonspecificity of the

animal models and to determine which model should most reliably predict the efficacy of individual therapies in humans, the distinguishing characteristics of the experimental models were com-

pared. Particular attention was paid to the size and morphologic structure of the treated artery, the susceptibility of the species to spontaneous and diet-induced arterial disease, the nature of the

Over the past decade, mechanical means of achieving percutaneous coronary artery revascularization have been greatly refined . The majority of atheromatous coronary stenoses can now be successfully treated by balloon dilation, excisional atherectomy . endoluminal stenting or Laser ablation (1-4). The long-term efficacy of each of these procedures remains limited, however, by recurrent stenosis at the site of intervention, which occurs in approximately 257 to 357 of treated stenoses (2-4.5) . This incidence of restenosis has proved to be refiactory to a large number of pharmacologic and mechanical interventions (6) and the histopathology of the recurrent lesion appears to be independent of the nature of the injury inflicted on the arterial wall (7,8) . As the number of percutaneous coronary interventions increases, the fiscal burden of recurrent corollary stenoses will climb . In 1992, the number of coronary angioplasty procedures performed in the United States is expected to be > 100,000.

Even a moderate reduction in the incidence of restenosis could he expected to have a major impact on the cost of these procedures. It has been estimated, for example . that a From the Division

of

Cardiology . DIP--

of In-al

Medicine .

University of Michigan Medical Center . An, Arbor . htiehlpn . Manuscript received April 15, 1991, revirod manuscript received lent 27 . 1991 .

accepted

July

17,

1991 .

Address far reorinls . David W. hl . Molten, MHny . Division of Cardiology . University of Michigan Medical Center, 9I F245, 151x7 East Siedical Center Drive, Ann Arbor, hlichigun 41109-0021--, V .1992 by

the American Collcac of Cardioloay

models, particularly in the extent and composition of the neoin-

lism and in the activity of the coagulation and fibrinnlylic systems, These differences may account for the variability in sensitivity of

these models to pharmacologic interventions and should be con . sidered carefully in the interpretation of experimental studies of new devices and of apparently effective pharmacologic therapies . (J Am Coll Cordial 1992 ;19 ;418-32)

reduction in the incidence of restenosis from 33% to 257, could result in annual savings of as much as $450 million 16) . It follows that a considerable research effort into the palhophysiology of and preventive measures against restenosis is urgently needed . However, clinical investigation of this entity has been impeded by several important limitations . First, until recently, the inaccessibility of human coronary arteries has limited the potential for antemortem histopathalogic analysis of postaetgioplasty recurrent stenoses. Now the excisional atherectomy catheter may provide valuable data regarding the cellular composition and immunocytochemical characteristics of the proliferative tissue at the site of, rior interventions (7,9 .10). A second limitation has Lecn the need for and cost of multicenter clinical trials of pharmacologic therapies in large numbers of patients . Analysis of the published findings of controlled clinical trials of treatment of restenosis suggests that the vast majority of these studies have lacked sufficient power to detect important differences among potentially valuable therapeutic agents (11) . A third limitation has been the need for complete angiographic follow-up in the study groups and a lack of consensus among investigators on the definitions of clinical and angiographic restenosis(6) . These considerations highlight the need for a reliable experimental model for investigating the pathophysioiogy of restenosis and evaluating pharmacologic and mechanical means of preventing its occurrence . The principal aim of this 0735-1097,91S5.00



2 February me'.als-)2 IACC Vnl. 19 . 4u.

MULLER ET AL . ANIMAL MODELS 'IF REStENOSIS

419

Figure I. Histolegic section of a human coronary

artery showing restenosis due to immal thickening 12 months after coronary balloon angiopla,ly . Note the clearly visible dissection nap

review, t herefore . i s to examine the animal models that hate been employed for this purpose and to consider whether any of these can be expected to reliably predict the impact of a given preventive measure on the incidence of restenosis in clinical trials . This will be approached b yy considering I I the

pathogenesis of human restenosis. 21 the distinguishing characteristics of each animal mode(, and 3) the reported etftcacies of specific drug therapies in the respective animal models and humans . Pathophysiology of Human Restenosis The predominant mechanism by which balloon angioplasty relieves obstructive coronary stenosis appears to be by fracture through the atheromatous plaque . often with extension of a dissection plane into the muscular medial layer(12-21) . A second mechanism may be stretching of less diseased segments of the arterial wall adjacent to an eccentric, relatively nonelastic atheromatous plaque 122) . the predominant mechanism for restenosis after balloon angioplasty is the formation of a fibrocellular neointima at the site of dilation . Postmortem studies (23 .24) in patients who died several months after apparently successful coronary angioplasty clearly showed evidence of previous fracturing of the atheromatous plaque and medial dissection, with extensive neointimal proliferative tissue filling the crevices . overlying the plaque and abstracting the coronary lumen. IFig . 1) . In the latter investigation (24) . immunoperoxidase studies showed no evidence of deposition of apolipoproteins in the neointima . suggesting that lipid deposition is not a significant pathogenic factor . In some patients . elastic recoil of overstretched, relatively disease-free segments of arterial wall may also cause recurrent stenosis without intimal proliferation (25,26). The factors that determine udhnther one nrtn.fied intitnat praiiierarian and restenosis occur after balloon angioplasty .

stenting . laser ablation or atherectomy remain poorly understood . It is generally believed that exposure of subintimal components of the vessel wall results in the deposition of platelets . thrombus formation and the release of growth factors from platelets . circulating monocytes, endothelium and . perhaps, smooth muscle cells themselves (5) . These mitogvns stimulate the transformation of medial smooth muscle cells from a resting contractile state to a synthetic phenotype and promote the proiiferation of these cells and their migration from the media to the intima (27,28) . Smooth muscle cell migration and proliferation are followed by the synthesis and secretion of extracellular connective tissue matrix . resulting in progressive lumen narrow,ag and obstruction to coronary flow .

Restenosis and Atherogenesis This sequence of events from endothelial cell and medial injury to the formation of a fibrocellular neointima is believed to he very similar to that which occurs over a much greater time frame during the development of atherosclerotic plaques . The "response to injury" hypothesis for the development of spontaneous atherosclerosis was first proposed by Virchow (29) and more recently modified by Ross (30) . Considerable data accumulated largely from animal models (31-35) suggest that chronic dietary hypercholesterolemia is associated with progressive endothelial cell injury and the eventual formation of atheromatous lesions . Although systematic evaluation of the genesis of atherosclerotic plaques in humans is more difficult . these findings are consistent with the observations of several studies (36,37) that have shown that fatty streaks develop very early in normal children at sites that correspond to the location of atheromatous plaques in the adult arterial tree . Unifying hypothesis of vascular injury . The preceding observations suggest that early atherogenesis and restenosis



420

Mi'I .I .ER I'r AL. ANIMAL MOD[ I S Of' RLS11 N-IS

JACC Vol, 19, N,, 2

after balloon angioplasty may have a similar pathogenic

and virtually acellular . A similar intimal structure is only

mechanism . A unifying t ypothesls was recently summarized

present in the larger arteries of humans during fetal and early

by Ip et al . (38).

Three grades of vascular injury were

neonatal life 146), Thereafter . the space becomes progres-

described . Type I injury result,, in no overt disruption of the

sively thicker because of infiltration of smooth muscle cells

endothelium . hut rather a more subtle functional change in

and the development of a musculoclastic layer consisting of

the integrity of the endothelial cells, Examples ofthis type of

elastic tissue . collagen and scattered smooth muscle cells .

injury include chronic hypercholesicrolemia (39) and injury

These age-related changes are also observed in the larger

induced by viral infection (4(1,41). Type 2 vascular injury .

mammals such as the nonhuman primates and swine .

hit[ no disruption of the internal elastic lamina or the

se-ad A rmwidrraoml is the soseepiihility of the selrcled speties to spontunrrrer or rliel-irtdured atherout lero-

muscular media . may occur during the early stages of

ss that morphologically and topographically resembles hu-

formation of atheromatous plaques, during balloon arterial

man atherosclerosis. This characteristic not only permits the

injury, in homocystioemia (42) and by immunologically

evaluation of pharmacologic and mechanical therapies ir :

mediated mechanisms (43-45) . Type 3 injury, which disrupts

diseased arteries . but also suggests that the intimal response

not only the endothelium but also the internal elastic lamina

to injury and its susceptibility to interventions may mimic

and media, is the predominant injury induced by balloon

the response seen in humans . Additional factors that may be

angioplasty, coronary alhereclomy, laser ablation and endo-

relevant to the study of restenosis include whether the artery

vascular sterling. This type of injury may also occur spon-

is or is not diseased at the time of injury. the nature of the

taneously in complicated atherosclerotic plaques and as a

underlying disease process (for example . foam cell infiltra-

consequence of increasing age (46) . Under this classification

lion vs . fibrocellutar thickening), the age (and perhaps gen-

of vascular injury . coronary artery restenosis along with

der) of the animal and the presence or absence of concomi-

coronary vein graft and cardiac allograft atherosclerosis can

tant hypercholesterolemia .

which results in endmhelial denudation and intimal damage

he considered to represent a much accelerated form of the

Arterial injury . Models employed to date have used a

same pathogenic process that results in spontaneous ather-

variety of endarterial injuries to induce intimal proliferation .

osclerosis .

These injuries may vary in the extent to which they stimulate a proliferative response . Factors to be considered in comparing the models include the severity of the injury (in particular. whether the internal clastic lamina is disrupted

Characteristics of Animal Models and the extent of injury to the underlying media) . the platelet of Restenosis response and formation of thrombus at the site of injury and Several factors raved to he considered in comparing the

rheologic factors including the presence of adjacent arterial

respective animal models . These include factors relating in

stenoses . branch points . changes in vasomotor tone and

1) the arterial substrate itself : 2) the nature of the arterial

blood flow disturbances .

injury : and 31 several practical and logistic considerations .

Practical considerations . These include the size of the

Arterial substrate, Ideally . the arterial bed of the animal

animal and the selected artery, the availability of ready

species chosen should have a size and structure that are

access to the arterial tree, .he time course of the develop-

histologically and biochemically similar to those of human

ment of intimal proliferation- the cost and procurabifity of

coronary arteries . The basic morphologic structure of mam-

the animal, the cost of its mousing and feeding, potential

malian arteries appears to be very similar in all species

difficulties in its handling, the robustness of the animal

examined . In elastic arteries such as the thoracic and ab-

species and its potential for disease transmission . The need

dominal aortas and the carotid and iliac arteries . a repealing

for fluoroscopic imaging is also a limiting factor in some

medial lamellar unit. originally described by Wolinsky and

models.

Giagov (47) . is present in all major vertebrate classes with only minor variations 148) . The number of lamellar units in the media of the adult mammalian aorta is proportional to the

radios

of the artery . and the

Specific Animal Models Several animal species have beam used to inver.tigate the

. regardless of age and per1^m!lais rkbyconta species 14711 . In

pathogenesis of atherosclerotic disease 150-521 . Many of

contrast, the amount of elastin in the media of coronary

these models are also useful for the study of restenosis . In

arteries. which are by definition muscular arteries . varies

contrast . avian models of atherosclerosis are not practical,

with the size of the animal . The elastin content of the dog.

principally because of the size and shape of the avian thorax .

pig and baboon coronary artery is very similar to that of the

but also because of differences in cardiopulmonary dynamics

human coronary artery and greater than apparent in smaller

and widespread involvement of small intramural coronary

species including rodents and fowl 149) . The thickness of the

arteries an the disease process (50) . For these reasons . avian

arterial intima also varies among species . In smaller animals

models will not he further considered and discussion will be

including rats and rabbits, the subendothelial space between

confined to mammalian species and consideration of in vitro

the endothelium and internal elastic lamina is very narrow

studies by cell culture .



MIJLLPR I:T Al ANIMAL M1IOUGLS OF aLS'ILNOSIS

JACC Vui . 19. N, . 2 Fmnaary Iwn41e-u+

421

SpecTable I . Characteristic, u( Animal L,1' Speaes Nut

57 .w

0-I ut

hcrhha

sri

ire Olsl

' 4

-'

\Saela

AJuh

itgi

Perch- Price' 151

n'-11 v --- 11_-211

Per D- Cu.I' i51 115_11 .75

itabbil Noes

/cdaeJ

r,-1I

5-h

WHIR,

f+-N

oar

It 7S

Pumarei0he .ua

Ill ID

J-J v

17U '_511

Lib

109

3'-595

2.300-3 .I9K1

4!94-5.(4

_nil-40,

I'll

0.95

'n-Jn

420

6.45

1_J I!

!4-t5

.15 1

S cu

Unlnr,hc Pig I' :< Minlpig it 'lneluJmpl¢,ehl :h,1g,, . he l,,hleheped,paemii ,hh,l

, -s ~munI.

' le g+dSL .hcs. .hmpesmnrmuiipidamlcdlcIl.WHHL-Wulanah,

r

Rats Response to arterial injury. Although )me inbred sirmns are susceptible to spontaneous or diet induced arterial disease (53 .54) . rats are generally considered to he relatively resistant to alherogenesis (51 .55) . Moreover. rat arteries differ morphologically from human arteries in that the} have no vasa vasorum . have a very much thinner suhintimal layer and have a relatively small elastin content in the media (49.551 . Despite this, the response of the rat arterial a all to injury is one of the most thoroughly investigated of any animal species . The principal advantages of the rat model are that the animals are readily available . easy to handle and inexpensive to purchase and maintain (Table I) and they develop a very predictable and rapid proliferative response to severai !'arms of endarterial injury inctudit :g air drying and balloon de,wdation (56-63) . Interestingly, endothelial denudation using a fire n,,Icn filament that does not damage the arterial media induces only a minor smooth muscle cell proliferative response (64 .65) . The use of air drying to cause extensive endothelial damage in the rat carotid artery was first described by Fishman et al . (56) . Although this model does not mechanically deform the arterial wall and does not cause any overt damage to the underlying medial smooth muscle, a dense layer of platelets rapidly forms over the denuded segment and intimal thickening due to smooth muscle cell proliferation occurs within 14 day's to a thickness of approximately three times that of the media (56 .571 . At later stages, a progressive decrease in thickness of the neoiulima occurs as a result of a decrease in the number of cells present and an increase in the density of the associated connective tissue stroma (561. The rat carotid artery balloon injury model . Balloon dilation and endothelial denudation of the rat carotid artery (Fig . 2A) . a modification of the aortic balloon injury model of Baume_artper (58), has been systematically evaluated by several investigators (59-63) . The passage of a Fogarty catheter along the length of a rat carotid artery is followed by a well defined series of events similar to that described in the model of Fishman et a1 . (56). Endothelial denudation is promptly followed by platelet adhesion and endothelial and

smooth muscle cell proliferation (59). The majority of the proliferating smooth muscle cells enter the growth cycle hciueen 24 and 72 b after injury (59) and begin to migrate from the media into the intima . Approximately 5067, of the migrating cells do not divide but nonetheless contribute to the increasing cellularity and thickness of the intima (60) . Neointirn al thickness peaks approximately 3 weeks after injury and remains relatively constant thereafter despite continuing cell proliferation . Several studies (59.61) have suggested a critical relation between the intact endothelium and medial smooth muscle cell proliferation . Regrowth of the rat endothelium ceases after approximately 6 weeks and may leave a large central area devoid of endothelium . Proliferation of the smooth muscle cells underlying these denuded sites may continue at rates of up to 50 times greater than control rates for periods of >12 months (62) . Interestingly . recent studies (66) have shown that rat carotid endoIheliai cells that have ceased replicating after balloon injury can he stimulated to recommence replication and cover the remaining denuded surface by the systemic administration of basic fibroblast growth factor . a potent endothelial mitogen and angiogcnic factor . The rat carotid artery model has also been used to evaluate the influence of a variety of physiologic changes on the extent of inamal proliferation after balloon endothelial denudation . For example . hypophysectomy 2 weeks before balloon denudation of the rat aorta has been shown (67) to completely inhibit the smooth muscle cell proliferative response without interfering with regrowth of the endothelium . In vitro studies Of have suggested that this effect is mediated through the influence of pituitary growth hormone on the release of somatomedin C . A second physiologic variable examined in the rat model is the influence of age on endothelial and smooth muscle cell proliferation . The rate of normal cell replication appears to be highest in the neonatal period (69) and in cell culture, immature cells proliferate in response to a stimulus sooner than do cells from older animals (70) . In vivo studies (71) suggest that although the smooth muscle cell proliferative response to endothelial denudation is indeed delayed in older animals . the response

Figure 2. Inumal prom.teratinn after balloon injury of a tai connid arery . Reprinted tom Yowell a al . (102) with per of the American Associavnn in, the Advanrea,ent of Sc,ence . B, futfmal prolife anion after hafkvnt argioplasty of a nlenoscd iliac artery in a hype reholeseerulemie rabbit . Note the p-evinas inrirral disseamn and marked eeointimal thickening due to treat cell infiltration. Reptrduced courtesy net Dr. David P. Lacer . C, Intimnl lhieketdng iIer balloon dilation of a

A

normal porcine carotid artery . The neeittima is eccentric and maximal at the site of dismprior of bush die imeraal and external elasnc laminar . Flit hbrocellular ncoiatima contains a large number of Iietnwidcuu •I adrn macrophages. D, Marked coronary sternums due m neoiotimal thickening 4 weeks lilac implanlaruen of en ovcrsirxd melodic stmt in a normal porcine coronary artery . Moslem wires penetrated both the art-al and external elastic laminae,



JAi l Vnl . Iv . Nu . . Icbni.us Ivv'_ .Jln-~:

in these animals may he more extensive and more suvnincd, Thus . the age of animals in restcnosis studies may hase an Important influence on the severity of the proliter :mre response . Rabbits On a normal vegetarian diet . most strains of rabbit do not develop spontaneous athernsclcrevs, but may drcJon d generative lesions in the arterial media that resemble blonckeherg's medial sclerosis (721 . In comrast to the rat . h--r, the rabbit appears to he highly su-ccptihle to the dcs'clopmem of intimal lesions when exposed to extremely high Itveb of serum cholesterol !51 ) . Sustained levels of chulesternl2 .000 mgJdl can be readily achieved by diet alone . This extreme hypercholesterolcmia mar result from an inherently poor eitabolic system for cholesterol 173) ;aid i s associated with widespread cholesterol deposition in the reticuloendothclial system . causing hepatic . splenic and lymph node enlargement, features trial are more typical of a lipid storage disease than atherosclerosis, Lesions that develop in the intima of the aorta and great arteries arc composed predominantly of lipid-laden macrophages (foam cells) and rarely display the features 01`163-d u!hcrosdl:rolic plaque, such as calcification. central necrosis . Ir .ation and thrombus formation . More typical atherosclerotic plaques occur in animals fed an intermittent hypercholesterolemic diet (50), in those fed a moderately hyperlipidemic diet after initial arterial injury 04 .75) and in certain breeds of rabbit that lack membrane receptors for low density lipuprotein (homozygous Watanabe heritable hyperlipidemic rabbit) (76) . Balloon denudation of aortic iliac arteries . 'I he use of the rabbit as an experimental model for balloon angioplasty was first described by Black el al . (77). and the long-term effects of balloon dilation in the atherosclerotic rabbit were later described by Faxon et al . (78) . In these studies . stenotic lesions consisting principally of concentric foam cell infiltration, of both the intima and media were initially induced in the aorta and iliac arteries by balloon endothelial denudation followed by a 6- to 12-week period of feeding with a 2%-cholesterol diet . Dilation of these lesions resulted in rupture of the neoinlima with platelet deposition and fibrin formation (77) and subsequent restenosis at the site of balloon angioplasty (78) . Histologic evaluation 4 weeks after angioplasty showed mar . ed concentric reduction in the arterial lumen as a resu t of the formation of a loose . relatively acellular . lipid-r ch . edematous connective tissue layer that filled the spaces created by rupture of the intima (Fig. 2B) . Organized ihrr mhus was apparent in approximately half of the animals The major advantages if the atherosclerotic rabbit iliac artery model are that the animals are easily handled and inexpensive, breed readily and reliably develop restenosis (Table I). In addition, unlit e the rat carotid artery and aorta, the adult rabbit iliac artery is approximately the same size as

NILt.LEO ET AL . ANIMAL Mtoni'LS OP R6s'rENOSIS

4 23

the human coronary artery. The model may be useful . therefore lit the evaluation of new percutaneous revascularit.dron devices (79-8 2_) . However, the major difference in consistency and composition of the obstructive material in ire hypcrchole ,a ,, ro!eroic rabbit front that of human atiterline .dgesn s that direct extrapolation of the efficacy of drug therapies and devices in this model should be undertaken with amuun . Thus. the principal disadvantage off he hyperorrlestcrolemie r bbit model is the prominent involvement of lipid not only in the initial plaque . but also in the restenosi, lesion . In addition, rabbit iliac arteries are highly ra,oretmtisc and have multiple side branches . An alternative model that produces iliac stenoses with a consistency more equivalent to that of human atheroma uses a chronically indwelling arterial catheter (74) . Whether this or the homorypous Watanabe heritable hyperlipidemic rabbit will provide better models of restcnosis remains to be evaluated . Balloon dilation of the normal rabbit iliac artery also results it a considerable smooth muscle cell proliferative response . Although this model has not been widely used to evaluate pharmacologic means of preventing restenosis, it has been used to examine the physiologic effects of balloon dilation . such as the relation berocen intimal thickening and chronic impairment of endothelial-dependent arterial relaxation (83 .84) . Day's Like all natural carnivores, the dog is very resistant to dic .iodoced atherosclerosis . In fact, arterial lesions rarely develop unless the animal is made simultaneously hypolhyreal by t1 yruidectomy or the administration of anlithyroid drug therapy (51,85( . The vessels that are first affected in thyroid- .suppressed hyperchotesterolemic dogs are predominantly small arteries including thee thyroid, iliac, intercostal and carotid arteries . rather than the larger vessels more commonly involved in humans . Coronary arteries become diseased late in the process. A second important difference from human atherosclerosis is the predominance of medial rather than intimal thickening and lipid accumulation . An additional factor that may confound the interpretation of drug studies in dogs is the considerably greater activity of the fibrinoiytic system in these animals . Circulating plasmin is often detectable in dogs (but not in humans) and fibrinogen levels may fluctuate widely in the postoperative period (86) . The activity of the fibrinolytic system at rest and in response to injury may be of great relevance because clinical studies (87) have suggested that fibrinolytic capacity may be impaired in patients who develop restenosis after coronary balloon angioplasty . Extent of intimal proliferation after arterial injury in the canine model . Despite these limitations, dogs have been commonly used as experimental models, principally because of the similarity in size between the coronary arteries of adult dogs and those of humans, the ready availability and low cost of procuring and keeping the animals and the ease



4 24

tACC not. 19 . No. 2 Felvwry 1 .2a1N st

IUL.t .t a i.i AI .. ANIMAL M0 1*15OF kxvtksiisls

of vascular access (Table 1) . Relatively few data are available on the extent of intimal proliferation after arterial injury in the canine model . In a recent study of balloon angioplasty of normal canine renal arteries, Orlandi et al . (88) observed that mural deoxyribonucleic acid (DNA) synthesis increased in the balloon . diluted segments 5 days after injury and persisted for c,2 weeks . This study and others (89,90) noted only a mild intimal proliferative response sat prominent proliferation of the vasa vasorum and the adventitia . The latter finding has. also been observed in association with human atherosclerotic diseaw (91), but its role in the pathogenesis of atherosclerosis and restenosis is al present unclear, Interestingly • moderate intimal proliferation (seen in association with disruption of the internal elastic lamina) has been reported (92) after external freezing of normal canine arteries, an observation in keeping with the concept that smooth muscle cell proliferation and intimal thickening are a nonspecific response to arterial wall injury . Evaluation of interventional devices. The most common use for the canine model has been in the evaluation of interventional devices including coronary scents, laser catheters and atherectomv devices (93-96) . The insertion of an appropriately sired cndovascular metallic stent results in the formation of a fine layer of thrombus . followed by rapid growth of a layer of endothelial and smooth muscle cells that cover the stent struts (93,941 . The neointima reaches a maximal thickness of 200 to 300 lam approximately 8 weeks after stent insertion and thereafter becomes progressively thinner over a period of 6 to 9 months (93) . E.milar changes have been observed after the implantation of porous vascular prostheses (86) . In contrast to the thick neointima that develops in the pig . the surface of the prosthesis becomes covered with a thin smooth layer of neointima within 8 to 12 weeks of implantation in the dog . These findings suggest that the canine arterial wall may be relatively nonrcactive to acute or chronic injury and, for this reason, extrapolation of the results of interventions in normal canine coronary arteries to diseased human arteries may be particularly difficult . Clearly . because of the difficulties in inducing atherosclerotic disease, the dog is not a good model for the evaluation of interventional devices in diseased arteries . One means of overcoming this limitation may be the use of an interposition graft of h •aman atherosclerotic coronary or peripheral artery anastomosed proximally and distally to a normal canine femoral artery (97) .

Nonhtrntmt Primates Spontaneous atherosclerosis has been reported in several species of nonhuman primate and in general the sequence of events and the topography of the lesions are similar to those observed in humans (50-52) . However, a broad spectrum of species-dependent changes has been described . The most susceptible species is the rhesus monkey (Mr .r a ca nwiarra), in which atherosclerotic plaques develop in the aorta and carotid, iliac and coronary arteries, both spontaneously over

a period of several years and in association with a hypercholesterolemic diet (98) . The changes seen resemble the features of advanced human atheroma including central medial necrosis, ulceration . calcification and thrombosis . In some species, atherosclerotic gangrene and fatal myocardial infarction have also been recorded (50) . Relatively little has been published. however, on the intimal proliferative response to arterial injury in nonhuman primates . As might be expected from the spectrum of atherosclerotic lesions seen, it does appear that the nature and extent of the proliferative response are both species and diet dependent . Balloon injury, for example, stimulates a predominantly fibromusculor intimal lesion in some species, particularly when performed in conjunction with a diet supplemented with peanut oil 1521 . In contrast, other species, particularly when fed a high fat • cholesterol-rich diet . may develop concentric . transmural, lipid-rich proliferative lesions (52) . Compounding this intraspecies difference in response to injury are the practical limitations of nonhuman primates as models of human atherosclerosis or restenosis (Table I ) . Not only are the animals expensive to obtain and maintain, but also they are highly susceptible to disease and death in captivity and are uncooperative and difficult to handle .

Swine Spontaneous and diet-induced atherosclerosis . Of all the animal species systematically examined to date, the pig is most similar to humans in its cardiovascular morphology and physiology and susceptibility to atherosclerosis. The coronary arteries of the adult pig approximate the size of human coronary arteries and have a very similar morphologic structure (99) . The animals appear to be susceptible to both spontaneous and diet-induced atherosclerosis- Pigs are omnivorous and readily consume a human diet . In pigs on a free range, unsuppiemented diet, atherosclerotic lesions develop within 4 to 8 years. The distribution and morphology of these lesions closely resemble those observed in humans, as does the sequence of events from lipid deposition and foam cell formation to fibrocellular prolif:•ration and calcification (50,51,1001. Although both cerebral and myocardial infarction have been observed in aged swine, the features of complicated atheroma including ulceration, thrombr its and plaque hemorrhage are uncommon in spontaneous porcine atherosclerosis (100) . The development of widespread atheromatous disease can be accelerated in domestic swine by dietary manipulation . In contrast to nonhuman primates, pigs fed a diet supplemented with a high content of saturated fats quickly develop changes in the serum lipid profile that mimic the changes seen in humans . These include markedly elevated serum cholesterol, elevated low density lipoproteins, a reversal in the ratio of high to low density lipoproteins (from >I to 400 kg (Table I) . The cost of maintaining such animals and the logistic difficulties involved in transporting and handling them make them unsuitable for long-term investigations . This problem has beer: largely overcome by using selectively bred strains of miniature ,wine . some vanelies of which reach a maximal weight of only 20 m 40 k 1102-106) . In adulthood. these animals develop atherosclerotic lesions that are indistinguishable from those of larger domestic pigs and very similar to the complex lesions typical of advanced human atherosclerosi- . The disease process can be considerably accelerated in selected arteries by using a combination of mechanical injury and diet-induced 'iypercholesterolemia (104,107). Although these special breeds of pig are more expensive than domestic swine (and are ne'- as readily available), the costs of feeding the latter increase progressively so that the cumulative costs of acquiring and maintaining a farm pig may be greater than that for miniature swine for investigations that last >2 or 3 months (107) . Balloon injury of the normal porcine artery . The events that follow balloon angioplasty of the normal porcine artery were investigated by Steele et aL (108) . Dilation of the carotid artery using an oversized balloon fballooniartery ratio approximately 1 .5 :1) resulted in complete endothelial denudation, marked platelet deposition and thrombus formation, dissection through the internal elastic lamina into the arterial media and necrosis of medial smooth muscle cells . Intimal thickening due to smooth muscle cell proliferation was apparent 7 days after injury and reached a mean maximal thickness of 85 pet at 14 days . Other studies (1091 using the same model have shown that arterial balloon dilation is a considerably stronger stimulus to smooth muscle cell proliferation than is endothelial denudation alone despite a similar degree of platelet deposition . Our own observations using this model suggest a critical role for disruption of the internal elastic lamina in the stimulation of intimal thickening . The lesions that occur after balloon injury of the porcine carotid artery are commonly eccentric and overlie breaches in the internal elastic lamina (Fig . 2C( . The remainder of the circumference n ; the arterial lumen, although exposed to the same pressure and degree of stretch, may be free of any apparent proliferative response . In this model, thrombus may play a more prominent role in the competition of the neointima than in some other models . This is, however . consistent with recent observations from histologic analysis of tissue retrieved by coronary atherectomy (110) or after stem implantation (28), which suggest that thrombus may play a more important role in the pathophysiology of human restenosis than had been previously appreciated.

MULLER ET AL . ANIMAL MOnFLS OF RFSTENOSIS

42 5

Enduiuminal slcuting of normal coronary arteries . An alternative means of creating a controlled, chronic arterial injury is endoluminal stenting . although implantation of appropriately sized . balloon-expandable metallic stents in porcine coronary arteries results in only a moderate degree of intimal thickening (Ill), the use of endovascular scents oversized by approximately 30% to 50% has been reported by several investigators to stimulate considerable intimal I'nickening both in normal porcine coronary arteries 1112 .1131 and is animals with diet-induced atherosclerosis (114 .115) . This appears to be a very effective means of promoting smooth muscle cell proliferation and concentric intimal thickening, whose histologic appearance is very similar to (hat of the proliferative neointimal tissue of human restenosis (112) (Fig . 2D). It has the advantage of allowing dilation of the coronary arteries and focal disruption of the infernal elastic lamina and media without the risk of acute closure and intractable ventricular fibrillation . Further characterization of this model is required, however, to better define the relative contributions of smooth muscle cell proliferation. thrombus formation and foreign body inflammatory reaction as principal components of the neointima . It is possible that this model may provide too great a stimulus to smooth muscle cell proliferation and that the pig may be too susceptible to this form of injury to permit evaluation of potentially valuable drug therapies . Unlike dogs and h umans. i n pigs, synthetic grafts rapidly and completely become endotheliahzed and subsequently develop a thick, often calcified capsule lining the lumen surface . This response is not dependent on the age, gender or strain of the pig, suggesting that these variables may be less important in this model than in other animal species (86).

Lessons From Cell Culture Considerable progress in the understanding of vascular biology has been made from the in vitro study of endothelial and vascular smooth muscle cells. Several primary cell culture systems have been described (116) . In one system, proteolytic enzymes such as collagenase and elastase are used to release individual medial smooth muscle cells from the underlying arterial substrate . Cells obtained using this method initially have the morphologically well differentiated appearance of the contractile phenotype . After a quiescent period of several days, the cells undergo a phenotypic change characterized by an increase in cytoplasmic organelle density and a decreaac in contractile elements (70) . This change is accompanied by proliferation of the cells until they become confluent- at which time the increased mitotic activity ceases. If the initial cell seeding is sufficiently dense to ailow a confluent monalayer of cells to form within five population doublings and 5 to 7 days of the original change in phenotype, the cells typically revert to the contractile state . Conversely, if the cell, are initially seeded sparsely, they typically remain in the synthetic phenotype after confluence has been achieved (70) . A second culture system (116) uses



42 6

MULLER ET AL . .ANIMAL MODELS OF RESTENOSIS

JACC Vol . 17 . N., z February 1'H2 :41h-12

Table 2. Comparison of Arterial Size and Susceptibility of Animal Species to Atherosdero,is Anenal Dmmeter rmml

Sueceplib,l,ly to Athero,cleru,n

Specie,

Coronary

Periphrrel

Rat Rabbit New Zealand WHHL Cog Prima¢, .S Domestic pig Mintpeg

0 .05-0.2

Aona 2 : aevind I

Rni,lam

Aorta Aonu .Anna A- .

Dietary chele,¢r.. . . Rapid . p--w, Re, .,tant Variable

0,14A 0.1-113 a7-2 .5 0 .6-2 .5 1-2 .5 1-2 .5

6. iliac 2-3 5 -. iliac 2-3 15 : femoral 2-4 6-10

Anna IN 24: -,d 3-5 Anna 15-1&. carnlld 2-3

Stmdar m human Similar to human

WHHL = Waanabe hedmble hyperlipidemic rabbit .

substrate-attached explants from arterial media . Cell outgrowth from the explant begins after approximately I week and the cells tend to retain a synthetic phenotype, even when mitotic activity returns to normal (116) . In vitro studies of response of endothelial and smooth muscle cell interactions . Cell culture techniques have become an increasingly important means of studying the behavior of individual components of the arterial wall . In vitro studies have highlighted interspecies differences in the response of endothelial and smooth muscle cells to a variety of . Adult rat aortic smooth muscle cells, for example . stint,!( undergo a more rapid change in phenotype and begin to proliferate earlier than adult rabbit or pig cells (70) . Cell culture studies have also proved valuable for the comparison of the growth kinetics of smooth muscle cells from normal, atherosclerotic and restenosing human arteries (9 .117), investigation of interactions between the constituent cells of normal and atherosclerotic arteries (69 .118) . investigation of factors that control cell migration (119) and cell adhesion to extracellular connective tissue matrix (120) and evaluation of cellular proliferative and metabolic responses to endogenous and exogenous mitogens and chemotactic factors (5,30,121123). Interpretation of the results of these studies and their extrapolation to the clinical situation are clearly limited by their isolation from the complex interactions that occur in vivo . Nonetheless, i:. addition to helping unravel the intricate series of events that characterize the intimal proliferative response to arterial injury, cell culture techniques may prove to he useful as a screening system for the development and evaluation of pharmacologic agents that inhibit smooth muscle cell proliferation and the synthesis of extracellular matrix without interfering with regrowth of the disrupted endothelium .

Pharmacologic Prevention of Restenosis in Animal Models A principal aim of this review is to compare the predictive accuracy of the commonly used animal models of human restenosis . Suck an undertaking requires a brief consideration of several statistical issues that are relevant to both

experimental animal and clinical restenosis studies . The overwhelming majority of clinical pharmacologic studies reported to date have failed to show a significant reduction in the incidence of restenosis (6). With few exceptions. however, these studies have lacked the power to detect a moderate therapeutic effect of the drug therapy . To detect a 33% reduction in the restenosis rate (assumed to be 307, in placebo-treated patients) at a significance level of p < 0 .05 with a power of 0.8, a randomized trial needs to enroll 320 patients with an angiographic follow-up rate of 100% . To detect a 25% reduction with the same degree of confidence, 596 patients with complete follow-up would be required. To date . however, only one published randomized study (124) has recruited >500 patients. The potential for false negative studies (type 11 error) also exists in experimental studies . Far fewer animals are required than in clinical trials, however, because histologic examination allows measurement of neointimai thickness as a continuous rather than a dichotomous variable and because complete follow-up is always possible . Of more concern in experimental studies is the consistency of the response to injury . The more predictable the severity of the inflicted injury and the extent of neointimal thickening in response to the injury . the fewer animals are required and the greater is the statistical power of the study . Some models of vessel wall injury may result in a regionally variable degree of arterial trauma. Ideally, in these models, some means of correcting the degree of intimal thickening for the local extent of mural injury should be performed . From the preceding discussion, it is clear that there are major differences among the animal models, particularly in terms of the nature of the arterial injury and the composition of the neointima . It could be expected, therefore, that a pharmacologic therapy that is effective in one animal model may be ineffective in another animal species or in humans . This has, indeed, been the experience with several classes of pharmacologic agents (Table 3) . Heparin . The agent that has been most extensively evaluated in animal models is heparin . In vitro studies 170) have suggested that a heparin-like glycosaminoglycan elaborated from endothelial cells maintains medial smooth muscle cells



Nu

Mtll .l

]ACC Vul. 19 . 2 1992.418-3?

February

4 27

En El' AI. .

ANIMAL MO'JEL.S OF RES1'EN(ISIS

Table 3 . Pha•alaoologic Interventions in Animal Models and Humans Drug Therapy I.

2.

3.

Antithmmb,n agent, Heparin Hmtdm Aellplalaiol agent, Aspirin Other Catourr, channel hlucken Ni0dinme Verzpamd

R :d Carotid

Colt Call-

+1

- 11 .121,

.1

I 1

-

Him.

0

+ 1173!

ll

-11321 II

IIN

+ 1134 .1371

- 11121

11371

1136 .1371

- 11651

+ 1141!

Human

Pie

+1!311

1 111, +1141, II • 11410

Dill-em

mmhdors Caplopnl Cilaeannl Immunoeuppre„1ve aeenr, Steroid, Cydo,por,nc Ammnogenic scent, Colchiaw Other fish oil, Growrh factor anta_comu, Anginpeptln Teopidil HMG-Co A eaduorasc i,h,b,lore Lovaaat!n

Rabbit

rWfi-loll

II

(139 .140) - 1166,

-11441 -

11461

-

11451

4 . ACE

5.

6-

7. 8.

9.

114'_1

-

11421

. -

115 .^1

11

0

+11481

11371

0

1169, 172,

-(149)

+ 1151,

0

- 11241

- 1168,

o

a

+ 11531

0

- 11541

1

11701

- 11711

0

1173,

0

x (174.175)

C

- 11591 -nssl

115!11

11591

0

-11771

11571

0

+ 116! .16 2 21

'1

The numbers in parenlt dicate the oiled ludy reference .

0 -

no Iady

+ 1163,

erpcn.-d

-

+ 1139,

•

11641

Iherapy -11-we, + = positive eesuhs reported : s = bath

positIleand negative -.It, rep.ucd,

in the contractile phenotype and inhibits smooth muscle cell proliferation. The growth-inhibitor} effect of heparin on quiescent smooth muscle cells in vitro is 50- to 100-fold greater than its effect on proliferating cells (125) . It appears to act by inhibition of cell transition from the prereplicative phase (GI) to the phase of DNA synthesis (S phase) and can be added as late as 12 to 18 h after stimulation without reduction in its antiproliferative activity (126.127) . In vivo studies (128-130) have consistently shown heparin to be effective by several routes of administration in the rat carotid model (at doses of 50 to 100 U/kg per h) and in the atherosclerotic rabbit iliac artery model (131) . The inhibitory activity of heparin in these models does not appear to depend on its anticoagulant activity because the nonanticoagulant subfraction of heparin is equally as effective as heparin with anticoagulant activity (128 .131) . In contrast . intravenous heparin titrated to a partial thrombuplastin time 1 .5 to 2 .5 times normal (approximately 10 to 16 U/kg per h) and given for 18 to 24 h after uncomplicated angioplasty was ineffective in the prevention of restenosis after balloon angioplasty in humans (132) . It is possible that more prolonged infusions, higher systemic doses of nonanticoagulant heparin or the use of high concentration local therapy may prove to be more effective in clinical studies ; one randomized, placebo-controlled 400-patient study of low molecular weight heparin (enoxanarin) in a dose of 40 mg/day subcutaneously for 28 days has recently completed enrollment .

Alternatively. other more potent thrombin inhibitors may prove to be effective in clinical studies . In one study (133), for example, intravenous recombinant hirudin was more effective than heparin in preventing recurrent stenosis after balloon angioplasty in atherosclerotic rabbit iliac arteries . Antip'arelet therapy . The role of platelets and plateletderived mitogenic factors in the genesis of atherosclerosis and restenosis has been much debated . Friedman et al . (134) noted suppression of the intimal proliferative response to balloon injury in severely thrombccytopenic rabbits . Other investigators have noted a reduction in the susceptibility of swine with van Willebrand's disease to atherosclerosis induced by atherogenic diet alone (135) or dietary manipulation and balloon endothelial injury (136) . In animal models of restenosis . however, antiplatelet therapy has been inconsistent in its efficacy (Table 3) . Although Faxon et al . (1371 observed a reduction in the incidence and severity of recurrent stenosis in the atherosclerotic rabbit model in animals treated with sulhnpyrazone cr aspirin and dipyridamole, antiplatelet therapy was not effective in the rat carotid model (138) or stented atherosclerotic porcine coronary model (115). In the study of Clowes et al. (138), although the antiplatelet agents (aspirin . reserpine and flurbiprofen) effecti .ely inhibited in vitro platelet aggregation, platelet adhesion to the denuded rat endothelium was unaffected . Because cyclooxygenase inhibitors such as aspirin block only thromboxane AZmediated platelet aggregation. i t is possible



428

tALC VOL IY.Nv.2 February 1992'410-3-'

M ULI .E R ET AI ANIMAL MODELS OF OCSTENOSIS

that more potent antiplalelet therapy directed, for example, at the platelet glycoprotein lih/91la receptor may be required to prevent smooth -ascle cell proliferation . This may explain the lack of impact of standard antipiatelel therapy on the incidence of restenosis after balloon angioplasty in clinical studies (139.140) . Calcium channel antagonist therapy . Several calcium channel antagonists have been evaluated in animal models of restenosis . Jackson et al . (141) noted a reduction in DNA synthesis after aortic balloon injury in rats treated with nifedipin, . verapamil, diltiazem or lanthanum. Nifedipine (10 mg/!:, ; . orally two limes a day) was also effective in reducing aortic neointimal formation in normolipidemic rabbits 14 days after balloon injury . In contrast, in other studies, verapamil 110(1 mg-kg per day) was ineffective in preventing intimal proliferation in the rat carotid artery (142) and nisoldipine (20 ng/day) was ineffective in preventing restenosis in the atherosclerotic rabbit iliac artery (143) . Although calcium channel antagonists are widely used clinically in the postangioplasty period, data from clinical trials have also been conflicting. Whereas two relatively small, randomized, angiographically controlled trials (144,145) showed no reduction in the incidence or severity of restenosis with calcium channel blocker therapy, preliminary results from an equally small study (146) suggested a beneficial effect of oral verapamil . Angiotensin-cnnverling enzyme inhibitors . After preliminary studies (147) soggcsting that , .local angiotensin sy°tent may regulate the vascular response to endothelial injury, Powell et al . (142) examined the effects of very high doses of the angiotensin-converting enzyme inhibitors cilazapril (10 mg)kg per day) and captoprii (100 mglkg per day) on neointimal proliferation in the rat carotid artery . Both agents were highly effective and concomitant heparin therapy appeared to exert a synergistic antiproliferative effect . Similarly, in the atherosclerotic rabbit iliac model, cilazapril (5 mgi.g per day) reduced the incidence of restenosis after balloon dilation (140). In contrast . Lam el al . (149) found no benefit of high dose cilazapril (20 mg/kg, twice a day) in the porcine carotid angioplasty model . Further data on the clinical efficacy of angiotensin-converting enzyme inhibition will be available in the near future from a large, multicenter cilazapril trial . However, the doses used in this clinical study (I to 10 mg, orally twice a day) . when adjusted for weight . are considerably lower than those used in the experimental studies . Other agents . Many other agents have been evaluated as potential inhibitors of restenosis . Several of these agents have been investigated in more than one animal model and in humans. Corticosteroids have been shown to inhibit smooth muscle cell proliferation in vitro (150) and to inhibit the development of stenosis in the atherosclerotic rabbit (151) . but they do not prevent restenosis after balloon angioplasty in humans 1124) . In in vitro studies colchicinc has inhibited many of the processes involved in restenosis, including smooth muscle cell proliferation, inflammatory cell chemo-

taxis and secretion of collagen (152) . and in the atherosclerotic rabbit, it reduced the incidence and severity of restencsis when given in very high doses (0 .2 mglkg per day) (153) . In a recently completed clinical study (154) . however, colchicine therapy (0.6 mg twice a day) was associated with a relatively high incidence of side effects but no reduction in the incidence of restenosis . Several grmrth fortor inhibitors have proved effective in animal models and are currently undergoing clinical evaluation . These include trapidil, a platelet-derived growth factor antagonist shown to be effective in cell culture (155). rats (1561. atherosclerotic rabbits (157) and one pilot clinical study (15K) . and angiopeptin . a somatostatin analogue that inhibited neointimal proliferation in rat carotid, rabbit iliac and nonhuman primate femoral angioplasty models (159) and in a heterotopic cardiac transplant model of accelerated atherosclerosis (16(1 . Clinical trials of angiopeptin are also currently in progress . Finally.lovastatin, which blocks the production of mevalonic acid and the synthesis of choleslerol by inhibiting the enzyme HMG-Co A. reductase, has beer .shown to inhibit both cell proliferation in culture 1161 .162) and restenosis after balloon angioplasty in rabbits (163) . Preliminary data from a small pilot clinical trial (164) suggest that lovastatiu might also reduce the incidence of restenosis in humans .

Conclusions It is evident that no single model of vessel wall injury has yet been shown to reliably predict the impact of pharmacologic interventions on the incidence of restenosis in humans . Moreover, there appears to be a considerable heterogeneity in the susceptibility of the individual models to these interventions, This is likely to reflect not only the use in most models of disease-free rather than atherosclerotic arteries and differences in the severity of the vessel injury, but also interspecics differences in the healing resporse to arterial injury and intrinsic differences in drug metabr ;aism, lipoprotein metabolism and the activity of the fibrinolytic system . It may also r eflect. in part, an inadequacy of sample size and angiographic follow-up in the clinical trials and the considerably lower drug doses used in these studies to avert the possibility of systemic side effects . Each of the models may be suitable for specific investigations- Cell culture studies remain a very valuable means of investigating smooth muscle cell and endothelial cell physiology and may allow rapid screening of multiple potentially effective pharmacologic agents . Balloon iniurv of the rat carotid artery or aorta may represent the simplest in vivo model for the study of pure smooth muscle cell proliferation . In contrast . evaluation of interventional devices requires the use of larger animals, including the atherosclerotic rabbit or swine, or interposition grafts in peripheral canine arteries . Which of the models will most reliably predict the efficacy of pharmacologic therapic-, in humans is not clear . To date . with few exceptions, published studies of pharmacologic



MULLER ET AL . ANIMAL MODELS OF RESTI'NOSIS

IA CC Vol, 19. No . 2 Feb- 0, 1972 :418-3?

interventions in the rat and rabbit models have been positive, but the agents investigated have been ineffective when tested in clinical trials . In contrast . although relatively few pharmacologic studies have been performed in nonltuman

pigs . with few exceptions . the reported sludhes in these models have shown no reduction in the severity of intimal proliferation . Of all the models studied, perhaps the primates and

most convenient and must histelogcally comparable to human restenosis is the normal porcine coronary artery with an oversized metallic tent . However, no therapeutic ma-

neuver has yet been shown to mtcence II'.e extent of intimtil thickening in this model .

The must import can Omit .1 iorpll,,Illicit rf rho,r nhaerrations is that considerable caution should he exercised in the extrapolation of positive results of pharmacologic studies in

experimental models to the likelihood of drug efficacy in humans . Ideally . promising new drug therapies should he proved effective in several animal models before expensive . large scale, clinical trials are initiated . Once therapeutic efficacy has been demonstrated in multiple models, clinical trials can then be undertaken with appropriate consideration of sample size and the adequacy c f angtographic follow-tip, It is hoped that this systematic approach will permit a more

r.;pid identification of unequivocally effective pharmacologic therapies .

References I . Deere K . Holubkov R. Kelsey' S. et al . Percutanena, tratnlommaO nary angioplasty in 1985-1996 and 1977-1981 . N East J Slot 19fiN318265-70. 2 . Satan RD . Gelbfi,h 1S . Erny RE . Schodt SJ. Schmldt DA . Baim US . Coronary atherectomy : clinical. angiecrunhic . and h :ntotogical findr,g, and observations regarding potential mecham,m, . Circulation 161 :82 . 69-79. 3 . MullerDWM .Elli,SG-Adaance,incoronaryanginpla,ty :erdo,a,wtar scents. Coronary Artery Disease 19911:1 :438-48 . 4. Karsch KR . team KB . Voelker W . Baumhach A . Mouser M . Seipel L. Percutaneous coronary excimer laser angioplasty in patient, with table and unstable angina rectons : acute results and incidence of resteno,i, during 6 month follow-up . Circulation 1990:81 :1849-59. 5. Liu MW. Roubin GS. King SB. Res:ccvs after coronary angioplasty . potential biologic determinants and role of intimal hyperplasia . Circulation 1989:79 :1374-87. 6. CotIff RM.Ohmao EM . Frid DJ. el al. Ro,teno,i : the clinical issues . fn : Topol ElL ed. Textbook of Interventional Cardiology. Philadelphis : WB Saunders . 1990:363-94 . 7 . Walter BF . Pinkenon r'A . Kerewhirs D. Luther S. Pin' . RP . Morphologic analysis of 566 coronary aih o,comy specimens f om 1071raticut, : histologically similar findings of restenosis followmc pnmaq balloon angioplasty versus primary atherectomy 'ahsirl . t . m Coll Cardiol 1990.15:147A. 8 . Walter BF . Pinkenon CA . Rothbaum D .A . et at . Restenn ... n following hot tip laser . excimer laser. primary atherectomy and balloon angioplasty procedures : hismlogically similar intimal:ratien prat in 33 athercmmy patents . Circulation Iwp'e'ISuppl HH :H7-3t2 . 9 . Dart .' PC. Voisard R. eaonr000l G. Hottng B. Her; E . Growth characfristics and cytosketelal organiaGan atculnred smooth muscle colt, from human Inanity ntenosing a, . •J resienosing lesions . Anerio,clee tests .990:10 :62-75. 10. Johnson DE. Hinohara l' . Salmon MR. Braden U. Simpson JB . Primary peripheral arterial stenoses and re i,ed by transluminal athcrctomy :'a hi-pathologic study . I Am Co11 Cardio 19a1 :15:4t9-

25.

429

II . Beml KI. Serruyc PW . Hugenholtz PG. Reeenosis alien wronmy angmpla,ly , new standards for clinical studies . J .Am Call Cattail 199 i . t r 491-8 . C . rotor CI'. Judkins MP . Transluminal treatment of anerienulerotic oh,Vucion : description of a new technique and a preliminary report of n, application . Circulation 1964 :30:654-70. n . Grucmoig AR . Transluminal dilanrlion of coronary artery scents, Linen 1978 :1263-6 . 14 . !€ y 1' C. Baughman KL, Pasternak RC. Fall- JT. Transluminal inµinplu,ty g correlation of morphological and angiagrapbic findings in an ..xpenmcnlal model. Circulation 1980 :61 :778-85. 1 1 . IW,lu, :an KL . Paoternak RC . Fallon J'L Block PC . Trensluminal naq a,g,opl,,Iy of postmmlem human hearts. Am 1 Cardiol 1981 ; anh al-7. 16 5ailiv. 11. . Tally WG . hlcClennan BL, Gilula LA . Peminaneous transluminal angioplasty : radiological-pathological correlation . Radiology 1981 :141651-4 . 11 Block PC .MytcrRK .StenurS.Felfn1T.Morphologyafter Iranslum. uv,l ingioplaaty in human being,- N Eugi I Mod 1981 :305 :382-5 . 18, . MlzunoK .Junta A.ImazekiN .Pa!hologicalfinding, after percataneous ,-For -I coronary angioplasty . Pr Heart J 1984 :52 :588-90. 19 Sm+ard AL. Eased CE . Semrys PW . Coronary arterial findings after accidental dealt immediately afie, suwes$ul peremaneoustfansluminal • nary angioplasty- Am 1 Cardiol 1985;56:794-5, tn. Wu11er BF . Garfinkel HL Rogers FL Kent KM, Robens WC . Early and late morphological changes in major epicardial coronary anterior after percurnocan transluminal coronary angioplasty . Am I Cardiol 1984:53: 470-7(' . 21 . RobinHis . Robins WC .Effectsofpercuraneoustrnsluminalcoronary angr tplu,ty on atherosclerocic plaques and relation of plaque composta., m,d arterial size to outcome. Am J Cardiol 1988 .62 .41-50. 22. w'a11er BF . Coronary luminal shape and the arc of disease-Gee wall : morphologic observations and clinical relevance . J Am Call Cardiol 1965 :5 1100-1 . 21 . L,ud CI :, Van den Brand M . Bukei AE . Trensluminal coronary soninplaav and early restenosis'r fibrocellular occlusion after wall lacerm . If, bean J 1983 :49:171-L 24 . Austin GE. Ratliff NR . Ilollman J . Tabei S. Phillips DF . Imimal protfertion of smooth muscle :ells as an explanation for recurrent • r artery srreosis after percutaneous transluminal coronary angeepla,ty, J Am Call Cardiol 1985 :6:369-75. 25 . Waller BF . McManus BM, Gmfinkel HJ, e1 al . Status of the major • nary anerie, NJ 1, 150 day, afar percmaneou, lan .daminal cam. anpmplasty : anaiyri, of 5 necropsy patients-Am 1 Cardiol 1983:51 : nt-a

26. Wa1kr HF . Pinkerton CA, Orr CM . Slack JD. VanTassel JW. Two dntinel types of restenosis lesions after coronary balloon angioplasty : intimal proliferation and atheroscieratic plaque analysis of 20 crnpsy patients (abstrl . Circulation 1990:82(sappl I11y111-314 . 27. Ueda M . Rocker AE, Rubde T, Numano F, Faiimoto T . Fihrocellular b,wc response after Fercutanems transluminal coronary angioplasty : an arm unocyiochemical analysis of the cellular composition . Circulation 1991.83.1327-32. 288 cander(rtessenWI .SemuysPW,vanBZUSekomHMM,eta) .Earlyand lira hi-logical features after stealing of human saphenons vein bypass graft, iabstrl . Eur Heart 1 1990:11:272 . 29, VirchowR .Deratheromatou.sProussderArteries. WienMedWacher,chr 1856 .6:825-9. 30 Ro„ R . The pathogenesis of atherosc!eeosis-an update . N Pent J Mcd 196 :314.488-500. II . Faaeimto A . Ross R . Harbor L. Studies of lrypercholeslerolemla In the on-human primate : I . Changes that lead to fatly streak formation . Aneriusclemsic 1984 .4 :923-40 3'_. Faggimto A, Ross R. Srudins of hvperchMesrerolemia in the nun-human primate: If. Fatty streak conversion In fibrous plaque . Anerieiclerosts 1984 :4:341-56 . 33. Gernty RG. Naito HK . Richard- M . Schwartz CJ . Dietary induced athcn,gcnesis in swine: morphology of the imima in prelesion stages . Am J Pathol 1979 :95 :775-92. 34. GcnnOs MG . The role of the mnnocyce in etherogenesis . 1. Transition of hit on I borne monocyms into foam cells in folly lesions . Am J Partial 1981 :103:181-90.



4 30

MULLED I!7 AL. . ANIMAL MODEL1 OP HCS1I.S 1515

35. Rosenfeld Me . 1 rggimlo A. Roar R . The rule of the m nuclear phagocytemprimateandrabhitmodehofothuro,dan,e+ . I Van Turin R, eel . Mononuclear Phagocyle'a Characteri,lin . P6y,inlogy and Fare Iron . The Have Mamma, Nilholf. 1985 :795-851. '6. Geer IC . fine structure of hum ;m annic in tonal Ihis kcerng and fall y ,,teak, . Lab Invert 1965 .14 :1764 .-87 . 37 . McGill III Ir . Peniaenr problem ~e the pathogenesis of amermdcm,is. Aneriowlerous 1984 :4:441-51 38 . It, 1H . lacier V . Bud I_ Bad- J, T'aubman 010 . Chicacbr. J1L Syndromes of 0ecelervted amero,a!ermi, . role of ve,cular injury and smooth muscle coil proliferation . J Am Coll Cordial 19990.15'.166,7-8?. 39 . Juck,un ST. Gouo AM Jr. Hypo,huic concerning men brane sfracture . cholc,ternl and nthero+elero,is . In. Paolelti R. Gouo AM Jr, oit, . Atherosdernais Reviews. New York Riven . 19'.6',1 :1-15 . 40 . Minick CR. Fahrican CG, Fahricanl l. Llrcnr MM . Alhcroaneno,dc" sI induced I infection with a herpe,vfru, . Am J Path,! 1979 .96 67341 . Melnick JL. Drcernrn GR. McColleD CH . Porno III ., BunIsl . DeBase p ME . Cylomegalnvims antigen within humus aerial smooth musce cell, . Lancet 1983 :2 :644-7 . 42 . Harker LA. Slichrer 5J . Scott CR . Rot, R . Borwcy,linemi, : vascular i jury' and arterial thrombosis . N Eng! 1 Mcd 19'4 ;291 :517-43. 43 . Barnhart GR, Pr- EA. Mills 5A . e t al. Accelerated coronary aneriecardiac Iransplanl recipients . Transplant Rev 1987:1 :31-46. 44 . Demernr AL Zerhe T, Banner B . Morphology of arhd Argon allograft anenopathy: Identification of proliferating monA cell 10,111 .11-1 . Transplant Pros N8921 :3667-9 . 45 . Hard,, NH . %iiniek K, Murphy G . Experimental mdudinn of erhem:m rennsclermw by the synergy ,f allergic injury I-rue,u :, ..Ad lipid Jiet. Am 1 P:dhol 1973 73 :351-!7 . St., 46. FH . Gavin 18. Vandeioce MA . The manna of human coronary arteries . Am Heart 1 1999.1IS32-9 . 47. Wulirnky H. Glagov 5. A lamellar unit of aortic medial nucture and function In mammal, . Circ Re, 1967 _1 .99-Is.l . 48. Bony CL. Organogenesi, of the arterial wall. In : Camlllen J.P. Harry CL. FeissingerJ"N . Rarimy J . eds . D.,ez,e, of the Arterial Wall . Berlin'. Springer-Verlag, 1989 :57-9 . 49 . Sims FIJ . A comparison of li dural features of the walk of coronary anede,Iron If ddTere~n nl l species. Pathology 1989:21 :115-24. 511 . Wittier Rill. Vesulinovitch D. Eup omenal model, of human olherosclcr„is. Ann NY Acad Sci 19hf .l49:907 22. SL Gnu, 1)R- Animal Model, m Cardmvascular Research . Novon : Slanim Nijhatf. 19R5'557-47. 52, Vesselinovitch D. Animal models and the swil of athero,dero,i,, Arch Pathol Lab Mad 1988 :112:1011-7 . 53. Kulesky S . 0be,e spontana,osly hypenendve rats : a model for the study of athennclerusis. E8p Mot Pathol 1973t19 :sr-nit. 54, Mikal EM . Were, JM, Schueberg SM. et al. Development of athero'cleratiolike lesions In the sand at LP mnuarn,, nlrrsrrl . Coronary Artery .,,,a,' 1990 :1 :469-76. 55. W'ilgnm GF. Ingle DJ . "Spontaneous" cardiovascular lesions in rats . In: Robert, IC Jr. Straw, R. eds . Comparative Alherusdcmsie Morphology of Spuotaneou, and Induced Athemsclerolic Lesions the in Animals and It, Relation m Human Disease . New York: Amer 1965 :07-91 . 56 . Fishman JA. Ryan GB . Karnovsky ML Endothelial regeneration in ,he rot carmiil artery end me significance of endothelial denodatien in the pmhngenesis of myointimal thickening . Lab Invest 1975 :32 :339-51 . 57 . CIA-, AW . Karnovsky 111 . Failure of certain antiplalelel drug, to ulfocr myointimal thickening following aneerial endothelial injury in the rat. Lab Invest 1977.36452-64. 58. Baurnganner HR. Eino neue Methods our Erzeugung der Thramben durch gueihe Cberdchnung derGefaswand . Z G,, Exp Paled 1953 :137 . 227-49. 59. CI-', AW . Reidy MA. Closes MIM . Kinetics of cellular proliferation after arterial injury. I . Smooth modde cell growth in the absence of endotheliun . : rest 19x349:327-33 . 60 . Closes AW, Schwan2 SM . Significance of quiescent smooth muscle migration in the injured rat carotid artery . Circ Re, 1985 :55:139-45, 61 . Haudenschild CC, Schwano SM. Endothelial regenerti,n . 11 . Restitulion of endothelial continuity . Lab Invest 1979 :41 :457-18.

1ACC Vol . 19. No . 2 February 199: :ale-S! 62. Clone, AW . Ch-, MM . Reify MA . Kineul, of cellufu pwliferalinn after arterial injury . 111 . EnJnthebal and emooth muscle cell growth in chemically denuded vcnel, . Lab Invert I9h6 :5a'395-103. 63. Cloor, AW . Ready 014. Cluwes MM . M,echani,m, of ,rnnn,n after an-al injury. Lab Invest 1983 :49.208-15 . 64. Fingerle J . .Au YPl' . Clew ., AW . Reide MA . Innmal boon formmmn in mid anerie, after cndnrbotral denudarion in ah,eace of medial mlua'rAneriosdumi, 1990 :10 :10x2-7 . 65 . Reidy MA, Silvor M . Endothelial rgeneratinn . VII . Lack of rnhmal . Am J Pmhni 1995 :119' preiiferuon a her defined iop, ,o rat con 171-7 . 66 . Linda,, V . Marck RA . Reidy M.,:. Bas¢ fibrobla,t growth factor ,limulalee endothelial regrowth and prc''feronon in denuded anenn . I Ch . Invest 199:1:85:-_sold-8 . 67 . "fiell fill .. Slemernan MR, 'ipael 'Ill . The influence of the pnmtary on arterial inGmal prohlcralion an the rat . Circ Ken 1910 .42 641-9 . 69 . Wcin,lcin R . Stemerm,sn MB . Slating T. Hor nal raquircmenb far gruwthnfanerialsmoothmuwleaellsiomra nenducnneapproach1. athemsderosis. Science 1981 :.12:810 10. 69, Schwamn SM. Headin EP. Aortic eedothcllal cell replicau .n . I . 6tfecr, of age and hypertension in the rat . Circ Roe 1977:41 :240-55. 70, Campbell GO, Campbell JH . Mandeeson JA, Homgan S . Rennick RE. Arsenal smooth mascle . a multifunctional mesenchymal cell . Arch Pathol Lab bled 1988 :112:977-86. 71, Stemermen MB, Weinstein R . Role 1 . Mating T . Four . R, Gardoce R . Vascular s mth muscle cell growth kinetics in vivo in aged rob . Pmt Null Acad Sci USA 198? 793863-6. ,!, Haul AID . More RH . Spomanoom lesions of ,he aorta in the rabbit . In get 15:1 s-75 . 73, P,llak 01. Experimental ancnepathks in the .bar (the devil's codcarolI n Bef55291-308. 74, Moore S . Thromboalheroscler,xis innrnnolipemic rabbits : a .,all of continued endothelial damage . Lab Invest 1973.29 :479-87. 75 . Farer DP. Weher VI, Haudenschild C. Golf man SO . McGmem WA. Ryan 11 . Acute effects of trnsluminal angioplasy in three model, of athnrrnclannsis . Aneoe,derosis 1982:2 :115-33 . 76, Bole LM, Kilo T. Goldstein IL . Watarabe Y . Brown MS . Cellular patholo_ey of progressive atherosclerrotis in the WHHL rabbit : an animal maJd of familial hyperchoiesterolemia. AneriA sclerosis 1983 ;3;87-101 . 77, Block PC . Baughman KL, Pstemak RC . Fallen IT . Trannluminal ngioplanty : correlation of morphologic and AngicPaphic findings in an e.perimental model . Circulation 1980:61 .778-85 . 78, bacon DP, Sanbom TA . Web., N J . e t al. Restenosis following tmnsluminal aAgiopla,Iy in eapenmenral atheroselerosis. Aneriosckrods 1984 :4 :189-95 . 79, SarembockII,LuVeacPa,SigaISL .etal.Influerceofinflationpm,,,m and balloon steer, the development of inrimal hyperplas:a after balloon angioplasty: a study in the atherosclerotic rabbit. Circulation 1989 :80: 1029-40. 80 . Pat. . JC . Windeler SAM, Gareia FM, Ti . F0. Sibbitt RR. Heater SR. Atherosclerotic rabbit conea: expandable tnlralumiaaigrBing. RadiolnSy 1986:IRS.723-6. 81 . Anderson HV, Zzamn GS. Roubm GS. Leimgraher PP. Groenlzig AR . Steeroble hberoptic catheter delivery of laser energy in atherosclerotic robbds . Am Heap 1 1986 :111 :1065-72 . 82 . Hansen DD. Au It DC . Vrooko B, Rilchir JL Rotational arhereaomy in arheroselerelic nbbil iliac arteries. Am Heart J 1988 :10 :160-5. 83, Consigny PM . Tulenko TN, Nicosia RE . Immediate and long-term effects of angioplasy-balloon dilation on normal rabbit iliac one* . An, riosclerosis 1986:k26`-76. 84, WeidingerFF,MrLenachanlM .Cybulsky61Lelci.Persistentdysfunc . lion of regeneroled endothelium after balloon aeuioplanty of rabbit iliac . Circulation 1990 ;81 :1667-79. -n' 85 . GA', JC . Guidry AMA. Experimental canine alherosderosis- In Ref 55 - 17C-85. 86. Sausage LR . Be:ger KE . Wood ST. Yates 36 . Smith JC . Mansfield PB. Intorspecies healing of porous arterial prostheses : observations from 1960 to 1974 . Arch Surg (974;109 :698-705. 87, Kirschnrein W, Simmner S. Demplle CE- et ni . Impaled fibroolyric careen" and tissue plasminogen release in patients with rest-is



IA CC Vol . 19. Ni, . 2 C h0a y 1982!418-32

perculanmu, Iran,luminal

MULLER ET AI_ ANIMAL MODELS OF RESTENOSIS 00000,00

angropla,ty IPTCAI .11-olb Ho,

must 1999 :62772-5 . 88 . Orlandi C . Slngh 1P, Boll FP . Scharb RG . Preldcrmoe and 'pad mnabolom response to balloon uagwpl,,ty in canine renal Ac,1 Am Call Cardiol 1990:13 1394-4410 . 89 Zollikofer CL, Redha FH . Bmhlmann WF . el al . Ace!, and long or ., elf"', of ma„ive balloon dllalmn on the aon,c wall and _a vusoono Radology 1987:164 :145-9 911 . William, (K, Arm0tmng ML . Heutad DO Lodolbebal deoudanaa u.In prol,Ioraton of va,a e avrtum 1 V,n c Mad biol 1990 2 .12-7 . 91 . Bangor AC . Beeevke, 12. Lamey LL . S,Ioernan Kl . Hypolhc,o : n a neovascularizanon of human coronary anenes . N F oo l I Mod 1994 .310 10175-7 . 92 . Gage AA .F eka,G .Rd.FEFr-mgmuryto!argebloodvc„eb •m day, Surgery 1967 .61 748-54 . 93 . Schal, RA . Palm, IC . (04 II) . (,raft S . Garaa U . RMOmr SR . 0a1o01 enpandahiP unraconmary

I-,

m the dull dog Cueu!aoon 1917,76 .

4511-6 94 Roubm GS . Robinson KA . K op SB . 0 al- Early and late ,corn, of in!naarenary arterial boding a fmt coronaq angaopla,y a dog, it . eulatmn 1987.76 :891-8. 95. Ham- DD. Aulh DC . Hall M . Rnchn 11 . halal orwkndanerencmy to rural canine cannery anenes . prclimmar; ,epos . 1 Am Cull Cardl 198% :11 :10773-7. 96. Filippo C . Abela G5 . Feech A . Smdh W . Pepme Cl . Conl, CIe . T-1- -d la, nadiation of a n Inca d,:,,- . angiographi, and morpholugm

lady ofaaca,-e effect, . Ar, I Carded

1985 57 . :171-4 . 97. AbnIa G5 . Co ., . R. Na-aim S . Fddman RL . l'epnc CJ A new model for esligaliun of !ramluminal re nalvaone human ather- ierouc 00100

q anery ncnngrafts Am 1 Cardtal 1984 :34 :2W-5 .

9B. Taylor CB. Eopenmemalr, induced anenosderoso is nonhuman an more, In Ref 55 .215-43 99. French JE . Jenning, MA . Floret' MIA . Morphrlagic:d ludic, en alherosderosis in some . Ann NY Acad Sci 156` .527789-94 . 100 Luginbuhl H . Jose, JET. The morphology and rnorptogene,u of amerosderosis in aged swine . Ann NY Acad Sui 1965 .1 227063-79 .

101. Moreland AF. Espenmemal athmnsckn,is of some . In Kef 5121-4 . 102. Mordand AF . Clank- TB . Loflard HB . Arherosclerosi, r "m-maa tire - swine. Arch Palhol 1963;76 :203-10. 103. Ro,tmae IS. Mahley RW, Fry DL. Yucatan miniature seine as a model for dietindoced atfrerosdcmso . Alhrro,dcron, 1982 :43119-1 2.

103, Nam SC. L nc WM .Jamnolych J, Lee Kl'. ., ThornWA . Rapid produo lion of advanced atheroschrosis in seine by a comh :nalien of endothehal injury and cholesterol feeding Enp Mo! Pallid 1973 :10 369-19 . 105 . Clopath P. Rapid production of advanced alherm:leroais in .are Artery 1931:3 .429-38 . 106 . Weiner BH. Dick- IS . Jarmolych L Fni, KE, Daoud AS . Comparison of pathologic and angoographic findings in a porcine preparation of contrary atherosclerosis. Circulation 1985 .72 :1011-6. 107 . Gal D, 900210ne AL Slovekai GA . el al. Alheroxlerolic Y'cataa mal model with high-gooda, fibncco,ha . nanfaay lesions suitable for testing calheter-based is,erventions . Am Horn 1 1990:119 :291-300. 108 . Steele PM . Chesebro JH, Slanson AW. et al . Balloon anpiaplaay! natural history of the palhophy,iolo2scal response n injury in a pig model. Circ Re , 1985 :57:105-12 . Ill . Weboer MW' , Chesebro 1H, Hems M . Mruk IS. Grill DE, IF.-, V . Effen of balloon inflation on smooth muscle colt oroliferalion in the porcine carotid artery labor . I Am Call Cordial (990 .15:165A . I Id . Dick R .1, Haadenschild CC. Popma 11, Yakuho, SL To,,,rl E2 . Insight, from quantitative hislological analysis of ezcisional athereclomy sp :ciens fabstrl . Circulation 1990IlliappI 111011-311 . Ill . van derGies,erWI,SemrysPW . .onBeusekomHMM .eIal .Coronary denting with a new . ndieopaque, bal!coneapandabic endoprnslhesis in pigs . Circulation 1991 :83 :1788-9R . 112 . Schwanz RS. Murphy JG . Edwards WD . Camrud AR . Vlierslra RE . Holmes DR . Restenosts after balloon angiop ,aay : a practical proiiferalive r.dal in ponnme coronary anenes . Circulation 1990 :82 ; 1 190-230.

113 . Karas SP. Gravanis MB. Robot- KA . Anderberg A . Hear. JA . King SM . Comparison of the response to coronary anery bull-,. ing.0 ad

43 1

vcn!ing in mal mode! of resrennnrs (ahstrl . Cinoolalbn IW(rb2;sappl 1111 :111.497 114 . Woolbcrt SC . Minor ST, Smith GS, et al. lnlracoronary denting ccelcraio, lummal narrowing 1n hypercholeolerolemic swine Izbstr . c0colutiun 1990 :821suppl1111.111656. ill Rodger . GP. Minor ST . Robinsm, K. et al, Adjuvant therapy for mnary dims: investigations in .,Iherosderotic swine, Circulation 19'%x2 .160-9 . I16 1 ho- WA, Kim ON . Urology of disease : atherosclerosis as a hyper, pilla uc anmdr neuplasuc process . Lab Invest 1983 :48:245-55 . 11! U:m,ch PC . Inching,, T, Be . E. Responses of cultured smooth muscle calls ham human rnnatherosderolic arteries and primary stenosing 10,5!000

after photoradiation : implications for pholodynamie Ihcnpy of

scalar stenoses . l Am Call Candid 1990451545--50. 1,0 I{,over GA. McCormick S . Kalant N . Effects of porcine aanic smooth e cell conditioned Tedium an endothelial cell replication . Anted .1989:9 :76-83 . !19. 11,11 ho' . : , Madn JA . EBe,, of platelet factors on migration of :oltuned n. aon;c endothelial and smooth muscle c,IN, Circ Rev 1989 ;65: 1157_0,0, 120 Clymun RL McDonald KA. Kramer RH. Inlegrin receptors on panic noaeh muscle cells mediate adhesion to hbrmnetin, laminin, and coil,sen .t- ire Res 1990:67:173-86. 121 . Klag,brun M . Edelman ER. Biological and biochemical properties of Mitchla,l growth fadols : implications for the pathogenesis of alhero,clean', . Anerioademsis 19899 .169-78 . 021 . Lbhy P. Warner SIC. Solomon RN . Birinyi LK . Production of plateletderned growth factor-like nritogen by smooth muscle cells from human alha . N End 1 :788.318:1491120. Ru,, 11 Flaielel-de-ived growth factor . Lancet. 19891 :1179-82 . 124 . Pepine C1. Iljrshfeld JW. Macdonald RG . eI al. A controlled trial of

conico,teroids to prevent restenwis after coronary angioplasty . C.W.- 1919 :81 :1753-61 . 510 Cauellol Jl . Wrighl IC, Kamovsky MJ. Regulation of vascular smooth tuck cell growth by heparin and hepare sull,nates . Semis Thrmnb Hems,, 1987 :13.499-503, 126. Majesky MW . Schwanz SM-Clowes MM. Clowes AW . Heparin regulat th muscle cell S phase entry in the injured rat carotid artery. 0c Re, 1987:61:296-300. 127. . . Chow AW . Kamovsky Ml. Suppression by herann of srrownh muscle cell proliferation in injured artenn. Nature 1977265 :625-6, 128. Carrot 1R. Rosenberg RD. Clown, AW, Kamovsky ML Inhibition of ratan,,i,l smooth muscle cell proliferation by heparin : in vivo dud ;vs with anticoagu;ant and nenanlicoagulant heparin. Circ Res 1980:46:61'5!4. C9. Clown, AW, Clowes MM . Rondos of cellular proliferation after arsenal injury . IV. Heparin inhibits rat smooth muscle mitogelresis and migen . Lion . Coo Res 1986:58839-45 . 130. Edelman ER- Adams DH . Kamovsky Ml . Ed t ofconnolled advem.tial 1lcparin delivery on smoalh muscle .711 prolfentinn following endothelial injury . 1000 Nall Aced Sao USA 1990 :87 :3773-7 . 131 . Currier 1W, Pool '1K, Haudensahild CC, Minihan AC, Faxon DP . Low molecular weight heparin (eno5aparen) reduces restenosis after iliac angiopla,ly in the hyperchaiesrerolemic rbbih I Am Call Cordial 1991 :17:1188-251. 132. Ellis SG, Radium GS, Wilentz 1 . Douglas IS, Kiag SB . Effect of 18-:4 hour heparin administration for prevention of restenosis after nncampli . -d caranary angioplady . Am Heart 1 1989;117!777-82 . 133 Sarembock H, Gene SD . Gimp!¢ SD. Powers E, Roberts WC. Angiographic aId pathologic study of (he effect of recombinant desulphatohirodin on re sic following balloon :mpoplasty in rabbits (abnlr) . Circulation 1990 :824suppl 1114:111.339 . 134. Fnedmzn RI . Stem,- MB. Wenz B. el al . The edit of Ihrombaytopenia on experimental arteriosclerotic lesion formation in rabbits . 1 Cur Invest 1977,601 135 . Faster V . Lie )T' . Badimon L. Rosemark IA, Badimon IJ . Bowie E1W . Spomaneous and diet-induced coronary atherosclerosis in normal swine and swine with vun Wilkbnnd disease. Arteriosclerosis 1985 :5 :67-73 . 136. Griggs TR, Reddick RL, Seltzer D, Bnnkhous KM . Susceplihility f atherusleros arenas and coronary anenes of swine with van Willebnand's disease . Am I Partial 1981 :102:137-45.



3ACC Vul . 19, No . 7 Felo-, 190 : 419-4z

'.1' 1, ER LT At, . 0141t1HLFIII II! SII Nr1SIS 11,101A1

432

117 . 10001111' .S:mh,.919 Huoden,cl,lIILI .R,an 11 llkcl,Lmopla!elet Ihee :pi 0n rcor,,- af! lmcnr 1 angl0pla,ly- Am 1 Cmdmi Is' 4 .'1 17( -?,( I IN- 0-, 1,19 K .o,co,I 611I lwlurr of oo'r, n anuplcl drug, nI 01,1 3011141444x1 Ihmkemny 10110x30€ :mrnal10doll1110! 111114:5 In me n L,h I10Cgl 1977,36 452-(A . 139 . Scull-, 1 .. Hn1,M(i . 00,1. ...01 1 . 101 . A,mnn and 4,111140 n 101 preyrntion of 111410„11 .11301 419 109014, u"mlum,nal 011.00111,1 . N I :ngl 1 AL:d !199..1IN .17,4 9. 1411 . 1~h 'rI,,, 111, Grooms.€ As . FIa11m,n 1 . King SlS. UOjet" 153013301101142111 n 130 .3030109. of recurrence 18111! I unit .al, 30110 ; tasty . :, 004,0 0J study . GrculaVUn I994 141 . to,k,nn ('! B -h P('. BOwye, DB . Inhlhnnry el(cn of "I" on, aall,,_, . . . On n on bA!,l!ln t0 . 4ndu0J 301 141 mocth am,cle cell pmbf0, ca1 nd 0mn 10A!h-1,1X9..,., 1939 .69- 115-' . 14 2 P0x1 1S. 010101 1-I'. \fuller R KS7. et a1 . I99,6,,lr, of ungi0lemim . en, , myurtmmal Prolderl,, r sfter vauu,ar in co....r^ e,!^ ma - . fury . Science (919.24i 136-N . 141-044,0!! Pemcrw,I,RM .I'euernnkH( FailureofnuoIdipmeIopleveut res!enosis 00e, argiuplauy In n!hero,cler0tr : 1066111 lob,ln . Circula6na I0 H21,uppl1117 :111.972 . 144 . %Vhitxanh HH . Hollman us 1 . el 41 . Flfcn of 308430301 Oa recnrreni,leno,i,afterps-I . l"ob" GS an, ttan,luminal1143011111angi,p!09l( 1 Am Colt Co dlol 1941 .3 271-6 145 . (grc'u1 T. Dav'ul PR . Vul P(i . 1t 0l . l adero of JJ -ear, to 101030 '4190,10, 4fter 410,00ncow o.mJummal m1,0011 ungmpl011! . Am Hexrl 1 19115 :109'926-31 . 146 . Hoberg F . 900 441! F- Schomig A- e, of P oon, 011 of roOenor,, 6y 91191. the 4010pmnil .901i1014, :y StUAy IVAS14lh,tnI Cirrnl4uon 19S"v ,1490 147 . Nafulan 911. plot RE . (loll, v1 InduCUOIl of plom1el-dcdved growth (11101 A