coronary intimal proliferation after balloon injury and ... · “guide for care and use of...

JACC Vol. 20. No. 2

August 1Y92:467-74 467

Restenosis occurs in about one thir of patients undergoing coronary angioplasty (1,2i. Altho gh the mechanism of restenosis is not fully understood, it appears that trauma to

the vessel wall during balloon inflation may trigger t migration of smooth muscle cells from the media to the intima, where they proliferate and produce extracellc!ar

matrix. This intimal hyperplasia may progress, leading to significant lumen narrowing (3,4).

Efforts to prevent restenosis by using various Fharmaco-

logic agents and mechanical devices have been largely

unsuccessful (5-9). Human studies of restenosis are expen- sive and may take years to complete, since they require large numbers of patients to gather meaningful data (IO). In addition, pathologic specimens are seldom available in human studies of restenosis (1 I ,1’2). Thus, an appropriate

aullilal lii&i ‘&tXru tJ be II=+II In screening potential thera- yy_._

From the Andreas Gruentzig Cardiovascular Center, Departments of Medicine (Cardiology) and Radiology and Department of Pathology, Emory Universi:y School of Medicine, Atlanta, Georgia.

Manuscript received October 22, 1991; revised manuscript received February 13, 1992. accepted March IO. 1992.

Address for corresoondence: Spencer B. King III, MD, Andreas Gruentzig Cardiovascular Center, Suite F 604, Emory University Hospital. 1364 Clifton Road, NE, Atlanta. Georgia 30322.

01992 by the American College of Cardiology

pies for restcnosis and may facilitate our understan

the cellular and molecular processes involved in the opment of restenosis.

Although several animal models have been studies of restenosis, the use of swine app

specific advantages (13-15). By injuring porcine coronary

arteries with balloon inflation or placement of a tantalum stent, we have attempted to develop a more precise experimental model of restenosis. This report describes the vascular responses to balloon injury and stent placement in swine and compares the histopathologic findings with those of

human restenosis.

This study conformed to the

guidelines specified in the National institutes of Health

“Guide for Care and Use of Laboratory Ani

was approved by the Animal Care and Use co

omestic swine (n = 2.5) weighing 25 lo 30 kg were

sedated with ketamine (II to I5 mglkgloody weight), intu- bated and ventilated with isoflurane, oxygen and nitrous

468 KARAS ET AL. BALLOON INJURY VERSUS STENTING IN SWINE

JACC Vol. 20. No. 2 August 1992:46?-74

oxide to maintain adequate anesthesia as determined by the absence of the limb withdrawal reff ex. Coronary arteriography Was performed by using an SF guiding catheter intro- duced from the right femoral artery. Vessel diameter was estimated from the arteriograms by using the catheter diameter as a standard. Attempts were made to use balloons and stents whose diameter was approximately 20% t0 30% greater than the baseline arterial diameter. Through the guiding catheter, a tantalum Wiktor stent mounted on a balloon catheter (Medtronic), was advanced to the midportion of the left anterior descending coronary artery In = 16) or the midportion of the left circumflex coronary artery (n = 19). The balloon was inflated twice for 33 s: then the balloon catheter was withdrawn, leaving the stent in place. The balloon catheter was subsequently advanced to the midportion of the unstented vessel and inflated three times to 8 atm for 30 sand withdrawn. After repeat coronary arteriography, all catheters were removed, the cutdown wound was re- paired and the animal was allowed to recover.

After 4 weeks, the pigs underwent repeat coronary arteriography. After euthanasia with an overdose of anesthesia, the heart was rapidly removed. The left coronary artery was perfusion-fixed with 10% buffered formalin at 100 mm Hg pressure for I5 min. The initial arteriograms were used to pinpoint the exact location of the stented and balloon-injured arterial segments, which were dissected free of the heart. In six pigs, the left coronary artery was perfusion fixed with glutaraldehyde for electron microscopic studies performed with previously published methods (16).

The swine were treated with aspirin (325 mglday), dipyr- idamoie (75 mg three times/day) and diltiazem (60 mg three times/day) beginning on the day before the initial study and continuing until the day of induced death. Anticoagulant therapy with heparin (200 NJ/kg intravenously hourly) was admiilistered during both studies. lntracoronary nitroglyc- erin G!oO pg) was given before arteriography. During the initial study, all pigs were given bretylium (5 mg/kg inti-iive- nously hourly) and nifedipine (10 mg sublingually).

Tissue ~~~~~~~tiQm. Balloon-injured arterial segments were embedded in paraffin, sectioned and stained with hematoxyin-eosin, Masson’s trichrome and Verhoeff-van Gieson elastic stains.

After the stented arterial segments were dehydrated and cleared, they were infiltrated and embedded in a combina-

tion of polymethyimethacrylate (PMMA), n-butyi phthalate and benzoyl peroxide. The arteries were then embedded on a prepolymerized layer of PMMA in glass scintillation vials. They were held at room temperature for 24 h and then Placed in a 40°C oven for curing. The resulting blocks, containing the stented arteries, were removed

from the vials, ground and polished. A tungsten-carbide knife was used to obtain 4-&m sections from the middle of the stented arteries. The sections were floated on a 70”~

waterbath to which 275 Bloom Gelatin had been added. The sections were picked up on ezg albumin-coated slides and

flattened with a roller. The slides were stacked and pressed

in a vise for 24 h at 60°C. odiiied hematoxylin-eosin and

e!cstin stains were used to stain the plastic-embedded

stented segments. Injured arterial segments were processed for tran

(n = 3) and scanning (n = 3) electron microsco previously published methods (16). lmmunohisto

staining using a monocional antibody to actin was performed on the injured arteri

dry. All sections were ex-

ysis of the most severely narrowed section of each injure arterial segment was performed by using a morphometri program (Cue-& Olym~us~. The areas within the lumen

(lumen area) and ide the internal el ic lamina (iatimal

area) were measur al thickness of the

neointima and intima. The degree of stenosis was expressed

as residual lumen area or the ratio of lumen area to i~timal area. In normal arteries, these areas are almost equal,

because the intimai thickness is negligible, and the ratio would be I. En contrast, when intinai proliferation has occurred, the area within the lumen is smaller than the area enclosed by the internal elastica and the residual lumen area (ralio of lumen area to ~ntima~ area) is < 1. More extensive intimal proliferation is associated with a smaller residua! lumen area.

lnitiai and final lrlmen diameter and maximal balloon

diameter for each injured segment were measured from arteriograms by an experienced ang~ograpber using digital eicctronic calipers.

St~~istic~~ analysis. Morphometric and arteriographic measurements from balloon-injured and stented vessels were compared by using paired I tests. Data are presented as

mean value -f SD. A p value CO.05 was considered statistically significant.

Sixteen pigs (Group I) underwent stent placement in the left anterior descending coronary artery and balloon injury

of the left circumflex coronary artery. Three of these pigs

died shortly after the initial procedure and necropsy con- firmed thrombosis of the stented arterial segment; the data from these three pigs are not included in this study. In nine

other pigs (Group 2), stent placement in the ieft circumflex coronary artery was attempted; in one of the nine, the stent

could not be successfully deployed and was withdrawn. These nine pigs all underwent balloon inflation in the left

anterior descending coronary artery and survived to the completion of the study. The difference in survival between Group 1 and Group 2 was not statistically significant (p > 0.05).

Because the study was designed to cause vascular injury by stretching with an oversized stent or balloon, arterial

JAW Vol. 20. No. 2 KARAS ET AL. 469 August 19923464-44 BALLOON BNU’JURY VERSUS SGEIJBNG IN SWLNE

esponse to Injury: Comparative istologic Grading Syskm

Stenting -

Grade 0

No hisfopathologic ChiMl!$S

Grade I lntima

EL

Media

Grade II

lntima

IEL

Media

Grade 111

Intima

EL

Media

52 rows of SMC internal to IEL

Focal reduplication

Minimal increase of fibrous tissue

Mild SMC prolifrration (! to 5 row:!

Fragmented or absent

Increase in collagen and MPS

Significant SMC proliferation

Absent

Replacement of (inner ‘/J) by myofibroblastic proliferation

Increased collagen and MPS

Grade IV

Intima

IEL

Media

Marked SMC proliferation,

3% reduction in lumen arca

AbWll

Complete replacement by myofbrobla~trc prolikra~uoa


Adventitia Round cell and histiocytic infiltration

No histopathologic changes

52 rows of SMC internal to EL

Intact

Thinning due to stent wire compression

Mild SMC proliferation (? to 5 TOWS)

Intact

Minimal increase in collagen iinner fi) with mild disorganization

Significant SMC proliferation

Absent at the site of stem wires

Replacement of inner ti by myofibrobli& prolifcriition at 3ilc

of stem wires

Incrce9ed collagen and MPS

Marked SMC proliferation,

?S% reduction in lumen iIWl

Absent at rite of stem wires

Complete replacement by myolibroblastic prohferation at site of

blent wires


Ir&tration by round cells. eosinophils and foreign body giant

cells around stent wires

EL = internal elastic lamina: MPS = mucopolysaccharide; SMC = smooth muscle cell.

segments in which the balloon/artery ratio was rl were eliminated from further analysis. This ratio occurred in five of the balloon-injured segments and in none of the stented segments. Three stented arterial segments and two balloon-injured segments were processed for electron microscopy and were not available for I

an y. All of the stented arterial segme

vealed marked intimal smooth muscle proli%rkm, an increase in extracellular matrix, destruction of the internal elastica and thinning of the media (Tables 1 and 2). These findings were most notable in the region of the stent wires (Fig. 1 and 2). In addition, several of these vessels demonstrated reactive inflammatory infiltrates surrounding the stent wires (Fig. 3). These infiltrates were composed predominantly of lymphocytes, histiocytes and eosinophils, which often extended into the adventitia. Occasionally, multinucleated giant cells were seen adjacent to stent wires, suggesting a foreign body type of reaction.

Table 2. Response to Injury Among Swine After Balloon Inflation and Stent Placement

Pathologic Grade Balloon Injury (n) Stenl (n)

I 2 0

II I 5

111 3 4

IV 9 9

~a~~oo~-~~~~red arterial segmems aiso showed intimal proliferation of smooth muscle cells and thinnin media (Fig. 4 aild 5), although in some vessels thes were iess prominent than those seen in the stented vessels. In addition, the lesions created by balloon i~~at~o~ tended to

an those created by 5tent placement.

igure B. Stented artery. Note medial compression at the site of the 3lent vkes and cxilmferential ii%imal thickening. B = intimal

proliferation; L = lumen; M = media. Hematoxylin-eosin x45,

reduced by 42%.


JACC Vol. 20, No. 2 August ~~2;46~-?4

Figure 2. Stcntcd artery showing marked myofibroblastic intimal proliferation with notable incrc:\se in extracellular matrix. Tne internal elastic lamina appears intact. Abbreviations as in Figure I. Hematoxylin-cosin x22S. rcduccd by 42%.

Inflammatory infiltrates were rarely seen in the balloon-

injured arteries. Transmission electron micrographs of the neointima of

both stented and balloon-injured vessels revealed proliferation of smooth muscle cells of the synthetic phenotype, characterized by large nuclei and abundant endoplasmic reticulum and relatively few contractile elements, sur- rounded by a loose connective tissue matrix. Imrr;unohisto- chemical staining revealed the presence of actin, confirming the smooth muscle origin of these cells.

By 4 weeks after arterial injury with either method, there was regrowth of a confluent, flow-dircctcd endothelial lining with scattered adherent microthrombi (Fig. 6). The histo-

Figure 3, Foreign body granulomatous reaction around stent wire. Hcmatoxylin=eosin x 225, reduced by 4 I %.

gwe a~~oon-insured coronary artery. Note circumferential intimal prolifccrauorr, complete absence of the media and internal elastic lamina involving approximately one half of the arterial circumference and indistinct external lamina. Abbreviations as in Figure I. f&sson’s ‘richrome x45, reduced by 44%.

pathologic features of the lesions in both stented and

balloon-injured vessels were found to be co those seen in human restenosis (Fig. 7).

~~ornet~y. The degree of intimal ~ro~iferat~o~ was sig antly greater in the stented arterial segments than in the vessels injured by balloon inflation (Table 3). Maximal intimal thickness was significantly greater and residual u- men area ~lun~en area/inti al area) was significantly less in the stented vessels. Despite this finding, both lumen and

intimal areas were significantly greater in the stented arteries.

Figure 5. Higher magnification of balloon-injured vessel shows the disruption of both the media and internal elastic lamina (E). Note the loose myofibroblastic intimal proliferation, extending also beneath the media at the previous site of dissection. Abbreviations as in Figure I. Masson’s trichrome x225, reduced by 43%.

JACC Vol. to, No. 2 August 1992:467-74 KARAS ET AL.

BALLOON INSURV VERSUS STENTlNG IN SWINE 441

re 6. Scanning electron micrograph of the lumen surface of an artery 28 days after stent placement. There is confluent linin composed of flow-directed endothelial cells with prominent micro- villi (x 1,950, reduced by 13%).

c analysis. The initial lumen diameter did not tly in stented and balloon-injured arterial

loon/artery ratio, final minimal lumen diameter and percent stenosis were also comparable in the two grou (Table 4).

Various animal models ha een u ntimal hyperplasia can be induced in the rat carotid artery or aorta by endothelial denudation by using a balloon (18). Others (19-21) have used similar balloon injury of the aorta in hypercholesterolemic rabbits as a model to study restenosis. Unfortunately, cholesterol loading in these animals results in lesions that contain predominantly lipid-laden foam cells rather than the smooth muscle cell proliferation seen in human restenotic lesions (13,19). In studies using the rabbit model (21), the degree of stenosis often did not differ significantly between dilated and nondilated segments, suggesting that the progres- sion of lesions may have been due to extreme ~yperl~p~demia rather than to balloon injury. Also, in contrast to angioplasty in humans, the method of injury in these models is primarily

enudation rather stretching of the vessel. The use of swine may

have certain advantages in the study of restenosis (13-15). In pigs, the coronary arteries are easily accessible with present

reduced by 49%.

catheterization and angioplasty techniques. Swine also re- semble humans with respect to coronary circulation (13,22,23), platelet coagulation system (13), spontaneous development of atherosclerosis (13,14,22) and lipoprotein metabolism (23). Swine have been used to study the imme- diate and long-term effects of coronary angioplasty (14).

Table 3. Morphometry of Injured Vascular Segments

Balloon injury Stenting

(n = 15) (n = 18)

Lumen area (LA) (mm*) 2.2 t 0.9 3.1 ?I 0.7 intimal area (IA) (mm21 2.8 -c 0.8 3.7 + 0.4 LAIIA 0.75 2 0.18 0.64 1 9.18 Maximal intimal thickness (mm) 0.4 f 0.3 0.6 ?I 0.3

Data are expressed as mean value 2 SD. All differences between values

after balloon injury and after stenting are significant (p < 0.05).


JACC Vol. 20. No. 2 August 1992:467-74

Table 4. Angiographic Analysis of Injured Vascular Segments

Balloon Injury Stenting

(n = 21) (n = 23) ___-

Initial vessel diameter (mm) 2.8 + 0.6 3.1 + 0.6

Maximal balloon or stent diameter (mm) 3.3 + 0.4” 3.8 + 0.4*

Balloon or stentlartery ratio I.2 + 0.2 I.2 i: 0.2

Minimal vessel diameter (mm) at 4 wk 2.4 k 0.7 2.4 k 0.6

Stenosis (% diameter reduction) 21 f II 23 2 II

*p < 0.05, balloon injury versus stenting. Values are mean value * SD.

Steele et al. (15) found that pigs devehped intimal hyperplasia after carotid artery angioplasty in the absence of elevated cholesterol levels.

Schwartz ct al. (24) recently created proliferative lesions in porcine coronary arteries. Their model involved intraarterial delivery of metal wire coils using grossly oversized balloons inflated to high pressures. Although these lesions resembled those of human restcnosis, the mechanisms involved may be quite different. Later studies by this group (25) suggest that these porcine lesions may result from the organization of large platelet-fibrin thrombi that are formed after severe vascular trauma. Although srri~e investigators favor an important role for thrombus in the genesis of restenosis, large space-occupying clots as described by Schwartz et al. (24) are rarely seen angiographically in patients after percutaneous transluminal coronary angioplasty (26). In addition, the severity of arterial injury and the resulting response may overwhelm the therapeutic effect of any potential treatment, thereby limiting the utility of this model as a screening tool.

Present model: normalipemic domestic swine. In the model described in this report, intimal hyperplasia was created by more modest arterial stretch injury. Balloon/ artery ratio averaged about I .3 for both balloon-injured and stented segments. More moderate degrees of vascular trauma may be associated with less extensive deposition of thrombus, possibly accounting for the relatively high survival rate (8898 seen in our pigs. Neovascularization and hemosiderin deposits, hallmarks of organizing thrombus, were not prominent features of these lesions.

Although histologically similar, these porcine lesions differed from those of human restenosis in that they devel- oped in previously normal vessels in the absence of atherosclerotic plaque. However, pathologic studies, indicate that the intimal smooth muscle proliferation seen in human restenosis is associated predominantly with the nonathero- matOUs portion of the vessel wall (27). Although the absence of atherosclerosis may lead to mechanical properties that are different from those of human restenotic vessels, the histologic similarities should permit the use of our porcine model to identify potential inhibitors of restenosis.

We used domestic swine to create our model of restenosis. Some groups (13,14,22,23) have advocated the use of cholesterol-loaded or other special strains of swine in their studies. Unfortunately, these animals are costly and scarce.

As Schwartz et al. (24) have arg restenosis must employ animals that are inexp available and easy to use and that develop fairly short period of time. Normoli~~mic d appear to be superior to the special varieties in all these respects.

g. In this study, we com- nse to two types of vascular injury: stenting

and balloon injury. We assessed the degree of this reaction using a pathologic grading syste that is based on changes in the various layers of the vessel wall. It appeal-s that the pathologic response induced by stenting was more uniformly severe than that see sections from stented art inflammatory reaction, po stituting the stent wires. This response may not be specific to tantalum stents. In pilot studies perfo~~~ed by our group (unpublished data), identical inflammatory reactions were observed in some animals after placement of stents made of stainless steel.

This model involves the de~loyme~~t of a bulloon- expnnduble stent. Therefore, the intimal reaction observed in stented vessels may have been caused, in part, by the balloon inflation necessary to deploy the stent. However, the stented vessels had significantly greater intimal proliferation than did the balloon-injured segments, suggesting stent created a greater stimulus for proliferation. In self-expanding stents, which do not require balloon delivery, alJo have been associated with intimal proliferation (8).

By embedding the stented segments in plastic and sectioning with a tungsten carbide blade, intact, high cross sections could be obtained for morphometr analysis suggested a greater proliferative response to stenting as compared with balloon injury, as demonstrated by significant differences in maximal intimal thickness and residual lumen area (lumen arealintimal a

Causes of greater lumen area in sten injured arteries. Although stenting w more smooth muscle proliferation than the final lumen area was greater in the stented vessels. Althor.gh this result might be expected if the stented vessels were initially larger than the balloon-injured vessels, the arteriographic data indicated that the stented and ballooa- injured vessels had a similar diameter before dilation. An understanding of the mechanisms by which stents and balloon inflation lead to vascular injury may resolve this appar- ent contradiction. Mechanical stretching associated with balloon inflation is followed by some degree of elastic recoil of the vessel once the balloon is deflated (28,29). However, a stent exerts continuous force on the vessel wall, maintain- ing a larger lumen than the surrounding normal vessel (8,16). Even though stenting is associated with more intimal proliferation, this proliferation occurs in vessels that, by the nature of the technique, are stretched beyond their original dimensions. Despite more pronounced intimal proliferation in stented segments, the final lumen area may be significantly

JACC Vol. 20, No. 2

coronary angioplasty: report from the Registry of the National Heart. load Institute. Am 3 Cardiot 1982;49:201 P-20.

3. Austin GE, Ratliff NB. Hollma~ 3. Tabei S. phihips DF. lntimal proliferation of smooth muscle cells as an explanation for recurrent coronary iWkry stenosis afler percutaneous lranslumiual coronary angioplasty. 3 Am Colt Cardiol P985:6:369-75.

4. Clowes AW, Schwartz SM. Signiticaace of quiescent smoorh muscle

ration in the injured rat carotid artery. Circ Res 1985;56:139-45.

5. nathan ES, Schwartz JS, Taylor L, et al. Aspirin and dipyridamole in prevention of acute coronary thrombosis complicating coronary

angioplasty. Circulation 39X7:76:325-34.

6. Thornton MA, Gruentzig AR, Hollman J, King SB, Douglas 3s. Couma- din and aspirin in the prevention of recurrence after transtuminal coronary angioplasty: a randomized study. Circulation 1984;69:721-7.

7. Whitworth HB, Roubin GS, ollman J, et al. Effect of nifedipine on recurrent stenosis after percuta:.eois transluminal coronary angioplasty.

ut a~~io~ra~h~~ r~ste~osis with .estenosis. An ment revealed a

9. US Directional Coronary Atherectomy Investigator Group: Restenosis following directional :oronery atherectomy in a multicenter experience. Circnlatioo t9YO;8Zt5u~~l 111):111-679.

IO.

II.

12.

a significantly larger lumen 13.

vation. Eccentric lesions may be missed on routine angiog- raphy, leading to exaggeration of the dimensions of a liar segment. This would be less likely to occur analyzing the stented segments, which were more concentric in morphology.

sious. Lesions similar to uman restenosis can be produced in porcine coronary arteries by stent placement and by balloon inflation. Intraarterial stent placement appears to be associated wi significantly more intimal smooth muscle proliferation n is found in arteries after balloon dilation. Inflammato filtrates s~rrow~diog stem wires are often seen in the stented vessels, suggesting a foreign body reaction. This model of the normolipemic swine may prove to be useful in screening potential inhibitors of restenosis.

14.

1.5.

16.

17.

18.

19.

20.

21.

We thank Cynthia Barnnowski for devising a better method of embedding. sectioning and staming stented arterial segments, Robert Apkarian, MS for his expertise with scanning and transmission electron microscopy and Eugene Malveaux for valuable technical support.

22.

23.

S 24.

1. Leimgruber PP, Roubin GS, Hollman J, et al. Restenosis after successful coronary angioplasty in patients with single-vessel disease. Circulation 1986;73:710-7.

25.

2. Kent KM, Bentivoglio LG, Block PC, et al. Percutaneous transluminal

KARAS~-AK_. BALLQONBNJURY VERSUSSBENFIRiGII\!SWlNE

473

Beatt KS, et al. Angiographic follow-up after placement of a self-expanding coronary-artery stem. N Engl J Mrd lYY1:324:13-7.

. Blackshear JL, O’Callaghan WG, Califf RM, Medical approaches 1.) prevention Of restenosis after coronary angioplasry. 9 Am Colt Cardiot 1987;9:834-8.

oubin GS. Douglas JS, Sbaw RE, Stertzer SM. In-hospital mortality after acute closure after coronary angioptasty: analysis of risk factors from 8.207 procedures. J Am Coll Cardiol 1988:11:21 l-6.

Ueda W Becker AE, Fujimoto T. Pathological changes induced by repeated percutaneous transluminal coronary angioplasty. Br Heart J 1987;58:635-43.

Fuster V. Fass DN, Kaye MP, Josa Zinmeister A, Bowie E. Arteriosclerosis in not mal and von Wille nd pigs: long-term prospec- tive study and aortic transplantation study. Circ Res 1982;51:587-93.

Weiner BH, Gckene IS, Jarmolych J, Fritz KE, Daoud AS. Comparison of pathologic and angiographic findings in a porcine preparation of athero rosis. Circulation 1985:72: 1081-6.

Steele . ChrsebroJH. Stanson AW. et al. Balloon angioplasty: ~.rrural history of the ;4thophysiologic response IO injury in a pig mode!. Circ Res 1985;57:105-12.

Robinson KA. Roubin GS. Siegel RJ, Black AJ, Apkarian RP, King SD Bl. Intra-arterial stenting in the atherosclerotic rabbit. Circulation 1988; 78646-53.

Giorno R. A comparison of two immunoperoxidase staining meihods based on avidin-biotin interaction, Diagn Immunol 1984;2: 1%6.

Lundergan C. Feogh ML, Va:gas R, et al. Inhibition of myointimal proliferation of the rat carotid by the peptides, argiopeptin and 5lM 23034. Atberoscleeosis 1989;80:49-55.

Block PC, Baughman KL. Pasternak RC. Fallon JT. Transluminal angioplasty: correlation of morphologic and angiographic findings in an experimental model. Circulation 1980:61:778-85.

Faxon DP, Sanbom TA. Haudenschild CC, Ryan TJ. Effect of antiplatelet therapy on restenosis after experimental angioplasty. Am J Cardiol 1984;53:72C -6C. Faxon DP. Sanbom TA, Weber VJ, et al. Restenosis following transluminal angioplasty in experimental atherosclerosis. Arteriosllerosis l984;4: 189-95.

French JE. Jennings MA, Florey HW. Morphological studies on atherosclerosis in swine. Ann NY Acad Sci 1965;1?7:780-99.

Fritz KE, Daoud AS, Augustyn JM, Jarmolych 9. ~o~hological and biochemical differences among grosslydefined types of swine aortic atherosclerosis induced by a combination of injur, and atherogenic diet. Exp Mol Pathol 1980;32:61-72. Schwartz RS, Murphy JG, Edwards WD, Camrud AR, VliClStra RE, Holmes D?!. Restenosis after balloon angioplasty: practical prohferative model in porcine coronary arteries. Circulation IYYO;82:2190-280.

Schwartz RS, Edwards WD, Murphv JG. Camrud AR, Holmes DR. Restenosis develops in four stages: serial histologic studies in a coronary injury model, J Am Coil Cardiol lY91;17:52A.

474 KARAS ET AL. SACQ’ Vol. 40. No. 2 BALLOON INJURY VERSUS STENTING ilN SWINE AugusP 199?:46?-74

26. Ellis SG, Roubin GS, King SB 111, et al. Angiographic and clinical predictors of acute closure after native vessel coronary angioplasty. Circulation 1988:77:3X&-9.

27. Gravanis MB. Roubin GS. Histopathologic phenomena at the site of percutaneous transluminal coronary angioplasty. Human Pathol 1989:X 471-85.

28. Serruys PW. Luijten HE. Beat1 KJ. et al. lncidcncc of restenosis after successful coronary angioplajty: a timerclated phenomenon. A quantt

tative angiographic study in 342 conwcutive patients at I, 2, 3 ;~nd 4 months. Circulation 1988:77:361-71.

29. Nvbuyoshi M. Kimura T, Nosaka H. et al. estenosis after successful percutaneous transluminal coronary angioplasty: serial angdographic follow-up of 229 patients. J Am Colt Cardiot 19X8:12:6&-‘3.

coronary intimal proliferation after balloon injury and ... · “guide for care and use of...

Documents