- Email: [email protected]
Angiographic predictors of recurrence of restenosis after Wiktor stent implantation in native coronary arteries
Angiographic Predictors of Recurrence of Restenosis After Wiktor Stent Implantation in Native Coronary Arteries Peter DeJaegere, MD, Patrick W. Serruys, MD, Michel Bertrand, MD, Volker Wiegand, MD, Jean Francois Marquis, MD, Matthias Vrolicx, MD, Jan Piessens, MD, Bernard Valeix, MD, Gisbert Kober, MD, Hans Bonnier, MD, Wolfgang Rutsch, MD, and Rainer Uebis, MD lntraconmaystenti~hasbeMproposed~an ~unct to balioon angioplasty to improve the im mediate Md kmgteml reslllts However, late lunlimdmwrowinghasbeen~followingtheim pbtaUonofavatietyofstenkOneofthestu& iesunwhtedwiththeWlktorstentisa~ tive regmty dedgned to evaluate the feasibiGty, safetyandeffkacyofekctivestentimpkmtation inpathb5withdocumentedrestemwSsofana the ammmuy autery. To Wentify mgiogrqGdc v& aMeapredlctingrecuwenceofreatenosls,the
~ofthenrst91patSentswithsulid stent inrplantation and without clinical evidence of (sub)acub tlmmbuc dent occlusion W8lWrndyzedWiththS~ASSisted~
gr@ic Analysis System using automaW edge was44%by ddecuon.lheiddenceofpatiemtmd4S%bystemtaccad~totheO.72 nmt!dteuw,ana2O%bypatientand29o/oby stentaccordi~tothe5O%diameterstenosis& lwstemdsforseveral Mgb tedon.Theriskfor m varlabler was deWmined using an univafiateanalysisandisexpreaQedasoddsratsowith co#ahceintewal.llleonlystati~ ~PpdktorOf-WNthe
reJatlvegainwhenitexceededOASusingthe 0.72 mm aftdon (odds ratio 2.7,95% confidence interval lA-6.4). Fu?thMnmu, thedationbe tween the dative gain (increase in minimal hunk naldbneternufnudixedtovesselsize)asangb grqshic index of vessei wall iduy and relative loss (decrease in minimal luminal diameter non matized to vessel size) as index of neointimal From the Catheterization Laboratory and Laboratory for Quantitative Angiographic Analysis (Cardialysis), Thoraxcenter, Rotterdam, the Netherlands: Department of Cardiology, Hopital Cardiologique, Lille, France;Department of Cardiology, Georg August UnivemitZt Gottingen, Germany; Department of Cardiology, University of Ottawa Heart Institute, Ottawa, Ontario, Canada;Departmentof Cardiology, University Hospital Gasthuisberg,Leuven, Belgium; Department of Cardiology, Clinique la Casamance,Marseille, France; Department of Cardiology, Klinikum der J.W. Goethe Universimt Frankfurt, Germany; Departmentof Cardiology, CatharinaHospital, Eindhoven, the Netherlands; Department of Cardiology, University Hospital Berlin, Germany; Department of Cardiology, Medical Clinic I RWTH Aachen, Germany. Manuscript received October 14, 1992; revised manuscript received and acceptedFebruary 22, 1993. Addressfor reprints: Patrick W. Serruys,MD, CatheterizationLaboratory and Laboratory for Quantitative Angiographic Analysis (Cardialysis), Thoraxcenter, ErasmusUniversity, P.O. Box 1738,300ODR Rotterdam,the Netherlands.
thickening was adyzed using a linear re#essh analysiaAPesrsonproduCtiMmMteorrelation . codhalt of 0.38 (p *o.OOl) was found. when usingthecategwical~toaddressre steno4bis,thereisaninaessed riskforrecunwnt restenosis when the mlative gain exceeds 0.48. TheCOlltillUOUSapproachU-thiScOrr
ceptbyindicatingaweakbutposRiverehtionbs tweenthefe4ativegainandrelativeloss. (Am J Cmliol1~72:16+170)
P
ercutaneoustransluminaJcoronary balloon angioplasty is now widely acceptedas a safe and effective treatmentin selectedpatients with obstructive coronary artery disease.’ However, acute vessel closure and late restenosis are inherent to balloon angioplasty and continue to compromiseits efficacy.2,3Although the exact pathophysiologic mechanism(s)and factors responsible for restenosis are largely unknown, some clinical and angiographic variables have been identified as potential risk factors.4 There is, furthermore, some evidence that repeat angioplasty of a previously dilated lesion in a native coronary artery is associated with a higher incidence of restenosiss9 The Wiktor stent has been used to treat such patients. The incidence of recurrence of restenosisfollowing Wiktor stent implantation in the first 50 consecutive patients has recently been published.tO The purpose of this study was to try to identify angiographic predictors of recurrence of restenosis after Wiktor stent implantation. This may not only have important clinical consequences,but may also broaden our understanding of the pathophysiology of restenosis. MRHODS PatienW Between January 1990 and May 1992, Wiktor stent implantation was attemptedin 193 patients becauseof recurrence of angina due to restenosisof a native coronary artery lesion after previous balloon angioplasty. The study population consisted of 91 of the 109 consecutivepatients with successfulstent implantation without clinical evidence of thrombotic stent occlusion during hospital stay and in whom the scheduledfollow-up angiography was completed at 5.7 Ifr 1.9 months (mean + SD) (Figure 1). Eighty-one patients were men (89%); the mean age (& SD) was 58 f 11 years. A first restenosiswas documented in 50 patients (55%), a second in 32 patients PREDICTORSOF RESTENOSISAFTERSTENTING 165
I
7
TABLE I Findings at Quantitative Angiography After Wiktor Stent Implantation (n = 91 patients)
Reference diameter (mm) Minimal luminal diameter (mm) Diameter stenosis (%)
Before PTCA
After PTCA
2.85 k 0.49
2.80 2 0.48 1.77 5 0.36 35211
1.13 -c 0.36
60 2 10
After Stent Implantation
At Follow-Up
3.02 2 0.42 2.45 + 0.34 19 2 7
2.98 f 0.47 1.73 + 0.67 42 2 21
All parameters are expressedas mean k SD. All changes were highly significant (p
(35%), a third in 8 patients (9%) and a fourth in one other patient (1%). In all patients, a single Wiktor stent was implanted except in one, in whom a single stent was implanted in both the left anterior descending and the right coronary artery. The target vessel was the left anterior descendingartery in 48 patients (52%), the right coronary artery in 31 patients (34%), and the left circumflex in 12 patients (13%). In another patient (l%), a single Wiktor stent was implanted in the left main stem. The size of the stent used was 3.4 + 0.4 mm (mean f SD): a 3.0 mm device in 38 patients (41%), a 3.5 mm device in 40 patients (44%) and a 4.0 mm device in 14 patients (15%). The stent is describedin detail elsewhere as well as the associateddrug protocoL1t wk
vmWks
and d&WKom of restene
sisr Based on the quantitative angiographic data, multiple variables were identified and recorded for each lesion. These variables were of a priori clinical interest on the basis of previously published reports on balloon angioplasty and intracoronaty stenting and of a more fundamental scientific interest based on experimental studies on stent implantation in animals.4J2-16The following continuous variables were selected: obstruction diameter before and immediately after stent implantation, baseline reference diameter and diameter stenosis, length of target lesion, plaque area and the relative gain defined by the increase in minimal luminal diameter immediately after stent implantation normalized to the 109 CONSECUTIVE PATIENTS 13 (SUBIACUTE OCCLUSIONS (6 CABG/ 8 re-PTCAI thrombolysis) i DEATH (HEART FAILURE) AT 3 months 3 REFUSALS
vessel size. In addition to these continuous variables, one discrete variable (left anterior descending artery) was selected. All coronary angiograms were analyzed with the computer-assistedCardiovascular Angiography Analysis System.Its principles and deft&ions of angiographic variables has been describedin a previous study on Wiktor stent implantation.11Sincethe primary objective of this study was to identify angiographic variables predicting restenosis,the process of restenosis was dichotomized and delined according to the 0.72 mm and 50% diameter stenosiscriteria.17 staU&ks A relative risk analysis was performed for the aforementionedangiographic variables. To avoid an arbitrary subdivision of data in the continuous variables, the median value was chosen as cutoff point. The selection of this value as cutoff point has the advantage of being consistent for all values, and thus avoids any bias in the selection of subgroupsthat might be undertaken to emphasize a particular point. All values are expressedas mean f SD. The risk for restenosisfor each parameteraccording to the 0.72 mm and 50% diameter stenosis criteria was determined by using an univariate analysis and is expressed as odds ratio with corresponding 95% confidence interval. An odds ratio of 1.0 for a particular variable implies that the presenceof that variable posesno additional risk for restenosis,an odds ratio >l.O or cl.0 implies additional or reduction in risk. Furthermore, since experimental animal data indicate that there is a relation between the severity of vessel wall injury and the extent of subsequent neointimal thickening, the relation between relative gain (as index of vessel wall injury) and relative loss (as index of neointimal thickening) was studied using a regressionanalysis. The relative gain is previously defined. The relative loss is the difference between the minimal luminal diameter immediately after stent implantation and at follow-up normalized to the vessel size.
1 MISSING
F91 CONSECUTIVE PATIENTS WITH ANGlOGRAPHIC FOLLOW-UP* *5.7 t 1.9 months a@ogr@lc PKwRELFlewdle@ml~ ptltknk~ecuteoreub ulelmt109ecubecclueiaecccuwsdbl~patlenb(~).Dftberemairr wllbeut illg98p0ti9llt!SWhOW~~ denceof(sub)acuteocclusion,9lundewantrepeat~ r&yat5.7~l.9(mean*sD)mcetbsefterstentinrpbntb ticn.cAB6= eoronayarteyW--A=repeat perwtaneoustradumlnalcoronmy~. 166
THE AMERICAN JOURNAL OF CARDIOLOGY
followap dinkd
in e*
VOLUME 72
RESULTS hcidmce d restamh The changesin stenosisgeometry are summarized in Table I. The incidence of restenosiswas 44% (40 of 91 patients) according to the 0.72 mm criterion and 30% (27 of 91 patients) according to the 50% diameter stenosis criterion. When the stent is used as a unit, the incidence of restenosiswas 45% (41 of the 92 stents)and 29% (27 of the 92 sterns), respectively (Figure 2). WC predictors: The relative risk and 95% confidenceintervals for each variable using either of the JULY 15,1993
TABLE II Angiographic Predictors of Restenosis After Wiktor Stent Implantation RestenosisCriteria 0.72 mm
Angiographic Variable Relative gain (mm) Minimal luminal diameter after (mm) Minimal luminal diameter before (mm)
20.48 22.38 11.08 2 6.48 ~2.77 2 7.47 261
Lesion length (mm) Reference diameter before (mm)
Plaque area (mm*) Diameter stenosis before (%) Left anterior descendingartery
2 criteria are listed in Table II. The only statistically signilicant predictor of recurrence of restenosis according to the 0.72 mm criterion was the relative gain when it exceeded0.48 (odds ratio 2.7, 95% confidence interval 1.1-6.4). Related to the relative gain is the obstruction diameter before and after stent implantation. Although their odds ratio exceeds 1.0, their corresponding 95% confidence interval precludes any ti conclusion. The same holds for lesion length, reference diameter and plaque area. The left anterior descendingartery was not identified as a risk factor for recurrence of restenosis. When using the 50% diameter stenosiscriterion, no angiographic predictor of restenosiswas evident. The number of balloon angioplasties performed before stent implantation did not influence the risk for sub-
4
MLD F/U (mm)
jc CHRON OCCL: 4(4%) 1 I
? 50%:
//
27(29%Y
/
50% Diameter Stenosis Criterion
Criterion
,
2.7 1.8 1.6 1.4 1.3 1.3 1.0 0.7
(1.1-6.4) (0.8-4.2) (0.7-3.7) (0.6-3.2) (0.6-2.9) (0.6-2.9) (0.4-2.3) (0.3-1.7)
1.1 0.5 2.7 1.8 1.4 1.1 1.3 1.1
(0.4-2.7) (0.2-1.4) (1.0-7.2) (0.7-4.6) (0.5-3.4) (0.4-2.7) (0.5-3.2) (0.5-2.8)
sequentrestenosisafter stenting. The odds for restenosis in caseof stent implantation for a second,third or fourth restenosis compared with stent implantation for a lirst restenosis was 0.8 (95% contidence interval 0.4-1.8) according to the 0.72 mm criterion and was 1.5 (95% cotidence 0.6-3.7) according to the 50% diameter stenosiscriterion. Regression analysis: The relation between relative gain, as index of vessel wall injury, and relative loss, as index of late neointimal hyperplasia as vessel wall responseto injury, is shown in Figure 3. A Pearsonproduct-moment correlation coefficient of 0.38 was found (p
, .4 Relative loss RL= .61 RG -.06 R= .38 (p=.OOl)
1.2l-
OO
0.81
0
0.8
0
I
1
2
3
4
MLD POST-STENT(mm)
0.72 mm LONQ
0
dSIlC8Of-S~llgtOthSSO%-~S
cdtdonwas29%(27ofthe92stents).Ofth8se,4stents oc&dedatfol~p,~shownonthex W-M dS.-OCCL=ClStWlkOCdUdan.
I
0 80
0.2 0 -0.2
fohwp(F/U)aftewWihtorstentbnphbth.7hediame terofeachse#nentimmdbMyaftershtimplmRa&nis cEmnderatfel~p.Thelinesoneadl mcrcdndsidaafthsnnaafmnfRy(~)~twisetha ViwiabilRy (OS% coddmca bntarval) of duplkata nlsasucB mants (a chmge of to.72 mm). A si@Mcmt tbcmass in nlinimalundndacxod~!tothe~O.72nuncdteuh wasfoundin4leftha92stwtsi@mIted(4s%).The~
00
0.4
TERM WU?IARILITY
FlGuRE2.chmgesinthemhimdlllmind-(MU))at
OOQ
OO
-0.4
0
I 0
0.2
0
0.4
0.8
0.8
1
1.2
Relative gain FlwRE1.maphkasplayoftherelatbnbe4waenthsrela tbe@h(RQ=hcfeuxwinminimaIluminald&wmterimmediinqhMthmwnlalizsdtothsvessslwall) w-mdrela#wb(RL=decrcwro inminlmalhdnddlmWer nomdkedtethevessalwdl).A~eOnear dfdkrwP ramillnwasfoundwmla-(R) of 0.38 (p<0.001)wilha~of0.81adani~onthey axis of 4.00. PREDICTORS OF RESTENOSIS AFTER STENTING
167
I
-PtO.Os----,
..
.
I
Schwartz Circulation
et al.
I
Nobuyoshi
‘90
I
‘91
Therefore, late loss or restenosiswas due to ingrowth of tissue into the lumen of the stented segment.t8 The lack of randomized studies precludes any conclusion as to whether stent implantation in this subsetof patients will reduce the incidence of subsequentrestenosis. Moreover, the exact incidence of recurrent restenosis after repeat balloon angioplasty is not known. There is some evidence that it increaseswith the number of repeat angioplasties. A second restenosis has been reported to occur in 25 and 34%, but amounts to 39 and 40% after a third or fourth angioplasty,respectively.~9 These data should be interpreted with caution considering the differencebetweenthe study populations, the dif-
-0.2 -0.4
O0
0.5
0.6 0.7 0.6 0.9
1
1.1 1.2 1.3 1.4
stent/artery
1.5 1.6
ratio
ApesRivelhemmlatienwasfeundwRha~cadi fhnt(R)[email protected])wRhadopeof0.6OtmnJan
ilMNqRof9~. 166
THE AMERICANJOURNALOF CARDIOLOGY VOLUME72
Beatt
et al.
J.Am.Coll.Cardiol.
I
et al.
J.Am.Coll.Cardiol.
‘92
ference in definition of restenosis,and in the time and completenessof follow-up angiography. In the present study in which stemswere implanted becauseof restenosis, the risk for recurrent restenosisalter stenting did not differ between patients with a first restenosis and patients with a second, third or fourth restenosis.Two other studies failed to show a statistically sign&ant difference in the restenosis rate for primary or de novo lesions comparedwith lesions that had undergoneprevious balloon angioplasty.12,13However, subgroup analysis of patients in whom a single Palmaz-Schatzstent was implanted showed a restenosis rate of 13% compared with 36% for patients with previous restenosis.*3 The development of restenosisremains incompletely understood. Histologic data have shown that any injury to the vessel wall, whatever its nature, will invariably be associatedwith neointimal hyperplasia as a nonspecilic tissue response leading to restenosis when excessive.14J9-23It is of both scientitic and utmost clinical importance to know and understandits developmentand to know the factors controlling the extent of the vessel wall response following injury. Experimental animal studies, postmortem pathologic observations in humans and angiographic studies have shown that the extent of neointimal hyperplasia is proportional to the degree of injury applied to the vessel wall (Figure 4).t4-16J4This is in accord with earlier work in the porcine model, describing a relation between the degree of injury and early platelet deposition.25~26 In the presentstudy,the relative gain was used as angiographic correlate of vessel wall injury. This parameterwas found to be the only statisticahy signilicant predictor of recurrenceof restenosis when using the 0.72 mm criterion. The drawback of the . _._ latter is, lirst, that the process of restenosis is dichotomized (present/not present) and second, that although
JULY 15,1993
statistically justifiable, cutoff points are used for the angiogmphic variables under investigation. However, the continuous approach underscoresthese Iindings by indicating a positive linear relation between the relative gain and relative loss. These observations were still valid when using absolute measurements.However, the correlation between the absolute gain and absolute loss was somewhatweaker (r = 0.21, p = 0.05). This is explained by the fact that the use of absolutevalues does not relate these changesto the vessel size. Although we found no difference in absolute loss between small and large vessels, an identical loss in minimal luminal diameter represents a larger relative loss in a small vessel than in a large vessel, and vice versa. This conlirms other studies using a categorical approach to detine restenosis after stenting, that small vessels are not, per se, more prone to restenosisthan large vessels.12J3,27 This concept, describing a proportional relation between vessel wall injury and neointimal thickening, has been reported in other studies using quantitative coronary angiography. Beatt et ali6 found that restenosis (0.72 mm criterion) was signilicantly correlated with both a greater improvement in obstruction diameter and a larger absolute dimension aher balloon angioplasty. Furthermore, in a recent study,28the absolute change in minimal luminal diameter was reported to be the greatest single determinant of late luminal narrowing after balloon angioplasty. In another rep~rt?~ it was found that interventions achieving a “bigger” lumen provoke a concomitantly larger relative loss, so that the ultimate end point of various treatment methods is similar. All these findings carry potential far-reaching clinical implications. Clearly the greater the improvement in minimal luminal diameter achieved by intervention, the greater the magnitude of subsequentluminal narrowing will be. Unfortunately, what cannot be extrapolatedfrom the data is how much damage the clinician may intlict on the vessel wall. On the one hand, a suboptimal angiographic result is associatedwith a higher risk of subacute occlusion due to rheologic factors and platelet deposition and a higher need for repeat balloon angioplasty but, on the other hand, improvement of the initial result may be at the price of more extensive late neointimal thickening.i4i6*30,31This indistinctness can be circumvented by properly matching the balloon size or device with the vessel wall using on-line quantitative coronary angiography. This is underscoredby the data displayed in Figure 5 indicating that the more the stent is oversized, the greater the loss in minimal luminal diameter will be. So far, there are 2 other angiographic studies that have attemptedto identify risk factors for restenosisor recurrenceof restenosis,and they found, as in the present study, that the vessel size and severity of stenosis, expressedas diameter stenosisand length of the lesion, not to be associatedwith an increasedrisk for restenosis.r2J3As mentioned, the indication (primary or secondary stenting) was not associatedwith an increasedrisk for restenosis.12J3 In accordancewith these 2 studies on stenting and 1 other study on balloon angioplasty, the target vessel was not associatedwith an increased risk for (recurrent) restenosis.12J3.32
SW limitations: The study limitations are essentially twofold. First, the precision of relative gain as an angiographic index of vessel wall injury and relative loss as an index of neointimal hyperplasia has not been studied. The coronary angiogram is a 2dirnensional echocardiogram describing the changesin stenosisgeometry but not the nature or extent of the pathology explaining these angiographic changes.Therefore, the measurement of the minimal luminal diameter, and consequently the relative gain and loss, are subject to potential imprecision. Second, the effect of other clinical, procedural and lesion-related characteristicson the development of late neointimal hyperplasia has not been considered. It is conceivable that if these characteristicsplay a role in the pathogenesisof neointimal thickening, they may not be uniformly distributed over the study population. ACkllOV . We thank Marie-Angele Morel for performing the quantitative analysis and for her assistancein the databasemanagementand statistical analysis.
1. De&e K, Holubkov R, Kelsey S, Bourassa M, Williams D, Holmes D, Dorms G, Faxon D, Myler R, Kent K, Cowley M, Cannon R, Robertson T, and the coinvestigators of the National Heart, Lung, and Blood Institute’s Petcutaoeous Tramhmid Coronary Angioplasty Registry. One-year follow-up results of the 19851986 National Heart, Lung, and Blood Imtihtte’s PI%A Regisby. Circulation 1989;SO: 421428. 2. de Feyter PJ, de Jaegere Pm, Murphy Es, Senuys PW. Abrupt coronary artery occlusion during penxtaoexx~s tmnsluminal coronary angioplasty. Am Heart .I 1992; 123:163>1642. 9. Sermys PW, Luijten HE, Beat KJ, Geuskebs R, de Feytez PJ, van den Brand M, Reiber JHC, ten Katen HJ, van Es GA, Hugenholtz PG. Incidence of restenosis after successful coronary angioplasty: a time-related phenomenon. Circulution 1988;77:361-371. 4. Botrassa M, Lesperance J, Eastwood C, Schwutz L, Cote G, Kazim F, Hudon G. clinical, physiologic, anatomic and procedural factors predictive of resteoosis after percutaneous traosluminal coronary angioplasty. J Am Coil Cardiol 1991; 18: 36M76. 1. Williams DO, Gtuentzig AR, Kent KM, Detre KM, Kelsey S, To T. Efficacy of repeat percutaneous tmnsluminal coronary angioplasty for coronary testenosis. Am J Cardiol 1984,53:32C-35C. 8. Meier B, King SB III, Gmentzig AR, Douglas JS, Hollman I, Ischinger T, Galao K, Tanker&y R. Repeat coronary angioplasty. J Am Co/l Car&l 1984;4:463466. 7. Black AJ, Anderson VH, Roubii GS, Powelson SW, Douglas JS, Kiig SB III. Repeat coronary angioplasty: conelates of a second restenosis. J Am Co/l Cardiol 1988;11:714-718. .B. Teintein F’S, Hoover CA, Ligon RW, Giorgi LV, Rutherford BD, McConahay DR. Johnson WL, Hart&r GO. Repeat coronary angioplasty: efficacy of a thii angioplasty for a second restenosis. J Am Co/l Curdiol 1989; 13:291-2%. 9. Quigley PI, Hlatky MA, Hinoham T, Rendall DS, Perez JA, Phillips HR. Calii RM, Stack RS. Repeat percutaneous tmnsluminal coronary aogioplasty and pmiictars of recunent restenosis. Am J Cardiol 1989;63:4w13. 10. de Jaegete P, Senuys PW, Bertrand M, Wiegaod V, K&r G, Marquis JF, Valeix B, Uebis R, P&ens J. Wiitor stent implantation in patients with restenw sis following balloon aogioplasty of a native coronary artery. Am J Cardiol 1992;69: 59M2. 11. Sermys PW, de Jaegere P, Bertrand M, K&r G, Marquis JF, Piessens J, U&ii R, Valeix B, Wiegand V. Morphologic change in coronary artery stenosis with the medtmnic Wiior stent. Initial results from the core laboratory for quantitative aogiogmphy. Cathet Cardiovasc Diagn 1991;14:237-245. l2. Strauss BH, Semtys PW, de Scheetder IK, Tijssen JG, Bertrand ME, Puel J, Meier B, Kaufmann U, Stauffer JC, Rickards AF, Sigwart U. Relative risk analysis of angiograpbic predictors of restenosis within the coronary wallstent. Circulation 1991;84:163&1643. 12. Fischman DL, Savage MI’, Ellis SG, Schatz RA, Leon MB, Baim D, Goldberg S. Restenosis after Palmaz-Schatz Stent implantation. In: Senuys PW, Strauss BH, Kiig SB III, eds. Restenosis After Intervention with New Mechanical Devices. Boston: Kluwer Academic Publishers, 1992; 191-205. i4. Schwartz RS, Mulphy JG, Edwards WD, Cammd AR, Vlietstm RE, Holmes DR. Restenosis after balloon angioplasty. A practical proliferative model in porcine coronary arteries. Circulation 1990$32:219&2200. $5. Nobuyoshi M, Kimura T, Ohishi H, Horiuchi H, Nosaka H, Hamasaki N, Yokoi H, Kim K. Restenosis after percutaneous transluminal comnary angioplasty: patbo
PREDICTORSOF RESTENOSISAFTERSTENTING 169
logic observationsin 20 patients.J Am Co// Cardiol 1991;17:433-439. 16. Beatt KJ, SenuysPW, Luijten HE, RensingBl, SwyapranataH, de Feyter PJ, van den Brand M, Lawman GJ, RoelandtI. Restenosisafter commuy angioplasty: the paradoxof increasedlumendiameterandrestenosis.J Am CON Cordial 199219:
24. SarembockIJ, LaVeau PJ, Sigal SL, Tiimis 1, SussmanJ, HaudenschildC, M. Infiuenceof inflation pressureand balloon size on the development of intimal hyperplasiaafter balloon angioplasty.A study in the atheroscleroticrat+ bit. Circulation 1989;80:102’+1040. 25a266. 25. SteelePM, ChesebmJH, StansonAW, Holmes DR. Dewanjec MK, Bodiman of short-, medium-,and L, FusterV. Balloon angioplasty:natural bistmy of the pathophysiologicresponse 17. Reiber JHC, SenuysPW, Kooijman CJ. Assessment to injury in a pig model. Circ Res 1985;57:105-112. long-term variations in arterial dimensionsfrom computer-assisted quantitationof 26. Wilentz JR, SanbomTA, HaudenschildCC, Valeri CR, Ryan TJ, Faxon DP. coronary cineangiograms. Circulation 1985;71:280-288. 18. de JaegereP, Bertrand M, Wiegand V, Marquis JF, Vrolicx M, P&ens J, Plateletaccumulationin experimentalangioplasty:time courseand relation to vasValeix B, Kober G, Bonnier H, RutschW, Uebis R. Recoil following Wiitor stent cular injury. Circulation 1987;75:636-642. implantation.Resultsof the EuropeanRegistry and of the core lab for quantitative 27. CarmzzaJ, SchatzR, Kuntz R, FischmanR, Bairn D. Variation in restenosis commuy angiography(abstr).Circulafion 1992;86:132OA. after Palm=-Schatzstentingare due to differencesin post-procedurelumen dim19. van dez GiessenWJ, SermysPW, van BeosekomHMM, van WcerkensLJ, &r. Circulation 1992;86:2116A. van Loon H, Lee Kie Soei, Straws BH, Beatt KJ, Verdouw P. Coronary stewing 22. RensingBJ, HermansWR, Vos J, Bean KJ, Bossuyt P, RutschW, Sermys with a new, mdiopaque,balloon-expandableendopmsthesisin pigs. Circulorion PW. On behalf of the Carport Smdy Group. Angiogmphicrisk factors of luminal 1991;83:1788-1798. narrowing after tommy balIoon angioplastyusingballoon mcasuremcnts to reflect 20. White IW, RameeSR, Banks AK, Mesa JE, Chokshi S, Isner JM. A new balstretchand elastic recoil at the dilatation site. Am J Cardiol 1992,69:584591. loon-expandabletantalumcoil stent: angiogmpbicpatencyand histologic fmdings 29. Foley DP, Sermys PW, HermansWR, de JaegereP, Morel MA, Rcelandt in an athemgenicswine model..I Am CON Cardiol 1992,19:87&876. JRTC. Is “bigger” really “better”? hnmediateand long-termoutcomefollowing 21. SchatzRA, PalmazC, Tio FO, GarciaF, Garcia0, ReuterSR.Balloon expand- balloonangioplasty,directionalatherectomyandstentimplantation(abstr). EurHeart J 199213:2226(A). able intmcomnarystentsin the adult dog. Circulation 1987;76:450-457. 22. Roubin GS, King SB, Douglas IS, Lemlw NJ, RobinsonKA. h~uacomnary 30. Nichols AB, Smith R, Berke A, Sblofmitz RA, PowersER. Importanceof balstentingduringpercutaneoustmnsluminalcomwuy angioplasty.Circulation 1990,81 loon size in coronary angioplasty.J Am Co/l Cardiol 1989;13:lC94-1100. (suppl IV):IV-2-IV-l@3 2l. Liu MW, Roubii GS,Kiig SB III. Restenosisafter coronaryangioplasty.Poten22. van BeusekomHMM, van der GiessenWJ, van Suylen RI, Bos E, Bosman tial determimmtsand role of intimal hyperplasia.Circulation 1989;79:13741387. Ff’, SermysPW. Histology after stentingof humansaphenousvein bypassg&is. 32. HermansWR, RensingBJ, Kelder JC, de Feyter PJ, SermysPW. Posta@+ Observationsfrom surgically excisedgmfts 3 to 320 days after stentimplantation. plasty trstenosisrate betweensegmentsof the major coronary arteries.Am J CarJ Am Co/l Cardio/ 1993;21:4>54. dial 1992:69:194-200.
170
THE AMERICANJOURNALOF CARDIOLOGY VOLUME72
Ezekowitz
JULY 15.1993
~ofthenrst91patSentswithsulid stent inrplantation and without clinical evidence of (sub)acub tlmmbuc dent occlusion W8lWrndyzedWiththS~ASSisted~
gr@ic Analysis System using automaW edge was44%by ddecuon.lheiddenceofpatiemtmd4S%bystemtaccad~totheO.72 nmt!dteuw,ana2O%bypatientand29o/oby stentaccordi~tothe5O%diameterstenosis& lwstemdsforseveral Mgb tedon.Theriskfor m varlabler was deWmined using an univafiateanalysisandisexpreaQedasoddsratsowith co#ahceintewal.llleonlystati~ ~PpdktorOf-WNthe
reJatlvegainwhenitexceededOASusingthe 0.72 mm aftdon (odds ratio 2.7,95% confidence interval lA-6.4). Fu?thMnmu, thedationbe tween the dative gain (increase in minimal hunk naldbneternufnudixedtovesselsize)asangb grqshic index of vessei wall iduy and relative loss (decrease in minimal luminal diameter non matized to vessel size) as index of neointimal From the Catheterization Laboratory and Laboratory for Quantitative Angiographic Analysis (Cardialysis), Thoraxcenter, Rotterdam, the Netherlands: Department of Cardiology, Hopital Cardiologique, Lille, France;Department of Cardiology, Georg August UnivemitZt Gottingen, Germany; Department of Cardiology, University of Ottawa Heart Institute, Ottawa, Ontario, Canada;Departmentof Cardiology, University Hospital Gasthuisberg,Leuven, Belgium; Department of Cardiology, Clinique la Casamance,Marseille, France; Department of Cardiology, Klinikum der J.W. Goethe Universimt Frankfurt, Germany; Departmentof Cardiology, CatharinaHospital, Eindhoven, the Netherlands; Department of Cardiology, University Hospital Berlin, Germany; Department of Cardiology, Medical Clinic I RWTH Aachen, Germany. Manuscript received October 14, 1992; revised manuscript received and acceptedFebruary 22, 1993. Addressfor reprints: Patrick W. Serruys,MD, CatheterizationLaboratory and Laboratory for Quantitative Angiographic Analysis (Cardialysis), Thoraxcenter, ErasmusUniversity, P.O. Box 1738,300ODR Rotterdam,the Netherlands.
thickening was adyzed using a linear re#essh analysiaAPesrsonproduCtiMmMteorrelation . codhalt of 0.38 (p *o.OOl) was found. when usingthecategwical~toaddressre steno4bis,thereisaninaessed riskforrecunwnt restenosis when the mlative gain exceeds 0.48. TheCOlltillUOUSapproachU-thiScOrr
ceptbyindicatingaweakbutposRiverehtionbs tweenthefe4ativegainandrelativeloss. (Am J Cmliol1~72:16+170)
P
ercutaneoustransluminaJcoronary balloon angioplasty is now widely acceptedas a safe and effective treatmentin selectedpatients with obstructive coronary artery disease.’ However, acute vessel closure and late restenosis are inherent to balloon angioplasty and continue to compromiseits efficacy.2,3Although the exact pathophysiologic mechanism(s)and factors responsible for restenosis are largely unknown, some clinical and angiographic variables have been identified as potential risk factors.4 There is, furthermore, some evidence that repeat angioplasty of a previously dilated lesion in a native coronary artery is associated with a higher incidence of restenosiss9 The Wiktor stent has been used to treat such patients. The incidence of recurrence of restenosisfollowing Wiktor stent implantation in the first 50 consecutive patients has recently been published.tO The purpose of this study was to try to identify angiographic predictors of recurrence of restenosis after Wiktor stent implantation. This may not only have important clinical consequences,but may also broaden our understanding of the pathophysiology of restenosis. MRHODS PatienW Between January 1990 and May 1992, Wiktor stent implantation was attemptedin 193 patients becauseof recurrence of angina due to restenosisof a native coronary artery lesion after previous balloon angioplasty. The study population consisted of 91 of the 109 consecutivepatients with successfulstent implantation without clinical evidence of thrombotic stent occlusion during hospital stay and in whom the scheduledfollow-up angiography was completed at 5.7 Ifr 1.9 months (mean + SD) (Figure 1). Eighty-one patients were men (89%); the mean age (& SD) was 58 f 11 years. A first restenosiswas documented in 50 patients (55%), a second in 32 patients PREDICTORSOF RESTENOSISAFTERSTENTING 165
I
7
TABLE I Findings at Quantitative Angiography After Wiktor Stent Implantation (n = 91 patients)
Reference diameter (mm) Minimal luminal diameter (mm) Diameter stenosis (%)
Before PTCA
After PTCA
2.85 k 0.49
2.80 2 0.48 1.77 5 0.36 35211
1.13 -c 0.36
60 2 10
After Stent Implantation
At Follow-Up
3.02 2 0.42 2.45 + 0.34 19 2 7
2.98 f 0.47 1.73 + 0.67 42 2 21
All parameters are expressedas mean k SD. All changes were highly significant (p
(35%), a third in 8 patients (9%) and a fourth in one other patient (1%). In all patients, a single Wiktor stent was implanted except in one, in whom a single stent was implanted in both the left anterior descending and the right coronary artery. The target vessel was the left anterior descendingartery in 48 patients (52%), the right coronary artery in 31 patients (34%), and the left circumflex in 12 patients (13%). In another patient (l%), a single Wiktor stent was implanted in the left main stem. The size of the stent used was 3.4 + 0.4 mm (mean f SD): a 3.0 mm device in 38 patients (41%), a 3.5 mm device in 40 patients (44%) and a 4.0 mm device in 14 patients (15%). The stent is describedin detail elsewhere as well as the associateddrug protocoL1t wk
vmWks
and d&WKom of restene
sisr Based on the quantitative angiographic data, multiple variables were identified and recorded for each lesion. These variables were of a priori clinical interest on the basis of previously published reports on balloon angioplasty and intracoronaty stenting and of a more fundamental scientific interest based on experimental studies on stent implantation in animals.4J2-16The following continuous variables were selected: obstruction diameter before and immediately after stent implantation, baseline reference diameter and diameter stenosis, length of target lesion, plaque area and the relative gain defined by the increase in minimal luminal diameter immediately after stent implantation normalized to the 109 CONSECUTIVE PATIENTS 13 (SUBIACUTE OCCLUSIONS (6 CABG/ 8 re-PTCAI thrombolysis) i DEATH (HEART FAILURE) AT 3 months 3 REFUSALS
vessel size. In addition to these continuous variables, one discrete variable (left anterior descending artery) was selected. All coronary angiograms were analyzed with the computer-assistedCardiovascular Angiography Analysis System.Its principles and deft&ions of angiographic variables has been describedin a previous study on Wiktor stent implantation.11Sincethe primary objective of this study was to identify angiographic variables predicting restenosis,the process of restenosis was dichotomized and delined according to the 0.72 mm and 50% diameter stenosiscriteria.17 staU&ks A relative risk analysis was performed for the aforementionedangiographic variables. To avoid an arbitrary subdivision of data in the continuous variables, the median value was chosen as cutoff point. The selection of this value as cutoff point has the advantage of being consistent for all values, and thus avoids any bias in the selection of subgroupsthat might be undertaken to emphasize a particular point. All values are expressedas mean f SD. The risk for restenosisfor each parameteraccording to the 0.72 mm and 50% diameter stenosis criteria was determined by using an univariate analysis and is expressed as odds ratio with corresponding 95% confidence interval. An odds ratio of 1.0 for a particular variable implies that the presenceof that variable posesno additional risk for restenosis,an odds ratio >l.O or cl.0 implies additional or reduction in risk. Furthermore, since experimental animal data indicate that there is a relation between the severity of vessel wall injury and the extent of subsequent neointimal thickening, the relation between relative gain (as index of vessel wall injury) and relative loss (as index of neointimal thickening) was studied using a regressionanalysis. The relative gain is previously defined. The relative loss is the difference between the minimal luminal diameter immediately after stent implantation and at follow-up normalized to the vessel size.
1 MISSING
F91 CONSECUTIVE PATIENTS WITH ANGlOGRAPHIC FOLLOW-UP* *5.7 t 1.9 months a@ogr@lc PKwRELFlewdle@ml~ ptltknk~ecuteoreub ulelmt109ecubecclueiaecccuwsdbl~patlenb(~).Dftberemairr wllbeut illg98p0ti9llt!SWhOW~~ denceof(sub)acuteocclusion,9lundewantrepeat~ r&yat5.7~l.9(mean*sD)mcetbsefterstentinrpbntb ticn.cAB6= eoronayarteyW--A=repeat perwtaneoustradumlnalcoronmy~. 166
THE AMERICAN JOURNAL OF CARDIOLOGY
followap dinkd
in e*
VOLUME 72
RESULTS hcidmce d restamh The changesin stenosisgeometry are summarized in Table I. The incidence of restenosiswas 44% (40 of 91 patients) according to the 0.72 mm criterion and 30% (27 of 91 patients) according to the 50% diameter stenosis criterion. When the stent is used as a unit, the incidence of restenosiswas 45% (41 of the 92 stents)and 29% (27 of the 92 sterns), respectively (Figure 2). WC predictors: The relative risk and 95% confidenceintervals for each variable using either of the JULY 15,1993
TABLE II Angiographic Predictors of Restenosis After Wiktor Stent Implantation RestenosisCriteria 0.72 mm
Angiographic Variable Relative gain (mm) Minimal luminal diameter after (mm) Minimal luminal diameter before (mm)
20.48 22.38 11.08 2 6.48 ~2.77 2 7.47 261
Lesion length (mm) Reference diameter before (mm)
Plaque area (mm*) Diameter stenosis before (%) Left anterior descendingartery
2 criteria are listed in Table II. The only statistically signilicant predictor of recurrence of restenosis according to the 0.72 mm criterion was the relative gain when it exceeded0.48 (odds ratio 2.7, 95% confidence interval 1.1-6.4). Related to the relative gain is the obstruction diameter before and after stent implantation. Although their odds ratio exceeds 1.0, their corresponding 95% confidence interval precludes any ti conclusion. The same holds for lesion length, reference diameter and plaque area. The left anterior descendingartery was not identified as a risk factor for recurrence of restenosis. When using the 50% diameter stenosiscriterion, no angiographic predictor of restenosiswas evident. The number of balloon angioplasties performed before stent implantation did not influence the risk for sub-
4
MLD F/U (mm)
jc CHRON OCCL: 4(4%) 1 I
? 50%:
//
27(29%Y
/
50% Diameter Stenosis Criterion
Criterion
,
2.7 1.8 1.6 1.4 1.3 1.3 1.0 0.7
(1.1-6.4) (0.8-4.2) (0.7-3.7) (0.6-3.2) (0.6-2.9) (0.6-2.9) (0.4-2.3) (0.3-1.7)
1.1 0.5 2.7 1.8 1.4 1.1 1.3 1.1
(0.4-2.7) (0.2-1.4) (1.0-7.2) (0.7-4.6) (0.5-3.4) (0.4-2.7) (0.5-3.2) (0.5-2.8)
sequentrestenosisafter stenting. The odds for restenosis in caseof stent implantation for a second,third or fourth restenosis compared with stent implantation for a lirst restenosis was 0.8 (95% contidence interval 0.4-1.8) according to the 0.72 mm criterion and was 1.5 (95% cotidence 0.6-3.7) according to the 50% diameter stenosiscriterion. Regression analysis: The relation between relative gain, as index of vessel wall injury, and relative loss, as index of late neointimal hyperplasia as vessel wall responseto injury, is shown in Figure 3. A Pearsonproduct-moment correlation coefficient of 0.38 was found (p
, .4 Relative loss RL= .61 RG -.06 R= .38 (p=.OOl)
1.2l-
OO
0.81
0
0.8
0
I
1
2
3
4
MLD POST-STENT(mm)
0.72 mm LONQ
0
dSIlC8Of-S~llgtOthSSO%-~S
cdtdonwas29%(27ofthe92stents).Ofth8se,4stents oc&dedatfol~p,~shownonthex W-M dS.-OCCL=ClStWlkOCdUdan.
I
0 80
0.2 0 -0.2
fohwp(F/U)aftewWihtorstentbnphbth.7hediame terofeachse#nentimmdbMyaftershtimplmRa&nis cEmnderatfel~p.Thelinesoneadl mcrcdndsidaafthsnnaafmnfRy(~)~twisetha ViwiabilRy (OS% coddmca bntarval) of duplkata nlsasucB mants (a chmge of to.72 mm). A si@Mcmt tbcmass in nlinimalundndacxod~!tothe~O.72nuncdteuh wasfoundin4leftha92stwtsi@mIted(4s%).The~
00
0.4
TERM WU?IARILITY
FlGuRE2.chmgesinthemhimdlllmind-(MU))at
OOQ
OO
-0.4
0
I 0
0.2
0
0.4
0.8
0.8
1
1.2
Relative gain FlwRE1.maphkasplayoftherelatbnbe4waenthsrela tbe@h(RQ=hcfeuxwinminimaIluminald&wmterimmediinqhMthmwnlalizsdtothsvessslwall) w-mdrela#wb(RL=decrcwro inminlmalhdnddlmWer nomdkedtethevessalwdl).A~eOnear dfdkrwP ramillnwasfoundwmla-(R) of 0.38 (p<0.001)wilha~of0.81adani~onthey axis of 4.00. PREDICTORS OF RESTENOSIS AFTER STENTING
167
I
-PtO.Os----,
..
.
I
Schwartz Circulation
et al.
I
Nobuyoshi
‘90
I
‘91
Therefore, late loss or restenosiswas due to ingrowth of tissue into the lumen of the stented segment.t8 The lack of randomized studies precludes any conclusion as to whether stent implantation in this subsetof patients will reduce the incidence of subsequentrestenosis. Moreover, the exact incidence of recurrent restenosis after repeat balloon angioplasty is not known. There is some evidence that it increaseswith the number of repeat angioplasties. A second restenosis has been reported to occur in 25 and 34%, but amounts to 39 and 40% after a third or fourth angioplasty,respectively.~9 These data should be interpreted with caution considering the differencebetweenthe study populations, the dif-
-0.2 -0.4
O0
0.5
0.6 0.7 0.6 0.9
1
1.1 1.2 1.3 1.4
stent/artery
1.5 1.6
ratio
ApesRivelhemmlatienwasfeundwRha~cadi fhnt(R)[email protected])wRhadopeof0.6OtmnJan
ilMNqRof9~. 166
THE AMERICANJOURNALOF CARDIOLOGY VOLUME72
Beatt
et al.
J.Am.Coll.Cardiol.
I
et al.
J.Am.Coll.Cardiol.
‘92
ference in definition of restenosis,and in the time and completenessof follow-up angiography. In the present study in which stemswere implanted becauseof restenosis, the risk for recurrent restenosisalter stenting did not differ between patients with a first restenosis and patients with a second, third or fourth restenosis.Two other studies failed to show a statistically sign&ant difference in the restenosis rate for primary or de novo lesions comparedwith lesions that had undergoneprevious balloon angioplasty.12,13However, subgroup analysis of patients in whom a single Palmaz-Schatzstent was implanted showed a restenosis rate of 13% compared with 36% for patients with previous restenosis.*3 The development of restenosisremains incompletely understood. Histologic data have shown that any injury to the vessel wall, whatever its nature, will invariably be associatedwith neointimal hyperplasia as a nonspecilic tissue response leading to restenosis when excessive.14J9-23It is of both scientitic and utmost clinical importance to know and understandits developmentand to know the factors controlling the extent of the vessel wall response following injury. Experimental animal studies, postmortem pathologic observations in humans and angiographic studies have shown that the extent of neointimal hyperplasia is proportional to the degree of injury applied to the vessel wall (Figure 4).t4-16J4This is in accord with earlier work in the porcine model, describing a relation between the degree of injury and early platelet deposition.25~26 In the presentstudy,the relative gain was used as angiographic correlate of vessel wall injury. This parameterwas found to be the only statisticahy signilicant predictor of recurrenceof restenosis when using the 0.72 mm criterion. The drawback of the . _._ latter is, lirst, that the process of restenosis is dichotomized (present/not present) and second, that although
JULY 15,1993
statistically justifiable, cutoff points are used for the angiogmphic variables under investigation. However, the continuous approach underscoresthese Iindings by indicating a positive linear relation between the relative gain and relative loss. These observations were still valid when using absolute measurements.However, the correlation between the absolute gain and absolute loss was somewhatweaker (r = 0.21, p = 0.05). This is explained by the fact that the use of absolutevalues does not relate these changesto the vessel size. Although we found no difference in absolute loss between small and large vessels, an identical loss in minimal luminal diameter represents a larger relative loss in a small vessel than in a large vessel, and vice versa. This conlirms other studies using a categorical approach to detine restenosis after stenting, that small vessels are not, per se, more prone to restenosisthan large vessels.12J3,27 This concept, describing a proportional relation between vessel wall injury and neointimal thickening, has been reported in other studies using quantitative coronary angiography. Beatt et ali6 found that restenosis (0.72 mm criterion) was signilicantly correlated with both a greater improvement in obstruction diameter and a larger absolute dimension aher balloon angioplasty. Furthermore, in a recent study,28the absolute change in minimal luminal diameter was reported to be the greatest single determinant of late luminal narrowing after balloon angioplasty. In another rep~rt?~ it was found that interventions achieving a “bigger” lumen provoke a concomitantly larger relative loss, so that the ultimate end point of various treatment methods is similar. All these findings carry potential far-reaching clinical implications. Clearly the greater the improvement in minimal luminal diameter achieved by intervention, the greater the magnitude of subsequentluminal narrowing will be. Unfortunately, what cannot be extrapolatedfrom the data is how much damage the clinician may intlict on the vessel wall. On the one hand, a suboptimal angiographic result is associatedwith a higher risk of subacute occlusion due to rheologic factors and platelet deposition and a higher need for repeat balloon angioplasty but, on the other hand, improvement of the initial result may be at the price of more extensive late neointimal thickening.i4i6*30,31This indistinctness can be circumvented by properly matching the balloon size or device with the vessel wall using on-line quantitative coronary angiography. This is underscoredby the data displayed in Figure 5 indicating that the more the stent is oversized, the greater the loss in minimal luminal diameter will be. So far, there are 2 other angiographic studies that have attemptedto identify risk factors for restenosisor recurrenceof restenosis,and they found, as in the present study, that the vessel size and severity of stenosis, expressedas diameter stenosisand length of the lesion, not to be associatedwith an increasedrisk for restenosis.r2J3As mentioned, the indication (primary or secondary stenting) was not associatedwith an increasedrisk for restenosis.12J3 In accordancewith these 2 studies on stenting and 1 other study on balloon angioplasty, the target vessel was not associatedwith an increased risk for (recurrent) restenosis.12J3.32
SW limitations: The study limitations are essentially twofold. First, the precision of relative gain as an angiographic index of vessel wall injury and relative loss as an index of neointimal hyperplasia has not been studied. The coronary angiogram is a 2dirnensional echocardiogram describing the changesin stenosisgeometry but not the nature or extent of the pathology explaining these angiographic changes.Therefore, the measurement of the minimal luminal diameter, and consequently the relative gain and loss, are subject to potential imprecision. Second, the effect of other clinical, procedural and lesion-related characteristicson the development of late neointimal hyperplasia has not been considered. It is conceivable that if these characteristicsplay a role in the pathogenesisof neointimal thickening, they may not be uniformly distributed over the study population. ACkllOV . We thank Marie-Angele Morel for performing the quantitative analysis and for her assistancein the databasemanagementand statistical analysis.
1. De&e K, Holubkov R, Kelsey S, Bourassa M, Williams D, Holmes D, Dorms G, Faxon D, Myler R, Kent K, Cowley M, Cannon R, Robertson T, and the coinvestigators of the National Heart, Lung, and Blood Institute’s Petcutaoeous Tramhmid Coronary Angioplasty Registry. One-year follow-up results of the 19851986 National Heart, Lung, and Blood Imtihtte’s PI%A Regisby. Circulation 1989;SO: 421428. 2. de Feyter PJ, de Jaegere Pm, Murphy Es, Senuys PW. Abrupt coronary artery occlusion during penxtaoexx~s tmnsluminal coronary angioplasty. Am Heart .I 1992; 123:163>1642. 9. Sermys PW, Luijten HE, Beat KJ, Geuskebs R, de Feytez PJ, van den Brand M, Reiber JHC, ten Katen HJ, van Es GA, Hugenholtz PG. Incidence of restenosis after successful coronary angioplasty: a time-related phenomenon. Circulution 1988;77:361-371. 4. Botrassa M, Lesperance J, Eastwood C, Schwutz L, Cote G, Kazim F, Hudon G. clinical, physiologic, anatomic and procedural factors predictive of resteoosis after percutaneous traosluminal coronary angioplasty. J Am Coil Cardiol 1991; 18: 36M76. 1. Williams DO, Gtuentzig AR, Kent KM, Detre KM, Kelsey S, To T. Efficacy of repeat percutaneous tmnsluminal coronary angioplasty for coronary testenosis. Am J Cardiol 1984,53:32C-35C. 8. Meier B, King SB III, Gmentzig AR, Douglas JS, Hollman I, Ischinger T, Galao K, Tanker&y R. Repeat coronary angioplasty. J Am Co/l Car&l 1984;4:463466. 7. Black AJ, Anderson VH, Roubii GS, Powelson SW, Douglas JS, Kiig SB III. Repeat coronary angioplasty: conelates of a second restenosis. J Am Co/l Cardiol 1988;11:714-718. .B. Teintein F’S, Hoover CA, Ligon RW, Giorgi LV, Rutherford BD, McConahay DR. Johnson WL, Hart&r GO. Repeat coronary angioplasty: efficacy of a thii angioplasty for a second restenosis. J Am Co/l Curdiol 1989; 13:291-2%. 9. Quigley PI, Hlatky MA, Hinoham T, Rendall DS, Perez JA, Phillips HR. Calii RM, Stack RS. Repeat percutaneous tmnsluminal coronary aogioplasty and pmiictars of recunent restenosis. Am J Cardiol 1989;63:4w13. 10. de Jaegete P, Senuys PW, Bertrand M, Wiegaod V, K&r G, Marquis JF, Valeix B, Uebis R, P&ens J. Wiitor stent implantation in patients with restenw sis following balloon aogioplasty of a native coronary artery. Am J Cardiol 1992;69: 59M2. 11. Sermys PW, de Jaegere P, Bertrand M, K&r G, Marquis JF, Piessens J, U&ii R, Valeix B, Wiegand V. Morphologic change in coronary artery stenosis with the medtmnic Wiior stent. Initial results from the core laboratory for quantitative aogiogmphy. Cathet Cardiovasc Diagn 1991;14:237-245. l2. Strauss BH, Semtys PW, de Scheetder IK, Tijssen JG, Bertrand ME, Puel J, Meier B, Kaufmann U, Stauffer JC, Rickards AF, Sigwart U. Relative risk analysis of angiograpbic predictors of restenosis within the coronary wallstent. Circulation 1991;84:163&1643. 12. Fischman DL, Savage MI’, Ellis SG, Schatz RA, Leon MB, Baim D, Goldberg S. Restenosis after Palmaz-Schatz Stent implantation. In: Senuys PW, Strauss BH, Kiig SB III, eds. Restenosis After Intervention with New Mechanical Devices. Boston: Kluwer Academic Publishers, 1992; 191-205. i4. Schwartz RS, Mulphy JG, Edwards WD, Cammd AR, Vlietstm RE, Holmes DR. Restenosis after balloon angioplasty. A practical proliferative model in porcine coronary arteries. Circulation 1990$32:219&2200. $5. Nobuyoshi M, Kimura T, Ohishi H, Horiuchi H, Nosaka H, Hamasaki N, Yokoi H, Kim K. Restenosis after percutaneous transluminal comnary angioplasty: patbo
PREDICTORSOF RESTENOSISAFTERSTENTING 169
logic observationsin 20 patients.J Am Co// Cardiol 1991;17:433-439. 16. Beatt KJ, SenuysPW, Luijten HE, RensingBl, SwyapranataH, de Feyter PJ, van den Brand M, Lawman GJ, RoelandtI. Restenosisafter commuy angioplasty: the paradoxof increasedlumendiameterandrestenosis.J Am CON Cordial 199219:
24. SarembockIJ, LaVeau PJ, Sigal SL, Tiimis 1, SussmanJ, HaudenschildC, M. Infiuenceof inflation pressureand balloon size on the development of intimal hyperplasiaafter balloon angioplasty.A study in the atheroscleroticrat+ bit. Circulation 1989;80:102’+1040. 25a266. 25. SteelePM, ChesebmJH, StansonAW, Holmes DR. Dewanjec MK, Bodiman of short-, medium-,and L, FusterV. Balloon angioplasty:natural bistmy of the pathophysiologicresponse 17. Reiber JHC, SenuysPW, Kooijman CJ. Assessment to injury in a pig model. Circ Res 1985;57:105-112. long-term variations in arterial dimensionsfrom computer-assisted quantitationof 26. Wilentz JR, SanbomTA, HaudenschildCC, Valeri CR, Ryan TJ, Faxon DP. coronary cineangiograms. Circulation 1985;71:280-288. 18. de JaegereP, Bertrand M, Wiegand V, Marquis JF, Vrolicx M, P&ens J, Plateletaccumulationin experimentalangioplasty:time courseand relation to vasValeix B, Kober G, Bonnier H, RutschW, Uebis R. Recoil following Wiitor stent cular injury. Circulation 1987;75:636-642. implantation.Resultsof the EuropeanRegistry and of the core lab for quantitative 27. CarmzzaJ, SchatzR, Kuntz R, FischmanR, Bairn D. Variation in restenosis commuy angiography(abstr).Circulafion 1992;86:132OA. after Palm=-Schatzstentingare due to differencesin post-procedurelumen dim19. van dez GiessenWJ, SermysPW, van BeosekomHMM, van WcerkensLJ, &r. Circulation 1992;86:2116A. van Loon H, Lee Kie Soei, Straws BH, Beatt KJ, Verdouw P. Coronary stewing 22. RensingBJ, HermansWR, Vos J, Bean KJ, Bossuyt P, RutschW, Sermys with a new, mdiopaque,balloon-expandableendopmsthesisin pigs. Circulorion PW. On behalf of the Carport Smdy Group. Angiogmphicrisk factors of luminal 1991;83:1788-1798. narrowing after tommy balIoon angioplastyusingballoon mcasuremcnts to reflect 20. White IW, RameeSR, Banks AK, Mesa JE, Chokshi S, Isner JM. A new balstretchand elastic recoil at the dilatation site. Am J Cardiol 1992,69:584591. loon-expandabletantalumcoil stent: angiogmpbicpatencyand histologic fmdings 29. Foley DP, Sermys PW, HermansWR, de JaegereP, Morel MA, Rcelandt in an athemgenicswine model..I Am CON Cardiol 1992,19:87&876. JRTC. Is “bigger” really “better”? hnmediateand long-termoutcomefollowing 21. SchatzRA, PalmazC, Tio FO, GarciaF, Garcia0, ReuterSR.Balloon expand- balloonangioplasty,directionalatherectomyandstentimplantation(abstr). EurHeart J 199213:2226(A). able intmcomnarystentsin the adult dog. Circulation 1987;76:450-457. 22. Roubin GS, King SB, Douglas IS, Lemlw NJ, RobinsonKA. h~uacomnary 30. Nichols AB, Smith R, Berke A, Sblofmitz RA, PowersER. Importanceof balstentingduringpercutaneoustmnsluminalcomwuy angioplasty.Circulation 1990,81 loon size in coronary angioplasty.J Am Co/l Cardiol 1989;13:lC94-1100. (suppl IV):IV-2-IV-l@3 2l. Liu MW, Roubii GS,Kiig SB III. Restenosisafter coronaryangioplasty.Poten22. van BeusekomHMM, van der GiessenWJ, van Suylen RI, Bos E, Bosman tial determimmtsand role of intimal hyperplasia.Circulation 1989;79:13741387. Ff’, SermysPW. Histology after stentingof humansaphenousvein bypassg&is. 32. HermansWR, RensingBJ, Kelder JC, de Feyter PJ, SermysPW. Posta@+ Observationsfrom surgically excisedgmfts 3 to 320 days after stentimplantation. plasty trstenosisrate betweensegmentsof the major coronary arteries.Am J CarJ Am Co/l Cardio/ 1993;21:4>54. dial 1992:69:194-200.
170
THE AMERICANJOURNALOF CARDIOLOGY VOLUME72
Ezekowitz
JULY 15.1993