The immediate and long-term effect of optimal balloon angioplasty on the absolute coronary blood...

European Heart Journal (2001) 22, 1725–1732doi:10.1053/euhj.2000.2587, available online at http://www.idealibrary.com on

The immediate and long-term effect of optimalballoon angioplasty on the absolute coronary blood

flow velocity reserve

A subanalysis of the DEBATE study

J. J. Piek1, E. Boersma2, M. Voskuil1, C. di Mario3, E. Schroeder4, C. Vrints5,P. Probst6, B. de Bruyne7, C. Hanet8, E. Fleck9, M. Haude10, E. Verna11,

V. Voudris12, H. Geschwind13, H. Emanuelsson14, V. Muhlberger15, H. O. Peels16

and P. W. Serruys2, on behalf of the DEBATE study group

1Academical Medical Center, Amsterdam, The Netherlands; 2Thoraxcenter Rotterdam, The Netherlands;3Centro Curore Columbus, Milano, Italy; 4Clinique Universitaire de Mont-Godinne, Yvoir, Belgium;

5Universitair Ziekenhuis Antwerpen, Belgium; 6Kardiologische Universitatsklinik Wien, Austria;7Onze Lieve Vrouwe Kliniek Aalst, Belgium; 8Clinique Universitaires de Saint-Luc, Brussel, Belgium;

9Deutsches Herzzentrum Berlin, Germany; 10Universitat Essen, Germany; 11Ospedale di Circolo, Varese, Italy;12Onassis Cardiac Surgery Center, Athens, Greece; 13Universite Paris XII Val de Marne, La Creteil, France;

14Sahlgrenska Hospital Goteborg, Sweden; 15Universitatsklinik fur Innere Medizin Innsbruck, Austria;16Academisch Ziekenhuis Groningen, The Netherlands

Revision submitted 19 December 2000, and accepted 20 December 2000.

Correspondence: Jan J. Piek, MD, Department of Cardiology, B2-108, Academic Medical Center, Meibergdreef 9, 1105 AZ Amsterdam,The Netherlands. P.O. Box 22700, 1100 DE Amsterdam, The Netherlands.

Background There are limited data regarding the immedi-ate and long-term effect of balloon angioplasty on thecoronary flow reserve evaluated in a multicentre setting.

Methods and Results A total of 86 patients with one-vessel disease and normal left ventricular function wereanalysed before and after optimal balloon angioplasty(diameter stenosis <35%) and at 6-month follow-up. Cor-onary flow reserve was assessed with a Doppler guide wire.A low coronary flow reserve (�2·5) after PTCA, due to anincreased baseline blood flow velocity, was encountered in42 of the 86 patients (49%). Recurrence of angina andtarget lesion revascularization were more frequent in thesepatients than in patients with a coronary flow reserve >2·5(46% vs 23% and 36% vs 16%, respectively; P<0·05) dueto a trend towards restenosis (29% vs 16%; P=0·15) or alow coronary flow reserve at follow-up due to persistentelevated baseline blood flow velocity. Patients withoutrestenosis showed a decrease or increase of coronary flowreserve during follow-up, determined by alterations ofhyperaemic blood flow velocity.

0195-668X/01/221725+08 $35.00/0

Conclusions Patients with an impaired coronary flowreserve directly after optimal balloon angioplasty showed ahigher target lesion revascularization rate compared topatients with a coronary flow reserve >2·5. This patientgroup consists of patients prone to develop restenosis, whileother patients are characterized by a persistently low cor-onary flow reserve, probably secondary to disturbed au-toregulation and/or diffuse mild coronary atherosclerosis.Coronary flow reserve alterations in patients withoutrestenosis were related to changes in hyperaemic bloodflow velocity, suggesting that this phenomenon relates toepicardial remodelling.(Eur Heart J 2001; 22: 1725–1732, doi:10.1053/euhj.2000.2587)� 2001 The European Society of Cardiology

Key Words: Angiography; coronary stenosis; angioplasty;Doppler flow; remodelling.

See page 1633 for the Editorial comment on this article

� 2001 The European Society of Cardiology

1726 J. J. Piek et al.

Introduction

There is controversy regarding the effect of balloonangioplasty on the distal coronary flow reserve. Severalstudies have indicated that balloon angioplasty results ina direct normalization of distal coronary flow reserve,while other studies reported temporary impairment ofcoronary flow reserve after balloon angioplasty[1–6]. Ithas been hypothesized that the spontaneous improve-ment of coronary flow reserve after balloon angioplastymay be related to remodelling of the epicardial segmentand/or slow recovery of autoregulation of the microvas-cular bed[3–6]. Recent insights from single-centre studiessuggest that an impaired coronary flow reserve afterballoon angioplasty may be due to a residual luminalobstruction resulting in a reduced hyperaemic responseor secondary to an elevated baseline blood flow velocityas a result of disturbed autoregulation of the microvas-cular bed[3,7,8]. A recently published multicentre trial(Doppler Endpoints Balloon Angioplasty Trial Europe,DEBATE) evaluated the effect of balloon angioplastyon distal coronary flow reserve in relation to the clinicaloutcome during a 6-month follow-up period[9]. The aimof this substudy was to evaluate the effect of balloonangioplasty on epicardial remodelling and adaptation ofthe microvascular bed in a selected cohort of patientswith optimal angiographical results after percutaneoustransluminal coronary angioplasty (PTCA). Thesephenomena were studied by analysis of arterial dimen-sions assessed by quantitative coronary angiographyand baseline and hyperaemic blood flow velocity altera-tions before and after PTCA and at 6-month follow-up.

Methods

Patient selection

The study population consisted of 86 out of 225 patientsundergoing PTCA for chest pain and/or documentedsigns of myocardial ischaemia (electrocardiographic,scintigraphic or echocardiographic) in the setting of amulticentre study (DEBATE). The details of this multi-centre study have been described elsewhere[9]. Briefly,patients scheduled for PTCA of one major native cor-onary artery and with a normal left ventricular functionwere included. Exclusion criteria were: multivessel dis-ease, previous transmural myocardial infarction in theterritory distributed by the vessel to be dilated, acutemyocardial infarction less than 1 week prior to PTCA,previous PTCA of the actual coronary segment, total orfunctional coronary occlusion, presence of left bundlebranch block, second or third degree atrioventricularblock, open bypass graft distal to the lesion to betreated, extreme tortuosity of the vessel to be dilated andintended alternative or additional interventional treat-ments such as directional and rotational atherectomy orstent implantation. The study protocol was approved bythe institutional review boards of the participating

Eur Heart J, Vol. 22, issue 18, September 2001

centres. All patients gave written informed consent.Patients were included in this analysis if they had had anoptimal result with balloon angioplasty, defined as adiameter stenosis <35% after PTCA determined byoff-line quantitative coronary angiography.

Angioplasty procedure and blood flowvelocity assessment

The study protocol included pre-medication of patientsbefore angioplasty with heparin and acetylsalicylic acid.A 0·014 inch Doppler tipped guide wire was used as theprimary angioplasty guide wire (FloWire; Endosonics,Rancho Cordova, CA, U.S.A.)[10]. The Doppler guidewire was advanced distal to the lesion and blood flowvelocity recordings were obtained under baseline andhyperaemic conditions. A distance from the stenosisgreater than five times the vessel diameter was main-tained in order to avoid post-stenotic turbulent flow.Maximal hyperaemia was induced by an intra-coronarybolus injection of adenosine, 12 �g for the right cor-onary artery and 18 �g for the left coronary artery. TheDoppler guide wire was left in place and blood flowvelocity was continuously monitored throughout theprocedure. Only angiographic criteria (diameter stenosis<50% in any angiographic view) were used to determinethe end-point of the angioplasty procedure. The site ofthe blood flow velocity measurements was recorded oncinefilm.

Quantitative coronary angiography

Cine-angiography was performed after the admin-istration of 0·1–0·3 mg nitroglycerin or 1–3 mg isosorb-ide dinitrate, intra-coronarily, to achieve maximalcoronary vasodilation. At least two cine-angiograms, inorthogonal projections, were obtained before coronaryangioplasty and repeated at late follow-up. All cinefilmswere analysed by an independent core laboratorywithout knowledge of the clinical information and theDoppler flow data. Matched views and frames wereselected for off-line quantitative computer-assistedanalysis (CAAS II system, Pie Medical Data,Maastricht, the Netherlands). Automatic edge-detectionof the luminal dimensions (minimum lumen diameter,reference diameter) was performed using the guidingcatheter filled with contrast as a scaling device.

Blood flow velocity analysis

During the angioplasty procedure, the Doppler flowvelocity signals were recorded continuously on video-tape using a Doppler flow spectral analyser (FloMap;Endosonics, Rancho Cardova, CA, U.S.A.). The bloodflow velocity measurements were used to calculate, bymeans of the time-averaged peak velocity (normalized to

A subanalysis of DEBATE 1727

the cardiac cycle), the distal blood flow velocity reserve,defined as the ratio between maximal blood flow velocityafter administration of adenosine and the baseline bloodflow velocity. The appropriateness of the Doppler flowmeasurements was verified by an independent corelaboratory.

Follow-up procedures

Patients were followed for 6 months (�4 weeks) fordocumentation of major adverse cardiac events, definedas cardiac death, myocardial infarction or the need fortarget lesion revascularization, in cases of recurrentangina and/or objective signs of myocardial ischaemiadue to coronary renarrowing with a diameter stenosisgreater than 50%. Quantitative coronary angiographywas performed at 6 months in the same views as duringthe initial procedure. The baseline average peak bloodflow velocity (b-APV) and hyperaemic average peakblood flow velocity (h-APV) recordings were obtained inthe same position as before PTCA.

Statistical analysis

Continuous variables are expressed as mean�standarddeviation. Chi-square analysis was used to detect adifference in categorical patient characteristics. A two-tailed paired t-test (or Wilcoxon test for parametricdata) was used to identify variation within subjects. Atwo-tailed unpaired t-test (or Mann–Whitney test fornon-parametric data) was used to assess differences incontinuous variables. These variables were subsequentlyentered in a univariate linear regression analysis todetermine significant predictors for changes in coronaryflow reserve at follow-up. Contributing factors wereentered into a multivariate linear regression analysis todetermine independence. For all tests, a P-value <0·05was considered statistically significant.

Results

A total of 86 patients of the 225 patients enrolled in theDEBATE study showed a percentage diameter stenosis<35% after balloon angioplasty and were analysed inthis study. Haemodynamic and angiographic data werecomplete in all 86 patients. Doppler data were completein 86 patients before PTCA and in 83 patients atfollow-up. A total of 19 of the 86 patients (22%) showedrestenosis at follow-up angiography.

Impaired coronary flow reserve directly afteroptimal balloon angioplasty

Baseline characteristicsA total of 42 of the 86 study patients (49%) showed animpaired coronary flow reserve after PTCA, while the

remaining 44 patients (51%) demonstrated a coronaryflow reserve >2·5. The baseline characteristics of bothpatient groups are depicted in Table 1. Patients with animpaired coronary flow reserve were older and showed asignificant difference with respect to the use of nitrates.

Table 1 Baseline characteristics of the 86 study patients

Characteristics CFR �2·5(n=42)

CFR >2·5(n=44) P

Age, years�SD 60�9 55�10 0·02Male 29 (69) 36 (82) nsSystemic hypertension 14 (33) 11 (25) nsCigarette smoking 9 (21) 14 (32) nsTotal cholesterol >6·5 mmol . l�1 20 (48) 14 (32) nsDiabetes mellitus 6 (14) 2 (5) nsPrevious myocardial infarction 9 (21) 9 (20) nsUnstable AP 21 (50) 21 (48) nsMedication

Beta-blocker 22 (52) 28 (64) nsCalcium antagonist 4 (10) 4 (9) nsNitrates 32 (76) 22 (50) 0·005

Coronary artery dilatedLAD 19 (45) 28 (64) nsRCA 11 (26) 8 (18) nsLCx 12 (29) 8 (18) ns

Type of lesion (ACC/AHA)A 7 (17) 5 (11) nsB 35 (83) 39 (89) nsC 0 (0) 0 (0) ns

Length of lesion >10 mm 6 (14) 6 (14) ns

CFR=coronary flow reserve; AP=angina pectoris; LAD=leftanterior descending coronary artery; RCA=right coronaryartery; LCx=left circumflex coronary artery; ACC=AmericanCollege of Cardiology; AHA=American Heart Association.Values are n (%).

Haemodynamic, angiographic and Doppler flowvariablesThere were no significant differences in heart rate, meanaortic pressure and angiographical variables betweeneither patient group before or after PTCA (Table 2).Coronary flow reserve before PTCA was lower inpatients with an impaired coronary flow reserve afterPTCA (1·47�0·7 vs 1·84�0·7; P<0·05). An impairedcoronary flow reserve after PTCA was associatedwith a significant increase in b-APV (17�7 to23�12 cm . s�1; P<0·05), while the increase in h-APVwas similar in both patient groups (26�16 to45�16 cm . s�1 vs 27�12 to 49�15 cm . s�1). At6-months follow-up, b-APV in patients with an impairedcoronary flow reserve normalized to a value of18�7 cm . s�1 (P<0·05), while the b-APV of thepatients with an unimpaired coronary flow reserveremained constant (15�6 cm . s�1). A multivariateanalysis of clinical, angiographic and haemodynamicvariables revealed that coronary flow reserve beforeangioplasty was the only independent predictor ofcoronary flow reserve directly after PTCA (P=0·03).



Impaired coronary flow reserve after optimalballoon angioplasty in relation to the clinical

outcome

Death or myocardial infarction did not occur duringthe 6-month follow-up period. Recurrence of anginaat follow-up was more frequent in patients with animpaired coronary flow reserve after PTCA thanin patients with a coronary flow reserve >2·5 (17 of37 patients, 46% vs 10 of 44 patients, 23%; P<0·05).Second, the incidence of target lesion revascularizationin patients with an impaired coronary flow reserve afterPTCA was higher than in patients with a coronary flowreserve >2·5 after PTCA (15 of 42 patients, 36% vs 7 of44 patients, 16%; P<0·05). Furthermore, there was atrend towards a higher restenosis rate in patients with animpaired coronary flow reserve after PTCA (12 of 42patients, 29% vs 7 of 44 patients, 16%; P=0·15). Theangiographic and Doppler flow data of the patients withan impaired coronary flow reserve directly after PTCAand target lesion revascularization at follow-up weredivided into two groups, with and without restenosis(Table 3). Patients with target lesion revascularization


but without restenosis had a low coronary flow reservedue to an increased baseline APV directly after PTCA,that remained elevated during follow-up. In contrast, anelevated baseline APV directly after PTCA in thepatients with restenosis decreased at follow-up.

Table 2 Quantitative coronary angiography and coronary blood flow velocity data of the 86 study patients

CFR �2·5 (n=42) CFR >2·5 (n=44)

Before BA After BA Follow-up Before BA After BA Follow-up

HR, beats . min�1 69�12 68�11 68�10 71�12 70�11 70�11MAP, mmHg 91�14 90�12 92�12 91�11 90�13 91�12% DS 60�10 29�6* 41�13* 60�9 30�4* 38�15*MLD, mm 1·08�0·27 1·92�0.28* 1·66�0·49* 1·15�0·33 2·01�0·39* 1·75�0·50*b-APV, cm . s�1 17�7 23�12* 18�7* 15�6 14�6† 15�6†h-APV, cm . s�1 26�16 45�16* 46�23 27�12 49�15* 47�18CFR 1·47�0·7 1·99�0·8* 2·49�0·9* 1·84�0·7† 3·47�0·8*† 3·11�0·9*†

*P<0·05 vs paired data of previous measurement.†P<0·05 vs patients with a coronary flow reserve �2·5 after balloon angioplasty.BA=balloon angioplasty; HR=heart rate; MAP=mean arterial pressure; DS=diameter stenosis; MLD=minimum lumen diameter;b-APV=baseline average peak velocity; h-APV=hyperaemic average peak velocity; CFR=coronary flow reserve.

Table 3 Quantitative coronary angiography and coronary flow velocity of the patients with an impaired coronary flowreserve and a target lesion revascularization at follow-up

With restenosis (n=10) Without restenosis (n=5)


HR, beats . min�1 71�11 70�12 72�10 71�12 72�9 71�8MAP, mmHg 94�8 93�10 95�11 87�11 88�10 85�15% DS 60�8 30�3* 59�7* 54�11 30�4* 42�6*†MLD, mm 1·09�0·34 1·92�0·31* 1·14�0·35* 1·27�0·49 1·97�0·56* 1·76�0·58*†b-APV, cm . s�1 18�6 22�11 16�5* 22�12 24�11 28�10†h-APV, cm . s�1 24�14 44�17* 25�12* 34�18 45�20 50�21†CFR 1·33�0·4 2·02�0·4* 1·50�0·3* 1·51�0·4 1·89�0·5* 1·82�0·6

*P<0·05 vs paired data of previous measurement.†P<0·05 vs patient restenosis.Abbreviations as for Table 2.

Coronary flow reserve alterations duringfollow-up in patients without restenosis

Baseline characteristicsOf the total of 86 patients, 67 did not demonstraterestenosis at 6 months follow-up. Angiographic andDoppler flow data were complete in 64 patients. Thirty-eight patients (59%) showed an increase of coronaryflow reserve at follow-up. The remaining 26 patients(41%) showed a decrease at follow-up. This analysis wasperformed independent of an initial cut-off value ofcoronary flow reserve. There were no significant differ-ences in baseline characteristics between patients show-ing an increase or decrease of coronary flow reserve atfollow-up (Table 4).


Haemodynamic, angiographic and Doppler flowvariablesThere was an inverse relationship between coronary flowreserve immediately after PTCA and the alterations ofcoronary flow reserve at follow-up (Fig. 1), i.e. initiallyhigh coronary flow reserve values showed a decrease atfollow-up (n=26) and initially low coronary flow reservevalues showed an increase at follow-up (n=38). Amultivariate analysis of all clinical, angiographic andhaemodynamic variables revealed the distal coronaryflow reserve after PTCA to be the only explanatoryvariable for the alterations of coronary flow reserve

Table 4 Baseline characteristics of the patients withoutrestenosis. Values are not significant

Characteristics CFR increase(n=38)

CFR decrease(n=26)

Age, years�SD 58�9 58�11Male 27 (71) 22 (85)Hypertension 14 (37) 6 (23)Cigarette smoking 13 (34) 5 (19)Total cholesterol >6·5 mmol . l�1 15 (39) 11 (42)Diabetes mellitus 2 (5) 2 (8)Previous myocardial infarction 10 (26) 6 (23)Unstable AP 19 (50) 13 (50)Medication

Beta-blocker 22 (58) 14 (54)Calcium antagonist 6 (16) 1 (4)Nitrates 26 (68) 15 (58)

Coronary artery dilatedLAD 16 (42) 16 (62)RCA 11 (29) 6 (23)LCx 11 (29) 4 (15)

Type of lesion (ACC/AHA)A 6 (16) 4 (15)B 32 (84) 22 (85)C 0 (0) 0 (0)

Length of lesion >10 mm 7 (18) 4 (15)

Abbreviations as in Table 1.

1·03·0

6·0

Change of CFR at follow-up

CF

R a

fter

PT

CA

–2·0

1·5

2·0

2·5

3·0

3·5

4·0

4·5

5·0

5·5

–1·5 –1·0 –0·5 2·52·01·51·00·50·0

Figure 1 Relationship between coronary flow reserve(CFR) immediately after PTCA and alterations of cor-onary flow reserve at follow-up in the patients withoutrestenosis (r= �0·67, P<0·0001).

at follow-up. There were no significant differences inhaemodynamic and angiographical data before or afterPTCA or at late follow-up in patients showing anincrease or decrease of coronary flow reserve at follow-up. Patients showing an increase of coronary flowreserve at follow-up were characterized by an initiallylow coronary flow reserve due to an impaired h-APVthat increased at follow-up (43�16 vs 55�19 cm . s�1;P<0·05). In contrast, patients with an initially highcoronary flow reserve directly after PTCA showed adecrease in coronary flow reserve at follow-up due toa reduction of h-APV (51�15 vs 45�8 cm . s�1;P<0·05). Minimum lumen diameter decreased inpatients with a coronary flow reserve decrease duringfollow-up, while in patients who showed an increase ofcoronary flow reserve, minimum lumen diameter did notshow a statistically significant change (Table 5).

Discussion

The present study demonstrates that an impaired cor-onary flow reserve (�2·5) is a frequent finding afteroptimal balloon angioplasty due to an increased b-APV,as compared to patients with a coronary flow reserve>2·5. An impaired coronary flow reserve directly afterPTCA is associated with a higher incidence of recurrentangina and target lesion revascularization during the6-month follow-up period. In patients without resteno-sis, there is an inverse relationship between coronaryflow reserve directly after PTCA and alterations ofcoronary flow reserve at follow-up, i.e. initially highcoronary flow reserve values decrease while low cor-onary flow reserve values increase. This phenomenon isassociated with changes in h-APV, suggesting that itrelates to epicardial remodelling.

The direct effect of optimal balloonangioplasty on distal coronary flow reserve

and its relation to clinical outcome

There is controversy in the literature regarding the effectof PTCA on coronary flow reserve. Several studiesreported a direct normalization of coronary flow reserveafter coronary intervention[1,2], while other studies re-ported a temporary decrease of coronary flow reserveafter PTCA followed by normalization within weeksafter the procedure[3–6]. The initial report of theDEBATE study indicated that an optimal angiographi-cal result (diameter stenosis <35%) as assessed by off-line quantitative coronary angiography in combinationwith a coronary flow reserve >2·5 is associated with alow target revascularization and restenosis rate[9]. Thepresent study shows that approximately 50% of thepatients with an optimal angiographical result in thisselected cohort of patients do not reach this arbitrarylevel of 2·5 due to an elevated b-APV. Multivariateanalysis revealed coronary flow reserve before PTCA as



Table 5 Quantitative coronary angiography and coronary flow velocity data of patients without restenosis

CFR increase (n=38) CFR decrease (n=26)


HR, beats . min�1 70�12 68�11 67�7 72�12 70�11 72�11MAP, mmHg 91�15 89�13 93�11 91�10 91�13 89�13% DS 61�10 28�6* 33�9* 59�9 30�4* 35�9*MLD, mm 1·05�0·27 1·93�0·28* 1·85�0·36 1·16�0·33 2·02�0·39* 1·85�0·49*b-APV, cm . s�1 16�7 20�12 18�7 15�6 16�6 17�8h-APV, cm . s�1 26�16 43�16* 55�19* 27�12 51�15*† 45�8*†CFR 1·62�0·7 2·33�0·8* 3·23�0·9* 1·87�0·6 3·50�0·9*† 2·73�0·6*‡

*P<0·05 vs paired data of previous measurement.†P<0·05 vs patients with an increase of coronary flow reserve at follow-up.Abbreviations as for Table 2.


The long-term effect of optimal balloonangioplasty on distal coronary flow reserve

Several previous clinical studies have demonstrated thatballoon angioplasty may result in a temporary reducedcoronary flow reserve that normalizes within severalweeks to months after the procedure. It has been sug-gested that this phenomenon relates to either epicardial

the only independent predictor of coronary flow reservedirectly after PTCA, suggesting that haemodynamicfactors before the intervention are responsible for theimpaired coronary flow reserve after PTCA ratherthan procedure related factors such as particulateembolization[4,11–13]. An enhanced baseline flow velocitynormalizes at late follow-up in patients without targetlesion revascularization, suggesting that an impairedcoronary flow reserve directly after PTCA is due totemporarily disturbed autoregulation. The incidence ofimpaired coronary flow reserve after PTCA is in accord-ance with a recent study of van Liebergen et al.[3] Thestudy of van Liebergen et al. was not designed toevaluate the relationship between the haemodynamicfindings after PTCA and clinical outcome of thepatients. The present study is the first clinical reportshowing that patients with an impaired coronary flowreserve, due to an elevated b-APV after optimal balloonangioplasty, show an increased incidence of recurrenceof angina, a higher target lesion revascularization rateand a trend towards a high restenosis rate during a6-month follow-up period. The higher target lesionrevascularization rate in this patient cohort is partly dueto patients developing restenosis. The association be-tween enhanced baseline APV directly after optimalballoon angioplasty and the development of restenosis isspeculative. Vascular shear stress is directly proportionalto blood flow velocity and inversely proportional tovessel diameter[14,15], indicating that patients with animpaired coronary flow reserve due to a higher b-APVare exposed to increased shear stress. The present find-ings suggest that shear stress is a relevant haemody-namic factor in the remodelling process followingballoon angioplasty. A recent study by Dirksen et al.evaluated cell type distribution of the atheroscleroticplaque in relation to shear stress[16]. This study showedthat smooth muscle cell proliferation predominantlyoccurs at the downstream part of the plaque exposed tolow shear stress in contrast to proliferation of macro-phages that occurs in the upstream regions exposed tohigh shear stress. One may speculate that turbulent flowacross a residual stenosis after PTCA results in lowshear stress at the distal site of the lesion and, hence,

preferential induction of smooth muscle cell prolifer-ation. Several recent clinical studies have demonstratedthat a reduced coronary flow reserve after balloonangioplasty may increase after stent implantationtowards values of the reference artery coronaryflow reserve[7,8]. It has been postulated that thisimprovement in coronary flow reserve is a result of anincrease in h-APV due to a reduction in the residualobstruction after stent implantation. Recent reportsby van Liebergen et al. and Kern et al. showed that acoronary flow reserve remains unaltered after stentimplantation in those patients with a reduced coronaryflow reserve after balloon angioplasty due to an elevatedb-APV[3,17].

Another category of patients warranting an interven-tion without angiographical restenosis show a low cor-onary flow reserve due to a persistent high baselineblood flow velocity, probably secondary to disturbedautoregulation and/or diffuse mild coronary atheroscler-osis[18]. It is conceivable that patients with an impairedcoronary flow reserve following optimal balloon angi-oplasty may benefit from stent implantation in order toreduce the incidence of angiographical restenosis atfollow-up. However, it is doubtful whether the clinicaloutcome of patients with a disturbed autoregulationand/or diffuse mild coronary atherosclerosis, as anunderlying mechanism of a reduced coronary flow re-serve, may be improved by stent implantation. Thishypothesis needs to be verified in subanalyses of recentlycompleted multicentre studies regarding optimal balloonangioplasty, guided by Doppler flow measurements, vsstent implantation (DESTINI, DEBATE-2)[19,20].


remodelling and/or slow adaptation of the microvascu-lar bed after balloon angioplasty[3–6]. This phenomenonwas studied in more detail in a recent report from vanLiebergen et al. showing that a reduced coronary flowreserve after PTCA is related to an initially high b-APVthat reduces at follow-up in association with a normal-ization of the coronary flow reserve of the target vessel[3].These findings were observed after both balloon angi-oplasty and stent implantation in patients without rest-enosis, suggesting that this phenomenon does not relateto epicardial remodelling, but rather to slow adaptationof the microvascular bed. The present study was per-formed in a selected cohort of patients with optimalballoon angioplasty in order to minimize the confound-ing effect of residual lumen obstruction after PTCA. Thefindings in the present study are in accordance with thereport of van Liebergen et al., i.e. initially high coronaryflow reserve values decrease, while low coronary flowreserve values increase within 6 months after the pro-cedure in patients without restenosis. Moreover, step-wise linear regression analysis revealed in both studiesthat the distal coronary flow reserve directly after PTCAwas the only independent explanatory variable for thealterations of coronary flow reserve at follow-up. Incontrast to the study of van Liebergen et al., the presentstudy demonstrates that the changes of coronary flowreserve after balloon angioplasty were positively associ-ated with alterations in h-APV. A reduction in coronaryflow reserve was observed in patients showing a signifi-cant decrease in h-APV, while an increase in coronaryflow reserve was noted in patients with a significantincrease in h-APV. These findings could not be attrib-uted to differences in haemodynamic variables or medi-cation, indicating that this phenomenon presumablyrelates to a variable response in epicardial remodelling,known to occur after coronary angioplasty[21]. Thehypothesis that the hyperaemic average peak velocity isrelated to epicardial remodelling is strengthened by theangiographical data in this study. Minimum lumendiameter decreased in patients with a coronary flowreserve decrease during follow-up, while in patients whoshowed an increase of coronary flow reserve, minimumlumen diameter did not show a statistically significantchange. There are several reports showing that an im-paired coronary flow reserve after stenting is associatedwith an enhanced baseline blood flow velocity similar tothe result after balloon angioplasty[3,17]. A residual ob-struction after balloon angioplasty, as compared to stentimplantation, is expected to be identified by a differencein hyperaemic blood flow velocity. It can be anticipatedthat the dynamic changes in hyperaemic blood flowvelocity, probably related to epicardial remodelling inthe present study, alter following stent implantation thatis associated with less remodelling as compared toballoon angioplasty. The study of van Liebergen et al.and the present study indicate two distinct haemody-namic responses during follow-up after balloon angi-oplasty, i.e. an increase in impaired coronary flowreserve directly after PTCA due to normalization ofautoregulation or epicardial remodelling. A recent study

by Dangas et al. demonstrated that the clinical outcomeafter balloon angioplasty depends, in part, on positive ornegative remodelling of the atherosclerotic plaque, asdetermined by intravascular ultrasound imaging beforecoronary intervention[22]. The present study adds tothese findings, showing that the clinical outcome afterballoon angioplasty may also be predicted using theinformation obtained on coronary haemodynamicsdirectly after PTCA.

Limitations

This study concerns a post-hoc analysis of a multicentrestudy. Since this was a prospective observational studywith investigators not being blinded to the Doppler flowvelocity and quantitative coronary angiography derivedresults, the incidence of target lesion revascularizationcould be driven by these measurements. However, theindication for target lesion revascularization was pre-defined based on patients angina status and/or theobjective documentation of ischaemia during a stresstest at follow-up. The clinical significance of the presentfindings on coronary haemodynamics following balloonangioplasty warrants evaluation in a prospective trial.Patients requiring stent implantation following balloonangioplasty were excluded from the trial. This restrictionof study protocol did not facilitate an analysis of cor-onary haemodynamics after PTCA with minimal influ-ence by epicardial remodelling. The changes in arterialdimensions were assessed by quantitative coronaryangiography without intravascular ultrasound imagingmandatory for appropriate evaluation of epicardial re-modelling, estimation of coronary volume flow changesor shear stress analysis. Blood flow velocity changesbefore and after PTCA and at follow-up were deter-mined by repeated measurements at the same site of theDoppler guide wire tip assessed by coronary angiogra-phy. It is possible that changes in medication, haemo-dynamic differences and variations in guidewire-tipposition at the time of the initial procedure and at6-month follow-up may have contributed to variationsnoted in coronary blood flow velocity parameters.Finally, this study involved a selected cohort of patientswith one-vessel disease and a normal left ventricularfunction, limiting extrapolation of the present findingsto other patient groups.

Clinical implications

The present study shows that an impaired coronary flowreserve (�2·5) is a frequent finding after optimal bal-loon angioplasty, related to an elevated b-APV. On theother hand, an impaired coronary flow reserve mayimprove during follow-up in those patients with a nor-mal b-APV following optimal balloon angioplasty. Thepresent findings indicate that the evaluation of coronaryhaemodynamics following balloon angioplasty is rel-evant for clinical decision making regarding adjunctive



stent implantation. Recent insight from clinical trialsindicate that a strategy of aggressive balloon angioplastyand bail-out stenting or optimal balloon angioplastyusing intravascular ultrasound, yields a clinical outcomesimilar to elective stent-implantation[23]. The presentstudy adds to these findings, showing that intracoronaryphysiological parameters are useful to avoid unnecessarystent implantation. However, these diagnostic tech-niques prolong procedure time and this, in addition toequipment costs, may explain why interventional cardi-ologists are reluctant to advocate a strategy of pro-visional stenting compared to elective or direct stentimplantation[24]. The results of the DEBATE trial led toa conduction of several multicentre studies (DEBATE 2,DESTINI, FROST) evaluating the cost-effectivenessof provisional stent implantation guided by analysisof coronary haemodynamics following balloon angi-oplasty[19,20,25]. The results of these studies underlinethe significance of intracoronary physiological par-ameters for evaluation of clinical outcome followingpercutaneous coronary interventions.

The DEBATE study was supported by a grant from Cardio-metrics, Inc. Jan J. Piek, MD, is clinical investigator for theNetherlands Heart Foundation (grant# D96.020). The authorsgratefully acknowledge the skilled assistance of Angelina de Jong,MSc, for preparation of the manuscript.

References

[1] Hodgson JMcB, Riley RS, Most AS, Williams DO. Assess-ment of coronary flow reserve using digital angiographybefore and after successful percutaneous transluminalcoronary angiography. Am J Cardiol 1987; 60: 61–5.

[2] Zijlstra F, Reiber JHC, Juilliere Y, Serruys PW. Normaliz-ation of coronary flow reserve by percutaneous transluminalcoronary angioplasty. Am J Cardiol 1988; 61: 55–60.

[3] van Liebergen RA, Piek JJ, Koch KT, de Winter RJ, Lie KI.Immediate and long-term effect of balloon angioplasty orstent implantation on the absolute and relative coronaryblood flow velocity reserve. Circulation 1998; 20: 2133–40.

[4] Nanto S, Kodama K, Hori M et al. Temporal increase inresting coronary blood flow causes an impairment of coronaryflow reserve after coronary angioplasty. Am Heart J 1992;123: 28–36.

[5] Wilson RF, Johnson MR, Marcus ML et al. The effect ofcoronary angioplasty on coronary flow reserve. Circulation1988; 77: 873–85.

[6] Uren NG, Crake T, Lefroy DC, de Silva R, Davies GJ, MaseriA. Delayed recovery of coronary resistive vessel function aftercoronary angioplasty. J Am Coll Cardiol 1993; 21: 612–21.

[7] Kern MJ, Dupouy P, Drury JH et al. Role of coronary arterylumen enlargement in improving coronary blood flow afterballoon angioplasty and stenting: a combined intravascularultrasound Doppler flow and imaging study. J Am CollCardiol 1997; 29: 1520–7.

[8] Haude M, Caspari G, Baumgart D, Brennecke R, Meyer J,Erbel R. Comparison of myocardial perfusion reserve beforeand after coronary balloon predilatation and after stentimplantation in patients with postangioplasty restenosis.Circulation 1996; 94: 286–97.


[9] Serruys PW, Di Mario C, Piek JJ et al., for the D.E.B.A.T.E.study group. Prognostic value of intracoronary flow velocityand diameter stenosis in assessing the short- and long-termoutcomes of coronary balloon angioplasty: The D.E.B.A.T.E.Study (Doppler Endpoints Balloon Angioplasty TrialEurope). Circulation 1997; 96: 3369–77.

[10] Doucette JW, Corl PD, Payne HM et al. Validation of aDoppler guide wire for intravascular measurement of cor-onary artery flow velocity. Circulation 1992; 85: 1899–911.

[11] Bates ER, McGillem MJ, Beals TJ et al. Effect of angioplastyinduced endothelial denudation compared with medialinjury on regional coronary blood flow. Circulation 1987; 76:710–6.

[12] Fischell TA, Bausback KN, McDonald TV. Evidence foraltered epicardial coronary artery autoregulation as a cause ofdistal coronary vasoconstriction after successful percutaneoustransluminal coronary angioplasty. J Clin Invest 1990; 86:575–84.

[13] Erbel R, Heusch G. Coronary Microembolization. J Am CollCardiol 2000; 36: 22–4.

[14] Gnasso A, Carallo C, Irace C et al. Association betweenintima-media thickness and wall shear stress in commoncarotid arteries in healthy male subjects. Circulation 1996; 94:3257–62.

[15] Chobanian AV. The influence of hypertension and otherhemodynamic factors in atherogenesis. Prog Cardiovasc Dis1983; 26: 177–96.

[16] Dirksen MT, van der Wal AC, van den Berg FM, van derLoos CM, Becker AE. Distribution of inflammatory cellsin atherosclerotic plaques related to the direction of flow.Circulation 1998; 19: 2000–3.

[17] Kern MJ, Puri S, Bach RG et al. Abnormal coronary flowvelocity reserve after coronary artery stenting in patients; roleof relative coronary reserve to assess potential mechanisms.Circulation 1999; 100: 2491–8.

[18] Anderson HV, Stokes MJ, Leon M, Abu-Halawa SA, StuartY, Kirkeeide RL. Coronary artery flow velocity is related tolumen area and regional left ventricular mass. Circulation2000; 102: 48–54.

[19] Serruys PW, de Bruyne B, Carlier S et al., on behalf of theDEBATE II Study Group. Randomized comparison of pri-mary and provisional balloon angioplasty guided by flowvelocity measurement. Circulation 2000; 102: 2930–7.

[20] Di Mario C, Moses JW, Anderson TJ et al., on behalf ofthe DESTINI Study Group. Randomized comparison ofelective stent implantation and coronary balloon angioplastyguided by online quantitative angiography and intracoronaryDoppler. Circulation 2000; 102: 2938–44.

[21] Mintz GS, Popma JJ, Pichard AD et al. Arterial remodelingafter coronary angioplasty: a serial intravascular ultrasoundstudy. Circulation 1996; 94: 35–43.

[22] Dangas G, Mintz GS, Mehran R et al. Preintervention arterialremodeling as an independent predictor of target-lesionrevascularization after nonstent coronary intervention: ananalysis of 777 lesions with intravascular ultrasound imaging.Circulation 1999; 99: 3149–54.

[23] Narins CR, Holmes DR Jr, Topol EJ. A call for provisionalstenting: the balloon is back. Circulation 1998; 97: 1298–305.

[24] Cantor WJ, Peterson ED, Popma JJ et al. Provisional stentingstrategies: systematic overview and implications for clinicaldecision-making. J Am Coll Cardiol 2000; 36: 1142–51.

[25] Lafont A, Dubois-Rande JL, Steg PG et al., the FROSTStudy Group. The French randomized optimal stenting trial: aprospective evaluation of provisional stenting guided by cor-onary velocity reserve and quantitative coronary angiography.J Am Coll Cardiol 2000; 36: 404–9.

The immediate and long-term effect of optimal balloon angioplasty on the absolute coronary blood...

Documents

Transcript of The immediate and long-term effect of optimal balloon angioplasty on the absolute coronary blood...