Chronic Arterial Responses to Overlapping Paclitaxel-Eluting Stents

CPI

JJEL

NO

Oc(

Btm

Mwmms

Rm0PmPsP

CpaC

FR&KMSWM

M

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 , N O . 2 , 2 0 0 8

© 2 0 0 8 B Y T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N I S S N 1 9 3 6 - 8 7 9 8 / 0 8 / $ 3 4 . 0 0

P U B L I S H E D B Y E L S E V I E R I N C . D O I : 1 0 . 1 0 1 6 / j . j c i n . 2 0 0 7 . 1 2 . 0 0 5

hronic Arterial Responses to Overlappingaclitaxel-Eluting Stents

nsights From Serial Intravascular Ultrasound Analyses in the TAXUS-V and -VI Trials

iro Aoki, MD, PHD,* Gary S. Mintz, MD, FACC,* Neil J. Weissman, MD, FACC,†. Tift Mann, MD, FACC,‡ Louis Cannon, MD, FACC,§ Joel Greenberg, MD, FACC,�berhard Grube, MD, FACC,¶ A. R. Zaki Masud, MD, FACC,# Joerg Koglin, MD,**azar Mandinov, MD, PHD,** Gregg W. Stone, MD, FACC*

ew York and Buffalo, New York; Washington, DC; Raleigh, North Carolina; Petoskey, Michigan;rlando, Florida; Seigburg, Germany; and Natick, Massachusetts

bjectives The purpose of this study was to use intravascular ultrasound (IVUS) to investigatehronic arterial responses at the site of and adjacent to overlapping paclitaxel-eluting TAXUS stentsPES) compared with overlapping bare-metal stents (BMS).

ackground Increased paclitaxel dose in the PES-overlap region might be associated with arterialoxicity expressed as excessive expansive remodeling, incomplete stent apposition, or aneurysm for-ation.

ethods In the TAXUS-V and -VI trials, 51 patients with overlapping stents (27 PES and 24 BMS)ere imaged with serial IVUS immediately after procedure and at 9 months. The IVUS measure-ents included intimal hyperplasia (IH), peri-stent plaque plus media (P&M), and external elasticembrane (EEM) areas. Vascular responses were assessed at the proximal and distal single stenttrut regions and the central overlap region.

esults Compared with BMS, all 3 PES stent regions showed: 1) significantly decreased IH (proxi-al: 0.97 � 1.06 mm2 vs. 3.12 � 2.40 mm2, overlap: 0.74 � 0.91 mm2 vs. 3.23 � 1.75 mm2, distal:.88 � 0.85 mm2 vs. 2.69 � 1.49 mm2, all p � 0.05); and 2) increased P&M and EEM areas (Delta&M; proximal: 0.96 � 1.36 mm2 vs. �0.02 � 1.48 mm2, overlap: 1.56 � 1.88 mm2 vs. 0.29 � 1.82m2, distal: 1.03 � 1.81 mm2 vs. 0.11 � 0.89 mm2, all p � 0.05). The IH and changes in EEM and&M areas were not significantly different in both the BMS and PES groups comparing the singletent strut and overlap regions. Incomplete stent apposition did not occur at the site of overlappingES in any patient.

onclusions Nine months after stent implantation, neointimal tissue growth was reduced and ex-ansive remodeling was greater with PES compared with BMS—effects that were not exaggeratedt the overlap region of PES. (J Am Coll Cardiol Intv 2008;1:161–7) © 2008 by the Americanollege of Cardiology Foundation

rom *Columbia University Medical Center and Cardiovascular Research Foundation, New York, New York; †Cardiovascularesearch Institute, Washington Hospital Center, Washington, DC; ‡Wake Heart Research, Raleigh, North Carolina; §Cardiac

Vascular Research Center of Northern Michigan, Petoskey, Michigan; �Florida Heart Institute, Orlando, Florida; ¶Helioslinikum, Siegburg, Germany; #Buffalo General Hospital, Buffalo, New York; and **Boston Scientific Corp., Natick,assachusetts. Drs. Mintz, Weissman, Mann, Cannon, Greenberg, Grube, and Stone have received research grants from Boston

cientific; Drs. Mintz, Weissman, and Cannon are on Boston Scientific Speaker’s Bureau and have received honoraria; Drs. Mintz,eissman, Cannon, Greenberg, and Stone are on the Boston Scientific Consultant/Advisory Boards; and Drs. Koglin andandinov are employed by Boston Scientific.

anuscript received October 12, 2007; revised manuscript received December 3, 2007, accepted December 10, 2007.

Gelmei-(bdsasmPc

(tD

tTsaidrsrsi

M

Pgbd1tNp(lwwnwww4tttaIdaaptoQfncooapCHt

tl(lm((Tcat

Aa

B

C

D

Em

I(

Ia

Iu

Ps

P(

Ss

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S , V O L . 1 , N O . 2 , 2 0 0 8

A P R I L 2 0 0 8 : 1 6 1 – 7

Aoki et al.

Chronic Arterial Response to Overlapping PES

162

rowing evidence supports the clinical benefit of paclitaxel-luting stents (PES) across a broad range of patient andesion types (1–4). However, stent overlap is associated with

ore frequent angiographic restenosis even with drug-luting stents (DES) (2). Whereas only 1 stent/lesion wasmplanted in the TAXUS-II and -IV trials, TAXUS-V andVI permitted overlapping stents for longer lesions1,2,5,6). An experimental study using balloon-injured rab-it iliac arteries showed that overlapping PES resulted inelayed healing compared with adjacent proximal and distalingle PES-layer segments (7) and greater inflammationnd fibrin deposition than overlapping sirolimus-elutingtents (SES). Conversely, studies in porcine coronary arteryodels have shown comparable healing for overlappingES (8). Finally, clinical outcomes after overlapping PESompared with overlapping SES were similar in some

real-world observational studies(9,10).

Vascular responses after stentimplantation may be assessedwith serial angiographic and in-travascular ultrasound (IVUS)measurements. Although per-cent in-stent obstruction in theoverlap region based solely onfollow-up IVUS analysis wasperformed in the TAXUS trials(11), peri-stent responses toDES (i.e., outside the stent orbetween the stent and the vesselwall) have been incompletelycharacterized (12). Early assess-ments of volumetric changesoutside DES reported a dose-dependent and partially revers-ible increase in arterial dimen-sions over time when comparedwith bare-metal stents (BMS)

13). These findings were hypothesized to reflect a con-rolled biologic response to the implantation of polymericES.Along with the comparison of different dose formula-

ions, such as slow- versus moderate-release PES in theAXUS-II trial (14), the implantation of overlapping

tents represents another internally controlled modelllowing examination of chronic vessel reactions to vary-ng local paclitaxel doses by comparing the proximal andistal single stent strut regions with the central overlapegion. The objective of the present analysis was to useerial (baseline and follow-up) IVUS studies from theandomized TAXUS-V and -VI multicenter studies totudy chronic arterial responses to overlapping PES

bbreviationsnd Acronyms

MS � bare-metal stent(s)

SA � cross-sectional area

ES � drug-eluting stent(s)

EM � external elasticembrane

H � intimal hyperplasiaarea)

SA � incomplete stentpposition

VUS � intravascularltrasound system

ES � paclitaxel-elutingtent(s)

&M � plaque and mediaarea)

ES � sirolimus-elutingtent(s)

mplantation. p

ethods

atient selection. The study designs and clinical and angio-raphic outcomes of the prospective, multicenter, double-lind, controlled TAXUS-V and -VI trials have beenescribed elsewhere (2,6). Briefly, TAXUS-V randomized,156 patients to either slow-release PES or visually indis-inguishable Express2 BMS (both Boston Scientific Corp.,atick, Massachusetts). The TAXUS-VI randomized 446

atients to either moderate-release PES or Express2 BMSboth Boston Scientific Corp.). Patients with single de novoesions, 10 to 46 mm in length, in a native coronary arteryith reference vessel diameter between 2.25 and 4.0 mmere enrolled in TAXUS-V, whereas patients with single deovo lesions 18 to 40 mm long in a native coronary arteryith reference vessel diameter between 2.5 and 3.75 mmere enrolled in TAXUS-VI. By protocol, multiple stentsere required for lesions �26 mm in length, in which casemm of stent overlap was specified in both studies to ensure

he absence of gaps. Of 1,602 total patients, 80 who werereated with overlapping stents were assigned to a substudyhat included baseline and follow-up IVUS imaging andnalysis. However, among these 80 patients, 19 did not haveVUS at implantation; and the overlap region could not beifferentiated from the adjacent single-PES layer in andditional 10 patients. The remaining 51 patients (27 PESnd 24 BMS) were included in the present study. Allatients provided written informed consent, and this inves-igation was approved at the local institutional review boardr ethics committee of each participating center.uantitative IVUS analysis. The IVUS imaging was per-

ormed after intracoronary administration of 0.1 to 0.2 mgitroglycerin with motorized pullback (0.5 mm/s) andontemporary commercial scanners. Images were recordednto s-VHS videotape or digitally onto CD or MO disc forffline core laboratory analysis. These images were analyzedccording to published standards, with computerizedlanimetry (Tapemeasure, Indec Inc., Mountain View,alifornia) by an independent core laboratory (Washingtonospital Center, Washington, DC) that remained blinded

o treatment allocation (15).The stented lesion was divided into 3 regions according

o location: 1) proximal single stent layer; 2) middle over-apping double stent layers; and 3) distal single stent layerFig. 1). External elastic membrane (EEM), stent, andumen cross-sectional areas (CSA) were measured for each

illimeter. Peri-stent plaque and media (P&M) CSAP&M � EEM � stent) and intimal hyperplasia (IH) CSAIH � stent � lumen) were calculated for each millimeter.he results are expressed as the average of the individual

ross-sectional slices within each region. Although everyttempt was made to analyze EEM CSA for each millime-er throughout the stented segment, it was not possible in all
atients, because in come cases the stent artifact obscured

ttortg

mepctfpqt

PIScvpac�

R

Tti

dle ov


A P R I L 2 0 0 8 : 1 6 1 – 7

Aoki et al.


163

he media-adventitia border. Therefore, for the purpose ofhe current analysis, we excluded segments in which �25%f the length of the EEM was not identifiable (16). As aesult, EEM analysis was possible in 26 patients (96.3%) inhe PES group and in 21 patients (87.5%) in the BMSroup.

Incomplete stent apposition (ISA) was defined as 1 orore stent struts clearly separated from the vessel wall with

vidence of blood flow behind the struts and not overlap-ing a side branch (13,17). Incomplete stent apposition waslassified as 1 of 3 types. Resolved ISA was defined as ISAhat was present at implantation but disappeared duringollow-up. Persistent ISA was defined as ISA that wasresent both at implantation and at follow-up. Late ac-uired ISA was defined as ISA that was absent at implan-ation but present at follow-up. When ISA was detected,

Figure 1. Three Regions in the Overlapping Stented Lesion

The stented lesion was divided into 3 regions (proximal single stent layer, mid

Table 1. Baseline Characteristics

Age, yrs 6

Gender male, %

Current smoking, %

Diabetes mellitus, %

Hypertension, %

Unstable angina, %

Prior myocardial infarction, %

Statin prescription at discharge, %*

Treated vessel

Left anterior descending coronary artery, %

Left circumflex coronary artery, %

Right coronary artery, %

Reference vessel diameter, mm

Lesion length, mm 2

Total stent length, mm 3

Average stent diameter, mm

*TAXUS-V patients only.

BMS � bare-metal stent; PES � paclitaxel-eluting stent.

&M area was calculated as EEM CSA � stent CSA �SA CSA.tatistical analysis. Discrete variables are displayed as per-entages and tested with Fisher exact test. Continuousariables are expressed as mean � SD and compared withaired or unpaired Student t test or analysis of variance asppropriate. Linear regression was performed to assess theorrelation between different IVUS outcomes. A value of p

0.05 was considered statistically significant.

esults

he PES and BMS groups were well-matched with respecto clinical and lesion characteristics (Table 1). Averagemplanted stent length, stent diameter, and overlap length

erlapping double stent layers, and distal single stent layer).

� 27) BMS (n � 24) p Value

8.53 (27) 63.75 � 9.41 (24) 0.94

(19/27) 66.7% (16/24) 1.00

(3/27) 25.0% (6/24) 0.28

(11/27) 37.5% (9/24) 1.00

(18/27) 58.3% (14/24) 0.57

(12/27) 37.5% (9/24) 0.78

(6/27) 33.3% (8/24) 0.53

(17/20) 72.2% (13/18) 0.44

(12/27) 25.0% (6/24) 0.24

(6/27) 16.7% (4/24) 0.73

(9/27) 58.3% (14/24) 0.10

0.43 (27) 2.82 � 0.55 (24) 0.20

7.47 (27) 26.68 � 9.09 (24) 0.23

6.69 (27) 36.97 � 5.64 (24) 0.66

0.35 (27) 3.19 � 0.48 (24) 0.23

PES (n

3.93 �

70.4%

11.1%

40.7%

66.7%

44.4%

22.2%

85.0%

44.4%

22.2%

33.3%

2.64 �

3.87 �

6.19 �

3.05 �

(0OnfCLa(

aa(p(OoPd

e and m


A P R I L 2 0 0 8 : 1 6 1 – 7

Aoki et al.


164

PES: 4.07 � 2.59 mm vs. BMS: 4.67 � 3.55 mm, p �.50) were also comparable between the 2 groups (Table 2).verlapping BMS. In lesions treated with BMS, there wereo statistically significant changes in serial (baseline vs.ollow-up) measures of the EEM, stent, or peri-stent P&MSA in the proximal, overlap, or distal stent regions.umen CSA significantly decreased in both the single stentnd overlap regions during the 9-month follow-up periodTable 2). The IH CSA and serial changes in EEM, stent,

Table 2. Intravascular Ultrasound Measurements at Post-Procedure and Fo

PES

Post-Procedure Follow-Up

Proximal region (single stent layer)

EEM area, mm2 15.17 � 4.30 (26) 16.11 � 4.72 (

Stent area, mm2 7.55 � 1.72 (27) 7.49 � 2.01 (

Lumen area, mm2 7.55 � 1.72 (27) 6.53 � 2.18 (

IH area, mm2 — 0.97 � 1.06 (

Peri-stent P&M area, mm2 7.61 � 3.03 (26) 8.57 � 3.09 (

Length, mm 15.89 � 5.80 (27) 15.96 � 4.92 (

Overlap region (double stent layers)

EEM area, mm2 14.23 � 3.91 (26) 15.67 � 4.41 (

Stent area, mm2 7.07 � 1.73 (27) 7.02 � 2.11 (

Lumen area, mm2 7.07 � 1.73 (27) 6.28 � 2.31 (

IH area, mm2 — 0.74 � 0.91 (


Length, mm 4.07 � 2.59 (27) 4.04 � 2.79 (

Distal region (single stent layer)

EEM area, mm2 12.23 � 3.10 (26) 13.39 � 3.53 (

Stent area, mm2 6.73 � 1.58 (27) 6.81 � 1.94 (

Lumen area, mm2 6.73 � 1.58 (27) 5.93 � 2.06 (

IH area, mm2 — 0.88 � 0.85 (


Length, mm 19.22 � 6.05 (27) 19.52 � 5.89 (

BMS � bare-metal stent; EEM � external elastic membrane; IH � intimal hyperplasia; P&M � plaqu

Table 3. Comparison of Vessel Responses Among the

Proximal Region

PES

IH area, mm2 0.97 � 1.06 (27)

∆Peri-stent P&M area, mm2 0.96 � 1.36 (26)

% ∆Peri-stent P&M area 15.6 � 21.3 (26)

∆EEM area, mm2 0.94 � 1.86 (26)

% ∆EEM area 6.8 � 12.8 (26)

BMS

IH area, mm2 3.12 � 2.40 (24)

∆Peri-stent P&M area, mm2 �0.02 � 1.48 (21)

% ∆Peri-stent P&M area 3.0 � 19.0 (21)

∆EEM area, mm2 �0.05 � 1.78 (21)

% ∆EEM area 0.7 � 11.1 (21)

Numbers in parenthesis indicate the number of segments available fo

ANOVA � analysis of variance; other abbreviations as in Table 2.

nd peri-stent P&M area were not significantly differentmong the proximal, overlap, and distal BMS regionsTable 3). At follow-up, ISA was found in 4.2% of theroximal regions but in none of the overlap or distal regionsp � 0.36).verlapping PES. Unlike BMS-treated lesions, there was anverall increase in EEM and peri-stent P&M areas in all 3ES segments (proximal, overlap, and distal). Stent CSAid not change, but lumen CSA decreased during the

p

BMS

p Value Post-Procedure Follow-Up p Value

0.02 16.89 � 4.45 (21) 16.83 � 4.11 (21) 0.89

0.71 8.38 � 2.44 (24) 8.42 � 2.42 (24) 0.84

0.0006 8.45 � 2.51 (24) 5.43 � 2.38 (24) �0.0001

— — 3.12 � 2.40 (24) —

0.001 8.84 � 2.70 (21) 8.81 � 2.26 (21) 0.94

0.88 16.29 � 4.29 (24) 16.92 � 3.91 (24) 0.35

0.004 15.82 � 4.57 (21) 15.77 � 4.74 (21) 0.93

0.78 7.74 � 2.23 (24) 8.02 � 2.51 (24) 0.27

0.008 7.75 � 2.23 (24) 4.79 � 2.12 (24) �0.0001

— — 3.23 � 1.75 (24) —

0.0003 8.24 � 2.93 (21) 8.52 � 2.84 (21) 0.48

0.91 4.67 � 3.55 (24) 4.71 � 3.34 (24) 0.92

0.008 14.42 � 4.71 (24) 14.32 � 4.90 (24) 0.71

0.52 7.56 � 2.12 (24) 7.39 � 2.35 (24) 0.23

0.0004 7.57 � 2.13 (24) 4.72 � 2.19 (24) �0.0001

— — 2.69 � 1.49 (24) —

0.008 6.80 � 2.88 (24) 6.91 � 2.80 (24) 0.54

0.48 19.54 � 5.96 (24) 18.71 � 6.37 (24) 0.12

edia; PES � paclitaxel-eluting stent.

gions

verlap Region Distal Region p ANOVA

.74 � 0.91 (27) 0.88 � 0.85 (27) 0.38

.56 � 1.88 (26) 1.03 � 1.81 (26) 0.21

9.3 � 38.4 (26) 30.1 � 66.5 (26) 0.28

.45 � 2.33 (26) 1.16 � 2.04 (26) 0.38

1.6 � 17.7 (26) 11.0 � 20.1 (26) 0.31

.23 � 1.75 (24) 2.69 � 1.49 (24) 0.85

.29 � 1.82 (21) 0.11 � 0.89 (24) 0.48

7.9 � 22.8 (21) 3.3 � 13.9 (24) 0.40

.05 � 2.64 (21) �0.10 � 1.30 (24) 1.00

1.0 � 16.8 (21) �0.8 � 8.5 (24) 0.95

ntravascular ultrasound analysis.

llow-U

26)

27)

27)

27)

26)

27)

26)

27)

27)

27)

26)

27)

26)

27)

27)

27)

26)

27)

3 Re

O

0

1

2

1

1

3

0

�0

r serial i

9qprECpipd

smnPi0�d(

wdBbI(srvwmCcede

war�rlp�

D

S[dhw

(olswIatpIDIopnpIsidI1ic


A P R I L 2 0 0 8 : 1 6 1 – 7

Aoki et al.


165

-month follow-up period (Table 2). More than three-uarters of PES-treated lesions showed an increase ineri-stent P&M area (proximal region: 73.1%; overlapegion: 84.6%; and distal region: 80.8%). The increase inEM CSA was similar to the increase in peri-stent P&MSA, indicating that vessel remodeling outside the stentaralleled the increase in peri-stent plaque (Table 3). Sim-lar to BMS-treated lesions, IH area and serial changes ineri-stent P&M and EEM CSA were not significantlyifferent among the 3 regions (Table 3).When the vascular responses to the slow-release PES

tents used in TAXUS-V were compared with theoderate-release PES stents in TAXUS-VI, there was a

on-significant trend toward an increase in peri-stent&M CSA in the moderate release PES-treated lesions

n all 3 regions: proximal region, 1.35 � 1.29 mm2 versus.82 � 1.39 mm2 (p � 0.39); central overlap region, 2.08

1.90 mm2 versus 1.37 � 1.89 mm2 (p � 0.40); andistal region 1.42 � 1.61 mm2 versus 0.88 � 1.90 mm2

p � 0.51).At 9 months follow-up among PES-treated lesions, ISA

as found in 11.1% of the proximal regions and in 7.4% of theistal regions but not in any of the overlap regions (p � 0.22).MS versus PES stents. Comparing the vascular responsesetween PES and BMS, PES-treated lesions had reducedH CSA in the proximal, overlap, and distal stent regionsTable 4). Although the difference did not reach statisticalignificance, IH area tended to be less in the PES overlapegion than in single strut PES regions (0.74 � 0.91 mm2

s. 0.93 � 0.92 mm2, p � 0.27), whereas the opposite trendas noted with BMS (3.23 � 1.75 mm2 vs. 2.90 � 1.72m2, p � 0.34). Furthermore, a greater increase in EEMSA and peri-stent P&M CSA was present with PES

ompared with BMS, consistent with a greater degree ofxpansive remodeling (Table 4). There were no significantifferences in the frequency of ISA with PES and BMS inither the single stent or multiple overlapping stent regions.

For lesions treated with BMS, significant correlationsere found between the degree of neointimal growth (IH)

nd the degree of outward remodeling (delta EEM) in all 3egions (R � 0.67, p � 0.001 in the proximal; R � 0.51, p

0.02 in the overlap; and R � 0.50, p � 0.01 in the distalegion). However, these correlations were not observed inesions treated with PES (R � �0.12, p � 0.54 in theroximal; R � �0.15, p � 0.46 in the overlap; and R �0.03, p � 0.89 in the distal region).

iscussion

everal previous studies (angiographic [18,19], IVUS20,21], angioscopic [22], and pathologic [7]) have ad-ressed the issue of DES overlap. To our knowledge, thereave been 2 published manuscripts in which IVUS analysis
as performed after overlapping DES implantations d
20,21). Kang et al. (20) presented a comparison of IVUSutcomes after different-DES overlap and same-DES over-ap. The IH and EEM CSA at the overlapping site were notignificantly different between the 2 groups. However, thereas no comparison between DES overlap and BMS overlap.

n addition, this analysis was based only on follow-up IVUSnalysis. Therefore, analyses of vessel remodeling and de-ailed ISA (late acquired, persistent resolved) were notossible. Kawaguchi et al. (21) presented angiographic andVUS outcomes after BMS overlap and 3 different types ofES overlap implantation in patients with diabetes. Percent

H was less in the DES arms than the BMS arm in bothverlap and non-overlap regions. The EEM area at post-rocedure and follow-up was analyzed. However, there waso serial analysis. In addition, ISA was not observed atost-procedure and follow-up in all patients. Detailed serialSA analysis was not possible. The present analysis repre-ents the first manuscript to investigate vessel responsenside and outside the overlapping stent region, includingetailed ISA analysis. The major findings of this serialVUS analysis of overlapping BMS and PES are as follows:) compared with BMS-treated lesions, neointimal growthn PES-treated lesions was significantly reduced in theentral overlap region as well as the adjacent proximal and

Table 4. Comparison of Vessel Responses Between PES and BMS

PES BMS p Value

∆Peri-stent P&M area, mm2

Proximal region 0.96 � 1.36 (26) �0.02 � 1.48 (21) 0.02

Overlap region 1.56 � 1.88 (26) 0.29 � 1.82 (21) 0.02

Distal region 1.03 � 1.81 (26) 0.11 � 0.89 (24) 0.03

∆EEM area, mm2

Proximal region 0.94 � 1.86 (26) �0.05 � 1.78 (21) 0.07

Overlap region 1.45 � 2.33 (26) �0.05 � 2.64 (21) 0.04

Distal region 1.16 � 2.04 (26) �0.10 � 1.30 (24) 0.01

IH area, mm2

Proximal region 0.97 � 1.06 (27) 3.12 � 2.40 (24) 0.0001

Overlap region 0.74 � 0.91 (27) 3.23 � 1.75 (24) �0.0001

Distal region 0.88 � 0.85 (27) 2.69 � 1.49 (24) �0.0001

Late acquired ISA, n (%)

Proximal region 3.7% (1/27) 4.2% (1/24) 1.00

Overlap region 0.0% (0/27) 0.0% (0/24) —

Distal region 7.4% (2/27) 0.0% (0/24) 0.49

Persistent ISA, n (%)

Proximal region 7.4% (2/27) 0.0% (0/24) 0.49

Overlap region 0.0% (0/27) 0.0% (0/24) —

Distal region 0.0% (0/27) 0.0% (0/24) —

Resolved ISA, % n (%)

Proximal region 0.0% (0/27) 8.3% (2/24) 0.22

Overlap region 0.0% (0/27) 0.0% (0/24) —

Distal region 0.0% (0/27) 4.2% (1/24) 0.47

ISA � incomplete stent apposition; other abbreviations as in Table 2.

istal non-overlapped regions; 2) in all 3 regions the

ilsssrwgPVpbmP(siPPsr

lpsitbdwfrideoedimmetspnvdabrp

siIcrrsPAmeaetoSsravswefbansvwePd2nw

C

Nndeccro

RCF


A P R I L 2 0 0 8 : 1 6 1 – 7

Aoki et al.


166

ncrease in EEM and peri-stent P&M areas in PES-treatedesions was significantly greater than in BMS-treated le-ions, and more than three quarters of PES-treated lesionshowed an increase in peri-stent P&M area in both singletrut and overlap regions; 3) among the 3 regions theelative changes in IH, EEM, and peri-stent P&M areasere not significantly different in the both BMS and PESroups; and 4) late acquired ISA was not observed in theES overlap region.ascular remodeling after BMS and PES. An increase ineri-stent P&M area after single PES implantation haseen observed in several studies with both slow- andoderate-release PES in TAXUS-II and with slow-releaseES in TAXUS-IV (13,14,23). Similarly, Petronio et al.

16) reported a slight increase in peri-stent P&M area afterlow-release PES implantation in the left anterior descend-ng coronary artery. The present study thus confirms thatES result in greater increases in EEM and peri-stent&M areas than with BMS, at both the site of overlappingtent struts and adjacent proximal and distal single stentegions.

Whereas swine studies have shown comparable endothe-ial cell coverage in the PES overlap zone (24), increasedaclitaxel or sirolimus elution from overlapping stents re-ults in delayed healing and arterial toxicity in the rabbitnjured iliac model (7). Concern has thus been raised abouthe potential for drug toxicity at the site of overlapping PES,ecause of greater dose at this site (25). In TAXUS-II theegree of peri-stent P&M area increase was exaggeratedith moderate-release PES compared with the slow-release

ormulation, potentially explained by the greater paclitaxelelease during either the burst phase within hours after stentmplantation or the 3- to 8-fold total increased paclitaxelose eluted (in vitro). In the present study, however,xaggerated vascular responses were not seen at the PESverlap site, although a trend was present toward greaterxpansive remodeling with the moderate-release PES. Highoses of paclitaxel might also be expected to result in anncreased incidence of late acquired ISA. The anti-

etabolic effect of higher-dose paclitaxel theoreticallyight induce focal necrosis or apoptosis and generate a new

mpty space between the struts and the vessel wall (26). Inhe present study, however, ISA also did not occur at theite of overlapping PES. Thus it would seem that overlap-ing the commercially available slow-rate release PES doesot result in excessive expansive remodeling or adverseascular responses, either because doubling the paclitaxelose remains below the toxic threshold or because therterial drug concentration in the overlap region might note twice as great as a single layer region (27). An additionaleason why vascular toxicity might not have been seen in the
resent report in contrast to the aforementioned animal
tudy was that the length of overlap was significantly shortern the current study (mean 4 mm vs. 9.8 mm) (7).H. In the current study, neointimal growth was signifi-antly inhibited for PES compared with BMS in the overlapegion and in the adjacent proximal and distal single strutegions. Although the difference did not reach statisticalignificance, the IH area was lowest in the overlap region ofES-treated lesions compared with non-overlap regions.lthough overlapping BMS struts might induce moreedial injury and a subsequent increase in IH (28,29), this

ffect seems to be counteracted by the increased drug densityt the overlap site of PES. Considering that immaturendothelial coverage of stent strut is a predictor of late stenthrombosis (30), a longer clinical follow-up of patients withverlapping stents should be reassuring in future studies.tudy limitations. Several limitations of the present studyhould be noted. Factors unrelated to the stent might play aole in vessel remodeling (31). Drugs such as statins andntihypertensive agents influence the change in plaqueolume (32–34). Although the prevalence of statin use wasimilar between PES and BMS groups, detailed lipid levelsere not measured and this influence cannot be totally

xcluded. Not all patients in the substudy had baseline andollow-up IVUS. The entire length of the EEM could note measured in each patient, because of stent shadowing orrtifact. Finally, this study comprised a relatively modestumber of patients. It might be possible that a larger sampleize would have been able to show significantly differentessel remodeling among the 3 regions in the lesions treatedith overlapping PES. In addition, the number of patients

nrolled receiving the slow-release and moderate-releaseES was insufficient to allow definitive conclusions to berawn regarding differential vascular responses between thePES types. Further IVUS investigation involving large

umbers of patients after DES overlap implantation isarranted.

onclusions

ine months after implantation of overlapping stents,eointimal tissue growth inside PES was significantly re-uced compared with BMS at the expense of greaterxpansive remodeling. In the PES group, exaggerated vas-ular responses were not present in the overlap regionompared with the adjacent proximal or distal single stentegions, and ISA was not observed at the site of PESverlap.

eprint requests and correspondence: Dr. Gary S. Mintz,ardiovascular Research Foundation, 111 East 59th Street, 11thloor, New York, New York 10022. E-mail: [email protected].

R

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3


A P R I L 2 0 0 8 : 1 6 1 – 7

Aoki et al.


167

EFERENCES

1. Stone GW, Ellis SG, Cox DA, et al. A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med2004;350:221–31.

2. Stone GW, Ellis SG, Cannon L, et al. Comparison of a polymer-basedpaclitaxel-eluting stent with a bare metal stent in patients with complexcoronary artery disease: a randomized controlled trial. JAMA 2005;294:1215–23.

3. Stone GW, Ellis SG, O’Shaughnessy CD, et al. Paclitaxel-elutingstents vs vascular brachytherapy for in-stent restenosis within bare-metal stents: the TAXUS V ISR randomized trial. JAMA 2006;295:1253–63.

4. Abizaid A, Chan C, Lim YT, et al. Twelve-month outcomes with apaclitaxel-eluting stent transitioning from controlled trials to clinicalpractice (the WISDOM Registry). Am J Cardiol 2006;98:1028–32.

5. Colombo A, Drzewiecki J, Banning A, et al. Randomized study toassess the effectiveness of slow- and moderate-release polymer-basedpaclitaxel-eluting stents for coronary artery lesions. Circulation 2003;108:788–94.

6. Dawkins KD, Grube E, Guagliumi G, et al. Clinical efficacy ofpolymer-based paclitaxel-eluting stents in the treatment of complex,long coronary artery lesions from a multicenter, randomized trial:support for the use of drug-eluting stents in contemporary clinicalpractice. Circulation 2005;112:3306–13.

7. Finn AV, Kolodgie FD, Harnek J, et al. Differential response ofdelayed healing and persistent inflammation at sites of overlappingsirolimus- or paclitaxel-eluting stents. Circulation 2005;112:270–8.

8. Wilson GJ, Polovick JE, Huibregtse BA, Poff BC. Overlappingpaclitaxel-eluting stents: long-term effects in a porcine coronary arterymodel. Cardiovasc Res 2007;76:361–72.

9. Chu WW, Kuchulakanti PK, Torguson R, et al. Comparison of clinicaloutcomes of overlapping sirolimus- versus paclitaxel-eluting stents inpatients undergoing percutaneous coronary intervention. Am J Cardiol2006;98:1563–6.

0. Aoki J, Ong AT, Rodriguez Granillo GA, et al. “Full metal jacket”(stented length � or �64 mm) using drug-eluting stents for de novocoronary artery lesions. Am Heart J 2005;150:994–9.

1. Weissman NJ, Ellis SG, Grube E, et al. Effect of the polymer-based,paclitaxel-eluting TAXUS Express stent on vascular tissue responses: avolumetric intravascular ultrasound integrated analysis from theTAXUS IV, V, and VI trials. Eur Heart J 2007;28:1574–82.

2. Mintz GS, Weissman NJ. Intravascular ultrasound in the drug-elutingstent era. J Am Coll Cardiol 2006;48:421–9.

3. Aoki J, Colombo A, Dudek D, et al. Peristent remodeling andneointimal suppression 2 years after polymer-based, paclitaxel-elutingstent implantation: insights from serial intravascular ultrasound analysisin the TAXUS II study. Circulation 2005;112:3876–83.

4. Tanabe K, Serruys PW, Degertekin M, et al. Chronic arterial responsesto polymer-controlled paclitaxel-eluting stents: comparison with baremetal stents by serial intravascular ultrasound analyses: data from therandomized TAXUS-II trial. Circulation 2004;109:196–200.

5. Mintz GS, Nissen SE, Anderson WD, et al. American College ofCardiology Clinical Expert Consensus Document on Standards forAcquisition, Measurement and Reporting of Intravascular UltrasoundStudies (IVUS). A report of the American College of Cardiology TaskForce on Clinical Expert Consensus Documents. J Am Coll Cardiol2001;37:1478–92.

6. Petronio AS, De Carlo M, Branchitta G, et al. Randomized compar-ison of sirolimus and paclitaxel drug-eluting stents for long lesions inthe left anterior descending artery: an intravascular ultrasound study.J Am Coll Cardiol 2007;49:539–46.

7. Mintz GS, Tinana A, Hong MK, et al. Impact of preinterventionalarterial remodeling on neointimal hyperplasia after implantation of
(non-polymer-encapsulated) paclitaxel-coated stents: a serial volumet-
ric intravascular ultrasound analysis from the ASian Paclitaxel-ElutingStent Clinical Trial (ASPECT). Circulation 2003;108:1295–8.

8. Minutello RM, Bhagan S, Feldman D, Sharma A, Hong MK, WongSC. Angiographic pattern of restenosis following implantation ofoverlapping sirolimus-eluting (Cypher) stents. Am J Cardiol 2006;97:499–501.

9. Burzotta F, Siviglia M, Altamura L, et al. Outcome of overlappingheterogenous drug-eluting stents and of overlapping drug-eluting andbare metal stents. Am J Cardiol 2007;99:364–8.

0. Kang WC, Oh KJ, Han SH, et al. Angiographic and intravascularultrasound study of the effects of overlapping sirolimus- and paclitaxel-eluting stents: comparison with same drug-eluting overlapping stents.Int J Cardiol 2007;123:12–7.

1. Kawaguchi R, Sabate M, Angiolillo DJ, et al. Angiographic and 3Dintravascular ultrasound assessment of overlapping bare metal stent andthree different formulations of drug-eluting stents in patients withdiabetes mellitus. Int J Cardiovasc Imaging 2008;24:125–32.

2. Ichikawa M, Yutani C, Hayashi T, et al. Angioscopic findings ofdelayed healing at sites of sirolimus-eluting stent overlap after 21-month implantation. Int J Cardiol 2007 Aug 15 [E-pub ahead ofprint].

3. Weissman NJ, Koglin J, Cox DA, et al. Polymer-based paclitaxel-eluting stents reduce in-stent neointimal tissue proliferation: a serialvolumetric intravascular ultrasound analysis from the TAXUS-IV trial.J Am Coll Cardiol 2005;45:1201–5.

4. Schwartz RS, Wilson GJ. Cypher versus Taxus stents: comparinginflammatory response in porcine coronary arteries. Am J Cardiol2006;98 Suppl 8A:36M.

5. Pires NM, Eefting D, de Vries MR, Quax PH, Jukema JW. Sirolimusand paclitaxel provoke different vascular pathological responses afterlocal delivery in a murine model for restenosis on underlying athero-sclerotic arteries. Heart 2007;93:922–7.

6. Serruys PW, Degertekin M, Tanabe K, et al. Intravascular ultrasoundfindings in the multicenter, randomized, double-blind RAVEL (RAn-domized study with the sirolimus-eluting VElocity balloon-expandablestent in the treatment of patients with de novo native coronary arteryLesions) trial. Circulation 2002;106:798–803.

7. Balakrishnan B, Tzafriri AR, Seifert P, Groothuis A, Rogers C,Edelman ER. Strut position, blood flow, and drug deposition: impli-cations for single and overlapping drug-eluting stents. Circulation2005;111:2958–65.

8. Farb A, Sangiorgi G, Carter AJ, et al. Pathology of acute and chroniccoronary stenting in humans. Circulation 1999;99:44–52.

9. Farb A, Weber DK, Kolodgie FD, Burke AP, Virmani R. Morpho-logical predictors of restenosis after coronary stenting in humans.Circulation 2002;105:2974–80.

0. Finn AV, Joner M, Nakazawa G, et al. Pathological correlates of latedrug-eluting stent thrombosis: strut coverage as a marker of endothe-lialization. Circulation 2007;115:2435–41.

1. Nicholls SJ, Tuzcu EM, Sipahi I, Schoenhagen P, Nissen SE.Intravascular ultrasound in cardiovascular medicine. Circulation 2006;114:e55–9.

2. Schoenhagen P, Tuzcu EM, Apperson-Hansen C, et al. Determinantsof arterial wall remodeling during lipid-lowering therapy: serial intra-vascular ultrasound observations from the Reversal of Atherosclerosiswith Aggressive Lipid Lowering Therapy (REVERSAL) trial. Circu-lation 2006;113:2826–34.

3. Nissen SE, Tuzcu EM, Libby P, et al. Effect of antihypertensive agentson cardiovascular events in patients with coronary disease and normalblood pressure: the CAMELOT study: a randomized controlled trial.JAMA 2004;292:2217–25.

4. Okazaki S, Yokoyama T, Miyauchi K, et al. Early statin treatment inpatients with acute coronary syndrome: demonstration of the beneficialeffect on atherosclerotic lesions by serial volumetric intravascularultrasound analysis during half a year after coronary event: the ES-
TABLISH Study. Circulation 2004;110:1061–8.