Clinical outcome after primary percutaneous coronary intervention with drug-eluting and bare metal...

Clinical Outcome After Primary Percutaneous CoronaryIntervention With Drug-Eluting and Bare Metal Stents in

Patients With ST-Segment Elevation Myocardial InfarctionLisette Okkels Jensen, MD, PhD; Michael Maeng, MD, PhD; Per Thayssen, MD, DMSci;

Anne Kaltoft, MD, PhD; Hans Henrik Tilsted, MD; Morten Bøttcher, MD, PhD;Jens Flensted Lassen, MD, PhD; Knud Nørregaard Hansen, MD; Lars Romer Krusell, MD;

Klaus Rasmussen, MD, DMSci; Knud Erik Pedersen, MD, DMSci; Lars Pedersen, MSc;Søren Paaske Johnsen, MD, PhD; Henrik Toft Sørensen, MD, PhD, DMSci; Leif Thuesen, MD; DMSci

Background—The use of drug-eluting stents (DESs) versus bare metal stents (BMSs) in primary percutaneous coronaryintervention (PCI) for ST-segment elevation myocardial infarction is a matter of debate. Therefore, we examined the riskof target lesion revascularization (TLR), stent thrombosis, myocardial infarction, and death after the implantation ofDES or BMS in primary PCI patients in Western Denmark.

Methods and Results—A total of 3756 consecutive patients with ST-segment elevation myocardial infarction treated withprimary PCI and stent implantation, recorded in the Western Denmark Heart Registry from January 2002 through June2005, were followed up for 2 years. We used Cox regression analysis to control for confounding. The 2-year incidenceof definite stent thrombosis was 1.9% in the DES group and 1.1% in the BMS group (adjusted relative risk [RR]�1.53;95% CI�0.84 to 2.78; P�0.17). Very late definite stent thrombosis (�12 months) was seen in 0.4% in the DES groupand 0.06% in the BMS group (adjusted RR�6.74; 95% CI�1.23 to 37.00; P�0.03). The 2-year incidence of myocardialinfarction was similar in the 2 groups, 5.2% in the DES group versus 6.3% in the BMS group (P�0.28; adjustedRR�1.13; 95% CI�0.81 to 1.59; P�0.47). All-cause 2-year mortality was 7.8% in the DES group and 11.4% in BMSgroup (P�0.004; adjusted RR�0.79; 95% CI�0.60 to 1.04; P�0.09). The 2-year incidence of target lesionrevascularization was 7.2% in the DES group and 8.7% in the BMS group (P�0.09; adjusted RR�0.70; 95% CI�0.52to 0.92; P�0.012).

Conclusions—In ST-segment elevation myocardial infarction patients treated with primary PCI, target lesion revascular-ization was reduced by 30% in patients treated with a DES. The risk of very late definite stent thrombosis was low butincreased in patients treated with DES. DES was not associated with increased risk of myocardial infarction or death,when compared with BMS. (Circ Cardiovasc Intervent. 2008;1:176-184.)

Key Words: mortality � myocardial infarction � primary PCI � restenosis � thrombosis

Primary percutaneous coronary intervention (PCI) is awell-established treatment for ST-segment elevation

myocardial infarction (STEMI).1,2 Drug-eluting stents(DESs) were developed to address the problem of in-stentrestenosis in patients who underwent bare metal stent (BMS)implantation. Randomized trials have shown that sirolimus-eluting stents are superior to BMS in STEMI patients inreducing the incidence of restenosis and target lesion revas-cularization (TLR).3–5 The beneficial effects of paclitaxel-eluting stents in STEMI patients are less well documented,

with only 1 randomized trial showing a trend toward betteroutcome with paclitaxel-eluting stents.6

The long-term beneficial effects of DES have been ques-tioned by reports of increased risks of stent thrombosis,7

myocardial infarction (MI), and death.8–10 However, thesepossible safety problems were not confirmed in other stud-ies.10,11 For STEMI patients treated with primary PCI, data onlong-term outcomes after DES implantation are limited. Aclinical database study of a relatively small STEMI patientcohort detected an initial beneficial effect that disappeared

Received May 27, 2008; accepted October 16, 2008.From the Department of Cardiology (L.O.J., P.T., K.N.H., K.E.P.), Odense University Hospital, Odense, Denmark; the Department of Cardiology

(M.M., A.K., M.B., J.F.L., L.R.K., L.T.), Skejby Sygehus, Aarhus University Hospital, Aarhus, Denmark; the Department of Cardiology (H.H.T., K.R.)and the Center for Cardiovascular Research (K.R., S.P.J.), Aalborg Hospital, Aarhus University Hospital, Aalborg, Denmark; the Department of ClinicalEpidemiology (L.P., S.P.J., H.T.S.), Aarhus University Hospital, Aarhus, Denmark; and the Department of Epidemiology (H.T.S.), Boston University,Boston, Mass.

Correspondence to Lisette Okkels Jensen, MD, PhD, Department of Cardiology, Odense University Hospital, Sdr. Boulevard 29, 5000 Odense C,Denmark. E-mail [email protected]

© 2008 American Heart Association, Inc.

Circ Cardiovasc Intervent is available at http://circinterventions.ahajournals.org DOI: 10.1161/CIRCINTERVENTIONS.108.794578

176 by guest on March 13, 2016http://circinterventions.ahajournals.org/Downloaded from

http://circinterventions.ahajournals.org/

over time. After 3 years of follow-up, the initial significantlylower target vessel revascularization rate in this group disap-peared, and an increased rate of stent thrombosis was ob-served in the DES group (BMS, 1.6%; sirolimus-eluting stent,2.7%; and paclitaxel-eluting stent, 2.9%).12 The current studyassessed the use of DES versus BMS in a large cohort ofprimary PCI patients who received dual antiplatelet therapyfor 12 months in accordance with recent recommendations.We used data from the Western Denmark Heart Registry(WDHR) to examine the rates of stent thrombosis, MI, death,and TLR after the implantation of DES or BMS in STEMIpatients treated with primary PCI.

MethodsSetting and DesignThe study was conducted using Western Denmark’s health caredatabases, which cover the entire population of the region of �3.0million inhabitants (55% of the Danish population). All patients werefollowed-up for 24 months. A detailed description of the databaseshas been reported previously.10

Patients and ProceduresPrimary PCI for STEMI has been the recommended treatment inDenmark after the publication of the DANish trial in Acute Myo-cardial Infarction-2 (DANAMI-2)1 study in 2003. To be eligible forprimary PCI, patients must meet the following criteria: (1) symptomspresent �12 hours from onset of pain to time of catheterization, and(2) ST-segment elevation (at least 1 mm in 2 or more standard leadsor at least 2 mm in 2 or more contiguous precordial leads) or a newleft bundle-branch block. We used the WDHR to identify all primaryPCIs performed from January 1, 2002 through June 30, 2005. Forinclusion in the current study, patients had to receive either DES orBMS. Those treated with balloon angioplasty without stent implan-tation or with a combination of BMS and DES were excluded (n�91;2.4%). During the first year of the study, only BMS were available.From 2003, DES or BMS was selected according to the preference ofthe PCI operator. PCI was performed according to the standardpractices of the participating centers. A glycoprotein IIb/IIIa receptorblocker was administered at the operator’s discretion. The postint-ervention antiplatelet regimen included lifelong acetylsalicylic acid(75 to 150 mg once daily) and clopidogrel with a loading dose of 300mg followed by maintenance with 75 mg daily. The recommendedduration of clopidogrel treatment was 3 to 12 months until November2002 and 12 months thereafter. In patients with definite stentthrombosis, medical records were reviewed to document the use ofantiplatelet therapy.

End PointsThe end points were the time to stent thrombosis (definite, probable,and possible), MI, all-cause mortality, cardiac death, and TLR. Allend points were assessed within 24 months of the PCI date andascertained from the WDHR, the Danish National Patient Registry(NPR) covering all Danish hospitals and the Danish Registry ofCauses of Death. All patients were followed-up for 24 months.

Stent ThrombosisWe defined stent thrombosis using the Academic ResearchConsortium’s definition, with a modification for probable stentthrombosis.13

Definite Stent ThrombosisAngiographic confirmation of stent thrombosis and at least 1 of thefollowing signs present within 48 hours: new onset of ischemicsymptoms at rest, new electrocardiographic changes suggestive ofacute ischemia, or typical increase and decrease of cardiacbiomarkers.

Probable Stent ThrombosisAny unexplained death within the first 30 days after intracoronarystenting.

Possible Stent ThrombosisAny unexplained death occurring from 30 days after intracoronarystenting until the end of the follow-up period. Stent thrombosis wasfurther characterized as acute (0 to �24 hours), subacute (�1 day to�30 days), late (�30 days to �1 year), and very late (�1 year to 24months).

We defined new MIs as hospitalization for MI occurring �28 daysafter the index PCI.14 We ascertained admissions and readmissionsfor MI (ICD-10 codes I21-I21.9) from the Danish National PatientRegistry10 and deaths from the Civil Registration System. Wevalidated the recorded cause of death using original death certificatesobtained from the National Registry of Causes of Death andclassified deaths according to their underlying cause.

From the WDHR, we ascertained TLR, defined as a repeat PCI ofthe index lesion or coronary artery bypass grafting, occurring within2 years after the index stent implantation. We assessed all clinicalend points occurring within 2 years of the index PCI. An expertcommittee reviewed relevant records and adjudicated the end pointsregarding stent thrombosis and cause of death. For all cases of stentthrombosis, we reviewed relevant medical records and catheteriza-tion films.

CovariatesFrom the WDHR, we retrieved data on potential predictors forsubsequent cardiovascular events, including the characteristics of thepatients, procedures, and lesions. For each patient, we obtained dataon all discharge diagnoses from the Danish National Patient Regis-try. We then computed the comorbidity scores using the Charlsoncomorbidity index,15 which covers 19 major disease categories,including diabetes mellitus, heart failure, cerebrovascular diseases,and cancer. The index value is a weighted summary of the diagnoses,with each weight calculated based on 1-year mortality associatedwith each disease in the original Charlson dataset.15

Data on all key patient and procedure characteristics were �95%complete, and ascertainment of end points (stent thrombosis, death,MI, and TLR) was 100% complete.

Statistical MethodsDistributions of continuous variables in the 2 groups were comparedusing either the 2-sample t test or the Mann–Whitney test, dependingon whether the data followed the normal distribution. Continuousvariables are presented either as median and interquartile range (25thto 75th) or mean�1 SD depending on whether the data followed thenormal distribution or not. Distributions of categorical variableswere compared using the �2 test.

We counted end point events that occurred during the follow-upperiod and compared their rates for the cohort of patients with DESand the cohort with BMS. Follow-up began on the date of the indexPCI procedure. In analyses with stent thrombosis as the outcome,follow-up continued until date of definite stent thrombosis, death,emigration, or until 24 months after implantation, whichever camefirst. In analyses with death as the outcome, follow-up continueduntil date of death, emigration or until 24 months after implantation.In analyses with MI as the outcome, follow-up continued until dateof MI, death, emigration, or until 24 months after implantation.

We constructed Kaplan–Meier curves for patients and lesionstreated with DES or BMS and used the life-table method to computethe 2-year incidence for each end point (proportion of the populationat risk with the outcome of interest). Cox proportional-hazardsregression was used to compute hazard ratios as estimates of therelative risk (RR) for each end point. Because the hazards were notproportional throughout the follow-up period, we computed theestimates of RR within separate time windows, where the propor-tionality assumption held. The RR in these analyses reflected the riskamong patients alive and at risk of the specific end point at the startof each time period (eg, after 30 days or 1 year of follow-up). Wecontrolled for age, gender, diabetes mellitus, and procedure time in

Jensen et al Primary PCI and Long-Term Outcome 177

by guest on March 13, 2016http://circinterventions.ahajournals.org/Downloaded from


all regression analyses. To improve precision of the risk estimates,we used the change-in-estimate method, which entailed retainingvariables that changed RR estimates for an outcome by �10%.16

Thus, in the lesion-specific analyses (stent thrombosis and TLR) wealso adjusted for glycoprotein IIb/IIIa receptor blockers, stent length,and reference vessel size. We also carried out Cox regression basedon propensity score estimated. The latter was estimated by logisticregression as predicted probability of use of BMSs as a function ofall measured potential confounding factors.17 All data analyses werecarried out using the SAS software version 9.13 (SAS Institute Inc).

ResultsDescriptive DataA total of 3756 STEMI patients (4469 lesions) treated withprimary PCI and stenting were followed up for 24 months. Ofthese, 2973 patients with 3486 lesions were treated withBMS, whereas 783 patients with 983 lesions received DES.The number of patients treated with DES was 783 (Cypherstent: n�465 [59.4%], Taxus stent: n�318 [40.6%]), andthose treated with BMS was 2973. The patients’ median agewas 63 years (interquartile range, 54 to 73 years), and 30.9%of patients were older than 75 years. Baseline patient andprocedure characteristics (Table 1) and lesion characteristics(Table 2) differed substantially between the DES and BMSgroups.

Stent Thrombosis

Definite Stent ThrombosisThe 2-year incidence of definite stent thrombosis did notdiffer significantly between patients treated with DES (n�19;1.9%) and BMS (n�38; 1.1%; adjusted RR�1.53; 95%CI�0.84 to 2.78; P�0.17). Definite stent thrombosis oc-curred within a week in 34 lesions (60%; DES, 0.9% [n�9];and BMS, 0.7% [n�25]) and within 30 days in 42 lesions

(74%; DES, 1.2% [n�12]; and BMS, 0.9% [n�30]). Mediantime to occurrence of definite stent thrombosis was 5 days(interquartile range, 1 to 37 days). Rates of acute, subacute,and late definite stent thrombosis were also similar in the 2groups (Figure 1A). The cumulative incidence of stentthrombosis over time showed an initial steep rise, followed byan almost linear slight increase up to 24 months. Very latedefinite stent thrombosis occurred in 4 lesions (0.4%) in theDES group and in 2 lesions (0.06%) in the BMS group(adjusted RR�6.74; 95% CI�1.23 to 37.00; P�0.03; Table3 and Figure 2A).

Predictors of Definite Stent ThrombosisTreatment with glycoprotein IIb/IIIa receptor blockers re-duced the risk of definite stent thrombosis (adjustedRR�0.47; 95% CI�0.27 to 0.83; P�0.008). Stent length(adjusted RR�1.03; 95% CI�1.00 to 1.06; P�0.07) andprocedure time (adjusted RR�1.01; 95% CI�1.00 to 1.02;P�0.12) did not influence the risk of definite stent thrombo-sis significantly. Neither infarct-related artery nor TIMI flowbefore intervention (adjusted RR�0.72; 95% CI�0.36 to1.56; P�0.37) and TIMI flow after intervention (adjustedRR�1.52; 95% CI�0.54 to 4.25; P�0.42) were related torisk of stent thrombosis.

Definite, Probable, or Possible Stent ThrombosisDefinite, probable, or possible stent thrombosis was found in28 patients treated with DES (2-year incidence, 3.6%) and in106 patients treated with BMS (2-year incidence, 3.6%;RR�0.99; 95% CI�0.65 to 1.50; P�0.96; Figure 1B). Aftercontrolling for covariates, the risk of stent thrombosis (defi-nite, probable, or possible) did not differ between the 2groups (adjusted RR�1.09; 95% CI�0.71 to 1.67; P�0.68;Table 4 and Figure 2B).

Definite Stent Thrombosis and Antiplatelet TherapyAmong the 57 patients who developed definite stent throm-bosis, 82.5% (n�47) were receiving dual antiplatelet therapy(aspirin and clopidogrel) at the time of the thrombotic event.In the 6 patients with very late stent thrombosis, 1 patient wason dual antiplatelet therapy (DES�1), 1 patient was treatedwith aspirin only (DES�1), and 4 patients (DES�2 andBMS�2) had discontinued both aspirin and clopidogrel after12 months.

MortalityAll-cause 2-year mortality was lower in DES than in BMSpatients (7.8% versus 11.4%; P�0.004; RR�0.67; 95%CI�0.51 to 0.88; P�0.0039; Figure 1C). This difference wasnot statistically significant after adjustment for covariates(adjusted RR, 0.79; 95% CI�0.60 to 1.04; P�0.09; Figure2C). All-cause mortality 12 to 24 months after initial stentingwas similar in the 2 groups (2.2% versus 2.4%; P�0.75;adjusted RR�1.04; 95% CI�0.59 to 1.83; P�0.90).

Cardiac mortality during the first 30 days was 4.6%, lowerin the DES group (3.2%) than in the BMS group (4.9%)(P�0.039; RR�0.65; 95% CI�0.43 to 0.98; P�0.04). Afteradjustment, risk of cardiac death during the first 30 days nolonger differed significantly between the 2 groups (adjustedRR�0.77; 95% CI�0.50 to 1.18; P�0.23). Cardiac mortality

Table 1. Characteristics of Patients With STEMI Treated WithPrimary PCI and DES or BMS in Western Denmark fromJanuary 2002 Through June 2005

DES BMS P

No. of patients 783 2973

Male, n (%) 573 (73.2) 2179 (73.3) 0.95

Age, years (interquartile range) 60.0 (50–69) 64 (55–73) �0.0001

Family history, n (%) 236 (30.1) 812 (27.3) 0.19

Smoking, n (%) 359 (45.9) 1268 (42.7) 0.10

Diabetes mellitus, n (%) 88 (11.2) 240 (8.1) �0.0001

Hypertension, n (%) 169 (21.6) 657 (22.1) 0.23

Previous coronary artery bypassgrafting, n (%)

18 (2.3) 34 (1.1) �0.001

Previous percutaneous coronaryintervention, n (%)

35 (4.5) 118 (4.0) 0.03

Previous MI, n (%) 67 (8.6) 341 (11.5) 0.007

Lipid lowering therapy, n (%) 116 (14.8) 317 (10.7) 0.0006

Glycoprotein IIb/IIIa receptorblocker, n (%)

514 (65.6) 1717 (57.8) �0.0001

Comorbidity index score, n (%) 0.40

0 547 (70.0) 2004 (67.4)

1–2 190 (24.3) 790 (26.6)

3� 46 (5.9) 179 (6.0)

178 Circ Cardiovasc Intervent December 2008



after 2 years was 6.6%, with lower cardiac mortality amongDES patients than among BMS patients (5.0% versus 7.0%;P�0.037; RR�0.70; 95% CI�0.50 to 0.99; P�0.04). Afteradjustment, risk of cardiac death did not differ significantlybetween the 2 groups (adjusted RR�0.83; 95% CI�0.59 to1.18; P�0.31).

Noncardiac 2-year mortality was 3.3%, which was similarin DES-treated (2.6%) and BMS-treated (3.5%) patients(RR�0.70; 95% CI�0.44 to 1.13; P�0.15). Controlling forcovariates did not change this result (adjusted RR�0.82; 95%CI�0.51 to 1.33; P�0.42).

During the second half of the follow-up period (12 to 24months), after discontinuation of the recommended dualantiplatelet therapy, neither all-cause mortality, nor cardiacmortality, nor noncardiac mortality differed between patientstreated with DES or with BMS (Table 4).

Myocardial InfarctionThe 2-year incidence of MI was similar in the 2 stent groups(6.0% in the DES group versus 4.9% in the BMS group;

P�0.28 [Figure 1D]; adjusted RR�1.13; 95% CI�0.81 to1.59; P�0.47 [Figure 2D]).

Target Lesion RevascularizationTLR within 2 years occurred slightly less frequently in DESpatients than in BMS patients (2-year incidence, 7.2% versus8.7%; P�0.09; Figure 1E). After controlling for age, gender,diabetes mellitus, stent length, reference vessel size, andglycoprotein IIb/IIIa, the risk reduction with DES was 30%(adjusted RR�0.70; 95% CI�0.52 to 0.92; P�0.012; Figure2E). Diabetes mellitus (RR�1.66; 95% CI�1.19 to 2.31;P�0.003) and stent length (RR [per 1 mm]�1.02; 95%CI�1.00 to 1.03; P�0.01) increased the risk of TLR within24 months. Reference vessel size was inversely related toTLR (RR [per 1 mm]�0.82; 95% CI�0.67 to 1.00; P�0.05).

Propensity Score AnalysisThe estimates of effect were unchanged when we used Coxregression with adjustment for propensity score (calculatedby logistic regression as predicted probability of use of BMSs

Table 2. Procedure Characteristics of Patients With STEMI Treated With PrimaryPCI and DES and BMS in Western Denmark, January 2002 Through June 2005

DES BMS P

No. of lesions 983 3,486

Infarct-related artery, n (%) �0.001

Left anterior descending artery, n (%) 607 (61.8) 1341 (38.5)

Left circumflex artery, n (%) 111 (11.3) 505 (14.5)

Right coronary artery, n (%) 235 (23.9) 1607 (46.1)

Left main, n (%) 30 (3.1) 32 (0.9)

Preintervention TIMI flow, n (%) �0.0001

Grade 0 475 (50.3) 1901 (55.7)

Grade 1 52 (5.5) 210 (6.2)

Grade 2 126 (13.4) 459 (13.4)

Grade 3 291 (30.8) 845 (24.7)

Lesion length, mm 15.0 (10.0–20.0) 13.0 (10.0–18.0) �0.0001

Reference segment, mm 3.3 (3.0–3.6) 3.5 (3.0–3.8) �0.0001

Minimum lumen diameter, mm 0 (0–0.30) 0 (0–0.20) 0.002

Stenosis, % of luminal diameter 100 (90–100) 100 (95–100) 0.001

Stent length, mm 18.0 (13.0–24.0) 18.0 (13.0–21.0) �0.0001

Stent, n (%) 0.01

1 790 (80.4) 2,892 (83.3)

2 155 (15.8) 482 (13.9)

3� 38 (3.9) 97 (2.8)

Max balloon pressure, atm 15.0 (14.0–18.0) 14.0 (12.0–17.0) �0.0001

Max balloon diameter, mm 3.3 (3.0–3.7) 3.6 (3.2–3.9) �0.0001

Final TIMI flow, n (%) 0.0795

Grade 0 12 (1.3) 32 (0.9)

Grade 1 8 (0.9) 32 (0.9)

Grade 2 44 (4.7) 258 (7.6)

Grade 3 881 (93.3) 3090 (90.5)

Procedure time, min 25.0 (15.0–38.0) 24.0 (15.0–35.0) 0.002

Flouro time, min 8.0 (5.0–13.0) 7.0 (4.0–11.3) �0.0001

Contrast, mL 136.0 (100.0–200.0) 120.0 (82.0–200.0) �0.0001




Figure 1. A, Risk of definite stent thrombosis among patients treated with DES or BMS. B, Risk of definite, probable, and possiblestent thrombosis among patients treated with DES or BMS. C, All-cause mortality among patients treated with DES or BMS. D, Risk ofMI among patients treated with DES or BMS. E, TLR among patients treated with DES or BMS (unadjusted data).




given measured potential confounding factors). Patients withDES and BMS did not significantly differ with respect totheir risk of definite stent thrombosis (adjusted RR�1.88;95% CI�0.95 to 3.72). Very late definite stent thrombosiswas seen more often in DES treated patients (adjustedRR�6.51; 95% CI�1.19 to 35.70). All-cause 2-year mortal-ity (adjusted RR�0.67; 95% CI�0.50 to 0.91) and cardiacmortality (adjusted RR�0.67; 95% CI�0.46 to 0.98) werelower in DES treated patients. The 2-year incidence of MIwas similar in the 2 stent groups (adjusted RR�1.31; 95%CI�0.88 to 1.95).

DiscussionThe current study shows that primary PCI patients treatedwith either DES or BMS in a real-world setting have similar2-year risks of stent thrombosis, MI, or death. Within thesame period, the use of DES was associated with 30%reduction in the risk of clinically driven TLR. To ourknowledge, our study represents the largest published data-base of STEMI patients treated with primary PCI. The 3interventional centers in our study are all high-volume PCIcenters, and one third of the PCI procedures are primary PCIincluding optimized transportation logistics.1

An important clinical dilemma exists as to whether thebenefit of reduced repeat revascularization with DES out-weighs the harm of a possible increased risk of the rare butserious complication of late stent thrombosis. Another con-cern is whether the net TLR reduction after DES is main-tained during longer-term follow-up, especially after cessa-tion of dual antiplatelet therapy. Our data show that 25% ofdefinite stent thromboses occurred acutely within 24 hoursand that another 50% occurred within 30 days. During thesefirst 30 days, the risk of definite stent thrombosis was similarin patients treated with DES or BMS. Previous studies onBMS implanted during primary PCI for MI have found highthrombosis rates. However, this is limited by the absence ofa uniform definition for stent thrombosis in these studies.Current studies comparing DES and BMS in this setting havealso found high rates of stent thrombosis in both groups. Stentthrombosis is therefore more frequent in primary PCI foracute MI compared with planned procedures and the rates ofstent thrombosis are similar for DES and BMS. However,longer follow-up is necessary to follow the risk of very latestent thrombosis to ensure that this risk will not outweigh thebenefit with reduction in TLR.

Several factors are associated with an increased risk ofstent thrombosis in STEMI patients treated with primary PCI,including the risk of incomplete stent apposition18 or under-expansion of the stent itself.19 Further, the increased plateletactivation in acute MI,20 coupled with delayed healing, riskof lack of endothelialization, and exposure to proinflam-matory and thrombogenic environment, might explain theincreased risk of stent thrombosis in STEMI patients.21

However, neither in our study nor in the 3 randomized studies(TYPHOON,4 PASSION,6 and SESAMI3) did these possiblepathophysiological processes result in an increased risk ofearly stent thrombosis after DES implantation. In the currentstudy, definite stent thrombosis during the first year after thecessation of dual antiplatelet therapy was more common afterDES than after BMS treatment. This difference, although itwas extremely small, is still important. At the same time,DES was associated with a marked and persistent reductionin TLR.

Another important finding in our study is the significantlylower risk of definite stent thrombosis in patients treated witha glycoprotein IIb/IIIa receptor blocker. In the CADILLACtrial,22 subacute thrombosis was reduced by 74% in patientstreated with abciximab. However, abciximab treatment didnot affect the composite end point at 1 year. Although theglycoprotein IIb/IIIa receptor blocker treatment in our studywas administered at the discretion of the PCI operator, weobserved a similar risk reduction of 53% in definite stentthrombosis after glycoprotein IIb/IIIa receptor blocker ad-ministration compared with abciximab administration. In theSTRATEGY trial,23 patients treated with DES plus tirofibanhad less angiographic stent thrombosis at 30 days comparedwith patients treated with the combination of BMS plusabciximab (0 versus 3 stent thromboses). In the recentlypublished MULTISTRATEGY trial,24 tirofiban therapy,when compared with abciximab, was associated with nonin-ferior resolution of ST-segment elevation at 90 minutes aftercoronary intervention, whereas sirolimus-eluting stent im-plantation was associated with a significantly lower thanuncoated stents risk of major adverse cardiac events within 8months after intervention. The ADMIRAL study25 did notpresent data on stent thrombosis, but it demonstrated thatearly inhibition of the platelet glycoprotein (GP) IIb/IIIareceptor with abciximab led to improved coronary patency,left ventricular function, and better clinical outcomes. Long-term follow-up in the ADMIRAL study25,26 showed thatadjunctive abciximab to primary stenting for STEMI elicits

Table 3. Risk of Definite Stent Thrombosis in Patients With STEMI Treated With Primary PCI and DES (n�983) or BMS(n�3486) in Western Denmark

DES BMS

RR* (95% CI) PNo. of Events No./100 Person-Year No. of Events No./100 Person-Year

All 19 1.05 38 0.61 1.53 (0.84–2.78) 0.17

Acute (24 h) 5 187.7 9 95.9 1.56 (0.48–5.09) 0.46

1 day to �30 days 7 9.1 21 7.8 1.04 (0.41–2.60) 0.94

30 days to 12 months 3 0.4 6 0.2 1.36 (0.26–7.03) 0.72

�12 months 4 0.5 2 0.1 6.74 (1.23–37.0) 0.03

*Adjusted for age, sex, diabetes mellitus, stent length, reference vessel size, procedure time, and glycoprotein IIb/IIIa.




Figure 2. A, Risk of definite stent thrombosis among patients treated with DES or BMS. B, Risk of definite, probable, and possiblestent thrombosis among patients treated with DES or BMS. C, All-cause mortality among patients treated with DES or BMS. D, Risk ofMI among patients treated with DES or BMS. E, TLR among patients treated with DES or BMS (adjusted data).




favorable clinical outcomes with the same absolute reduc-tions in risks of “hard” clinical outcomes from 30 days up to3 years of follow-up.

Mortality in the current study was generally higher forpatients treated with DES or BMS compared with mortalityestimates reported in both a meta-analysis27 and in random-ized clinical trials of DES versus BMS in primary PCI.3–6

This may be important for the interpretation of our data,because randomized trials (and the meta-analysis based onthese trials) usually include highly selected patients. Thus,our study extends the results of the randomized trials to areal-world setting. Clinically driven TLR was lower in ourstudy than in the randomized trials, probably because routineangiographic follow-up in such trials increases the TLR rate.

The current data did not demonstrate significant differ-ences in the cumulative rates of death or MI after 2 years forpatients treated with DES or BMS. Furthermore, all-causemortality and cardiac mortality for patients treated withprimary PCI were similar to the findings in the DANAMI-2study.28

LimitationsThe validity of our findings depends on data quality and theability to control for potential confounding. Our design isbased on computerized registries with complete nationwidecoverage, enabling study of a well-defined large populationwith complete follow-up. However, like all observationalstudies, our study is prone to biases related to unmeasuredfactors. A bias attributable to an unknown variable cannot beeliminated.

We collected data over a 3-year period, during which theprevalence of DES stents implanted increased from zero to

53%. To reduce bias during this transition period, wefollowed-up every patient for up to 24 months. It also isimportant to note that although we report 12 months as therecommended duration of dual antiplatelet treatment, we hadonly access to the actual duration of this treatment amongpatients with definite stent thrombosis. Finally, according tothe ARC definition of probable and possible stent thrombosis,we categorized all unexplained deaths as stent thromboses.However, it is unlikely that stent thrombosis was the cause ofdeath in all cases of unexplained death.

ConclusionsThe results of the current study indicated that patients treatedwith stent implantation during primary PCI the overall ratesof stent thrombosis, MI, and death were of similar magnitudein patients receiving DES or BMS after 2 years. Comparedwith BMS, DES improved the clinical outcome by reducingthe risk of clinically driven TLR.

DisclosuresDr Jensen has received lecture fees from Abbott, AstraZeneca,Cordis, and Merck; Dr Maeng has received lecture fees from Cordisand consultant fees from Novo Nordisk and Medtronic; Dr Kaltofthas received lecture fees from Cordis; Dr Lassen has received lecturefees from Bristol-Meyers Squibb, Nycomed, Boston Scientific,GlaxoSmithKline, and Cordis; and Dr Thuesen has received lecturefees from Medtronic, Abbot, Cordis, and Boston Scientific, consult-ing fees from Abbott, is in the advisory board of Boston Scientific,and has received research grant support from Medtronic, Abbot,Cordis, and Boston Scientific.

References1. Andersen HR, Nielsen TT, Rasmussen K, Thuesen L, Kelbaek H,

Thayssen P, Abildgaard U, Pedersen F, Madsen JK, Grande P, VilladsenAB, Krusell LR, Haghfelt T, Lomholt P, Husted SE, Vigholt E, Kjaergard

Table 4. Relative Risk Estimates for Death, MI, and Stent Thrombosis (Definite, Probable, and Possible) in Patients With STEMITreated With Primary PCI and DES (n�783) or BMS (n�2973) in Western Denmark

Patients With DES Patients With BMS

RR* (95% CI) PNo. of Events No./100 Person-Years No. of Events No./100 Person-Years

All-cause death 61 4.1 339 6.3 0.79 (0.60–1.04) 0.09

�12 months 45 6.0 275 10.0 0.73 (0.53–1.00) 0.05

�12 months 16 2.2 64 2.4 1.04 (0.59–1.83) 0.90

Cardiac 39 2.6 209 3.9 0.83 (0.59–1.18) 0.31

�12 months 32 4.3 190 6.9 0.75 (0.51–1.10) 0.14

�12 months 7 1.0 19 0.7 1.66 (0.69–3.99) 0.25

Noncardiac 20 1.4 105 1.9 0.82 (0.51–1.33) 0.42

�12 months 12 1.6 70 2.6 0.75 (0.40–1.38) 0.35

�12 months 8 1.1 35 1.3 0.97 (0.45–2.09) 0.94

MI 47 3.5 145 2.9 1.13 (0.81–1.59) 0.47

28 days to 12 months 31 4.7 113 4.6 0.94 (0.62–1.41) 0.75

�12 months 16 2.3 32 1.3 1.85 (1.01–3.40) 0.05

Overall stent thrombosis 28 2.73 106 2.87 1.09 (0.71–1.67) 0.68

Acute (24 h) 6 282.8 19 237.3 1.09 (0.44–2.66) 0.86

1 day to � 30 days 11 17.9 54 23.4 0.96 (0.50–1.85) 0.90

30 days to 12 months 6 0.89 23 0.93 0.95 (0.36–2.50) 0.91

�12 months 5 0.7 10 0.4 2.06 (0.70–6.04) 0.19

*Adjusted for age, sex, diabetes mellitus, and procedure time.




HK, Mortensen LS. A comparison of coronary angioplasty with fibrinolytictherapy in acute myocardial infarction. N Engl J Med. 2003;349:733–742.

2. Keeley EC, Boura JA, Grines CL. Comparison of primary and facilitatedpercutaneous coronary interventions for ST-elevation myocardialinfarction: quantitative review of randomised trials. Lancet. 2006;367:579–588.

3. Menichelli M, Parma A, Pucci E, Fiorilli R, De FF, Nazzaro M, GiuliviA, Alborino D, Azzellino A, Violini R. Randomized trial of sirolimus-eluting stent versus bare-metal stent in acute myocardial infarction(SESAMI). J Am Coll Cardiol. 2007;49:1924–1930.

4. Spaulding C, Henry P, Teiger E, Beatt K, Bramucci E, Carrie D, SlamaMS, Merkely B, Erglis A, Margheri M, Varenne O, Cebrian A, Stoll HP,Snead DB, Bode C. Sirolimus-eluting versus uncoated stents in acutemyocardial infarction. N Engl J Med. 2006;355:1093–1104.

5. Valgimigli M, Campo G, Arcozzi C, Malagutti P, Carletti R, Ferrari F,Barbieri D, Parrinello G, Percoco G, Ferrari R. Two-year clinicalfollow-up after sirolimus-eluting versus bare-metal stent implantationassisted by systematic glycoprotein IIb/IIIa inhibitor infusion in patientswith myocardial infarction: results from the STRATEGY study. J AmColl Cardiol. 2007;50:138–145.

6. Laarman GJ, Suttorp MJ, Dirksen MT, van HL, Kiemeneij F, SlagboomT, van der Wieken LR, Tijssen JG, Rensing BJ, Patterson M. Paclitaxel-eluting versus uncoated stents in primary percutaneous coronary inter-vention. N Engl J Med. 2006;355:1105–1113.

7. Stone GW, Moses JW, Ellis SG, Schofer J, Dawkins KD, Morice MC,Colombo A, Schampaert E, Grube E, Kirtane AJ, Cutlip DE, Fahy M,Pocock SJ, Mehran R, Leon MB. Safety and efficacy of sirolimus- andpaclitaxel-eluting coronary stents. N Engl J Med. 2007;356:998–1008.

8. Lagerqvist B, James SK, Stenestrand U, Lindback J, Nilsson T, WallentinL. Long-term outcomes with drug-eluting stents versus bare-metal stentsin Sweden. N Engl J Med. 2007;356:1009–1019.

9. Pfisterer M, Brunner-La Rocca HP, Buser PT, Rickenbacher P, HunzikerP, Mueller C, Jeger R, Bader F, Osswald S, Kaiser C. Late clinical eventsafter clopidogrel discontinuation may limit the benefit of drug-elutingstents: an observational study of drug-eluting versus bare-metal stents.J Am Coll Cardiol. 2006;48:2584–2591.

10. Jensen LO, Maeng M, Kaltoft A, Thayssen P, Hansen HH, Bottcher M,Lassen JF, Krussel LR, Rasmussen K, Hansen KN, Pedersen L, JohnsenSP, Soerensen HT, Thuesen L. Stent thrombosis, myocardial infarction,and death after drug-eluting and bare-metal stent coronary interventions.J Am Coll Cardiol. 2007;50:463–470.

11. Tu JV, Bowen J, Chiu M, Ko DT, Austin PC, He Y, Hopkins R, TarrideJE, Blackhouse G, Lazzam C, Cohen EA, Goeree R. Effectiveness andsafety of drug-eluting stents in Ontario. N Engl J Med. 2007;357:1393–1402.

12. Daemen J, Tanimoto S, Garcia-Garcia HM, Kukreja N, van de SM,Sianos G, de Jaegere PP, van Domburg RT, Serruys PW. Comparison ofthree-year clinical outcome of sirolimus- and paclitaxel-eluting stentsversus bare metal stents in patients with ST-segment elevation myocardialinfarction (from the RESEARCH and T-SEARCH Registries). Am JCardiol. 2007;99:1027–1032.

13. Cutlip DE, Windecker S, Mehran R, Boam A, Cohen DJ, Van Es GA,Steg PG, Morel MA, Mauri L, Vranckx P, McFadden E, Lansky A,Hamon M, Krucoff MW, Serruys PW. Clinical end points in coronarystent trials: a case for standardized definitions. Circulation. 2007;115:2344–2351.

14. Tunstall-Pedoe H, Kuulasmaa K, Amouyel P, Arveiler D, RajakangasAM, Pajak A. Myocardial infarction and coronary deaths in the WorldHealth Organization MONICA Project: Registration procedures, eventrates, and case-fatality rates in 38 populations from 21 countries in fourcontinents. Circulation. 1994;90:583–612.

15. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method ofclassifying prognostic comorbidity in longitudinal studies: developmentand validation. J Chronic Dis. 1987;40:373–383.

16. Greenland S. Modeling and variable selection in epidemiologic analysis.Am J Public Health. 1989;79:340–349.

17. Sturmer T, Joshi M, Glynn RJ, Avorn J, Rothman KJ, Schneeweiss S.A review of the application of propensity score methods yieldedincreasing use, advantages in specific settings, but not substantially dif-ferent estimates compared with conventional multivariable methods.J Clin Epidemiology. 2008;59:437–447.

18. Cook S, Wenaweser P, Togni M, Billinger M, Morger C, Seiler C, VogelR, Hess O, Meier B, Windecker S. Incomplete stent apposition and verylate stent thrombosis after drug-eluting stent implantation. Circulation.2007;115:2426–2434.

19. Fujii K, Carlier SG, Mintz GS, Yang YM, Moussa I, Weisz G, Dangas G,Mehran R, Lansky AJ, Kreps EM, Collins M, Stone GW, Moses JW,Leon MB. Stent underexpansion and residual reference segment stenosisare related to stent thrombosis after sirolimus-eluting stent implantation:an intravascular ultrasound study. J Am Coll Cardiol. 2005;45:995–998.

20. Ault KA, Cannon CP, Mitchell J, McCahan J, Tracy RP, Novotny WF,Reimann JD, Braunwald E. Platelet activation in patients after an acutecoronary syndrome: results from the TIMI-12 trial. Thrombolysis inMyocardial Infarction. J Am Coll Cardiol. 1999;33:634–639.

21. Luscher TF, Steffel J, Eberli FR, Joner M, Nakazawa G, Tanner FC,Virmani R. Drug-eluting stent and coronary thrombosis: biological mech-anisms and clinical implications. Circulation. 2007;115:1051–1058.

22. Tcheng JE, Kandzari DE, Grines CL, Cox DA, Effron MB, Garcia E,Griffin JJ, Guagliumi G, Stuckey T, Turco M, Fahy M, Lansky AJ,Mehran R, Stone GW. Benefits and risks of abciximab use in primaryangioplasty for acute myocardial infarction: the Controlled Abciximaband Device Investigation to Lower Late Angioplasty Complications(CADILLAC) trial. Circulation. 2003;108:1316–1323.

23. Valgimigli M, Percoco G, Malagutti P, Campo G, Ferrari F, Barbieri D,Cicchitelli G, McFadden EP, Merlini F, Ansani L, Guardigli G, Bettini A,Parrinello G, Boersma E, Ferrari R. Tirofiban and sirolimus-eluting stentvs abciximab and bare-metal stent for acute myocardial infarction: arandomized trial. JAMA. 2005;29:2109–2117.

24. Valgimigli M, Campo G, Percoco G, Bolognese L, Vassanelli C, Col-angelo S, de CN, Rodriguez AE, Ferrario M, Moreno R, Piva T, SheibanI, Pasquetto G, Prati F, Nazzaro MS, Parrinello G, Ferrari R. Comparisonof angioplasty with infusion of tirofiban or abciximab and with implan-tation of sirolimus-eluting or uncoated stents for acute myocardialinfarction: the MULTISTRATEGY randomized trial. JAMA. 2008;299:1788–1799.

25. Montalescot G, Barragan P, Wittenberg O, Ecollan P, Elhadad S, VillainP, Boulenc JM, Morice MC, Maillard L, Pansieri M, Choussat R, PintonP. Platelet glycoprotein IIb/IIIa inhibition with coronary stenting for acutemyocardial infarction. N Engl J Med. 2001;344:1895–1903.

26. Three-year duration of benefit from abciximab in patients receiving stentsfor acute myocardial infarction in the randomized double-blindADMIRAL study. Eur Heart J. 2005;26:2520–2523.

27. Kastrati A, Dibra A, Spaulding C, Laarman GJ, Menichelli M, ValgimigliM, Di LE, Kaiser C, Tierala I, Mehilli J, Seyfarth M, Varenne O, DirksenMT, Percoco G, Varricchio A, Pittl U, Syvanne M, Suttorp MJ, Violini R,Schomig A. Meta-analysis of randomized trials on drug-eluting stents vs.bare-metal stents in patients with acute myocardial infarction. EurHeart J. 2007;28:2706–2713.

28. Busk M, Maeng M, Rasmussen K, Kelbaek H, Thayssen P, Abildgaard U,Vigholt E, Mortensen LS, Thuesen L, Kristensen SD, Nielsen TT,Andersen HR. The Danish multicentre randomized study of fibrinolytictherapy vs. primary angioplasty in acute myocardial infarction (theDANAMI-2 trial): outcome after 3 years follow-up. Eur Heart J. 2008;29:1259–1266.




and Leif ThuesenRasmussen, Knud Erik Pedersen, Lars Pedersen, Søren Paaske Johnsen, Henrik Toft SørensenMorten Bøttcher, Jens Flensted Lassen, Knud Nørregaard Hansen, Lars Romer Krusell, Klaus

Lisette Okkels Jensen, Michael Maeng, Per Thayssen, Anne Kaltoft, Hans Henrik Tilsted,and Bare Metal Stents in Patients With ST-Segment Elevation Myocardial Infarction

Clinical Outcome After Primary Percutaneous Coronary Intervention With Drug-Eluting

Print ISSN: 1941-7640. Online ISSN: 1941-7632 Copyright © 2008 American Heart Association, Inc. All rights reserved.

Avenue, Dallas, TX 75231is published by the American Heart Association, 7272 GreenvilleCirculation: Cardiovascular Interventions

doi: 10.1161/CIRCINTERVENTIONS.108.7945782008;1:176-184Circ Cardiovasc Interv.

http://circinterventions.ahajournals.org/content/1/3/176World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circinterventions.ahajournals.org//subscriptions/

is online at: Circulation: Cardiovascular Interventions Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Answer

Permissions and Rights Question andunder Services. Further information about this process is available in thepermission is being requested is located, click Request Permissions in the middle column of the Web pageClearance Center, not the Editorial Office. Once the online version of the published article for which

can be obtained via RightsLink, a service of the CopyrightCirculation: Cardiovascular Interventionsin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:


http://circinterventions.ahajournals.org/content/1/3/176

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circinterventions.ahajournals.org//subscriptions/


Clinical outcome after primary percutaneous coronary intervention with drug-eluting and bare metal...

Documents

Transcript of Clinical outcome after primary percutaneous coronary intervention with drug-eluting and bare metal...