doi:10.1016/j.jcin.2008.03.019 2008;1;369-378 J. Am. Coll. Cardiol. Intv.

RoeWhite, Philip E. Aylward, James J. Ferguson, Robert M. Califf, and Matthew T. Mark Y. Chan, Kenneth W. Mahaffey, Lena J. Sun, Karen S. Pieper, Harvey D.

Have Angiographically Documented Significant Coronary DiseasePatients With Non�ST-Segment Elevation Acute Coronary Syndromes Who

Prevalence, Predictors, and Impact of Conservative Medical Management for

This information is current as of May 21, 2011

http://interventions.onlinejacc.org/cgi/content/full/1/4/369located on the World Wide Web at:

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

by on May 21, 2011 interventions.onlinejacc.orgDownloaded from

PCWSS

MKJM

D

Oma

Bbs

MsEtc

R1atmd(

CrhpNC

FHLS

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 . 4 , 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 8 . 0 3 . 0 1 9

revalence, Predictors, and Impact ofonservative Medical Management for Patientsith Non–ST-Segment Elevation Acute Coronary

yndromes Who Have Angiographically Documentedignificant Coronary Disease

ark Y. Chan, MD, MHS,* Kenneth W. Mahaffey, MD, FACC,* Lena J. Sun, MS,*aren S. Pieper, MS,* Harvey D. White, MD, FACC,† Philip E. Aylward, MD, FACC,‡

ames J. Ferguson, MD, FACC,§ Robert M. Califf, MD, FACC,*atthew T. Roe, MD, MHS, FACC*

urham, North Carolina; Auckland, New Zealand; Adelaide, Australia; and Houston, Texas

bjectives We sought to characterize the utilization and impact of a conservative medical manage-ent strategy for patients with non–ST-segment elevation acute coronary syndromes (NSTE ACS)nd significant coronary artery disease on early angiography.

ackground Practice guidelines recommend an early invasive management strategy for NSTE ACS,ut revascularization procedures may not always be performed after early angiography, even whenignificant coronary artery disease is present.

ethods We evaluated 8,225 intermediate- to high-risk NSTE ACS patients with at least 1 coronary le-ion �50% stenosis on early angiography from the SYNERGY (Superior Yield of the New Strategy ofnoxaparin, Revascularization, and Glycoprotein IIb/IIIa Inhibitors) trial (2001 to 2003), comparing patientsreated with conservative medical management with those who underwent in-hospital percutaneousoronary intervention (PCI) or coronary artery bypass grafting (CABG) within 7 days of randomization.

esults A total of 2,633 patients (32%) were medically managed, 4,294 (52%) underwent PCI, and,298 (16%) underwent CABG. The strongest independent predictors of conservative medical man-gement versus any intervention were prior CABG, lower body weight, lack of a reinfarction be-ween randomization and catheterization, and 3-vessel disease. With conservative medical manage-ent, the cumulative risk of 1-year mortality after discharge increased rapidly during the first 90ays and thereafter remained higher at 7.7% compared with that seen in patients treated with PCI3.6%) or CABG (6.2%).

onclusions One-third of patients with NSTE ACS and significant coronary disease on early angiog-aphy were managed without in-hospital revascularization in the SYNERGY trial, and these patientsad an increased risk of late mortality. These findings highlight the need for novel treatment ap-roaches for NSTE ACS patients who are not candidates for revascularization. (SYNERGY trial;CT00043784) (J Am Coll Cardiol Intv 2008;1:369–78) © 2008 by the American College ofardiology Foundation

rom the *Duke Clinical Research Institute, Durham, North Carolina; †Green Lane Cardiovascular Service, Auckland Cityospital, Auckland, New Zealand; ‡Flinders Medical Center, Adelaide, Australia; and the §Texas Heart Institute, St.uke’s-Episcopal Hospital, Houston, Texas. Dr. Chan receives salary support from the National Medical Research Council,ingapore. The SYNERGY trial was funded by Sanofi-Aventis.

anuscript received February 14, 2008; accepted March 19, 2008.

by on May 21, 2011 interventions.onlinejacc.orgDownloaded from

Cmrsdiptfciic8adspe

M

SyrNcpDootpshSlsh

el(railcalwOwptvtcowaSsiCircpcustCri

Aa

Cb

Cd

D

GA

M

Nsc

Pc

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 . 4 , 2 0 0 8

A U G U S T 2 0 0 8 : 3 6 9 – 7 8

Chan et al.

Conservative Management of Significant Coronary Disease for NSTE ACS

370

linical practice guidelines recommend an early invasiveanagement treatment strategy for intermediate- to high-

isk patients with non–ST-segment elevation acute coronaryyndromes (NSTE ACS) (1,2). However, angiography is aiagnostic procedure that does not confer treatment benefit

f patients do not undergo a subsequent revascularizationrocedure (3). In 4 randomized trials comparing conserva-ive medical management with early invasive managementor NSTE ACS, the majority of patients had significantoronary disease identified on angiography in the earlynvasive treatment arms (75% to 87%), but the use ofn-hospital percutaneous coronary intervention (PCI) ororonary artery bypass grafting (CABG) varied from 59% to2% (4–7). Data from a contemporary registry showed thatpproximately 25% of NSTE ACS patients with 3-vesselisease did not undergo revascularization (8). These datauggest that many NSTE ACS patients in contemporaryractice with significant coronary artery disease (CAD) onarly angiography do not undergo early revascularization,

but the long-term consequencesof a conservative medical man-agement approach have not beenfully explored.

We, therefore, sought to de-scribe the prevalence and predic-tors of a conservative medicalmanagement treatment strategyfor patients with NSTE ACS andsignificant CAD on early angiog-raphy, and to investigate the long-term risk of adverse outcomes inthese patients compared with thatseen in patients who underwentrevascularization.

ethods

tudy design and subjects. We performed a subgroup anal-sis of patients with significant disease on coronary angiog-aphy enrolled in the SYNERGY (Superior Yield of theew Strategy of Enoxaparin, Revascularization, and Gly-

oprotein IIb/IIIa Inhibitors) trial, which randomized 9,978atients from 12 countries between August 2001 andecember 2003 (9,10). The SYNERGY trial was an

pen-label trial designed to evaluate the efficacy and safetyf enoxaparin versus unfractionated heparin when adminis-ered with established guidelines-recommended therapy inatients with NSTE ACS planned for an early invasivetrategy. Patients were eligible for study participation if theyad �2 of the following high-risk features: age �60 years,T-segment changes, and a troponin or creatine kinase-MB

evel exceeding the upper limit of normal. The majority ofubjects underwent cardiac catheterization during the initial

bbreviationsnd Acronyms

ABG � coronary arteryypass grafting

AD � coronary arteryisease

ES � drug-eluting stent(s)

RACE � Global Registry ofcute Coronary Events score

I � myocardial infarction

STE ACS � non–ST-egment elevation acuteoronary syndromes

CI � percutaneousoronary intervention

ospitalization (92%).interventions.onlinDownloaded from

We included only patients who underwent cardiac cath-terization during the initial hospitalization with 1 or moreesions �50% severity in a native epicardial coronary arteryleft main, left anterior descending, circumflex artery, oright coronary artery) or a major side branch of a nativertery. Angiographic lesion severity was recorded by sitenvestigators without utilization of an angiographic coreaboratory. We excluded patients who did not undergoatheterization during the initial hospitalization (n � 880)nd patients with normal coronaries on angiography or allesions on angiography �50% (n � 873), as these patientsould not be candidates for revascularization (Fig. 1).utcomes ascertainment. Data on in-hospital outcomesere obtained by reviewing clinical records for all enrolledatients. Outcomes after discharge were captured by con-acting patients by telephone or through scheduled patientisits through 1 year, and 1-year mortality was ascertainedhrough medical records or national death indexes. Adjudi-ation was performed for all nonfatal myocardial eventsccurring through 6 months. The total number of subjectsho were lost to follow-up for the mortality end point at 6

nd 12 months were 16 and 35, respectively.tatistical methods. UNADJUSTED ANALYSIS. We divided thetudy population into 3 groups: 1) those who underwentn-hospital PCI; 2) those who underwent in-hospitalABG; and 3) those who were medically managed without

n-hospital revascularization performed within 7 days ofandomization. Data were displayed as percentages forategorical variables and medians (25th percentile, 75thercentile) for continuous variables. Comparison of baselineharacteristics, angiographic results, in-hospital medicationse, and outcomes was performed using the Pearson chi-quare test for categorical variables and the Kruskal-Wallisest for continuous variables. The Global Registry of Acuteoronary Events (GRACE) score has been established as a

obust tool for predicting 6-month post-discharge mortalityn patients with ACS (11). In order to provide a single

2633 MM 4294 PCI 1298 CABG

83 Underwent catheterization after 7

days

9188 Underwent in-hospital

catheterization

8225 Significant CAD

880 Normal coronaries/minor

CAD

9105 Underwent in-hospital

catheterization within 7 days

Figure 1. Subject Flow

CABG � coronary artery bypass grafting; CAD � coronary artery disease;MM � conservative medical management; PCI � percutaneous coronary

intervention.

by on May 21, 2011 ejacc.org

ngs3lMmP

dpvspNctddwocP

bocwmmCemcatrMapaaCmcPlormcMmfc

T

Cmmaoatf

ctI

R

Upaetp(Tic6(c(Ciclvtnsa1AgpoidlgCbate

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 . 4 , 2 0 0 8

A U G U S T 2 0 0 8 : 3 6 9 – 7 8

Chan et al.

Conservative Management of Significant Coronary Disease for NSTE ACS

371

umerical estimate of risk for each of the 3 treatmentroups, we calculated the GRACE risk scores for allubjects studied. Unadjusted in-hospital events rates in the

groups were compared using the chi-square test, whileong-term event rates were estimated using the Kaplan-

eier method with pairwise comparisons between groupsade using the log-rank test.

REDICTORS OF IN-HOSPITAL REVASCULARIZATION. Weeveloped a logistic regression model to identify significantredictors of in-hospital conservative medical managementersus in-hospital revascularization (Model 1) using bothtepwise and backward selection techniques on 37 candidateredictors derived from a large observational registry ofSTE ACS patients (12). Nonfatal clinical outcomes

aptured before catheterization were entered as predictors inhe model. The spline transformation method was used toetermine the functional form for continuous variables thatid not meet linearity assumptions. Whenever appropriate,e applied piecewise linear splines as the preferred methodf transformation. A p value of 0.05 was used as theriterion for variable entry and retention in the model.OST-DISCHARGE MORTALITY. To evaluate the associationetween conservative medical management and long-termutcomes, we developed a proportional hazards modelomparing the conservative medical management groupith the PCI and CABG groups individually (conservativeedical management vs. PCI and conservative medicalanagement vs. CABG), as patients undergoing PCI andABG often have very different clinical profiles (13). We

xamined the association between conservative medicalanagement as an independent variable and death (all-

ause mortality) or myocardial infarction (MI) at 6 monthss a dependent variable occurring after a discharge landmarkime point, defined as time of discharge, or 7 days fromandomization if hospitalization was prolonged (Model 2).

odeling outcomes using a landmark time point enabledccounting for the ascertainment bias expected when com-aring early MI rates between conservative medical man-gement patients (in whom markers were drawn only whensuspected recurrent ischemic event occurred) and PCI orABG patients (in whom serial markers were recom-ended to be routinely drawn post-procedure); as such a

omparison artificially inflates the relative rate of MI in theCI and CABG groups (14). The use of a discharge

andmark further permitted the modeling of nonfatal eventsccurring before the landmark time point, including recur-ent ischemia, recurrent MI, acute heart failure, stroke, andajor bleeding, as time-updated covariates reflecting

hanges in dynamic risk status (15). We then adjustedodel 2 for the propensity for in-hospital conservativeedical management by including propensity scores derived

rom data captured before catheterization (Model 1) as a

ovariate in Model 2. f

interventions.onlinDownloaded from

IME-DEPENDENT COVARIATE ANALYSIS. We developed aox proportional hazards model that included conservativeedical management versus PCI and conservative medicalanagement versus CABG as time-dependent covariates

nd death from catheterization through 1 year as theutcome variable (Model 3). Time-dependent covariatesccount for the timing of PCI and CABG relative to theime of catheterization and the outcome of interest, there-ore serving as a means to reduce survivor bias (16).

All p values were two-tailed and alpha set at 0.05 withonfidence intervals calculated to the 95th percentile. Sta-istical analyses were performed using SAS version 8.2 (SASnstitute Inc., Cary, North Carolina).

esults

tilization of revascularization procedures. A total of 8,225atients were found to have significant CAD on earlyngiography, and 4,752 of these patients (57.8%) werenrolled from U.S. sites. In-hospital revascularization beforehe discharge landmark time point was performed in 5,592atients (68%), with 4,294 patients undergoing PCI52.2%) and 1,298 patients undergoing CABG (15.8%).he median time from randomization to cardiac catheter-

zation was 21 (6, 42) h, while the median times fromatheterization to PCI and CABG were 0.5 (0.2, 1.2) h and4.8 (23.5, 136.1) h, respectively. A total of 2,633 patients32%) were medically managed without in-hospital revas-ularization before the discharge landmark time pointFig. 1).linical characteristics. Compared with patients undergoing

n-hospital revascularization (PCI/CABG), patients in theonservative medical management group were older, moreikely to be women, and more likely to have peripheralascular disease, prior stroke, hypertension, diabetes melli-us, hypercholesterolemia, prior CABG, and lower creati-ine clearance values (Table 1). The median GRACE riskcore was similar for the CABG group at 124 (110, 141)nd conservative medical management group at 121 (106,36), but was lower for the PCI group at 103 (90, 119).ngiographic results. Patients in the CABG group had thereatest frequency of left main and 3-vessel disease, whileatients in the PCI group had the greatest frequency of 1-r 2-vessel disease (Table 2). More than half of the patientsn the conservative medical management group had 3-vesselisease. Median left ventricular ejection fraction values were

owest in the conservative medical management and CABGroups.oncomitant medication utilization. The use of aspirin,eta-blockers, angiotensin-converting enzyme inhibitors,nd statins was high in all groups during the initial hospi-alization, at discharge, and at 30 days (Table 3). The use ofvidence-based medications was highest in the PCI group

ollowed by the conservative medical management group.

by on May 21, 2011 ejacc.org

Pvmcbud

RrdiuomlvtphIrt(pPoac

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 . 4 , 2 0 0 8

A U G U S T 2 0 0 8 : 3 6 9 – 7 8

Chan et al.

Conservative Management of Significant Coronary Disease for NSTE ACS

372

redictors of conservative medical management. Severalariables were independently associated with conservativeedical management (Table 4). The strongest predictors of

onservative medical management were prior CABG, lowerody weight, lack of a recurrent MI from randomizationntil the time of catheterization, and 3-vessel diseaseemonstrated on angiography.

Table 1. Baseline Characteristics

CharacteristicMM

(n � 2,633)

Median age, yrs 15.3 (9.5, 21.0)

Female gender, % 32.8

Race, %

White 83.6

Black 7.3

Asian 1.9

Hispanic 4.6

U.S. enrollment, % 44.4

Hypertension, % 70.9

Diabetes, % 33.3

Smoking, % 35.6

Hypercholesterolemia, % 61.8

Prior MI, % 34.2

Prior heart failure, % 12.2

History of stroke, % 6.1

History of PVD, % 11.9

Prior CABG, % 23.0

Prior PCI, % 22.5

Hours from symptom onset (quartiles 1 and 3) 15.3 (9.5, 21.0)

Killip class II–IV, % 14.8

Systolic BP, mm Hg 132 (118, 150)

Heart rate, beats/min 70 (62, 81)

Creatinine clearance, ml/min* 71.1 (53.6, 92.5)

ST-segment depression on ECG, % 57.5

Elevated cardiac enzymes, %† 64.4

Values in parentheses are 25th, 75th percentile. *Creatinine clearance calculated using the Cockrof

BP � blood pressure; CABG � coronary artery bypass grafting; ECG � electrocardiogram; MI � m

PVD � peripheral vascular disease.

Table 2. Angiography and LV Function

MM(n � 2,633)

PCI(n � 4,294)

CABG(n � 1,298) p Value

Angiographic data,number (%)*

�0.001

Left main disease 9.8 3.9 28.1

1-vessel disease† 24.2 37.2 3.4

2-vessel disease† 22.9 33.7 17.1

3-vessel disease† 52.9 29.1 79.5

LVEF, %‡ 50 (40, 60) 55 (45, 61) 50 (40, 60) �0.001

Values in parentheses are 25th, 75th percentile. *Number of patients with �50% stenosis in each

of the following epicardial arteries: left main, left anterior descending, circumflex, or right

coronary artery; †Excluding left main disease; ‡Left ventricular ejection fraction (LVEF)

determined by echocardiography, cardiac catheterization, or radionuclide angiography.

tLV � left ventricular; other abbreviations as in Table 1.

interventions.onlinDownloaded from

evascularization procedures during follow-up. Overall,ates of late or repeat revascularization occurring afterischarge were low (Fig. 2A). Patients who underwentn-hospital CABG before 7 days were the least likely tondergo repeat revascularization procedures, while the ratef late revascularization was similar in the conservativeedical management and PCI groups. The median time to

ate revascularization (PCI) after discharge for the conser-ative medical management group was substantially shorterhan the median time to repeat revascularization for thoseatients who had received PCI/CABG during the indexospitalization (Fig. 2B).n-hospital outcomes. Unadjusted in-hospital events occur-ing between catheterization and the discharge landmarkime point occurred most frequently in the CABG groupTable 5). However, the in-hospital mortality curves ap-eared to be similar for the 3 comparison groups (Fig. 3A).ost-discharge outcomes. Unadjusted post-discharge eventsccurred most frequently in the conservative medical man-gement group (Table 6). The post-discharge mortalityurves demonstrated an early hazard for both the conserva-

PCI(n � 4,294)

CABG(n � 1,298) p Value

67 (60, 74) 67 (61, 74) �0.001

31.7 25.9 �0.001

�0.001

87.4 85.5

5.4 6.3

0.7 0.8

4.7 5.5

59.5 79.1 �0.001

65.8 70.3 �0.001

26.2 33.2 �0.001

34.2 36.1 �0.001

59.4 60.0 0.144

25.9 23.4 �0.001

6.5 6.8 �0.001

3.8 5.2 �0.001

8.7 9.1 �0.001

16.1 6.7 �0.001

21.6 15.2 �0.001

14.4 (8.1, 20.5) 14.3 (8.0, 20.4) �0.001

8.9 13.1 �0.001

130 (117, 146.0) 130 (117, 147) 0.033

70 (62, 80) 72 (64, 83) �0.001

79.2 (60.6, 101.8) 77.4 (57.8, .98.6) �0.001

53.0 59.6 �0.001

70.2 69.9 �0.001

formula; †troponin or creatine kinase-MB fraction above the upper limit of normal.

l infarction; MM � conservative medical management; PCI � percutaneous coronary intervention;

t-Gault

yocardia

ive medical management and CABG groups, but after 90 by on May 21, 2011 ejacc.org

dh3htirml

(amcmOohraastwtad

D

OhmsinaomUos

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 . 4 , 2 0 0 8

A U G U S T 2 0 0 8 : 3 6 9 – 7 8

Chan et al.

Conservative Management of Significant Coronary Disease for NSTE ACS

373

ays, the curves diverged and the mortality hazard appearedigher in the conservative medical management group (Fig.B). Because the conservative medical management groupad both an increased rate of post-discharge revasculariza-ion procedures and an earlier median time to revascular-zation compared with the PCI and CABG groups, weepeated the landmark analysis excluding all medicallyanaged patients with revascularization performed after the

andmark time point (discharge or 7 days) through 30 days

Table 3. Evidence-Based Medication Use at Baseline, Discharge,and 30 Days

MM(n � 2,633)

PCI(n � 4,294)

CABG(n �1,298) p Value

In-hospital

Aspirin, % 94.8 96.0 94.8 0.032

Beta-blocker, % 85.7 87.2 91.2 �0.001

ACE inhibitor, % 67.6 61.3 62.6 �0.001

Statin, % 71.6 71.9 71.2 0.869

Clopidogrel, % 52.8 84.3 34.8 �0.001

At discharge

Aspirin, % 90.7 96.1 89.7 �0.001

Beta-blocker, % 85.4 90.1 80.2 �0.001

ACE inhibitor, % 75.2 77.5 55.8 �0.001

Statin, % 82.4 86.7 70.9 �0.001

Clopidogrel, % 37.9 84.2 14.2 �0.001

At 30 days

Aspirin, % 90.3 93.5 88.7 �0.001

Beta-blocker, % 85.4 87.0 78.7 �0.001

ACE inhibitor, % 72.2 72..9 57.5 �0.001

Statin, % 82.1 85.5 74.1 �0.001

Clopidogrel, % 35.5 68.8 13.9 �0.001

ACE � angiotensin-converting enzyme; other abbreviations as in Table 1.

Table 4. Independent Predictors of Conservative Medical Management (Mo

Predictor Wa

History of prior CABG

Body weight per 10 kg decrease

MI from randomization to time of catheterization

3-vessel disease

Non-U.S. site

History of prior heart failure

Time from symptom onset to presentation (per hour increase)

Black race

Killip class II–IV

History of prior MI

Diabetes mellitus

Age (per 10-yr increase)

The following variables were entered into the multiple logistic regression model: age (per 10-year

10-kg decrease; diabetes; hypertension; hyperlipidemia; family history of coronary artery disease; his

percutaneous coronary intervention and prior peripheral vascular disease; pre-randomization use o

Hg decrease; ST-T changes on electrocardiography; positive cardiac biomarkers; triple-vessel disease

([re]-MI, stroke, new-onset heart failure, cardiogenic shock, TIMI [Thrombolysis In Myocardial Infarc

bleeding). Model chi-square � 367.6, degrees of freedom � 17, c-index � 0.64, Hosmer and Lemeshow go

interventions.onlinDownloaded from

Table 6, bottom). We observed that the unadjusted anddjusted risk of MI declined in the conservative medicalanagement group when events through 30 days were

ensored, but the adjusted risks of 6-month and 1-yearortality remained similar.verall outcomes. We analyzed overall mortality at 1 yearver a period of observation that included both the in-ospital and post-discharge time periods (Fig. 4). The earlyisk of death from CABG was seen in the first 90 days, but,fter that period, the risk in the medical cohort increased atgreater rate and the 2 curves cross. Thus, for those who

urvived the initial surgery, the risk became less than that ofhe medically treated patients with time. Overall mortalityas highest in the medically managed group and lowest in

he PCI group, with the difference persisting even afterdjusting for conservative medical management as a time-ependent covariate (Table 7).

iscussion

ur study showed that, in a fairly contemporary cohort ofigh-risk NSTE ACS managed with an early invasiveanagement strategy, the majority of which were from U.S.

ites, a substantial proportion of patients did not receiven-hospital revascularization despite documentation of sig-ificant coronary disease on angiography. Medically man-ged patients had a higher risk of long-term adverseutcomes despite widespread utilization of evidence-basededications beyond the initial hospitalization.se of revascularization after early catheterization. Mostbservational studies and randomized controlled trialstudying early invasive management strategies for NSTE

)

-Square Odds Ratio 95% Confidence Interval

7 1.44 1.25–1.64

1 1.10 1.05–1.14

5 0.21 0.11–0.42

0 1.33 1.17–1.50

7 1.69 1.33–2.14

7 1.48 1.24–1.77

8 1.01 1.00–1.01

0 1.42 1.17–1.74

7 1.33 1.13–1.56

0 1.22 1.08–1.36

6 1.15 1.03–1.29

3 1.07 1.01–1.13

); gender; race; nationality; smoking status; creatinine clearance (per 10 ml/min); body weight per

rior myocardial infarction (MI); prior heart failure; prior coronary artery bypass grafting (CABG); prior

, clopidogrel, warfarin, and/or beta-blockers; Killip class II to IV; systolic blood pressure per 10 mm

nary angiogram; and in-hospital events occurring from randomization to the time of catheterization

ow grade major, or GUSTO [Global Use of Strategies To Open Occluded Coronary Arteries] severe

del 1

ld Chi

27.

25.

20.

20.

18.

18.

15.

12.

11.

11.

6.

6.

increase

tory of p

f aspirin

on coro

tion] fl

odness-of-fit test, p � 0.8.

by on May 21, 2011 ejacc.org

Arpamwwircifctdw

dcomRscdnsgarpcCtvsrpsAtawiu

rolon

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 . 4 , 2 0 0 8

A U G U S T 2 0 0 8 : 3 6 9 – 7 8

Chan et al.

Conservative Management of Significant Coronary Disease for NSTE ACS

374

CS have focused on early catheterization rather than earlyevascularization as a management strategy (4–7). Theotential consequences of this approach were highlighted by recent meta-analysis of clinical trials of early invasiveanagement in NSTE ACS in which the survival benefitith an early invasive strategy was most evident when thereas proportionally greater use of revascularization in the

nvasive versus conservative treatment arms (17). While aecent analysis from the SYNERGY trial examined out-omes in NSTE ACS patients undergoing CABG forntermediate- to high-risk NSTE ACS, this analysis did notocus upon patients with angiographically defined signifi-ant coronary disease, so the medical management popula-ion was diluted with patients with insignificant coronaryisease (18). Our study demonstrates that many patientsith NSTE ACS and significant coronary disease are

Late or repeat revascularization rates after discharge*

9.0%

3.8%

7.4%

2.3%

0.2%

1.6%

0%

2%

4%

6%

8%

10%

PCI afterdischarge*

CABG afterdischarge*

CABG PCI

A

Figure 2. Late or Repeat Revascularization

*Discharge landmark time point of discharge, or 7 days if hospitalization was p

Table 5. In-Hospital Events*

MM(n � 2,633)

PCI(n � 4,294)

CABG(n � 1,298)

pValue

Death 0.9 0.5 1.2 0.027

MI 2.0 9.6 26.6 �0.001

Stroke 0.3 0.3 1.3 �0.001

Cardiogenic shock 0.6 0.9 3.5 �0.001

New onset heart failure 9.9 4 15.5 �0.001

GUSTO severe bleed 1.0 0.9 5.9 �0.001

TIMI major bleed 1.6 2.1 34.3 �0.001

*Events occurring from catheterization through discharge, or 7 days if hospitalization was pro-

longed.

uPCI � percutaneous coronary intervention; other abbreviations as in Table 4.

interventions.onlinDownloaded from

etermined to not be candidates for revascularization pro-edures, and these patients have a significantly increased riskf adverse events, despite widespread use of evidence-basededications.easons for nonuse of revascularization procedures. Pos-ible reasons for patients not undergoing in-hospital revas-ularization despite significant coronary disease includeeath before planned revascularization, unfavorable coro-ary anatomy, serious comorbidities, resource availability,ocioeconomic disparities, and patient refusal. Practiceuidelines recommend consideration of early invasive man-gement for patients with NSTE ACS who have had priorevascularization (1), but our data show that patients withrior CABG were least likely to undergo in-hospital revas-ularization after early catheterization. Patients with priorABG comprise a subgroup in which repeat revasculariza-

ion may be technically challenging due to diffuse nativeessel and graft disease (19). Despite the availability of 1ubgroup analysis of a randomized trial to guide selection ofevascularization strategies in these patients (20), the risk oferiprocedural adverse events with any revascularizationtrategy remains high for patients with prior CABG.dditionally, many unmeasured variables likely contributed

o decision-making regarding revascularization procedures,s reflected in the relatively poor c-index (0.64) of the modele developed to determine predictors of conservative med-

cal management. This matter requires prospective study tonderstand whether there are opportunities to expand the

0 50 100 150

an toI

an toG

ical management

71 (13, 123) d 71 (13, 123) d

75 (29, 124) d75 (29, 124) d

33 (13, 97) d33 (13, 97) d

1 (0, 2) d 1 (0, 2) d ††102 (48, 144) d102 (48, 144) d

27 (11, 56) d27 (11, 56) d

Time (days)

iming of late or repeat evascularization after ischarge*

ged; †Includes re-exploration. Abbreviations as in Figure 1.

Meditime

PC

MeditimeCAB

Med

Trd

B

tilization of PCI and CABG for these patients. by on May 21, 2011 ejacc.org

owCssisrcppr

cDitDiptErti

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 . 4 , 2 0 0 8

A U G U S T 2 0 0 8 : 3 6 9 – 7 8

Chan et al.

Conservative Management of Significant Coronary Disease for NSTE ACS

375

Data from the CRUSADE (Can Rapid risk stratificationf Unstable angina patients Suppress ADverse outcomesith Early implementation of the American College ofardiology/American Heart Association guidelines) trial

howed that the widespread availability of drug-elutingtents (DES) from 2003 to 2005 was associated with a sharpncrease in PCI use in patients with NSTE ACS andignificant CAD, while the rate of CABG proceduresemained relatively stable (8). This was accompanied by aoncordant decline in the proportion of medically managedatients, from 49% in 2002 to 36% in 2005, suggesting thathysicians had become more willing to offer percutaneous

Figure 3. In-Hospital and Post-Discharge Mortality

(A) Death from catheterization through discharge. (B) Death from discharge*p � 0.0001; conservative medical management versus CABG, p � 0.08. *Landmtions as in Figure 1.

evascularization options to patients who might have been rinterventions.onlinDownloaded from

onsidered poor candidates for PCI before the advent ofES. The high rate of off-label DES usage since its

ntroduction, primarily for complex lesion subsets, supportshis conclusion (21). Given recent safety concerns aboutES and uncertainty over the actual benefit of an early

nvasive strategy (22), the proportion of medically managedatients may remain stable, or actually even increase overime.vidence-based medications. The risk-treatment paradoxefers to progressive diminution in the use of evidence-basedherapies as baseline risk and future probability of deathncreases (23). However, treatment intensification in high-

gh 1 year. Log-rank p values: conservative medical management versus PCI,ime point of discharge, or 7 days if hospitalization was prolonged. Abbrevia-

throuark t

isk patients remains a major challenge because these by on May 21, 2011 ejacc.org

ptmcPUttwtIPt

stm

ibaPatedi

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 . 4 , 2 0 0 8

A U G U S T 2 0 0 8 : 3 6 9 – 7 8

Chan et al.

Conservative Management of Significant Coronary Disease for NSTE ACS

376

atients have an increased susceptibility to drug toxicity andreatment complications (24). The greatest difference inedication use among the comparison groups was with

lopidogrel, which was targeted for patients who underwentCI. A subgroup analysis of the CURE (Clopidogrel innstable Angina to Prevent Recurrent Ischemic Events)

rial showed a similar benefit of clopidogrel in patientsreated with conservative medical management comparedith the benefit seen in patients treated with revasculariza-

ion (25). However, in the recent TRITON (Trial to Assessmprovement in Therapeutic Outcomes by Optimizinglatelet Inhibition With Prasugrel) study, older patients and

hose with lower body weight were also more likely to have

Figure 4. Overall Mortality From Randomization Through 1 Year

Table 6. Post-Discharge Events*

OutcomeMM

(n � 2,633)PCI

(n � 4,294)

Adjusted HR � 95%(Mode

Adjusted for BaselineCharacteristics �In-Hospital Events

All patients

Death/MI at 6 months 11.7% 4.7% 2.01 (1.61–2.51)

MI at 6 months 7.5% 3.3% 1.99 (1.52–2.60)

Death at 1 yr 7.7% 3.6% 1.30 (0.87–1.95)

Censored patients†

Death/MI at 6 months 8.0% 4.7% 1.43 (1.10–1.86)

MI at 6 months 3.7% 3.3% 1.02 (0.72–1.44)

Death at 1 yr 8.0% 3.6% 1.45 (0.96–2.19)

*Only events occurring from discharge, or 7 days if hospitalization was prolonged, were considered; †

censored in this end point analysis. Outcomes for the in-hospital PCI and CABG groups are the sam

CI � confidence interval; HR � hazard ratio; other abbreviations as in Table 1.

Log-rank MM versus PCI, �0.0001; MM versus CABG, 0.52. Abbreviations as in Figure

interventions.onlinDownloaded from

erious bleeding complications and less likely to derive a netreatment benefit with more potent thienopyridine treat-ent (26).Similar challenges face the widespread implementation of

ntensive lipid-lowering in the medically managed populationecause of their greater prevalence of comorbid conditions. Inmeta-analysis of 4 large clinical trials that included the

ROVE-IT–TIMI 22 (Pravastatin or Atorvastatin Evaluationnd Infection Therapy–Thrombolysis In Myocardial Infarc-ion 22) trial (27), investigators confirmed the safety andfficacy of high-dose statin therapy compared with standard-ose statin therapy (28). However, patients with serum creat-

nine �2.5 mg/dl were excluded from the aforementioned

MM vs. PCI

CABG(n � 1,298)

Adjusted HR � 95% CI for MM vs. CABG(Model 2)

ted for Baselineracteristics �spital Events �ensity for MM

Adjusted for BaselineCharacteristics �

In-Hospital Events†

Adjusted for BaselineCharacteristics �

In-Hospital Events �Propensity for MM

9 (1.59–2.49) 4.9% 3.80 (2.65–5.47) 3.47 (2.40–5.02)

2 (1.46–2.52) 1.7% 6.65 (3.74–11.83) 6.27 (3.50–11.23)

2 (0.88–0.97) 6.2% 1.77 (0.98–3.19) 1.83 (1.00–3.37)

8 (1.06–1.80) 4.9% 2.67 (1.81–3.94) 2.36 (1.58–3.53)

4 (0.66–1.35) 1.7% 2.99 (1.61–5.55) 2.66 (1.41–5.00)

8 (0.97–2.26) 6.2% 2.36 (1.24–4.50) 2.54 (1.30–4.97)

ts in the conservative medical management group who underwent PCI or CABG within 30 days were

er the “All Patients” analysis in the top rows.

CI forl 2)

AdjusCha

In-HoProp

1.9

1.9

1.3

1.3

0.9

1.4

Patien

e as und

1.

by on May 21, 2011 ejacc.org

cpmmemlmmmeSawaniSdedtctittawlmboaca7tin

C

AtutoCrnapd

ATtt

R2C

R

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 . 4 , 2 0 0 8

A U G U S T 2 0 0 8 : 3 6 9 – 7 8

Chan et al.

Conservative Management of Significant Coronary Disease for NSTE ACS

377

linical trials, limiting generalizability of the findings to thisopulation. Impaired renal function is a key concern, and theajority of NSTE ACS patients with renal impairment areanaged medically (29). Hence, questions remain about the

fficacy, safety, and correct dosing of guidelines-recommendededications in medically managed patients excluded from

andmark randomized controlled trials because of excess co-orbidity. Dedicated trials with alternative dosing strategiesay be a step toward resolving concerns regarding drugetabolism and the increased likelihood of medication side

ffects in this population.tudy limitations. Several limitations were present with thisnalysis. First, detailed lesion-level angiographic findingsere not captured, so we were not able to characterize

ngiographic suitability for revascularization. Second, sig-ificant lesions in the major coronary arteries were recorded

n a dichotomized approach (�50% vs. �50%) on theYNERGY case report form, so we could not assessifferent classification schemes for significant coronary dis-ase such as stenoses �70%. Third, details about physicianecision-making regarding revascularization were not cap-ured. However, patients with clinical contraindications toatheterization and revascularization were excluded fromhe SYNERGY trial, so the study population represented andeal group of patients for early invasive management at theime of randomization (30). Fourth, despite our best effortso adjust for confounding, the conservative medical man-gement group represents a particularly high-risk cohort inhom survival is unlikely to approximate that of the

ower-risk PCI group because of unmeasured confoundersaking it difficult to draw definitive conclusions about the

enefits of early revascularization procedures. Nonetheless,ur intent was not to show that early revascularization wasssociated with improved outcomes, but rather to fullyharacterize the short- and long-term risks associated with conservative medical management strategy. Last, the-day landmark analysis, although necessary for the inves-igation of MI outcomes in this particular study, mayntroduce biases associated with left censoring of longitudi-

Table 7. Time-Dependent Model Combining In-Hospital and Post-Discharge

TreatmentGroup

Unadjusted1-Year Mortality*, %

TreatmentComparison

Adjusted for MM asTime-Dependent

Covariate†

MM 8.9

PCI 4.4 MM vs. PCI 2.30 (1.88–2.82)

CABG 8.2 MM vs. CABG 1.16 (0.91–1.49)

*All events occurring from the time of catheterization through 1 year were considered; †Conser

management versus coronary artery bypass grafting (CABG) modeled as time-dependent covariate

Abbreviations as in Tables 1 and 6.

al data.interventions.onlinDownloaded from

onclusions

lmost one-third of patients with NSTE ACS documentedo have significant coronary disease on angiography did notndergo revascularization procedures in the SYNERGYrial. These patients have a higher risk of long-term adverseutcomes compared with those who undergo PCI orABG despite widespread utilization of guidelines-

ecommended medications. These findings highlight theeed for novel strategies to mitigate the increased risk ofdverse outcomes in this population and for additionalrospective studies to examine patient and physicianecision-making regarding revascularization decisions.

cknowledgmentshe authors would like to thank the following people for

heir invaluable contributions: Amanda McMillan for edi-orial assistance and Kimberly Ferrer for graphics assistance.

eprint requests and correspondence: Dr. Matthew T. Roe,400 Pratt Street, Terrace Level, Room 0311, Durham, Northarolina 27705. E-mail: matthew.roe@duke.edu.

EFERENCES

1. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007guidelines for the management of patients with unstable angina/non–ST-elevation myocardial infarction: a report of the American Collegeof Cardiology/American Heart Association Task Force on PracticeGuidelines (Writing Committee to Revise the 2002 Guidelines for theManagement of Patients With Unstable Angina/Non–ST-ElevationMyocardial Infarction) developed in collaboration with the AmericanCollege of Emergency Physicians, the Society for CardiovascularAngiography and Interventions, and the Society of Thoracic Surgeonsendorsed by the American Association of Cardiovascular and Pulmo-nary Rehabilitation and the Society for Academic Emergency Medi-cine. J Am Coll Cardiol 2007;50:e1–157.

2. Bassand JP, Hamm CW, Ardissino D, et al. Guidelines for thediagnosis and treatment of non–ST-segment elevation acute coronarysyndromes. The Task Force for the Diagnosis and Treatment ofNon–ST-Segment Elevation Acute Coronary Syndromes of the Eu-ropean Society of Cardiology. Eur Heart J 2007;28:1598–660.

lity*

Adjusted HR � 95% CI (Model 3)

Adjusted for MM asTime-Dependent

variate � Baseline Characteristics

Adjusted for MM asTime-Dependent

Covariate � Baseline Characteristics �Propensity for MM

1.72 (1.40–2.11) 1.70 (1.37–2.10)

1.01 (0.78–1.30) 0.99 (0.77–1.28)

edical management versus percutaneous coronary intervention (PCI) and conservative medical

Morta

Co

vative m

s.

3. Shavelle DM, Parsons L, Sada MJ, French WJ, Every NR. Is there abenefit to early angiography in patients with ST-segment depression

by on May 21, 2011 ejacc.org

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

Ka

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 . 4 , 2 0 0 8

A U G U S T 2 0 0 8 : 3 6 9 – 7 8

Chan et al.

Conservative Management of Significant Coronary Disease for NSTE ACS

378

myocardial infarction? An observational study. Am Heart J 2002;143:488–96.

4. Fox KA, Poole-Wilson PA, Henderson RA, et al. Interventional versusconservative treatment for patients with unstable angina or non–ST-elevation myocardial infarction: the British Heart Foundation RITA 3randomised trial. Randomized Intervention Trial of unstable Angina.Lancet 2002;360:743–51.

5. FRagmin and Fast Revascularisation during InStability in Coronaryartery disease Investigators. Invasive compared with non-invasivetreatment in unstable coronary-artery disease: FRISC II prospectiverandomised multicentre study. Lancet 1999;354:708-15.

6. McCullough PA, O’Neill WW, Graham M, et al. A prospectiverandomized trial of triage angiography in acute coronary syndromesineligible for thrombolytic therapy. Results of the medicine versusangiography in thrombolytic exclusion (MATE) trial. J Am CollCardiol 1998;32:596–605.

7. Effects of tissue plasminogen activator and a comparison of earlyinvasive and conservative strategies in unstable angina and non–Q-wavemyocardial infarction. Results of the TIMI IIIB trial. Thrombolysis inMyocardial Ischemia. Circulation 1994;89:1545–56.

8. Gogo PB Jr., Dauerman HL, Mulgund J, et al. Changes in patterns ofcoronary revascularization strategies for patients with acute coronarysyndromes (from the CRUSADE Quality Improvement Initiative).Am J Cardiol 2007;99:1222–6.

9. Ferguson JJ, Califf RM, Antman EM, et al. Enoxaparin vs unfraction-ated heparin in high-risk patients with non–ST-segment elevationacute coronary syndromes managed with an intended early invasivestrategy: primary results of the SYNERGY randomized trial. JAMA2004;292:45–54.

0. Mahaffey KW, Cohen M, Garg J, et al. High-risk patients with acutecoronary syndromes treated with low-molecular-weight or unfraction-ated heparin: outcomes at 6 months and 1 year in the SYNERGY trial.JAMA 2005;294:2594–600.

1. Eagle KA, Lim MJ, Dabbous OH, et al. A validated prediction modelfor all forms of acute coronary syndrome: estimating the risk of6-month postdischarge death in an international registry. JAMA2004;291:2727–33.

2. Bhatt DL, Roe MT, Peterson ED, et al. Utilization of early invasivemanagement strategies for high-risk patients with non–ST-segmentelevation acute coronary syndromes: results from the CRUSADEQuality Improvement Initiative. JAMA 2004;292:2096–104.

3. Monteiro P, Portuguese Registry on Acute Coronary Syndromes. Impactof early coronary artery bypass graft in an unselected acute coronarysyndrome patient population. Circulation 2006;114 1 Suppl:I467–72.

4. Wang TY. Challenges of using peri-procedural myocardial infarctionas a quality measure among patients undergoing percutaneous coronaryintervention (abstr). Circulation 2007;116:II–X.

5. Anderson JR, Cain KC, Gelber RD. Analysis of survival by tumorresponse. J Clin Oncol 1983;1:710–9.

6. Zhou Z, Rahme E, Abrahamowicz M, et al. Survival bias associatedwith time-to-treatment initiation in drug effectiveness evaluation: acomparison of methods. Am J Epidemiol 2005;162:1016–23.

7. Bavry AA, Kumbhani DJ, Rassi AN, Bhatt DL, Askari AT. Benefit of

early invasive therapy in acute coronary syndromes: a meta-analysis of s

interventions.onlinDownloaded from

contemporary randomized clinical trials. J Am Coll Cardiol 2006;48:1319–25.

8. Chew DP, Huang Z, Pieper KS, et al. Patients with non–ST-elevationacute coronary syndromes undergoing coronary artery bypass grafting inthe modern era of antithrombotic therapy. Am Heart J 2008;155:239–44.

9. Eagle KA, Guyton RA, Davidoff R, et al. ACC/AHA 2004 guidelineupdate for coronary artery bypass graft surgery: summary article. Areport of the American College of Cardiology/American Heart Asso-ciation Task Force on Practice Guidelines (Committee to Update the1999 Guidelines for Coronary Artery Bypass Graft Surgery). J Am CollCardiol 2004;44:e213–310.

0. Morrison DA, Sethi G, Sacks J, et al. Angina With Extremely SeriousOperative Mortality Evaluation. Percutaneous coronary intervention ver-sus repeat bypass surgery for patients with medically refractory myocardialischemia: AWESOME randomized trial and registry experience withpost-CABG patients. J Am Coll Cardiol 2002;40:1951–4.

1. FDA Statement on Coronary Drug-Eluting Stents (September 14,2006). Available at: http://www.fda.gov/cdrh/news/091406.html. Ac-cessed February 19, 2008.

2. Ko DT, Mamdani M, Alter DA. Lipid-lowering therapy with statinsin high-risk elderly patients: the treatment-risk paradox. JAMA 2004;291:1864–70.

3. de Winter RJ, Windhausen F, Cornel JH, et al. Invasive versusConservative Treatment in Unstable Coronary Syndromes (ICTUS)Investigators. Early invasive versus selectively invasive management foracute coronary syndromes. N Engl J Med 2005;353:1095–104.

4. Alexander KP, Chen AY, Roe MT, et al. Excess dosing of antiplateletand antithrombin agents in the treatment of non–ST-segment eleva-tion acute coronary syndromes. JAMA 2005;294:3108–16.

5. Fox KA, Mehta SR, Peters R, et al. Benefits and risks of thecombination of clopidogrel and aspirin in patients undergoing surgicalrevascularization for non–ST-elevation acute coronary syndrome: theClopidogrel in Unstable angina to prevent Recurrent ischemic Events(CURE) trial. Circulation 2004;110:1202–8.

6. Wiviott SD, Braunwald E, McCabe CH, et al. Prasugrel versusclopidogrel in patients with acute coronary syndromes. N Engl J Med2007;357:2001–15.

7. Cannon CP, Braunwald E, McCabe CH, et al. Pravastatin or atorvastatinevaluation and infection therapy—Thrombolysis in Myocardial Infarction22 Investigators. Intensive versus moderate lipid lowering with statins afteracute coronary syndromes. N Engl J Med 2004;350:1495–504.

8. Cannon CP, Steinberg BA, Murphy SA, Mega JL, Braunwald E.Meta-analysis of cardiovascular outcomes trials comparing intensiveversus moderate statin therapy. J Am Coll Cardiol 2006;48:438–45.

9. Han JH, Chandra A, Mulgund J, et al. Chronic kidney disease inpatients with non–ST-segment elevation acute coronary syndromes.Am J Med 2006;119:248–54.

0. SYNERGY Executive Committee. The SYNERGY trial: study designand rationale. Am Heart J 2002;143:952–60.

ey Words: conservative medical management � coronaryrtery disease � non-ST-segment elevation acute coronary

yndrome � prognosis.

by on May 21, 2011 ejacc.org

doi:10.1016/j.jcin.2008.03.019 2008;1;369-378 J. Am. Coll. Cardiol. Intv.

RoeWhite, Philip E. Aylward, James J. Ferguson, Robert M. Califf, and Matthew T. Mark Y. Chan, Kenneth W. Mahaffey, Lena J. Sun, Karen S. Pieper, Harvey D.

Have Angiographically Documented Significant Coronary DiseasePatients With Non�ST-Segment Elevation Acute Coronary Syndromes Who

Prevalence, Predictors, and Impact of Conservative Medical Management for

This information is current as of May 21, 2011

& ServicesUpdated Information

http://interventions.onlinejacc.org/cgi/content/full/1/4/369including high-resolution figures, can be found at:

Supplementary Material

1http://interventions.onlinejacc.org/cgi/content/full/1/4/369/DCSupplementary material can be found at:

References

BLhttp://interventions.onlinejacc.org/cgi/content/full/1/4/369#BIfree at: This article cites 27 articles, 17 of which you can access for

Rights & Permissions

http://interventions.onlinejacc.org/misc/permissions.dtltables) or in its entirety can be found online at: Information about reproducing this article in parts (figures,

Reprints http://interventions.onlinejacc.org/misc/reprints.dtl

Information about ordering reprints can be found online:

by on May 21, 2011 interventions.onlinejacc.orgDownloaded from

CORRECTIONS

Chan MY, Mahaffey KW, Sun LJ, Pieper KS, White HD, Aylward PE, Ferguson JJ, Califf RM, RoeMT. Prevalence, Predictors, and Impact of Conservative Medical Management for Patients WithNon-ST-Segment Elevation Acute Coronary Syndromes Who Have Angiographically DocumentedSignificant Coronary Disease. J Am Coll Cardiol Intv 2008;1:369-78.

This article contained the following errors:

1. In the abbreviations section on page 370, “GRACE � Global Registry of Acute Coronary Eventsscore” should read “GRACE � Global Registry of Acute Coronary Events” without the word score.

2. On page 372, Table 1, the median age of the MM group reads “15.3 (9.5, 21.0).” It should read “69(62, 75).” The corrected table is printed below.

3. On page 373, the following sentence currently reads “The early risk of death from CABG was seen inthe first 90 days, but, after that period, the risk in the medical cohort increased at a greater rate and the2 curves cross” it should read “The early risk of death from CABG was seen in the first 90 days, but,after that period, the risk in the medical cohort increased at a greater rate and the 2 curves crossed.”

Table 1. Baseline Characteristics

CharacteristicMM

(n � 2,633)PCI

(n � 4,294)CABG

(n � 1,298) p Value

Median age, yrs 69 (62, 75) 67 (60, 74) 67 (61, 74) �0.001

Female gender, % 32.8 31.7 25.9 �0.001

Race, % �0.001

White 83.6 87.4 85.5

Black 7.3 5.4 6.3

Asian 1.9 0.7 0.8

Hispanic 4.6 4.7 5.5

U.S. enrollment, % 44.4 59.5 79.1 �0.001

Hypertension, % 70.9 65.8 70.3 �0.001

Diabetes, % 33.3 26.2 33.2 �0.001

Smoking, % 35.6 34.2 36.1 �0.001

Hypercholesterolemia, % 61.8 59.4 60.0 0.144

Prior MI, % 34.2 25.9 23.4 �0.001

Prior heart failure, % 12.2 6.5 6.8 �0.001

History of stroke, % 6.1 3.8 5.2 �0.001

History of PVD, % 11.9 8.7 9.1 �0.001

Prior CABG, % 23.0 16.1 6.7 �0.001

Prior PCI, % 22.5 21.6 15.2 �0.001

Hours from symptom onset (quartiles 1 and 3) 15.3 (9.5, 21.0) 14.4 (8.1, 20.5) 14.3 (8.0, 20.4) �0.001

Killip class II–IV, % 14.8 8.9 13.1 �0.001

Systolic BP, mm Hg 132 (118, 150) 130 (117, 146.0) 130 (117, 147) 0.033

Heart rate, beats/min 70 (62, 81) 70 (62, 80) 72 (64, 83) �0.001

Creatinine clearance, ml/min* 71.1 (53.6, 92.5) 79.2 (60.6, 101.8) 77.4 (57.8, .98.6) �0.001

ST-segment depression on ECG, % 57.5 53.0 59.6 �0.001

Elevated cardiac enzymes, %† 64.4 70.2 69.9 �0.001

Values in parentheses are 25th, 75th percentile. *Creatinine clearance calculated using the Cockroft-Gault formula; †troponin or creatine kinase-MB fraction above the upper

limit of normal.

BP � blood pressure; CABG � coronary artery bypass grafting; ECG � electrocardiogram; MI � myocardial infarction; MM � conservative medical management; PCI �

percutaneous coronary intervention; PVD � peripheral vascular disease.

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 . 5 , 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 .

by on May 21, 2011 interventions.onlinejacc.orgDownloaded from

4. On page 376, Table 6, the hazard ratio and 95% confidence interval for death at 1 year for all patients under the AdjustedHR � 95% CI for MM vs. PCI (Model 2) and Adjusted for Baseline Characteristics � In-Hospital Events � Propensityfor MM headings (row 4, column 5) reads “1.32 (0.88–0.97).” It should read “1.32 (0.88–1.97)”. The corrected table isprinted below.

We apologize for these errors.

doi:10.1016/j.jcin.2008.03.019

Brodie BR, Stuckey T, Downey W, Humphrey A, Bradshaw B, Metzger C, Hermiller J, Krainin F, Juk S, Cheek B,Duffy P, Smith H, Edmunds J, Varanasi J, Simonton CA, for the STENT (Strategic Transcatheter Evaluation of NewTherapies) Group. Outcomes and Complications With Off-Label Use of Drug-Eluting Stents Results From theSTENT (Strategic Transcatheter Evaluation of New Therapies) Group. J Am Coll Cardiol Intv 2008;1:1405–14.

In this article, there was an error in the right side of Figure 4B. A corrected Figure 4B is printed below. The authorsapologize for this error.

doi:10.1016/j.jcin.2008.06.005

Table 6. Post-Discharge Events*

OutcomeMM

(n � 2,633)PCI

(n � 4,294)

Adjusted HR � 95% CI for MM vs. PCI(Model 2)

CABG(n � 1,298)

Adjusted HR � 95% CI for MM vs. CABG(Model 2)

Adjusted for BaselineCharacteristics �In-Hospital Events

Adjusted for BaselineCharacteristics �

In-Hospital Events �Propensity for MM

Adjusted for BaselineCharacteristics �

In-Hospital Events†

Adjusted for BaselineCharacteristics �

In-Hospital Events �Propensity for MM

All patients

Death/MI at 6 months 11.7% 4.7% 2.01 (1.61–2.51) 1.99 (1.59–2.49) 4.9% 3.80 (2.65–5.47) 3.47 (2.40–5.02)

MI at 6 months 7.5% 3.3% 1.99 (1.52–2.60) 1.92 (1.46–2.52) 1.7% 6.65 (3.74–11.83) 6.27 (3.50–11.23)

Death at 1 yr 7.7% 3.6% 1.30 (0.87–1.95) 1.32 (0.88–1.97) 6.2% 1.77 (0.98–3.19) 1.83 (1.00–3.37)

Censored patients†

Death/MI at 6 months 8.0% 4.7% 1.43 (1.10–1.86) 1.38 (1.06–1.80) 4.9% 2.67 (1.81–3.94) 2.36 (1.58–3.53)

MI at 6 months 3.7% 3.3% 1.02 (0.72–1.44) 0.94 (0.66–1.35) 1.7% 2.99 (1.61–5.55) 2.66 (1.41–5.00)

Death at 1 yr 8.0% 3.6% 1.45 (0.96–2.19) 1.48 (0.97–2.26) 6.2% 2.36 (1.24–4.50) 2.54 (1.30–4.97)

*Only events occurring from discharge, or 7 days if hospitalization was prolonged, were considered; †Patients in the conservative medical management group who underwent PCI or CABG within 30 days were

censored in this end point analysis. Outcomes for the in-hospital PCI and CABG groups are the same as under the “All Patients” analysis in the top rows.

CI � confidence interval; HR � hazard ratio; other abbreviations as in Table 1.

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 . 5 , 2 0 0 8

O C T O B E R 2 0 0 8 : 6 0 0 – 1

Corrections

601

by on May 21, 2011 interventions.onlinejacc.orgDownloaded from