[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.163.129.96. Please contact the publisher to request reinstatement.
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Download PDF
Figure 1.
Extent, Distribution, and Location of Obstructive Non–Infarct-Related Artery Disease Among Patients With ST-Elevation Myocardial Infarction
Extent, Distribution, and Location of Obstructive Non–Infarct-Related Artery Disease Among Patients With ST-Elevation Myocardial Infarction

Obstructive coronary artery disease was defined as more than 50% stenosis in major epicardial coronary artery.

aThe data are from the Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) I trial,11 the GUSTO IIb trial,12 the GUSTO III trial,13 the Integrilin to Minimise Platelet Aggregation and Coronary Thrombus in Acute Myocardial Infarction (IMPACT-AMI) trial,14 the GUSTO V trial,15 the Integrilin and Tenecteplase in Acute Myocardial Infarction (INTEGRITI) trial,16 the Caldaret in ST Elevation Myocardial Infarction (CASTEMI) trial,17 and the Assessment of Pexelizumab in Acute Myocardial Infarction (APEX AMI) trial.18

bPresence or absence of obstructive coronary artery disease on each vessel territory (LAD, LCX, RCA, or LMCA) was counted. The denominators include patients at risk for non-IRA disease in the specified artery who did not have missing data on angiogram-estimated stenosis necessary to determine non-IRA status.

Figure 2.
Kaplan-Meier Curve for 30-Day Mortality Between Patients With and Without Obstructive Non–Infarct-Related Artery (Non-IRA) Disease
Kaplan-Meier Curve for 30-Day Mortality Between Patients With and Without Obstructive Non–Infarct-Related Artery (Non-IRA) Disease

Among 28 282 patients, 15 patients with missing information on mortality (6 patients were missing any indication of death and 9 did not have data on time to death) were excluded (n = 28 267).

Table 1.  
Summary of Trials Used in the Study
Summary of Trials Used in the Study
Table 2.  
Baseline Clinical and Anatomic Characteristics by Presence or Absence of Non–Infarct-Related Coronary Artery Disease (CAD)
Baseline Clinical and Anatomic Characteristics by Presence or Absence of Non–Infarct-Related Coronary Artery Disease (CAD)
Table 3.  
Unadjusted and Adjusted 30-Day Mortality for Patients With or Without Non–Infarct-Related Coronary Artery Disease
Unadjusted and Adjusted 30-Day Mortality for Patients With or Without Non–Infarct-Related Coronary Artery Disease
Supplement.

eTable 1. Baseline characteristics of patients included in study population vs those excluded

eTable 2. Observed 30-day mortality, according to the extent, distribution, and location of obstructive infarct related and non-infarct related artery disease

eTable 3. Unadjusted and adjusted 30-day mortality models according to the extent of non-infarct-related obstructive coronary artery disease in the overall cohort and in each cohort stratified by type of reperfusion trials

eTable 4. Baseline clinical and anatomic characteristics and therapeutic strategy by presence or absence of non-infarct-related obstructive coronary artery disease in the KAMIR registry

eTable 5. Unadjusted and adjusted 30-day and 1-year mortality models according to the presence or absence of non-infarct-related obstructive coronary artery disease in the KAMIR registry

eTable 6. Unadjusted and adjusted 30-day and 1-year mortality models according to the extent of non-infarct-related obstructive coronary artery disease in the KAMIR registry

eTable 7. Unadjusted and adjusted 30-day and 1-year mortality models according to the presence or absence of non-infarct-related obstructive coronary artery disease in the Duke cohort

eFigure 1. Kaplan-Meier curve for 1-year mortality between patients with and without obstructive non-infarct related artery disease

eFigure 2. Extent, distribution, and location of obstructive non-infarct related artery disease among patients presented with ST-elevation myocardial infarction in the KAMIR registry

eFigure 3. Kaplan-Meier curve for 30-day and 1-year mortality between patients with and without obstructive non-infarct related artery disease in the KAMIR registry

eFigure 4. Extent, distribution, and location of obstructive non-infarct related artery disease among patients presented with ST-elevation myocardial infarction in the Duke cardiovascular databank

eFigure 5. Kaplan-Meier curve for 30-day and 1-year mortality between patients with and without obstructive non-infarct related artery disease in the Duke cohort

1.
Virmani  R, Kolodgie  FD, Burke  AP,  et al.  Lessons from sudden coronary death. Arterioscler Thromb Vasc Biol. 2000;20(5):1262-1275.
PubMedArticle
2.
Keeley  EC, Grines  CL.  Primary coronary intervention for acute myocardial infarction. JAMA. 2004;291(6):736-739.
PubMedArticle
3.
Naghavi  M, Libby  P, Falk  E,  et al.  From vulnerable plaque to vulnerable patient. Circulation. 2003;108(14):1664-1672.
PubMedArticle
4.
Goldstein  JA, Demetriou  D, Grines  CL,  et al.  Multiple complex coronary plaques in patients with acute myocardial infarction. N Engl J Med. 2000;343(13):915-922.
PubMedArticle
5.
Kotani  J, Mintz  GS, Castagna  MT,  et al.  Intravascular ultrasound analysis of infarct-related and non-infarct-related arteries in patients who presented with an acute myocardial infarction. Circulation. 2003;107(23):2889-2893.
PubMedArticle
6.
Kim  DH, Burton  JR, Fu  Y,  et al.  What is the frequency and functional and clinical significance of complex lesions in non-infarct-related arteries after fibrinolysis for acute ST-elevation myocardial infarction? Am Heart J. 2006;151(3):668-673.
PubMedArticle
7.
Cavender  MA, Milford-Beland  S, Roe  MT,  et al.  Prevalence, predictors, and in-hospital outcomes of non-infarct artery intervention during primary percutaneous coronary intervention for ST-segment elevation myocardial infarction (from the National Cardiovascular Data Registry). Am J Cardiol. 2009;104(4):507-513.
PubMedArticle
8.
Corpus  RA, House  JA, Marso  SP,  et al.  Multivessel percutaneous coronary intervention in patients with multivessel disease and acute myocardial infarction. Am Heart J. 2004;148(3):493-500.
PubMedArticle
9.
Hannan  EL, Samadashvili  Z, Walford  G,  et al.  Culprit vessel percutaneous coronary intervention versus multivessel and staged percutaneous coronary intervention for ST-segment elevation myocardial infarction patients with multivessel disease. JACC Cardiovasc Interv. 2010;3(1):22-31.
PubMedArticle
10.
Wald  DS, Morris  JK, Wald  NJ,  et al.  Randomized trial of preventive angioplasty in myocardial infarction. N Engl J Med. 2013;369(12):1115-1123.
PubMedArticle
11.
GUSTO Investigators.  An international randomized trial comparing four thrombolytic strategies for acute myocardial infarction. N Engl J Med. 1993;329(10):673-682.
PubMedArticle
12.
Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) IIb Investigators.  A comparison of recombinant hirudin with heparin for the treatment of acute coronary syndromes. N Engl J Med. 1996;335(11):775-782.
PubMedArticle
13.
Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO III) Investigators.  A comparison of reteplase with alteplase for acute myocardial infarction. N Engl J Med. 1997;337(16):1118-1123.
PubMedArticle
14.
Ohman  EM, Kleiman  NS, Gacioch  G,  et al.  Combined accelerated tissue-plasminogen activator and platelet glycoprotein IIb/IIIa integrin receptor blockade with integrilin in acute myocardial infarction. Circulation. 1997;95(4):846-854.
PubMedArticle
15.
Topol  EJ; GUSTO V Investigators.  Reperfusion therapy for acute myocardial infarction with fibrinolytic therapy or combination reduced fibrinolytic therapy and platelet glycoprotein IIb/IIIa inhibition. Lancet. 2001;357(9272):1905-1914.
PubMedArticle
16.
Giugliano  RP, Roe  MT, Harrington  RA,  et al.  Combination reperfusion therapy with eptifibatide and reduced-dose tenecteplase for ST-elevation myocardial infarction. J Am Coll Cardiol. 2003;41(8):1251-1260.
PubMedArticle
17.
Bär  FW, Tzivoni  D, Dirksen  MT,  et al.  Results of the first clinical study of adjunctive CAldaret (MCC-135) in patients undergoing primary percutaneous coronary intervention for ST-elevation myocardial infarction. Eur Heart J. 2006;27(21):2516-2523.
PubMedArticle
18.
Armstrong  PW, Granger  CB, Adams  PX,  et al.  Pexelizumab for acute ST-elevation myocardial infarction in patients undergoing primary percutaneous coronary intervention. JAMA. 2007;297(1):43-51.
PubMedArticle
19.
Chen  KY, Rha  SW, Li  YJ,  et al.  Triple versus dual antiplatelet therapy in patients with acute ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention. Circulation. 2009;119(25):3207-3214.
PubMedArticle
20.
Eisenstein  EL, Anstrom  KJ, Kong  DF,  et al.  Clopidogrel use and long-term clinical outcomes after drug-eluting stent implantation. JAMA. 2007;297(2):159-168.
PubMedArticle
21.
Ghali  WA, Quan  H, Brant  R,  et al.  Comparison of 2 methods for calculating adjusted survival curves from proportional hazards models. JAMA. 2001;286(12):1494-1497.
PubMedArticle
22.
Thomas  L, Peterson  ED.  The value of statistical analysis plans in observational research. JAMA. 2012;308(8):773-774.
PubMedArticle
23.
Gibson  CM, Kirtane  AJ, Murphy  SA,  et al.  Distance from the coronary ostium to the culprit lesion in acute ST-elevation myocardial infarction and its implications regarding the potential prevention of proximal plaque rupture. J Thromb Thrombolysis. 2003;15(3):189-196.
PubMedArticle
24.
Wang  JC, Normand  SL, Mauri  L, Kuntz  RE.  Coronary artery spatial distribution of acute myocardial infarction occlusions. Circulation. 2004;110(3):278-284.
PubMedArticle
25.
Widimsky  P, Holmes  DR  Jr.  How to treat patients with ST-elevation acute myocardial infarction and multi-vessel disease? Eur Heart J. 2011;32(4):396-403.
PubMedArticle
26.
Toma  M, Buller  CE, Westerhout  CM,  et al.  Non-culprit coronary artery percutaneous coronary intervention during acute ST-segment elevation myocardial infarction. Eur Heart J. 2010;31(14):1701-1707.
PubMedArticle
27.
Jernberg  T, Johanson  P, Held  C,  et al.  Association between adoption of evidence-based treatment and survival for patients with ST-elevation myocardial infarction. JAMA. 2011;305(16):1677-1684.
PubMedArticle
28.
Puymirat  E, Simon  T, Steg  PG,  et al.  Association of changes in clinical characteristics and management with improvement in survival among patients with ST-elevation myocardial infarction. JAMA. 2012;308(10):998-1006.
PubMedArticle
29.
Stone  GW, Maehara  A, Lansky  AJ,  et al.  A prospective natural-history study of coronary atherosclerosis. N Engl J Med. 2011;364(3):226-235.
PubMedArticle
Original Investigation
November 19, 2014

Extent, Location, and Clinical Significance of Non–Infarct-Related Coronary Artery Disease Among Patients With ST-Elevation Myocardial Infarction

Author Affiliations
  • 1Division of Cardiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
  • 2Duke Clinical Research Institute, Durham, North Carolina
  • 3Duke Translational Medical Institute, Durham, North Carolina
  • 4University Hospitals, Leuven, Belgium
  • 5Stanford University, Stanford, California
  • 6University of Alberta, Edmonton, Canada
  • 7Chonnam National University, Gwangju, Korea
JAMA. 2014;312(19):2019-2027. doi:10.1001/jama.2014.15095
Abstract

Importance  Little information exists about the anatomical characteristics and clinical relevance of non–infarct-related artery (IRA) disease among patients with ST-segment elevation myocardial infarction (STEMI).

Objectives  To investigate the incidence, extent, and location of obstructive non-IRA disease and compare 30-day mortality according to the presence of non-IRA disease in patients with STEMI.

Design, Setting, and Participants  Retrospective study of patients pooled from a convenience sample of 8 independent, international, randomized STEMI clinical trials published between 1993 and 2007. Follow-up varied from 1 month to 1 year. Among 68 765 patients enrolled in the trials, 28 282 patients with valid angiographic information were included in this analysis. Obstructive coronary artery disease was defined as stenosis of 50% or more of the diameter of a major epicardial artery. To assess the generalizability of trial-based results, external validation was performed using observational data for patients with STEMI from the Korea Acute Myocardial Infarction Registry (KAMIR) (between November 1, 2005, and December 31, 2013; n = 18 217) and the Duke Cardiovascular Databank (between January 1, 2005, and December 31, 2012; n = 1812).

Main Outcomes and Measures  Thirty-day mortality following STEMI.

Results  Overall, 52.8% (14 929 patients) had obstructive non-IRA disease; 29.6% involved 1 vessel and 18.8% involved 2 vessels. There was no substantial difference in the extent and distribution of non-IRA disease according to the IRA territory. Unadjusted and adjusted rates of 30-day mortality were significantly higher in patients with non-IRA disease than in those without non-IRA disease (unadjusted, 4.3% vs 1.7%, respectively; risk difference, 2.7% [95% CI, 2.3% to 3.0%], P < .001; and adjusted, 3.3% vs 1.9%, respectively; risk difference, 1.4% [95% CI, 1.0% to 1.8%], P < .001). The overall prevalence and association of non-IRA disease with 30-day mortality was consistent with findings from the KAMIR registry (adjusted, 3.6% for patients with non-IRA disease vs 2.5% in those without it; risk difference, 1.1% [95% CI, 0.6% to 1.7%]; P < .001), but not with the Duke database (adjusted, 4.7% with non-IRA disease vs 4.3% without it; risk difference, 0.4% [95% CI, −1.4% to 2.2%], P = .65).

Conclusions and Relevance  In a retrospective pooled analysis of 8 clinical trials, obstructive non-IRA disease was common among patients presenting with STEMI, and was associated with a modest statistically significant increase in 30-day mortality. These findings require confirmation in prospectively designed studies, but raise questions about the appropriateness and timing of non-IRA revascularization in patients with STEMI.

Introduction

Acute ST-segment elevation myocardial infarction (STEMI) is the leading cause of sudden cardiac death and typically arises from thrombotic occlusion of a coronary artery.1,2 Given that plaque instability is caused by a pathophysiological inflammation process and diffuse nature of coronary atherosclerosis exists in acute MI,3 important angiographic information for patients presenting with STEMI is not only about the status of infarct-related artery (IRA) but also about the atherosclerotic disease burden and disease severity of non-IRA vessels.

Previous studies have reported that approximately 40% to 65% of patients presenting with STEMI had multivessel coronary artery disease (CAD; ie, concomitant obstructive non-IRA disease)47 and showed conflicting results with conservative or aggressive treatment for non-IRA disease.810 Because the presence of significant non-IRA disease in patients with STEMI has been found to be associated with recurrent angina, repeat revascularization, and worse prognosis, a comprehensive estimation of the extent, severity, and location of obstructive non-IRA disease may be clinically important information for physicians in the development of optimal treatment strategies. However, large-sized, systematic data on the anatomic status and clinical relevance of obstructive non-IRA disease are lacking. Therefore, the purpose of our study was to (1) determine the prevalence, extent, distribution, and location of obstructive non-IRA disease in patients with STEMI and (2) evaluate 30-day mortality following STEMI according to the presence or absence of non-IRA disease in a large database of clinical trials with external validation using 2 observational registries, which represent the real-world population with STEMI.

Methods
Study Population and Outcome Measures

Patients were pooled from a convenience sample of 8 independent, international, previously published randomized STEMI clinical trials (Global Use of Strategies to Open Occluded Coronary Arteries [GUSTO] I,11 GUSTO IIb,12 GUSTO III,13 Integrilin to Minimise Platelet Aggregation and Coronary Thrombus in Acute Myocardial Infarction [IMPACT-AMI],14 GUSTO V,15 Integrilin and Tenecteplase in Acute Myocardial Infarction [INTEGRITI],16 Caldaret in ST Elevation Myocardial Infarction [CASTEMI],17 and Assessment of Pexelizumab in Acute Myocardial Infarction [APEX AMI]18) between 1993 and 2007 whose databases are maintained at the Duke Clinical Research Institute, Durham, North Carolina. Detailed information on study designs, methods, and primary results were previously published.1118 Summary data from each trial used in this study appear in Table 1. For the analysis, demographic characteristics, clinical risk factors or coexisting conditions, cardiac manifestations, measures of the patient’s hemodynamic status, left ventricular function, angiographic data on IRA and non-IRA territories, in-hospital cardioactive medications, and mortality data were used as recorded in the database for each clinical trial. Follow-up varied from 1 month to 1 year. The current analysis was performed as part of subanalyses of the Duke Clinical Research Institute clinical trials database and was approved by the Duke University institutional review board, with a waiver of the requirement for written informed consent.

Patients were excluded if they (1) had previously undergone coronary artery bypass graft surgery, (2) had missing information on IRA disease definition, (3) had left main IRA, (4) had IRA not located in left anterior descending artery (LAD), left circumflex artery (LCX), or right coronary artery (RCA), (5) had nonsignificant IRA disease (diameter stenosis <50%), or (6) had inadequate information to classify non-IRA disease. Obstructive CAD was defined as stenosis of 50% or more of the diameter of the major epicardial artery. The disease extent of non-IRA CAD was defined as the sum of the number of coexistent obstructive coronary arteries involved among the LAD, the LCX, the RCA, or the left main coronary artery (LMCA). If stenosis was greater than 50% in the LMCA, then this was counted as obstructive disease in 2 vessels (ie, in place of the LAD and the LCX). Irrespective of IRA and non-IRA territory, nonsignificant or mild coronary artery stenosis (<50% diameter stenosis) in all vessels was disregarded.

The primary end point of the study was all-cause mortality within 30 days of enrollment. Information on 1-year mortality was only available in the 3 fibrinolytic trials (GUSTO I, GUSTO IIb, and GUSTO III). To determine whether the presence of non-IRA CAD was associated with long-term mortality, 1-year mortality in this limited population was also provided.

To assess the generalizability of trial-based results in real-world settings, external validation was performed using observational data from the Korea Acute Myocardial Infarction Registry (KAMIR)19 and the Duke Cardiovascular Databank.20 Briefly, KAMIR is a nationwide, prospective, multicenter online registry designed to reflect the real-world practice in Korean patients presenting with acute MI during the era of drug-eluting stent use, with support from the Korean Circulation Society since November 2005. A total of 71 university or community hospitals that are high-volume centers with facilities for primary percutaneous coronary intervention (PCI) and onsite cardiac surgery have participated in this national registry. In addition, we evaluated data from consecutive patients with STEMI who had cardiac catheterization at Duke University Medical Center between January 1, 2005, and December 31, 2012.

Statistical Analysis

The prevalence, extent, and locations of obstructive non-IRA disease were summarized by the IRA territory of the LAD, the LCX, or the RCA. Baseline characteristics, including patient demographics, risk factors, clinical presentation, cardiac status, in-hospital medications, and anatomic characteristics were described for patients with and without non-IRA disease. Categorical variables were presented as counts (proportions) and continuous variables were presented as medians (interquartile ranges).

The number and percentage of patients who had died at 30 days (entire study population) and at 1 year (limited study population from only GUSTO I, GUSTO IIb, and GUSTO III trials) were presented for patients with and without non-IRA disease. To describe risk associated with the presence of non-IRA disease, Cox proportional hazards models of 30-day and 1-year mortality were used. After unadjusted analyses were initially performed, multivariable Cox regression analyses were performed to adjust potential confounders identified by the investigators using a literature search and based on data available across all trials in our study. These covariates included age, sex, body mass index (calculated as weight in kilograms divided by height in meters squared), systolic blood pressure, diastolic blood pressure, heart rate, prior MI, prior PCI, geographic region, smoking status, race, and baseline Killip class. Time-dependent variables on the pathway between CAD and mortality, such as subsequent interventions, were not explored.

Continuous variables were checked for linear associations with mortality, and spline transformations were used to account for nonlinearity of these adjustment covariates whenever appropriate. The risk associated with the presence of non-IRA disease was described by the hazard ratio and corresponding 95% confidence interval. Risk differences and corresponding 95% confidence intervals were computed for unadjusted results using Kaplan-Meier methods; for adjusted models, the corrected group prognosis method was used.21 Kaplan-Meier event rate plots are provided to describe survival over time from randomization.

These data consisted of multiple trials that were conducted at different times. To account for differences in treatment, trial enrollment criteria, and changes in standards of care over time, all models (unadjusted and adjusted) were stratified by clinical trial of origin. A likelihood ratio test assessed heterogeneity of the estimated relationship between non-IRA disease and 30-day mortality across different studies by comparing a model with non-IRA disease by trial interactions with the simplified model assuming a homogenous association. We observed homogeneity in the estimated relationship (P = .52); therefore, it was appropriate to conduct a pooled, patient-level, aggregate analysis stratified by trial. We also assessed interactions between non-IRA disease and date of randomization to explore whether the relationship between the presence of non-IRA disease and mortality has changed over the time frame in which the trials were conducted.

To determine whether the relationship between the presence of non-IRA CAD and 30-day mortality was consistent regardless of primary reperfusion therapy for STEMI, we conducted sensitivity analyses according to the type of reperfusion trials (a cohort of fibrinolytic trials and a cohort of primary PCI trials). Interactions by non-IRA disease severity explored whether increased severity of non-IRA disease was further associated with mortality first by the number of obstructive non-IRA disease vessels and second by the maximal stenosis of the nonculprit regions.

A complete case analysis was conducted. Due to missing data on various clinical covariates, the fully adjusted models in the overall cohort included 25 454 patients for 30-day mortality and the limited cohort (GUSTO I, GUSTO IIb, and GUSTO III) with available 1-year information included 20 484 patients for 1-year mortality. The variables with the most missing data (>1% missing) were body mass index (4.2%), blood pressure (1.8%), heart rate (1.2%), and race (2.4%).

This is post hoc secondary analysis. To carefully define the population of interest and to minimize data-dredging processes, we prespecified study objectives, a hypothesis, and a statistical approach using a statistical analysis plan.22 All reported P values are 2-sided, and those less than .05 were considered statistically significant. For all statistical analyses, SAS version 9.2 (SAS Institute Inc) was used.

Results
Prevalence, Extent, and Distribution of Obstructive Non-IRA Disease

Of the 68 765 patients with STEMI enrolled in the 8 trials of merged data sets, 28 282 (41%) had valid angiographic data and met the study inclusion criteria (Figure 1). Baseline characteristics of the patients included vs those excluded in the study sample are summarized in eTable 1 in the Supplement.

Among the 28 282 patients included in the study, 11 647 (41.2%) had an infarct on the LAD territory, 3502 (12.4%) on the LCX territory, and 13 133 (46.4%) on the RCA territory. In all patients, 14 929 (52.8%) had obstructive non-IRA disease; 8367 (29.6%) had involvement with 1 vessel and 5328 (18.8%) had involvement with 2 vessels. In addition, the exact data required to determine whether non-IRA disease involved 1 or 2 vessels were missing for 1234 patients (4.4%).

There were no substantial differences in the extent of obstructive non-IRA disease according to the IRA location (Figure 1). On average, approximately one-third of patients had non-IRA disease in each epicardial territory beyond the IRA territory. Among patients with IRA disease of the LAD territory and non-IRA CAD, 32.7% were on the LCX territory and 35.6% were on the RCA territory. This pattern was similar in patients with IRA disease who had the infarct on the LCX or the RCA territory. In total, 3.1% of patients had non-IRA disease of the LMCA. Overall, there was no predisposition regarding location and distribution of non-IRA disease according to the IRA location.

Patient Characteristics According to Obstructive Non-IRA Disease

The baseline clinical and anatomical characteristics of patients with and without non-IRA disease appear in Table 2. Compared with patients without non-IRA disease, those with non-IRA disease were generally older; more likely to be men; and more likely to have diabetes, hypertension, or hypercholesterolemia; and to have prior history of MI, congestive heart failure, coronary angioplasty, or cerebrovascular accident. The time from onset of qualifying symptom to randomization and the time from onset of qualifying symptom to reperfusion therapy was longer in patients with non-IRA disease than in those without non-IRA disease. Compared with patients without non-IRA disease, those with non-IRA disease had a lower ejection fraction and a higher baseline Killip class. Use of in-hospital cardiac-related medications was comparable between the 2 groups (with vs without non-IRA disease). There were no considerable differences in anatomic location and maximal stenosis of IRA disease between patients with vs without non-IRA disease.

Mortality According to Obstructive Non-IRA Disease

A total of 28 267 patients had available mortality data. The unadjusted and adjusted rates of 30-day mortality by non-IRA disease (with vs without) in the overall cohort and in each cohort stratified by type of reperfusion trials appear in Table 3. Overall, patients with non-IRA disease had a higher unadjusted 30-day mortality vs those without non-IRA disease (4.3% vs 1.7%, respectively; risk difference, 2.7% [95% CI, 2.3%-3.0%], P < .001; Figure 2). After multivariable adjustment for clinical covariates, non-IRA disease was significantly associated with an increased risk of 30-day mortality (3.3% vs 1.9% without non-IRA disease; risk difference, 1.4% [95% CI, 1.0%-1.8%]; P < .001). When we assessed the relationship between time of clinical trials and the association of non-IRA disease with mortality, there was no significant interaction between the presence of non-IRA disease and date of randomization for 30-day mortality (P = .36). Observed rates of 30-day mortality according to the extent, distribution, and location of obstructive IRA and non-IRA disease appear in eTable 2 in the Supplement.

In sensitivity analyses according to the type of reperfusion trials, non-IRA disease was also significantly associated with adjusted risks of 30-day mortality in a cohort of thrombolytic trials (3.0% vs 1.8% without non-IRA disease; risk difference, 1.3% [95% CI, 0.9%-1.7%]) and in a cohort of primary PCI trials (4.5% vs 2.5%, respectively; risk difference, 2.0% [95% CI, 1.0%-2.9%]). The relationship between the extent of non-IRA disease and mortality was also assessed (eTable 3 in the Supplement). In unadjusted and adjusted analyses, there was an incremental relationship with mortality according to the extent of non-IRA disease.

We conducted additional analyses to assess whether maximal stenosis of non-IRA disease was further associated with increased risk of mortality. In an adjusted model, total occlusion in patients with non-IRA disease was associated with increased risk of 30-day mortality vs patients without non-IRA disease (4.0% vs 1.9%, respectively; risk difference, 2.1% [95% CI, 1.6%-2.7%]). Lack of total occlusion in patients with non-IRA disease was associated with less mortality risk, but risk was still significantly increased (2.7% vs 1.9% patients without non-IRA disease; risk difference, 0.8% [95% CI, 0.3%-1.2%]).

In the limited population of the GUSTO I, GUSTO IIb, and GUSTO III trials, in which 1-year follow-up data were available (n = 23 014), patients with non-IRA disease had a higher unadjusted 1-year mortality rate than those without non-IRA CAD (7.0% vs 3.0%, respectively; risk difference, 3.9% [95% CI, 3.3%-4.5%]; eFigure 1 in the Supplement). After multivariable adjustment, the presence of non-IRA disease was significantly associated with an increase of 1-year mortality (5.4% vs 3.5% with absence of non-IRA disease; risk difference, 1.9% [95% CI, 1.3%-2.5%]).

External Validation Using Observational Registries

Between November 1, 2005, and December 31, 2013, a total of 40 851 patients with acute MI were enrolled in the KAMIR registry. The follow-up was performed through February 28, 2014. Among these patients, 18 217 with STEMI and who had valid angiographic information were used as an external validation sample for same-pattern analyses such as trial-based results (eFigure 2 in the Supplement). Among the 18 217 patients with STEMI, 9523 (52.3%) had an infarct on the LAD territory, 1795 (9.9%) on the LCX territory, and 6899 (37.9%) on the RCA territory. Similar to the trial-based findings, 9416 patients (51.7%) had non-IRA disease, 5365 (29.5%) had involvement with 1 vessel, and 4051 (22.2%) had involvement with 2 vessels.

The location of obstructive non-IRA disease was relatively balanced in each IRA territory. Baseline clinical and anatomical characteristics of patients with and without non-IRA disease appear in eTable 4 in the Supplement. Consistent with the trial-based findings, patients with non-IRA disease had a significantly higher adjusted 30-day mortality (3.6% vs 2.5% those without non-IRA disease; risk difference, 1.1% [95% CI, 0.6%-1.7%], P < .001) and adjusted 1-year mortality (5.4% vs 4.0%, respectively; risk difference, 1.4% [95% CI, 0.7%-2.0%]) (eTable 5 and eFigure 3 in the Supplement). There was also an incremental relationship with mortality according to the extent of non-IRA disease (eTable 6 in the Supplement).

Between January 1, 2005, and December 31, 2012, 41 432 patients received coronary catheterization at Duke University Medical Center. Among them, 1812 patients with STEMI were eligible for the current analysis (eFigure 4 in the Supplement). Similar to trial-based results, 753 patients (41.6%) had infarct on the LAD, 270 (14.9%) on the LCX territory, and 789 (43.5%) on the RCA territory. Coinciding with trial-based findings, overall incidence of obstructive non-IRA disease was 53.4% (968/1812) of patients; 31.5% (n = 570) had involvement with 1 vessel and 22.0% (n = 398) had involvement with 2 vessels. There was not a significant difference for 30-day mortality between patients with vs without non-IRA disease (adjusted 30-day mortality, 4.7% vs 4.3%, respectively; risk difference, 0.4% [95% CI, −1.4% to 2.2%]; P = .65) (eTable 7 and eFigure 5 in the Supplement). However, patients with non-IRA disease had a significantly higher 1-year mortality vs those without non-IRA disease (adjusted 1-year mortality, 11.5% vs 7.6%, respectively; risk difference, 3.9% [95% CI, 1.3% to 6.5%]; P = .004).

Discussion

In a pooled analysis of 8 clinical trials, we found that 53% of patients had non-IRA disease and there were no substantial differences in the extent and location of non-IRA disease according to the IRA territory. The presence of non-IRA disease was significantly associated with increased 30-day mortality. These findings were also validated in the KAMIR registry (1 of the 2 observational registries representing real-world patients with STEMI), but there was no difference in 30-day mortality found in the Duke registry.

The primary purpose of this study was to systematically evaluate the angiographic extent and distribution of non-IRA disease according to the culprit IRA location. Similar to findings from the current study, previous studies46 suggested that combined unstable coronary lesions in multiple vessels were common and they might often involve a larger area beyond the culprit artery in acute MI. Although there have been several reports determining anatomic and spatial characteristics of the culprit lesion in patients with STEMI,23,24 information regarding anatomic extent, distribution, and location of lesions occurring in noninfarct arteries is lacking. In the present study, we found that the location of non-IRA disease was relatively balanced without a strong and predominant propensity to the LAD territory. However, because mapping the distribution of angiographic sites of noninfarct lesions along each coronary tree was not available in this data set, we could not determine whether non-IRA disease had a tendency to cluster within the proximal tracts of coronary vessels such as the infarct lesions.

A second objective of the study was to determine the relationship between the presence of non-IRA disease and mortality in patients with STEMI. In this study, the presence of non-IRA disease was significantly associated with an increased risk of 30-day mortality. The elevated risk of mortality associated with non-IRA disease was maintained for up to 1 year in the subpopulation with available 1-year follow-up. This relationship between non-IRA disease and mortality was also consistent in the observational KAMIR registry, but there was no difference in 30-day mortality found in the Duke registry. However, we cannot address whether the presence of non-IRA disease has direct causality for mortality or if it functions as a marker of more severe coronary atherosclerosis and combined high-risk clinical comorbidity. In addition, our study does not delineate the exact mechanism linking non-IRA disease and mortality in patients with STEMI. Further studies are needed to understand the downstream effects and mechanisms of non-IRA disease leading to increased mortality.

Conflicting data exist regarding treatment strategies of non-IRA disease in STEMI and its clinical effect on outcome,79,25 and there is no definite, large randomized trial to answer this clinically important question. Because valid information regarding treatment of non-IRA disease is limited in these data sets, the present study does not allow us to speculate whether the relationship between non-IRA disease and mortality would be considerably different according to several strategies of non-IRA treatment. A secondary analysis of the APEX AMI trial, which is one of the trials merged in this study, showed that an acute multivessel intervention was significantly associated with an increased risk of 90-day mortality.26 By contrast, in the Preventive Angioplasty in Acute Myocardial Infarction trial,10 immediate preventive PCI for non-IRA disease significantly reduced the risk of adverse cardiovascular events compared with PCI limited to the IRA. This clinical issue warrants further investigation to determine the best clinical management, which would ideally be confirmed through large, randomized clinical trials with long-term follow-up.

Potential limitations of this study warrant discussion. First, as a retrospective observational analysis, residual confounding or selection bias cannot be completely excluded. In addition, because this is a secondary data analysis, results should be considered as hypothesis generating. Second, because the database merged several clinical trials, interstudy variability in care may exist that could have influenced the results. To account for differences in treatment, as well as changes in standard of care over time, clinical trial of origin was included as a stratification variable in all models; however, because the merged databases in the trials were accrued over decades, during which adjunctive pharmacological drugs, use of procedure, and guidelines for diagnostic and therapeutic strategies have all evolved, we cannot exclude the possibility that disease patterns, clinical presentations, and the prognosis for the STEMI population have changed over time.27,28

To address these limitations, external validation was attempted using 2 recent observational registries. The prevalence of non-IRA disease was similar in the 2 registries. However, the association of non-IRA disease with 30-day mortality was significant in the KAMIR registry but not in the Duke registry. The lack of confirmatory validation regarding 30-day mortality in the Duke registry might be due either to limited power for a short-term mortality effect of non-IRA disease or differential care patterns present at a single center.

Third, because catheterization was not mandated as part of the protocols of the trials, patient selection may have introduced a potential referral bias and survival bias. Although the clinical scenario of patients seen in the catheterization laboratory with non-IRA disease is well addressed by the current analysis, the present report still represents approximately 40% of the patients in the combined database. Fourth, atherosclerotic plaques that lead to acute MI often occur at sites of angiographically mild stenosis.29 However, because we excluded patients with nonsignificant CAD (diameter stenosis <50%), we cannot evaluate the extent and prognostic value of such lesions in patients with STEMI. Although the arbitrary definition of 50% diameter stenosis on angiography might not reflect functionally significant stenosis of CAD, it was used because it represents a sensitive convention for significant obstructive coronary disease. In addition, because this database did not have valid information on applied treatments for non-IRA disease, we were unable to address the optimal therapeutic strategy for patients with STEMI and non-IRA disease.

Conclusions

In a retrospective pooled analysis of 8 clinical trials, obstructive non-IRA disease was common among patients presenting with STEMI, and was associated with a modest statistically significant increase in 30-day mortality. These findings require confirmation in prospectively designed studies, but raise questions about the appropriateness and timing of non-IRA revascularization in patients with STEMI.

Back to top
Article Information

Corresponding Author: Manesh R. Patel, MD, Duke Clinical Research Institute, Duke University Medical Center, 2400 Pratt St, Durham, NC 27715 (manesh.patel@duke.edu).

Author Contributions: Drs Park and Patel had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Park, Pieper, Ohman, Harrington, Armstrong, Granger, Jeong, Patel.

Acquisition, analysis, or interpretation of data: Park, Clare, Schulte, Pieper, Shaw, Califf, de Werf, Hirji, Harrington, Jeong, Patel.

Drafting of the manuscript: Park, Clare, Armstrong, Jeong.

Critical revision of the manuscript for important intellectual content: Park, Clare, Schulte, Pieper, Shaw, Califf, Ohman, de Werf, Hirji, Harrington, Granger, Jeong, Patel.

Statistical analysis: Park, Clare, Schulte, Shaw, Jeong.

Obtained funding: Harrington, Jeong.

Administrative, technical, or material support: Park, Califf, Ohman, Harrington, Jeong.

Study supervision: Park, Pieper, de Werf, Hirji, Harrington, Armstrong, Jeong, Patel.

Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Park reported receiving grants from the Cardiovascular Research Foundation (Seoul, Korea) outside this study. Ms Pieper reported receiving personal fees from AstraZeneca and GlaxoSmithKline. Dr Califf reported being a board member of Portola; receiving personal income for serving as a consultant (however, 100% of this income is donated to a non-profit organization) from the following: Novartis, Johnson & Johnson-Scios, Bayer, Roche, Bristol-Myers Squibb, Regeneron, CV Sight, DSI-Lilly, Gambro, Heart.org, Kowa, Genentech, GlaxoSmithKline, Les Laboratories Servier, Amgen, and Regado NJ; receiving research grants or contracts awarded to Duke University, which support his salary, from the following: Amylin, Johnson & Johnson-Scios, Merck/Schering Plough, Novartis, Bristol-Myers Squibb Foundation, and Eli Lilly & Co; receiving income from Portola; and holding equity in N-30 Therapeutics. Dr Harrington reported receiving grant funding from AstraZeneca, Bristol-Myers Squibb, Merck, GlaxoSmithKline, Portola, Regado, The Medicines Company, sanofi-aventis, and Novartis; and receiving personal fees from Merck, The Medicines Company, Amgen, Gilead, Novartis, Medtronic, and MyoKardia. Dr Armstrong reported receiving grant support and personal fees from Boehringer Ingelheim and Merck; receiving personal fees from Eli Lilly, Roche, and AstraZeneca; and grant support from GlaxoSmithKline. Dr Granger reported receiving grant support and personal fees from Boehringer Ingelheim, Bristol-Myers Squibb, GlaxoSmithKline, Pfizer, sanofi-aventis, Takeda, The Medicines Company, AstraZeneca, Daiichi Sankyo, and Janssen Pharmaceuticals; personal fees from Hoffmann-LaRoche, Eli Lilly, Ross Medical Corporation, Salix Pharmaceuticals, and Gilead; and grant support from Medtronic Foundation, Merck, and Bayer. Dr Patel reported serving as a consultant to Bayer; receiving grant support from Johnson & Johnson, the National Heart, Lung, and Blood Institute, the Agency for Healthcare Quality, AstraZeneca, and GlaxoSmithKline; and serving on advisory boards for Genzyme, Janssen, Otsuka, and AstraZeneca. No other disclosures were reported.

Funding/Support: This study was supported in part by the John Bush Simson Fund (awarded to Dr Patel). The statistical portion of the manuscript was funded by the Duke Clinical Research Institute.

Role of the Funders/Sponsors: The John Bush Simson Fund and the Duke Clinical Research Institute had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

A full list of study investigators and coordinators for each trial has been published previously.1118

References
1.
Virmani  R, Kolodgie  FD, Burke  AP,  et al.  Lessons from sudden coronary death. Arterioscler Thromb Vasc Biol. 2000;20(5):1262-1275.
PubMedArticle
2.
Keeley  EC, Grines  CL.  Primary coronary intervention for acute myocardial infarction. JAMA. 2004;291(6):736-739.
PubMedArticle
3.
Naghavi  M, Libby  P, Falk  E,  et al.  From vulnerable plaque to vulnerable patient. Circulation. 2003;108(14):1664-1672.
PubMedArticle
4.
Goldstein  JA, Demetriou  D, Grines  CL,  et al.  Multiple complex coronary plaques in patients with acute myocardial infarction. N Engl J Med. 2000;343(13):915-922.
PubMedArticle
5.
Kotani  J, Mintz  GS, Castagna  MT,  et al.  Intravascular ultrasound analysis of infarct-related and non-infarct-related arteries in patients who presented with an acute myocardial infarction. Circulation. 2003;107(23):2889-2893.
PubMedArticle
6.
Kim  DH, Burton  JR, Fu  Y,  et al.  What is the frequency and functional and clinical significance of complex lesions in non-infarct-related arteries after fibrinolysis for acute ST-elevation myocardial infarction? Am Heart J. 2006;151(3):668-673.
PubMedArticle
7.
Cavender  MA, Milford-Beland  S, Roe  MT,  et al.  Prevalence, predictors, and in-hospital outcomes of non-infarct artery intervention during primary percutaneous coronary intervention for ST-segment elevation myocardial infarction (from the National Cardiovascular Data Registry). Am J Cardiol. 2009;104(4):507-513.
PubMedArticle
8.
Corpus  RA, House  JA, Marso  SP,  et al.  Multivessel percutaneous coronary intervention in patients with multivessel disease and acute myocardial infarction. Am Heart J. 2004;148(3):493-500.
PubMedArticle
9.
Hannan  EL, Samadashvili  Z, Walford  G,  et al.  Culprit vessel percutaneous coronary intervention versus multivessel and staged percutaneous coronary intervention for ST-segment elevation myocardial infarction patients with multivessel disease. JACC Cardiovasc Interv. 2010;3(1):22-31.
PubMedArticle
10.
Wald  DS, Morris  JK, Wald  NJ,  et al.  Randomized trial of preventive angioplasty in myocardial infarction. N Engl J Med. 2013;369(12):1115-1123.
PubMedArticle
11.
GUSTO Investigators.  An international randomized trial comparing four thrombolytic strategies for acute myocardial infarction. N Engl J Med. 1993;329(10):673-682.
PubMedArticle
12.
Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) IIb Investigators.  A comparison of recombinant hirudin with heparin for the treatment of acute coronary syndromes. N Engl J Med. 1996;335(11):775-782.
PubMedArticle
13.
Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO III) Investigators.  A comparison of reteplase with alteplase for acute myocardial infarction. N Engl J Med. 1997;337(16):1118-1123.
PubMedArticle
14.
Ohman  EM, Kleiman  NS, Gacioch  G,  et al.  Combined accelerated tissue-plasminogen activator and platelet glycoprotein IIb/IIIa integrin receptor blockade with integrilin in acute myocardial infarction. Circulation. 1997;95(4):846-854.
PubMedArticle
15.
Topol  EJ; GUSTO V Investigators.  Reperfusion therapy for acute myocardial infarction with fibrinolytic therapy or combination reduced fibrinolytic therapy and platelet glycoprotein IIb/IIIa inhibition. Lancet. 2001;357(9272):1905-1914.
PubMedArticle
16.
Giugliano  RP, Roe  MT, Harrington  RA,  et al.  Combination reperfusion therapy with eptifibatide and reduced-dose tenecteplase for ST-elevation myocardial infarction. J Am Coll Cardiol. 2003;41(8):1251-1260.
PubMedArticle
17.
Bär  FW, Tzivoni  D, Dirksen  MT,  et al.  Results of the first clinical study of adjunctive CAldaret (MCC-135) in patients undergoing primary percutaneous coronary intervention for ST-elevation myocardial infarction. Eur Heart J. 2006;27(21):2516-2523.
PubMedArticle
18.
Armstrong  PW, Granger  CB, Adams  PX,  et al.  Pexelizumab for acute ST-elevation myocardial infarction in patients undergoing primary percutaneous coronary intervention. JAMA. 2007;297(1):43-51.
PubMedArticle
19.
Chen  KY, Rha  SW, Li  YJ,  et al.  Triple versus dual antiplatelet therapy in patients with acute ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention. Circulation. 2009;119(25):3207-3214.
PubMedArticle
20.
Eisenstein  EL, Anstrom  KJ, Kong  DF,  et al.  Clopidogrel use and long-term clinical outcomes after drug-eluting stent implantation. JAMA. 2007;297(2):159-168.
PubMedArticle
21.
Ghali  WA, Quan  H, Brant  R,  et al.  Comparison of 2 methods for calculating adjusted survival curves from proportional hazards models. JAMA. 2001;286(12):1494-1497.
PubMedArticle
22.
Thomas  L, Peterson  ED.  The value of statistical analysis plans in observational research. JAMA. 2012;308(8):773-774.
PubMedArticle
23.
Gibson  CM, Kirtane  AJ, Murphy  SA,  et al.  Distance from the coronary ostium to the culprit lesion in acute ST-elevation myocardial infarction and its implications regarding the potential prevention of proximal plaque rupture. J Thromb Thrombolysis. 2003;15(3):189-196.
PubMedArticle
24.
Wang  JC, Normand  SL, Mauri  L, Kuntz  RE.  Coronary artery spatial distribution of acute myocardial infarction occlusions. Circulation. 2004;110(3):278-284.
PubMedArticle
25.
Widimsky  P, Holmes  DR  Jr.  How to treat patients with ST-elevation acute myocardial infarction and multi-vessel disease? Eur Heart J. 2011;32(4):396-403.
PubMedArticle
26.
Toma  M, Buller  CE, Westerhout  CM,  et al.  Non-culprit coronary artery percutaneous coronary intervention during acute ST-segment elevation myocardial infarction. Eur Heart J. 2010;31(14):1701-1707.
PubMedArticle
27.
Jernberg  T, Johanson  P, Held  C,  et al.  Association between adoption of evidence-based treatment and survival for patients with ST-elevation myocardial infarction. JAMA. 2011;305(16):1677-1684.
PubMedArticle
28.
Puymirat  E, Simon  T, Steg  PG,  et al.  Association of changes in clinical characteristics and management with improvement in survival among patients with ST-elevation myocardial infarction. JAMA. 2012;308(10):998-1006.
PubMedArticle
29.
Stone  GW, Maehara  A, Lansky  AJ,  et al.  A prospective natural-history study of coronary atherosclerosis. N Engl J Med. 2011;364(3):226-235.
PubMedArticle
×